70912 A WORLD BANK STUDY Concentrating Solar Power in Developing Countries R E G U L AT O R Y A N D F I N A N C I A L I N C E N T I V E S FOR SCALING UP Natalia Kulichenko and Jens Wirth A W O R L D B A N K S T U D Y Concentrating Solar Power in Developing Countries Regulatory and Financial Incentives for Scaling Up Natalia Kulichenko and Jens Wirth © 2012 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1234 15 14 13 12 World Bank Studies are published to communicate the results of the Bank’s work to the development community with the least possible delay. The manuscript of this paper therefore has not been prepared in accordance with the procedures appropriate to formally edited texts. This work is a product of the staff of The World Bank with external contributions. Note that The World Bank does not necessarily own each component of the content included in the work. The World Bank therefore does not warrant that the use of the content contained in the work will not infringe on the rights of third parties. The risk of claims resulting from such infringement rests solely with you. The ï¬?ndings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are speciï¬?cally reserved. Rights and Permissions This work is available under the Creative Commons A ribution 3.0 Unported license (CC BY 3.0) h p://creativecommons.org/licenses/by/3.0. Under the Creative Commons A ribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: A ribution—Please cite the work as follows: Kulichenko, Natalia and Jens Wirth. 2012. Concentrating Solar Power in Developing Countries: Regulatory and Financial Incentives for Scaling Up. Washington, D.C.: World Bank. DOI: 10.1596/978-0-8213-9607-0. License: Creative Commons A ribution CC BY 3.0 Translations—If you create a translation of this work, please add the following disclaimer along with the a ribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. All queries on rights and licenses should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. ISBN (paper): 978-0-8213-9607-0 ISBN (electronic): 978-0-8213-9608-7 DOI: 10.1596/978-0-8213-9607-0 Library of Congress Cataloging-in-Publication Data Kulichenko, Natalia. Concentrating solar power in developing countries : regulatory and ï¬?nancial incentives for scaling up / Natalia Kulichenko, Jens Wirth. p. cm. Includes bibliographical references. ISBN 978-0-8213-9607-0 (alk. paper); 978-0-8213-9608-7 1. Energy development--Developing countries. 2. Solar power--Developing countries. 3. Renewable energy sources--Developing countries. 4. Energy industries--Developing countries--Finance. I. Wirth, Jens. II. World Bank. III. Title. HD9502.D442K85 2012 333.792’3091724--dc23 2012020118 Contents Foreword ..................................................................................................................................viii Acknowledgments ....................................................................................................................ix Acronyms and Abbreviations .................................................................................................. x Executive Summary ................................................................................................................ xii Regulatory Frameworks ...........................................................................................xii Cost Reduction Potential and Sustainability Assessment....................................xv Economic Analysis of Reference CST Plants ....................................................... xvii Potential for Cost Reduction through Local Manufacturing ........................... xviii Assessment of Procurement Practices ...................................................................xxi PART I INTRODUCTION AND TECHNOLOGY BRIEF .............................................. 1 Chapter 1 Context, Relevance, and Audience .................................................................... 3 Chapter 2 Overview of Concentrating Solar Thermal Technologies ............................ 4 PART II FINANCIAL AND REGULATORY SCHEMES—THE CURRENT SITUATION ........................................................................................................................ 7 Chapter 3 Policy Instruments Used to Promote CST in Developed Countries .......... 9 Regulatory Framework and Financial Incentive Options ...................................... 9 Investment Trajectories in Spain and the United States....................................... 15 Analysis and Conclusions ........................................................................................ 16 Chapter 4 Renewable Energy Schemes Supporting CST in Developing Countries .....20 MENA Incentive Schemes ........................................................................................ 20 India’s Incentive Schemes ......................................................................................... 23 South Africa’s Incentive Schemes ............................................................................ 28 PART III FINANCING CST—HOW TO BRING TECHNOLOGY COSTS DOWN .................................................................................................................. 33 Chapter 5 Cost Drivers and Cost Reduction Potential................................................... 35 LCOEs for CST in Speciï¬?c Developing Country Markets ................................... 35 Overview of the Cost Structure ............................................................................... 36 Assessment of the Cost Drivers for CST ................................................................. 37 Technical and Scale-Related Cost Reduction Potential ........................................ 43 Economic Analysis of Reference CST Plants .......................................................... 53 Financial Sustainability Assessment of Financial and Regulatory Incentives.... 45 Chapter 6 Assessment of Local Manufacturing Capabilities for CST ........................ 59 Local Manufacturing Capabilities in MENA ......................................................... 59 Local Manufacturing Capabilities in South Africa................................................ 66 iii iv Contents Chapter 7 Assessment of Procurement Practices ............................................................ 72 Tendering Models and Practices .............................................................................. 72 Bid Selection Criteria ................................................................................................. 75 PPA Structuring.......................................................................................................... 79 Appendix A Overview of Concentrating Solar Thermal Technologies ..................... 84 Parabolic Trough ............................................................................................................... 84 Technological Maturity ..................................................................................................... 88 Linear Fresnel ..................................................................................................................... 88 Power Tower ...................................................................................................................... 92 Dish-Engine ........................................................................................................................ 96 Power Blocks ...................................................................................................................... 98 Thermal Storage Options ................................................................................................. 99 Hybridization ................................................................................................................... 100 Appendix B Tables and Figures ....................................................................................... 103 References................................................................................................................................ 129 Chapter Bibliographies ......................................................................................................... 134 Tables Table ES.1: Recommended Bid Selection Criteria for CST in Developing Countries ...xxi Table 3.1: Policy Instruments, Characteristics, Advantages, and Disadvantages in Implementation .................................................................................................. 10 Table 3.2: FiTs versus RPS Schemes .......................................................................................11 Table 3.3: Currently Installed CST Capacity (MW) .............................................................12 Table 4.1: Gujarat Tariff Rates for Solar Projects ..................................................................24 Table 5.1: Estimate of Capital Expenditures—Parabolic Trough ......................................37 Table 5.2: Estimate of Capital Expenditures—Reference Power Tower...........................38 Table 5.3: Estimate of Operational Expenditures—Reference Parabolic Trough ...........39 Table 5.4: Estimate of Operational Expenditures—Reference Power Tower ..................40 Table 5.5: Overview of Cost Elements and Cost Drivers ....................................................42 Table 5.6: Local Content Sensitivities—Middle East and North Africa Case Study.......42 Table 5.7: Cost Reduction Potential of Economies of Scale / Volume Production ..........43 Table 5.8: Deï¬?nitions Used .....................................................................................................46 Table 5.9: Sensitivity Analysis India—Cost-Effectiveness of Regulatory Approaches ..... 50 Table 5.10: Sensitivity Analysis Morocco—Cost-Effectiveness of Regulatory Approaches ............................................................................................................ 51 Table 5.11: Sensitivity Analysis South Africa—Cost-Effectiveness of Regulatory Approaches ............................................................................................................ 51 Table 5.12: Economic Analysis for CST Reference Plants in India ....................................54 Table 5.13: Economic Analysis for CST Reference Plants in Morocco ..............................55 Table 5.14: Economic Analysis for CST Reference Plants in South Africa .......................56 Contents v Table 5.15: Performance and Cost Penalties .........................................................................57 Table 5.16: Impacts of Dry Versus Wet Cooling Technologies...........................................57 Table 6.1: SWOT Analysis of MENA Industries Suitable for CST ....................................61 Table 6.2: Possible Local Content by Component of CST Power Plants ..........................62 Table 6.3: Direct and Indirect Local Economic Impact in Scenarios A, B, and C ............66 Table 6.4: SWOT Analysis CST Value Chain in South Africa ............................................67 Table 6.5: Estimated Economic Impacts for Different CST Technologies ........................70 Table 6.6: Estimated Job Creation up to 2020 for Different CST Plant Technologies .....71 Table 7.1: Solicitation Types Summary .................................................................................72 Table 7.2: Procurement Methods Summary .........................................................................73 Table 7.3: Pricing Structure Summary ...................................................................................74 Table B.1: Overview of the Main Technical Characteristics of CST Technologies ........104 Table B.2: Overview of the Main Commercial Characteristics of CST Technologies ...105 Table B.3: Parabolic Trough Power Plant Projects .............................................................106 Table B.4: Demonstration Central Receiver Projects .........................................................106 Table B.5: Commercial Central Receiver Projects ..............................................................107 Table B.6: Demonstration Parabolic Dish Collector Projects ...........................................107 Table B.7: Component Speciï¬?c Cost Reduction Potential—Parabolic Trough..............108 Table B.8: Component-Speciï¬?c Cost Reduction Potential—Power Tower ....................108 Table B.9: Component-Speciï¬?c Cost Reduction Potential—Linear Fresnel...................109 Table B.10: Component-Speciï¬?c Cost Reduction Potential—Dish Engine ....................109 Table B.11: Main Financial and Regulatory Assumptions for LCOE Analysis .............110 Table B.12: Impact Assessment of Different Regulatory Incentives in India .................111 Table B.13: Impact Assessment of Different Regulatory Incentives in Morocco...........112 Table B.14: Impact Assessment of Different Regulatory Incentives in South Africa ....113 Table B.15: Economic Analysis—Main Cost Assumptions ..............................................114 Table B.16: Global CST Value Chain Analysis....................................................................115 Table B.17: Technical and Economic Barriers to Manufacturing CST Components .....116 Table B.18: Action Plan for Stimulation of Production of CST Products in MENA .....118 Table B.19: Component-Speciï¬?c Local Manufacturing Prospects in South Africa .......121 Table B.20: Capacity to Manufacture CST Components and Provide CST-Related Services in South Africa..................................................................................... 123 Table B.21: G20 and Select Nonmembers’ Producer Price Inflation (% over previous year) ..................................................................................................... 124 Table B.22: Select MENA Wholesale Price Inflation (% over previous year) ................125 Figures Figure 2.1: Markets and Applications for Solar Power .........................................................5 Figure 5.1: LCOEs for Parabolic Trough and Power Tower in India, Morocco, and South Africa ........................................................................................................... 36 vi Contents Figure 5.2: CAPEX Breakdown—Parabolic Trough (100 MW—13.4 h TES—US$914 m) .....41 Figure 5.3: CAPEX Breakdown—Power Tower (100 MW—15 h TES—US$978 m) .......41 Figure 5.4: Cost Reduction Potential for CST Technologies ..............................................44 Figure 5.5: LCOE Reduction Potential for CST ....................................................................45 Figure 5.6: Impact Assessment of Different Regulatory Approaches on LCOE in India ................................................................................................................... 47 Figure 5.7: Impact Assessment of Different Regulatory Approaches on LCOE in Morocco ............................................................................................................. 48 Figure 5.8: Impact Assessment of Different Regulatory Approaches on LCOE in South Africa ........................................................................................................... 49 Figure 5.9: Balance Sheet versus Off-Balance-Sheet Financing Effects on LCOE in India ....................................................................................................... 52 Figure 6.1: Components and Services for CST.....................................................................60 Figure 6.2: Interrelations Between MENA Home Market Size, Possible Export Volume and Focus of Support for Local Industries ........................................ 63 Figure 6.3: Potential Roadmap for EPC and Services in MENA CST Projects ................64 Figure 6.4: Potential Roadmap for the Production of Glass Mirrors in RSA ..................69 Figure 7.1: Contract Type Characteristics.............................................................................74 Figure A.1: Markets and Applications for Solar Power ......................................................85 Figure A.2: Illustration of Parabolic Trough Collectors and Sun Tracking .....................85 Figure A.3: Basic Scheme of a Parabolic Trough Power Plant ...........................................86 Figure A.4: Linear Fresnel System Diagram ........................................................................89 Figure A.5: Views of Linear Fresnel Reflector Arrays.........................................................89 Figure A.6: Example of a CFLR System Source ...................................................................90 Figure A.7: Schematic of Open Volumetric Receiver Power Tower Plant with Steam Turbine Cycle ........................................................................................... 92 Figure A.8: North Field Layout Mills ....................................................................................92 Figure A.9: Surround Field Layout Mills..............................................................................93 Figure A.10: Dish-engine Photo with Major Component Identiï¬?cation ..........................96 Figure A.11: Storage Concepts for CST ...............................................................................100 Figure A.12: Saturated Steam Hybrid Plant Conï¬?guration .............................................101 Figure A.13: Basic Scheme of an ISCCS ..............................................................................102 Figure B.1: Possible Evolutions of Local CST Industries for Key Components in MENA ............................................................................................................. 126 Figure B.2: Potential Roadmap for the Production of CST Mirrors in the MENA Region .................................................................................................... 127 Figure B.3: Potential Roadmap for the Production of Metal Structures for CST in RSA .................................................................................................................. 128 Boxes Box ES.1: Recommended PPA Elements for CST Projects in Developing Countries .... xxii Box 3.1: Germany’s Recent FiT Reform ................................................................................11 Box 3.2: The Renewable Energy Reverse Auction Mechanism ..........................................13 Contents vii Box 5.1: LCOE Structure..........................................................................................................36 Box 6.1: Estimating Employment Generation of CST Development ................................67 Box 6.2: Illustrative Industrial Development in RSA: Automotive Industry...................68 Box 7.1: Recommended Bid Selection Criteria for CST in Developing Countries ..........75 Box 7.2: Recommended PPA Elements for CST Projects in Developing Countries .......80 Foreword C oncentrating solar thermal (CST) technologies have a clear potential for scaling up renewable energy at the utility level, thereby diversifying the generation portfolio mix, powering development, and mitigating climate change. A recent surge in demand for solar thermal power generation projects in several World Bank Group (WBG) partner countries shows that CST could indeed become an important renewable energy technol- ogy that would be able to provide an alternative to conventional thermal power genera- tion based on the central utility model. The WBG is supporting the development of the technology in several partner coun- tries. In the Middle East and North Africa, the World Bank, the International Finance Corporation (IFC), and the Clean Technology Fund (CTF) are working with Algeria, the Arab Republic of Egypt, Jordan, Morocco, and Tunisia to assist them on the ï¬?nancing of the construction of a series of CST facilities. South Africa’s government has sought funding support from the CTF and technical advice from the World Bank for a 100 MW power tower CST plant in the Kalahari Desert. In addition the WBG is assisting India on a CST program that supports the Jawaharlal Nehru National Solar Mission (JNNSM). To assist our partner countries be er, there is a need to analyze the experience of developed and developing countries in designing and implementing regulatory frame- works supporting the deployment of this technology and to draw relevant lessons for emerging markets. We expect that this report will provide insights for policy makers, stakeholders, private ï¬?nanciers, and donors in meeting the challenges of scaling up the deployment of renewable energy—and CST in particular. Lucio Monari Manager, Energy Anchor Unit (SEGEN) Sustainable Energy Department viii Acknowledgments T he broad scope of this report was drawn extensively from more than 300 documents related to past and ongoing projects and from analytical experience in the ï¬?eld of concentrating thermal solar (CST) power. Natalia Kulichenko (Task Team Leader) and Jens Wirth of the Sustainable Energy Department led the preparation of this report under the guidance of Lucio Monari, Sector Manager, Energy, Sustainable Energy Department. Eleanor Ereira, Brian Klein, and Victor Loksha provided valuable contributions to the chapter addressing CST regulatory frameworks in India, South Africa, and countries of the Middle East and Northern Africa region, as did Silvia Martinez Romero for the overview of CST technologies chapter. This report also beneï¬?ted from advice, suggestions, and corrections about the numerous technical, ï¬?nancial, economic, and regulatory issues involved in the develop- ment and deployment of concentrating solar thermal (CST) power. The authors would like to express their gratitude to the following colleagues inside and outside the World Bank Group (WBG): Suman Babbar, Roger Coma Cunill, Gabriela Elizondo Azuela, Chan- drasekar Govindarajalu, Rohit Khanna, Tobias Maerz, Silvia Pariente-David, Michael Toman, Philippe Roos, Gevorg Sargsyan, and Chandrasekeren Subramaniam at the World Bank; Dana Younger at the International Finance Corporation (IFC); David Kearney, IFC consultant; Charles Kutscher, Michael Mendelsohn, and Paul Gilman at the National Renewable Energy Laboratory (NREL), U.S. Department of Energy; and Luiz Crespo at Protermosolar, Spain. Several chapters are partly based on the work of external con- sultants, including Ynï¬?niti/Nexus/CENER (Chapters 2 and 5); Fichtner (Chapters 2 and 6); Anil Markandya at Metroeconomica (economic analysis in Chapter 5); and Ernst & Young and Fraunhofer (Chapter 6); and NOVI Energy (Chapter 7). The authors bear sole responsibility for any errors and omissions. The co-ï¬?nancing by the Africa Renewable Energy Access Program (AFREA) is grate- fully acknowledged. AFREA—a Trust Fund Grant Program funded by the Kingdom of the Netherlands through the Clean Energy Investment Framework (CEIF) Multi Donor Trust Fund (MDTF) recipient-executed and technical assistance window established by the Energy Sector Management Assistance Program (ESMAP). These funds are ear- marked to support analytical and advisory activities executed by the Africa Energy Unit (AFTEG) and also to provide recipient-executed technical assistance and pre-investment grants that would help accelerate deployment of renewable energy systems. The ï¬?nancial support of ESMAP is also gratefully acknowledged. ESMAP is a global knowledge and technical assistance trust fund program administered by the World Bank that helps low- and middle-income countries increase the know-how and institutional capacity to achieve environmentally sustainable energy solutions for poverty reduc- tion and economic growth. ESMAP is governed and funded by a Consultative Group comprised of official bilateral donors and multilateral institutions, representing Austria, Australia, Denmark, Finland, Germany, Iceland, the Netherlands, Norway, Sweden, the United Kingdom, and the WBG. ix Acronyms and Abbreviations AET average electricity tariff AfDB African Development Bank BUB back-up boiler CAPEX capital expenditure CCGT combined cycle gas turbine CDM Clean Development Mechanism CERC Central Electricity Regulatory Commission CHP combined heat and power CIF climate investment funds CLFR Compact Linear Fresnel Reflector CoC cost of capital CPV concentrating photovoltaic (subset of CSP) CREB Clean Renewable Energy Bond CSP concentrating solar power (includes CST and CPV) CST concentrating solar thermal (subset of CSP) CTF Clean Technology Fund DISCO distribution company DNI direct normal irradiation DSCR debt service coverage ratio DSG direct steam generation EBIT earnings before interest and taxes ENPV economic net present value EPC engineering, procurement, and construction ERR economic rate of return ESTELA European Solar Thermal Electricity Association EIB European Investment Bank FiT feed-in tariff GDP gross domestic product GEF Global Environment Facility GW gigawa GWh gigawa -hour HTF heat transfer fluid IBRD International Bank for Reconstruction and Development IEA International Energy Agency IPP independent power producer ISCC integrated solar combined cycle ISCCS integrated solar combined cycle system JNNSM Jawaharlal Nehru National Solar Mission KfW Kreditanstalt für Wiederau au kW kilowa kWh kilowa -hour LCOE levelized cost of electricity MASEN Moroccan Agency for Solar Energy MAT minimum alternative tax MDB Multilateral Development Bank MENA Middle East and North Africa MW megawa x Acronyms and Abbreviations xi MWh megawa -hour NERSA National Energy Regulator of South Africa NREL National Renewable Energy Laboratory NTPC National Thermal Power Corporation Ltd. O&M operation and maintenance OBA output-based approach OPEX operational expenditure PCU power conversion unit PPA power purchase agreement PPP public-private partnership PSA Plataforma Solar de Almería R&D research and development RAM reverse auction mechanism REC Renewable Energy Certiï¬?cate REFIT renewable energy feed-in tariff RFI Request for Information RFP Request for Proposal RPO Renewable Purchase Obligation RPS Renewable Purchase Standard RSA Republic of South Africa SEGS solar energy generating system SWOT strengths, weaknesses, opportunities, and threats TES thermal electric storage TSP Tunisian Solar Plan USDA United States Department of Agriculture WACC weighted average cost of capital WTP willingness-to-pay ZAR South African rand Executive Summary C oncentrating solar thermal power (CST) has a tremendous potential for scaling up renewable energy at the utility level, diversifying the generation portfolio mix, pow- ering development, and mitigating climate change. A recent surge in demand for solar thermal power generation projects using different CST technologies in various countries shows that CST could become an important renewable energy technology that would provide an alternative to conventional thermal power generation based on the central utility model. At present, different CST technologies have reached varying degrees of commercial availability. This emerging nature of CST means that there are market and technical impediments to accelerating its acceptance, including cost competitiveness, an under- standing of technology capability and limitations, intermi ency, and beneï¬?ts of electric- ity storage. Many developed and some developing countries are currently working to address these barriers in order to scale up CST-based power generation. Given the considerable growth of CST development in several World Bank Group (WBG) partner countries, there is a need to assess the recent experience of developed countries in designing and implementing regulatory frameworks and draw lesson that could facilitate the deployment of CST technologies in developing countries. Merely rep- licating developed countries’ schemes in the context of a developing country may not generate the desired outcomes. Against this background, this report (a) analyzes and draws lessons from the efforts of some developed countries and adapts them to the characteristics of developing econo- mies; (b) assesses the cost reduction potential and economic and ï¬?nancial affordability of various technologies in emerging markets; (c) evaluates the potential for cost reduction and associated economic beneï¬?ts derived from local manufacturing; and (d) suggests ways to tailor bidding models and practices, bid selection criteria, and structures for power purchase agreements (PPAs) for CST projects in developing market conditions. Regulatory Frameworks Based on an assessment of the experiences of regulatory frameworks that are in place in developed markets and an assessment of regulatory incentives proposed and employed in developing markets to incentivize the development of CSP, the following general con- clusions can be drawn: 1. In nearly all cases analyzed in this report, including in India, Morocco, and South Africa, the levelized cost of electricity (LCOE) for parabolic trough and power tower projects is still too high in relation to the tariffs available for CST-generated electricity to allow for full cost recovery and to meet ï¬?nancing constraints. 2. Further modiï¬?cations of regulatory frameworks that are currently in place in emerging markets should be considered to at least partly mitigate these con- straints and thereby ensure large-scale CST deployment and the creation of local manufacturing and service capacities. xii Executive Summary xiii 3. A feed-in tariff (FiT) seems to be the most appropriate instrument if large-scale CST deployment and the maximization of local inputs are the main drivers behind the establishment of the incentive framework and if cost considerations are not pivotal. This is because of the demonstrated ability of FiTs to trigger large-scale investments in a relatively short timeframe. If properly designed, FiTs are the most straightforward way to provide investors with the security necessary to overcome otherwise prohibitive development risks and ensure adequate ï¬?nancial returns. 4. Any FiT scheme could beneï¬?t from several recent lessons learned regarding its design to reduce high societal costs. A FiT scheme should entail at the minimum (a) an annual and overall capacity cap based on a realistic and affordable policy goal, and (b) predetermined tariff revisions for new capacities and ultimately a phase-out schedule to keep tariffs in line with decreasing capital and invest- ment costs. While preserving the main beneï¬?ts of a FiT for developers—its sim- plicity and predictability—these measures can help keep societal costs under control and minimize them. 5. An alternative scheme involves a combination of a FiT with a reverse auctioning mechanism. Such mechanisms could have the following minimal features: (a) an annual and overall capacity cap based on a realistic and affordable policy goal, (b) the possibility for developers to bid on the eligible capacity within a given timeframe and offer the delivery of the electricity at a ï¬?xed tariff level below the original FiT, and (c) a mechanism assuring the technical and ï¬?nancial feasibility of the submi ed bids. While offering similar beneï¬?ts as a FiT for developers, this approach could lower societal costs. 6. A Renewable Portfolio Standard (RPS) scheme that combines a variety of other regulatory and ï¬?nancial incentives could also be a viable option. An RPS scheme could be successful in triggering investments in CST if it is combined with (a) sovereign guarantees for PPAs signed with utilities or a single buyer to ensure bankable sources of revenue; and (b) signiï¬?cant amounts of conces- sional ï¬?nancing, which tend to be the most cost-efficient way of incentivizing CST investments. 7. The recent experience on RPS schemes and/or FiT frameworks shows that both developers and commercial banks assign a higher overall risk proï¬?le to proj- ects with cash flows based on a typical PPA arrangement under an RPS scheme instead of a FiT. This might be different if PPAs reflect competitive tariffs and are signed with single buyers or utilities under explicit or implicit sovereign backing. RPS schemes currently seem to be preferable to FiTs only if (a) societal cost considerations are the prevailing issue for policy makers; (b) there are no ï¬?xed targets for CST capacity to be installed; and (c) building local capacity for component manufacturing and service delivery is somewhat less of a priority. 8. Incentive frameworks should be tailored to the speciï¬?c circumstances to allow developers to use the respective CST capacity in the most efficient way possible. This could includes avoiding capacity limits on individual plants, because of the considerable economies of scale for individual plants that can be achieved, and limits on the use of storage. The la er is particularly important, since an optimal amount of storage decreases the LCOEs of individual plants and therefore the cost of CST-generated electricity on a per-kilowa -hour basis. xiv Executive Summary In addition to these general conclusions, the report provides a review and detailed analyses and recommendations on the incentive schemes for CST currently in place in some of the major emerging markets as described below. Middle East and North Africa Region In the context of MENA, the current support schemes are centered on either public sector projects or public-private partnership (PPP) models. Experience to date shows that (a) the region is not quite ready to embrace FiTs or RPSs, although efforts to champion the introduction of such schemes are ongoing; (b) independent power producer and power purchase agreement (IPP/PPA) schemes have not worked well in the past, as illustrated in projects supported by the Global Environment Facility (GEF), which had to be restruc- tured into public projects; and (c) a new PPP scheme is being tried out for an individual, large-scale projects (Morocco), and it seems to have a be er chance of success than the earlier a empts to engage the private sector through a pure IPP concept. The approach currently taken to scale up CST deployment in MENA with the support of the Clean Technology Fund (CTF) assumes that guaranteed source of sub- sidies will help address, to a certain degree, issues related to both high capital costs and uncertainties regarding the policy and regulatory frameworks. The expectation is that, with more clarity in the policy framework for CST development in the MENA countries in the midterm, the need for subsidies will be reduced. Over the longer term, and in order to achieve transformational effects and replicability goals, these invest- ments need to be accompanied by appropriate national policies, such as FiTs and/or RPS quotas combined with other regulatory and ï¬?nancial incentives in the respective jurisdictions. India The Government of India has made a strategic choice to promote grid-connected solar power and put in place the needed incentive packages. The Government of India’s policy on CST is designed to be largely private sector–driven, with the government creating an enabling environment for investors. Despite criticisms on the FiT guidelines, private developers are active participants in the early bidding stages to strategically position themselves in India’s emerging CST market. This could explain the oversubscription of the ï¬?rst bidding round for CST projects under Phase 1 of the JNNSM. Over the long term, the regulatory framework could beneï¬?t from improving the consistency among instruments (the current process mixes RPS and FiT elements), and the coordination between state-level and central government–level incentives. Given the great degree of uncertainty about the required (or justiï¬?ed) level of capital costs for CST projects in developing countries in general, and in India in particular, an approach involving competitive procurement of speciï¬?ed amounts of CST capacity may be a good choice. A combined RPS/FiT scheme with a built-in reverse auction mecha- nism may not be as aggressive a strategy as a pure FiT in securing a massive expansion of solar power capacity. However, it facilitates the price discovery process be er than a pure FiT system. This may result in substantial cost savings both for the public sector and for the rate payer. By contrast, doubts remain as to whether the tariffs offered by winning bidders are not undervalued. The overall effectiveness of the incentives frame- work for solar power development is still to be demonstrated by ï¬?nancial closures for the concluded PPAs. Executive Summary xv South Africa The proposed framework of the renewable energy feed-in tariff (REFIT) is not yet opera- tional in South Africa. One can only speculate as to how successful it will be in encourag- ing investments in both CST and other renewable energy technologies. There are concerns over the lack of a deï¬?ned structure of the REFIT, uncertainty over what the ï¬?nal tariffs will be, and how they could a ract or deter potential IPPs. However, many of these concerns could be addressed once the National Energy Regulator of South Africa (NERSA) and the national utility (Eskom), as a single buyer, ï¬?nalize the process for arranging the PPAs. This will happen once tariff levels are decided and the role of the single buyer (Eskom or an independent party) is be er deï¬?ned. It is conceivable that the REFIT may encourage more investment for certain technolo- gies than for others. In the same way that an RPS scheme induces investments predomi- nantly in the cheapest technology, the REFIT may only promote signiï¬?cant investments in more established and less risky technologies, such as wind power, rather than CST. The fact that the vast majority of applications received by Eskom so far have been for wind projects indicates the disparity of the effectiveness of the policy across different technologies. The combination of a CTF-funded, large-scale CST project, a planned solar park project, and the introduction of a FiT system may well succeed in mobilizing private sector investments in CST technology in South Africa. However, the process is still ongo- ing and various steps need to be completed before electricity generated from renewable technologies will be sold at the prescribed tariff. Cost Reduction Potential and Sustainability Assessment Different CST technologies have, at present, reached varying degrees of commercial availability. While parabolic trough and, to a slightly lesser degree, power tower are basi- cally close to full commercial state, clear commercial cost data have yet to be established for the Linear Fresnel and Dish Stirling technologies. A detailed LCOE analysis based on the existing incentive schemes and various assumptions regarding country speciï¬?c natural and economic characteristics was conducted for some of the major emerging markets for CST—India, Morocco, and South Africa—comparing parabolic trough and power tower technologies (as the most mature technologies). The report also presents a review of typical cost structures for parabolic trough and power tower plants, which was derived from projects developed or under preparation in Spain and the United States speciï¬?cally for this report, and an in-depth assessment of the respective cost drivers. Based on these analyses, the report provides (a) technology-speciï¬?c LCOE reduction potentials and (b) an assessment of effects on public sector resources from different regulatory and ï¬?nancial incentives used to lower the LCOEs in various emerging market conditions. Component-, Technical-, and Scale-Related Cost Reduction Potential Detailed analyses of potential for component-speciï¬?c cost reductions are given in the report. This was based on a detailed assessment of the respective cost drivers for each component and the underlying development in the respective industries producing these components. Among parabolic trough components, the most potential for cost reduc- tion in the timeframe until 2020 is demonstrated for reflectors (18–22 percent), reflector xvi Executive Summary mounting structures (25–30 percent), receivers (15–20 percent), heat transfer systems (15–25 percent), and molten salt systems (20 percent). Power tower system components showing the most cost reduction potential are reflector mounting structures (17–20 percent), heat transfer systems (15–25 percent), and molten salts as heat transfer fluids (20 percent). Components for Linear Fresnel systems showing the most cost reduction potential include reflector mounting structures (25–35 percent) and receivers (15–25 per- cent), while for the Stirling Dish engine system, it is the reflectors (35–40 percent) and reflector mounting structures (25–28 percent). The overall cost reduction potential for each CST technology was derived by model- ing reference plants based on the assumed component speciï¬?c cost reduction potentials. For these reference plants, the individual cost reduction potentials of components were deducted from the component speciï¬?c cost data available from developed markets for CST. The la er were chosen, since they were seen to be more established than the com- ponent speciï¬?c cost data available from emerging markets for CST. Sustainability Analysis of Financial and Regulatory Incentives A basic sustainability analysis was conducted for a variety of regulatory and ï¬?nancial incentives granted in three of the major emerging markets for CST—India, Morocco, and South Africa—based on the incentives’ impact on the LCOEs of 100 MW reference plants in these markets. The primary aim was to estimate the impacts of speciï¬?c regulatory and ï¬?nancial incentives on CST generation cost and the societal cost expressed in ï¬?nancial terms. The analysis was carried out to Determine the ï¬?nancial cost-effectiveness of different regulatory incentives and approaches in terms of their impact on LCOEs and hence their ability to facilitate investments per dollar spent. The tested incentives ranged from tax holidays to favorable depreciation schemes and the use of concessional ï¬?nancing schemes, such as through the International Bank for Reconstruction and Development (IBRD), CTF, and GEF. The following observations can be derived: 1. The accuracy of solar resource assessment in measuring site-speciï¬?c levels of direct normal irradiation (DNI) is essential as the robustness of the ï¬?nancial analysis for a CST plant is heavily dependent on the quality of the DNI data. Given the inverse relationship between the DNI and LCOE for CST plants, data measured on the ground at the actual site of the project over the course of at least a full year are required to provide sufficient grounding for a solid ï¬?nancial model. 2. For all technologies in all three scenarios considered, the LCOEs for stand- alone projects are most likely currently too high to allow for cost recovery and meeting ï¬?nancing constraints. This is especially the case when the LCOEs are compared to the FiTs available for CST-generated electricity in Phase 1 of the JNNSM in India and the FiTs that have been proposed for Phase 2 of the REFIT scheme in South Africa. 3. LCOE calculations based on balance sheet ï¬?nancing might be considerably lower than estimates based on nonrecourse (off–balance sheet) ï¬?nancing assumptions, such as the ones made for this analysis. However, balance sheet ï¬?nancing increases the risk proï¬?le of a company’s investments and might Executive Summary xvii require cross-subsidization among projects within the company’s portfolio, since the ï¬?nancial viability of a stand-alone project is no longer guaranteed. 4. Financial and regulatory incentives, as well as concessional ï¬?nancing schemes, can signiï¬?cantly lower LCOEs. Within the range of considered ï¬?nancial and reg- ulatory incentives, simple tax reductions and exemptions tend to have the lowest impact and are likely to be the least cost-effective incentives in ï¬?nancial terms (not considering economic opportunity cost). By contrast, concessional ï¬?nanc- ing schemes tend to have the highest impact and are likely to be the most cost- effective incentives in terms of their impact on LCOE on a per-dollar-spent basis. With regard to the other incentives considered, accelerated depreciation, especial- ly when compared to simple tax reductions or exemptions, seems to be the superior option. Although far from cheap, it might be worth considering in cases where—as seen in the case of South Africa—the existing regulatory incentive framework just needs to be moderately adjusted to lower LCOEs to the threshold where stand-alone projects become ï¬?nancially viable. Economic Analysis of Reference CST Plants The report provides an economic analysis based on current investment costs for refer- ence 100 MW CST plants—both parabolic trough and power tower—in the three respec- tive countries considered in the report—India, Morocco, and South Africa. Sensitivity analyses are provided for higher investment costs, project delays, lower load factors, and a higher value of the power generated. The following important observations can be made across all three countries: 1. In none of the countries does the economic rate of return (ERR) achieve a rate required for infrastructure projects of more than 10 percent. Excluding carbon and other environmental beneï¬?ts, the ERR ranges from −0.65 percent to 4.8 per- cent for the power tower and from −2.55 percent to 3.8 percent for the parabolic trough. Including the economic beneï¬?t of reducing carbon emissions, the ERR ranges from 2.1 percent to 8.8 percent for the power tower and from 1.1 percent to 7.4 percent for the parabolic trough reference plants. 2. The carbon values that are needed to make projects achieve an ERR are implausi- bly large in India and Morocco. In South Africa they are also quite high, but one could argue that carbon reduction projects with costs in that range (US$80–100/ ton CO2) have been undertaken in other sectors. The sensitivity analysis shows approximately a 1 percent reduction in the ERR for a 10 per- cent higher project cost and a further 1 percent reduction for an additional 10 percent higher project cost. A reduction in the load factor by 20 percent has a bigger impact—reducing the ERR by 2.5 percent to 3 percent. In the case of India, the results show that parabolic trough has a higher return than power tower, and that a ï¬?ve-year delay increases the ERR by nearly 3 percent. In the case of Morocco, the delay is not as effective in increasing the ERR (possible because the increases in power value are more modest). Even with carbon and local pollutant beneï¬?ts, the ERR is well below a test rate. In Morocco, power tower appears to exhibit slightly be er economics than parabolic trough. For the South African case, because of the higher value of power and carbon beneï¬?ts, a 12 percent ERR can be exceeded with xviii Executive Summary both technologies, although the power tower has a higher return by 1–2 percent. Includ- ing the beneï¬?ts of reduced local pollutants would increase the ERR further—potentially by up to 1 percent. The analysis indicates that while power tower technology has a slightly higher return than parabolic trough, and the use of wet cooling can slightly improve the ERR, CST proj- ects at current investment costs have low ERRs that would be unable to meet commer- cial infrastructure investment requirements. However, investment costs are projected to decrease considerably over the coming years—a development that is expected to largely alter the economics of CST technologies. Therefore, the decision to uptake CST technol- ogy might not necessarily be based on economic considerations alone, but might include other aspirations, such as gaining market leadership and experience through technology development or targeting the building-up of a local manufacturing industry. Potential ways also exist for improving the economics of CST, even under current investment cost assumptions through, for example, hybridization and the large-scale application of storage—areas that, however, are outside the scope of this report. Potential for Cost Reduction through Local Manufacturing To realize the cost reduction trajectories projected in this report, a major scale-up of CST developments would be necessary, both in the already-established markets, as well as in emerging markets in the MENA region, India, and South Africa. A major increase in CST capacity in emerging markets, however, is likely only when the countries concerned beneï¬?t from the technology for their economic development in general. One of the pri- mary means to foster development could be the establishment of local manufacturing and assembly capacities. Local manufacturing might have the added beneï¬?t of reduc- ing the cost of local projects in the near term and bringing down the cost for a variety of components and CST-related services in the mid- to long term. By looking at local manufacturing capabilities in several emerging markets for CST, including the MENA region and South Africa, several general conclusions on incentivizing and supporting the buildup of local capacities to manufacture components and provide CST-related ser- vices can be made: 1. The implementation of a stable and sustainable regulatory framework is the key precondition for the development of a market for CST projects that is needed to create investment conditions for local manufacturing and service capacities in emerging markets. 2. In the medium to long term, the annually installed capacity should be on the highest scale possible in order to incentivize the development of production lines, particularly in the case of mirrors and receivers. 3. Regulatory incentive frameworks must be in line with general national strate- gies for industrial development, and national energy policies should be well coordinated and involve clear targets for the market diffusion of CST, sub- stantial research and development (R&D) efforts, strategy funds for industrial development of CST industry sectors, and—in most cases—a stronger regional integration of policies. 4. The provision of low-interest loans and grants speciï¬?cally designed for local manufacturing of renewable energy components might help local companies Executive Summary xix raise funds for R&D to support product innovation or provide risk capital for new start-up companies. 5. The buildup of local industries could further be facilitated by introducing local content clauses within CST bids and other support instruments. Local content requirements, however, need to be set at realistic levels while being allowed to increase over time, according to the speed at which local industries can be developed. 6. Business models should build on the comparative advantages of particular sectors in the respective country and should involve international cooperation agreements, for example, in the form of joint ventures and licensing. In the case of receivers, for example, subsidiaries of foreign companies will most likely be relevant business models in the beginning. Furthermore, obvious areas for local manufacturing capacity development include investments in new, highly auto- mated production lines for the mounting structure and glass production, as well as the adaptation of techniques for coating and bending mirrors. With regard to CST-related services, the local assembly of plants and involvement of local EPC contractors are important initial steps to maximize the local value contribution. 7. Establishing local manufacturing capacity will have to involve comprehensive education and training programs for the industrial workforce in relevant sec- tors. Universities and technical schools should be encouraged to teach CST technology–based courses to educate the potential workforce, particularly engi- neers and other technical graduates. 8. Ultimately, to ensure regional and international quality requirements and to strengthen the competitiveness of future local CST industries, implementing quality assurance standards for CST components should be considered. Speciï¬?c assessments of the local capabilities were conducted for two of the major emerg- ing markets for CST—the MENA region and South Africa. Based on an in-depth assess- ment of the local CST value chain, the report provides component-speciï¬?c projections for local manufacturing, draws roadmaps and action plans in order to maximize local content generation in the industry, and estimates the immediate economic beneï¬?ts of local manufacturing, especially with regard to employment generation. For the MENA region, an important ï¬?nding concerning the status quo and future perspectives of local manufacturing is that, while several parts of the piping system in the solar ï¬?eld—for the interconnection of collectors and power block—can already be produced locally by regional suppliers, a further scale-up of local manufacturing capabilities in certain sectors—especially mirrors—has signiï¬?cant potential. For this potential to be reached, however, the countries would have to aggressively build on the know-how gained from the successful construction of the integrated solar combined cycle (ISCC) projects, while at the same time encouraging the involvement of interna- tional companies to build up local production facilities. A certain specialization in each country would be beneï¬?cial because local demand will probably be relatively low in the short to medium term. In South Africa the currently possible proportion of local manufacturing for CST power plant projects is expected to be up to 60 percent, depending on whether speciï¬?c CST components, such as receiver tubes, heat transfer fluid (HTF) pumps, and swivel xx Executive Summary joints, can be developed and manufactured locally. Depending on the uptake of the CST industry, however, this share can be considerably lower for construction and compo- nents or can increase further. Local mirror and receiver production are seen as starting as early as 2015 in the accelerated scenario, which also projects the local production of other specialized, high-precision steel accessories for CST applications. Beyond 2020, the share of local manufacturing would increase even more as a result of further technology transfer and knowledge sharing through the realization of more CST plants in South Africa, since the learning effect is expected to play out fully around this time. This would also lead to a drop in the cost of locally manufactured CST components because of tech- nological advancements, economies of scale, and competition in the CST component manufacturing sector. Roadmaps for the Development of Local Manufacturing of CST Components The report identiï¬?es potential routes for the development of local manufacturing capacities for different components for both MENA countries and South Africa, and sets out the main milestones required for the establishment of both local and export markets. The approach is to deï¬?ne a set of actions to be implemented among stakehold- ers who may bring about an activation of CST component manufacturing in the respec- tive jurisdictions. Potential Economic Beneï¬?ts of Developing a CST Industry in MENA and South Africa The economic and employment beneï¬?ts of developing a CST industry estimated in the report are gross estimates and therefore do not consider the potential cost of scaling down or not strengthening other industries providing other technologies that could supply the same amount of energy. In general, the economic beneï¬?ts are strongly related to the market size of CST. For the MENA region, an accelerated scenario—assuming 5 GW of installed capacity by 2025—would create a local economic impact of US$14.3 billion, roughly half of which would be from indirect impacts of the CST value chain (excluding component exports), compared to only US$2.2 billion in a business-as-usual scenario, assuming no replication effects from the uptake of 1 GW of capacity as envisaged by the CTF Investment Plan for region. The impact on labor generation would be a per- manent workforce of 4,500–6,000 local employees regionally by 2020 under a business- as-usual scenario based on the CTF Investment Plan. In contrast, in the accelerated scenario in 2025, the number of permanent local jobs could rise to between 65,000 and 79,000 (46,000–60,000 jobs in the construction and manufacturing sector plus 19,000 jobs in operation and maintenance). Additional impacts on job creation and growth of gross domestic product (GDP) could come from export opportunities for CST components. Exporting the same components that are manufactured for local markets to the Euro- pean Union, United States, or MENA (2 GW by 2020, 5 GW by 2025) could lead to addi- tional revenues of more than US$3 billion by 2020 and up to US$10 billion by 2025 for local CST industries. For South Africa the accelerated scenario creates a local economic impact of US$25.9 billion compared with US$4.1 billion in the same business-as-usual scenario as described for the MENA region. In terms of employment generation, the impact would be 66,800–83,100 permanent jobs for local employees by 2020 under the accelerated sce- nario and 11,000–14,800 permanent jobs under the business-as-usual scenario based on Executive Summary xxi the CTF Investment Plan. Exporting components could lead to additional revenues of more than US$3.6 billion by 2030. Assessment of Procurement Practices The report concludes by describing and analyzing various bidding models, practices, and the bid selection criteria typically used for CST projects based on information avail- able from the developers and utilities in developed markets. The report then provides recommendations on tailoring these practices, criteria, and PPA structuring for develop- ing country markets to help facilitate business transactions for CST projects. Recommen- dations are provided for primary elements of each subtopic. Bidding Criteria The report provides guidance on the best-practice structuring of bidding criteria—from both a regulator’s point of view under, for example, a FiT scheme, and a utility’s or single buyer’s point of view under an RPS scheme. In addition, it provides recommendations on how to design PPAs under an RPS scheme. With regard to bidding selection criteria, the report suggests a weighted bid matrix for CST projects, as shown in table ES.1. The weighted bid matrix provides a set of recommended bid selection criteria. The weights associated with each criterion should be assessed by individual respective entities respon- sible for bid criteria design based on the relative importance placed on each factor. Table ES.1: Recommended bid selection criteria for CST in developing countries Cost-Based • Level of concessional ï¬?nancing necessary Feasibility-Based • Company and team experience • Company ï¬?nancial stability • Technology maturity • Interconnection feasibility • Site control • Environmental approvals • Ability to raise ï¬?nancing • Levelized cost of electricity (LCOE) Policy-Based • Speed of implementation (schedule) Value-Based (Optional) Source: NOVI Energy 2011. Elements of Power Purchase Agreements Ultimately, the report provides recommendations on components that should be included in an optimally balanced PPA for CST projects to adequately reflect the interests of both the developer and the utility (or a single buyer). When selecting the recommended PPA elements, considerations should include characteristics of solar technologies, as well as aspects that may be applicable to projects in emerging markets for CST, such as per- xxii Executive Summary Box ES.1: Recommended PPA elements for CST projects in developing countries • Fixed dispatch with sharing of curtailment risk • Energy payment using PPI/CPI/exchange rate/LIBOR • Time of delivery factors for energy payments • Renewable energy credits bundled with energy (if applicable) • Seller development security (refunded at commercial operations) • Seller performance security (throughout the term of the PPA) • Buyer payment security (throughout the term of the PPA) • Opportunities to rectify default before contract termination • Seller re-pricing and exit on incentive cancellation • “Politicalâ€? force majeure provisions Source: NOVI Energy 2011. ceived risks over the reliability of transmission and distribution systems, off taker credit strength, and the sustainability of a respective government policy, particular in regard the executed contracts and promised government incentives. The recommended elements were selected to help reduce the risk perception and thus to improve the a ractiveness of PPAs for investors and ï¬?nanciers, while meeting the needs of buyers (see box ES.1). PART I Introduction and Technology Brief CHAPTER 1 Context, Relevance, and Audience C oncentrating solar power (CSP) refers to several different technologies that use mirrors to focus, or concentrate, the sun’s rays to generate electricity. The two sub- categories of CSP are (a) concentrating photovoltaic (CPV), which focuses the sun’s rays onto photovoltaic panels to generate electricity directly and (b) different Concentrating solar thermal (CST) technologies, all of which—with the exception of Dish Stirling— work on the same principle of focusing solar radiation to generate heat, which is then used to drive an engine or turbine to generate electricity. CST technologies have tremendous potential for scaling up renewable energy at the utility level, diversifying the generation portfolio mix, powering development, and mitigat- ing climate change. A recent surge in demand for solar thermal power generation projects using different CST technologies in Spain, the United States, and a handful of other countries shows that CST could become a key renewable energy technology that is able to provide an alternative to conventional thermal power generation based on the central utility model. With respect to World Bank Group (WBG) partner countries, several countries in the Middle East and North Africa (MENA)—Algeria, the Arab Republic of Egypt, Jordan, Morocco, and Tunisia—are pursuing regional CST investment projects to be ï¬?nanced by the World Bank, IFC, and Clean Technology Fund (CTF). The plan for these installa- tions is to supply power across the region and potentially to Europe. The South African government has sought funding support from the CTF and technical advice from the World Bank for a 100 MW power tower CST plant in the Kalahari Desert. The WBG is also providing technical assistance to the Government of India on certain aspects of the implementation of the Jawaharlal Nehru National Solar Mission (JNNSM). At present, the different CST technologies have reached varying degrees of commercial maturity. This emerging nature of CST means that there are market impediments that need to be overcome to accelerate its acceptance, including cost competitiveness, awareness of technology capabilities and limitations, intermi ency, and the need for electricity storage. Given the considerable pace of CST development in several WBG partner countries, there is a need to review the recent experience in developed countries in designing and implementing regulatory frameworks to draw relevant lessons for emerging markets. Adaption of these lessons to speciï¬?c developing country circumstances will be neces- sary, since the mere replication of developed countries’ schemes may not generate the desired outcomes. After providing a brief overview of the current state of CST technologies (Chapter 2), the report evaluates recent experiences with regard to regulatory frameworks in some of the developed countries, as well as those developing countries that have started establish- ing regulatory frameworks targeted at CST deployment (Chapters 3 and 4); assesses the cost reduction potential and economic and ï¬?nancial affordability of various technologies in emerging markets (Chapter 5); evaluates the potential for cost reduction resulting from local manufacturing and associated economic beneï¬?ts (Chapter 6); and ultimately sug- gests ways of tailoring bidding models and practices, bid selection criteria, and power purchase agreement (PPA) structuring to speciï¬?cs of CST projects (Chapter 7). 3 CHAPTER 2 Overview of Concentrating Solar Thermal Technologies A pplications of solar thermal technologies are best suited for regions that experience high levels of direct normal irradiation (DNI). These regions are typically located in dry areas such as deserts, which also have the advantage of plentiful land unsuitable for agricultural or industrial purposes. According to a recent report,1 among the various solar technologies, the CST is pri- marily suited for larger scale installations, while PV-based technologies are be er matched for smaller-scale or distributed generation applications (ï¬?gure 2.1). Photovoltaic panel the- oretically has wider geographical applications, even if a certain level of diffuse radiation is needed in order to make the electricity generation economically viable. Solar thermal technologies have geographical limitations, and can potentially be economically viable only in regions that possess high DNI to ensure high energy yields. The main advantages of CST applications include less intermi ency because of the system thermal inertia, and the option to integrate thermal storage, thus making power generation possible during extended hours (when the sun doesn’t shine) and to use CST in utility scale operations. The following factors are typically cited as drawbacks of the current application of CST technologies: â–  CST-based plants are presently characterized by high upfront investment resulting in increased electricity generation costs, which could be decreased by further technological innovations and economies of scale, including volume production and larger-sized units. â–  Locations with irradiations of more than 2,000 kWh/m2/year are suitable to make solar thermal performance economically justiï¬?able (Viebahn and others 2008). The primary CST technologies include â–  Parabolic trough â–  Power tower (central receiver) â–  Linear Fresnel â–  Parabolic Dish (Dish Stirling) The Parabolic Dish technology differs signiï¬?cantly from the other three in both technical and economic terms. The parabolic trough, power tower, and Linear Fresnel technolo- gies, although based on the same technical principals, vary with regard to their reliability, maturity and operational experience in utility scale conditions. Relevant design features of each technology are discussed in more detail in Appendix A, along with a summary of the maturity status of each technology. Every technology has advantages and disadvan- 4 Concentrating Solar Power in Developing Countries 5 Figure 2.1: Markets and applications for solar power Category Small Medium Large Installation size < 10kW 10 to 100kW 100kW to 1 MW 1 to 10MW 10 to 100MW > 100 MW Technology mix in each market 100% PV 99% PV, 1% CSP 20% PV, 80% CSP 2007 Share of worldwide solar market 7GW (84%) 0.7 GW (9%) 0.5 GW (7%) (installed capacity and % of installed capacity) Distributed generation Installation type Central generation Residential Commercial Markets served Utility Base (50%), intermediate (40%), peak (10%) Non-tracking PV PV Non tracking PV based dispatchable CPV Dish-engine Thermal Dispatchable trough based (with storage) tower LFR Installation size <10kW 10 to 100 kW 100kW to 1MW 1 to 10 MW 10 to 100 MW > 100 MW Legend: best suited suitable Source: Adapted from Grama, Wayman, and Bradford 2008. tages, and the suitability of each one should be assessed carefully depending on the needs and requirements of every site and project. The summary results of the technical and commercial assessments of the technologies, as per literature and operational experience reviews, are summarized in tables B.1 and B.2 in Appendix B. Regarding operational experience and technological maturity, parabolic trough and, to a lesser extent, power tower are closest to commercial maturity state. Fresnel and Dish Stirling technologies are still at earlier development levels. Therefore, the techno- logical risk is considered to be the lowest for parabolic through and again to a slightly lesser degree for power tower plants. Investment and operating and management costs (O&M) costs are also be er known for these two technologies thus reducing the related ï¬?nancing risks. Tables B.3–B.6 in Appendix B include lists of projects developed for each technology. Storage has allowed CTS technologies to considerably increase their capacity factors and meet the dispatchability requirements demanded by utilities and regulators. Hybrid- ization, independent of whether it is combined with storage or fuels (such as natural gas, diesel, and biomass), can increase the reliability and the capacity factor of CST plants in general at a potentially lower capital investment cost than storage. Note 1. Grama, Wayman, and Bradford 2008: A guide to the impact CSP technologies will have on the solar and broader renewable energy markets through 2020. PART II Financial and Regulatory Schemes—The Current Situation CHAPTER 3 Policy Instruments Used to Promote CST in Developed Countries S everal countries—principally in the OECD area—have established dedicated regula- tory frameworks and incentives to encourage CST deployment. There are a wide range of regulatory measures and ï¬?nancial incentives that can be used to encourage development in the renewable energy sector (table 3.1). This chapter reviews the experi- ence of the prevailing regulatory and ï¬?nancial approaches for CST in the two largest markets —Spain and the southwestern United States. Both the Spanish FiT regime and the regimes combining Renewable Portfolio Standards (RPSs) with a variety of other instruments, which are in use in the southwestern United States, were hence evaluated against a set of four indicators: (a) the overall investment trends in the renewable energy sector; (b) the total CST capacity installed as a consequence of the introduction of a par- ticular framework or combination of incentives; (c) a share of CST generation in the over- all electricity supply mix; and (d) a structure of ï¬?nancial arrangements and the amount of private sector investments leveraged into the respective projects by the applied frame- work or a combination of incentives. Regulatory Framework and Financial Incentive Options The two principal options for the promotion of renewable energy are schemes centered on the FiT and RPS. An RPS is typically combined with several other incentives listed in table 3.2. The actual design, however, usually varies from jurisdiction to jurisdiction. A review1 of the literature suggests that the ability of a particular regulatory regime or instrument to trigger investments into the particular technology at the lowest pos- sible societal cost depends on the set policy objectives. If the stated policy objective is to increase the share of energy generated from renewable sources and to facilitate the development of respective industries, FiT schemes have been the most successful instrument employed by policy makers so far. In Europe in particular, the FiT regimes of Denmark, Germany, and Spain (see box 3.1) have won high praise, especially with regard to wind and solar photovoltaic power expansion. Meanwhile, quota systems applied in other European countries (such as Belgium, Italy, Sweden, and the United Kingdom) are largely considered by experts to have failed to bring about the desired levels of capacity growth in the renewable energy sector. This might lead to an assumption that FiTs are the best policy option available to date. However, recent modiï¬?cations of FiTs available for solar photovoltaics in Europe suggest that this might not always be the case. Different regulatory experiences in the United States where the RPS scheme prevails as the framework of choice also support 9 10 A World Bank Study Table 3.1: Policy instruments, characteristics, advantages, and disadvantages in implementation Objectives and Policy instruments characteristics Advantages Disadvantages Subsidy/tax incentive Fiscal instrument to reduce Easy to understand and High administrative costs. costs for renewable energy implement. Use of May not be cost effective. consumers or producers government funds to meet Needs effective monitoring particular policy objectives mechanisms to minimize risks. No guarantee of meeting quantitative targets. Renewable energy fund Financial instrument to Increase efï¬?ciency and Lack of experiences in fund support renewable energy, reduce management cost management. How to either in R&D, fund transfer, through professional fund combine public and private or in market-based management. interest/beneï¬?t through applications. effective management. Voluntary green electricity Mobilize consumers’ interest Generate additional funds Effectiveness depends on scheme and support. Provide from consumers, less use electricity prices and flexibility. of government resources, a consumers’ access to tool for engaging public and information and awareness. private sector participation. Not cost-effective. No guarantee for meeting quantitative target. High administrative costs. RPS/Green certiï¬?cate scheme Combines obligation for Encourages competition and May not do much for producers/consumers to cost effectiveness. Relies high-cost technologies. use green electricity with on market mechanism for Transaction costs can certiï¬?cation of green resource utilization and be high. Transparency production. (within green) technology and veriï¬?cation systems choice. needed. Sovereign Loan Guarantees Government shares some Can substantially lower High administrative costs. of ï¬?nancial risk of projects ï¬?nancing costs for a Amount of guarantees that otherwise would not particular project and tip provided might be limited. yet be supported in the the bankability of a stand- commercial marketplace. alone project. Feed-in tariffs Financial scheme ensuring a premium payment to eligible electricity production. Can ensure long-term return for investors, and is relatively simple to implement and flexible (for example, different technologies can be provided with different tariffs and contract lengths) May not ensure a long-term target. Requires good monitoring mechanism. Transparency needed. Not necessarily cost-effective. Source: Adapted from Gan et al. 2007. this argument. FiT schemes generally are not favorites of U.S. policy makers, who have instead often opted for RPSs coupled with various investment and production tax incentives, grants, and loan guarantees. Indeed, 36 U.S. states and the District of Columbia now have RPSs enacted, while only a handful of U.S. state jurisdictions are implementing FiTs—with none of them currently considering a FiT tailored for CST (U.S. DOE 2011). Regarding the speciï¬?c incentives for CST, the European and the U.S. experience are both very relevant and must be taken into account. This chapter will review the regula- tory incentive frameworks of Spain and several western and southwestern U.S. states (see table 3.3), in which CST penetration has been most signiï¬?cant (see tables B3.3–B.6 in Appendix B). Concentrating Solar Power in Developing Countries 11 Table 3.2: FiTs versus RPS schemes FiTs FiT regimes usually guarantee a payment to suppliers for energy generated from a speciï¬?ed source (such as renewable energy) at a deï¬?ned rate over an extended period. Quite often the FiT regime also provides preferential access to the grid. Tariff levels are usually set at a predeï¬?ned level or as a premium above the market price. FiT can further be tailored to the cost speciï¬?cs of a particular technology, as well as to different sites and characteristics of the energy resource (such as reflecting the level of intermittency or seasonal resource availability). Ideally, tariff levels are sufï¬?ciently high to mitigate the risk of high up-front investment cost and potential regulatory changes. The period, for which FiT payments are guaranteed, is also long enough to provide developers with adequate incentives to overcome otherwise prohibitive development risks—such as the cost of research, land leases, permitting, construction, guarantees, and warrantees. In most cases, utilities are required to off take all output generated at the respective technology-speciï¬?c tariff level, but are also usually allowed to pass the cost difference on to ï¬?nal consumers. FiTs can theoretically lead to societal gains in terms of reduced market prices, reduced levels of GHG emissions, and a decrease in fossil fuel consumption and/or imports. By contrast, FiTs also come at a societal cost, since they usually lead to an increase in the overall price of electricity per customer or to an increase in government’s subsidies. RPSs The prevailing regulatory framework in the United States and several other OECD countries (Belgium, Sweden, and the United Kingdom) is based on a quota system, generally referred to as an RPS combined with a variety of investment and production tax incentives, loan guarantees, ï¬?nancing from renewable energy funds, and voluntary purchases of renewable power by utilities. RPSs are designed to maintain or increase the contribution of renewables to the overall supply mix by obliging retail suppliers to reserve a speciï¬?ed amount or percentage of renewable energy to their individual supply mix. These obligations generally increase over time with suppliers being required to demonstrate compliance on a year-to-year basis. To fulï¬?ll their obligations, utilities usually have to rely, at least partly, on generation from their own facilities while being able to make up for shortfalls by purchasing renewable power from independent power producers (IPPs). In some jurisdictions, utilities are also allowed to meet at least a part of their obligations by trading in so-called Green Certiï¬?cates (GCs), which are created when a unit of energy is generated from a renewable source and which work much like tradable emission permits. Source: Authors’ data. Box 3.1: Germany’s recent FiT reform Germany introduced FiTs for a variety of renewable energies through its Erneuerbare-Energien- Gesetz (Renewable Energy Sources Act) in 2000. The law guaranteed renewable power gen- erators priority access to the grid and required utilities to off take any electricity produced by renewable sources at predeï¬?ned tariffs. The latter, and the period they were guaranteed for, were tailored to the respective capital and investment costs of each individual technology, with actual tariff levels decreasing at a certain percentage rate per year to set an incentive for cost reduction. Utilities were allowed to pass the additional cost above the nonrenewable AET through to ï¬?nal consumers. In addition, FiTs were combined with a variety of incentives like sub- sidized investment loans and tax credits to aggressively increase the share of renewable energy in the overall power portfolio to 30 percent by 2020. The law jump-started markets for renewable energies—especially for wind and solar PV—causing the share of renewable energies in ï¬?nal electricity consumption to increase from 6.3 percent in 2000 to 15.1 percent in 2008, with wind supplying more than 40,000 GWh and PV supplying around 4,000 GWh in 2008. According to Germany’s government, the FiT-based approach reaped considerable societal beneï¬?ts of approximately EUR 9.3 billion in 2006 from decreased spot-market prices because of the merit- order effect (del Rio and Gual 2007), avoided GHG emissions, and decreased fossil fuel imports, as well as adding around 280,000 new “greenâ€? jobs (BMU 2009). By contrast, the overall cost for ï¬?nal consumers rose to EUR 4.5 billion in 2008 (equivalent to EUR 1.1cent/kWh, or 5 percent of the average retail price), and is projected to have peaked at EUR 8.5 billion in 2010 and to decrease after until reaching zero by 2020. The recent spike in consumers’ cost has partly been caused by a larger-than-expected number of installations using renewable technologies, namely rooftop solar PV. According to the Association of Consumer Protection Agencies, rooftop PV capacity installed in 2009 will most likely cost ï¬?nal consumers EUR 10 billion over the course of their lifetime as opposed to the planned EUR 2.4 billion (VZB 2010). As a reaction to this devel- opment, the government recently decided to decrease FiTs for new PV-based capacity by up to 16 percent, with the stated aim of bringing tariffs in line with decreased investment and produc- tion costs and limiting the impact on consumers. 12 A World Bank Study Table 3.3: Currently installed CST capacity (MW) Regulatory Total under scheme Main features Total operating construction Total planned FiT—Spain • EUR 26.9375 cents/kWh over whole life 382.48 1,540 497 cycle or premium over market wholesale price up to EUR 34.3976 cents/kWh • Guaranteed grid access/off take RPS—U.S. total Federal incentives: 432.46 1,077 9,912 • Accelerated depreciation • Investment tax credit or renewable energy grants • Federal loan guarantees • Rural energy grants • Clean renewable energy bonds • Manufacturing investment tax credits • Production incentive payments California RPS 33% by 2020 + 363.8 718 6,896.8 Federal incentives + • Property tax exemption Nevada RPS 25% by 2025 + 64 0 2,184 Federal incentives + • Property tax abatement Arizona RPS 15% by 2025 + 2.6 280 1,010 Federal incentives + • Corporate tax credit • Property tax reductions • Business tax incentives Florida Federal incentives + 10 75 0 • Corporate tax credit • Renewable energy technology grants Source: Adapted from CSP Today 2010. Database of State Incentive for Renewables & Efficiency. Spanish Feed-in Tariffs The Spanish FiT for renewable energy is widely considered the most successful—at least until recently—and as such is certainly the most studied example. In 1998, the Royal Decree on the Special Regime (RD 2818/1998) gave renewable energy generators two options: (a) a ï¬?xed premium on top of the electricity market price or (b) a ï¬?xed total price (ï¬?xed feed-in) (del Rio and Gual 2007). The amended Royal Decree 436/2004 allowed renewable energy producers to sell their electricity to distributors or directly to the mar- ket. In both cases, support was tied to the AET.2 The 2007 modiï¬?cation, reflected in Royal Decree 661/2007, ultimately decoupled renewable energy support from the AET, tied it to the Consumer Price Index (CPI), and instituted a cap-and-floor system for the pre- mium on top of the electricity market price. Solar thermal electricity was ï¬?rst identiï¬?ed for the FiT support in the RD 436/2004 with the stated aim of developing a local CST industry. The 2007 reform increased the ï¬?xed FiT rate to EUR 26.9375 cents/kWh, and set a price range for the premium above the AET between 25.4038 and EUR 34.3976 cents/kWh for electricity generated by plants with up to 50 MW capacity. Either the ï¬?xed rate or the premium is guaranteed for 25 years for all electricity supplied to the grid under the scheme until 2013, adjusted annually according to the changed CPI minus 1 percent, and dropping uniformly to Concentrating Solar Power in Developing Countries 13 EUR 21.5 cents/kWh after 25 years of operation. Renewable energy projects including CST are also granted priority access to the grid. In theory, the consumer pays the incre- mental price increase, since utilities are allowed to pass on the cost difference to ï¬?nal consumers. However, this mechanism has not been applied. Only part of the cost dif- ference is passed through, resulting in a situation when the government must partially reimburse utilities for the additional cost related to the FiT. The ï¬?rst Spanish CST installation—Solucar PS-10, a tower system of 11 MW capacity— was connected to the grid in 2006. Ten more installations have since come online, bring- ing the total CST generation capacity in Spain close to 383 MW. Fifty-one installations are now under construction or planned. When completed, they will add more than 2,037 MW of CST generation capacity to the grid (CSP Today 2010). This tremendous increase in capacity and the need to reimburse utilities for the cost difference prompted the govern- ment to implement some modiï¬?cations of the FiT scheme starting in 2009. The primary motivation behind these changes—besides the need to deflate the investment bubble— was most likely to limit the societal cost of the FiT, especially in terms of restricting ï¬?s- cal reimbursements to utilities. The government’s Royal Decree 6/2009 established a pre- assignment register, for which developers need to sign up to be granted approval for their individual projects. A 500 MW annual cap for capacity eligible for the FiT was introduced. This translated into a 2.5 GW cap until 2013 based on the ï¬?rst-come-ï¬?rst-served principle (Boletín Oï¬?cial del Estado 283/2009). No plant is subsequently allowed to choose the ï¬?xed premium variant of the FiT during its ï¬?rst year of operation. While these steps will contribute to controlling societal costs, they most likely will not be sufficient to deflate the investment bubble, since FiTs remain relatively generous for capacity coming online until 2013. At the same time, there is a considerable degree of insecurity in the market since the current framework only extends to 2013. Some modiï¬?cations, such as annual capacity caps, could further help deflate the investment bubble and avoid unnecessarily high societal costs. The most crucial modi- ï¬?cation could be to align the FiTs with actual capital and investment costs. A reverse auctioning mechanism (as outlined in box 3.2) for a set amount of capacity eligible for Box 3.2: The renewable energy reverse auction mechanism A potential way to assure maximum cost efï¬?ciency of the CST capacity installed under a RPS scheme could be in the application of so called Renewable Energy Reverse Auction Mechanisms (RAMs). Already being used for wind power under RPS schemes in New England and proposed for solar PV by the California Public Utilities Commission under the Californian RPS (CPUC 2009), RAM would require developers to bid the lowest possible price per kilowatt-hour, under which they would still be willing to develop a CST project, with utilities accepting the lowest-cost projects up to the total capacity cap. While setting a long-term investment signal, this approach has the beneï¬?t of securing the most cost-efï¬?cient investment while avoiding any potential windfalls to developers at the expense of ratepayers. However, RAMs would require setting up a standardized procurement system under which utilities would be able to rank individual bids, including their cost-efï¬?ciency characteristics. The least-cost projects would then be offered to sign PPAs with utilities for up to the general capacity cap or the target established under the RPS. RAMs would thereby secure preapproved utility cost recovery, cost certainty, and a minimum cost impact for consumers while still presenting regulatory certainty for developers (Kubert and Sinclair 2010). 14 A World Bank Study the FiT in a given year could be a potential solution in this regard. The experience shows that caps on individual plants’ capacity are likely to lead to inefficiencies. The la er is linked to considerable gains to be realized from increasing the scale of individual CST plants, which can be foregone by limiting the maximum amount of capacity of a single plant eligible for the FiT scheme. Renewable Portfolio Standards and CST in the United States Of the 36 U.S. states that enacted the RPS scheme by 2010, 16 have provisions requiring a speciï¬?c level of solar power in the supply mix. These states include Nevada (1.5 percent by 2025), Arizona (4.5 percent by 2025), and New Mexico (4 percent by 2020). Usually the RPSs are combined with a variety of other incentives, such as federal loan guarantees, investment and production tax credits, renewable energy grants, property and sales tax breaks, and Clean Renewable Energy Bonds coming from federal and state governments (see also table 3.3 above). The major downside of the RPS scheme with regard to CST seems to be its inabil- ity to a ract nonresource ï¬?nancing terms for project development without the avail- ability of loan guarantees at scale. In most cases, small and mid-scale developers are unable to secure nonrecourse ï¬?nancing. For this very reason, until recently, most plants that received construction permits in the United States were based on balance-sheet ï¬?nancing. This is rather different from the Spanish case where nearly every project was ï¬?nanced on a nonrecourse basis. This situation has, however, changed with the availability of relative large-scale fed- eral loan guarantees starting in 2009, providing the opportunity to improve the bank- ability of an individual project. The U.S. Department of Energy (U.S. DOE) is authorized to issue loan guarantees up to the total amount of US$10 billion to projects in the ï¬?eld of renewable energy, energy efficiency, and advanced transmission and distribution. CST is one of the eligible technologies under the current U.S. DOE loan guarantee program. The amount of the provided guarantees varies among individual projects, but the total project value is usually higher than US$25 million. The full repayment is required over a period not exceeding 30 years or 90 percent of the projected useful life of the physical asset. BrightSource, a California-based company, was one of the ï¬?rst awardees of the fed- eral loan guarantee program that secured a US$1.6 billion loan guarantee for its 383 MW Ivanpah power tower project in California. The Spanish developer Abengoa secured another US$1.45 billion in guarantees for its 250 MW Solana plant in Arizona. In both cases, the respective guarantees covered around 75 percent of the total expected project cost. Currently there are apparently another ï¬?ve–six CST projects in the pipeline being evaluated for receiving a loan guarantee. Though loan guarantees are apparently crucial for improving the bankability of proj- ects, for smaller and mid-size developers, such an incentive comes at a certain adminis- trative and compliance cost, including obligations on the use of local manufacturing and services and labor and environmental requirements. In addition, as already mentioned, the processes to secure the guarantee can be fairly slow, with no assurance that the cur- rent scheme will be extended once the US$10 billion has been allocated (which at the current pace of awarding could happen relatively soon). By contrast, proponents usually indicate the hands-off character of the loan guaran- tee program, allowing the market to make decisions as opposed to governments actively picking winners. Another discussed advantage is that fees charged for the guarantees Concentrating Solar Power in Developing Countries 15 can technically be set at a sufficiently high level to cover expected losses from the guar- antee program (depending on the expected rate of default). Investment Trajectories in Spain and the United States To assess both regulatory approaches in terms of their ability to provide sufficient incen- tives for developers to deploy CST, the following trends were analyzed: Overall Investment Trends in the Renewable Energy Sector Spain is a signiï¬?cant player in the renewable energy sector with overall investments of US$10.4 billion in 2009, down by approximately 50 percent from 2008 because of the ï¬?nancial crisis. The largest chunk of these investments went to wind (34.2 percent or US$3.5 billion) and solar (60.6 percent or US$6.3 billion) power generation. Total invest- ments have grown at about 80 percent over the last ï¬?ve years with total installed renew- able capacity having grown by 9.1 percent in the same period, reaching 22.4 GW or 30.1 percent of total installed electricity capacity (PEW Charitable Trusts 2010). In 2010, wind and solar (both PV and CST) accounted for 23 percent of the total installed capac- ity and 18 percent of total electricity generation. Total renewable capacity installed was 23 GW. This impressive investment trend is probably the result of the relatively gener- ous terms of the FiT framework. The United States recently dropped to the second rank globally in terms of overall investments in renewables, losing their leading position to China. The same happened with regard to the technology in review, CST, in which the United States just lost its top rank to Spain. Overall renewable investments in the United States stood at US$18.6 billion in 2009, down by 42 percent from 2008, also because of the ï¬?nancial crisis, but were set to have increased considerably in 2010 when roughly one-third of the clean energy stimulus funding was spent. The largest chunk of the overall investments went to wind (43.1 percent or US$8.0 billion), biofuels (22.1 percent or US$4.1 billion), and solar (17.4 percent or US$3.2 billion, both PV and CST). Total investments have grown by over 100 percent over the previous ï¬?ve years with the total installed renewable capacity having grown by 24.3 percent in the same period, reaching 53.4 GW or 4 percent of total power capacity (PEW Charitable Trusts 2010). Total CST Capacity Installed as a Consequence of the Framework Installed With regard to Spain, most of the installed CST generation capacity came online after the landmark Royal Decree 661/2007, even though projects were previously developed because of the tailoring of the FiT to CST applications in 2004. The overall capacity added since the introduction of the FiT has since reached nearly 383 MW with a further 1,540 MW under construction. Regarding the United States, one would have to subtract the nine SEG plants, which came online in the late 1980s and early 1990s from current installed capac- ity. New capacity coming online since 2006—the year in which the ï¬?rst of the cited RPS frameworks was introduced—has added up to 78.7 MW, with 1,077 MW currently under construction (CSP Today 2010). However, the United States has announced a considerably higher amount of capacity to be developed—9,912 MW compared to 497 MW in Spain. The Share of CST Generation in the Overall Electricity Supply Mix Despite the recent considerable increase in plants in operation, the overall share of CST in the electricity supply mix of both the United States and Spain is still relatively small. 16 A World Bank Study The most recent yearly overall electricity generation data available from the Interna- tional Energy Agency (IEA) for Spain and the United States, for 2008, shows total Span- ish electricity supply at 311,130 GWh and total U.S. electricity supply at 4,343,820 GWh (IEA 2010). Assuming a capacity factor for installed generation of around 22–24 per- cent, the overall CST-based output would be equal to 761.1 GWh in Spain and 860.5 GWh in the United States in 2010. Even compared to the 2008 supply data, this would mean that the share of CST generation in the overall electricity supply mix amounts to approxi- mately 0.25 percent for Spain and 0.02 percent for the United States. Assuming that all capacity currently under construction or in development would come online, the over- all share, relative to the 2008 supply data, would increase to 1.6 percent for Spain and 0.52 percent for the United States. The Structure of Financial Arrangements and the Amount of Private Sector Investments Leveraged into the Respective Projects Using Incentive Mechanisms With regard to Spain, the tailoring of the FiT to CST in 2004 already triggered the ï¬?rst devel- opment proposals, but it was not until modiï¬?cation of the FiT by Royal Decree 661/2007— which considerably increased tariff rates and premiums and decoupled them from market reference prices—that a large number of projects became bankable. Although actual data with regard to ï¬?nancial structures are hard to come by—developers are fairly secretive in this regard in both countries—most, if not all Spanish projects, seem to have triggered limited recourse or nonrecourse ï¬?nancing. Currently, more than 1.5 GW of capacity has received either limited recourse or nonrecourse ï¬?nancing from domestic or international commercial banks. This contrasts with plant developments in the United States, where, until the recent large-scale provision of loan guarantees, apparently only very few projects were based on limited recourse or nonrecourse ï¬?nancing. Analysis and Conclusions Both the United States and Spain have seen a rapid uptake of CST technology over the past several years, and the trend is likely to continue, despite minor modiï¬?cations of the Spanish FiT. Based on the investment trends analyzed above, the following conclusions can be drawn: 1. FiTs have been the most successful incentive for jump-starting renewables’ market penetration and encouraging rapid development of domestic CST companies. Spain is regarded as the leader in the CST ï¬?eld, and it is likely to continue in this role because of the continuing success of the FiT scheme. The Spanish FiT has triggered a considerable number of projects in a relatively short time and enabled rather favorable ï¬?nancing terms compared to the RPS schemes in the United States. Although coming at a considerable ï¬?scal cost, the overall net societal beneï¬?ts in the form of reduced spot market prices for electricity, lower GHG emissions, a reduced need for fuel imports and net contributions to GDP seem to substantial (APPA 2009). 2. FiTs have encouraged large, integrated infrastructure companies to enter the CST market, providing be er opportunities for large-scale project development. The large, integrated infrastructure companies of Spain were motivated to pursue CST because of the secure cash flow revenue streams guaranteed by Concentrating Solar Power in Developing Countries 17 the FiT scheme. In the United States, start-up companies, not large developers, have ï¬?rst brought the technology to construction. However, as the technology matures, it seems that large companies would become involved. The Spanish giant Abengoa, for example, has made its way into the U.S. market by securing a US$1.45 billion in guarantees for its 280 MW Nevada-based Solana project. This incentive scheme is likely to beneï¬?t large companies, which are gener- ally in a be er position to ï¬?nance larger installations, and to take advantage of economies of scale—one of the primary assumed drivers for cost reduction for CST technologies. 3. When coupled with well-designed power purchasing agreements, tax incen- tives, grants and especially loan guarantees, RPSs can also be an adequate incentive for CST industry growth. The success of RPSs seems to be associated with the provision of simultaneous schemes, such as well-designed PPAs, tax incentives, grants, and especially loan guarantees that make CST projects a ractive for developers and commercial banks. More than 80 percent of the cost of a CST installation lies in initial con- struction and connection costs, making it important for developers to receive assistance in ï¬?nancing the upfront costs associated with large-scale CST devel- opment until the technology can reap its high, cost-reduction potential. Loan guarantees can be a powerful complementary instrument under an RPS scheme, as evidenced in the United States. However, this set of policy instruments imposes high administrative costs on developers and on the governments. 4. The details of any incentive scheme—whether FiT or RPS—are critical to its success, perhaps more critical than the choice of a particular incentive scheme to apply. For example, FiTs that deviate too much from the “market clearingâ€? price are either likely to fail to a ract sufficient private sector investment if they are set too low or set for too short a timeframe, or to grant a potential windfall to developers and investors at the expense of consumers and/or taxpayers if they are set too high or guaranteed for too long. Potential solutions for these problems include, for example, a reverse auc- tion mechanism, which in theory could result in a tariff reflecting the conï¬?- dence of a developer to implement the project at the bid price that should be close to the actual technology cost. An additional advantage of a reverse auc- tion would be that FiTs would not necessarily have to be reviewed regularly to align them both to investors’ interest and the public interest. If technology- speciï¬?c tariffs are set by the regulator, periodic tariff reviews would undermine the main advantage of FiTs—their predictability for investment decisions. Under a classic FiT regime, a scheduled phase-out of the granted FiT by a cer- tain amount every year could also be a potential solution. However, if a sched- uled phase out is applied; it might be problematic to ï¬?nd a reduction rate for the FiT that brings it in line with the actual technology cost reduction rate. The Spanish experience also shows the importance of introducing a capacity ceiling to control societal costs. As with stand-alone RPS schemes, concerns are raised with respect the high administrative cost on developers and that it may not provide sufficient incen- tives to overcome the high investment costs. It is therefore of utmost importance 18 A World Bank Study that RPS schemes not be overly burdensome in terms of administrative com- pliance cost and that incentives be tailored toward the characteristics of CST. Even if the RPS scheme is appropriately tailored, there might still be the need to provide loan guarantees on a large scale to buy down the real and perceived technology risk. The fact that investments in the technology in the United States only took off after the introduction of a comprehensive and generous loan guar- antee program seems to support this conclusion. 5. Continuity is essential for the success of any policy instrument. Developers and investors are more likely to assume the ï¬?nancial risk of a CST project if the support scheme in place is credibly guaranteed for a certain period. This is especially important with regard to the timeframe for FiTs, since they were usually able to trigger nonrecourse, project ï¬?nancing. As the la er are obviously based on consistent cash flow projections, any insecurity with regard to the level or timeframe of a FiT will most likely deteriorate conditions for this type of ï¬?nancing and hence for CST development under the respective frame- work. This can present a problem, since even when periodic tariff reviews or a scheduled phase-out are enshrined in the FiT framework, a sudden change in government priorities or a reassessment of the respective policy goals might well trigger a modiï¬?cation of the tariff framework. Such a modiï¬?cation—regardless of whether or not it is justiï¬?ed from an economic point of view—might have a negative effect on the overall investment trends in the market. In the case of RPS schemes, best-practice PPAs should provide for a com- parable long-term predictability of cash flows. However, the experience of the developers in the United States suggests that, so far, PPAs alone have not been able to trigger large-scale investment in the technology, let alone nonrecourse ï¬?nancing for CST plants. This highlights the need to ensure predictability for both developers and investors. This could be obtained by establishing off take arrangements that allow for a viable and predictable income stream, which in turn would make these projects bankable (see section 7.3 on PPA Structuring in Chapter 7). However, unless the public sector provides additional reliable incentives to cope with the large upfront investments, PPAs alone are unlikely to provide the necessary cash-flow security. 6. Particular conditions of a country will determine the best approach. Both FiTs and RPS schemes are ultimately funded by consumers—be it in their capacity as taxpayers or rate payers—and, as such, will only be appro- priate in jurisdictions with well-established governance and electricity regula- tory frameworks. Based on the material reviewed in this evaluation, it seems likely that, given the potentially higher administrative costs associated with a complex array of incentives, such as tax incentives and grants, which usually go along with RPS schemes, a FiT combined with concessional and noncon- cessional loans might, in theory, be a preferable option for jump-starting industry development, because of its simplicity and predictability. The relative flexibility of FiTs in targeting different technologies might well prove superior to RPS schemes. By contrast, one must keep in mind that the methodology for designing and structuring technology-speciï¬?c FiTs is rather a “try and adjustâ€? approach, requiring keeping track of technology developments and evolvement of manufacturing markets to produce CST components locally (see Chapter 6). Concentrating Solar Power in Developing Countries 19 The tremendous downside of a FiT from a public policy maker’s point of view is certainly its considerable societal cost. Incentives should be aligned with the overall affordability of consumers and taxpayers. This holds true for both developed and developing countries, although in the former the impact is less immediate because of higher income levels of the population. There are potential options to minimize the societal cost in the form of a cap on the overall capacity eligible for a FiT, and conducting periodical tariff reviews to adjust FiTs to changes in the investment and production costs or simply sched- ule the phase-out of the tariff over a certain timeframe. Nevertheless, in situ- ations where the political economy rules out the use of a FiT, or where it is politically inacceptable to pass the full cost increase on to the end user, a strong RPS combined with a variety of incentives might also be effective in promoting CST development, although potentially at a slower pace. In any case, one can assume that a comprehensive sovereign loan guarantee program would have to be launched in order to trigger desired investments under an RPS scheme, espe- cially in emerging markets where investors still perceive project risk as higher than in the developed markets. Notes 1. The literature review included the following sources: Durrschmidt 2008; Rowlands 2004; Astrad 2006; Fouquet and Johansson 2008; del Rio and Gual 2007; Nilsson and Sundqvist 2006; Lorenzoni 2003; Nielsen and Jeppesen 2003. 2. Meaning the average between different electricity tariffs that tend to vary for residential, business, and industrial customers, and for any single class depending on the time of day or by the capacity or nature of the supply circuit even within a single region or power district. CHAPTER 4 Renewable Energy Schemes Supporting CST in Developing Countries A variety of approaches have been taken in developing countries to incentivize investment in renewable energy in general and CST in particular. This chapter will review and analyze those currently under planning or implementation in the Middle East and North Africa (MENA) region, India, and South Africa. MENA Incentive Schemes Algeria Algeria stands out as a notable example of a country within the region that has taken steps to introduce price incentives for renewable energy. In 2004, the Algerian govern- ment issued a decree instituting FiTs. Under the decree, premiums are to be granted for electricity produced from renewable energy resources. The premiums are expressed on the percentage of the average wholesale price set by the market operator based on bids from generators and buyers of electricity, as deï¬?ned in the law on gas and electricity (GOA 2002). The tariffs are differentiated by technology and do include a tariff for CST. For plants producing electricity exclusively from solar energy (including both CST and CPV), the premium is 300 percent of the average wholesale price. For hybrid solar- gas power plants with solar energy contributing at least 25 percent of the plant’s output, the premium is 200 percent. For smaller proportions of solar energy in the plant out- put, the premium is set at lower levels—for example, 180 percent if solar generation is between 20 and 25 percent (JORADP 2004). Even though the tariff level can vary over time (because of the connection to the price set by the market operator), the size of the premium in relation to the average system price is guaranteed for the full lifetime of a project (FuturePolicy.org 2010). While the introduction of a feed-in-tariff (FiT) scheme in Algeria is an encouraging example that holds promise for the future, the price incentives along with the entire structure of the scheme do not seem to be a ractive enough for investors in solar energy. The proponents of the Algerian renewable energy projects currently in the pipeline (including CST projects) appear to put more faith into leveraging concessional capital from sources such as the CTF and large European Union–sponsored initiatives, such as Desertec (Fenwick 2011)—the only plant currently under construction is an integrated solar combined cycle (ISCC) plant at Hassi R’Mel with a 25 MW parabolic trough CST component in combination with a 125 MW combined cycle gas turbine, which was ï¬?nanced by Kreditanstalt für Wiederau au (KfW)—the German bilateral development bank, and the European Investment Bank (EIB). Part of the reluctance of the private 20 Concentrating Solar Power in Developing Countries 21 sector to embrace the Algerian FiT scheme may be caused by the lack of protection from the wholesale market price volatility and the influence of domestic fuel subsidies on the whole sale electricity pricing. Egypt Egypt has no speciï¬?c price support mechanism yet in place for renewable energy. How- ever, the need to cover additional costs for renewable energy projects through tariffs has been recognized by the country’s Supreme Energy Council, and some other policy measures have been initiated to promote renewables and especially CST. These include (a) an exemption from customs duties on wind and CST equipment; (b) the ï¬?nalization of the land use policy for wind and CST developers; (c) the acceptance of foreign currency denominated PPAs; (d) the conï¬?rmation of central bank guarantees for all build-own- operate (BOO) projects; and (e) the support for developers with respect to environmental, social, and defense permits and clearances (CIF 2010). Despite the lack of speciï¬?c price support mechanisms, an ISCC plant with a 20 MW CST component is already operating at El-Kureimat, located roughly 100 kilometers south of Cairo. The construction of this plant was ï¬?nanced by JBIC and again supported by a grant from the Global Environment Facility (GEF), for which the World Bank was the executing agency. Morocco Morocco does not have price incentives yet in place for renewable energy. Nevertheless, the country is aiming to have 2,000 MW of solar power generation capacity installed by 2020, starting with the ambitious Ouarzazate 500 MW CST project. The project is expected to utilize parabolic trough technology equipped with storage. The legal, regu- latory, and institutional framework is being set up with several laws enacted in early 2010, including the renewable energy law, the law creating the dedicated Moroccan Solar Agency (MASEN) to implement the Morocco Solar Plan and the law se ing up the Energy Efficiency Agency. Morocco’s recently issued Renewable Energy Law (REL) (Dahir 2010) and the Moroccan Agency for Solar Energy (MASEN) Law (Dahir 2010) are intended to scale up the development of renewable energy with special focus on solar technologies. MASEN is entrusted by the government to develop at least 2,000 MW of grid-connected solar power by 2020, and in particular to conduct technical, economic, and ï¬?nancial studies, as well as to support relevant research and fundraising, to seek utilization of local indus- trial inputs in each solar project and to establish associated infrastructure. While the generated electricity must be sold in priority to the national electric utility ONE (Office National de l’Electricité) for the domestic market, the law allows MASEN, under condi- tions speciï¬?ed in the convention signed with the government (described below), to sell electricity to other public or private operators on national or export markets. An obvious export market would be the European Union. EU Directive 2009/28/ EC allows EU member states to import renewable energy–generated electricity from projects in third countries using their respective incentive mechanisms in order to fulï¬?ll the respective national targets by 2020 if a variety of conditions are fulï¬?lled. This could be the framework for the establishment of major export markets, which could ensure a viable income stream for a major scale-up of CST in Morocco. In reality, the export option, especially at the desired FiT level, is rather difficult to realize for a variety of reasons, including the following: (a) the directive needs to be transferred into national 22 A World Bank Study laws, which has so far experienced delays in most cases; (b) approvals in each respective jurisdiction are required to use the electricity generated in nonmember countries against the country compliance with the RE targets; and (c) the EU Directive itself, which in Article 9 sets up certain time limitations on when renewable energy generated in non- member countries can count toward domestic renewable energy targets. Notwithstanding these potential limitations with regard to export markets, the US$9 billion Morocco Solar Plan, launched in November 2009, calls for the commission- ing of ï¬?ve solar power generation plants between 2015 and 2020, for a total capacity of 2,000 MW. With this plan, 4,500 GWh annually will be produced from solar energy alone. In October 2010, conventions were signed between MASEN and the government on the one hand, to stipulate state support for the Moroccan Solar Plan, and MASEN and ONE on the other hand, to cover the conditions for connection and operation of solar power plants and for sales of electricity. According to the convention, the state will compensate MASEN for the “gapâ€? between the two PPAs. ONE is already operat- ing an ISCC plant with a 30 MW solar-assisted combined cycle gas turbine (CCGT) at Ain Beni Mathar (northeastern Morocco), which is ï¬?nanced by the African Develop- ment Bank (AfDB) and supported by a grant from the Global Environment Facility executed by the World Bank. Issues Related to Regulatory Frameworks in the MENA Region Information on the enabling policies for CST in MENA countries remains scarce. Morocco’s commitment to a racting private sector participation in CST development on a project- speciï¬?c PPP basis, and Algeria’s decree of 2004 introducing technology-speciï¬?c premi- ums for renewable energy are notable exceptions. However, the lack of implementation mechanisms in the case of Algeria and Morocco and the lack of deï¬?ned incentive policies in the case of other countries to support CST (and other renewables) generate regulatory uncertainty that, if not rectiï¬?ed, may become a serious deterrent to future private invest- ments in the sector. The individual bilateral and multilateral projects to build up solar power capacity in MENA may expedite, but cannot substitute the development of such national policies. This is especially true since the ï¬?rst CST projects in MENA are expected to come on line in 2014–15, and even then export opportunities could be limited, and thus generation would essentially focus on domestic markets. Given the circumstances, while there is a strong rationale for strengthening mech- anisms and institutions to enable investments, certain large-scale investment projects may be justiï¬?ed on a stand-alone basis. Support schemes for these projects are highly customized, but usually involve such common features as (a) a long-term PPA between the power utility, or another form of a single buyer, and the generator; (b) a competi- tive bidding process for the generators; and (c) commitments from the government and ï¬?nanciers, sometimes including international donors, to support the project. Under the CTF-supported program to scale up CST in MENA, the PPA model is being utilized for the Ouarzazate project in Morocco, among others. For a large donor- supported project, the project model is innovative, since it relies on the private sector— not as just a supplier of equipment, but as an integral partner in the implementation scheme under a public-private partnership. The rationale for stand-alone projects (as opposed to policies driving investments in projects) needs to pass a reasonable test of sustainability and replicability. A large stand- alone project may enjoy a high-proï¬?le status that allows it to receive an unprecedented Concentrating Solar Power in Developing Countries 23 level of support from the government and the donors. As a result, the project may create a ractive incentives for private sector participation, but such conditions may not be easy to replicate. At the same time, large-scale demonstration projects can be essential for reaching the critical mass of investment in new technology, such as CST. The success of the Ouarzazate project in Morocco in a racting private sector inves- tor participation in the project on a PPA basis could be a considerable breakthrough, since the PPP model for CST deployment Most of the previous a empts to a ract pri- vate sector investment in CST have failed not only in MENA, but in India and Mexico as well. In MENA, the ISCC projects in El-Kureimat (Egypt) and Ain Beni Mathar (Morocco) were either designed as public sector projects from the beginning, as in the case of El-Kureimat, or had to be restructured because the original project design based on the IPP concept did not work, as in the case of Ain Beni Mathar. MENA Incentive Conclusions There are four or ï¬?ve models (depending on classiï¬?cation details) to be considered for supporting CST in the MENA region. The models given most a ention in the developed country markets are the FiT and RPS models. In the MENA context, however, the cur- rently relevant choices are largely between the pure public project model (supported by concessional ï¬?nancing) and the PPP model. The MENA experience to date shows that â–  The region is not quite ready to embrace FiTs or RPS, although efforts to cham- pion the introduction of such schemes are ongoing. â–  IPP/PPA schemes have not worked well in the past, as illustrated by the GEF projects that had to be restructured into public sector projects. â–  Combined PPA/PPP schemes are being tried out for some individual large projects (Morocco), and they have a be er chance of success than the earlier a empts to engage the private sector that used a pure IPP concept. The CST investments planned in the Middle East and North Africa for the next decade and beyond are, to a large extent, driven by individual projects supported by the Euro- pean Union, and by multilateral and bilateral sponsors. The policies initiated domes- tically to a ract investment that would serve the domestic markets are few, although Morocco’s commitment to test the PPP model and Algeria’s FiT scheme launched in 2004 are encouraging examples. The approach currently taken under the CTF-supported CST scale-up program in MENA assumes that concessional ï¬?nancing will help address the issues of both high capital costs and the existing uncertain policy and regulatory framework. The expec- tation is that, with more clarity in the policy framework for CST development in the Middle Eastern and North African countries by 2015 or so, the need for concessional ï¬?nancing will be reduced (CIF 2010). However, these investments will require to be fol- lowed by appropriate national policies, such as FiTs or RPS/quotas combined with other supporting instruments to achieve a transformational impact in the long term. India’s Incentive Schemes Over the last few years, India has introduced incentive schemes for solar power, both at the central and state level. Among the states, the most advanced are Gujarat and to a 24 A World Bank Study lesser degree Rajasthan, where project developers had concluded PPAs and are prepar- ing to close the deals with ï¬?nanciers. State-Level Incentives At the state level, Gujarat has emerged as the frontrunner in a racting private invest- ment in solar power. The Gujarat government has laid out the norms of the Renewable Purchase Obligation (RPO) policy and has set the ambitious target of installing 1,000 MW of solar power capacity by the end of 2012 and 3,000 MW in the following ï¬?ve years. According to the Solar Power Policy issued by Gujarat’s government in January 2009, each PPA shall include a speciï¬?c levelized ï¬?xed tariff per kilowa -hour and is concluded for a period of 25 years as shows in table 4.1. Table 4.1: Gujarat tariff rates for solar projects Tariff for Photovoltaic Tariff for Solar Thermal Sr. No. Date of Commissioning Projects (INR/kWh) Projects (INR/kWh) I Before December 31, 2010 13.00 for the ï¬?rst 12 years and 10.00 for the ï¬?rst 12 years and 3.00 during years 13–25 3.00 during years 13–25 II Other projects commissioned 12.00 for the ï¬?rst 12 years and 9.00 for the ï¬?rst 12 years and before March 31, 2014 3.00 during years 13–25 3.00 during years 13–25 Source: Adapted from Government of Gujarat 2009. Recent reports indicate that the state-owned utility GUVNL has signed PPAs with as many as 54 solar power generation companies for 537 MW. The total solar power installation commitments signed via Memoranda for Understanding with the Govern- ment of Gujarat have been reported at 933.5 MW, which is close to the installation target of 1,000 MW by 2012 (Panchabuta 2010a). Central Government Level Incentives—Jawaharlal Nehru National Solar Mission The Government of India (GOI) announced the JNNSM in January 2010, which set a target of 20,000 MW of solar power installed by 2022. The target for the ï¬?rst phase (by 2013) is 1,000 MW of grid-connected solar power capacity, of which 500 MW should be solar thermal projects and 500 MW solar PV.1 An additional 3,000 MW is targeted by the end of the second phase in 2017. It is understood that the ambitious target of 20,000 MW or more by the end of the third phase in 2022 will be dependent on the learning success of the ï¬?rst two phases (MNRE 2009). Since the central government issued guidelines for switching from state supported schemes to JNNSM (CERC 2009), most of the discussion about incentives for solar energy in India has focused on this new initiative by the central government. The available information on the projects whose developers have chosen to switch (“migrateâ€?) from the state-level schemes in both Gujarat and Rajasthan to JNNSM shows that 16 projects with a total capacity of 84 MW have officially “migrated.â€? Of these, only three projects with a total capacity of 30 MW were CST projects. RENEWABLE PURCHASE OBLIGATION Under the JNNSM, investment in the grid-connected solar power will be supported “through the mandatory use of the renewable purchase obligation by utilities backed with a preferential tariff.â€? The key driver for promoting solar power will be a renew- Concentrating Solar Power in Developing Countries 25 able purchase obligation (RPO) mandated for power utilities (distribution companies, or DISCOMs) with a speciï¬?c solar component. This is expected to drive utility scale power generation, both solar PV and solar thermal. The solar-speciï¬?c RPO will be gradually increased, while the tariff ï¬?xed for solar power purchase will decline over time (MNRE 2009). The MNRE guidelines mention a national level solar RPO of 0.25 percent of the total annual electricity purchased by the utilities by the end of the ï¬?rst phase and 3 per- cent by 2022. The state governments are responsible for se ing solar RPOs in their respec- tive states. Related to the RPO targets are the government procurement quotas used under the NNSM. For the ï¬?rst round of competitive bidding, implemented through the reverse auc- tion mechanism and conducted in 2010 to advance the progress toward the 0.25 percent target, the government solicited bids for 150 MW of PV and 470 MW of CST projects. In conjunction with the RPO targets, the government mandate to procure the solar power capacity is the ï¬?rst and foremost element of the Indian incentive scheme for solar power. PREFERENTIAL TARIFF The preferential tariff is the second element in the scheme. The Central Electricity Regu- latory Commission (CERC) guidelines published in July 2010 (CERC 2010b) specify INR 15.31/kWh (or about US$0.34/kWh, converting at 45 INR/US$) as the levelized total (single-part) wholesale tariff for CST in the ï¬?rst phase of the JNNSM. Provided the capi- tal costs of CST plant construction in India will be consistent with the capital expenditure (CAPEX) norm set by CERC 2010a at INR 153 million per MW (US$3400/kW),2 the target (pretax) return on an equity basis on this levelized tariff is calculated to be 19 percent per year for the ï¬?rst 10 years and 24 percent per year from the 11th year onward. Solar energy priced at INR 15.31/kWh stands out as much more expensive than con- ventional power, which tends to cost on average about INR 2.5/kWh or less in India. Power from grid-connected PV is even more expensive, with the levelized CERC approved tariff for Phase 1 at INR 17.91/kWh. To sell this energy to distribution utilities, the nodal agency—NTPC Vidyut Vyapar Nigam Ltd. (NVVN), the trading arm of the national power utility National Thermal Power Corporation Ltd. (NTPC)—will be bundling solar power with electricity from coal and possibly nuclear plants. In one useful illustration (IDFC 2009), the proportions between solar and conventional energy bundled by NVVN for sale to state distribution utilities could be 1:4,3 with the electricity from the unallocated quota costing INR 2.5/kWh. This would result in an overall (weighted average) price of about INR 5–6/kWh. It should be noted, however, that the levelized tariff of INR 15.31/kWh for CST (as well as the respective tariff for PV) is not intended to be used as a guaranteed, European- style FiT. The price eventually included in the PPA between the solar power producer and NVVN is reduced by the competitive procurement procedure mentioned earlier. The bidding round completed in November 2010 for the ï¬?rst 470 MW of CST capacity saw investors offering discounts in the range of 20–31 percent from the ceiling price of INR 15.31/kWh. As many as 66 bids for CST projects were received by the government by the closing date (in addition to 363 for solar PV),4 while only 7 CST companies were eventually short-listed (Panchabuta 2010b). In the bidding scheme to procure the ï¬?rst 470 MW of CST capacity, the preferential tariff of INR 15.31/kWh was used as a ceiling price with many bidders have offering prices below that level. The seven winning bids were between INR 10.49 and 12.24/kWh. 26 A World Bank Study OTHER INCENTIVES Besides the RPO, the competitive procurement scheme and the preferential tariff, another element of the incentive scheme included in the guidelines is the Renewable Energy Certiï¬?cate (REC) mechanism. The certiï¬?cates will be speciï¬?c to solar energy and will be bought and sold by utilities and solar power generation companies to meet their solar power purchase obligations (MNRE 2008). In addition to the core elements of the incentive scheme already mentioned, other incentives available to CST developers in India include (a) accelerated depreciation and (b) generation-based incentives (MNRE 2008).5 In both cases, the CERC position is that such incentives and subsidies should be taken into account when calculating the appli- cable tariff. In other words, these incentives should not be additional to the preferential tariffs offered under the JNNSM. Finally, a peculiar feature in India is the Clean Development Mechanism (CDM) beneï¬?t-sharing provision, under which CDM credits earned by renewable energy proj- ects must be shared between the project developer and the buyer of renewable energy. In Tamil Nadu, for example, the regulator issued guidelines under which CDM credits would accrue to the developer in the ï¬?rst year, but then the developer’s share would decrease by 10 percent every year in favor of the power purchaser until it reaches a 50:50 ratio (TNERC 2010). The concept of CDM sharing has been criticized by those who believe that CDM beneï¬?ts should belong only to the developers, who deserve them by virtue of going through the cumbersome process of CDM, including required additional tests for their projects (Sarangi and Mishra 2009). Issues Related to India’s Incentive Schemes As described in the previous chapter, the regulatory environment for deployment of solar energy in India is rapidly evolving and can be characterized as both relatively advanced and rather complex. In fact, the multiplicity of the incentive instruments intro- duced under the JNNSM can be a source of confusion about the nature and role of each instrument. Under the NNSM, as long as a sufficient number of suppliers are willing to bid below the ceiling price (which so far has been the case), the incentive scheme oper- ates as a quantity-based scheme that is closer to an RPS than a FiT scheme.6 A tendering scheme or auction could be a more accurate description of the Indian incentive framework for CST. Like RPO/RPS, tenders and auctions are quantity-based instruments—that is, the required quantity is speciï¬?ed in advance and the price is set by the market. The process of an RPO/RPS, however, is somewhat different from that of a tender—for example, an RPO/RPS does not usually involve sealed ï¬?nancial bids. Instead, the price is agreed on between the supplier and off taker through negotiation. In the international practice, auctions have often been used as the basis for long- term PPAs. Bidders are usually asked to compete on the basis of price per kilowa -hour, with the starting (ceiling) price announced in advance. The capacity to be built by each supplier, as speciï¬?ed in the bid, becomes part of the contract for the winning bidders. Each winning bidder gets the off take price at the level that was bid.7 The procurement procedure used in India for CST is essentially the same—that is, an auction for a certain aggregate CST plant capacity to be built by several winning bidders. Tendering procedures and auctions have worked well in many cases in developed markets (such as in Europe), at least to kick-start the market. One of the system’s draw- backs, however, is that if competition is too strong, the prices offered are sometimes Concentrating Solar Power in Developing Countries 27 very low and thus pose a risk of projects not being implemented. By contrast, it has the advantage of fast deployment to kick-start the market in a speciï¬?c technology sector. However, it is not well suited for a large and rapidly growing market because of its high administrative costs, the risk of unrealistic bids and the potential for creating adminis- trative barriers (World Bank/ESMAP 2010). It is too early to evaluate the effectiveness of the incentive scheme in terms of its ability to a ract the investment capital to the most promising locations, and select projects and companies most likely to deliver results. In both the PV and CST tenders, new entrants dominated the list of successful candidates. Many established players have been unable to win. This may be a good result if the new entrants can deliver, thus becoming estab- lished players themselves and making the solar thermal industry more competitive. By contrast, if the new entrants fail to fulï¬?ll their contractual obligations, the effectiveness of the process will be questioned for its failure to accommodate the established players at a higher off take price. It is clear that some new entrants may not even be able to secure the needed loans, whereas established players would have an advantage because of their balance-sheet strength. A survey of 25 potential CST project developers in a World Bank– commissioned study showed that many of the interviewed developers felt that in the PPAs concluded with NVVN, the buyer would not be “bankableâ€?—(that is, ï¬?nancial closure would be unlikely)—unless the PPAs are guaranteed by the GOI, or backed by some other dedicated source of funds. In their view, the banks might not be convinced that the PPA alone is a bankable source of revenue (World Bank/ESMAP 2010). The comparison of the incentives under the JNNSM in regard to those available at the state level may require further analysis. As noted earlier, the GOI has offered the state-level developers the option of switching (“migratingâ€?) to the JNNSM. However, relatively few developers have taken this opportunity, and only 16 projects with a total capacity of 84 MW (of which 30 MW is CST) have migrated. It is important to note that the state-level schemes, such as the one in Gujarat, do not involve competitive bidding. Thus, developers and investors might have felt that the competitive bidding (the reverse auction) under the JNNSM might eliminate the initial price advantage while at the state level, procurement is of the type “what you see is what you get.â€? Secondly, the pro- cess of switching to the JNNSM was competitive as well, and the time window for such migration was rather short. Concerns have also been expressed on the bundling scheme introduced under the JNNSM. First of all, this is fundamentally a cross-subsidy scheme with its inherent eco- nomic distortions. Secondly, the cost of bundled (solar plus coal or nuclear) power is still above the average system cost. At INR 5–6/kWh, while much more affordable than “pureâ€? CST power costing three times as much as an average wholesale rate, as such this cost may still be a challenge for the distribution utilities. Many of the state distribution utilities are in a poor ï¬?nancial state to begin with (World Bank/ESMAP 2010). The difference between this cost and the average cost of conventional power (about INR 2.5/kWh) must be cov- ered either by the rate payers, or through an incremental cost recovery mechanism, which, however, does not seem to be explicitly funded. India Incentive Conclusions The GOI has made a strategic choice to promote grid-connected solar power, and the introduced incentive package is impressive. India has a vibrant economy, and has a good chance to emerge as a major player in the CST industry. 28 A World Bank Study India’s policy on CST is designed to be largely private sector–driven, with the gov- ernment creating an enabling environment for investors. For all the concerns on the guidelines, developers still see success in the early bidding stages as important for stra- tegic positioning in the market. This may explain why the ï¬?rst round of bidding for CST under Phase 1 of the JNNSM was oversubscribed. However, it remains to be seen how effective the whole package of incentives will be. Over the longer term, it needs to be well integrated and coherent—in terms of the instruments (the current process mixes RPO and FiT elements), as well as coordination between state and central governments. Given a great degree of uncertainty about the required (or “justiï¬?edâ€?) level of capi- tal costs for CST projects in India, the quantity-based approach may be a good choice. An RPO scheme may not be as aggressive a strategy as a FiT in securing a massive expansion of solar power capacity, but it facilitates the price discovery process be er than a FiT system. This may result in substantial cost savings both for the public sector and for the ï¬?nal consumer. At the same time, the support schemes available at the state level (notably, in Gujarat) have demonstrated the effectiveness of ï¬?xed FiTs (rather than tariff-se ing schemes involving competitive bidding) in a racting private investors into PPAs. Overall, the effectiveness of the incentives for solar power development is still to be demonstrated by ï¬?nancial closures for concluded PPAs. South Africa’s Incentive Schemes The 2003 White Paper on Renewable Energy (Departments of Minerals and Energy Republic of South Africa 2003) set a target of 10,000 GWh, to be produced from biomass, wind, solar, and small-scale hydro by 2013. The South African Department of Energy, in consultation with the National Energy Regulator of South Africa (NERSA) and Eskom, the national utility, developed a plan for capacity additions called the Integrated Resource Plan 1 (IRP1), which was signed by the Department of Energy on December 16, 2009. IRP1 laid out additional capacity that is required to reach the objective of 10,000 GWh of renewable by 2013 (Department of Energy 2009). A draft version of the new Integrated Resource Plan, named IRP2010, was published in October 2010. It details the plan for capacity additions for the next 20 years in South Africa (Integrated Resource Plan for Electricity 2010). The plan included 1,025 MW from wind, CST, landï¬?ll, and small hydro, supported by the renewable energy feed-in-tariff (REFIT). In March 2011, the ï¬?nal version of IRP2010 was approved by the cabinet, specify- ing that over the next 20 years, 17.8 GW should come from renewable sources (Engineer- ing News 2011). Speciï¬?cally, 1 GW of CST, 8.4 GW of solar PV, and 8.4 GW of wind are expected to be added between 2010 and 2030 (Integrated Resource Plan for Electricity 2010–2030, 2011). The contribution of renewables supported by the REFIT was similar to the draft, although an additional requirement of a solar program of 100 MW each year from 2016 to 2019 was added. Feed-in Tariff In March 2009, NERSA announced Phase I of the REFIT. Similar to standard FiTs, the REFIT requires Eskom, the national utility, to buy electricity from eligible generating units at a tariff set by NERSA that can be passed on to the rate payers. As part of the REFIT phase I, on March 31, 2009, NERSA set the REFIT tariff for parabolic trough plants with 6 hours’ storage per day at ZAR 2.1/kWh, which is equivalent to approximately US30¢/kWh, assuming an exchange rate of ZAR 7 to the U.S. dollar (NERSA 2009b). On Concentrating Solar Power in Developing Countries 29 November 2, 2009, NERSA announced Phase II of the REFIT, expanding eligibility for more technologies under the policy. The announcement added two further tariffs for CST at ZAR 3.14/kWh (US45¢/kWh) for parabolic trough without storage, and ZAR 2.31/kWh (US33¢/kWh) for power tower with 6 hours’ worth of storage per day (NERSA 2009a). Fossil backup for CST is permi ed, but must be limited to 15 percent of the total primary energy input. Eskom’s Single Buyer Office acts as the Renewable Energy Power Purchase Agency (REPA) and, as such, is obliged to buy power through PPAs regulated by NERSA. The tariff was based on LCOE calculations, and will be reviewed annually for the ï¬?rst ï¬?ve years after implementation, which will begin once all conditions of the REFIT and the ï¬?nal regulatory structure are ï¬?nalized, and then every three years thereafter. At the time of writing, NERSA was still in discussions with the Department of Energy, the National Treasury, the Department of Public Enterprises, the Department of Environ- mental Affairs, and Eskom to ï¬?nalize the PPA rules that will govern the operation of the REFIT. NERSA has already published Regulatory Guidelines, a draft PPA, and rules on selection criteria for projects under the REFIT. On September 30, 2010, the Department of Energy announced the start of the procurement process and the government’s intentions to ensure an investor-friendly enabling environment by developing a set of standardized procurement documentation for the PPA. The Department of Energy also announced an official Request for Information (RFI) aimed at potential private power developers to gain understanding on the progress of their projects under the REFIT. The RFI was intended as a “market soundingâ€? to obtain information on projects that will be ready and able to add capacity (MW) and energy to the system before March 2016 (Department of Energy 2010b). The Department of Energy stated that before the procurement documentation is ï¬?nalized and released, a “ministerial determinationâ€? regarding the buyer under the REFIT, as given in the Electricity Regulation Act, would be undertaken ï¬?rst (Aphane 2010). The RFI received 384 responses, identifying a total of approximately 20 GW of REFIT technologies, although less than 30 had received an indicative quote and a preliminary timeframe for connection (Department of Energy 2010a). In March 2011, the cabinet approved the Independent System and Market Operator Bill for tabling in parliament, which is intended to ensure that IPPs are included in the addition of new generation capacity in South Africa, rather than just from Eskom. Although this is not a bill exclu- sively for IPPs under the REFIT, its purpose is to promote the role of IPPs that are the entities that will beneï¬?t from the REFIT once it gets underway. The IRP2010 resolves the uncertainties around long-term capacity addition targets, and includes the recommendation to ï¬?nalize the REFIT process as quickly as possible. Although the PPA process is still being ï¬?nalized, Eskom claims to have received 156 appli- cations from IPPs already, representing a combined total capacity of 15,154 MW, 13,252MW of which is wind (Van de Merwe 2010). This leaves 1,902 MW of different technologies under the REFIT, which include the three CST technologies, namely trough, power tower, and power tower with storage, and also solar PV, solid biomass, biogas, land-ï¬?ll gas, and small hydro, among which the distribution of applications is as yet unannounced. The RFI shed light on the breakdown of potential IPP projects, to be supported by the REFIT and broken down by technology. Of the 384 RFI responses, one-third were wind projects, one-third were solar PV projects, and 5 percent of responses with 10 percent of capacity came from CST projects. The remainder consisted of biomass, hydro, landï¬?ll gas and biogas, and cogeneration. 30 A World Bank Study Aside from the REFIT, US$350 million of the US$500 million CTF investment plan for South Africa has been awarded to Eskom to develop wind and CST projects. The IBRD and AfDB are also proposing loans each of US$260 million to further co-ï¬?nance the projects. Combining the CTF, IBRD, AfDB contributions with those from other bilat- eral and commercial lenders, the project’s total budget is US$1.228 billion. The CST component is estimated to require US$783 million, while the wind component will cost US$445 million. The CST project will be located in Upington in the Northern Cape Province, where Direct Normal Insolation (DNI) is approximately 2,800kWh/m2 per year, one of the highest levels of solar potential in the world. Eskom has indicated that the preferable technology is power tower with storage, although the decision on the technology to be used has yet to be ï¬?nalized. SOUTH AFRICA INCENTIVE ISSUES The REFIT program is not yet fully established as the procurement process remains under discussion. As a result, concerns have been raised concerning REFIT’s effectiveness in encouraging investments in CST and other renewables. The issues raised include whether the targeted goal of 10,000 GWh from renewable sources in 2013 acts as a capacity “capâ€? of PPAs eligible for the REFIT, whether NERSA will assess the eligibility criteria for projects, and whether Eskom’s Single Buyer Office can process all applications effi- ciently. In addition, the question remains whether NERSA’s proposed tariffs are high enough to induce investment (Bukala 2009). In March 2011, one week after the government passed IRP2010, which speciï¬?ed that 17,000 MW should come from renewable energy, NERSA announced a review of the REFIT tariffs and proposed that they should be cut. The announcement of high renew- able energy targets, combined with the cut in tariffs that are in place to reach this tar- get, could be interpreted as somewhat conflicting, since lower tariffs could a ract fewer renewable project developers. Parabolic trough with storage faces a cut of 41.5 percent, which is one of the largest cuts of all REFIT tariffs. The paper also speciï¬?es that the tariff for power tower technology should be reduced by 39.4 percent, and CST trough without storage should fall by 7.3 percent (NERSA 2009b). NERSA predicts that the tariff review procedure will be completed by the end of May 2011, when the ï¬?nal approved tariffs will replace the original ï¬?gures developed in Phases I and II. The discussion over changing the tariffs is likely to further delay the awarding of PPAs as IPPs as project developers wait for the ï¬?nal announcement and plan investments accordingly. One goal of the Upington CST project, funded with support of the MDBs, is to resolve some uncertainties over cost and risk, thereby encouraging IPPs to enter into PPAs under the REFIT. It is believed that the general visibility of CST will rise with the national utility running a large-scale CST project, signaling that the government is com- mi ed to a future with renewable energy technologies. Without Eskom’s participation and a visibly successful large-scale project, the private sector is unlikely to make signiï¬?- cant investments to allow for rapid diffusion of CST technology in South Africa. SOUTH AFRICA INCENTIVE CONCLUSIONS Since the REFIT is not yet operational in South Africa, it is premature to predict how successful it will be in encouraging investments in CST, and the other energy tech- nologies it covers. There are concerns over the lack of a deï¬?ned structure of the REFIT, and uncertainty over what the ï¬?nal tariffs will be. However, many of these concerns could be addressed once NERSA and Eskom ï¬?nalize the process for arranging the Concentrating Solar Power in Developing Countries 31 PPAs, tariff levels are decided, and the role of the single buyer as Eskom or an indepen- dent third party is determined. During the consultation processes of se ing the tariffs, NERSA received a signiï¬?cant number of comments, demonstrating the sensitivity of the process and the importance of the outcomes for stakeholders. It is conceivable that the REFIT may encourage more investment for certain technologies than for others. In the same way that an RPS scheme induces investments predominantly in the cheapest tech- nology, the REFIT may only promote signiï¬?cant investments in more established and less risky technologies, such as wind power, rather than CST. The fact that the vast majority of applications, which Eskom has received so far, have been for wind projects could indicate the disparity in effectiveness of the policy across different technologies. Notes 1. The capacity of CST projects supported under NSM is speciï¬?ed as between 5 MW and 100 MW. 2. The methodology for arriving at the tariff level of INR 15.31/kWh involves assumptions, such as the normative CAPEX of INR 153 million/MW (about US$3.4 million/MW), a project life of 25 years, a debt-to-equity ratio of 70:30 with debt of 10-year maturity available at 12 percent, and a capacity utilization factor of 23 percent. No thermal storage is assumed. 3. NSM documents stipulate that for each megawa of solar capacity signed by NVVN, an equiva- lent megawa of capacity from the unallocated quota of NTPC stations shall be allocated. Hence, during the ï¬?rst phase, 1 GW of solar capacity will be coupled with 1 GW of NTPC coal plants. However, the amounts of electricity produced by coal plants may be four times as much as that coming from solar plants, because of a much higher plant load factor. 4. According to EVI 2011, 66 bids were received. 5. Generation-based incentives (GBIs) have been introduced by MNRE, in a scheme separate from the JNNSM, ï¬?rst for wind and then in January 2008 for grid-connected solar power, including CST. Under this scheme, the ministry would provide an incentive of a maximum of INR 12/kWh for PV and INR 10/kWh for CST. The maximum amount of incentive applicable for a project would be determined after deducting the power purchase rate for which a PPA has been signed by the utility with a project developer from a notional amount of Rs. 13/kWh. This incentive would be provided to project developers at a ï¬?xed rate for a period of 10 years, but the maximum amount of GBI offered for new plants would be decreasing over time. The scheme was designed mainly to support smaller entrepreneurs with a total proposed plant capacity of 5 MW or less. 6. By adopting RECs as a mechanism supplementary to RPOs, the Indian system adopts another feature typical of the schemes in the United States and United Kingdom. 7. A recent report on auctions (World Bank/ESMAP 2011a) classiï¬?es such auctions as “pay-as-bidâ€? or “discriminatoryâ€? auctions. This is a form of a sealed-bid auction in which each bidder submits a schedule of prices and quantities (that is, a supply function). The auctioneer gathers together all the bids, creating an aggregate supply curve, and matches it with the quantity to be procured. The clearing price is determined when supply equals demand. The winners are all bidders whose bids, or sections of their bids, offered lower prices than the clearing price. The winners receive different prices based on their ï¬?nancial offers. The auctions for electricity contracts carried out in Panama and Peru have used a pay-as-bid design. Mexico also uses a pay-as-bid design for its auctions for PPAs. PART III Financing CST—How to Bring Technology Costs Down CHAPTER 5 Cost Drivers and Cost Reduction Potential D ifferent CST technologies have, at present, reached varying degrees of commercial availability. While commercial cost data exist for parabolic trough, and to a slightly lesser degree for power tower, such cost data has yet to be established for the Fresnel and Dish Stirling technologies. Under these circumstances, a thorough assessment of the main cost drivers and the cost reduction potential will be key when considering the economic viability of CST in general and different CST technologies in particular. Based on an assessment of LCOEs for different CST technologies in some of the main emerg- ing markets for CST—India, Morocco, and South Africa—and a review of typical cost structures for parabolic trough and power tower plants derived from projects developed or under preparation in developed markets, this chapter provides (a) an assessment of the main cost drivers, (b) an affordability assessment of different regulatory and ï¬?nan- cial incentives used to lower LCOEs in various emerging market conditions, and (c) an economic analysis of reference CST plants in the main emerging markets for CST that are considered. LCOEs for CST in Speciï¬?c Developing Country Markets A common way to assess the ï¬?nancial cost of a particular power technology and/or com- pare the ï¬?nancial cost of alternative technologies is to express the cost of producing elec- tricity for a certain plant as the LCOE (see box 5.1). The la er allows se ing all the costs incurred by a particular plant over its lifetime (ï¬?xed capital cost elements, as well as variable O&M cost elements) in relation to the value of total electricity produced over its lifetime. LCOE is usually highly sensitive to changes in the underlying variables. There- fore, future variations of any of the cost elements for CST might well have an impact on the actual CST technology-speciï¬?c LCOEs. A detailed ï¬?nancial LCOE analysis was conducted for some of the major emerging markets for CST—India, Morocco, and South Africa—comparing parabolic trough and power tower technologies. The assumptions used in the analysis are listed in table B.11 in Appendix B. The results of the analysis are shown in ï¬?gure 5.1. The analysis was based on a set of assumptions regarding the economic parameters (for example, interest rate and inflation), and the technical conditions prevalent in each country. Although LCOEs for CST are highly sensitive to the site-speciï¬?c solar resource, DNI, there is no clear pat- tern of the sensitivity to the DNI resources available for analysis1 because of widely dif- fering ï¬?nancial conditions in each scenario considered. Generally however—under the assumption that the optimal amount of storage (the amount of storage which minimizes LCOE for each plant) is available—power tower technology offers lower LCOEs com- pared to parabolic trough in all three scenarios. Notwithstanding the lack of compre- hensive data for power tower plants with the amount of storage assumed here (because 35 36 A World Bank Study Box 5.1: LCOE structure LCOE generally represents the cost of generating electricity for a particular plant or system. The concept is basically a ï¬?nancial assessment of all the accumulated costs of the plant over its life cycle relative to the total energy produced over its life cycle. More speciï¬?cally, LCOE is a ï¬?nancial annuity for the capital amortization expenses, including ï¬?xed capital costs (for ex- ample, equipment, real estate purchase, and lease) and variable O&M expenses (for thermal plants mostly consisting of fuel expenses and O&M expenses, for CST plants mostly of O&M expenses), taking into account the depreciation and the interest rate over the plant’s life cycle, divided by the annual output of the plant adjusted by the discount rate. If the discount rate is assumed to be equal to the rate of return LCOEs reflect the price that would have to be paid to investors to cover all expenses incurred (for example, capital and O&M) and hence the mini- mum cost recovery rate at which output would have to be sold to break even (Kearney 2010): N ∑I + M t− 1 (1+ r) t LCOE = N Et ∑ (1+ r) t t− 1 where r = discount rate â?? N = the life cycle of the plant â?? t = year â?? It = Investment costs in year t â?? Mt = O&M costs in year t â?? Et = electricity generation in year t of a limited number of these plants having been constructed so far—see Chapter 2), the lower LCOEs for power tower are mainly because of certain technical advantages, like for example, the ability to reach higher operating temperatures and higher operating rates (for more information see Chapter 2). Overview of the Cost Structure Internal cost structures of CST projects are often not readily available. However, exam- ples for potential cost breakdowns with regard to total CAPEX and operational expen- Figure 5.1: LCOEs for parabolic trough and power tower in India, Morocco, and South Africa 60 US$ Cents/kWh 40 20 0 India Morocco South Africa Parabolic trough (air-cooled) Parabolic trough (wet-cooled) Power tower (air-cooled) Power tower (wet-cooled) Source: Authors’ data, using Solar Advisor Model (NREL). Concentrating Solar Power in Developing Countries 37 ditures (OPEX) for reference parabolic trough and power tower plants with 100MW and 50 MW capacity, and different amounts of Thermal Electricity Storage (TES), could be presented as in tables 5.1–5.4 and ï¬?gures 5.2 and 5.3. Assessment of the Cost Drivers for CST The cost elements listed in table 5.5, which comprise the typical cost structure of a CST project, are influenced by a variety of cost drivers, including the production and compe- tition related issues, available ï¬?nancing conditions, changes in the underlying prices for key input commodities, and for land and labor inputs. Their respective impact has been assessed accordingly. Local Inputs: Changes in Land and Labor Prices Land-related expenses for a plant can account for a considerable share of the overall investment costs for most CST technologies. The actual share, however, will depend on land availability, ownership, and taxation issues. The second major issue will be the actual amount and price of local labor, relative to the total labor inputs needed to build and maintain the plant. The actual price of labor will obviously depend on local labor market conditions, but in nearly all cases and for nearly all parts of the value chain (project development; components; engineering, procurement, and construction (EPC); and O&M), will be lower in emerging market conditions. The share of local labor inputs partly depends on the chosen technology, the degree to which local services can be employed in different stages of the project value chain and on the degree of local manu- facturing of the CST component. A detailed assessment of the potential of local manufac- turing potential to reduce CST investment costs in several emerging markets is provided Table 5.1: Estimate of capital expenditures—parabolic trough Option parabolic trough 100 MWe 50 MWe Item Unit TES 4.5 h TES 9.0 h TES 13.4 h TES 9.0 h Nominal plant size Exchange rate Euro/US$ 1.40 1.40 1.40 1.40 Rated electric power, gross MWe 100 100 100 50 EPC Contract Costs mln US$ 704.2 721.1 872.7 388.8 Solar ï¬?eld mln US$ 323.6 284.4 334.2 142.5 HTF system mln US$ 68.1 59.9 70.3 30.0 Thermal energy storage mln US$ 62.7 123.6 184.4 62.7 Power block mln US$ 107.7 107.7 107.7 67.3 Balance of plant mln US$ 45.0 46.0 55.7 24.2 Engineering mln US$ 36.4 37.3 45.1 29.4 Contingencies mln US$ 60.7 62.2 75.2 32.7 Owners’ costs mln US$ 33.4 34.2 41.4 21.6 CAPEX grand total (± 20%) mln US$ 737.6 755.3 914.1 410.4 Speciï¬?c CAPEX US$ / kW 7,376 7,553 9,141 8,207 Source: Fichtner 2010. 38 A World Bank Study Table 5.2: Estimate of capital expenditures—reference power tower Option central receiver 100 MWe 50 MWe Item Unit TES 9.0 h TES 12.0 h TES 15.0 h TES 15.0 h Nominal plant size Exchange rate Euro/US$ 1.40 1.40 1.40 1.40 Rated electric power, MWe 100 100 100 50 gross EPC contract costs mln US$ 679.7 798.0 926.7 501.0 Site preparation mln US$ 27.0 33.0 42.4 19.9 Heliostat ï¬?eld mln US$ 218.3 267.6 323.3 165.4 Receiver system mln US$ 106.4 125.8 144.3 85.8 Tower mln US$ 15.0 15.0 15.0 8.8 Thermal energy storage mln US$ 58.7 77.1 95.3 49.3 Power block mln US$ 110.0 110.0 110.0 65.4 Balance of plant mln US$ 40.7 47.6 55.0 30.0 EPC contractors mln US$ 46.1 54.1 62.8 34.0 engineering Contingencies mln US$ 57.6 67.6 78.5 42.5 Owners’ costs mln US$ 37.4 43.9 51.0 27.6 CAPEX grand total (± 20%) mln US$ 717.1 841.9 977.7 528.6 Speciï¬?c CAPEX US$/kWT 7,171 8,419 9,777 10,572 Source: Fichtner 2010. in Chapter 6. Current local content sensitivities and local staffing demand for a reference 100 MW parabolic trough plants in the Middle East and North Africa region (MENA) are given in table 5.6. Changes in Underlying Commodity Prices As in most energy industries, CST’s cost structure depends, to a certain degree, on price fluctuations of the underlying nonfuel commodity inputs. The impact of price fluctua- tions of these commodities on the actual cost structure is partly determined by both the respective CST technology’s commodity needs and the degree to which commodities can be supplied locally. Concrete and steel for all Spanish plants and for El-Kureimat plant in Egypt were, for example, supplied locally, resulting in lower investment costs. Commodities used for CST components include steel, concrete, sand, glass, plastic, and a variety of different metals, such as silver, brass, copper, or aluminum, as well as nitrates or molten salts for storage systems and a variety of other chemicals. Several input commodities—such as steel or concrete—are difficult to substitute for. Sharp price movements for these commodities can lead to potential fluctuations in the ï¬?nal costs of plant components and/or O&M expenses. Economies of Scale and Volume Production Mass production of components would most likely make CST technologies more eco- nomically viable because of the high standardization potential of several components, Concentrating Solar Power in Developing Countries 39 Table 5.3: Estimate of operational expenditures—reference parabolic trough Option parabolic trough 100 MWe 50 MWe Item Unit TES 4.5 h TES 9.0 h TES 13.4 h TES 9.0 h Technical-ï¬?nancial constraints Exchange rate EURO/US$ 1.4 1.4 1.4 1.4 Power generation GWh/a 441.1 492.4 583.8 237.2 Number of operating staff — 60 60 75 45 Manpower cost (average) 1000 $/a 58.8 58.8 58.8 58.8 Price diesel fuel $/liter 1.1 1.1 1.1 1.1 Fuel consumption 1000 Liter/a 200 200 200 120 Raw water US$/m3 0.70 0.70 0.70 0.70 Annual raw water 1000* m3/a 132,330 147,720 175,140 71,160 consumption HTF Consumption t/a 61 54 64 26 HTF price US$/t 3,000 3,000 3,000 3,000 Annual OPEX (costs as of 2009) Fixed O&M Costs: mln US$ 13.4 13.6 16.5 8.0 Solar ï¬?eld & storage mln US$ 4.5 4.7 5.9 2.4 system Power block mln US$ 2.3 2.3 2.5 1.4 Personnel mln US$ 3.5 3.5 4.4 2.6 Insurance mln US$ 3.0 3.1 3.8 1.6 Variable O&M Costs mln US$ 1.2 1.2 1.4 0.6 (Consumables): Fuel mln US$ 0.2 0.2 0.2 0.1 Water mln US$ 0.1 0.1 0.1 0.0 HTF mln US$ 0.2 0.2 0.2 0.1 Other consumables & mln US$ 0.7 0.7 0.9 0.4 residuesa Total OPEX mln US$ 14.6 14.9 17.9 8.6 Percentage of CAPEX % 1.97% 1.97% 1.96% 2.10% Source: Fichtner 2010. a. Electricity import, HTF, nitrogen, chemicals. including most of the reflecting devices.2 However, different cost reduction mechanisms will most likely apply to each component. In the case of parabolic trough and Fresnel, receiver costs will depend largely on the size scale-up, production volume, and increased competition, which could result in a 45 percent cost reduction by 2025 (Kearney 2010). The cost reduction of reflectors will largely depend on alternative or new material com- positions and production methods for mirrors, with overall prices expected to come down by 20 percent until 2020 for parabolic trough and 25 percent until 2025 for power tower and Fresnel (Kearney 2010). Considering general experience curve concepts and progress ratios quantifying the effect of cost decrease for increased production and expe- rience, a range of the cost scale-down from 5 percent to 40 percent can potentially be expected, according to different estimates (Kearney 2010). 40 A World Bank Study Table 5.4: Estimate of operational expenditures—reference power tower Option central receiver 100 MWe 50 MWe Item Unit TES 9.0 h TES 12.0 h TES 15.0 h TES 15.0 h Technical-ï¬?nancial constraints Exchange rate EURO/US$ 1.4 1.4 1.4 1.4 Power generation (net) GWh/a 430.8 538.3 629.6 315.5 Number of operating staff - 60 68 77 52 Manpower cost (average) 1000 $/a 59 59 59 59 Price diesel fuel $/liter 1.1 1.1 1.1 1.1 Fuel consumption 1000 Liter/a 300 300 300 150 Raw water US$/m3 0.7 0.7 0.7 0.7 Annual raw water 1000* m3/a 116,323 145,340 169,982 85,183 consumption Annual OPEX (costs as of 2009) Fixed O&M Costs: mln US$ 12.29 14.19 16.24 9.47 Solar ï¬?eld & storage mln US$ 3.83 4.71 5.63 3.00 system Power block mln US$ 2.26 2.37 2.48 1.43 Personnel mln US$ 3.53 3.98 4.50 3.06 Insurance mln US$ 2.67 3.14 3.64 1.98 Variable O&M Costs mln US$ 1.32 1.57 1.78 0.89 (Consumables) Fuel mln US$ 0.34 0.34 0.34 0.17 Water mln US$ 0.08 0.10 0.12 0.06 Other consumables & mln US$ 0.90 1.13 1.32 0.66 residuesa Total OPEX mln US$ 13.6 15.8 18.0 10.4 In percent of CAPEX % 1.90% 1.87% 1.84% 1.96% Source: Fichtner 2010. a. Electricity import, HTF, nitrogen, chemicals. A potentially important side effect would be that, unlike most components for fos- sil fuel plants that require skilled labor, mass-manufactured CST components could be designed to minimize the need for highly skilled labor for assembly, and hence open the opportunity for local manufacturing in several emerging markets, providing an oppor- tunity for further potential cost decreases (Shinnar and Citro 2007). While the basic val- ues are provided in table 5.7, a more detailed discussion on cost reduction potential in several emerging markets is provided in Chapter 6. Monopoly Rents and Supply Chain Bottlenecks for CST Components Monopolistic or oligopolistic market situations, especially in terms of the supply of criti- cal, CST-speciï¬?c components, might cause the respective components to be overpriced, thereby negatively affecting the overall investment costs and hence the CST-speciï¬?c Concentrating Solar Power in Developing Countries 41 Figure 5.2: CAPEX breakdown—parabolic trough (100 MW—13.4 h TES—US$914 m) Owner's cost, 5% Contingencies, 8% Engineering, 7% Balance of Solar field, 35% plant, 6% Power block, 17% HTF system, 7% Thermal energy Storage, 15% Source: Fichtner 2010. Figure 5.3: CAPEX breakdown—power tower (100 MW—15 h TES—US$978 m) Owner's cost, 5% Site Contingencies, preparation, 4% 8% EPC contractors engineering, 6% Balance of plant, 6% Heliostat field, 33% Power block, 11% Receiver system, 15% Thermal energy storage, 10% Tower, 2% Source: Fichtner 2010. 42 A World Bank Study Table 5.5: Overview of cost elements and cost drivers Cost elements Cost drivers Cost of land • Space availability and cost • Taxation issues • Financing conditions available Cost of solar ï¬?eld • Cost of commodities • Monopoly/oligopoly rents • Economies of scale in production • Financing conditions available • Market demand Cost of power block • Cost of commodities • Financing conditions available • Market demand Transmission connection cost • Regulation • Distance from load centers • Technology • Financing conditions available Storage • Cost of commodities • Monopoly/oligopoly rents • Economies of scale in production • Financing conditions available O&M costs • Local content sensitivities • Local labor costs • Water availability and cost Source: Authors’ data. Table 5.6: Local content sensitivities—Middle East and North Africa case study Local stafï¬?ng demand (person years/ Local content (%) Foreign share (%) 1,760 hrs/yr) Project development 0–10% 90–100% 6–20 Engineering planning 30–50% 50–70% 75–95 Technology (procurement) 30–60% 40–70% 145–220 Construction and site 100% 0% 320 improvement Operations and 90–100% 0–10% 40–45 maintenance Source: Kearney 2010. Concentrating Solar Power in Developing Countries 43 Table 5.7: Cost reduction potential of economies of scale/volume production Component Reduction potential Cost drivers Receivers 45% by 2025 (for parabolic trough and Fresnel) • Size scale-up • Production volume • Increased competition Reflectors 20% until 2020 (for parabolic trough) • New material compositions 25% until 2025 (for power tower and Fresnel) • Production methods Source: Kearney 2010. LCOEs. Such an inflated cost proï¬?le might seriously slow the development of the tech- nology in general and in particular in an emerging market se ing. This is because the more specialized and technically challenging the respective component is, the fewer the number of qualiï¬?ed competitors. For example, there are very few companies special- izing in production of receiver tubes for parabolic trough, and Fresnel (Scho Solar and Siemens—formerly Solel) basically share the market and have relatively high earnings before interest and taxes (EBIT) margins of around 20–25 percent (Ernst & Young and Fraunhofer Institute 2010) or in supplying heat storage systems, thermal oils and central control systems. Also, as CST technologies are reaching a higher degree of commercial- ization, market consolidation has already taken place and is expected to progress. This would reduce the number of players in each segment of the value chain even further. With regard to developers, the ï¬?rst consolidation round has already taken place as large integrated infrastructure companies started buying up smaller start-ups to get access to their respective technologies. For example, Areva had bought Ausra (now Areva Solar), Siemens had acquired Solel Solar, Acciona had secured a majority share in Solargenix, and Alstom has a strategic relationship with BrightSource Energy. Financing Conditions Available The availability and type of ï¬?nancing for CST as for any other major energy installment will depend on the following: (a) the technology-speciï¬?c overall capital requirements; (b) the perceived performance risk by investors and lenders, which in turn will depend on available performance data, the ï¬?nancial position of developers and the provision of performance assurance by developers; (c) the creditworthiness of the off taker; and (d) the regulatory and ï¬?nancial framework of the respective jurisdiction. The la er will not only determine the applicable taxation rates, but also the availability, viability, and predictability of any ï¬?nancial incentive provided, whether in the form of a FiT or the different incentives provided under an RPS regime. How these incentives are designed will have a considerable influence on the availability of ï¬?nancing as a properly designed regulatory framework can help mitigate risks and increase considerably investment for developers. Technical and Scale-Related Cost Reduction Potential Component-Speciï¬?c Cost Reduction Potential Detailed component-speciï¬?c cost reduction potentials for each CST technology are given in tables A.7–A.10 in Appendix A. These estimates are based on a detailed assessment of 44 A World Bank Study the respective cost drivers for each component and the underlying situation in the respec- tive industries producing these components (YES/Nixus/CENER 2010). In summary, parabolic trough components showing the most potential for cost reduction include the reflectors (18–22 percent), reflector mounting structures (25–30 percent), receivers (15–20 percent), the heat transfer system (15–25 percent), and molten salt system (20 per- cent). Power tower system components showing the most cost reduction potential are the reflector mounting structures (17–20 percent), heat transfer system (15–25 percent) and molten salts (20 percent). Linear Fresnel system components showing the most cost reduction potential are the reflector mounting structures (25–35 percent) and receivers (15–25 percent), while for the Dish Stirling engine, it is the reflectors (35–40 percent) and reflector mounting structures (25–28 percent). Technology-Speciï¬?c LCOE Cost Reduction Potential Based on these cost reduction potentials for individual components, the overall cost reduction potential for each CST technology is described in ï¬?gure 5.4. The respective reduction potential was assessed through the modeling of reference plants, whereby cal- culations were performed without accounting for any costs related to the connection to the transmission system, costs related to the purchase of land or the use of water. A com- prehensive picture of the actual cost reduction potential in each case emerges through the assessment of the cost reduction potential of all components for a speciï¬?c technology provided in table B.7–B.10 in Appendix B. Figure 5.4: Cost reduction potential for CST technologies 30 LCOE in $/kWH 25 20 15 10 2010 2015 2020 Parabolic trough Power tower Fresnel Dish stirling Source: YES/Nixus/CENER 2010. Note: Numbers converted at EX US$1.35/Euro, based on averages of LCOE percentage cost reduction by 2015 and 2020. Overall LCOE Cost Reduction Potential A. T. Kearney (2010) performed a slightly different cost reduction potential evaluation on the basis of initial investment cost and performance data for a series of seven different ref- erence plants spanning all CST technologies available, with the aim of calculating LCOE as the minimum required tariff necessary to ensure coverage of project ï¬?nancing, based Concentrating Solar Power in Developing Countries 45 on a 25-year plant runtime. This calculation took ï¬?nancing prerequisites (such as a typical debt service coverage ratio (DSCR) of 1.4) into account to derive cost reduction potentials for respective minimum required tariff CST-based output needed to repay debt, earn an adequate return on invested capital, and secure long-term ï¬?nancing. Figure 5.5 shows upper and lower estimates for LCOE reductions until 2025. The respective cost reduction projections can also be used to evaluate CST’s future position within the overall supply mix (ï¬?gure 5.5). In the best case scenario, CST might, for example, in the long term be able to substitute CCGT and potentially other fossil fuel–based plants as a peak to mid-load provider, depending on future fossil fuel prices. The hybridization of CST and the intro- duction of a carbon price could increase the likelihood of such a replacement. Figure 5.5: LCOE reduction potential for CST 0.35 0.3 0.25 LCOE in $/kWH 0.2 0.15 0.1 0.05 0 2012 2015 2020 2025 Lower limit Average Source: Kearney 2010. Financial Sustainability Assessment of Financial and Regulatory Incentives In the near to midterm, well-tailored and appropriately designed regulatory and ï¬?nan- cial incentives will not only be necessary to ensure a particular project’s ï¬?nancial via- bility, but most likely remain crucial in order to realize the projected cost reduction trajectories outlined above. Without such incentives, a major rollout of the technology seems uncertain or would most likely be delayed, which could alter the cost reduction trajectories considerably. By contrast, regulatory and ï¬?nancial incentives always entail a societal cost, either in terms of a ï¬?scal expenditure or lost ï¬?scal revenues, or in terms of increased electricity tariffs for consumers, if the cost of incentives is directly passed through to ï¬?nal consumers. Even though these societal costs can be limited by applying recent lessons learned when designing the respective incentive framework—especially with regard to the design of FiTs (see Chapters 3 and 4)—most incentives granted to stimulate investment will still cause a more or less considerable societal cost burden which, depending on the respective jurisdiction, is ultimately to be borne by either the taxpayer or the ï¬?nal 46 A World Bank Study consumer, or both. Limiting the societal cost of incentives is therefore central to ensuring the sustainability of the incentives granted. This is even more crucial under developing country conditions where the overall ï¬?scal position and individual income levels in most cases limit the overall resources that can be allocated to scaling up renewable energies. The following pages entail a basic affordability and sustainability analysis for a variety of regulatory and ï¬?nancial incentives granted in three major emerging markets for CST—India, Morocco, and South Africa3—based on their impact on the LCOEs of 100MW reference plants in these markets. The main aim of this analysis is to ï¬?nd ways of optimizing regulatory and ï¬?nancial incentives in order to minimize both CST generation cost and the societal cost in purely ï¬?nancial terms. The tested incentives range from tax holidays to more favorable depreciation schemes and the use of concessional ï¬?nancing schemes (such as the IBRD, CTF, GEF, donor-supported output-based approach (OBA), and others). The analysis therefore generally aims to (see also table 5.8): ➩ Determine the cost-effectiveness of different regulatory incentives and approaches in terms of their impact on LCOEs and hence their ability to facilitate investments per dollar spent. Assessments were made for parabolic trough and power tower technologies, as well as both wet- and air-cooling methods, although, with the scaling up of CST in most emerg- ing markets, the authors expect the majority of future plants in emerging markets to be air-cooled. All scenarios are based on the optimal amount of thermal electrical storage (TES)4 for each reference plant,5 which is determined by the combination of storage and solar multiple that minimizes LCOEs for parabolic trough and the optimal combination of storage and tower height and receiver dimensions for the power tower systems. Table 5.8: Deï¬?nitions used Impact of a policy instrument Impact of a regulatory incentive or approach on lowering LCOEs and hence facilitating investments Cost-effectiveness of a policy instrument Impact of a regulatory incentive or approach on lowering LCOEs and hence facilitating investments per dollar spent. Societal cost Total additional expenses caused by a particular policy instrument to either the taxpayer and/or the ï¬?nal rate payer. Source: Authors’ deï¬?nitions. Assumptions regarding prevailing capital and O&M costs, as well as macroeco- nomic, ï¬?nancial, and regulatory conditions in both markets, are outlined in table B.11 in Appendix B and were based on a variety of sources: (a) information regarding the actual capital and O&M costs and the ï¬?nancial and regulatory conditions faced in a particular jurisdiction, provided by developers;6 (b) respective applicable regulatory documents in the cases of India and South Africa (CERC 2009a); (c) ï¬?nancial assumptions made for an internal analysis for an IBRD co-ï¬?nanced CST development in the MENA region, for the Moroccan case; and (d) informed assumptions by World Bank staff. The analysis gener- ally assumes nonrecourse ï¬?nancing. Impact Assessment of Different Regulatory Approaches to Lower LCOEs To determine the impact of different regulatory incentives and approaches in terms of their ability to lower LCOEs, and thereby facilitate investments, sensitivity analyses were run for the following incentives under the outlined assumptions: Concentrating Solar Power in Developing Countries 47 â–  Tax holidays/reductions lowering the applicable corporate income tax rate by 50 percent. â–  VAT exemptions lowering the amount of direct cost to which VAT applies from 100 percent to 70 percent. â–  Accelerated depreciation schemes allowing for straight line depreciation over seven years. â–  Concessional loan terms allowing for loan terms of 25 years. â–  Concessional loan rates lowering the applicable debt interest rate by 3 percent, by blending concessional and commercial ï¬?nancing.7 INDIA In the Indian case, the concessional ï¬?nancing terms—especially the concessional loan terms—have a far larger impact on LCOEs than simple tax reductions or exemptions. While relatively substantial tax cuts and exemptions only lower LCOEs by less than a percentage point, more favorable depreciation schemes can lower LCOEs by several percentage points. Concessional schemes, however, have the highest impact, with a 3 percent lower debt interest rate resulting in an approximately 7.3 percent lower LCOE in all four cases. The speciï¬?c impact of each incentive for each technology in terms of their ability to lower LCOEs and facilitate investments is shown graphically in Figure 5.6 and numerically in table B.12 in Appendix B. Figure 5.6: Impact assessment of different regulatory approaches on LCOE in India 38 36 Ceiling tariff 34 32 30 28 26 24 Effective tariff 22 20 Current Tax VAT Accelerated Longer loan Concessional Concessional Accelerated scenario reduction exemption depreciation term financing loan term + depreciation rates + Concessional loan term + Parabolic trough (air-cooled) Parabolic trough (wet-cooled) rates Power tower (air-cooled) Power tower (wet-cooled) Source: Authors’ data, using solar advisor model (NREL). Given the current nominal CERC FiT, only power tower technology would currently pose a ï¬?nancially viable option. However, because of the program’s reverse auction mechanism, the lowest bidding criteria lower the effective FiT available to a minimum of Rs. 10.49, or US$23.3 cents (which was the lowest winning bid in the recently concluded Phase I of the JNNSM). At this level, a modiï¬?cation of the current ï¬?nancial and regulatory 48 A World Bank Study incentive framework would be needed to allow LCOEs to drop under the threshold of the effective FiT level. A combination of concessional loan terms and rates is the single most effective incentive in ensuring that LCOEs—at least for power tower—would drop below the threshold. MOROCCO Under the Moroccan scenario, results are similar (see also ï¬?gure 5.7), as concessional schemes again have a larger impact in terms of lowering LCOEs than simple tax reduc- tions or exemptions. A combination of concessional loan terms and rates would lower LCOEs in all four cases by around 19 percent, whereas tax reductions or exemptions only lower LCOEs by 1–2 percent (see numerical presentation in table B.13 in Appendix B). The important difference, however, is that, opposed to the Indian case, accelerated depreciation proves to have a higher impact on lowering LCOEs in this scenario because of the much higher assumed corporate income tax rate in Morocco (accelerated deprecia- tion creates a large tax shield in the ï¬?rst years of operation, which lowers the NPV of the total amount of taxes paid over the project’s lifetime). Under our assumption of straight- line depreciation over seven years, LCOEs drop by around 14.5 percent in all four cases. Figure 5.7: Impact assessment of different regulatory approaches on LCOE in Morocco 40 35 30 US$ cents/kWh 25 20 15 10 5 0 Current Tax VAT Accelerated Longer loan Concessional Concessional AD + scenario reduction exemption depreciation term financing loan term + concessional rates loan term + rates Parabolic trough (air-cooled) Parabolic trough (wet-cooled) Power tower (air-cooled) Power tower (wet-cooled) Source: Authors’ data, using solar advisor model (NREL). SOUTH AFRICA Regarding South Africa, the same picture as in Morocco was observed (see also ï¬?gure 5.8). In all four cases, the effect of the accelerated depreciation is a 12.5 percent lower LCOE, slightly larger than the one of combined concessional loan terms and rates, whereas again tax reductions or exemptions only have a minor impact on levelized cost (table B.14 in Appendix B). This would be even more important, given the slightly higher capital costs and less favorable ï¬?nancial conditions assumed for South Africa. To allow power tower plants to become ï¬?nancially viable, a tariff of around ZAR 2.5 would be sufficient under the assumptions taken for this analysis. The tariff of ZAR 2.31 Concentrating Solar Power in Developing Countries 49 that would theoretically be available for power tower under phase two of the REFIT is already relatively close to this level, but is only guaranteed for 20 years—shorter than the expected lifetime of the plant. In addition, the REFIT tariff would only allow for power tower plants with up to six hours of storage which, based on this analysis, would not allow for the use of the optimal amount of storage to minimize LCOE for a particular power tower plant in South Africa. The tariff offered for parabolic trough under phase two of the REFIT at ZAR 3.14 seems unlikely to ensure the ï¬?nancial viability of any para- bolic trough plant under the assumed circumstances. Figure 5.8: Impact assessment of different regulatory approaches on LCOE in South Africa 45 REFIT for power tower w/ 6hrs TES 40 35 US$ cents/kWh 30 25 20 15 10 5 0 Current Tax VAT Accelerated Longer loan Concessional Concessional AD + scenario reduction exemption depreciation term financing loan term + concessional rates loan term + rates Parabolic trough (air-cooled) Parabolic trough (wet-cooled) Power tower (air-cooled) Power tower (wet-cooled) Source: Authors’ data, using solar advisor model (NREL). Cost-Effectiveness of Different Regulatory Approaches to Lower LCOEs Ultimately the ï¬?nancial cost-effectiveness of each incentive has to be determined in terms of its impact on LCOEs and hence its ability to facilitate investments per dollar spent. In order to provide more illustrative numbers, cost effectiveness was calculated in terms of the dollar amount that would have to be spent or the tax revenue that would have to be foregone in order to lower LCOE by 1 percent. By assessing cost-effectiveness, the report aims to provide policy makers with the information they need to choose a set of regulatory incentives that can both (a) maximize the impact on LCOEs and therefore facilitate investments; and (b) limit the overall societal cost in ï¬?nancial terms by maxi- mizing impact per dollar spent. To represent the ï¬?nancial burden of an incentive pro- gram be er, costs were extrapolated for 500 MW capacity, which was expected to come in the form of ï¬?ve individual 100 MW plants. The actual composition of the societal cost mainly comes in the form of lower tax revenues (when tax reductions, VAT exemptions, and/or accelerated depreciation are granted) or in the form of additional expenditures (when concessional loan terms and/or rates are provided—in our example by blending concessional and commercial ï¬?nancing so as to lower the applicable debt interest rate for the debt share of each individual plant 50 A World Bank Study by 3 percent). The ï¬?nal value was calculated as the NPV of the difference in cash flows for income tax payments (for tax reduction and accelerated depreciation), the difference in upfront VAT payments on total direct costs (VAT exemptions) and the indicative cost of upfront fees and guarantees (in the case of concessional loan terms and rates). In the la er case, it was assumed that concessional ï¬?nancing would be channeled to developers through a government intermediary that would cover expenses related to upfront fees and the purely administrative cost of providing the necessary guarantees. Under the assumption of a zero percent probability of default and not accounting for their economic opportunity cost, guarantees would under this framework have a rela- tively low societal cost in ï¬?nancial terms.8 The analysis, however, quantiï¬?es the amount of guarantees that would have to be granted to allow for an easy calculation of societal cost if a higher probability of default is to be assumed. The overview of the results for India, Morocco, and South Africa are provided in tables 5.9–5.11. Since the differences between wet- and air-cooled assumptions are negligible, we omi ed the wet-cooled cases to allow for a be er overview. All three concessional schemes—with longer loan terms (25 years in all three scenar- ios) combined with lower loan rates (3 percent, lower applicable debt interest by blending concessional and commercial ï¬?nancing)—are the most cost-effective ways of lowering LCOEs for both technologies in ï¬?nancial terms, as long as the assumed probability of default is less than 25 percent. The amount of concessional ï¬?nancing necessary to lower applicable loan rates would, however, be considerable—from around US$877 million for Table 5.9: Sensitivity analysis India—cost-effectiveness of regulatory approaches Incentive Reduction in Cost impact for US$ per Technology granted LCOE (%) Cost effect 500 MW (US$) 1% LCOE Parabolic trough Tax reduction – 00.96 Lower tax 81.7 million 85.1 million (Air-cooled—with revenues storage) VAT exemption – 0.96 Lower tax 47.2 million 49.1 million revenues Accelerated – 4.16 Lower tax 149.2 million 35.9 million depreciation revenues Concessional – 16.12 Upfront fees 2.2 milliona 0.14 million loan terms and guarantees (877 million in guarantees) Power tower Tax reduction – 0.97 Lower tax 88.1 million 90.8 million (Air-cooled—with revenues storage) VAT exemption – 0.97 Lower tax 50.9 million 52.5 million revenues Accelerated – 4.17 Lower tax 160.8 million 38.6 million depreciation revenues Concessional – 16.19 Upfront fees 2.4 milliona 0.15 million loan terms and guarantees (945 million in guarantees) Source: Authors’ data. a. These numbers were calculated assuming that the societal cost of guarantees, in ï¬?nancial terms and not accounting for economic opportunity cost, would consist of the front-end fee of 0.25 percent of the total loan amount. The actual loan amounts were calculated to cause a 3 percent drop in the cost of debt for the total debt capital share, based on a concessional ï¬?xed LIBOR + 1.5 percent rate. Concentrating Solar Power in Developing Countries 51 Table 5.10: Sensitivity analysis Morocco—cost-effectiveness of regulatory approaches Incentive Reduction Cost impact for US$ per 1% Technology granted in LCOE (%) Cost effect 500 MW (US$) LCOE Parabolic trough Tax reduction – 1.21 Lower tax revenues 156.3 million 129.2 million (Air-cooled—with VAT – 1.93 Lower tax revenues 117.9 million 61.1 million storage) exemption Accelerated – 14.31 Lower tax revenues 296.1 million 20.7 million depreciation Concessional – 18.77 Upfront fees and 3.0 milliona 0.16 million loan terms guarantees (1,189 million in guarantees) Power tower Tax reduction – 1.20 Lower tax revenues 188.4 million 157.0 million (Air-cooled—with VAT – 1.98 Lower tax revenues 142.3 million 71.9 million storage) exemption Accelerated – 14.48 Lower tax revenues 357.0 million 24.7 million depreciation Concessional – 19.04 Upfront fees and 3.6 milliona 0.19 million loan terms guarantees (1,434 million in guarantees) Source: Authors’ data. Table 5.11: Sensitivity analysis South Africa—cost-effectiveness of regulatory approaches Incentive Reduction Cost impact for US$ per Technology granted in LCOE (%) Cost effect 500 MW (US$) 1% LCOE Parabolic trough Tax reduction – 1.75 Lower tax 144.0 million 82.3 million (Air-cooled—with revenues storage) VAT – 2.01 Lower tax 126.2 million 62.8 million exemption revenues Accelerated – 12.41 Lower tax 262.0 million 21.1 million depreciation revenues Concessional – 12.03 Upfront fees 2.4 milliona 0.2 million loan terms and guarantees (967 million in guarantees) Concessional Upfront fees loan rates and guarantees Power tower Tax reduction – 1.77 Lower tax 168.1 million 95.0 million (Air-cooled—with revenues storage) VAT – 2.05 Lower tax 146.6 million 71.2 million exemption revenues Accelerated – 12.60 Lower tax 306.0 million 24.3 million depreciation revenues Concessional – 12.24 Upfront fees 2.8 milliona 0.23 million loan terms and guarantees (1,124 million in guarantees) Source: Authors’ data. 52 A World Bank Study parabolic trough plants in India to more than US$1.4 billion for power tower plants in the case of Morocco, assuming a total capacity of 500 MW. Compared to simple tax reductions or exemptions that proved to be by far the least cost-effective incentive across all scenarios and technologies, requiring up to US$90 million in order to reduce LCOEs by 1 percent, accelerated depreciation seems by far a superior option. Although at US$21 to US$38 million per 1 percent reduction in LCOE is not that inexpensive, they might be worth considering in cases where—as seen in the case of South Africa—the existing regulatory incentive framework just needs to be moderately adjusted to lower LCOEs to the threshold where stand-alone projects become ï¬?nancially viable. Balance Sheet vs. Off-Balance-Sheet Financing All LCOE calculations in this chapter assumed largely nonrecourse or off-balance-sheet ï¬?nancing under the applicable ï¬?nancial and regulatory conditions in the respective jurisdiction, albeit complete nonrecourse project ï¬?nancing may be unrealistic for the ï¬?rst generation of such projects, since lenders may seek some limited recourse to the assets of the sponsor, particularly until the construction phase is completed and any cost over- runs have been fully accounted for and paid by the sponsor. LCOE estimates, however, can in theory drop considerably if plants are ï¬?nanced on balance sheet, depending on the ï¬?nancial standing of the respective company. If a plant is to be ï¬?nanced on balance sheet, the assumption would be that the weighted average cost of capital (WACC) for the project would equal the general cost of capital of the respective company, which might be lower than the commercial loan rate a stand-alone project could receive. In addi- tion, balance sheet ï¬?nancing might also avoid the need to cope with other constraints that nonrecourse ï¬?nancing entails, including the need to fulï¬?ll a minimum debt service coverage ratio (DSCR) and requirements for positive cash flows. By contrast, balance sheet ï¬?nancing increases the risk proï¬?le of a company’s investments and might require cross-subsidization between projects, since the ï¬?nancial viability of a project on a stand- alone basis is no longer guaranteed. In the case of India (see ï¬?gure 5.9), LCOEs would Figure 5.9: Balance sheet versus off-balance-sheet ï¬?nancing effects on LCOE in India 40 35 Nominal CERC FIT 30 Effective CERC FIT 25 20 15 10 5 0 Non-recourse financing scenario On balance sheet financing scenario Parabolic trough (air-cooled) Parabolic trough (wet-cooled) Power tower (air-cooled) Power tower (wet-cooled) Source: Authors’ data. Concentrating Solar Power in Developing Countries 53 drop considerably by around 33 percent for each technology under the assumption of a WACC based, for example, on a cost of capital of 8 percent for a large integrated infra- structure company, a repayment period that would stretch over the plant’s economic lifetime (25 years), and no minimum DSCR requirements. This would bring LCOEs under the threshold of the effective CERC FiT (based on lowest bid), but would not nec- essarily make projects ï¬?nancially viable on a stand-alone basis. Conclusions Based on the above results, the following observations can be made: â–  DNI accuracy ma ers—any underlying ï¬?nancial analysis for a CST plant is only as good as the quality of the DNI data the plant is modeled on. Given the inverse relationship between DNI and LCOE for CST plants, any analysis not based on data measured on the ground at the actual site of the project over the course of at least a full year will not provide sufficient grounding for a diligent ï¬?nancial model. â–  For all technologies in all three scenarios considered, the LCOEs for stand-alone projects are most likely too high to allow for cost recovery and meeting ï¬?nanc- ing constraints at present. This is speciï¬?cally the case when the LCOEs are com- pared to the FiTs available for CST-generated electricity in Phase 1 of the JNNSM in India and the FiTs that have been proposed for Phase 2 of the REFIT scheme in South Africa. LCOE calculations based on balance-sheet ï¬?nancing might be considerably lower than calculations based on nonrecourse (off-balance-sheet) ï¬?nancing assumptions, such as the ones made for this analysis. However, balance-sheet ï¬?nancing increases the risk proï¬?le of a company’s investments and might require cross-subsidization between projects, since the ï¬?nancial via- bility of a stand-alone project is no longer guaranteed. â–  Financial and regulatory incentives, as well as concessional ï¬?nancing schemes, can signiï¬?cantly lower LCOEs. Within the range of considered ï¬?nancial and reg- ulatory incentives, simple tax reductions and exemptions tend to have the lowest impact and are most likely the least cost-effective incentives in ï¬?nancial terms (not considering economic opportunity cost). By contrast, concessional ï¬?nanc- ing schemes tend to have the highest impact and are likely to be the most cost- effective incentives in terms of their impact on LCOE on a per-dollar spent basis. â–  With regard to the other incentives considered, accelerated depreciation, espe- cially when compared to simple tax reductions or exemptions, seems to be the superior option. Although far from cheap, it might be worth considering in cases where—as seen in the case of South Africa—the existing regulatory incentive framework just needs to be moderately adjusted to lower LCOEs to the threshold where stand-alone projects become ï¬?nancially viable. Economic Analysis of Reference CST Plants This section presents an economic analysis, based on current investment costs, for ref- erence 100 MW CST plants—both parabolic trough and power tower—in the respec- tive three countries considered for the analysis—India, Morocco, and South Africa. The economic analysis consists of estimating full economic costs and beneï¬?ts of individual projects, and calculating the economic net present value (ENPV) at a 10 percent discount 54 A World Bank Study rate and the internal economic rate of return (ERR). In addition, a sensitivity analysis was performed for the following scenarios: (a) 10 percent and 20 percent higher total project cost; (b) a 20 percent lower load factor; and (c) a 60 percent higher value of power. The main cost assumptions are provided in table B.15 in Appendix B, which in general summarizes the assumptions used in the analysis. The main results for the three coun- tries are given in tables 5.10–5.12, respectively, for India, Morocco, and South Africa. The following general observations can be made across all three countries: 1. In none of the countries does the ERR achieve a rate required for infrastructure projects of over 10 percent. Without the carbon and other environmental ben- eï¬?ts the ERR ranges from -0.65 percent to 4.8 percent for the power tower and from -2.55 percent to 3.8 percent for the parabolic trough. With carbon (and local pollutant beneï¬?ts for Morocco), the ERR ranges from 2.1 percent to 8.8 percent for the power tower and from 1.1 percent to 7.4 percent for the parabolic trough. 2. Valuing carbon using the wider social costs of carbon rather than a single value increases the ERR by 1–2 percent (South Africa). If a single value is used the ERR goes up by about 0.5 percent. 3. The carbon values needed to achieve an ERR would be implausibly large in India and Morocco. In South Africa they would also be quite high, but one could argue that carbon emissions reduction projects with costs in that range (US$80–100/ton CO2) have been undertaken in other sectors. 4. The sensitivity analysis shows approximately a 1 percent reduction in the ERR for a 10 percent higher project cost and a further 1 percent reduction for a 10 percent Table 5.12: Economic analysis for CST reference plants in India Sensitivity analysis for the base case India: central receiver power tower Cost overrun Load factor Value of power Base Case 5Yr Delay 10% 20% 20% Lower 60% Higher No carbon beneï¬?ts 0.00% 2.39% −0.74% −1.39% −2.64% 5.55% Revised carbon 3.95% 6.88% 3.10% 2.34% 1.30% 8.38% beneï¬?ts Carbon price for 12% IRR US$/Ton CO2 153.3 97.0 174.7 196.0 215.4 97.0 Sensitivity analysis for the base case India: central receiver-parabolic trough Cost overrun Load factor Value of power Base Case 5Yr Delay 10% 20% 20% Lower 60% Higher No carbon beneï¬?ts 2.11% 3.83% 1.47% 0.90% −0.19% 7.00% Revised carbon 5.57% 7.95% 4.81% 4.14% 3.23% 9.53% beneï¬?ts Carbon price for 12% IRR US$/Ton CO2 137.8 87.3 159.0 178.5 196.0 81.5 Source: Macroeconomica 2011. Note: the carbon price is for 2012 or 2017 in the case of the 5-year delay. The central value for 2012 is US$38.8/ton and the central value for 2017 is US$43.1/ton. Concentrating Solar Power in Developing Countries 55 Table 5.13: Economic analysis for CST reference plants in Morocco Sensitivity analysis for the base case Morocco: Central receiver power tower Cost overrun Load factor Value of power Base Case 5Yr Delay 10% 20% 20% Lower 60% Higher No carbon beneï¬?ts −0.65% 1.46% −1.46% −2.18% −3.45% 5.27% Original carbon 1.77% 3.94% 0.90% 0.13% −0.98% 6.93% beneï¬?ts Revised carbon 2.07% 4.76% 1.19% 0.40% −0.70% 7.15% beneï¬?ts Carbon price for 12% IRR US$/Ton CO2 252.3 159.0 291.1 302.40 357.1 157.2 Sensitivity analysis for the base case Morocco: Parabolic trough Cost overrun Load factor Value of power Base Case 5Yr Delay 10% 20% 20% Lower 60% Higher No carbon beneï¬?ts −2.93% −0.02% −3.54% −4.07% −6.66% −2.93% Original carbon 0.23% 2.14% −0.45% −1.06% −2.85% 0.23% beneï¬?ts Revised carbon 0.87% 2.82% 12.04% −0.45% −2.12% 8.65% beneï¬?ts Carbon price for 12% IRR US$/Ton CO2 295.0 217.40 333.7 368.7 411.4 201.0 Source: Macroeconomica 2011. Note: the carbon price is for 2012 or 2017 in the case of the 5-year delay. The central value for 2012 is US$38.8/ton and the central value for 2017 is US$43.1/ton. higher project cost. A reduction in the load factor of 20 percent has a bigger impact—reducing the ERR by 2.5–3 percent. 5. The value of power is a critical factor in the ERR. Ideally it should be measured as the willingness-to-pay for the additional power. Using the market price as a proxy would result in an underestimated willingness-to-pay, since it ignores the consumer surplus, but the adjustment is small if the project adds only a small amount to the total generation and does not supply individuals who are currently without power or with limited access to electricity. In countries with power short- ages, some adjustment for this factor has to be warranted. In any event, if the power supplied has a higher value, the ERR goes up a lot and can even exceed 12 percent (see, for example, table 5.12). 6. A delay in starting the project has two effects. First, there is a reduction in cost because of technology developments, and second there is an increase in the val- ue of power, as consumers’ willingness-to-pay increases. Decreases in the capital costs are assumed to be around 10 percent in the case of the parabolic trough and around 8 percent in the case of the power tower over the ï¬?ve years of delay assumed. The results of a ï¬?ve-year delay are to increase the ERR by 1–3 percent, depending on how much future power beneï¬?ts rise (see tables 5.13 and 5.14). 56 A World Bank Study Table 5.14: Economic analysis for CST reference plants in South Africa Sensitivity analysis for the base case Load Value of South Africa: Central receiver power tower Cost overrun factor power Base case 5Yr delay 10% 20% 20% Lower 60% Higher No carbon beneï¬?ts 4.80% 5.55% 3.76% 2.85% 1.63% 12.00% Original carbon 7.04% 7.88% 5.92% 4.94% 3.80% 13.65% beneï¬?ts Revised carbon 8.81% 11.96% 7.65% 6.62% 5.55% 14.93% beneï¬?ts Carbon price for 12% IRR US$/Ton CO2 76.9 62.1 95.1 112.50 128.1 0.0 Sensitivity analysis for the base case Load Value of South Africa: Central receiver-parabolic trough Cost overrun factor power Base case 5Yr delay 10% 20% 20% Lower 60% Higher No carbon beneï¬?ts 3.80% 4.31% 2.97% 2.24% 1.04% 9.93% Original carbon 5.72% 6.39% 4.81% 4.02% 2.94% 11.33% beneï¬?ts Revised carbon 7.41% 8.63% 6.47% 5.65% 4.76% 12.52% beneï¬?ts Carbon price for 12% IRR US$/Ton CO2 104.8 78.7 124.2 143.6 158.9 31.1 Source: Macroeconomica 2011. Note: the carbon price is for 2012 or 2017 in the case of the 5-year delay. The central value for 2012 is US$38.8/ton and the central value for 2017 is US$43.1/ton. Country-speciï¬?c observations include the following: 1. In the case of India, the results show that a parabolic trough has a higher return than power tower; a ï¬?ve-year delay increases the ERR by nearly 3 percent. 2. In the Moroccan case study, the delay is not as effective in increasing the ERR (possible because the increases in power value are more modest). Even with car- bon and local pollutant beneï¬?ts, the ERR is well below a test rate. Power tower appears to exhibit slightly be er economics than parabolic trough. 3. For the South African case, because of the higher value of power and the re- vised carbon beneï¬?ts, a 12 percent ERR can be exceeded with both technologies, although the power tower has a higher return by 1–2 percent. Including beneï¬?ts of reduced local pollutants would increase the ERR further—by up to 1 percent. When comparing air- and wet-cooling technologies, it becomes evident that there are clear differences between the technologies with respect to performance and cost, which are as summarized in table 5.15. To indicate the impacts of the technologies on the ERR, the base case for each coun- try has been rerun with the alternative technology. The results are given in table 5.16. Concentrating Solar Power in Developing Countries 57 Table 5.15: Performance and cost penalties Technology Process Performance Penalty Cost Penalty Power tower Wet cooling None None Air cooling 1–3% 5% Parabolic trough Wet cooling None None Air cooling 4.5–5% 2–9% Source: Macroeconomica 2011. Table 5.16: Impacts of dry versus wet cooling technologies India Morocco South Africa Parabolic trough Dry-Cooling 5.6% −0.5% 7.4% Wet-Cooling 6.7% 0.9% 8.9% Power Tower Dry-Cooling 4.0% 1.8% 8.8% Wet-Cooling 4.2% 2.1% 9.1% Source: Macroeconomica 2011. Wet-cooling technology increases the ERR in the case of the parabolic trough by around 1.5 percent and 0.2 percent in the case of the power tower. The analysis presented here indicates that while power tower technology has a slightly higher return than parabolic trough, and the use of wet cooling can slightly improve the ERR, CST plants in general, assuming current prices, do not have an ERR that would meet commercial infrastructure investment requirements. However, invest- ment costs are projected to decrease considerably over the coming years—a develop- ment that is expected to largely alter the economics of CST technologies. Further on, the decision to uptake CST technology might not necessarily be based on economic consid- erations alone, but might include other aspirations, such as gaining market leadership and experience through technology development or targeting the building-up of a local manufacturing industry. There are also potential ways of improving the economics of CST even under current investment cost assumptions through, for example, hybridiza- tion and the large-scale application of storage—areas that, however, remain outside the scope of this report. Notes 1. The necessary physical weather data with regard to Direct Normal Irradiation (DNI) were taken from the U.S. Department of Energy’s EnergyPlus Energy Simulation Software weather database. 2. An often-cited example of the lack of economies of scale in production is that the relatively high estimated LCOE for Dish Stirling at US$0.28–0.35/kWh will only be feasible with production levels above 500 Dish Stirling per year, which is unlikely in the short term. This leaves an increased inter- est in Dish Stirling as a source of distributed, off-grid generation in areas where fuel costs and fuel supply costs would make Dish Stirling competitive relative to fossil-based capacity. 3. To perform the affordability and sustainability analyses, this report relied on the Solar Advisory Model (SAM)—Version 2010.11.9—provided by the U.S. National Renewable Energy Laboratory (NREL) in cooperation with Sandia National Laboratories and the U.S. Department of Energy 58 A World Bank Study Solar Energy Technologies Program (SETP). The model is widely used for planning and evaluating research, and developing cost projections and performance estimates, and it relies on NREL’s and Sandia’s long-standing experience with CSP. The necessary physical weather data with regard to DNI were taken from the U.S. Department of Energy’s EnergyPlus Energy Simulation Software weather database. When no site-speciï¬?c DNI data were available, mock DNI data for comparable sites and DNI resources were chosen. 4. The respective combination of storage and solar multiple/tower height and receiver dimensions was identiï¬?ed by running parametric simulations for a range of solar multiple, tower height, and receiver dimensions values. 5. The optimal amount of storage for each parabolic trough plant was based on the parametric simulation for a range of solar multiple values are the following: India, 6 hours with a solar mul- tiple of 2.5; Morocco, 3 hours with a solar multiple of 1.75; and South Africa, 3 hours with a solar multiple of 1.75. For power tower plants, optimal storage is 15 hours in all three cases with a solar multiple of 3. 6. This information was provided by developers active in the respective country on a nondisclo- sure basis to bank staff. It reflects the assumed actual ï¬?nancial and regulatory conditions indepen- dent developers would be facing when considering the construction of a reference 100 MW CSP plant in their respective jurisdiction. 7. This assumes that concessional ï¬?nancing can be blended with commercial ï¬?nancing up to the amount of concessional ï¬?nancing necessary to lower the overall interest rate of the debt share of an individual plant by 3 percent, whereby the actual amount of concessional ï¬?nancing needed to reach a 3 percent reduction of the average debt interest rate depends on the commercial rate avail- able. The assumption for concessional ï¬?nancing was a LIBOR + 1.5% interest rate. 8. In economic terms, guarantees indeed have an opportunity cost, since the money could have been used for activities with a higher economic rate of return. However, given that the use of avail- able concessional ï¬?nancing is often limited to the ï¬?nancing of renewables, this opportunity cost can be regarded as relatively negligible. Likewise, the effect of guarantees on a respective country’s balance sheet—potentially affecting a country’s general interest rate—might not be sizeable in the case study countries considered for this analysis. CHAPTER 6 Assessment of Local Manufacturing Capabilities for CST T o realize the cost reduction trends described in Chapter 5, a major scale-up of CST developments would be necessary, both in the already established markets, as well as in emerging markets in the MENA region, India, and South Africa. A major increase in CST capacity in emerging markets is, however, only likely when the countries con- cerned beneï¬?t from the technology for their economic development in general. One of the primary means to foster development could be the establishment of local manufac- turing capacities. Local manufacturing would have the added beneï¬?t of reducing the cost of local projects in the near term and bringing down the cost for a variety of com- ponents and CST-related services in the mid- to long term. This chapter assesses local manufacturing capabilities in several emerging markets for CST, including the MENA region and South Africa. It also provides some estimates on the economic beneï¬?ts and potential employment opportunities that could be generated. It should be noted that such estimates have been carried out on a gross basis, without considering the cost for reducing or not expanding alternative technologies. Local Manufacturing Capabilities in MENA1 The CST Value Chain in MENA An evaluation of the MENA region’s potential for developing a home base for CST requires a detailed analysis of the CST value chain: the technologies and services, the production processes, and the main industrial players. It is also important to review the cost of CST and contributions from individual components of the CST value chain. Based on the complexity level and the potential for local manufacturing, as well as the share of added value in the CST value chain, a number of key components and services can be identiï¬?ed that are most promising: key components include mounting structures, mirrors, and receivers, while key services range from assembling and EPC to operation and maintenance (O&M). Single countries within the MENA region have already developed some production capabilities of secondary components—including electronics, cables, and piping—which might contribute to the local supply of future CST projects, although their share in the overall value chain might yet be of minor importance. Figure 6.1 shows the different components and services linked to the pro- duction and use of CST. Based on a detailed analysis of these components, it seems evident that there are a variety of opportunities for local manufacturing and the local provision of services all along the value chain. 59 60 A World Bank Study Figure 6.1: Components and services for CST Low or Potential for Cost share medium local in value complexity manufacturing chain Mounting CSP key components Mirrors Receivers structure Road Action CSP key services Assembling O&M EPC map plan CSP secondary Electronics Cable Piping components CSP other Trackers, HTF, pumps, storage, power components block, control system, etc. Source: Ernst & Young and Fraunhofer 2010. Drawing on a detailed analysis of (a) the global CST value chain (an overview is provided in table B.16 in Appendix B) and (b) a detailed assessment of the opportunities for MENA industries to manufacture CST components in the value chain, including an analysis of technical and economic barriers for local manufacturing (see table B.17 in Appendix B), the following SWOT analysis of MENA industries illustrating the respec- tive strengths, weaknesses, opportunities, and threats for the industries with regard to participating in the CST value chain can be provided (see table 6.1). â–  Aside from the SWAT analysis, the following general conclusions can be drawn: A growing market has been identiï¬?ed for all groups in the value chain (raw materials, components, engineering, engineering, procurement and construc- tion contractors, operator, owner, investors and research institutions). â–  High-technological know-how and advanced manufacturing processes are nec- essary for some key components, such as parabolic mirrors or receivers, which nevertheless offer the highest reward in terms of value added. â–  Some sectors and companies, such as receiver suppliers, strongly depend on CST market demand and growth. Other ï¬?rms have built their production and manufacturing capacities to respond to the demand of other markets (CST is a niche for them). â–  Some components (piping, HTF, electronics, power block) can be produced by companies without extensive CST know-how or background because this equipment is used for many other applications (chemical, electronic, and elec- tric industries). â–  The potential of MENA CST may be achieved by the manufacture of compo- nents by local, regional, and international companies, and the construction of CST plants in MENA by local construction companies and subsidiaries of inter- national CST companies. Concentrating Solar Power in Developing Countries 61 Table 6.1: SWOT analysis of MENA industries suitable for CST Strengths Weaknesses • Low labor cost (especially for low-skilled workers) • Insufï¬?cient market size • One of the highest solar potentials in the world • Administrational and legal barriers • Strong GDP growth over the past ï¬?ve years in all MENA • Lack of ï¬?nancial markets for new ï¬?nancing countries • Higher wages for international experts and engineers • High growth in the electricity demand will require large • Higher capital costs investments in new capacities • Energy subsidized up to 75% in some countries • Strong industrial sector in Egypt • Weak or nonexistent ï¬?scal, institutional, and legislative • Particular proximity of Spain and Morocco frameworks for RE development • Existing float glass sector in Algeria • Despite regulations, implementation and enforcement of • Large export industry in Tunisia and Morocco with long environmental regulations often deï¬?cient experience with Europe (for example, the automotive • Need for network of business and political connections industry and, to a lesser extent, aeronautics) • Lack of specialized training programs for renewable • SCCS plants in three countries constructed by 2010 • Partly insufï¬?ciently developed infrastructure Opportunities Threats • Further cost reduction of all components • Training of workforce and availability of skilled workers • Attractive to external investors insufï¬?cient • Solar energy: Moroccan Solar Plan (2 GW), Tunisian • Technical capacities of local engineering ï¬?rms Solar Plan, and premises of an Egyptian Solar Plan, for • Low awareness of management of CST opportunities example • Access to ï¬?nancing for new production capacities • Possibility of technology transfer or spillover effects from • Competition with foreign stakeholders: German players and foreign stakeholders in MENA strong interest of the United States in the Egyptian market • Political will to develop a local renewables industry • Higher costs compared to international players • Export potential (priority given to export industries) • High costs because of insufï¬?cient infrastructure Source: Ernst & Young and Fraunhofer 2010. â–  Production capabilities for some key components (mirrors and receivers) moved to the current CST markets in Spain and the United States as soon as the market (or prospects for the market) had a ained a sufficient size. They could move to MENA when the CST market takes off in the region. Potential for Local Manufacturing In the near- to midterm, international companies will have an important role to play in the development of local industries. EPC companies and project developers already active in the region have local offices in MENA countries close to the CST projects and their customers. The companies employ local and international workers and engineers for projects in the countries. Comparable with conventional power plants, CST compa- nies also expect a large share of project development, management, and engineering from international companies with extensive technical expertise and project experience. Table 6.2 provides an overview of the possible local content of different parts in the value chain as seen by international players. Several industrial sectors with the potential to integrate the CST value chain in the MENA region are dynamic and competitive on a regional, and sometimes international, scale. The glass industry, for example, particularly in Egypt and Algeria, has been a regional leader for a long time and is still increasing its production capacity. The cable, electrical, and electronic industry can also claim the same position, especially in Egypt, Morocco, and Tunisia. The success of these industries is facilitated by the development of joint ventures between large international companies and local ï¬?rms, as well as by the local implantation of subsidiaries of international players. In the past, the develop- ment of MENA industries was driven by the low cost for labor and energy (the la er in 62 A World Bank Study Table 6.2: Possible local content by component of CST power plants Component Local Manufacturing Possible? Services and Power Block Local Manufacturing Possible? Mirrors Yes, large market Civil works Yes, up to 100% Receivers Yes, long-term Assembling Yes, up to 100% Metal structure Yes, today Installation works (solar ï¬?eld) Partly, up to 80% Pylons Yes, today Power block No Trackers Partly Grid connection Yes, up to 100% Swivel joints Partly Project development Partly, up to 25% HFT systems No, except pipes EPC Partly, up to 75% Source: Ernst & Young and Fraunhofer 2010. particular for Algeria and Egypt) and by the geographic proximity to Europe. To posi- tion themselves for the CST market, MENA industries face several challenges, mainly in adapting their capacity to higher technology content. The landscape is already changing; the situation of pure subcontracting is now shifting toward more local R&D and the pro- duction of high-tech components. MENA countries are aiming to be considered centers of excellence instead of low-cost and low-skilled workshops. Key ï¬?ndings on the status quo and future perspectives of local manufacturing include the following: â–  Successfully constructed integrated solar combined cycle system (ISCCS) projects have increased CST experience and know-how in MENA. â–  Some components and parts for the collector steel structure were supplied by the local steel manufacturing industry (Algeria, Egypt, and Morocco). â–  The workforce has been trained on the job; engineering capacities have also seen progress. â–  Specialization of each country would be beneï¬?cial because local demand will probably be relatively low in the short and medium terms. â–  Several parts of the piping system in the solar ï¬?eld—for the interconnection of col- lectors and power block—can already be produced locally by regional suppliers. â–  The development of a CST mirror industry in MENA countries has signiï¬?cant potential. â–  Involvement of international companies will play an important role in the mid- term development of the CST industry in MENA countries because it will build up local production facilities. â–  Minimum factory outputs have to be taken into consideration for local manu- facturing of special components (glass, receivers, salt, thermal oil). The prospects for local manufacturing can be summarized for each component: â–  Construction and civil works: In the short term, all construction at the ï¬?nal plant site with the basic infrastructure, installation of the solar ï¬?eld, and con- struction of the power block and storage system could be accomplished by local companies (17 percent of total CST investment for a reference plant or approxi- mately US$1 million per megawa ). â–  Mounting structure: The mounting structure can be supplied locally if local companies can adapt manufacturing processes to produce steel or aluminum components with the required high accuracy. Concentrating Solar Power in Developing Countries 63 â–  CST-speciï¬?c components with higher complexity: In the short to medium term, local industry is generally capable of adapting production capacities and creating the technological knowledge to produce mirrors (glass bending, glass coating, and possibly float glass process) of high quality and to a high technical standard, as required for parabolic mirrors in parabolic trough plants. This might require inter- national cooperation for speciï¬?c manufacturing steps in the short term. Later, local provision of components could include high-quality mirrors, receivers, electronic equipment, insulation, and skills for project engineering and project management. In particular, for the receiver (absorber) technology, the most promising option will be for international companies to move closer to the rapidly increasing markets. Possible evolutions of local CST industries for some of the key components (mirrors, mounting structure, and electrical and electronic equipment) in the MENA region are provided in Figure B.1 in Appendix B, taking into account the market size for different components. Scenarios for Local Manufacturing in MENA Countries It is assumed that the volume of installed CST capacity within the MENA region (the home market volume) is a main precondition for the emergence of local manufacturing. Thus, the scenarios represent critical levels of market development for local manufac- turing. The home market volume and the potential amount of export (external market volume) are regarded as indicators for the development of a successful policy scheme. The scenarios chosen here therefore represent critical levels of market development for local manufacturing (for an overview, see ï¬?gure 6.2). Scenario A—Stagnation: The home market volume amounts to only 0.5 GW. Strong obstacles to local manufacturing of CST components remain in the country Figure 6.2: Interrelations between MENA home market size, possible export volume and focus of support for local industries MENA home market volume 0.5 GW 1 GW 5 GW Scenarios A B C v v v Potential foreign trade 0 2 GW Focus of support Enhancing the provision Adaptation of international Strengthening the of products and services production and service innovative capacity for with low barriers by standards for components CSP components and existing companies with medium barriers services Source: Ernst & Young and Fraunhofer 2010. 64 A World Bank Study markets, and most components, particularly those whose production requires high investment costs, are imported from more advanced markets. Scenario B—No-replication: The home market volume amounts to 1 GW in 2020. In this scenario, the market offers some opportunities for the development of local manufacturing of CST components and provision of CST services. This scenario aims at an adaptation of international production standards and techniques in existing industries, and leads to a region-wide supply of suitable CST compo- nents produced locally in the MENA region. Scenario C—Transformation: The home market volume of the ï¬?ve countries amounts to 5 GW, and the export of components reaches a volume correspond- ing to 2 GW installed CST capacity. National CST promotion plans have been developed quickly, international initiatives are strongly represented, and/or private investors are notably active in the region. Policy actions should support innovations and the development of intellectual property rights in the ï¬?eld of CST components. Roadmaps for the Development of Local Manufacturing of CST Components in the MENA Region Based on the assessment and identiï¬?cation carried out of existing and potential domes- tic and foreign players, potential routes to developing local manufacturing capabilities were identiï¬?ed. The aim of the roadmap is to show possible technological and entrepre- neurial developments in the regional manufacturing of each component in the short, medium, and long term and to identify overall, long-term objectives in these ï¬?elds. Fig- ure 6.3 provides a detailed roadmap for EPC services in CST projects. A further roadmap for key mirrors is to be found under ï¬?gure B.3 in Appendix B. Figure 6.3: Potential roadmap for EPC and services in MENA CST projects Status Quo Short Term Mid-Term Overall Goal Business development Few large EPC Subcontracts in Large regional EPC Project Engineering & contractors are CSP projects given contractors with management is construction are active in MENA. to local companies comprehensive know- carried out by completed by local First experiences in by international how in the field of CSP MENA companies companies only CSP projects have EPC contractors Local service are active in MENA and already been providers gain supra-regional. Other gained. profound project Positive spill-over sectors benefit from experience & effects on other their profound Logistics are local workforce service sectors experience. organized locally receives extensive training All civil works, on-site Civil works and on- Assembly is carried Independent jig- assembly, logistics and site assembly are out locally (under and field assembly maintenance works are partly performed by supervision of by local companies accomplished by the local workforce. experienced EPC local workforce. contractors) Policy framework & market development A well-trained Clearly formulated No national targets Strong focus on Coordinated workforce for the political targets. for development of Long-term, stable education & national strategies CSP service sector Extensive availability of CSP and related policy framework training related to defined for service is widely available training centers, well- service sector, no is implemented & CSP services sector development Facilitated transport trained workforce specific training and energy targets public funds made Extensive upgrade available available of transport & of CSP components Well-developed infra- communication leads to more structure assures Infrastructure partly efficiency of logistic transport services and underdeveloped infrastructure procedures communication Source: Ernst & Young and Fraunhofer 2010. Concentrating Solar Power in Developing Countries 65 A detailed action plan for stimulating CST manufacturing and service provision in the MENA region was developed for all relevant actors (see also table B.18 in Appendix B) summarizing the potential measures addressed to different actors to stimulate the pro- duction of CST components and provide CST-related services in the MENA region that most likely would have to include the following: â–  The creation of a stable policy framework and sustained domestic market for CST is a key precondition for the development of local manufacturing in MENA countries. Long term, the annually installed capacity should be on a gigawa scale for the development of production lines, particularly in the case of mirrors and receivers. â–  National strategies for industrial development and energy policy should be well coordinated and involve clear targets for the market diffusion of CST, sub- stantial R&D efforts, strategy funds for industrial development of CST industry sectors, and stronger regional integration of policies. â–  A provision of low-interest loans and grants speciï¬?cally designed for local manufacturing of renewable energy components might help local compa- nies raise the funds for the innovation of production lines or new company start-ups. â–  Another direct political measure to foster a long-term demand for CST compo- nents would be the introduction of local (domestic) content clauses within CST tenders and other support instruments. â–  To enhance the innovative capacity of the industrial sectors, the creation of a larger number of technology parks or clusters and regional innovation plat- forms should be pursued. This would particularly help small and medium-size ï¬?rms overcome innovation barriers and gain access to the latest technological advancements. â–  Business models should build on the comparative advantages of certain sectors in MENA countries and also involve international cooperation agreements, for example, in the form of joint ventures and licensing. In the case of receivers, sub- sidiaries of foreign companies will most likely be the relevant business model in the beginning. Governments could assist the private sector in the matchmaking process leading to such cooperation. â–  The investment in new production lines based on highly automated processes for the mounting structure and glass production, as well as adaption of tech- niques for coating and bending mirrors, will be a crucial ï¬?rst step. â–  Establishing local manufacturing will involve comprehensive education and training programs for the industrial workforce in relevant sectors. Uni- versities should be encouraged to teach CST technology-based courses to educate the potential workforce, particularly engineers and other technical graduates. â–  Additionally, to ensure regional and international quality requirements and to strengthen the competitiveness of future Middle Eastern and North African CST industries, implementing quality assurance standards for CST components should be considered in the medium to long term. â–  For the service sector, local assembly of the plants and involvement of local EPC contractors are important initial steps for increasing the local component. 66 A World Bank Study Potential Economic Beneï¬?ts of Developing a CST Industry in North Africa The economic beneï¬?ts of developing a CST industry were evaluated for the three CST scenarios (stagnation, no replication, and transformation) for northern Africa. The economic impact on GDP is depicted in table 6.3—economic impact is strongly related to the market size of CST in the Middle East and North Africa region. Scenario C creates a local economic impact of US$14.3 billion, roughly half of which is from indi- rect impacts in the CST value chain (excluding component exports), compared to only US$2.2 billion in scenario B. Table 6.3: Direct and indirect local economic impact in scenarios A, B, and C in Mio US$ Local Share Cost Reduction (cumulated) 2012 2015 2020 2025 by 2025 by 2026 Scenario A 30 193 916 1,498 25.7% ∼16% Direct 20 125 571 946 Indirect 10 68 344 551 Scenario B 61 465 2,163 3,495 30.6% −16% Direct 39 251 1,167 1,959 Indirect 22 213 996 1,535 Scenario C 368 2,803 14,277 45,226 56.6% ∼40% Direct 206 1,403 6,999 21,675 Indirect 162 1,401 7,278 23,551 Source: Ernst & Young and Fraunhofer 2010. The impact in terms of labor generation would be a permanent workforce of 4,500 to 6,000 local employees by 2020 under scenario B (for more information on estimating employment generation, see box 6.1). In contrast, in scenario C in 2025, the number of permanent local jobs could rise to between 65,000 and 79,000 (46,000 to 60,000 jobs in the construction and manufacturing sector plus 19,000 jobs in operation and maintenance). Additional impacts for job creation and growth of GDP could come from export oppor- tunities for CST components. Exporting the same components that are manufactured for local markets to the European Union, United States, or MENA (2 GW by 2020, 5 GW by 2025) could lead to additional revenues of more than US$3 billion by 2020 and up to US$10 billion by 2025 for local CST industries. Local Manufacturing Capabilities in South Africa The Potential CST Value Chain in South Africa2 Based on an in-depth analysis of the main CST related companies and sectors in South Africa—assessed were the glass, steel and allied industries, electronics, and cable man- ufacturing industries, as well as engineering consulting and project management and EPC ï¬?rms, in order to determine the respective component-speciï¬?c potential for local manufacturing (for details see table B.19 in Appendix B)—a SWOT analysis of RSA’s potential CST value chain is shown in table 6.4. Concentrating Solar Power in Developing Countries 67 Box 6.1: Estimating employment generation of CST development One of the main justiï¬?cations for providing ï¬?nancial incentives not only to CST, but to emerging energy technologies in general, is the employment generated by the speciï¬?c energy sector. The actual amount of employment generated, however, can be estimated in different ways, mak- ing simple comparisons between studies of employment generated by a particular incentive framework potentially misleading. A recent World Bank paper by Robert Bacon and Masami Kojima (2011) describes the various measures of employment generation that are widely used and discusses the deï¬?nitions and methodologies used. The paper compares for example approaches focusing on (a) estimating the incremental employment created by a speciï¬?c project vs. (b) evaluating the total employment supported by an energy subsector at a moment in time; (c) evaluating the incremental employment effects of different forms of a stimulus program in which the energy sector is one possible recipient of government spending; or (d) comparing the employment creation of alternative energy technologies to achieve the same goal, whether it be the amount of power delivered or million dollars of expenditure. Generally the paper categorizes employment generated as either direct (those employed by the project itself), indirect (those employed in supplying the inputs to the project), or induced (those employed as a result of spending from the incomes of the direct and indirect employment), while a further distinction is made between employment for construction, installation, and manufacture (CIM), and employment for operation and maintenance (O&M). This report relies on studies that capture both the direct (project associated) as well as indirect (resulting from increased local manufacturing) employment. Table 6.4: SWOT analysis of CST value chain in South Africa Strengths Weaknesses • High growth in electricity demand resulting in substantial • Sensitivity of local currency investments in the energy sector • Deï¬?cient transport and energy infrastructure • Low labor costs • Administrative barriers and delays • Diversiï¬?ed industry and strong ï¬?nancial institutions • Shortage of skilled employees and insufï¬?cient training • Well-regulated public sector ï¬?nances of workforce • Comparably high DNI • Scarcity of ground water resulting in cooling and wash • High manufacturing capabilities for float and bend glass, water limitations as well as for glass coatings • Strong presence of large power plant equipment manufacturers with signiï¬?cant manufacturing facilities • South Africa hosts some of Africa’s largest steelworks and electrical cable manufacturers • Well-established supply industry—three of Africa’s largest EPC companies • Highly reputable R&D institutions and universities staffed by highly rated scientists and engineers Opportunities Threats • Renewable Energy FiT encouraging CST activities • Restrictive labor regulations • CST project pipeline of up to 5 GW, indicating high potential • Difï¬?culties regarding access to ï¬?nancing of CST implementation • Lack of CST track record • Export potential to Sub-Saharan countries • Lack of bankable PPAs for renewable energy projects • South African leadership in CRS technologies in the long • Energy policy uncertainty regarding the role of IPPs in the term in case of successful implementation renewable energy sector, as well as power sector reform • High potential for cost-effective CST component manufacturing • Governmental support for potential CST component • Attractiveness to external investors, developers, and manufacturers unclear manufacturers by large market demand • Competition with other emerging countries • Improvement of energy security Source: Fichtner 2011. 68 A World Bank Study Box 6.2: Illustrative industrial development in RSA: automotive industry The potential of local industries in South Africa to develop CST activities is conï¬?rmed by the phenomenal success of the automotive industry in South Africa established in the 1920s, which manufactures 83 percent of Africa’s vehicle output (DTI, State of the Automotive Industry Report, September 2003), employs more than 200,000 people (NAAMSA Statistics), and has a local content ratio of at least 60 percent, meaning there are signiï¬?cant beneï¬?ts to the local downstream industries, such as the ï¬?tting and turning factories within South Africa (NAAMSA statistics). Most importantly, the great majority of the more than 200 component manufacturers are South African companies. Several lessons learned are identiï¬?able from the automotive sector experience that could be rather valuable for CST manufacturing in South Africa, including the following: (1) Lack of bank ï¬?nancing or fundraising might inhibit the industry’s growth: The understanding of the ï¬?nancing of CST projects is still low in South Africa. The raising of ï¬?nance on the local market could be a challenge. (2) CST development might be more capital intensive than automotive sector investments. It would be difï¬?cult for the state to ï¬?nance a CST project without adversely affecting its sovereign credit rating. (3) There is no clarity on the administrative requirements yet for CST projects from the Depart- ments of Public Enterprises and Energy. (4) Despite the preliminary research that has been done on CST technologies, the CST indus- try is still in its infancy in South Africa. It will take several years before the knowledge of CST technology is widespread and able to sustain CST plants locally. (5) Clarity on the contribution of CST to the power generation mix is required. The IRP2 has allocated a ï¬?gure for renewable power generation that is being contested by most orga- nizations. Finality of this issue is required so as to send a signal to potential CST power plant developers. (6) Clarity on the role of IPPs in the power sector is urgently needed. Most of the people inter- viewed as part of this research have indicated that IPPs are expected to drive investment in future power plants. The power sector regulatory framework needs to be clariï¬?ed urgently by the Department of Energy in order to give investment signals to investors. Source: Fichtner 2011. Potential for Local Manufacturing As in the MENA region, the uptake of local manufacturing capabilities will be partly driven by major international CST industry players that have already established a pres- ence in South Africa and are assembling land, organizing permits, and developing local partnerships, in order to prepare themselves to get involved on a signiï¬?cant scale in large-scale CST projects in South Africa. The report has analyzed the status quo of the manufacturing capacity for CST com- ponents and the capacity to provide CST-related professional services, including EPC services (an overview is provided in table B.20 in Appendix B). The overall current pro- portion of local manufacturing for power plant projects is expected to be up to 60 per- cent, depending on whether speciï¬?c CST components—for example, receiver tubes, HTF pumps, and swivel joints—can be locally developed and manufactured. For the “stagna- tion scenario,â€? the local share is expected to be considerably lower for construction and components. Under scenario C—the accelerated scenario—the local share in some projects could increase further. Local mirror and receiver production is seen as starting as early as 2015 Concentrating Solar Power in Developing Countries 69 for the acceleration scenario, which would also see the local production of other special- ized, high-precision steel accessories for CST applications. Beyond 2020, the share of local manufacturing would increase even further because of more technology transfers and knowledge sharing through the realization of more CST plants in South Africa, since the learning effect is expected to fully play out around this time. This would also lead to a drop in the cost of locally manufactured CST components because of technological advancements, economies of scale, and competition in the CST component manufactur- ing sector. The modeling for the local share of manufacturing does not include the model- ing of local content requirements set out by the South African government, which would require foreign contractors to procure some material locally. A stable market and large market demand, as well as incentives for investors to venture into the renewable energy sector, will influence many investment decisions on the local production of CST components. Roadmaps for the Development of Local Manufacturing of CST Components in South Africa Figure 6.4 identiï¬?es potential routes for the development of local manufacturing capaci- ties for glass mirrors in the short (up to 5 years), medium (between 5 and 10 years), and long term (beyond 10 years), se ing out the main milestones required to provide both the local and export market. A roadmap for metal structures can be found as ï¬?gure B.3 in Appendix B. Potential Economic Beneï¬?ts of Developing a CST Industry in RSA New CST projects in South Africa will add valuable economic beneï¬?ts to the country’s economy and could support signiï¬?cantly the industrialization of South Africa and Sub- Saharan Africa, as well as the political endeavor of creating jobs. The creation of jobs will Figure 6.4: Potential roadmap for the production of glass mirrors in RSA Status Quo Short-term Mid-term Long-term Overall goal Technology development Adaptation of Adaptation of Single float glass Single float glass Upgrading of Application of One or two large High availability of production lines of the production factory(s) are factory(s) are additional alternative suppliers of white raw material and main producer for line of main adapted to upgraded for production materials & float glass and production of white CSP flat glass and producer for produce CSP flat flat mirrors line(s) for designs (e.g., several mirror float glass by one coating for the CSP bent glass (coating) bending process polymers, thin manufacturers in major provider to required production mirrors gas, aluminum) RSA produce CSP the big local quality mirrors at automotive industry competitive prices No current Supply of white glass Provision of highly precise Supply of white glass Provision of highly precise Exploring capacity production of and flat mirrors for CSP bent mirrors for a major and flat mirrors for bent mirrors in RSA for of all types of mirrors with CSP in RSA possible for a part of RSA demand heliostats and demand and surplus mirrors in RSA quality. major part of RSA Fresnel in RSA for demand demand and surplus Business development Independent One major producer production of CSP of white float glass Foundation of joint mirrors in RSA. with high capacities ventures Newly emerging mirror companies Automotive industry Positive spill-over effects and strong increase set track record of Acquisition of Comprehensive Investment in High level of licenses training of upgrade of sophistication is on other glass sectors of overall sectoral quick industry (e.g., PV) potential creation employees production lines reached Important R&D Strong focus on Applied Techniques and Patented innovations in Growing of sector with existing R&D in the field of research materials reflector designs & intellectual property cooperation for a reflector design, accompanying adapted to maintenance equipment with regard to CSP CSP pilot project coatings & ongoing projects specific needs in RSA mirrors. Profit for maintenance & testing plants and resources innovations Policy framework & market development Establishment of Strategy funds for National targets for industrial upgrade CSP industry are still institutions/ Clear political goals associations to are provided to be agreed upon regarding industrial Consolidated define and policy and exports FiT have recently national support RSA R&D Large number of been adjusted, but strategies for and funding R&D competence are still to be agreed industrial framework Focused support for clusters created industrial upon developments and energy Long-term, stable Intense trade of CSP Growing export of development of CSP Ongoing discussion Favorable tax rates mirrors with RSA CSP mirrors from RSA mirror industry targets defined policy framework exist for mirrors on implementation is implemented neighboring countries of new single-buyer Continuous and Substantial CSP Definition of long- Growing CSP Growing level of confidence Minimum of 100 MW of Minimum installed stable growth of CSP project pipeline term CSP objectives pipeline in CSP technology installed capacity per year capacity of 2 GW market in RSA Source: Fichtner 2011. 70 A World Bank Study enhance the number of people with disposable income, which means an increased pur- chasing power of goods and services, which in turn increases the Foreign Direct Invest- ment (FDI) by foreign companies wanting to take advantage of the improved disposable income in South Africa. The socioeconomic and foreign trade impacts from CST plant development and com- ponent manufacturing in South Africa were analyzed based on a multistage modeling approach incorporating component speciï¬?cations, based on technology requirements, as well as country and project-related assumptions for local manufacturing of components and plant construction.3 The model applied used a cost build-up approach, which con- siders the effect of cost, economic and job effects on a component by component basis. The approach considered the same three scenarios as for the MENA region including scenarios, stagnation, and acceleration. The numbers indicated below are modeled for individual 100 MW reference CST plants using different technologies. DIRECT AND INDIRECT ECONOMIC IMPACTS Direct and induced economic impact values were calculated for each of the three scenarios using NREL’s Jobs and Economic Development Impact (JEDI) model4 and are depicted for a single 100 MW plant in table 6.5. In addition to the local manufacturing of components and the construction of CST plants, O&M services will also have a con- siderable positive impact. Direct economic impacts are related to the design, construc- tion, operation, and maintenance of the CST power plants. Induced effects are economic impacts because of increased demand in the supply value chain, as well as multiplier effects resulting from increased disposable income. Table 6.5: Estimated economic impacts for different CST technologies Stagnation Base case Acceleration scenario scenario scenario Parameter CST technology (EUR million) (EUR million) (EUR million) Estimated PTC without storage 140 180 280 Direct and induced economic PTC with storage 374 412 475 impacts over the project life cycle (project development, CRS without storage 182 230 334 construction, O&M phase) CRS with storage 358 392 448 Source: Fichtner 2011. IMPACT IN TERMS OF LABOR GENERATION O&M services for CST plants will add a considerable number of jobs over a longer period once a particular plant is constructed. Wages and the number of employees were adapted to South Africa’s lower wages and low mechanization of tasks, leading to more workers being employed over the lifetime of the plant. The increasing use of automated plant condition monitoring systems in power plants over time could, however, lower the number of jobs created during the O&M phase. The estimated results of the job impact assessment per single 100 MW plant are given in table 6.6. TRADE IMPACT With regard to the trade impact of CST component manufacturing in South Africa, the model is based on the assumption that exports will only take place if local demand exists Concentrating Solar Power in Developing Countries 71 Table 6.6: Estimated job creation up to 2020 for different CST plant technologies Stagnation Base case Acceleration Parameter CST technology scenario scenario scenario Estimated number of jobs created PTC without storage 956 1,257 1,479 over the project lifecycle (project PTC with storage 1,023 1,480 1,662 development, manufacturing, construction, O&M) CRS without storage 867 1,107 1,337 CRS with storage 945 1,330 1,592 Source: Fichtner 2011. in the region. Respectively, the modeling for this aspect considered only scenario C, under which components like mirrors or receivers are exported to markets in the Euro- pean Union, United States, and MENA. If industry competition increases and costs of components are reduced after 2020, exports are expected to begin soon after 2020. In such a scenario, labor generation and direct economic impacts would increase signiï¬?- cantly. It is expected that after extrapolating the CST capacity curve for the “acceleration scenarioâ€? beyond 2020, more than US$3.6 billion could be earned by exporting CST com- ponents to CST projects in Sub-Saharan Africa and the global market by 2030. Notes 1. This section is based on the report of Ernst & Young and Fraunhofer 2010. 2. This section is based on the Fichtner report 2011. 3. Further assumptions included the following: • The job creation impact assessment has been done on an economy-wide basis. • The Jobs and Economic Development Impact (JEDI) model developed by the National Renew- able Energy Laboratory (NREL) of the United States has been used as reference for this study, but the input ï¬?gures have been changed to suit South Africa. • Effects of an internal CST market growth are considered to be linked with the export of CST components to the world market, such as to other Sub-Saharan African countries. • Scenarios cover the different cases of market development that will have different implications on the economic beneï¬?t and implementation of local supply and component manufacturing in factories of South Africa. • The JEDI model has been used to analyze the impacts for both the PTC and CRS technologies, with and without thermal storage. • The capacity factors assumed are less than 30 percent without thermal storage and 56 percent with storage. • The basis of the modeling is the impacts accruing from one CST plant, which is 100 MW. • The level of job mechanization has been taken to be low. • The DNI ï¬?gures for the Northern Cape Province in South Africa have been used for modeling. • The job market in South Africa is highly influenced by low labor costs, limited availability of skilled workers, and lower productivity of the workforce. As a result, twice as many workers as needed are used for construction. Low worker productivity is due to low mechanization of construction-related tasks in South Africa’s construction industry. The South African gov- ernment has outlined its intention of creating jobs in its New Growth Path (NGP) economic policy. Labor Intensive Construction (LIC) methods are recommended for use by the South African Government on all large-scale projects. 4. Here a link to NREL’s JEDI website and some information would have to be provided. CHAPTER 7 Assessment of Procurement Practices T his chapter describes and analyzes various tendering models, practices, and the bid selection criteria typically used for CST projects based on current information avail- able from the developers and utilities in developed markets, and then provides recom- mendations on tailoring these practices, criteria, and PPA structuring for developing country markets to help facilitate business transactions for CST projects. Recommenda- tions are provided for key elements of each subtopic.1 Tendering Models and Practices The procurement process should be examined in the context of the type of solicitation that is desired. Solicitations can be grouped into two main types: power procurement and project development. Power Procurement involves the purchasing of power by a regulated or public sector utility. This is a hands-off approach where the solicitor does not get deeply involved in the project details. Project Development, by contrast, requires signiï¬?cant involvement and expertise from the solicitor. The characteristics, as well as the advantages and disadvantages of each, are highlighted in table 7.1. Table 7.1: Solicitation types summary Solicitation types Power procurement Pros: Cons: Simpliï¬?ed role for solicitor—no detailed engineering or Potentially higher ï¬?nal cost because of mark-ups in construction requirements generated value chain Minimal expertise in project development needed Little control over project Project development Pros: Cons: Increased control over project structure and implementation More time and effort from solicitor necessary to develop bid packages, evaluate bidders, and oversee construction and implementation Potential for lower cost because of fewer steps in value chain Signiï¬?cant expertise in project development required Source: NOVI Energy 2011. Once the motivations for the procurement are established, the next step is to deter- mine the procurement process that will be used to implement the project. Options include procuring by Sole Source or by Competitive Bidding. Sole Source procurements involve selecting one contractor to perform the scope of work without holding a com- petitive bid. This is prevalent in the industry in the form of conglomerate companies 72 Concentrating Solar Power in Developing Countries 73 taking on multiple roles in a project (owner/developer/EPC). Competitive Bidding is the alternative to Sole Source where requests for proposals (RFPs) are circulated, and multiple bidders respond with proposals. Each of these methods has been used in the past for CST and other renewable energy projects, and each has its advantages and dis- advantages as summarized in table 7.2. Table 7.2: Procurement methods summary Procurement methods Sole source Pros: Cons: Minimal time spent on the selection process Lack of competitive pricing that may result in higher project cost Repeated use may prevent new entrants into the industry Competitive bidding Sealed bidding Pros: Cons: Competitive pricing Potential to under-design systems to satisfy low price, which may affect performance and longevity Transparency Inability to discuss complex procurements to make sure bid offering covers solicitation requirements Less time consuming than Open Bidding Open bidding Pros: Cons: Competitive bidding of the entire construction Bid clariï¬?cations and negotiations can be very time contract provides the lowest cost for the design consuming requirements speciï¬?ed Provides the best assurance that bid content meets RFP requirements and is not over/under designed Source: NOVI Energy 2011. The next step in the procurement process is to determine the contract structure that will be used for the procurement. Although there are numerous options for con- tract structuring, contracts used in renewable energy projects can be grouped into two broad categories: EPC Contracts and Multiple Contracts. The main character- istic of an EPC contract is that it offers protection to the owner from performance and/or cost overrun risks by bundling multiple services into one contract with these risks taken on by the contractor. However, this comes at the price of a risk premium charged by the EPC contractor. The Multiple Contracts approach minimizes the risk premium, but requires the owner to have expertise in managing multiple contractors to deliver the plant on time and within the budget and requires the owner to bear most of the risk. Pricing Structure (table 7.3) also plays an important role in the procurement process. Pricing structures can be manipulated to shift risk from the owner to the contractor or vice versa, depending on the needs of the various players involved in the project. Pric- ing structures used in the renewable energy industry (presented in ï¬?gure 7.1) include ï¬?rm-ï¬?xed-pricing, time-and-materials pricing, and hybrids of the two that are meant 74 A World Bank Study Table 7.3: Pricing structure summary Pricing structure Firm-ï¬?xed-price Pros: Cons: Developer-owner completely protected from cost overrun risk Highest risk premium from contractor may lead to highest overall project cost Fewer contractors may be willing to bid with this type of pricing structure because of unwillingness to take on risk Quality of subcontractors and products may be reduced in order to minimize cost overruns Time-and-materials Pros: Cons: No risk premium; therefore, potential for lowest project cost Highest cost overrun risk, no deï¬?ned cap on the expenses incurred by the contractor No incentive for the contractor to stay within a project budget Hybrid pricing Pros: Cons: Allows optimal balancing of cost overrun risk between parties Some level of risk premium will be included in project cost Maintains incentive for contractor to stay within budget Quality of subcontractors and products may be reduced in order to minimize cost overrun Source: NOVI Energy 2011. to reallocate risk between the parties to accomplish certain objectives (such as incentive alignment). Renewable energy based incentives are usually designed to achieve certain key policy goals and are usually developed in consideration with their se ing. Renewable energy incentives affect the procurement behavior of utilities and in turn influence implemen- tation of renewable energy projects. The schedule sensitivity of expiring incentives and availability of ï¬?nancing, as well as the mitigation of the numerous risks inherent in renew- Figure 7.1: Contract type characteristics Potential for low cost Low risk of cost overrun Owner control of scope Open book EPC Closed book Open book Open book Multiple EPC major eqp., conceptual, contracts closed book closed book BOP EPC Source: NOVI Energy 2011. Concentrating Solar Power in Developing Countries 75 able energy projects, also influence the procurement and implementation of CST projects in developing nations. Bid Selection Criteria The choice of bid selection criteria is critical to the success of the procurement process. Effectively designed criteria help convey the needs of the solicitor and allow bidders to make optimal tradeoffs when developing project proposals. Multiple categories of bid selection criteria were considered for the planned and implemented CST projects, including cost-based, feasibility-based, value-based, and policy-based. Any one of these categories taken alone is insufficient to ensure an optimal match between the proposed projects and the solicitor’s needs. Given the limited experience on bid selections in developing countries analyzed in this report, solicitors should be allowed to consider a range of project a ributes and select the project that represents the best combination of tradeoffs for the solicitor’s needs, by varying the weight applied to each factor. Thus, a recommended option for bid selection criteria design for CST projects in developing countries would be the weighted matrix evaluation approach. The weighted matrix evaluation method also allows the solicitor to more clearly convey their needs by way of published matrix weights as part of the RFP, thus increasing the likelihood that bidders will make appropriate tradeoffs. Without an advanced notice of bid matrix weights, bidders with the capability to provide an optimized proposal may fail to sub- mit it because they would not know that it was, in fact, an optimal balance of the solicitor’s needs. Minimum recommended criteria from each subcategory that should be included in a weighted bid matrix for CST projects in the case study of developing countries are provided in box 7.1. The weights should be selected by each individual solicitor to best reflect the relative importance they place on each factor, and therefore no weight recommendations are provided in box 7.1. Box 7.1: Recommended bid selection criteria for CST in developing countries Cost-based Level of concessional ï¬?nancing Feasibility-based Company/team experience* Company ï¬?nancial stability* Technology maturity Interconnection feasibility Site control Environmental approvals Ability to raise ï¬?nancing Levelized cost of electricity (LCOE) Policy-Based Speed of implementation (schedule) Value-Based (optional) Source: NOVI Energy 2011. *These criteria are optional as separate requirements if “Ability to Raise Financingâ€? is an included criterion. 76 A World Bank Study Cost-Based If a FiT is the primary incentive granted in a particular jurisdiction, choosing the lowest level of concessional ï¬?nancing as the cost-based criterion can be recommended. Since the payment to the winning bidder under a FiT is set regardless of the cost of their project (“guaranteed payment rateâ€?), using a cost-based criterion, such as lowest up-front CAPEX or LCOE to choose the winning bidder would not be effective. The result of using one of these criteria would be that all bidders would understate their up-front and/or O&M costs so that their bid would appear to be the lowest, knowing that they would receive the guaranteed payment rate regardless of the cost they report. This incentive misalignment makes it difficult (if not impossible) to select the project with the lowest cost. Evaluat- ing bids based on the lowest level of concessional ï¬?nancing provides an alternative that minimizes this issue. Bidders will want to use the highest level of concessional ï¬?nancing possible to maximize their project returns. However, they will want to use the lowest level in order to be selected as the winning bidder. This healthy competition will serve to minimize the likelihood that a bidder will understate the level of concessional ï¬?nanc- ing required. Use of this criterion will help maximize the beneï¬?t from the concessional ï¬?nancing available through organizations offering such ï¬?nancing. The use of this crite- rion should not affect the a ractiveness of the procurement to potential bidders. Bidders will be a racted to the procurement if the FiT is high enough to make a project proï¬?t- able. Requiring bidders to use the lowest level of concessional ï¬?nancing possible will just change the way they structure their project. It is worth noting that if the FiT were structured as a “cost-plusâ€? payment, where it pays a set premium over the selected bidder’s LCOE, this would reduce the incen- tive for bidders to understate their costs and make the LCOE measure more useful as a cost-based bid selection criterion. This could be a consideration of incentive design. However, this solution is not without its drawbacks. Structuring the FiT as “cost plusâ€? may make it less desirable for the more cost-efficient bidders, since their lower costs will no longer result in a greater proï¬?t. For example, the level of the FiT could be set based on the understanding by the tariff se er (for example, a regulator) of what an average plant of the type considered should cost to set up and operate. Since the FiT is ï¬?xed for all bid- ders, the regulating body should pick this average value (or somewhere above the low- est value) because they do not want to excessively limit the number of bidders who will ï¬?nd the tariff a ractive. In the case of a ï¬?xed, average-cost FiT, the lowest cost generator will realize a greater proï¬?t from the FiT than an average cost generator, incentivizing the low-cost generator to develop as many projects as possible (good for the country). If a “cost plusâ€? tariff were implemented, both the low cost and average cost generators would have a similar incentive to participate. Another potential option is that taken by India’s JNNSM bid selection criteria. The JNNSM guidelines contain a provision that requires bidders to propose a discount to the offered FiT. Using these proposed discounts, the solicitor chooses the projects equal- ing the desired capacity with the largest discount offered. While it is not a method of determining the underlying cost of the project or selecting the bidder with the lowest cost structure, it results in lower-priced electricity for customers, as long as the winning bidders can actually deliver the bid capacity at the respective discount they offer. This method would only work, however, if bidders are offering more capacity than desired, because otherwise, the risk of nondelivery can undermine the targeted policy goals regarding the total installed capacity. Concentrating Solar Power in Developing Countries 77 Feasibility-Based Consideration of feasibility-based criteria is critical to ensure that time and money are not wasted by selecting projects with a low likelihood of success. Company and team experience should be considered, since it has a direct effect on the likelihood of project success. If a similar project has been successfully completed by the team, the chances of their completing the next project successfully are increased. Financial stability of the bid- der is also important to assure that the project won’t be jeopardized by bankruptcy and/ or other ï¬?nancial issues with the project developer. While CST technology is constantly evolving and improving, some consideration should be given to the maturity of the proposed technology to minimize risk. The weight applied to this factor can be small if the solicitor feels that the beneï¬?ts of improved tech- nology efficiency outweigh the risks of successful implementation. It is recommended that early phases of CST program implementation for a given country place a higher weight on technological maturity to ensure that the program has a successful start. Once several successful projects have been completed and the country has experience imple- menting CST projects, they should consider reducing the weight of technological matu- rity. This will allow for newer, more efficient technologies to be employed and reduce the average capital cost per MW and O&M expenses (and thus the LCOE) of the indus- try. A failure of a new technology would not be as damaging to the program after it has already been implemented in other projects, since it would be if one of the ï¬?rst projects had failed. This appears to be the approach taken by India in its JNNSM. The techni- cal requirements state that during Phase I only CST technologies “which have been in operation for a period of one year or [. . .] for which ï¬?nancial closure of a commercial plant has already been obtainedâ€? will be considered. While it is not explicitly stated in the documentation, the notice that these requirements apply for Phase I, could infer that less mature technologies may be eligible for the Phase II implementation. Some consideration should be given to the ability to raise ï¬?nancing. An assess- ment will have to be made regarding the project’s “bankability.â€? Factors, such as the types of contracts and pricing used (for example, Full-Wrap EPC with Firm-Fixed- Price vs. Multiple Contracts with Time-and-Materials), the maturity of the technology, and the security of the off take agreement (resulting from a stable legal and regula- tory structure), will help determine the ability to secure project ï¬?nancing. The solicitor should also consider any existing commitments from debt or equity providers and their terms and conditions. If a project proposal shows that it can raise ï¬?nancing (that is, the project already has ï¬?rm debt and equity commitments), the above criteria regarding team experience and company ï¬?nancial stability can be considered optional. This is because equity providers and lenders typically go through substantial due diligence to examine team experience and company ï¬?nancial stability before agreeing to provide capital for a project. While LCOE is typically used as a cost-based measure, the previous discussion high- lighted why it should not be used as one in the case of a procurement offering a guaran- teed payment rate (FiT or generation-based incentive), as is the case in Algeria and South Africa. However, it can effectively be used as a feasibility-based criterion to under- stand if the project developer will be able to implement the project at the cost reported. By requiring bidders to submit their estimated LCOE, the solicitor will be able to use its previous experience, an outside contractor (such as the owner’s engineer), or a com- parison with other bidders’ responses to make a judgment regarding the feasibility of 78 A World Bank Study achieving the cost presented. If costs appear to be unrealistically low, the score for this criterion can be lowered. Policy-Based The only policy-based criterion called out in the minimum recommended bid matrix is speed of implementation (“scheduleâ€?). However, more policy-based criteria should be included in the evaluation, depending on the speciï¬?c policy goals of each individual solicitor. The project schedule should be considered by all solicitors, since it will directly affect the achievement of their phased renewable energy policy goals. It is important that the weight of the schedule criterion be chosen carefully by the solicitor. If too much weight is given to the schedule, it can drive up the project cost. It was not prudent to provide a minimum recommendation for other policy-based criteria because of the variability and range of potential policy goals that different solici- tors may wish to factor into their evaluation. Examples include (but are not limited to) local employment and content requirements, preferences for certain technologies and preference for distributed generation over large centralized plants. In considering other policy-based criteria, the solicitor must be careful not to create overly restrictive policy- based requirements. To ensure that the maximum number of bidders respond to the RFP, restrictive criteria, such as minimum domestic content or required use of local labor, should be used sparingly and with caution. In many cases, the project economics will drive the developer to use domestic content and local labor; however, in other cases these restrictive criteria may reduce the a ractiveness of the RFP and discourage qualiï¬?ed bid- ders from responding. Value-Based Value-based criteria are considered optional in the minimum recommended bid matrix criteria for CST projects in developing nations. Examples of value-based criteria include grid stabilization (for example, variability management, known as VAR management), dispatchability and ramp rates (fast start-up), black start capability, and time of day of power supply. While this category can theoretically add value to the bid selection process, if the solicitor does not see value in the characteristics presented or does not anticipate variation among bids, this category might add unnecessary complexity to the bidding and evaluation process. For example, if the solicitor cannot easily quantify the beneï¬?t of VAR reduction or if the nature of the transmission and distribution system in the country necessitates that all of the bids submi ed have black start capability (because of frequent blackouts), it would not be necessary to include these characteristics. Additional Considerations FOSTERING COMPETITION When choosing bid selection criteria, the solicitor should consider each criterion’s affect on increasing or reducing the pool of eligible and willing bidders. Feasibility-based cri- teria are primarily employed to ensure that the probability is high that the project will be successful, enabling the policy goals of the solicitor to be met. If no feasibility-based criteria are employed, the solicitor may end up choosing project proposals with li le chance of success because of the immaturity of the technologies proposed or to devel- oper inexperience. However, if the feasibility-based criteria chosen are too restrictive, they may eliminate many potential bidders and leave the solicitor to choose from only a Concentrating Solar Power in Developing Countries 79 few options. This would most likely result in higher project costs and suboptimal realiza- tion of policy goals. An example of this would be if the solicitor required a high experi- ence threshold for potential bidders, such as experience with multiple projects that have been in operation for several years, using the proposed technology in the proposed scale. REDUCING PROJECT COST As discussed above, it is difficult to control the cost of a project and ensure that the lowest-cost projects are selected when the incentive offered is a ï¬?xed FiT- or generation- based incentive that is not based on the speciï¬?c project’s cost of power (as is the case in Algeria and South Africa). With this incentive structure, the IPP will receive a predeter- mined amount per kilowa -hour regardless of the actual cost to produce power. There- fore, there is no incentive for them to report accurate cost information as part of the bid process. If the FiT were structured as a “cost-plusâ€? tariff as suggested above, this would allow the solicitor to use the LCOE method to choose the lowest-cost project because the bidder would have incentive not to overestimate or underestimate their cost of genera- tion. So unless the incentive structures are revised in the case study countries, it would be difficult for them to choose bid selection criteria that effectively reduce project costs and result in selection of the lowest cost bids. PPA Structuring From the prospective of a project developer (seller), the primary purpose of a PPA is to provide revenue security to the project. A well-crafted PPA assures that if the project is built and operated properly, the electricity it generates will be purchased by an off taker at a predetermined price. Given the large capital cost required and the speciï¬?city of gen- eration assets, such a revenue guarantee is required to secure ï¬?nancing for the project.2 This is especially the case with regard to projects structured with high levels of non- or limited-recourse debt. For balance sheet ï¬?nancing (owner or utility ï¬?nanced), the need for a PPA is dependent on speciï¬?c circumstances.3 From a buyer’s prospective, the primary purpose of the PPA is to provide power supply assurance at the lowest possible cost. Therefore, from a buyers’ point of view, the PPA should warrant that the project is completed on schedule and that it delivers the promised capacity and energy generation. With these primary purposes identiï¬?ed, PPAs were analyzed along with other indus- try feedback to determine the different ways the goals of the seller and buyer could be met by the PPA, and recommendations are provided for the components that should be included in an optimal PPA for CST projects. Considerations when selecting the recom- mended PPA elements included characteristics of solar technologies, as well as aspects that may be applicable to projects in developing countries, such as concerns over trans- mission and distribution system reliability, off taker credit strength and the stability of the government, which will determine whether the executed contracts or promised govern- ment incentives are honored. The recommended elements were chosen to help alleviate these concerns and ultimately make a PPA more a ractive to sellers and ï¬?nanciers, while still meeting the needs of buyers. These recommended elements are shown in box 7.2. Dispatch Agreement Based on the various PPAs reviewed, including both CST and other types of renewable energy generation, the best practice for solar PPAs is to include a ï¬?xed dispatch agreement 80 A World Bank Study Box 7.2: Recommended PPA elements for CST projects in developing countries Fixed dispatch with sharing of curtailment risk Energy payment adjusted using PPI/CPI/exchange rates/LIBOR Time of delivery factors for energy payments Renewable energy credits bundled with energy Seller development security (refunded at commercial operations) Seller performance security (throughout term of PPA) Buyer payment security (throughout term of PPA) Opportunities to rectify default before contract termination Seller repricing or exit on incentive cancellation “Politicalâ€? force majeure provisions Source: NOVI Energy 2011. that allows the project to deliver power whenever the solar resource is available (sub- ject to transmission constraints and energy caps). The risks associated with an intermit- tent resource with a variable dispatch agreement would make it particularly difficult to ï¬?nance the project. As thermal storage systems mature, allowing longer storage times and more control over when the power can be delivered, it is recommended that any CST PPA be structured as a ï¬?xed or “as-availableâ€? dispatch agreement to help minimize revenue risk. The risk allocation of curtailment should be addressed by the PPA as well. If the buyer has responsibility for the transmission system, the buyer should bear at least some (if not all) of the risk that the project would be curtailed because of transmission system constraints or problems. This is especially important in developing countries because of limitations with respect to transmission and distribution systems, and the seller may not have control over those issues. Energy Payment PPAs for projects in developing countries may need several forms of adjustment to pro- tect both the buyer and the seller from large operating costs, exchange rates, and interest rate changes. It can be recommended that adjustment clauses in CST PPAs use indexes that track the cost of labor, if available, since it is typically the greatest component of CST operating costs. If a labor cost index is unavailable, an alternative would be to use a consumer price index (CPI) as a proxy for labor cost. Along with the labor cost index, a targeted PPI should also be used to adjust a portion of the payment if operating costs other than labor may vary signiï¬?cantly over the term of the agreement. The buyer and seller should also consider currency exchange rate adjustments if input costs or debt are in a foreign currency to protect against appreciation of the input cost or debt currency relative to the revenue currency. Additionally, LIBOR-based (or the locally applicable interest rate benchmark) adjustments should be considered if the debt interest rate is variable. If the renewable energy incentive present in the market is a FiT (and therefore not subject to adjustment), the seller can reduce its exposure to exchange rate risk by sourcing equipment from the local area and securing capital denominated in the local currency. Interest rate risk can be mitigated by ï¬?nancing the debt with a ï¬?xed interest rate. Concentrating Solar Power in Developing Countries 81 A ï¬?xed escalation percentage based on historical price inflation can be used; how- ever, the volatility (or standard deviation) of the historical inflation is a key factor. If volatility is high,4 a ï¬?xed escalation percentage would leave the seller exposed to large potential input cost increases, which would make the PPA less a ractive to the seller and potential sources of ï¬?nancing. Algeria, India, Morocco, South Africa, and Tunisia all have moderate PPI/Wholesale Price volatility (see table B.21 (Producer Prices) and table B.22 (World Bank) in Appendix B), which may allow for agreements on a negotiated ï¬?xed escalation percentage, while Egypt and Jordan have relatively high volatility, making adjustments using an index more appropriate for these markets. The energy payment should also be structured to account for the time of day and time of year that the project supplies energy (time of delivery factors). This allows the buyer to communicate to potential sellers the value of energy provided at different times of the day and allows CST sellers to receive the justiï¬?ed premium for their power since it is typically generated during peak demand periods. Capacity Payment None of the PPAs reviewed (including one project with thermal energy storage) con- tained capacity payment provisions, since capacity payments are typically designed to cover the ï¬?xed costs of the project. Solar generating facilities have high ï¬?xed costs with low variable costs (fuel is free) and therefore, if a capacity payment covering the majority of the project’s ï¬?xed costs was included in a CST PPA, the seller would have less incen- tive to produce any energy. However, having some portion of the ï¬?xed costs covered by a capacity payment guaranteed by a PPA would serve the purpose of reducing project risk and increasing the likelihood of securing ï¬?nancing. As a result, the inclusion of capacity payments that pay for a portion of the upfront ï¬?xed costs should be considered by both the seller and the buyer. Renewable Energy Credits Renewable energy credits can either be bundled with the energy sold to the buyer or can be retained by the seller to be sold through third-party contracts or in the spot market. Given the relatively unknown price volatility of green a ributes, it is recommended that any renewable energy credits be sold along with the energy from the project to lock in those revenues and help reduce the overall risk of the project. Non-performance and Default DEVELOPMENT SECURITY The existence of a development security in the PPA is a good incentive to help ensure that bidders don’t overpromise and underdeliver. It also prevents the seller from being granted rights resembling a put option where the seller could walk away from the PPA and sell its output to another off taker if electricity prices increased (abandon the option). In the event of decreasing electricity prices, the seller could “exerciseâ€? the put option and receive the “strike priceâ€? (also known as the PPA energy payment rate) by delivering under the PPA (Lund and others 2009). This would be unacceptable to buyers since their long-term capacity planning would be affected if a seller were to walk away from the PPA and would then have to procure the shortfall at now-higher market prices. Addi- tionally, a development security helps to ensure that the project remains on schedule and becomes operational in time for the buyer to meet customer obligations. 82 A World Bank Study PERFORMANCE SECURITY A performance security would help ensure that the buyer receives the energy prom- ised by the seller throughout the term of the PPA. This security could be provided in the form of a le er of credit from the seller or an escrow account. The escrow account could be funded by withholding a small portion of each monthly payment due to the seller. Once an agreed-upon escrow account cap is reached, there would be no more withholding unless an event occurred that required withdrawal from the account. A drawback of the proposed escrow account is that it builds over time and a large amount would not be available at the start of commercial operations. However, smaller devel- opers may have difficulty securing a le er of credit to provide this security, so alterna- tives such as an escrow account should be considered. While it was not observed in any PPAs reviewed, a combination of an escrow account and a le er of credit could also be used to mitigate these issues. Penalties for non-performance can be viewed as a substitute for easy exit clauses, since they both provide incentives to perform. How- ever, performance penalties are more palatable from the perspective of potential lend- ers, since PPA termination puts debt service in serious jeopardy, while performance penalties (assuming they are not overly severe) will still allow the project to recover and remain in operation. PAYMENT SECURITY In situations where the buyer’s credit quality is weak, it is recommended that a payment security be included in the PPA, similar to the provisions in the JNNSM template PPA. These could include an irrevocable le er of credit and/or an escrow account to provide security that those payments will be made. The escrow account in this case could be funded by diverting some portion of the buyer’s revenues (from other activities not part of this PPA) into the account, up to an agreed-upon cap. This would help reduce the buyer’s default risk and would help secure project ï¬?nancing. EXIT CLAUSES Exit clauses should not allow for too easy of an exit for either party. If the buyer could easily exit from the PPA, ï¬?nancing the project would be difficult. If the seller could easily exit, it would have rights resembling a put option. However, a speciï¬?c exit clause related to the uncertainty around any government incentives should be considered to allow the seller to reprice or terminate the contract if planned incentives are not implemented. In general, it is be er to use performance penalties to provide assurance that the seller meets its obligations than allowing the buyer to terminate the PPA at the ï¬?rst sign of default. Substitution Rights The need for substitution rights in a PPA can be determined by the severity of the exit clauses and performance penalties mentioned above. If the buyer is unwilling to give sufficient time5 for the seller to rectify any issues that lead to a loss of generation or imposes high penalties for non-performance, the contract should include some form of substitution rights to allow the seller to fulï¬?ll its obligations through another means. If the seller is given reasonable time to prevent any defaults prior to the buyer being able to terminate, contract substitution rights would not be necessary. This is the preferred method, since it avoids introducing operational, delivery and reliability concerns that may result from substituted power coming from an uncertain or changing source. Concentrating Solar Power in Developing Countries 83 Force Majeure A good force majeure clause should include separate lists of events that are and are not force majeure to help reduce ambiguity that can be present in this clause. Additionally, force majeure should only be used when events are out of both parties’ control and should not be used to remove the risk from a party that is primarily responsible for the outcome (Lund and others 2009). Force majeure typically includes acts of war and natural disasters. However, events that may occur in developing countries (such as government failure to act, a change in law, or a boyco or embargo of the country by others) should be captured as “politicalâ€? force majeure to protect both buyer and seller. Purchase Obligation While not mandatory, a purchase obligation requiring the buyer to purchase the proj- ect under certain circumstances (for example, prolonged force majeure) would serve to improve the project’s chances of obtaining ï¬?nancing, since it would give potential lend- ers the assurance that the debt service would still be covered if unexpected events occur. However, the value of this type of obligation is entirely dependent on the credit quality of the buyer. Notes 1. This chapter is based on the NOVI Energy report 2011. 2. Assets can be considered “speciï¬?câ€? when they can only be used for one purpose (cannot make other products or products cannot easily be sold to other buyers). Solar generation assets are highly speciï¬?c because they are often located in remote areas with limited off taker options, and are not easily moved. 3. There are many combinations of ï¬?nancing structures that will have different needs with regard to revenue security. If a utility is building its own self-ï¬?nanced plant and “sellingâ€? to them, a PPA may not be necessary. The key point is that the purpose of a PPA is to provide revenue security when necessary, given the speciï¬?c ï¬?nancial and ownership structure of the project. 4. The deï¬?nition of “highâ€? will depend on the risk tolerance of the seller and its ï¬?nancing sources. Developed nations typically have PPI volatility in the range of 1–4 percent (see table B.23 in Appendix A). 5. The length of time that qualiï¬?es as “sufficientâ€? will be different, depending on the cause of the default. The key point here is that if the buyer is unwilling to allow some flexibility regarding the curing of a default, the seller should negotiate for substitution rights to be included in the contract. Source: NOVI Energy 2011. APPENDIX A Overview of Concentrating Solar Thermal Technologies A pplications of solar thermal technologies (including CST) are best suited for regions that experience high levels of DNI. These regions are typically located in dry areas such as deserts, which also have the advantage of plentiful land unused for agricultural or industrial purposes, see ï¬?gure A.1. The Prometheus Institute investigated the use of solar technologies and found that CST technologies are primarily suited for larger scale installations, while PV-based tech- nologies are more suited for smaller scale or distributed generation applications (Grama, Wayman, and Bradford 2008). Photovoltaic panel theoretically are applicable wider geo- graphically, but a certain level of diffused radiation is needed in order to make the elec- tricity generation economically viable. Solar thermal technologies also have geographical limitations and work only in regions that possess a certain level of DNI, not lower than 2,000 kWh/m2/year. The main advantages of CST applications include less intermi ency because of the system inertia; the possibility to use CST in a utility scale operations and the option to integrate thermal storage, thus making power generation possible during extended hours when the sun doesn’t shine. The following factors are typically cited as drawbacks of the current application of CST technologies: â–  CST-based plants are presently characterized with high electricity generation costs, which can be decreased by technological innovations, and economies of scale, that is, volume production, and larger-sized units. â–  Only locations with irradiations of more than 2,000 kWh/m2/yr are suited to a reasonable economic solar thermal performance (Viebahn and others 2008). The four primary CST technologies differ signiï¬?cantly from one another, not only with regard to technical and economic aspects, but also in relation to reliability, maturity and operational experience in utility scale conditions. Given the different levels of tech- nological maturity of the technologies, the biggest experience is accumulated through implementation of projects using the parabolic trough technology and, to a lesser extent, the central receiver application. The main results of the technical assessment of the tech- nologies are summarized in tables B.1 and B.2 in Appendix B In the sections below, relevant design features of each technology are briefly dis- cussed and a review of the status of technological maturity is presented. Parabolic Trough Overview1 Parabolic trough power plants consist of many parabolic trough collectors, an HTF sys- tem, a steam generation system, a Rankine steam turbine/generator cycle and optional thermal storage and/or fossil-ï¬?red backup systems. The collector ï¬?eld is made up of a 84 Concentrating Solar Power in Developing Countries 85 Figure A.1: Markets and applications for solar power Category Small Medium Large Installation size < 10 kW 10 to 100 kW 100 kW to 1 MW 1 to 10 MW 10 to 100 MW > 100 MW Technology mix in each market 100% PV 99% PV, 1% CSP 20% PV, 80% CSP 2007 Share of worldwide solar market 7 GW (84%) 0.7 GW (9%) 0.5 GW (7%) (installed capacity and % of installed capacity) Distributed generation Installation type Central generation Residential Commercial Markets served Utility Base (50%), intermediate (40%), peak (10%) Non-tracking PV PV Non tracking PV based dispatchable CPV Dish-engine Thermal Dispatchable trough based (with storage) tower LFR Installation size <10 kW 10 to 100 kW 100 kW to 1 MW 1 to 10 MW 10 to 100 MW > 100 MW Legend: best suited suitable Source: Grama, Wayman, and Bradford 2008. large number of single-axis-tracking parabolic trough solar collectors. The solar ï¬?eld is modular in nature and comprises many parallel rows of solar collectors, normally aligned on a north-south horizontal axis. Each solar collector has linear parabolic-shaped mirrors that focus the sun’s direct beam radiation on a linear absorber pipe located at the focus of the parabola. The collectors track the sun from east to west during the day to ensure that the sun is continuously focused on the linear absorber (see Figure A.2). An HTF is heated up as it circulates through the absorber and returns to a steam generator of a conventional steam cycle. Figure A.2: Illustration of parabolic trough collectors and sun tracking Source: Radiant & Hydronics 2006. 86 A World Bank Study Figure A.3: Basic scheme of a parabolic trough power plant Solar Field Solar Steam Turbine Superheater HTF Heater Boiler (optional) (optional) Condenser Fuel Thermal Fuel Energy Storage (optional) Steam Generator Solar Preheater Low Pressure Deaerator Preheater Solar Reheater Expansion Vessel Source: Ecostar 2005. The basic scheme of a parabolic trough power plant can be observed in ï¬?gure A.3. The system can be divided into the following three parts: â–  The solar ï¬?eld. â–  The power block (with optional re-heater). â–  The piping and heat exchangers. In this scheme, two optional elements of a CST plant are also represented: the Thermal Energy Storage (TES) and the back-up boiler (BUB), usually working with natural gas. Both of them increase the capacity factor of the system, allowing the plant to operate even when there is not enough direct solar radiation, and sometimes to ï¬?t to a demand curve. Introducing one of these systems allows solar thermal power plants to deliver reliable, dispatchable, and stable electrical energy to the grid. Moreover, it improves the use and amortization of the power block (YES/Nixus/CENER 2010). Parabolic trough solar ï¬?elds are modular; they can be implemented at any capacity, which provides a great versatility. Even so, the optimal capacity for current technology is estimated to be about 150–200 MW. The key components of parabolic trough systems are the receiver tubes, curved mir- ror assemblies (concentrators) and HTF. RECEIVER TUBES The receiver is the component where solar energy is converted to thermal energy in the form of sensible or latent heat of the fluid that circulates through it. It is a critical com- ponent for the performance of the solar power plant because it is where thermal losses are produced. This makes it probably the most important component in the system. Currently, the vacuum tube receiver is the only type of receiver available for parabolic trough power plants. The main providers are Scho and Siemens (Solel Solar Systems), Concentrating Solar Power in Developing Countries 87 but new manufacturers like Archimede Solar (from the Angelatoni Group) and China entrants have also emerged lately. CURVED MIRROR ASSEMBLIES The purpose of the concentrator mirrors is to concentrate solar radiation on the receiver located in the line of focus. Their parabolic geometry and optical reflectivity are extremely important because they are the basic properties that make it possible to concentrate the solar energy efficiently. For this reason the mirrors usually have a support structure to give them the rigidity they require and on which a ï¬?lm of a highly reflective material is deposited. In general, the support structure that provides the rigidity to the parabolic- trough mirror is a metal, glass or plastic plate, while the reflective material is usually silver or aluminum. The material most commonly used to date for collector reflector mirrors is the glass substrate mirror with silver deposition, which reaches maximum reflectivity of around 93.5 percent. HEAT TRANSFER FLUID The purpose of the HTF is to absorb the energy provided by the absorber tube in the form of enthalpic gain by increasing in temperature as it goes through the solar ï¬?eld collector loops. The hot HTF goes to a heat exchanger to heat water and generate steam at a certain pressure and temperature. The solar ï¬?eld outlet temperature is restricted by the HTF properties, and this means that the fluids that can perform these functions are also limited. Experience over the years has shown that by increasing the solar ï¬?eld outlet temperature, the performance of the power block and thereby the whole plant also increases signiï¬?cantly. The commercially proven technology is limited to a tempera- ture of around 400°C, after which, in addition to degrading the fluid, thermal losses increase and the selective coatings also may be degraded. Therefore, there are sev- eral lines of R&D today directed at studying both working fluids and the rest of the components. The fluid currently in use in commercial plants is synthetic oil. Synthetic oil’s advan- tages include a much lower vapor pressure than water at the same given temperature, so pressures required in the system are much lower, which allows simpler facility and safety measures. Furthermore, current oils have responded very well to the current needs of commercial plants, as their maximum temperature coincides with the optimum collector operating temperature. Disadvantages include a high price, and a maximum working temperature below 400°C, which limits the power cycle temperature and, therefore, its electrical conversion efficiency. Molten salt is another alternative HTF. The salt most commonly used in solar applications is nitrate salt with advantages including low corrosion effects on materi- als used for solar ï¬?eld piping, high thermal stability at high temperatures, low steam pressure making it possible to operate at relatively low pressures in its liquid state and its availability and low cost. The main disadvantage is the high freezing point of the salt, which may range from 120° to 200°C depending on the type used. The freeze-protection strategy is very important in this case, and several different tech- niques are necessary to maintain the fluid above a certain temperature: constant cir- culation of salt, auxiliary heating and heat tracing throughout the piping (Kearney and others 2004). 88 A World Bank Study Technological Maturity2 Compared to all other CST technologies, parabolic trough is the most mature. Built between 1984 and 1991, the largest operating group of solar plant systems in the world— with a total capacity of 354 MW—is the Solar Energy Generating Systems (SEGS) I–IX parabolic trough plants, in the Mohave Desert in Southern California now owned by Next Era Energy (owned by Florida Power & Light). In 2007, the ï¬?rst new large parabolic trough power plant, Acciona Solar’s Nevada Solar One, started operation in the United States. Nevada Solar One has a net electric output of 64 MW and is a solar-only Rankine cycle power plant generating approxi- mately 130 GWh of peak power a year (equals a capacity factor of about 23 percent). In 2009, the ï¬?rst large European parabolic trough power plant, Andasol-1, started operation. This was a milestone in the development of the parabolic trough system, since Andasol-1 is the ï¬?rst large-scale, commercial parabolic trough power plant equipped with thermal energy storage. Andasol-1 has a total net electric output of 50 MW and is equipped with a two-tank molten salt storage system with a thermal capacity of 1,050 MWh in combination with an oversized solar ï¬?eld, which enables storage charg- ing during daytime full-load operation, and additional night time operation of up to 7.5 hours. Because of the large storage and a proportionally larger solar ï¬?eld, the 50 MW Andasol I power plant will generate approximately 170 GWh per year, signiï¬?cantly more than the larger Nevada Solar One power plant without storage and with a smaller solar ï¬?eld. Therefore the capacity factor could be increased to above 39 percent. Andasol-1 was the ï¬?rst of around 50 CST plants under construction or development in Spain. Because of the Spanish FiT for CST plants, there was a CST capacity of more than 2,300 MW preregistered in Spain before the end of 2009, with most of the power plants using parabolic trough technology. At present there is approximately 1.2 GW of CST plants in operation divided nearly equally between Spain and the United States. Besides Spain and the United States, there are also several other parabolic trough power plants in advanced development stages throughout the world. An outline of parabolic trough power plants under operation and construction or development is given in table B.3 in Appendix B. Linear Fresnel Overview3 Linear Fresnel power plants consist of many Linear Fresnel reflectors, an HTF system, a steam generation system (if not direct steam generating), a Rankine steam turbine/generator cycle and optional thermal storage and/or fossil-ï¬?red backup systems (see ï¬?gure A.4). The main difference between the parabolic trough technology and the Fresnel tech- nology is the reflector conï¬?guration. Similar to the parabolic trough, the Fresnel collector is designed as single-axis tracking. Therefore, the Linear Fresnel reflectors concentrate sunlight using long flat-plane mirror strips that are grouped in a mirror ï¬?eld close to the ground. The sunlight is focused onto a linear ï¬?xed absorber located above this mirror ï¬?eld and optionally equipped with an additional secondary reflector located above the absorber (see ï¬?gures A.5 and A.6). While the Linear Fresnel concept could use an oil HTF, the conï¬?gurations in devel- opment are mainly based on direct steam generation (DSG), that is, circulating water/ Concentrating Solar Power in Developing Countries 89 Figure A.4: Linear Fresnel system diagram Source: U.S. Department of Energy n.d. steam in the receiver serves as a heat transfer medium (HTF). Hence, a separate steam generation system is not required in the case of DSG. Those Fresnel trough systems are currently operating with saturated steam parameters of up to 55 bar/270°C, but in the medium and long term, superheated steam generation is proposed. Similar to the para- bolic trough system, the Linear Fresnel system can also be operated with HTFs based on molten salt or synthetic oil. The latest development is called the Compact Linear Fresnel Reflector, which is a new conï¬?guration to overcome the limited ground coverage of classical LFR systems. The classical LFR system has only one raised linear absorber, and therefore there is no choice about the direction of orientation of a given reflector. However, for technology Figure A.5: Views of linear Fresnel reflector arrays Source: Morrison 2006. 90 A World Bank Study Figure A.6: Example of a CFLR system source Source: YES/Nixus/CENER 2010. supplying electricity in the multi-megawa range, there will be many linear absorbers in the system. If the absorbers are close enough, then individual reflectors can direct reflected solar radiation onto at least two adjacent absorbers. The additional variable in reflector orientation allows much more densely packed arrays with minimal shading and blocking. The Linear Fresnel technology may be a lower cost alternative to parabolic trough technology for the production of solar steam for power production. The main advantages, compared to parabolic trough technology, are seen as: â–  Inexpensive planar mirror and simple tracking system. â–  Fixed absorber tubes with no need for flexible high pressure joints. â–  No vacuum technology and no metal-to-glass sealing and thermal expansion bellows for absorber tubes for lower temperature conï¬?gurations. â–  Absorbers tubes similar to troughs likely for higher temperature designs. â–  Because of the planarity of the reflector strips and the low construction above ground, wind loads and material usage are substantially reduced. â–  Because of direct steam generation (DSG) within the absorber tubes, no separate steam generator is necessary. â–  Efficient use of land. â–  Lower maintenance requirements are postulated. However, there is also a signiï¬?cant drawback related to the LFR technology. LFR sys- tems suffer from a performance drawback because of higher intrinsic optical losses (ï¬?xed absorber) compared to parabolic trough systems. Different studies evaluated a reduction in optical efficiency of around 30–40 percent compared to parabolic trough technology, which then must be compensated for by lower total investment costs. Technological Maturity4 Fresnel technology is still at an early development level compared to other CST tech- nologies like parabolic trough. That is why there are only a few examples of small scale Concentrating Solar Power in Developing Countries 91 pilot and demonstration projects employing the Fresnel technology. Some existing proj- ects are highlighted in the paragraphs below. The Liddell Power Station is located in New South Wales, Australia. This power plant is coal powered, with four 500 MW GEC (UK) steam driven turbo alternators for a combined capacity of 2,000 MW. In 2004, AUSRA developed the world’s ï¬?rst solar ther- mal power collector system for coal-ï¬?red power augmentation, called the John Marcheff Solar Project. In a ï¬?rst phase, this solar module generated one megawa equivalent (MW) of solar generated steam. This facility was expanded in 2008 with the construction of a second phase, which has a power capacity of 3 MW. Another project, known as Fresdemo, is the ï¬?rst LF demonstration power plant built in Spain. It is located in the PSA, Almería. The demonstration LF system, which has a 100-meter-long collector, generates 1 MWh (peak) and is designed as a modular system. The pilot plant was built by Ferrostaal in collaboration with Solar Power Group and the aim of the plant is to produce evidence that electricity can be generated more competi- tively, proving that Fresnel technology is commercially viable for large-scale projects. It was put into operation in July 2007 and the trial period lasted two years. The results of the operation and testing that took place at the PSA identiï¬?ed several key areas where substantial improvements must be achieved before the technology can be considered ready for commercial deployment. It is unclear, at this stage of development, if the cost reduction of this technology in relation to conventional parabolic trough technology can compensate for its lower solar-to-electricity yearly conversion efficiencies (Bernhard and others 2009). The 5 MW Kimberlina Solar Thermal Power Plant in Bakersï¬?eld, California, started operation in 2008 and is the ï¬?rst commercial solar thermal power plant built by Ausra. Kimberlina uses Ausra’s LF technology. It supplies steam to an existing thermal power plant located nearby. Puerto Errado 1, promoted by Novatec Biosol (now Novatec Solar), is the most recent LF plant put into operation. It has an installed power capacity of 1.4 MW, tak- ing up 18,000 m2 of mirrored area. This plant will generate an estimated annual electric energy of 2 GWh by using the DSG technology. Novatec has developed its own patented collector technology—the collector Fresnel NOVA-1—which has been implemented for the ï¬?rst time in this power plant that was connected to the grid in 2009. The Puerto Errado 1 plant is, to our knowledge, the only commercial grid-connected plant using dry cooling in Spain. Besides projects already operating, there are very few announced Linear Fresnel projects in the pipeline. Novatec Solar has a project pipeline, including an additional Linear Fresnel project, included in the register of the Spanish Ministry of Industry. This project, Puerto Errado 2, which is the second phase of the already operating Puerto Errado 1, will have a total installed power of 30 MW and will also be built in Murcia. The largest pipeline belongs to Areva (Ausra), which has announced a project pipeline with a total power capacity of 337 MW, consisting of several projects located in Australia, Chile, Jordan, and Portugal (Emerging Energy 2010). To some market observers Linear Fresnel technology is increasingly being used for steam generation to meet niche market applications that may not depend primarily on power generation (for example, steam flooding for enhanced oil recovery and steam for industrial process use). 92 A World Bank Study Power Tower Overview5 In power tower (central receiver) power plants, a ï¬?eld of heliostats (large two-axis track- ing individual mirrors) is used to concentrate sunlight onto a central receiver mounted at the top of a tower (see ï¬?gure A.7). Figure A.7: Schematic of open volumetric receiver power tower plant with steam turbine cycle open grid volumetric receiver superheater duct burner (optional) reheater vaporizer economizer turbine generator blower condenser cooling tower feedwater pump Source: Fichtner 2010; Quaschning 2003. The ï¬?eld of heliostats, which all move independently of one another, can either sur- round the tower (Surround Field) for larger systems or be spread out on the shadow side of the tower (North Field) in the case of smaller systems (see ï¬?gures A.8 and A.9). Because of the high concentration ratios, high temperatures and hence higher effi- ciencies can be reached with power tower systems. Within the receiver, an HTF absorbs the highly concentrated radiation reflected by the heliostats and converts it into ther- mal energy to be used in a conventional power cycle. The power tower concept can be incorporated with either a Rankine steam turbine cycle or a Brayton gas turbine cycle, depending on the applied HTF and the receiver concept, respectively. Figure A.8: North ï¬?eld layout mills Source: Mills and others 2002. Concentrating Solar Power in Developing Countries 93 Figure A.9: Surround ï¬?eld layout mills Source: Mills et al. 2002. Major investigations during the last 25 years have focused mainly on four plant conï¬?gurations depending on the applied technology and HTF system: â–  Water/steam solar tower (Rankine cycle) â–  Molten salt solar tower (Rankine cycle) â–  Atmospheric air solar tower (Rankine cycle) â–  Pressurized air solar tower (Brayton cycle) Besides the four mentioned plant conï¬?gurations, liquid metals (mainly sodium) were also investigated as a possible HTF. However, because of different hazards (especially ï¬?re) R&D efforts on liquid metals is currently out of focus. Therefore, only the four main plant conï¬?guration options are described below. WATER/STEAM SOLAR TOWER Water/steam offers the beneï¬?t that it can be directly used in a Rankine cycle without further heat exchange. The production of superheated steam in a solar receiver yields higher efficiencies and has been demonstrated in several prototype projects like the Solar One or CESA-1 projects. However, the operational experience showed some problems related to the control of zones with dissimilar heat transfer coefficients, like evaporators and super-heaters. Difficult to handle were also the start-up and transient operation of the system, leading to local changes of the cooling conditions in the receiver tubes, in particular in the receiver’s superheating section. Because of the abovementioned problems related to superheating steam in central receivers, the ï¬?rst commercial water/steam receiver power plants are producing only saturated steam. The ï¬?rst such plants are the PS-10 and PS-20 power plants built by Abengoa Solar, with 10 MW and 20 MW, respectively. MOLTEN SALT SOLAR TOWER Molten salt mixtures combine the beneï¬?ts of being both an excellent heat transfer and a good high temperature energy storage fluid. Because of a very good heat transfer, the applied heat flux at the receiver surface can be higher compared to other central receiver designs, yielding higher receiver efficiencies. As the molten salt can be stored directly at high temperatures, the speciï¬?c storage costs are the lowest under all CST technologies. This means that molten salt power tower technology, when proven, will be the preferred choice for applications that require a storage component. 94 A World Bank Study Depending on the speciï¬?c composition, the molten salt liqueï¬?es at a temperature between 120°C and 240°C (in the current state of the technology this is the upper end) and can be used in conjunction with metal tubes for temperatures up to 600°C with- out imposing severe corrosion problems. As discussed earlier with regard to parabolic trough systems, the challenge is to avoid freezing of the salt in any of the valves and piping of the receiver, storage and steam generation system at any time. The operating range of the state-of-the-art molten nitrate salt, a mixture of 60 percent sodium nitrate and 40 percent potassium nitrate, matches the operating temperatures of modern Ran- kine cycles. In a molten salt power tower plant, the cold salt (290°C) is pumped from the cold tank to the receiver, where the salt is heated up to 565°C by the concentrated sun- light. This hot salt is then pumped through a steam generator to generate superheated steam that powers a conventional Rankine cycle steam turbine. The solar ï¬?eld is gener- ally sized to collect more power than demanded by the steam generator system and the excess energy can be accumulated in the hot storage tank. With this type of stor- age system, solar tower power plants can be built with annual capacity factors of up to 70 percent. Several molten salt development and demonstration experiments have been conducted over the past two-and-a-half decades in the United States and Europe to test the entire system and develop components. The largest demonstration of a molten salt power tower was the 10 MW Solar Two project located near Bartow, California. ATMOSPHERIC AIR SOLAR TOWER Air offers the beneï¬?t of being nontoxic, having no practical temperature constraints and is available for free. However, air is a poor heat transfer medium because of its low den- sity and low heat conductivity. In a central receiver solar power plant with an atmospheric air heat transfer circuit, based on the so-called PHOEBUS scheme, a blower transports ambient air through the receiver, which is heated up by the concentrated sunlight. The receiver consists of wire mesh, ceramic or metallic materials in a honeycomb structure, and air is drawn through this and heated up to temperatures between 650°C and 850°C. On the front side, cold, incoming air cools down the receiver surface. Therefore, the volumetric structure pro- duces the highest temperatures inside the receiver material, reducing the heat radiation losses on the receiver surface. The hot air is used in a heat recovery steam generator to produce steam at 480 to 540°C/35 to 140 bar. The PHOEBUS scheme also integrates several equivalent hours of ceramic thermocline thermal storage, able to work in charging and discharging modes by reversing air flow with two axial blowers. Current heat storage capacity restrictions lead to designs with a limited number of hours (between 3 and 6). Therefore, higher annual capacity factors can only be reached with backup from a duct burner between the receiver and steam generator. Another option is to use sand as a storage media. How- ever, the heat transfer from air to the sand is poor and the technology has not yet been demonstrated on a larger scale. PRESSURIZED AIR SOLAR TOWER In this concept, pressurized air (around 15 bar) from the compressor stage of a gas tur- bine is heated up (to 1100°C) in a pressurized volumetric receiver (REFOS receiver) and then used to drive a gas turbine. At the moment, the concept needs additional fuel to increase the temperature above the level of the receiver outlet temperature. In the Concentrating Solar Power in Developing Countries 95 future, a solar-only operation at higher receiver outlet temperatures and the use of ther- mal energy storage might be possible. The waste heat of the gas turbine goes to a heat recovery steam generator that generates steam to drive an additional steam-cycle pro- cess. This pressurized air solar tower/CCGT process can reach high efficiencies of over 50 percent. These systems have the additional advantage of being able to operate with natural gas during start-up and with a high fossil-to-electric efficiency when solar radiation is insufficient. Hence, no shadow capacities of fossil fuel plants are required and high- capacity factors are provided. In addition, the speciï¬?c cooling water consumption is reduced in comparison with Rankine cycle systems. Technological Maturity Although power towers are commercially less mature than parabolic trough systems, a number of component and experimental systems have been ï¬?eld tested around the world in the last few years, demonstrating the technical feasibility and economic poten- tial of different power tower concepts. Furthermore, the already operating power tower plants have proven their feasibility on an entry-commercial scale at small plant capaci- ties The most experience has been collected through several European projects, mainly in Spain at the Plataforma Solar de Almería (PSA) and the Plataforma Solucar of Aben- goa Solar near Seville, as well as earlier in the United States (U.S. DOE’s Solar One and Solar Two that have since been decommissioned). An outline of solar tower demonstra- tion projects is given in table B.4 in Appendix B. In 2007, the ï¬?rst commercial power tower plant started operation in Spain. The PS-10 power plant, built by Abengoa Solar, uses saturated steam as the HTF and has a net electrical output of 10 MW. Based on the same receiver concept, the PS-20 plant located in close vicinity to the PS-10 plant has been in commercial operation since 2009 with 20 MW electrical output. Other plants already in operation are the Sierra Sun Tower in California of eSolar, with an electrical output of 5MWe and the Solar Tower Jülich with 1.5 MW. These plants represent demonstration/pilot plants for the latest developments on the basis of super- heated steam (eSolar) and the volumetric air concept (Solar Tower Jülich). A 1.5 MW eSolar plant is currently also undergoing commissioning in India by Acme. The Solar Tres plant (17 MW), with completion expected in 2011, will operate with molten salt as the HTF and storage medium (direct storage). After an intermediate scale up to 10–20 MW of capacity, solar tower developers now feel conï¬?dent that grid-connected central receiver plants can be built up to a capacity of 200 MW solar only units. The largest new solar power tower project currently being constructed is the 392 MW Ivanpah project of BrightSource Energy, Inc. in California. The two dominating solar tower systems being developed and commercialized by several companies are the ones using water/steam and molten salt as HTFs. While the system using atmospheric air as HTF is expected to be commercially available in the near term, further R&D is required for the commercialization of medium- and large-sized solar tower systems based on the pressurized air receiver concept. The main disadvan- tage of the power tower system using the atmospheric air is that the storage option cannot be easily integrated, and will most likely be inefficient because of high thermal losses in air-to-water heat exchangers. An overview of already realized and upcoming commercial-scale power tower projects is given in table B.5. 96 A World Bank Study Dish-Engine Overview6 The dish-engine is unique among CST systems in directly heating the working fluid of the power unit rather than an intermediate fluid to produce electricity. Dish-engine systems consist of a mirrored dish that collects and concentrates sunlight onto a receiver mounted at the focal point of the dish. The receiver is integrated into a high-efficiency engine (the Stirling engine is the most commonly used heat engines because of high efficiency). Solar Parabolic Dish-engine sys- tems include two main parts: a large Parabolic Dish, and a power conversion unit (PCU). Figure A.10: Dish-engine photo with major component identiï¬?cation Source: Bill Brown Climate Solutions 2009. The PCU is held at the focal point of the concentrator dish and includes a receiver, as well as a heat engine and generator assembly for converting the collected thermal energy to electricity (see ï¬?gure A.10). Typically, a high-efficiency Stirling engine is used. Individual units range in size from 3 to 25 kW and are self-contained and air-cooled, thus eliminating a cooling water requirement, which is a signiï¬?cant advantage of Dish Stirling systems. At the same time, an inherent issue with these systems is that electri- cal production ceases immediately upon loss of sun. In that respect, they are similar to solar photovoltaic plants. Currently, no concept for commercial thermal storage has been demonstrated and implemented for dish engine systems. Compared to the other CST technologies, the main advantages of dish-engine sys- tems are as follows: â–  Water usage is limited to operational and maintenance activities (such as mirror washing). â–  It has a ained efficiencies as high as 30 percent in the testing facility at the Sandia Laboratories. â–  Its modularity allows for a range of system sizes, from several megawa s to hundreds of megawa s. â–  Central or decentralized operations are possible with the scale between 3 kW and several 100 MW. Concentrating Solar Power in Developing Countries 97 â–  High energy density, lower land use. â–  Short construction times. The main disadvantages of dish-engine systems are higher investment costs, lack of existing storage and hybridization solutions, and a concern about higher O&M costs because of the large number of the kW-scale engines in a multi-MW installation. The two major components of dish-engine systems are the reflective dish and the receiver, or the PCU. REFLECTIVE DISH The concentrator dish is made up of a parabolic shaped reflector, which concentrates the incident solar irradiation into a receiver located at the dish focal point. The ideal shape of the concentrator is a parabloid of revolution, although most designs approximate this shape by using multiple spherical mirrors. Reflectors used in concentrators consist of a glass or plastic substrate with a thin alu- minum or silver layer deposited over it. The most durable material known to the present is the current silver/glass thick mirror, which reaches reflectivity values typically close to 94 percent (Solar Dish Engine n.d.). However, silvered polymer solar reflectors (thin mirror) are ï¬?nding increasing use in dish concentrator applications (Harrison 2001). An innovative trend toward a new concept that would allow be er optical efficiencies was introduced in the 1990s: the stretched membrane mirror, implemented in the SBP design. The size of the Parabolic Dish is mainly determined by two factors: â–  Thermal power demand of the power block (Stirling engine) in nominal conditions. â–  Wind loads: restricting the economical viability of large installations. POWER CONVERSION UNIT The power conversion unit is the element that absorbs concentrated solar energy and converts it to thermal energy that heats the working fluid (gas) inside the typically 3 kWe to 30 kWe engine. These receivers usually adopt the cavity geometric conï¬?guration, with a small aperture and its own isolation system. In order to carry out this energy transfor- mation, it is necessary to reach a high temperature and high levels of incident radiation fluxes while minimizing every possible loss (Gener). Many different conï¬?gurations of receivers have been proposed, adapted to different HTFs. These conï¬?gurations can be gathered in two main groups: â–  Direct Interchange Receiver (DIR): Fluid absorbs the radiation being directly applied to it. â–  Indirect interchange receivers: There is an additional element, which transforms solar radiation into heat and then delivers it to the HTF through convection. Technological Maturity At the moment, dish-engine systems for large scale applications are considered commer- cially less mature than other solar power generation systems. A number of component and pilot systems have been ï¬?eld tested around the world in the last 25 years, dem- onstrating the technical feasibility and the economic potential of the Parabolic Dish col- lector for small-scale applications and/or remote locations. Dish Stirling systems are under development and prototype testing in the United States and Europe (for example, by such companies as Tessera Solar/SES, EuroDish, and 98 A World Bank Study EnviroDish). In addition, the use of small solar driven gas turbines at the focus of dishes (dish/Brayton systems) has been investigated. This would offer the potential for high- efficiency operation, with lower maintenance requirements than for the Dish Stirling cycle. An outline of Parabolic Dish collector plants realized and/or under operation, is given in table B.6. To date, there are no operating commercial plants based on the Parabolic Dish tech- nology. Tessera Solar—a developer, builder, operator and owner of large utility-scale solar power plants—deployed the SunCatcherâ„¢ solar Dish Stirling system, using the technol- ogy developed and manufactured by the Tessera Solar affiliate Stirling Energy Systems Inc. (SES), headquartered in Sco sdale, Arizona. The company’s ï¬?rst plant, Maricopa Solar, began operations in Arizona in January 2010. The other planned projects, such as Calico (850 MW) reportedly had trouble securing ï¬?nancing and the PPA was lost. The project was in part sold to PV developer, but reserved 100 MW of the phase II implementation for SES’s Dish Stirling technology with the rest (750 MW) consisting of solar PV technology. Power Blocks8 All CST technologies discussed above, with the exception of the dish-engine type, use a power block to convert the heat generated to electricity. The components that make up the power block in a solar thermal power plant are generally equivalent to the compo- nents of conventional thermal power plants. However, certain characteristics of power blocks in CST plants call for speciï¬?c considerations. The incorporation of the Rankine cycle into a solar thermal power plant introduces additional operational requirements as a consequence of the cyclical nature of solar energy. While transients can be minimized transients through the use of thermal storage and use of an auxiliary boiler, daily stoppage is prevalent because of legislative limita- tions on gas consumption or low demand needs at night. Therefore, it is important to keep in mind a series of additional considerations, both in the design of the equipment and in operational practices of the plant. These considerations include: â–  Since the plant is not going to operate 24 hours a day, it is important to utilize high efficiency steam turbine cycles to make the project economically feasible. This leads to larger turbines with optimized feed water heating, in turn result- ing in a reduced solar ï¬?eld size, which translates into a reduction in investment costs, and, therefore, of the cost of the power generated. â–  The thermodynamic cycle can also include a reheat stage depending on the quality of the steam at which it is going to operate. This could improve the effi- ciency and reduce problems of erosion, corrosion and humidity. â–  The annual plant production is affected by turbine start-up time because of the daily starts. Both the daily cyclicality and variations in temperature require spe- cial a ention. One important characteristic of the turbine is the total mass of its components. Optimizing the mass of machine rotors and cladding can shorten start-up time. â–  Another important factor, especially for plants that do not include storage, is the turbine turn-down ratio, which will affect the number of plant operating hours. By being able to operate the turbine at a lower part-load level power gen- eration hours can be gained, although the system is penalized by the reduced efficiency of the turbine at partial loads. Concentrating Solar Power in Developing Countries 99 Thermal Storage Options9 A distinct advantage of solar thermal power plants compared with other renewable ener- gies, such as PV and wind, is the possibility of using thermal energy storage systems that are substantially cheaper than other current systems for storing electricity. Since there are new storage technologies under development to store electricity on a large scale (such as compressed air and utility scale Na-S ba eries), and smart-grids are emerging, the long-term success of CST technology will also depend on the availability of inexpen- sive and highly efficient thermal energy storage systems for solar thermal power plants. The basis, on which the use of thermal energy storage systems is determined for solar thermal power plants, depends strongly on the daily and annual variation of irra- diation and on the electricity demand proï¬?le. The main options for the use of TES are discussed below. Buffering The goal of a buffer is to smooth out transients in the solar input as a result of passing clouds, which can have a signiï¬?cant impact on the operation of a solar thermal power plant. The efficiency of electrical production will degrade with intermi ent insulation, largely because the turbine-generator will frequently operate at partial loads and in a transient mode. If regular and substantial cloudiness occurs even over a short period, turbine steam conditions and/or flow can degrade enough to force turbine trips if there is no supplementary thermal source to “ride throughâ€? the disturbance. Buffer TES sys- tems would typically require small storage capacities (typically 1–2 equivalent full-load hours depending on weather conditions). Delivery Period Displacement Thermal energy storage can also be used for delivery period displacement, which requires the use of a larger storage capacity. The storage shifts some or all of the energy collected during periods with sunshine to a later period with higher electricity demand or tariffs (electricity tariffs can be a function of the hour of day, the day of the week and the season). This type of TES does not necessarily increase either the capacity factor or the required collection area, as only solar heat that would have otherwise been used directly throughout the day is stored for later use. The typical storage capacity ranges from three to six hours of the full operational load. Delivery Period Extension The size of a TES for delivery period extension will be of similar size (3 to 12 hours at full load). However the purpose of the TES in this case is to extend the period during which the power plant operates using solar energy. Such TES increases the capacity factor of the solar power plant and requires larger solar ï¬?elds than a system without storage. The optimal storage capacity is site and system dependent. Therefore, a detailed sta- tistical analysis of system electrical demand and weather pa erns at a given site, along with a comprehensive economic tradeoff analysis, are desirable in a feasibility study to select the storage capacity for a speciï¬?c application. There are a number of storage concepts for CST power plants, which have been either successfully tested and are now commercially available, or which are still under development. An overview on the most promising storage concepts and their status is presented in ï¬?gure A.11. Current parabolic trough systems are “indirect,â€? in that the oil 100 A World Bank Study HTF flowing through the solar ï¬?eld both charges and discharges molten-salt-ï¬?lled stor- age tanks via an oil-to-salt heat exchanger. “Directâ€? systems are those in which the HTF system and storage medium are the same fluid, without an intermediate heat exchange process. Molten salt power towers and parabolic troughs with a molten salt HTF are examples of such systems. Figure A.11: Storage concepts for CST Thermal oil storage Direct Storage tank (PT) Steam accumulator (FT,ST) Molten salt tank (ST) Sensible storage Indirect Storage (temperature change) Molten salt tank (PT) Sand or ceramics (ST) Latent storage (phase change) Ionic liquids PT – Parabolic trough Concrete Combi- FT – Fresnel trough nation ST – Solar tower Chemical storage for DSG – Direct steam generation Phase change material (PCM) DSG - Commercially available Source: Fichtner 2010. Hybridization From an environmental point of view, solar-only conï¬?gurations are the best as only heat from the solar ï¬?eld is used to generate steam. However, as no mature TES solutions are available for all the CST technologies, hybridization is an interesting alternative to increase the capacity factor of the power plants, increasing their commercial viability. Usually, this type of designs allow three operational modes (solar, fossil or hybrid) pro- viding great levels of versatility and dispatchability. Hybridization Options HYBRIDIZATION WITH A FOSSIL FUEL BOILER PLACED IN PARALLEL TO THE SOLAR FIELD. This option can be used with parabolic trough and Lineal Fresnel power plants (see ï¬?g- ure A.2 and ï¬?gure A.8). CONVENTIONAL RANKINE CYCLE WITH SOLAR PREHEATING This concept aims at adding a solar preheater to big fossil power plants in order to reduce their fuel consumption and gases emissions (see ï¬?gure A.12). It has been demonstrated at Liddell coal power plant in New South Wales, Australia. The annual solar fraction (amount of solar energy in the total thermal energy of the plant) is usually lower than Concentrating Solar Power in Developing Countries 101 Figure A.12: Saturated steam hybrid plant conï¬?guration Superheated Steam 2 3 4 Condensate 40–70°C 1 5 Feedwater 140°C 6 1 Solar Field 3 Turbine 5 Deaerator/Feedwater Tank 2 Gas Furnace 4 Air-cooled Condenser 6 Feedwater Pump Source: Novosol. 5 percent. However, solar energy is converted to power with high efficiencies and the investment cost is low, so it can be a relevant option to retroï¬?t existing fossil fuel plant already in operation and introduce CST technologies to the market. No solar energy is lost during start-up and shut-down periods. INTEGRATED SOLAR COMBINED CYCLE SYSTEMS (ISCCSS) These systems consist in integrating solar energy into a combined cycle power plant, as shown in ï¬?gure A.13. They have been primarily considered for parabolic trough collectors, but the characteristics of Linear Fresnel collectors (low cost, low temperature, DSG) made them very relevant for ISCC systems. They can result very effective, in particular if stable and continuous power production is needed. Solar thermal energy is delivered to the Heat Recovery Steam Generator (HRSG) of the combined cycle, thus the steam turbine receives higher heat input than in classical combined cycles, resulting in higher efficiencies. ISCCS beneï¬?t from the high efficiencies of combined cycles: some studies assess annual fuel-to-power efficiencies of about 60 percent. Besides, as the investment cost for gas turbines is lower than for steam turbines, ISCCS are more cost-effective than hybrid solar Rankine cycles. As in conventional Rankine cycle with solar preheating, no solar energy is lost during start-up and shut-down periods. The Martin Next Generation Solar Energy Center is a hybrid 75 MW parabolic trough solar energy plant, built by Florida Power & Light Company (FPL). The solar plant is a component of the 3,705 MW Martin County Power Plant, which is currently the single largest fossil fuel burning power plant in United States. The facility will also be the ï¬?rst hybrid facility in the world to connect a solar facility to an existing combined cycle power plant. It is located in western Martin County, Florida. Construction began in 2008 and was completed by the end of 2010. ISCC plants are also being constructed in Algeria (Hassi R’Mel) and Morocco (Ain Beni Mathar) in collaboration with Abengoa Solar. Abengoa Solar is providing the design and will act as the technician of the solar ï¬?eld. The ISCC of El-Kureimat, in Egypt, is being developed by New and Renewable 102 A World Bank Study Figure A.13: Basic scheme of an ISCCS Option B - Low Pressure Solar Steam Solar Steam Generator Expansion Vessel Low Steam Turbine Pressure Feedwater Steam Fuel Flue Gas Gas Turbine Condenser Waste Heat Recovery System Option A - Low Pressure Solar Steam High Pressure Steam Solar Steam Expansion Generator Vessel Low Pressure Feedwater Preheater Deaerator Source: ECOSTAR. Energy Authority (NREA), and is expected to start production at the end of 2012. Other projects are under development in Mexico (Agua Prieta) and Iran (Iazd). In addition to the options above, there are other lines of research in order to develop other hybrid options. As an example, the company AORA-Solar has developed an advanced solar-hybrid power generation unit. A pilot project was built in 2009 in Kib- bu Samar, in the southern desert of Israel. The system offers a modular solution, com- prising small Base Units of 100 kWe (comprised by heliostat and solar tower with a micro turbine) that can be strung together, building up into a large power plant. When the available sunlight is not sufficient, the system can operate on any alternative fuel source (fossil fuel, bio fuel). Hybridization and Regulatory Framework In Spain, the development of the solar thermal technology has risen because of a favor- able regulatory framework. In addition to a FiT policy, it was regulated the possibility of building hybrid plants. However, the range of hybridization was limited to 12–15 percent (fraction of fossil fuel energy in the total thermal energy of the plant) by the legal frame- work. In the United States, this fraction can reach up to 25 percent. Notes 1. Based on Fichtner (2010) 2. Based on Fichtner (2010). 3. Based on Fichtner (2010) 4. Based on YES/Nixus/CENER (2010). 5. Based on Fichtner (2010). 6. Based on YES/Nixus/CENER (2010). 7. Fichtner (2010). 8. Based on YES/Nixus/CENER (2010). 9. Fichtner (2010). APPENDIX B Tables and Figures 103 Table B.1: Overview of the main technical characteristics of CST technologies Technology Units Parabolic trough Fresnel trough Molten salt solar tower Water steam solar tower Parabolic dish Item {ART} Plant Size, envisaged [MWe] 50–300a 30–200 10–200a 10–200 0.01–850 Plant Size, already realized [MWe] 50 (7.5 TES), 80 (no TES) 5 20 20 1.5 (60 units) Collector/Concentration [-] Parabolic trough (70–80 suns) Fresnel trough / > 60 suns, Heliostat ï¬?eld / > 1,000 suns Heliostat ï¬?eld / > 1,000 suns Single Dish / > 1,300 suns depends on secondary reflector Receiver/Absorber [-] Absorber ï¬?xed to tracked Absorber ï¬?xed to frame, External tube receiver External or cavity tube receiver, Multi receiver system collector, complex design no evacuation, secondary multi receiver systems reflector Storage System [-] Indirect two-tank molten salt Short-time pressurized Direct two-tank molten salt Short-time pressurized steam No storage for dish (380°C; dT = 100K) steam storage (<10min) (550°C; dT = 300K) storage for saturated steam Stirling, chemical storage (<10min) under development Hybridisation [-] Yes, indirect (HTF) Yes, direct (steam boiler) Yes Yes, direct (steam boiler) Not planned Grid Stability [-] medium to high (TES or medium (back-up ï¬?ring high (large TES) medium (back-up ï¬?ring low hybridisation) possible) possible) Cycle [-] Rankine steam cycle Rankine steam cycle Rankine steam cycle Rankine steam cycle Stirling cycle, Brayton cycle, Rankine cycle for distributed dish farms Steam conditions [°C/bar] 380°C / 100 bar 260°C / 50 bar 540°C / 100–160 bar up to 540°C / 160 bar up to 650°C / 150 bar Land requirementsb [km2] 2.4–2.6 (no TES) 1.5–2 (no TES) 5–6 (10–12 h TES) 2.5–3.5 (DPT on the lower site) 2.5–3 4–4.2 (7h TES) Required slope of solar ï¬?eld [%] < 1–2 <4 < 2–4 (depends on ï¬?eld design) < 2–4 (depends on ï¬?eld design) >10% Water requirementsc [m3/MWh] 3 (wet cooling) 3 (wet cooling) 2.5–3 (wet cooling) 2.5–3 (wet cooling) 0.05–0.1 (mirror washing) 0.3 (dry cooling) 0.2 (dry cooling) 0.25 (dry cooling) 0.25 (dry cooling) Annual Capacity Factor [%] 25–28% (no TES) 22–24% 55% (10h TES), larger TES 25–30% (solar only) 25–28% 40–43% (7h TES) possible Peak Efï¬?ciency [%] 22–25% 16–18% 18–22% 31% Annual Solar-to-Electricity [%] 14–16% 9–10% (saturated) 14–16% 15–17% 20–22% Efï¬?ciency (net) Source: Fichtner 2010. a Maximum/optimum depends on storage size. b 100 MWe plant size. c Depends on water quality. Table B.2: Overview of the main commercial characteristics of CST technologies Technology Units Parabolic trough Fresnel trough Molten salt solar tower Water steam solar tower Parabolic dish Item Maturity [-] - Proven technology on large - Demonstration projects, ï¬?rst - Demonstration projects, - Saturated steam projects - Demonstration projects, scale commercial projects under ï¬?rst commercial projects in operation ï¬?rst commercial projects (ï¬?rst - Commercially viable today construction under construction - Superheated steam units) in 2011 - Commercially viable 2011 - Commercially viable 2011 demonstration projects, - Commercially viable 2012 onwards onwards ï¬?rst commercial projects onwards under construction - Commercially viable 2012 onwards Total Installed Capacity (in [MWe] 1,000 7 10 10 (superheated/demo) 1.7 operation Q4 2010) 30 (saturated steam) Estimated total Installed [MWe] 3,000–4,000 200–300 200–400 400–500 500–1,000 Capacity (in operation 2013) Number of Technology [-] high (>10), Abengoa Solar/ medium (3–4), Areva, Novatec medium (2–5) SolarReserve medium (3–4), Abengoa medium (4–5), Abengoa Provider Abener, Acciona, ASC Cobra/ Biosol AG, Sky Fuels, Solar and Torresol others like Solar, BrightSource Solar, Inï¬?nia, SES / Tessera Sener, Albiasa Solar, Aries Power Group, etc. Abengoa Solar and eSolar, Energy, eSolar etc. Solar, SB&P, Wizard Power Ingeniera, Iberdrola, MAN SolarMillenium are planning SolarMillenium, Samca, Solel/ entry Siemens, Torresol etc. Technology Development Risk [-] low medium medium medium medium Investment costs for 100MW [$/KW] 4,000–5,000 (no storage) 3,500–4,500 (no storage) 8,000–10,000 (10h storage) 4,000–5,000 (no storage) 4,500–8,000 (depending on 6,000–7,000 (7–8h storage) volume production) O&M Costs [m $/a] 6–8 (no storage) 5.5–7.5 7–10 (molten salt with TES) 5–7 (water steam, no TES) 10–15 (water steam, no TES) Source: Fichtner 2010. 106 A World Bank Study Table B.3: Parabolic trough power plant projects (Estimated) Thermal energy ï¬?rst year of Peak output storage/ Project name/location Country Developer operation [MWel] dispatchibility Nevada Solar One, USA Acciona Solar Power 2007 74 None Boulder City Andasol I–III Spain ACS Cobra / Sener 2008–2011 3 × 50 Molten Salt Thermal Solar Millennium Storage Solnova I–V Spain Abengo Solar 2009–2014 5 × 50 Gas heater ExtreSol I–III Spain ACS Cobra / Sener 2009–2012 3 × 50 Gas heater Kurraymat Egypt Iberdrola / Orascom 2010 20 (solar) ISCC & Flagsol Ain Beni Mathar Morocco Abener 2010 20 (solar) ISCC Shams 1 UAE Abengoa Solar 2012 100 Gas ï¬?red superheater Beacon Solar Energy USA Beacon Solar 2012 250 Gas heater Project, Kern County Blythe USA Solar Millennium 2013–2014 4 × 250 Gas heater Source: Fichtner 2010. Table B.4: Demonstration central receiver projects First year of Electrical Thermal energy Name/location/country operation output (MWel) HTF storage SSPS, Spain 1981 0.5 Liquid sodium Sodium EURELIOS, Italy 1981 1 Water/steam Salt/water SUNSHINE, Japan 1981 1 Water/steam Salt/water Solar One, USA 1982 10 Water/steam Synthetic oil/rock CESA-1, Spain 1983 1 Water/steam Molten salt MSEE/Cat B, USA 1983 1 Molten salt Molten salt THEMIS, France 1984 2.5 Molten salt (hitec) Molten salt SPP-5, Ukraine 1986 5 Water/steam Water/steam TSA, Spain 1993 1 Atmospheric air Ceramics Solar Two, USA 1996 10 Molten salt Molten salt Consolar, Israel 2001 0.5* Pressurized air No (fossil hybrid) Solagte, Spain 2002 0.3 Pressurized air No (fossil hybrid) Solair, Spain 2004 3* Atmospheric air — CO-MINIT, Italy 2005 2 × 0.25 Pressurized air No (fossil hybrid) CSIRO Solar Tower Australia 2006 1* Other (gas reformation) Chemical (solar gas) DBT-550, Israel 2008 6* Water/steam (superheated) — STJ, Germany 2008 1.5 Atmospheric air Ceramics Eureka, Spain 2009 2* Water/steam (superheated) — Source: Fichtner 2010. Concentrating Solar Power in Developing Countries 107 Table B.5: Commercial central receiver projects Initial operation Name / location Company Concept Size (MWe) year/status PS 10 / Seville, Spain Abengoa Solar Water/Steam 10 2007 Solar Tower Jülich / Jülich, Kraftanlagen München Volumetric Air 1,5 2008 Germany PS 20 / Seville, Spain Abengoa Solar Water/Steam 20 2009 Sierra SunTower / California, USA eSolar Water/Steam 5 2009 Solar Tres / Seville, Spain Sener Molten Salt 17 2011/Under Construction Ivanpah 1–3 / California, USA BrightSource Energy Water/Steam 1 × 126 / 2 × 133 2013/Under Construction Geskell Sun Tower, Phase I–II / eSolar Water/Steam 1 × 105 / 1 × 140 Planning California, USA Alpine Power SunTower / California, eSolar/NRG Energy Water/Steam 92 Planning USA Cloncurry Solar Power Station / Ergon Energy Water/Steam 10 2010/on hold Queensland, AUS Upington / Upington, South Africa Eskom Molten Salt 100 2014/Announced Rice Solar Energy Project / Solar Reserve Molten Salt 150 Planning California, USA Tonopah / Nevada, USA Solar Reserve Molten Salt 100 Planning Source: Fichtner 2010. Table B.6: Demonstration parabolic dish collector projects First year of Net output Heat transfer Name/location/country operation [MWel] fluids/PCU Remark Rancho Mirage, USA 1983 0.025 Stirling motor individual-facet Vanguard Los Angeles, USA 1984 0.025 individual-facet, MDAC-25 Warner Springs, USA 1987 individual stretched membrane facets Osage City, USA 1987 Saudia Arabia 1984 2 × 0.05 Stirling motor SBP, stretched membrane Freiburg, Germany 1990 ï¬?xed focus, Bomin Solar Lampoltshausen, Germany 1990 Stirling motor SBP, stretched membrane, 2nd generation Almeria, Spain 1992–1996 6 × 0.01 Stirling motor SBP, stretched membrane Europe (Seville, Milano, etc.) 2002–2004 6 × 0.01 Stirling motor SBP, stretched membrane EuroDish/ EnvrioDish Johannesburg, South Africa 2002 0.025 Stirling motor SES & Eskom, multi-facets ALBUQUERQUE, New 2006–2008 8 × 0.025 Stirling motor SES & SNL, multi-facets Mexico, USA MARICOPA, Phoenix 2010 1.5 Stirling Motor SES, multi facets Source: Fichtner 2010. 108 A World Bank Study Table B.7: Component speciï¬?c cost reduction potential—parabolic trough Midterm cost reduction Long-term potential cost reduction Subsystem Component Reduction factor (%) potential (%) Solar ï¬?eld Reflectors New mirror concept 8–10 18–22 Mounting structure Mass production and material savings 12–20 25–30 Standardization 6–12 — Tracking system Experience curve 13–15 Receiver Operational improvements 15–20 Size increases 15 — Heat transfer system Experience curve 15–25 Thermal storage Molten salts Thermocline concept 20 — Fluid handling system Thermocline concept 10 — Power block Power block Experience curve 0–1 Balance of plant (bop) Experience curve 5–10 Source: YES/Nixus/CENER 2010. Table B.8: Component-speciï¬?c cost reduction potential—power tower Midterm cost Long-term reduction cost reduction Subsystem Component Reduction factor potential (%) potential (%) Solar ï¬?eld Reflectors New mirror concept 4–5 6–8 Mounting structure Mass production and material savings 15–18 17–20 Standardization 6–12 — Tracking system Experience curve 13–15 Receiver Experience curve 5–10 Heat transfer system Experience curve 15–25 Thermal storage Molten salts Thermocline concept 20 — Fluid handling system Thermocline concept 10 — Power block Power block Experience curve 0–1 Balance of plant Experience curve 5–10 Source: YES/Nixus/CENER 2010. Concentrating Solar Power in Developing Countries 109 Table B.9: Component-speciï¬?c cost reduction potential—linear fresnel Midterm cost Long-term reduction cost reduction Subsystem Component Reduction factor potential (%) potential (%) Solar ï¬?eld Reflectors Mass production 4–5 6–8 Mounting structure Mass production and material savings 20–25 25–35 Standardization 6–12 — Tracking system Experience curve 13–15 Receiver Wide operational improvement 15–25 Size increase 10 — Power block Power block Experience curve 0–1 Balance of plant Experience curve 5–10 Source: YES/Nixus/CENER 2010. Table B.10: Component-speciï¬?c cost reduction potential—dish engine Midterm cost Long-term reduction cost reduction Subsystem Component Reduction factor potential (%) potential (%) Solar ï¬?eld Reflectors Process automation and mass production 20–25 35–40 Mounting structure Mass production and material savings 17–20 25–28 Standardization 6–12 — Solar to energy Receiver/electric Experience curve 5–10 conversion motor and BOP Source: YES/Nixus/CENER 2010. 110 A World Bank Study Table B.11: Main ï¬?nancial and regulatory assumptions for LCOE analysis Main ï¬?nancial and regulatory assumptions India— Morocco— South Africa— South parabolic India—power parabolic Morocco— parabolic Africa— trough tower trough power tower trough power tower Plant size 100 MW 100 MW 100 MW Analysis 25 years 25 years 25 years period Inflation rate* 5.5% 2.15% 6.0% Real discount 11.25% 8.25% 10.5% rate Applicable 19.93% (MAT) 30% with Tax Holiday of 5 years, 28% tax rate from year 1 of construction (3 years construction + 2 of operation) Property tax 0% 0% 0% Vat 5% 14% 14% Depreciation 7% ï¬?rst 10 years—2% thereafter 25 years straight line 25 years straight line schedule Loan term 14 years 18 years with 4 years grace 20 years (commercial) period Loan rate 11.75% 9% 12% (commercial) Debt / equity 70 / 30 80 / 20 70/30 ratio Roe 19% 15% 17% Min required 15% 15% 15% irr Insurance 0.5% 0.5% 0.5% Exchange rate 45 Rs/US$ 8.2 Dhs/US$ ZAR 10/US$ Capital cost US$4,500/ US$5,000/ US$4,500/kW US$5,000/kW US$4,700/kW US$5,200/ kW (excluding kW (excluding (excluding (excluding (excluding kW (excluding storage) storage) storage) storage) storage) storage) O&M cost US$32/kW-yr US$30/kW-yr US$35/kW-yr US$33/kW-yr US$70/kW-yr US$66/kW-yr (including (plus Dhs (plus Dhs Variable cost) 15 million/ 15 million/year year rent) rent) Optimal 6 hours TES 15 hours TES 3 hours TES 15 hours TES 3 hours TES 15 hours TES storage Total installed US$7,707/kW US$8,306/kW US$7,385/kW US$8,909/kW US$7,900/kW US$9,171/kW cost Capacity factor 38.5% 52.7% 32.5% 62% 35% 67.9% (air-cooled) Annual mwh 337,341 MWh 461,592 MWh 284,891 MWh 543,348 MWh 306,269 MWh 595,008 MWh generated (air-cooled) Assumed dni 2,262 kWh/m2/year 2,578 kWh/m2/year 2,916 kWh/m2/year System 0.25–0.5% (0.425% assumed) 0.25–0.5% (0.425% assumed) 0.25–0.5% (0.425% assumed) degradation Source: Macroeconomica 2011. *Average CPI-Inflation from 2000 to 2009. Concentrating Solar Power in Developing Countries 111 Table B.12: Impact assessment of different regulatory incentives in India LCOE % Current after Change Technology LCOE Incentive applied incentive in LCOE Parabolic trough (Air-cooled— 35.54 Tax reduction 35.20 −0.96 with storage) VAT exemption 35.20 −0.96 Accelerated depreciation 34.06 −4.16 Concessional loan terms 33.36 −6.13 Concessional loan rates 32.94 −7.32 Concessional loan terms + rates 29.81 −16.12 AD + concessional loan terms + rates 28.32 −20.32 Power tower (Air-cooled— 27.85 Tax reduction 27.58 −0.97 with storage) VAT exemption 27.58 −0.97 Accelerated depreciation 26.69 −4.17 Concessional loan terms 26.13 −6.18 Concessional loan rates 25.80 −7.36 Concessional loan terms + rates 23.34 −16.19 AD + concessional loan terms + rates 22.16 −20.43 Parabolic trough (Wet-cooled— 33.27 Tax reduction 32.95 −0.96 with storage) VAT exemption 32.95 −0.96 Accelerated depreciation 31.89 −4.16 Concessional loan terms 31.23 −6.13 Concessional loan rates 30.84 −7.32 Concessional loan terms + rates 27.91 −16.11 AD + concessional loan terms + rates 26.51 −20.32 Power tower (Wet-cooled— 26.67 Tax reduction 26.41 −0.97 with storage) VAT exemption 26.42 −0.94 Accelerated depreciation 25.56 −4.16 Concessional loan terms 25.03 −6.15 Concessional loan rates 24.71 −7.35 Concessional loan terms + rates 22.35 −16.20 AD + concessional loan terms + rates 21.23 −20.40 Source: Authors’ analysis. 112 A World Bank Study Table B.13: Impact assessment of different regulatory incentives in Morocco Technology Current LCOE Incentive applied LCOE after incentive % Change in LCOE Parabolic trough 37.25 Tax reduction 36.80 −1.21 (Air-cooled—with VAT exemption 36.53 −1.93 storage) Accelerated depreciation 31.92 −14.31 Concessional loan terms 34.49 −7.41 Concessional loan rates 33.68 −9.58 Concessional loan 30.26 −18.77 terms + rates AD + concessional loan 24.82 −33.37 terms + rates Power tower 23.27 Tax reduction 22.99 −1.20 (Air-cooled—with VAT exemption 22.81 −1.98 storage) Accelerated depreciation 19.90 −14.48 Concessional loan terms 21.52 −7.52 Concessional loan rates 21.00 −9.76 Concessional loan 18.84 −19.04 terms + rates AD + concessional loan 15.40 −33.82 terms + rates Parabolic trough 34.52 Tax reduction 34.11 −1.19 (Wet-cooled—with VAT exemption 33.85 −1.94 storage) Accelerated depreciation 29.58 −14.31 Concessional loan terms 31.96 −7.42 Concessional loan rates 31.21 −9.59 Concessional loan 28.04 −18.77 terms + rates AD + concessional loan 23.00 −33.37 terms + rates Power tower 22.11 Tax reduction 21.85 −1.18 (Wet-cooled—with VAT exemption 21.68 −1.94 storage) Accelerated depreciation 18.91 −14.47 Concessional loan terms 20.45 −7.51 Concessional loan rates 19.96 −9.72 Concessional loan 17.91 −19.00 terms + rates AD + concessional loan 14.64 −33.79 terms + rates Source: Authors’ analysis. Concentrating Solar Power in Developing Countries 113 Table B.14: Impact assessment of different regulatory incentives in South Africa Technology Current LCOE Incentive applied LCOE after incentive % Change in LCOE Parabolic trough 42.32 Tax reduction 41.58 −1.75 (Air-cooled—with VAT exemption 41.47 −2.01 storage) Accelerated depreciation 37.07 −12.41 Concessional loan terms 41.18 −2.69 Concessional loan rates 38.78 −8.36 Concessional loan 37.23 −12.03 terms + rates AD + concessional loan 31.91 −24.60 terms + rates Power tower 24.92 Tax reduction 24.48 −1.77 (Air-cooled—with VAT exemption 24.41 −2.05 storage) Accelerated depreciation 21.78 −12.60 Concessional loan terms 24.24 −2.73 Concessional loan rates 22.80 −8.51 Concessional loan 21.87 −12.24 terms + rates AD + concessional loan 18.69 −25.00 terms + rates Parabolic trough 38.90 Tax reduction 38.21 −1.77 (Wet-cooled—with VAT exemption 38.11 −2.03 storage) Accelerated depreciation 34.07 −12.42 Concessional loan terms 37.85 −2.70 Concessional loan rates 35.64 −8.38 Concessional loan 34.22 −12.03 terms + rates AD + concessional loan 29.33 −24.60 terms + rates Power tower 23.76 Tax reduction 23.34 −1.77 (Wet-cooled—with VAT exemption 23.27 −2.06 storage) Accelerated depreciation 20.77 −12.58 Concessional loan terms 23.11 −2.74 Concessional loan rates 21.73 −8.54 Concessional loan 20.85 −12.25 terms + rates AD + concessional loan 17.82 −25.00 terms + rates Source: Authors’ analysis. 114 A World Bank Study Table B.15: Economic analysis—main cost assumptions Parabolic trough Power tower Item Unit India Morocco S. Africa India Morocco S. Africa Capacity (gross) MW 100 100 100 100 100 100 Generation net gWh/a. 397 264 440 388 388 493 Degradation of generation % p.a. 0.0 0.5 0.0 0.0 0.5 0.0 Capacity factor % 50% 30% 56% 49% 31% 63% CAPEX US$Mn. 738 600 861 717 717 786 Cons. period Years 6 3 6 6 6 6 Lifetime of plant Years 25 25 25 20 20 20 Variable O&M costs Fuel US$Mn. 0.2 0.30 0.30 0.3 0.3 0.3 Water US$Mn. 0.12 0.12 0.12 0.11 0.08 0.08 Fixed O&M costs US$Mn. 14.2 15.1 16.6 14.5 12.3 16.3 Personnel US$Mn. 4.4 4.5 4.4 2.7 3.5 4.5 Non-personnel US$Mn. 9.8 10.6 12.2 11.8 8.8 11.8 CO2 Eq. saved Kg/kWh 1.03 0.64 1.03 1.03 0.64 1.03 Local pollutants SO2 Kg./kWh n.a. 0.011 n.a. n.a. 0.011 n.a. NOx Kg./kWh n.a. 0.003 n.a. n.a. 0.003 n.a. PM10 Kg./kWh n.a. 0.001 n.a. n.a. 0.001 n.a. Escalation factors Value of electricity % p.a. 3.64 2.15 0 3.64 2.15 0 O&M costs % p.a. 1.0/5.0 2.15 1.0/5.0 1.0/5.0 2.15 1.0/5.0 CO2 & other ext. values 0 2.15 0 0 2.15 0 Value of electricity US¢/kWh 8.0 11.1 17.5 8.0 11.1 17.5 Value of CO2 in 2014 Original US$/ton — 31.3 29.0 — 31.3 29.0 Modiï¬?ed US$/ton 40.5 40.5 40.5 40.5 40.5 40.5 Value local pollutants SO2 US$/ton n.a. 267 n.a. n.a. 267 n.a. NOx US$/ton n.a. 1,156 n.a. n.a. 1,156 n.a. PM10 US$/ton n.a. 711 n.a. n.a. 711 n.a. Source: Macroeconomica 2011. Note: Escalation of O&M costs was 1% for nonpersonnel and 5% for personnel costs in S. Africa & India. The escalation of the value of CO2 was only in the original case. n.a. = not available. Concentrating Solar Power in Developing Countries 115 Table B.16: Global CST value chain analysis Industry structure Economics and costs Project â–« Small group of companies with technological â–« Mainly labor-intensive engineering activities and development know-how activities to obtain permits. â–« International actors have fully integrated activities of concept engineering; often with project development, engineering, ï¬?nancing. EPC â–« Strong market position for construction, energy, â–« Large infrastructure companies (high turnover) contractors transport and infrastructure projects. Parabolic â–« Few, large companies, often from the automotive â–« Large turnover for a variety of mirror and glass mirrors sector products â–« Large factory output Receivers â–« Two large players â–« Large investment in know-how and machines â–« Factories also in CST markets in Spain and the required United States Metal support â–« Steel supply can be provided locally â–« High share of costs for raw material, steel or structure â–« Local and international suppliers can produce aluminum the parts Market structure and trends Key competiveness factor Project â–« Strongly depending on growth/expectations of â–« Central role for CST projects development individual markets â–« Technology know-how â–« Activities worldwide â–« Access to ï¬?nance EPC â–« Maximum 20 companies â–« Existing supplier network contractors â–« Most of the companies active on markets in Spain and the United States Parabolic â–« A few companies share market, all have increased â–« Bending glass mirrors capacities â–« Manufacturing of long-term stable mirrors with â–« High mirror price might decline high reflectance â–« Inclusion of upstream float glass process Receivers â–« Strongly depending on market growth â–« High-tech component with specialized production â–« Low competition today; new players about to and manufacturing process enter the market Metal support â–« Increase on the international scale expected â–« Price competition structure â–« Subcontractors for assembling and materials â–« Mass production / Automation Strengths Weaknesses Opportunities Threats Project â–« Reference projects â–« Dependency on â–« Projects in pipeline â–« Price competition with development â–« Technology know-how political support other renewables EPC â–« Reference projects â–« High cost â–« Projects in pipeline â–« Price competition with contractors â–« Well-trained staff â–« Achieve high cost other renewables â–« Network of suppliers reduction Parabolic â–« Strong position of â–« Cost of factory â–« New CST markets â–« Unstable CST market mirrors few players â–« Continuous demand â–« Barriers for market â–« Flat mirror technology â–« High margins (high required entry (Fresnel/tower) cost reduction potential) Receivers â–« High margins (high â–« Dependency on â–« High cost reduction â–« Unstable CST market cost reduction potential) CST market potential through â–« Low market demand â–« High entry barrier competition â–« Strong market position for new players of few players; hard for (know-how/invest) new players to become commercial Metal support â–« Experience â–« High cost competition â–« Increase of efï¬?ciency â–« Volatile CST market structure â–« New business and size opportunities for structural steel â–« Low entry barriers Source: Ernst & Young and Fraunhofer 2010. 116 A World Bank Study Table B.17: Technical and economic barriers to manufacturing CST components Level of Components Technical barriers Financial barriers Quality Market Suppliers barriers Civil work Low technical skills Investment in large Standard quality Successful market Existing supplier Low required shovels and trucks of civil works, players will pro- structure can be exact works vide these tasks used for materials EPC Very highly skilled — Quality manage- Limited market Need to build up Medium engineers professionals: ment of total site of experienced their own network and project engineers and has to be done engineers managers project managers with university degrees Assembly Logistic and Investment in Accuracy of Collector assembly Steel parts Low management skills assembly-building process, low fault has to be located transported over necessary for each site, production during close to site longer distance investment in train- continuous large Lean manufacturing, Competitive ing of work force output automation suppliers often Low skilled also local ï¬?rms workers Receive Highly specialized High speciï¬?c High process Low market Supplier network High coating process with investment for know-how for opportunities to not strongly high accuracy manufacturing continuous high sell this product required Technology- process quality to other industries intensive sputtering and sectors step Float glass Float glass process Very capital- Purity of white Large demand is Supplier network High production is the state-of-the- intensive glass (raw required to build not strongly (for flat art technology but products) production lines required and curved large quantities mirrors) and highly energy intensive Complex manufac- turing line Highly skilled workforce to run a line Mirror flat Complex manufac- Capital-intensive Long-term stability High quality flat Supplier network High (float glass) turing line of mirror coatings mirrors have not strongly Highly skilled limited further required workforce to run markets a line Large demand is required to build production lines Mirror See flat mirrors See flat mirrors See flat mirrors Large demand Supplier network High parabolic Plus: + bending devices High geometric is required to not strongly Bending highly precision of build production required automated bending process lines production Parabolic mirrors can only be used for CST market (Table continues on next page) Concentrating Solar Power in Developing Countries 117 Table B.17: (continued) Level of Components Technical barriers Financial barriers Quality Market Suppliers barriers Mounting Structure and Automation is For tracking and Markets with large Raw steel Low structure assembly are usually capital-intensive mounting: stiff- and cheap steel market important proprietary know- Cheap steel ness of system Transformation how of companies is competitive required industries are Standardization/ advantage highly competitive automation by robots or stamping reduces low skilled workers, but increases process know-how HTF Chemical industry Very capital- Standard Large chemical Not identiï¬?ed High with large production. intensive product, heat companies pro- However, the oil is resistant duce thermal oil not highly speciï¬?c Connection Large and intensive Capital-intensive High precision Large quantities Not identiï¬?ed Medium piping industrial steel production line and heat transformation resistance processes Process know-how Storage Civil works and Not identiï¬?ed Not identiï¬?ed Low developed Not identiï¬?ed Medium system construction is done market, few locally project developers Design and archi- in Spain tecture Salt is provided by large suppliers Electronic Standard cabling Not identiï¬?ed Not identiï¬?ed Market demand Often supplier Low equipment not difï¬?cult of other industries networks because Many electrical necessary of division components special- ized, but not CST speciï¬?c equipment; Equipment not pro- duced for CST only Source: Ernst & Young and Fraunhofer 2010. 118 Table B.18: Action plan for stimulation of production of CST products in MENA Potential Implementation Goals Intermediate steps Necessary processes/assistance Target groups actors timeframe A World Bank Study Upgrade & increase of Provision of information on CSP market Implementation of national and regional CSP Current and potential future producers Δ ♦◊ Short to medium industrial and service size and opportunities of production associations that foster networking, accelerate of intermediate products and CSP term capacities and service adjustment business contacts and provide information components, research organizations Establishment of superordinated national See above Δ Short to medium institutions responsible for CSP targets to term enhance and coordinate policy development in the regional context and to provide assistance Creation of internet platforms, newsletters on See above Δ Short to medium technical issues and market development, term information centers and other informational support Assessment of technical feasibility for Foundation of consorts of technical experts Current producers of intermediate products Δ Short to medium ï¬?rms to upgrade current production that support companies which show interest and CSP components term to CSP component production and in CSP manufacture or provision of funds to service provision consult external technical experts Implementation of investment support Financial support of a certain share of the Current local producers of intermediate Δ Short to medium mechanisms for adaptation of necessary investment for implementation products term production lines of upgrade of production facilities (e.g. “renewable energy innovation fundâ€?) Provision of long-term low-interest loans for Current local producers of intermediate Δ Short to medium companies willing to invest in innovation of products and potential future producers term production lines Facilitation of foreign investments by International players Δ Short to medium simpliï¬?cation of bureaucracy and assistance term Price incentives Tax incentives for production/export of CSP Local producers, national and international Δ Medium term components (e.g. reduction or exemption on companies customs duties for raw materials, parts or spare parts of CSP components, refund of customs duties with export) Tax credits or deductions for investments National and international companies Δ Medium term in production lines related to CSP and investments in R&D Lowered trade barriers for RE/CSP See above Δ Medium term components and intermediate products to accelerate the trade of components Tax credits on ï¬?rm-level training measures See above Δ Short to medium term Further incentives Local and regional content obligations for See above Δ Medium term components and services in CSP projects Foster integration of secondary components See above Δ Short term suppliers in region Activation of further Strong focus in national and regional Formulation of clear national targets regarding National and international industrial players Δ Short to medium potential market players industrial policy on CSP development the development of CSP industries in general term and service providers Provision of administrative and legislative support National and international industrial players Δ Short to medium Concentrating Solar Power in Developing Countries for company start-ups and foreign investments, in general term and formation of relevant institutions Financial support mechanisms for national National players Δ Short to medium company start-ups in the sector of renewable term energy manufacturing Introduction of regional quality assurance National and international companies Δ ♦◊ Medium to long standards for CSP products to decrease term uncertainty Awareness raising Awareness-raising initiatives (e.g. conferences, National and international industrial players Δ ♦ Medium to long workshops, other marketing activities) and in general term formation of relevant institutions Facilitation of skill Promote creation of joint ventures Facilitation of networking and knowledge Regional and international manufacturers Δ♦◊ Short to medium enhancement and between existing manufacturers and transfer by creating networking platforms and term knowledge transfer potential regional newcomers organization of business fairs Support of training activities for local Review of existing national training facilities, Δ Short to medium workforce upgrade/creation of speciï¬?c institutions if needed term Provision of short basic training courses for civil Regional companies, particularly low-skilled Δ Short to medium 119 workers (e.g. involved in assembly activities) workforce term (Table continues on next page) Table B.18: (continued) 120 Potential Implementation Goals Intermediate steps Necessary processes/assistance Target groups actors timeframe Support the training of regional workforce by Regional companies, international companies Δ Short to medium A World Bank Study ï¬?nancial support if external training facilities term are involved Promotion of ï¬?nancial incentives for ‘train the Regional companies, international companies Δ Short to medium trainers’ programs term Support of higher education Establishment of study courses with regard Regional students and engineers, O&M Δ Short to medium to solar energy techniques/CSP and other workforce term required skills related to RE/CSP Creation of master programs at foreign Regional students Δ Short to medium universities and student exchange programs term with regard to RE/CSP Review of management and project planning Students, potential CSP workforce (e.g. Δ Medium to long capabilities and creation of training courses existing EPC contractors) term Support of private and public R&D Improvement of renewable energy related Manufacturers, private and public research Δ Short to medium R&D legislation, and national legislation institutions (e.g. universities) term exchange (e.g. through RCREE) Foundation of research institutions and See above Δ ♦◊ Medium to long technology clusters with regard to CSP term technologies, to foster regional knowledge distribution and innovation Implementation of CSP testing plants and CSP-project developer, national and Δ ♦◊ Short to medium project-parallel research activities at CSP sites international CSP component producers, term public and private research facilities Promotion of international science networks Scientists at national and international Δ Medium to long and exchange of scientiï¬?c experts in the institutions term ï¬?eld of CSP component design (particularly important for collectors and receivers) Enhancement of links between industry and Scientists at national and international Δ ♦◊ Medium to long research facilities (universities) institutions, regional companies, international term companies Source: Ernst & Young and Fraunhofer 2010. Actors/ï¬?nancers: Δ = national authorities, = international donors, â—Š = national CST players, ♦ = international CST players Concentrating Solar Power in Developing Countries 121 Table B.19: Component-speciï¬?c local manufacturing prospects in South Africa Potential for manufacture within CST system/component South Africa Remarks Structural steel High Up to 100% of steel required can be provided locally. Concrete High Up to 100% of concrete required can be provided locally. Steel piping High Up to 80% of all the steel piping can be provided locally. CST-shaped glass Medium in the short to medium term High in the long term Electrical and Control cabling and High Up to 100% of all cabling can be manufactured locally. accessories Pressure vessels and storage tanks High All pressure vessels and storage tanks and vessels can be manufactured locally. Shaped steel sections High All shaped steel sections can be provided locally. Medium voltage and low voltage High All MV and LV motors can be manufactured locally. electric motors DC motors High All DC motors can be manufactured locally. Valves and actuators High Valves and actuators can be manufactured locally. Distribution and power transformers High All transformers can be manufactured locally. (Oil-ï¬?lled and dry type) Lead Acid and nickel cadmium High All batteries can be manufactured locally. batteries Battery chargers, UPSs and inverters High This equipment can be manufactured locally. Variable speed drives (low voltage) High VSDs for LV motors can be manufactured locally. Variable speed drives (medium voltage) Low MV drives will be imported into the long term. Steam turbines Low Heat exchangers High All heat exchangers can be manufactured locally. Instruments High All instruments can be manufactured locally. Programmable logic controllers, Low plant information systems and DCS equipment Nitrogen systems Low Most of the Nitrogen gas will need to be imported. Aluminum conductor for overhead High All Aluminum conductors for overhead lines can be lines manufactured locally. Molten salts Low Oil-based HTF Low Diesel generator sets Low Diesel generator sets can be assembled in South Africa, but alternators and diesel engines, as well as the controls, will be imported into the long term. Pumps High Most of the pumps can be manufactured locally. It is very likely that HTF pumps can be supplied locally in the medium term since there are existing suppliers of large pumps for the petrochemical industry. (Table continues on next page) 122 A World Bank Study Table B.19: (continued) Potential for manufacture within CST system/component South Africa Remarks Water treatment plants High All water treatment plants can be designed and assembled locally. Chemicals for water treatment High All chemicals can be manufactured locally. Heaters High Heating, ventilation and air Medium conditioning equipment (HVAC) Fencing material High All fencing material can be provided locally. Fireï¬?ghting equipment High CST steel structures Medium Low in the short term. High in the medium to long term. Tracking systems Medium Low in the short to medium term. High in the long term. Automotive component manufacturers have got the machining equipment to manufacture high-precision structures. The machining equipment can be used to manufacture tracking systems in the long term. Weather measurement equipment High Telecommunications and telecontrol Medium equipment MV and LV switchgear Medium Source: Fichtner 2011. Concentrating Solar Power in Developing Countries 123 Table B.20: Capacity to manufacture CST components and provide CST-related services in South Africa Research & Potential of entry Financial development by international Sector strength potential ï¬?rms into sector Remarks Steel High High Medium. The 2 ï¬?rms have a dominant role in the steel manufacturing sector in South Africa. South Africa’s Industrial Large local Both Arcelor Policy Action Plan (IPAP) is proposing incen- ï¬?rms Arcelor Mittal and Evraz tives for foreign investors into South Africa Mittal and have got large Evraz Highveld R&D divisions Steel dominate and also beneï¬?t this sector from the R&D capabilities of parent companies. Automotive High Low High Most ï¬?rms have small R&D capabilities and component rely on industry bodies to coordinate R&D manufacturers efforts. Capacity to manufacture CST steel structures and components low in the short term, but there is potential for increase in the long term. Glass High Medium High The capacity to manufacture CST glass in manufacturing the short to medium term is limited for PG sector Glass Industries. Electrical High High High This sector is dominated by the Big 5 multi- equipment national ï¬?rms: GE, ABB, Siemens, Alstom and Groupe Schneider. Potential exists for other international players to enter this market for speciï¬?c electrical equipment, such as MV Vari- able Speed Drives, which are currently being imported, as well as for large transformers and DCS equipment for power plants. Electronics Medium Medium High Most of the local electronics components equipment manufacturing ï¬?rms are small. This market is dominated by Siemens, Alstom and ABB. EPC ï¬?rms High Medium High The local EPC ï¬?rms do not have experience in doing EPC on CST projects. There is scope For the big for them to work as subcontractors to large 3 ï¬?rms (Murray international EPC CST plant developers such & Roberts, as Abengoa. Group 5 and Grinaker LTA) Professional High Medium High Local engineering consulting and project services management ï¬?rms do not have experience (engineering in executing CST projects. There is scope for consulting entry of international consulting ï¬?rms in this and project area and subcontract work to local ï¬?rms. management) Cement and High High Low This sector is dominated by a few large concrete companies with a large market share. The oligopolistic nature of the industry presents signiï¬?cant entry barriers to new entrants. Source: Fichtner 2011. Table B.21: G20 and select nonmembers’ producer price inflation (% over previous year) 124 Country 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Average Std Dev Argentina .. .. 0.4 4.3 2.9 −1.1 −3.4 −4.0 3.7 −2.0 78.3 19.6 7.7 8.4 11.0 11.8 14.5 4.8 16.5 10.2 18.9 Australia 1.5 2.0 0.8 4.2 0.3 1.2 −4.0 −0.9 7.1 3.1 0.2 0.5 4.0 6.0 7.9 2.3 8.3 −5.4 2.2 2.2 3.6 A World Bank Study Brazil 987.8 2050.1 2311.6 57.5 6.3 10.1 3.5 16.6 18.1 12.6 16.7 27.6 10.5 5.6 0.8 5.6 13.7 −0.2 5.7 292.6 703.2 Canada 0.5 3.6 6.1 7.4 0.4 0.7 0.4 1.8 4.3 1.0 0.1 −1.2 3.2 1.6 2.3 1.5 4.3 −3.5 1.0 1.9 2.5 China .. .. .. .. .. .. .. .. 2.8 −4.0 0.4 3.0 7.1 3.2 3.1 3.1 6.9 −5.4 5.5 2.3 4.0 Hong Kong, SAR, China 1.8 0.7 2.1 2.8 −0.1 −0.3 −1.8 −1.6 0.2 −1.6 −2.7 −0.3 2.3 −7.9 2.2 3.0 5.6 −1.7 6.0 0.5 3.2 Euro Area 1.5 1.5 2.0 4.1 0.3 1.0 −0.7 −0.5 5.1 2.2 −0.2 1.4 2.3 4.3 5.4 2.7 6.3 −5.2 2.8 1.9 2.6 France −1.5 −1.5 1.1 6.1 −2.6 −0.6 −0.9 .. 4.4 1.2 −0.2 0.9 2.1 3.1 2.9 2.3 4.8 −5.6 3.0 1.0 2.9 Germany .. 0.2 0.6 1.7 −1.2 1.2 −0.4 −1.0 3.3 3.0 −0.6 1.7 1.6 4.3 5.4 1.3 5.5 −4.2 1.6 1.3 2.4 India 11.9 7.5 10.5 9.3 4.5 4.5 5.9 3.5 6.6 4.8 2.5 5.4 6.6 4.7 4.7 4.8 8.7 2.1 9.4 6.2 2.7 Indonesia 5.2 3.7 5.4 11.4 7.9 9.0 101.8 10.5 12.5 13.0 4.4 3.4 7.4 15.3 13.7 14.7 21.5 4.6 3.1 14.1 21.8 Italy 1.9 3.8 3.7 7.9 1.9 1.3 0.1 −0.3 6.0 1.9 −0.2 1.6 2.7 4.0 5.6 3.5 4.8 −4.7 3.0 2.6 2.8 Japan −0.9 −1.6 −1.6 −0.8 −1.7 0.7 −1.5 −1.5 0.0 −2.3 −2.1 −0.8 1.3 1.7 2.2 1.7 4.6 −5.3 −0.2 −0.4 2.1 Korea, Republic of 2.2 1.5 2.7 4.7 3.2 3.9 12.2 −2.1 2.0 −0.4 −0.3 2.2 6.1 2.1 0.9 1.4 8.6 −0.2 3.8 2.9 3.3 Mexico 12.3 7.4 6.1 38.6 33.9 17.5 16.0 14.2 7.8 5.0 5.1 7.5 9.3 4.2 6.6 3.6 6.5 5.9 3.3 11.1 9.8 Netherlands 1.8 0.1 0.5 1.5 2.0 1.8 −0.2 1.0 4.8 3.0 0.8 1.7 2.8 3.2 3.6 4.5 5.1 −3.8 2.6 1.9 2.1 Russian Federation .. 943.8 337.0 236.5 50.8 15.0 7.0 58.9 46.5 18.2 10.4 16.4 23.4 20.6 12.4 14.1 21.4 −7.2 12.2 102.1 228.0 Saudi Arabia 1.3 0.6 1.8 7.3 −0.3 0.0 −1.9 0.4 0.4 −0.1 0.0 0.9 3.1 2.9 1.2 5.7 9.0 −3.0 4.3 1.8 3.0 Singapore −4.4 −4.4 −0.4 0.0 0.1 −1.2 −3.0 2.1 10.1 −1.6 −1.5 2.0 5.1 9.7 5.0 0.3 7.5 −13.9 4.8 0.9 5.6 South Africa 9.2 7.0 8.8 9.9 7.1 8.1 4.4 4.9 6.7 7.6 13.5 2.2 2.3 3.6 7.7 10.9 14.3 0.0 6.0 7.1 3.7 Spain 1.3 2.5 4.3 6.4 1.7 1.0 −0.7 0.7 5.4 1.7 0.6 1.4 3.4 4.7 5.4 3.6 6.5 −3.4 3.2 2.6 2.5 Switzerland 0.7 0.4 −0.5 −0.1 −1.8 −0.7 −1.2 −1.0 0.9 0.5 −0.5 0.0 1.2 0.8 2.1 2.4 3.4 −2.1 −0.1 0.2 1.4 Turkey 62.1 58.0 121.3 86.0 75.9 81.8 71.8 53.1 51.4 61.6 50.1 25.6 14.6 5.9 9.3 6.3 12.7 1.2 8.5 45.1 34.4 United Kingdom 3.1 4.0 2.5 4.0 2.6 1.0 0.0 0.6 1.4 −0.3 −0.1 0.6 1.0 2.0 2.0 2.3 6.8 1.6 4.2 2.1 1.8 United States 0.6 1.5 1.3 3.6 2.3 −0.1 −2.5 0.8 5.8 1.1 −2.3 5.3 6.2 7.3 4.7 4.8 9.8 −8.8 6.8 2.5 4.3 Source: Bureau of Labor Statistics 2010. Table B.22: Select MENA wholesale price inflation (% over previous year) Country 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 Average Std Dev Algeria .. .. .. .. .. .. 3.9 2.5 2.5 4.8 2.3 4.5 4.3 3.1 2.0 3.9 8.5 3.5 .. 3.8 1.7 Concentrating Solar Power in Developing Countries Egypt, Arab Rep. .. 6.4 6.0 5.7 8.9 3.3 1.6 1.6 1.5 1.5 6.0 14.1 17.3 5.3 7.0 10.3 21.2 (5.6) .. 6.6 6.5 Jordan .. .. .. .. .. .. 0.0 (2.2) (3.3) (1.1) (3.4) 2.4 5.8 9.9 16.0 8.6 56.3 (16.8) .. 6.0 17.9 Morocco .. 4.9 2.4 5.7 5.4 (2.1) 3.2 (2.0) 4.2 0.0 2.0 (4.9) .. .. .. .. .. .. .. 1.7 3.5 Tunisia .. 6.1 2.9 5.6 3.9 2.5 2.5 1.2 2.4 2.3 3.4 2.2 3.2 4.2 7.0 3.7 11.7 2.4 .. 4.0 2.5 Source: Bureau of Labor Statistics 2010. 125 126 A World Bank Study Figure B.1: Possible evolutions of local CST industries for key components in MENA Current situation 2015 2020 2030 MENA market (capacity, nb of 70 MW, 3 ISCC, ~0.5 300 MW, 3-5 plants, ~2 ~1 GW, 10-20 plants, ~8 ~2 GW, 20-30 plants, ~15 plants, size in billion $) Egyptian Glass Company Flabeg Potential ~ same players Dr Greiche new entrants Rioglass solar Sphinx Glass Sphinx Glass Saint-Gobain SIALA Dr Greiche Guardian Ind. EGC Pilkington Mirrors Flabeg Flabeg Rioglass solar SIALA Rioglass solar Saint-Gobain Guardian Industries Saint-Gobain Guardian Ind. Pilkington Market size (MS) 25 100 PPG 400-500 800-1000 - All ISCC supplied by - Large international companies - Regional market size becoming - Developments of local assets, international companies getting interested in MENA market significant increase of local producers’ market Description - MENA market size not - Part of value chain produced - Integration of full value chain by share driven by call for tenders’ and drivers large enough for local CSP locally (coating) by local glass local players local production clauses glass and mirror production transformers - Implementation of a large - Consolidation of all « historical » - MENA market size still too small international firm’s affiliate and stakeholders’ positions for full value chain integration development of reconversion of assets by pure local players NSF Engineering Very limited NSF Engineering Ynna Holding nb of intl. DLM Delattre Levivier Maroc DLM companies AOI NSF Engineering NSF Engineering AOI Mounting Ynna Holding structures El Fouladh Areva + other new Acciona entrants Abengoa Solar Abengoa Solar Market size (MS) ~50 ~225 800-1000 1500-1700 - Abengoa supplied Aïn Beni - Increased interest of company - Cost reduced by local production - Perfect command of mounting Mathar and Hassi R’mel with R&D capacity and already (escpecially through low transport structure construction techniques Description - NSF engineering designed producing complex metallic and low labor costs) - Previous economical drivers still and drivers and prduced the Kuraymat structures (roofs, wind towers) - Development of local knowledge influential mounting structure - International developers and experiece gained in first - Only « pure » local production as preferring « standard » mounting MENA projects no need for know-how transfer structure design already - Very specific needs of anymore and industrialization of implemented in other CSP plants international developers production Developer’s ~ same players El Sewedy Cables International Groupe Elloumi El Sewedy suppliers Electric and TECI Cables electronic Elloumi equipment TECI Leoni Cables, Developer’s Leoni Cables, + new Delphi, Yakazl, International suppliers Delphi, Yakazl, Sumitomo, entrants Sumitomo, ~ same players Market size (MS) 2 Nexans + new entrants 5-10 ~30 Nexans + new entrants ~50 - Local players used to comply - Consolidation of market shares - Share of the market between top - High-tech components with stringent requirements from by local players local firms (competitive on supplied by conventional international clients developing Description international suppliers -Decrease of components’ import international markets) and and drivers specific CSP components because of the combination of international firms having - Low added-value - New entrants in high-tech components (cables, etc.) competitive local products and developed local capacity because components (trackers, for local production clauses in call for of low labor cost and strategic supplied by local companies example), as aeronautical or automotive companies tenders location : Import in MENA : “Pureâ€? local production (current local players) : Local production (implantation of international players) Source: Ernst & Young and Fraunhofer 2010. Figure B.2: Potential roadmap for the production of CST mirrors in the MENA Region Potential roadmap for the production of CSP mirrors in the MENA region Status Quo Short-Term Mid-Term Long-Term Overall Goal Technology development High availability of Single float glass Mirror companies Production Application of One or two large raw materials but factories in MENA in MENA possess facilities and skills alternative suppliers of white glass currently no are upgraded for skills for are upgraded for materials & design and several mirror production of high- production of high- production of CSP bending process (e.g., polymers, thin manufacturers in quality white glass quality white glass mirrors (coating) glass, aluminum) MENA produce highly or parabolic mirrors precise CSP reflectors in MENA at a competitive price Mirrors for all types of All reflectors for Supply of white Provision of linear Provsion of highly CSP projects in MENA CSP plants in MENA glass for potential reflectors for precise parabolic region can be supplied are imported from (foreign) mirror Fresnel plants or mirrors for solar by regional companies abroad factories in MENA solar towers trough plants plus export of mirrors possible possible possible Business development Concentrating Solar Power in Developing Countries Subsidiary of Predominantly foreign company Independent medium-sized production of CSP mirror companies mirrors in MENA. with no activity in Newly emerging mirror Foundation of Positive spill-over CSP technology so companies and strong joint ventures Comprehensive Investments in High level of effects on other far increase of overall training of upgrade of sophistication is glass sectors (other sectoral potential employees production lines reached special purpose Acquisition of glasses, solar glass— licenses e.g., Photovoltaic) Growing intellectual Poorly developed Strong focus on Techniques and Patented Innova- property with regard to intellectual property Applied research R&D in the field of materials adapted tions in reflector CSP mirrors. Profit rights in MENA, high accompanying reflector design, to specific needs designs & mainten- from innovative dependency on ongoing projects coatings & main- and resources of ance equipment designs, materials, and market leaders & testing plants tenance the countries in MENA cleaning methods Policy framework & Strategy funds for High level of market development industrial upgrade regional Superordinate are provided integration of the No national targets Coordinated Region-wide clear institutions are CSP value chain political goals for development of national strategies Large number of established realized in MENA regarding industrial CSP mirror industry for industrial R&D competence development and clusters created policy Institutional energy targets responsibilities and defined Long-term, stable Favorable tax rates License trade of Growing export of Focused support for budgetary powers policy framework exist for CSP CSP mirrors in the CSP mirrors from industrial development partly fragmented is implemented mirrors MENA region MENA of CSP mirror industry CSP market development in Definition of long- MENA uncertain, Growing number Growing level of Minimum of 4 GW term objectives for Continuous & stable small number of of CSP projects in confidence in CSP added CSP capacity CSP development growth of CSP market projects in pipeline pipeline technology in MENA per year in MENA in MENA 127 Source: Ernst & Young and Fraunhofer 2010. Figure B.3: Potential roadmap for the production of metal structures for CST in RSA 128 Status Quo Short-Term Mid-Term Long-Term Overall goal A World Bank Study Technology development RSA companies are Major production of Adaptation of metal able to raw steel available processing manufacture metal production lines to Mass production Cost reduction structures of CSP Steel processing CSP products and quality at a techniques available required quality competitive price in automotive industry. No experience with CSP components Provision of complete Enhancing of structure for CSP projects production Exporting capacity RSA provide most of capacities to cover to meet the metal structures export demands neighboring for coal power plants demand Business development Independent production of CSP Foundation of joint metal structures for Huge automotive solar fields in RSA. industry with limited ventures Emerging companies experience in and overall increase cooperating with of industrial other industrial Acquisition of Comprehensive High level of Positive spill-over effects training of sophistication is on other sectors (e.g., PV) potential for CSP players licenses employees reached Use of innovative designs and Important R&D Focus on R&D for Techniques and Patented innovations in materials with the sector with existing design, weight materials reflector designs & aim of enhanced cooperation for a reduction and adapted to maintenance equipment intellectual CSP pilot project accuracy of tracking specific needs in RSA properties and resources Policy framework & market development Strategy funds for National targets for Establishment of industrial upgrade CSP industry are still institutions/ are provided associations to Clear political goals to be agreed upon regarding industrial define and Consolidated support RSA R&D policy and exports FIT have recently national and funding Large number of been adjusted, but strategies for framework R&D competence are still to be agreed industrial clusters created Focused support for upon developments industrial and energy Long-term, stable Intense trade of CSP Growing export of development of CSP targets defined policy framework Favorable tax rates mirrors with RSA CSP mirrors from RSA mirror industry Ongoing discussion exist for mirrors on implementation is implemented neighboring countries of new single-buyer Continuous and Substantial CSP Definition of long- Growing CSP Growing level of confidence Minumum of 100 MW of Minimum installed stable growth of CSP project pipeline term CSP objectives pipeline in CSP technology installed capacity per year capacity of 2 GW market in RSA Source: Fichtner 2011. References Aphane, Ompi. 2010. “South African Government Information, Speeches and Statements, Comments by Mr Ompi Aphane, acting Deputy-Director-General in the Depart- ment of Energy.â€? At press conference on the publication of a request for informa- tion for potential developers of renewable energy projects under the Renewable Energy Feed-in Tariff (REFIT) programme in South Africa, September 30. h p:// www.info.gov.za/speech/DynamicAction?pageid=461&sid=13536&tid=21858. Astrad, Kerstin. 2006. “Examining Influences of EU Policy on Instrument Choice: The Selection of a Green Certiï¬?cate Trading Scheme in Sweden.â€? Bacon, Robert, and Masami Kojima. 2011. “Issues in Estimating the Employment Gener- ated by Energy Sector Activities.â€? Washington, DC: World Bank. Bernhard, R., J. De Lalaing, and others. 2009. “Linear Fresnel Collector Demonstration at the PSA—Operation and Investigation.â€? SolarPACES 2009, Berlin, Germany. Bill Brown Climate Solutions. 2009. “Our New Energy Economy.â€? January. h p:// billbrownclimatesolutions.blogspot.com/2009_01_25_archive.html. Boletín Oï¬?cial del Estado 283/2009. Sec. III, p. 99848. BMU. 2009. “Erneuerbare Energien in Zahlen: Nationale und Internationale Entwick- lungen.â€? Berlin: BMU. Bukala, Thembani. 2009. Interview by Engineering News. February 5. h p://www. engineeringnews.co.za/article/proposed-renewable-tariffs-too-low-to-entice- investors-2009-02-05. CERC. 2009a. “Determination of Generic Levelized Generation Tariff under Regula- tion 8 of the Central Electricity Regulatory Commission (Terms and Conditions for Tariff Determination from Renewable Energy Sources) Regulations, 2009.â€? Petition No. 284/2009 (Suo Motu). — — —. 2009b. “Statement of Objectives and Reasons on CERC Regulations: Terms and Conditions for Tariff determination from Renewable Energy Sources.â€? h p://www.cercind.gov.in/Regulations/Final_SOR_RE_Tariff_Regulations_to_ upload_7_oct_09.pdf. CIF. 2010. “Update on the CSP-MNA Investment Plan. CTF Investment Plan for Con- centrated Solar Power in the Middle East and North Africa Region. Supple- mental Document.â€? October 28, 2010. CTF/TFC.6/Inf.2. October 29. h p://www. climateinvestmentfunds.org/cif/sites/climateinvestmentfunds.org/ï¬?les/CTF%20 Inf%202%20CSP%20MNA%20IP%20nov2010.pdf. CPUC. 2009. System-Side Renewable Distributed Generation Pricing Proposal. h p:// docs.cpuc.ca.gov/eï¬?le/RULINGS/106275.pdf. CSP Today. 2010. Global Concentrated Solar Power Industry Report 2010–2011. London: FC Business Intelligence. Dahir. 2010. Dahir no. 1-10-16 du 26 Safar 1431. February 11. Database of State Incentive for Renewables & Efficiency. Washington, D.C: U.S. DOE. h p://www.dsireusa.org/. del Rio, P., and M. A. Gual. 2007. An Integrated Assessment of the Feed-in Tariff System in Spain. Energy Policy 35:994–1012. 129 130 A World Bank Study Department of Energy. 2009. “Determination Regarding the Integrated Resource Plan and New Generation Capacity.â€? December 31. h p://www.info.gov.za/view/ DownloadFileAction?id=114667. — — —. 2010a. “Procurement Process for Renewable Energy Independent Power Pro- ducers (IPPs) under the Reï¬?t Program Gets Underway with the Analysis of 384 Responses to the Request for Information.â€? December 6. h p://www.energy. gov.za/files/media/pr/DoE%20REFIT%20RFI%20Results%20%20Press%20 Release%20%2024%20November%202010%20%20Draft%203%2000%20 %282%29.pdf. — — —. 2010b. “Republic of South Africa.â€? Request for Information Issued to Potential Developers of Renewable Energy Projects under the REFIT Programme in South Africa (on-shore wind, solar, biomass, biogas, small hydropower and landï¬?ll gas) and Potential Developers of Cogeneration Projects. September 30. Departments of Minerals and Energy Republic of South Africa. 2003. “White Paper on Ren- ewable Energy.â€? November. h p://unfccc.int/ï¬?les/meetings/seminar/application/ pdf/sem_sup1_south_africa.pdf. Durrschmidt, Busgen. 2008. “The Expansion of Electricity Generation from Renewable Energies in Germany: A Review Based on the Renewable Energy Sources Act Progress Report 2007 and the New Feed-in Legislations.â€? Berlin: Federal Ministry of the Environment. Ecostar. 2005. “European Concentrated Solar Thermal Road-Mapping.â€? ECOSTAR. Emerging Energy Research. 2010. “Global Concentrated Solar Power Markets and Strat- egies: 2010–2025.â€? April 2010. Engineering News. 2011. March 22. h p://m.engineeringnews.co.za/article/nersa-moves- to-cut-reï¬?t-tariffs-just-as-sa-promises-to-boost-renewables-2011-03-22. Ernst & Young and Fraunhofer Institute. 2010. “MENA Assessment of the Local Manu- facturing Potential for Concentrated Solar Thermal Power (CSP) Projects.â€? Draft report prepared for the World Bank. EU Directive 2009/28/EC. h p://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=OJ:L: 2009:140:0016:0062:en:PDF. Ernst & Young and Fraunhofer Institute. 2010. “MENA Assessment of the Local Manu- facturing Potential for Concentrated Solar Thermal Power (CSP) Projects.â€? Draft report prepared for the World Bank. Fenwick, Samuel. 2011. “Algeria—170 MW from Concentrated Solar Energy by 2015.â€? January 17. h p://www.evwind.es/noticias.php?id_not=9605. Fichtner. 2010. “Technology Assessment of Concentrated Solar Thermal Power Tech- nologies for a Site Speciï¬?c Project in South Africa.â€? Final Report 5442P10/FICHT- 6156498-v1 for World Bank Project P118730. — — —. 2011. “Assessment of Local Manufacturing Capacities for Solar Power Technolo- gies in South Africa.â€? Fichtner Report 6543P09/FICHT-7130099 for World Bank Project P118730. World Bank. Washington DC. Fouquet, D., and T. B. Johansson. 2008. “European Renewable Energy Policy at Crossroads—Focus on Electricity Support Mechanisms.â€? Energy Policy 36:4079–92. FuturePolicy.org. 2010. Last accessed December 3, 2010. h p://www.futurepolicy.org/ 2689.html. Gan and others. 2007. “Green Electricity Market Development: Lessons from Europe and the US.â€? Energy Policy 35:144–55. Concentrating Solar Power in Developing Countries 131 Gener, A. “Doctoral Thesis.â€? GOA (Government of Algeria). 2002. “Loi No. 02-01 du 22 Dhou El Kaada 1422 cor- respondant au 5 février 2002 relative à l’électricité et à la distribution du gaz par canalizations.â€? h p://www.mipi.dz/doc/fr/55.pdf. Government of Gujarat. 2009. “Government of Gujarat Solar Policy 2009.â€? Resolu- tion No. SLR-11-2008-2176-B. January 6. h p://www.geda.org.in/pdf/Solar%20 Power%20policy%202009.pdf. Grama, Sorin, Elizabeth Wayman, and Travis Bradford. 2008. Concentrating Solar Power— Technology, Cost, and Markets. Cambridge, MA: Prometheus Institute. www. greentechmedia.com. Harrison, John. 2001. “Investigation of Reflective Materials for the Solar Cooker.â€? IDFC. 2009. “The Jawaharlal Nehru National Solar Mission: Implementation Issues on Bundling Scheme.â€? Presentation by Manisha Gulati. h p://www.idfc.com/pdf/ publications/The_Bundling_Scheme_under_the_National_Solar_Mission.pdf. IEA. 2009. “Quarterly Statistics: Oil, Gas, Coal & Electricity.â€? Third Quarter. Paris: OECD. Integrated Resource Plan for Electricity 2010. 2010. Revision 2, Draft Version 8, October 8. h p://www.doe-irp.co.za/content/INTEGRATED_RESOURCE_PLAN_ ELECTRICITY_2010_v8.pdf. Integrated Resource Plan for Electricity 2010–2030. 2011. Revision 2, Final report, March 25. h p://www.doe-irp.co.za/content/IRP2010_2030_Final_Report_20110325.pdf. JORADP (Journal Officiel de la Republique Algerienne). 2004. No. 19. 7 Safar 1425. March 28. h p://www.futurepolicy.org/ï¬?leadmin/user_upload/Axel/Algeria_ FIT_2004.pdf. Kearney, A. T. 2010. “Solar Thermal Electricity 2025. Clean Electricity on Demand: A ractive STE Cost Stabilize Energy Production.â€? Report commissioned by Euro- pean Solar Thermal Electricity Association (ESTELA). Kearney, D., and others. 2004. “Engineering Aspects of a Molten Salt Heat Transfer Fluid in a Parabolic Trough Solar Field.â€? Energy 29:861–70. Kubert, Mark, and Charles Sinclair. 2010. A Review of Emerging State Finance Tools to Advance Solar Generation. Clean Energy States Alliance. Lorenzoni, A. 2003. “The Italian Green Certiï¬?cates Market between Uncertainty and Opportunities.â€? Energy Policy 31:33–42. Lund, Morten, William H. Holmes, Stephen Hall, Jennifer Martin. 2009. “Lex Helius: The Law of Solar Energy.â€? Portland, Oregon: Stoel Rives LLP. h p://www.stoel.com/ webï¬?les/lawofsolarenergy.pdf. MNRE (Ministry of New and Renewable Energy). 2008. “Guidelines for Generation Based Incentive: Grid Interactive Solar Thermal Power Generation Projects.â€? MNRE, January. h p://www.mnre.gov.in/pdf/guidelines_spg.pdf. — — —. 2009. “Jawaharlal Nehru National Solar Mission: Towards Building Solar India.â€? h p://www.indiaenvironmentportal.org.in/content/jawaharlal-nehru- national-solar-mission-towards-building-solar-india. NERSA. 2009a. “NERSA Decision on Renewable Energy Feed-In Tariffs (REFITS) Phase II, Media Statement.â€? November 2. — — —. 2009b. “NERSA Media Statement, NERSA Decision on Renewable Energy Feed-in Tariff (REFIT).â€? March 31. Nielsen, L., and T. Jeppesen. 2003. “Tradable Green Certiï¬?cates in Selected European Countries—Overview and Assessment.â€? Energy Policy 31:3–14. 132 A World Bank Study Nilsson and Sundqvist. 2006. “Using the Market at a Cost: How the Introduction of Green Certiï¬?cates in Sweden Led to Market Inefficiencies.â€? Utilities Policy 15:49–59. NOVI Energy. 2011. “Regulatory and Financial Incentives for Scaling up Concentrating Solar Power in Developing Countries—Procurement Practices Analysis.â€? Report prepared for World Bank Project P118730. World Bank, Washington, DC. Panchabuta. 2010a. “Gujarat Urja Vikas Nigam Ltd (GUVNL) Signs PPA for 537 MW with 54 Solar Power Generation Companies Taking Total Signed to 933.5 MW.â€? h p:// panchabuta.wordpress.com/2010/12/28/gujarat-urja-vikas-nigam-ltd-guvnl-signs- ppa-for-537mw-with-54-solar-power-generation-companies-taking-total-signed- at-933-5mw/. — — —. 2010b. “Solar PV and Thermal Developers Shortlisted for the First Phase of JNNSM (National Solar Mission).â€? h p://panchabuta.wordpress.com/2010/11/17/solar- pv-and-thermal-developers-shortlisted-for-the-first-phase-of-jnnsm-national- solar-mission/. PEW Charitable Trusts. 2010. “Who’s Winning the Clean Energy Race? Growth, Com- petition and Opportunity in the World’s Largest Economies.â€? Washington, DC: PEW Charitable Trusts. Producer Prices. “Principal Global Indicators.â€? h p://www.principalglobalindicators. org/default.aspx. Quaschning, Volker. 2003. “Solar Thermal Power Plants—Technology Fundamentals.â€? Erneuerbare-Energien-und-Klimaschu .de. h p://www.volker-quaschning.de/ articles/fundamentals2/index.php. Radiant & Hydronics. 2006. “AZ Solar Station to Harnesses Sun’s Heat for Power.â€? http://www.radiantandhydronics.com/Archives/BNP_GUID_9-5-2006_A_ 10000000000000267639. Rowlands, Ian. 2004. “Envisaging Feed-in Tariffs for Solar Photovoltaic Electricity: European Lessons for Canada.â€? Department of Environment and Resource Stud- ies, Faculty of Environmental Studies, University of Waterloo, Waterloo, Canada. Sarangi, Gopal K., and Arabinda Mishra. 2009. “Environmental Innovation in Electricity Sector in India: Role of Electricity Regulators.â€? TERI. h p://www.dime-eu.org/ ï¬?les/active/0/Sarangi_Dime-Workshop-Paper%20_Draft-Final_.pdf. Shinnar, R., and F. Citro. 2007. “Solar Thermal Energy: The Forgo en Energy Source.â€? Technology in Society 29:261–70. Solar Dish Engine. N.d. h p://www.solarpaces.org/CSP_Technology/docs/solar_dish.pdf. TNERC (Tamil Nadu Electricity Regulatory Commission). 2010. “Consultative Paper on Comprehensive Tariff Order for Solar Photovoltaic and Solar Thermal Power Plants up to 3 MW having grid connectivity below 33 kV level.â€? h p://tnerc.gov. in/Concept%20Paper/2010/Solar%20CP%20on.pdf. U.S. DOE (U.S. Department of Energy). N.d. “Linear Concentrator Systems for Con- centrating Solar Power.â€? h p://www.eere.energy.gov/basics/renewable_energy/ linear_concentrator.html. — — —. 2011. Database of State Incentives for Renewables & Efficiency (DSIRE). Washing- ton, DC: U.S. DOE. h p://www.dsireusa.org/.US. Van de Merwe, C. 2010. “Wind Developers Show Strong Appetite for SA Projects.â€? October 8. h p://www.engineeringnews.co.za/article/second-wind-seminar-held- in-johannesburg-2010-10-08-1. Concentrating Solar Power in Developing Countries 133 Viebahn, Peter, Stefan Kronshage, Franz Trieb (DLR), and Yolanda Lechon (CIEMAT). 2008. “Final Report on Technical Data, Costs, and Life Cycle Inventories of Solar Thermal Power Plants. New Energy Externalities Developments for Sustain- ability.â€? Integrated Project. DLR, CEIMAT. h p://www.needs-project.org/RS1a/ RS1a%20D12.2%20Final%20report%20concentrating%20solar%20thermal%20 power%20plants.pdf. VZB (Vebraucherzentralen Bundesverband eV). 2010. “Die künftige Förderung von Solarstrom über das Erneuerbare-Energien-Gese .â€? Berlin: VZB. World Bank. World Development Indicators (WDI) & Global Development Finance (GDF). World Bank Databank. h p://databank.worldbank.org/. World Bank/ESMAP. 2010. “Study on Barriers for Solar Power Development in India.â€? Washington, DC: World Bank, July. — — —. 2011a. “Electricity Auctions: An Overview of Efficient Practices.â€? Luiz Maurer, Luiz Barrozo, and others. Conference Edition. Washington, DC: World Bank, February. YES/Nixus/CENER. 2010. “Activity 1.1. Review of CST Technologies.â€? Final report pre- pared for the World Bank under Project Number P119536. Washington, DC. Chapter Bibliographies Chapter 2 Bibliography The chapter was based on the following reports: YES/Nixus/CENER. 2010. “Activity 1.1. Review of CST Technologies.â€? Final report prepared for the World Bank under Project Number P119536. World Bank, Washington DC. Fichtner. 2010. “Technology Assessment of Concentrated Solar Thermal Power Tech- nologies for a Site Speciï¬?c Project in South Africa.â€? Final Report 5442P10/FICHT- 6156498-v1 for World Bank Project P118730. World Bank, Washington DC. Bibliography AECON Constructors. “C.N. Tower Limited.â€? Albiasa. 2009. h p://www.albiasasolar.com. Allen, N., and M. Edge. 1992. “Fundamentals of Polymer Degradation and Stabilization.â€? London and New York: Elsevier. Assessment of the World Bank/GEF Strategy for the Market Development of Concentrat- ing Solar Thermal Power. 2005. GEF Council. June 3–8, 2005. AUSRA. 2009. “The Liddell Solar Thermal Station.â€? Baker, A. F. 1989. “US-Spain Joint Evaluation of the Solar One and Cesa-1 Receiver and Storage Systems.â€? Bautista, M. C., and A. Morales. “Silica Antireflective Films on Glass Produced by the Sol-Gel Method.â€? Solar Energy Materials and Solar Cells 80(2003):217–25. Benson, B. A. 1985. “Silver/Polymer Films for Concentrators.â€? Solar Thermal Research Program Annual Conference. SERI/CP-251-2680. Benz, N., and others. 2008. “Advances in Receiver Technology for Parabolic Troughs.â€? In Proceedings of 14th International SolarPACES Symposium on Solar Thermal Concentrating Technologies, Las Vegas, NV. Bernhard, R., J. De Lalaing, and others. 2009. “Linear Fresnel Collector Demonstration at the PSA—Operation and Investigation.â€? SolarPACES 2009, Berlin, Germany. Brost, R., Al Gray, F. Burkholder, T. Wendelin, and D. White. 2009. “Skytrough Optical Evaluations Using Vshot Measurement.â€? SolarPACES 2009, Berlin, Germany. Burbidge, D., Mills, D., and others. 2000. Stanwell Solar Thermal Power Project. 10th SolarPACES International Symposium of Solar Thermal Concentrating Technolo- gies, Sydney. Burgaleta, J. I. 2009. “Gemasolar: Tecnología de torre central y almacenamiento con sales fundidas.â€? 3a Cumbre de concentración solar termoeléctrica, November (Torresol Energy). Carmichael, D. C., G. B. Gaines, F. A. Sliemers, C. W. Kistler, and R. D. Igou. 1976. “Review of World Experience and Properties of Materials for Encapsulation of Terrestrial Photovoltaic Arrays.â€? ERDA/JPL/954328-76/4. Carreras, L., and F. Montalá. 2003. “Actualidad Industrial de las Técnicas de Recu- brimientos de Capas Duras Finas.â€? Ibérica de Tecnología 404(June). Castañeda, Nora. 2006. “Sener parabolic trough collector design and testing.â€? 134 Concentrating Solar Power in Developing Countries 135 Davenport, R., and R. Taylor. 2009. “Low Cost Glass-Reinforced Concrete Heliostats.â€? SolarPACES 2009, Berlin, Germany. de Vries, Eize. 2009. “Concrete-Steel Hybrid Tower from ATS.â€? Renewable Energy World, October. Dersch, J., and others. 2004. “Trough Integration into Power Plants—A Study on the Performance and Economy of Integrated Solar Combined Cycle Systems.â€? Energy 29(2004):947–59. Econoticias. “Lanzamiento de SkyTrough.â€? h p://www.ecoticias.com/20081022- lanzamiento-de-skytrough.htm. Emerging Energy Research. 2010. “Global Concentrated Solar Power Markets and Strat- egies: 2010–2025.â€? April 2010. Fabrizi, F. 2007. “Trough Molten Salt HTF Field Test Experience: Experimental Remarks on Behaviour during Operation and Thermal Fluid Dynamics in Transition States of Molten Salt Mixtures.â€? NREL Parabolic Trough Technology Workshop, March 8–9, 2007, NREL Denver West Business Park, Golden, CO. Falcone, P. K. 1986. “A Handbook for Solar Central Receiver Design.â€? SAND86-8009, Sandia National Laboratories, Livermore, CA. Fernández-García, A., E. Zarza, and others. “Parabolic-Trough Solar Collectors and Their Applications.â€? Renewable and Sustainable Energy Reviews. In press, cor- rected proof. García, G., A. Egea. 2000. “El Helióstato Autónomo.â€? Madrid: Ciemat, p. 82. Gener, A. “Tesis Doctoral.â€? Gereffi, G., and K. Dubay. 2008. Concentrating Solar Power—Clean Energy for the Elec- tric Grid. Center on Globalization, Governance and Competitiveness. Geyer, M., and others. 2006. “Dispatchable Solar Electricity for Summerly Peak Loads from the Solar Thermal Projects Andasol-1 and Andasol-2.â€? Proceedings Solar- PACES2006 A4-S2, 13th International Symposium on Concentrating Solar Power and Solar Energy Technologies, Seville, Spain, June 20, 2006. Harrison, John. 2001. “Investigation of Reflective Materials for the Solar Cooker.â€? Herrarte, M. 1976. “Estudio Comparativo de Encofrados Metálicos.â€? Escuela de Ingeni- ería, Universidad Mariano Galvez de Guatemala. Hoyer, C., and others. 2009. “Performance and Cost Comparison of Linear Fresnel and Parabolic Trough Collectors.â€? SolarPACES 2009, Berlin, Germany. Inï¬?nia. h p://www.inï¬?niacorp.com/applications/solar/iss_index.html. Jorgensen, G. 1993. “Reflective Coatings for Solar Applications.â€? NREL/TP-471-5536. Jorgensen, G., and R. Govindarajan. 1991. “Ultraviolet Reflector Materials for Solar Detoxiï¬?cation of Hazardous Waste.â€? SERI/TP—257-4418. Kearney, D., and others. 2004. “Engineering Aspects of a Molten Salt Heat Transfer Fluid in a Parabolic Trough Solar Field.â€? Energy 29:861–70. Kennedy, C. E., and H. Price. 2006. “Progress in Development of High-Temperature Solar-Selective Coating.â€? American Society of Mechanical Engineers, New York, NY 10016-5990. Kennedy, C. E., K. Terwilliger. 2005. “Optical Durability of Candidate Solar Reflectors.â€? Transactions of the ASTME 127:262–69. Kennedy, C. E., Michael Milbourne, Hank Price, and Kent Terwilliger. 2003. “Summary of Status of Most Promising Candidate Advanced Solar Mirrors and Absorber Materials (Testing and Development Activities).â€? 136 A World Bank Study Kennedy, C. E., R. V. Smilgys. 2002. “Durability of Solar Reflective Materials with an Alum- ina Hard Coat Produced by Ion-Beam-Assisted Deposition.â€? NREL/CP-520-32824. Lerchenmüller, H., G. Morin, and others. 2004. “Plug-in Strategy for Market Introduc- tion of Fresnel-Collectors.â€? 12th SolarPACES Intern. Symposium, Mexico. Lopez, C., and K. Stone. 1993. “Performance of the Southern California Edison Company Stirling Dish.â€? SAND93-7098, Sandia National Laboratories, Albuquerque, NM. Lovegroove, K. 2010 “A 500 m2 Paraboloidal Dish Solar Concentrator.â€? Lüpfert, E., and others. 2000. “Eurotrough—A New Parabolic Trough Collector with Advanced Light Weight Structure.â€? In Solar Thermal 2000 International Confer- ence. Sydney, Australia. Maccari, A. 2008. “The ENEA’s Way to Concentrating Solar Power.â€? Solar Innovation Today, Embassy of the Kingdom of the Netherlands, 30/05/2008. Mancini, T. R. 2000. “Technical Report No. III—1/00 Catalog of Solar Heliostats.â€? IEA-Solar Power and Chemical Energy Systems. Task III: Solar Technology and Applications, June. Marie a Silos, LLC. 2010. www.marie asilos.com (consulted in April). Mavis, C. L. 1989. “A Description and Assessment of Heliostat Technology.â€? SAND87- 8025, Sandia National Laboratories, January. Mills, D., and G. Morrison. 2000. “Compact Linear Fresnel Reflector Solar Thermal Powerplants.â€? Solar Energy 68(3):263–83. Mills, D., G. Morrison, and others. 2002. Project Proposal for a Compact Linear Fres- nel Reflector Solar Thermal Plant in the Hunter Valley. ANZSES Annual Confer- ence—Solar Harvest, Newcastle, Australia. Mills, D. R., P. Schramek, C. Dey, D. Briue, I. Imenes, B. S. Haynes, and G. L. Morrison. 2002. “Multi Tower Solar Array Project.â€? Paper 1b4, ANZSES Annual Conference— Solar Harvest, Newcastle. Moens, L., D. Blake D. 2008. “Mechanism of Hydrogen Formation in Solar Parabolic Trough Receivers.â€? NREL TP-510-42468. Monterreal, R., M. Romero, G. García, and G. Barrera. 1997. “Development and Testing of a 100 m2 Glass-Metal Heliostat with a New Local Control System.â€? In D. E. Claridge and J. E. Pacheco (eds.), Solar Engineering, 1997, New York: ASME, ISBN: 0-7918-1556-0, pp. 251–59. Morin, G., W. Pla er, and others. 2006. Road Map towards the Demonstration of a Lin- ear Fresnel Collector Using a Single Tube Receiver. 13th SolarPACES International Symposium, Seville, Spain. Morrison, G. 2006. “Large Scale Solar Thermal Electricity.â€? Nepveu, F. 2008. “Production décentralisée d’électricité et de chaleur par système Parabole/Stirling: Application au système EURODISH.â€? Perpignan, France: Uni- versité de Perpignan, p. 279. Novatec Biosol. 2009. “Novatec Solar Field Improved 50 MW Project Economics through the Use of Novatec Technology.â€? November. Ortiz Vives, F., M. Meyer-Grünefeldt. 2009. Flexible Hose System—Rotationflex® Con- nection to HCE of Parabolic Collectors. SolarPACES 2009, Berlin, Germany. Osuna, R., F. Cerón, M. Romero, and G. García. 1999. “Desarrollo de un Prototipo de Helióstato para la Planta Colón Solar.â€? Energía XXV(6):71–79. Pincemin, S., R. Olives, and others. 2008. “Highly Conductive Composites Made of Phase Change Materials and Graphite for Thermal Storage.â€? Solar Energy Materials and Solar Cells 92(6):603–13. Concentrating Solar Power in Developing Countries 137 Pi -Paal, R., and others. 2005. “ECOSTAR Road Map Document.â€? Porter, M. E. 1985. Competitive Advantage. The Free Press, New York, 1985. Price, H., and V. Hassani. 2002. “Modular Trough Power Plant Cycle and Systems Analy- sis.â€? Report No. NREL/TP-550-31240, NREL, Colorado. Relloso, S., N. Castañeda, and M. Domingo. 2008. “New Senertough Collector Develop- ment in Collaboration with Key Components Suppliers.â€? In SolarPACES 2008, 14th International Symposium on Concentrating Solar Power and Chemical Energy Technologies, Las Vegas, NV. Riffelmann, K.-J., J. Kö er, P. Nava, F. Meuser, G. Weinrebe, W. Schiel, G. Kuhlmann, A. Wohlfahrt, A. Nady, and R. Dracker. 2009. Heliotrough—A New Collector Gen- eration for Parabolic Trough Power Plants. SolarPACES 2009, Berlin, Germany. Romero, M., and E. Zarza. “Handbook of Energy Efficiency and Renewable Energy.â€? Romero, M., E. Conejero, and M. Sánchez. 1991 “Recent Experiences on Reflectant Mod- ule Components for Innovative Heliostats.â€? Solar Energy Materials 24:320–32. Sánchez, M. 1990. “Desarrollo y caracterización de primera versión de polímero con alta reflectividad en el UV.â€? Informe CIEMAT. IER/R2D04/IT. — — —. 1995. “Caracterización y degradación de polímeros reflectantes para aplica- ciones solares mediante métodos espectroscópicos.â€? Memoria Tesis Doctoral, CIEMAT-UCM. Schissel, P., G. Jorgensen, R. Pi s. 1991. “Application Experience and Field Performance of Silvered Polymer Reflectors.â€? NREL/TP—257-4146. Schissel, P., H. H. Neidlinger, A. W. Czanderna. 1985. “Polymer Reflector Research during FY 1985.â€? SERI/PR—225-2835. Schlaig Bergermann und Partner. 2001. “Eurodish–Stirling. System Description.â€? Siemens. 2008. “Solar Thermal Power Plants. Industrial Steam Turbines.â€? Silberstein, Elon, and others. “Brightsource Solar Tower Pilot in Israel’s Negev Opera- tion at 130 Bar @ 530°c Superheated Steam.â€? Singh, A. N. 2007. “Concrete Construction for Wind Energy Towers.â€? Indian Concrete Journal (August). Skyfuel. 2010. “Design of a High-Temperature Molten Salt Linear Fresnel Collector.â€? Presentation. February 10, 2010. Slipform International. 2010. www.slipform-int.com (consulted in April). Smilgys, R. V. 2005. “Production of Solar Reflective Materials Using a Laboratory-Scale Roll Coater.â€? NREL/SR-520-37007. Solar Dish Engine. N.d. h p://www.solarpaces.org/CSP_Technology/docs/solar_dish.pdf. Status Report on Solar Trough Power Plants. 1996. German Federal Minister for Educa- tion, Science, Research and Technology. Contract No. 0329660. Steinmann, W.-D., D. Laing, and others. 2008. “Latent Heat Storage Systems for Solar Thermal Power Plants and Process Heat Applications.â€? SolarPaces 2008. Las Vegas, NV. Ubach, J., F. Miranda, C. Castañon, F. Ainz, and J. Martínez. 2009. “Rioglass Solar’s Glass Tempered Solar Mirrors: A Successful Approach.â€? SolarPACES 2009. Berlin, Germany. Vazquez, J., and others. 2008. “Sener Parabolic Trough Collector Design and Testing.â€? SolarPaces 2008. 14th International Symposium on Concentrating Solar Power and Chemical Energy Technologies, Las Vegas, NV. 138 A World Bank Study Winter, C.-J., and R. L. Sizmann. 1991. “Solar Power Plants.â€? Winter, C.-J., R. L. Sizmann, L. L. Vaut Hull. 1991. “Solar Power Plants: Fundamentals— Technology—Systems—Economics.â€? Heidelberg: Springer-Verlag Berlin. Chapter 3 Bibliography APPA. 2009. “Estudio del impacto macroeconómico de las energías renovables en España.â€? Barcelona: Asociación de Productores de Energías Renovables. Astrad, Kerstin. 2006. “Examining Influences of EU Policy on Instrument Choice: The Selection of a Green Certiï¬?cate Trading Scheme in Sweden.â€? BMU. 2009. “Erneuerbare Energien in Zahlen: Nationale und Internationale Entwick- lungen.â€? Berlin: BMU. Durrschmidt, Busgen. 2008. “The Expansion of Electricity Generation from Renewable Energies in Germany: A Review Based on the Renewable Energy Sources Act Progress Report 2007 and the New Feed-in Legislations.â€? Berlin: Federal Ministry of the Environment. CPUC. 2009. System-Side Renewable Distributed Generation Pricing Proposal. h p:// docs.cpuc.ca.gov/eï¬?le/RULINGS/106275.pdf. CSP Today. 2010. Global Concentrated Solar Power Industry Report 2010–2011. London: FC Business Intelligence. Database of State Incentive for Renewables & Efficiency. Washington, D.C: U.S. DOE. h p://www.dsireusa.org/. del Rio, P., and M. A. Gual. 2007. An Integrated Assessment of the Feed-in Tariff System in Spain. Energy Policy 35:994–1012. Fouquet, D., and T. B. Johansson. 2008. “European Renewable Energy Policy at Cross- roads—Focus on Electricity Support Mechanisms.â€? Energy Policy 36:4079–92. Gan and others. 2007. “Green Electricity Market Development: Lessons from Europe and the US.â€? Energy Policy 35:144–55. Grama, Sorin, Elizabeth Wayman, and Travis Bradford. 2008. Concentrating Solar Power—Technology, Cost, and Markets. Cambridge, MA: Prometheus Institute. www.greentechmedia.com. Greenpeace International and others. 2009. “Concentrating Solar Power: Global Outlook 09.â€? Amsterdam: Greenpeace International. Haas and others. 2004. “How to Promote Renewable Energy Systems Successfully and Effectively.â€? Energy Policy 32:833–39. h p://www.lgprogram.energy.gov. IEA. 2009. “Quarterly Statistics: Oil, Gas, Coal & Electricity.â€? Third Quarter. Paris: OECD. Kubert, Mark, and Charles Sinclair. 2010. A Review of Emerging State Finance Tools to Advance Solar Generation. Clean Energy States Alliance. Lorenzoni, A. 2003. “The Italian Green Certiï¬?cates Market between Uncertainty and Opportunities.â€? Energy Policy 31:33–42. Nielsen, L., and T. Jeppesen. 2003. “Tradable Green Certiï¬?cates in Selected European Countries—Overview and Assessment.â€? Energy Policy 31:3–14. Nilsson and Sundqvist. 2006. “Using the Market at a Cost: How the Introduction of Green Certiï¬?cates in Sweden Led to Market Inefficiencies.â€? Utilities Policy 15:49–59. PEW Charitable Trusts. 2010. “Who’s Winning the Clean Energy Race? Growth, Com- petition and Opportunity in the World’s Largest Economies.â€? Washington, D.C.: PEW Charitable Trusts. Concentrating Solar Power in Developing Countries 139 Public Interest Energy Research Program. 2007. “Government Actions and Innovation in Clean Energy Technologies: The Cases of Photovoltaic Cells, Solar Thermal Elec- tric Power, and Solar Water Heating.â€? REN21. 2009. Renewables Global Status Report: 2009. Update. Paris: REN21 Secretariat. Rowlands, Ian. 2004. “Envisaging Feed-in Tariffs for Solar Photovoltaic Electricity: Euro- pean Lessons for Canada.â€? Department of Environment and Resource Studies, Fac- ulty of Environmental Studies, University of Waterloo, Waterloo, Canada. VZB (Vebraucherzentralen Bundesverband eV). 2010. “Die künftige Förderung von Solar- strom über das Erneuerbare-Energien-Gese .â€? Berlin: VZB. Wiser and others. 2005. “Evaluating Experience with Renewable Portfolio Standards in the United States.â€? Berkeley: Berkeley National Laboratory. Chapter 4 Bibliography Aphane, Ompi. 2010. “South African Government Information, Speeches and Statements, Comments by Mr Ompi Aphane, acting Deputy-Director-General in the Depart- ment of Energy.â€? At press conference on the publication of a request for informa- tion for potential developers of renewable energy projects under the Renewable Energy Feed-in Tariff (REFIT) programme in South Africa, September 30. h p:// www.info.gov.za/speech/DynamicAction?pageid=461&sid=13536&tid=21858. Balch, Oliver. 2010. “India’s Solar Mission: Phase One Guidelines Fall Short of Industry Expectations.â€? CSP Today. h p://social.csptoday.com/industry- insight/india%E2%80%99s-solar-mission-phase-one-guidelines-fall-short- industry-expectations. Boletín Oï¬?cial del Estado 283/2009. Sec. III, p. 99848. Bukala, Thembani. 2009. Interview by Engineering News. February 5. h p://www. engineeringnews.co.za/article/proposed-renewable-tariffs-too-low-to-entice- investors-2009-02-05. CERC. 2009. “Statement of Objectives and Reasons on CERC Regulations: Terms and Conditions for Tariff determination from Renewable Energy Sources.â€? h p:// www.cercind.gov.in/Regulations/Final_SOR_RE_Tariff_Regulations_to_ upload_7_oct_09.pdf. — — —. 2010a. “Notiï¬?cation Dated February 25, 2010, in Respect of Petition No. 13/2010.â€? h p://www.indiaenvironmentportal.org.in/ï¬?les/tariff.pdf. — — —. 2010b. “Petition No. 53/2010.â€? h p://www.cercind.gov.in/2010/ORDER/February 2010/53-2010_Suo-Motu_RE_Tariff_Order_FY2010-11.pdf. CIF. 2009. “Clean Technology Fund Investment Plan for Concentrated Solar Power in the Middle East and North Africa Region.â€? CTF/TFC.IS.1/3. November 10. h p:// www.climateinvestmentfunds.org/cif/sites/climateinvestmentfunds.org/files/ mna_csp_ctf_investment_plan_kd_120809.pdf — — —. 2010. “Update on the CSP-MNA Investment Plan. CTF Investment Plan for Con- centrated Solar Power in the Middle East and North Africa Region. Supplemental Document.â€? October 28, 2010. CTF/TFC.6/Inf.2. October 29. h p://www.climate- investmentfunds.org/cif/sites/climateinvestmentfunds.org/ï¬?les/CTF%20Inf%20 2%20CSP%20MNA%20IP%20nov2010.pdf. Creamer, Terence. 2010. “First Solar Park Deals Possible in Early 2011, First Power by End of 2012. September 28.â€? Engineering News. h p://www.engineeringnews.co.za/article/ ï¬?rst-solar-park-deals-possible-in-early-2011-ï¬?rst-power-by-end-2012-2010-09-28. 140 A World Bank Study Department of Energy. 2009. “Determination Regarding the Integrated Resource Plan and New Generation Capacity.â€? December 31. h p://www.info.gov.za/view/ DownloadFileAction?id=114667. — — —. 2010a. “Procurement Process for Renewable Energy Independent Power Producers (IPPs) under the Reï¬?t Program Gets Underway with the Analysis of 384 Respons- es to the Request for Information.â€? December 6. h p://www.energy.gov.za/ï¬?les/ media/pr/DoE%20REFIT%20RFI%20Results%20%20Press%20Release%20%20 24%20November%202010%20%20Draft%203%2000%20%282%29.pdf. — — —. 2010b. “Republic of South Africa.â€? Request for Information Issued to Potential Developers of Renewable Energy Projects under the REFIT Programme in South Africa (on-shore wind, solar, biomass, biogas, small hydropower and landï¬?ll gas) and Potential Developers of Cogeneration Projects. September 30. Departments of Minerals and Energy Republic of South Africa. 2003. “White Paper on Renewable Energy.â€? November. h p://unfccc.int/ï¬?les/meetings/seminar/applica tion/pdf/sem_sup1_south_africa.pdf. Deshmukh, Ranjit, Ashwin Gambhir, and Girish Sant. 2010. “Need to Realign India’s National Solar Mission.â€? Economic & Political Weekly XLV(12), March 20. h p:// www.indiaenvironmentportal.org.in/ï¬?les/Need%20to%20Realign%20Indias%20 National%20Solar%20Mission.pdf. Dii. 2010. “Energy from the Desert.â€? Dii Industry Initiative Annual Conference & Expo 2010. h p://www.dii-eumena.com/annual-conference/conference-programme.html. Edkins, M., H. Winkler, A. Marquard. 2009. “Large-Scale Rollout of Concentrating Solar Power in South Africa.â€? Climate Strategies, September. Embassy of France in the United States. 2010. “Transgreen Project: Developing Renew- able Energy in Africa and Europe.â€? Communiqué issued by the Ministry for Ecol- ogy, Energy, Sustainable Development and Marine Affairs, Responsible for Green Technology and Climate Negotiations. Paris, July 5. h p://ambafrance-us.org/ spip.php?article1746. Engineering News. 2010. “Upington Solar Park Could Deliver First Power by the End of 2012.â€? October 1. h p://www.engineeringnews.co.za/article/ï¬?rst-solar- park-deals-possible-in-early-2011-ï¬?rst-power-by-end-2012-2010-10-08. — — —. 2011. March 22. h p://m.engineeringnews.co.za/article/nersa-moves-to-cut-reï¬?t- tariffs-just-as-sa-promises-to-boost-renewables-2011-03-22. Eskom. “Company Information.â€? h p://www.Eskom.co.za/live/content.php?Category_ ID=59. — — —. “Key Facts.â€? h p://www.Eskom.co.za/annreport10/proï¬?le_key_facts.htm. EVI 2011. “Summary of Tariff Bidding Process for Solar Projects under NSM.â€? Emergent Ventures Publication (EVI). January. h p://docs.google.com/viewer?a=v&q=cach e:gA1w2RyBcsoJ:www.emergent-ventures.com/UploadedFiles/Videos/Summary %2520of%2520Tariff%2520Bidding%2520Process%2520For%2520Solar%2520Proj ects%2520under%2520NSM.pdf+Aurum+Renewables+Rajasthan+National+Solar +Mission&hl=en&gl=us&pid=bl&srcid=ADGEEShCtD4xEGNnXhkV3bIAGLTox cycz-nNDiSHQk5yQqG2S73IirbRhc1Fwx-pBBUj2MbO4e9rhJiRGA0nKc6XOJrigx IOcbUMaTGR2a7cAit0yxXQFdQVPNPse0UHzZcF-nF2VMUD&sig=AHIEtbRe-pS tquSbWTVlWZveo5ph9pvgFQ&pli=1. Fenwick, Samuel. 2011. “Algeria—170 MW from Concentrated Solar Energy by 2015.â€? January 17. h p://www.evwind.es/noticias.php?id_not=9605. Concentrating Solar Power in Developing Countries 141 Fichtner. 2010. “Technology Assessment of Concentrated Solar Thermal Power Tech- nologies for a Site Speciï¬?c Project in South Africa.â€? Final Report 5442P10/FICHT- 6156498-v1 for World Bank Project P118730. FuturePolicy.org. 2010. Last accessed December 3, 2010. h p://www.futurepolicy. org/2689.html. GET FiT. 2010. “GET FiT Program: Global Energy Transfer Feed-in Tariffs for Devel- oping Countries.â€? DB Climate Change Advisors. Deutsche Bank Group, April. h p://www.dbcca.com/dbcca/EN/_media/GET_FiT_Program.pdf. Gipe, Paul. 2009. “Algerian Feed-in Tariff Decree 2004.â€? h p://www.wind-works.org/ FeedLaws/Algeria/AlgerianFeed-inTariffDecree2004.html. GOA (Government of Algeria). 2002. “Loi No. 02-01 du 22 Dhou El Kaada 1422 cor- respondant au 5 février 2002 relative à l’électricité et à la distribution du gaz par canalizations.â€? h p://www.mipi.dz/doc/fr/55.pdf. GOI. 2010. “Jawaharlal Nehru National Solar Mission: Building Solar India. Guidelines for Selection of New Grid Connected Solar Power Projects.â€? h p://www.mnre. gov.in/pdf/jnnsm-gridconnected-25072010.pdf. Government of Gujarat. 2009. “Government of Gujarat Solar Policy 2009.â€? Resolu- tion No. SLR-11-2008-2176-B. January 6. h p://www.geda.org.in/pdf/Solar%20 Power%20policy%202009.pdf. h p://www.nersa.org.za/Admin/Document/Editor/ï¬?le/Electricity/Legislation/Regula tory%20Rules/RULES%20FOR%20SELECTION%20CRITERIA%2019%20Feb10. pdf. IDFC. 2009. “The Jawaharlal Nehru National Solar Mission: Implementation Issues on Bundling Scheme.â€? Presentation by Manisha Gulati. h p://www.idfc.com/pdf/ publications/The_Bundling_Scheme_under_the_National_Solar_Mission.pdf. Integrated Resource Plan for Electricity 2010. 2010. Revision 2, Draft Version 8, October 8. h p:// www.doe-irp.co.za/content/INTEGRATED_RESOURCE_PLAN_ELECTRICITY_ 2010_v8.pdf. Integrated Resource Plan for Electricity 2010–2030. 2011. Revision 2, Final report, March 25. h p://www.doe-irp.co.za/content/IRP2010_2030_Final_Report_20110325.pdf. JMEMR (Jordan Ministry of Energy and Mineral Resources). 2010. “Renewable Energy and Energy Efficiency in Jordan.â€? h p://www.narucpartnerships.org/Documents/ Jordan%20RE%20Opportunities-%20Ziad-%20May%202010.pdf. JORADP (Journal Officiel de la Republique Algerienne). 2004. No. 19. 7 Safar 1425. March 28. h p://www.futurepolicy.org/ï¬?leadmin/user_upload/Axel/Algeria_ FIT_2004.pdf. Koch-Weser, Caio, and George Soros. 2010. “New Green Drivers of Growth.â€? November 9. h p://www.project-syndicate.org/commentary/kochweser1/English. Le Treut, H., R. Somerville, U. Cubasch, Y. Ding, C. Mauri en, A. Mokssit, T. Peter- son, and M. Prather. 2007. “Historical Overview of Climate Change.â€? In Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor and H. L. Miller (eds.). Climate Change 2007: The Physical Science Basis. Contribution of Work- ing Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge, United Kingdom and New York, NY: Cambridge University Press. h p://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1- chapter1.pdf. 142 A World Bank Study “Long Term Mitigation Scenarios Technical Summary.â€? 2007. Energy Research Centre, October. h p://www.environment.gov.za/HotIssues/2009/LTMS2/LTMSTechnical Summary.pdf. MEDEMIP. 2009. “MEDRING Update.â€? Second Coordination Meeting. Presentation by Dr. Albrecht Kaupp, Tunis, Tunisia, December 13. h p://www.medemip.eu/Calc/ FM/MED-EMIP/MEDRING_Study_Update/2nd_meeting/MEDRING-2nd_ Coordination_Meeting.pdf. MNRE (Ministry of New and Renewable Energy). 2008. “Guidelines for Generation Based Incentive: Grid Interactive Solar Thermal Power Generation Projects.â€? MNRE, January. h p://www.mnre.gov.in/pdf/guidelines_spg.pdf. — — —. 2009. “Jawaharlal Nehru National Solar Mission: Towards Building Solar India.â€? h p://www.indiaenvironmentportal.org.in/content/jawaharlal-nehru-national- solar-mission-towards-building-solar-india. — — —. 2010. “Guidelines for Migration of Existing Under Development Grid Connected Solar Projects from Existing Arrangements to the Jawaharlal Nehru National Solar Mission (JNNSM).â€? h p://www.mnre.gov.in/pdf/migration-guidelines- jnnsm.pdf. MSP 2010. “Mediterranean Solar Plan: Strategy Paper.â€? Document examined by the MSP Experts Group on 10/02/2010. h p://ec.europa.eu/energy/international/ international_cooperation/doc/2010_02_10_mediterranean_solar_plan_strategy_ paper.pdf. National Response to South Africa’s Electricity Shortage. 2008. January. h p://www. info.gov.za/view/DownloadFileAction?id=77837. NERSA. 2008. “NERSA Consultation Paper, Renewable Energy Feed-in Tariff.â€? December. h p://www.NERSA.org.za/Admin/Document/Editor/ï¬?le/NERSA%20REFIT%20 %20consultation%20paper%2002%20Dec%202008.pdf. — — —. 2009a. “NERSA Decision on Renewable Energy Feed-In Tariffs (REFITS) Phase II, Media Statement.â€? November 2. — — —. 2009b. “NERSA Media Statement, NERSA Decision on Renewable Energy Feed-in Tariff (REFIT).â€? March 31. — — —. 2010. “NERSA Rules on Selection Criteria for Renewable Energy Projects under the REFIT Programme.â€? February 19. — — —. N.d. “NERSA, Electricity, Overview.â€? h p://www.NERSA.org.za/#. — — —. N.d. “NERSA, Renewable Energy Feed-in Tariffs Guidelines.â€? h p://www. google.com/url?sa=t&source=web&cd=1&ved=0CBgQFjAA&url=h p%3A%2F% 2Fwww.ameu.co.za%2Flibrary%2Findustry-documents%2FNERSA%2FREFIT% 2520Reasons%2520for%2520Decision%25202%2520310309.pdf&rct=j&q=reï¬?t%20 too%20low%20south%20africa&ei=6Rj0TNiDGYGClAeRoPmjDQ&usg=AFQjC NH9_yI6gM60poc34jtSn_1w2 6DA&sig2=_sJX6HDIvWe79lAu-_bwIA&cad=rja. Panchabuta. 2010a. “Gujarat Urja Vikas Nigam Ltd (GUVNL) signs PPA for 537MW with 54 solar power generation companies taking total signed to 933.5MW.â€? h p:// panchabuta.wordpress.com/2010/12/28/gujarat-urja-vikas-nigam-ltd-guvnl-signs- ppa-for-537mw-with-54-solar-power-generation-companies-taking-total-signed- at-933-5mw/. — — —. 2010b. “Solar PV and Thermal Developers shortlisted for the ï¬?rst phase of JNNSM (National Solar Mission).â€? h p://panchabuta.wordpress.com/2010/11/17/solar- Concentrating Solar Power in Developing Countries 143 pv-and-thermal-developers-shortlisted-for-the-first-phase-of-jnnsm-national- solar-mission/. Peters, Dipuo. 2010. Media statement by the Minister of Energy, South Africa, Ms. Dipuo Peters at the Official Announcement of the Solar Park Investors Conference. h p:// www.energy.gov.za/ï¬?les/media/pr/media_release2010.pdf. Project Finance NewsWire 2010. Renewables Spread to the Near East. By Sohail Barkatali and Yasser Yaqub. Project Finance NewsWire, September 2010. RAD (Renewable Energy Ads). 2010. “Solar Mission May Shortlist Project Developers in a Week.â€? November 4. h p://ap.renewableenergyads.com/blog.php?month=201011. Saleh, Heba. 2010. “Egypt to Reign in Energy Subsidies . . . in Due Course.â€? Financial Times. June 15. h p://blogs.ft.com/beyond-brics/2010/06/15/egypt-to-reign-in-energy- subsidies-in-due-course/. Sarangi, Gopal K., and Arabinda Mishra. 2009. “Environmental Innovation in Electricity Sector in India: Role of Electricity Regulators.â€? TERI. h p://www.dime-eu.org/ ï¬?les/active/0/Sarangi_Dime-Workshop-Paper%20_Draft-Final_.pdf. Sewall, Adam. 2010. “New Law in Jordan to Boost Solar Energy, Other Renewables.â€? GetSolar. January 14 h p://www.getsolar.com/blog/new-law-in-jordan-to-boost- solar-energy-other-renewables/3152/. South Africa Department of Energy. N.d. “Independent Power Producers.â€? h p://www. energy.gov.za/ï¬?les/electricity_frame.html. TNERC (Tamil Nadu Electricity Regulatory Commission). 2010. “Consultative Paper on Comprehensive Tariff Order for Solar Photovoltaic and Solar Thermal Power Plants up to 3 MW having grid connectivity below 33 kV level.â€? h p://tnerc.gov. in/Concept%20Paper/2010/Solar%20CP%20on.pdf. Van de Merwe, C. 2010. “Wind Developers Show Strong Appetite for SA Projects.â€? October 8. h p://www.engineeringnews.co.za/article/second-wind-seminar-held- in-johannesburg-2010-10-08-1. Winkler, Harald. N.d. “Long Term Mitigation Scenarios.â€? Project Report. h p://www. erc.uct.ac.za/Research/LTMS/LTMS_project_report.pdf. World Bank. 2007. “Project Appraisal Document on a Proposed Grant from the Global Environment Facility Trust Fund in the Amount of US$43.2 Million to the Office National de L’Elecricite (National Electricity Utility) of the Kingdom of Morocco for an Integrated Solar Combined Cycle Power Project.â€? Washington, D.C.: World Bank, February 20. h p://web.worldbank.org/external/projects/main?Projectid=P 041396&theSitePK=40941&piPK=64290415&pagePK=64283627&menuPK=642821 34&Type=Overview. — — —. 2008. “Update to the Development Commi ee on Key Issues and World Bank Group Activities.â€? DC2008-0010. Washington, DC: World Bank, October 10. h p://siteresources.worldbank.org/DEVCOMMINT/Documentation/21936688/ DC2008-0010(E)PresNote.pdf. — — —. 2009. “Project Appraisal Report for Tunisia Energy Efficiency Project.â€? Report No: 48704-TN. Washington, D.C.: World Bank, June 3. h p://www-wds.worldbank. org/external/default/WDSContentServer/WDSP/IB/2009/06/15/000333037_200906 15001220/Rendered/PDF/487040PAD0P104101Official0Use0Only1.pdf. — — —. 2010. “Project Appraisal Document on a Proposed IBRD Loan in the Amount of US$200 Million and a Clean Technology Fund Loan in the Amount of US$100 144 A World Bank Study Million to the Moroccan Agency for Solar Energy for the Ouarzazate Concen- trated Solar Power Plant.â€? Draft. Washington, DC: World Bank, December 13. World Bank/ESMAP. 2009. “Pricing and Contracting Mechanisms for Renewable Energy Development in the Philippines.â€? Component I Report. Prepared by Mercados– Energy Markets International. Washington, DC: World Bank, October. — — —. 2010. “Study on Barriers for Solar Power Development in India.â€? Washington, DC: World Bank, July. — — —. 2011a. “Electricity Auctions: An Overview of Efficient Practices.â€? Luiz Maurer, Luiz Barrozo, and others. Conference Edition. Washington, DC: World Bank, February. — — —. 2011b. “Middle East and North Africa Region Assessment of the Local Manufac- turing Potential for Concentrated Solar Power (CSP) Projects.â€? Washington, DC: World Bank, January. h p://arabworld.worldbank.org/content/dam/awi/pdf/ CSP_MENA__report_17_Jan2011.pdf. World Bank/GEF. 2006. Assessment of the World Bank/GEF Strategy for the Market Development of Concentrating Solar Thermal Power. Washington, DC: World Bank-GEF. Chapter 5 Bibliography The chapter was partly based on the following reports: YES/Nixus/CENER. 2010. “Review of CSP Technologies.â€? Final report prepared for the World Bank under Project Number P119536, p. 18. Bibliography Bhatia, B., and M. Gulati. 2004. “Reforming the Power Sector, Controlling Electricity Theft and Improving Revenue.â€? Note no. 272. Private Sector Development Vice Presidency, World Bank Group, September. Bloomberg New Energy Finance. 2010. “Crossing the Valley of Death: Solutions to the Next Generation Clean Energy Project Financing Gap.â€? — — —. 2010. “Week in Review 13–19 July 2010.â€? Bloomberg New Energy Finance VI(43). Central Electric Authority, Ministry of Power, India. 2009. CERC. 2009. “Determination of Generic Levelized Generation Tariff under Regulation 8 of the Central Electricity Regulatory Commission (Terms and Conditions for Tar- iff Determination from Renewable Energy Sources) Regulations, 2009.â€? Petition No. 284/2009 (Suo Motu). CSP Today. 2010. “Global Concentrated Solar Power Industry Report 2010–2011.â€? Lon- don: FC Business Intelligence. Dersch and Richter. 2007. “Water Saving Heat Rejection for Solar Thermal Power Plants.â€? Institute of Technical Thermodynamics. Emerging Energy Research. 2010. “Global Concentrated Solar Power Markets and Strat- egies: 2010–2025.â€? April 2010. Ernst & Young and Fraunhofer Institute. 2010. “MENA Assessment of the Local Manu- facturing Potential for Concentrated Solar Thermal Power (CSP) Projects.â€? Draft report prepared for the World Bank. Fichtner. 2010. “Technology Assessment of Concentrated Solar Power Technologies for a Site Speciï¬?c Project in South Africa.â€? Final report prepared for the World Bank under Project Number P118730. Concentrating Solar Power in Developing Countries 145 Gereffi, G., K. Dubay. 2008. “Concentrating Solar Power—Clean Energy for the Electric Grid.â€? Center on Globalization, Governance and Competitiveness. IEA. 2009. “Concentrated Solar Power Technology Roadmap.â€? IEA, April. Kearney, A. T. 2010. “Solar Thermal Electricity 2025. Clean Electricity on Demand: A rac- tive STE Cost Stabilize Energy Production.â€? Report commissioned by ESTELA. Kistner, R., and H. Price. 1999. “Financing Solar Thermal Power Plants.â€? NREL/CP-550- 25901, p. 5. Kolb, G. J., S. A. Jones, M. W. Donnelly, D. Gorman, R. Thomas, R. Davenport, and R. Lumia. 2007. Heliostat Cost Reduction Study. Sandia report SAND2007-3293. Kutscher, Chuck. 2008. “Cooling Options for Geothermal and Concentrating Solar Ther- mal Power Plants.â€? NREL presentation at EPRI Workshop on Advanced Cooling Technologies, June 9, 2008. Kutscher, C., and A. Buys. 2006. “Analysis of Wet/Dry Hybrid Cooling for a Parabolic Trough Power Plant.â€? NREL. Madrigal, M., and B. Arizu Jablonski. 2010. “Transmission Expansion for Renewable Energy: Challenges and Approaches and the Case of the Philippines.â€? Washing- ton, DC: World Bank. McDonnell Douglas Astronautics Company. 1981. “Final Report—Second Generation Heliostat with High Volume Manufacturing Facility Deï¬?ned by General Motors.â€? SAND81-8177. Sandia National Laboratories, Albuquerque, NM, April. Novi Energy. 2010. “Procurement Practices Analysis: Part A Tendering Models and Prac- tices.â€? Prepared for the World Bank Group under Project Number P118730, CSP, Regulatory and Financial Incentives. Pi -Paal, R., J. Dersch, and B. Milow. 2005. “Ecostar RoadMap.â€? SES6-CT-2003-502578. European Concentrated Solar Thermal Road-Mapping. Price, H., and D. Kearney. 2003. “Reducing the Cost of Energy from Parabolic Trough Solar Power Plants.â€? NREL/CP-550-33208, pp. 5 and 7. — — —. 2003. “Reducing the Cost of Energy from Parabolic Trough Solar Power Plants.â€? NREL/CP-550-33208, p. 6. Shinnar, R., and F. Citro. 2007. “Solar Thermal Energy: The Forgo en Energy Source.â€? Technology in Society 29:261–70. Sioshansi, R., and P. Denholm. 2010. “The Value of Concentrating Solar Power and Ther- mal Energy Storage.â€? NREL-TP-6A2-45833. Solgate Final Technical Report. 2004. “Solar Hybrid Gas Turbine Electric Power System.â€? Contract no. ENK5-CT-2000-00333. Stoddard and others. 2006. “Economic, Energy, and Environmental Beneï¬?ts of Concen- trating Solar Power in California.â€? NREL/SR-550-39291, p. 10. U.S. DOE. 2007. “Concentrating Solar Power Commercial Application Study: Reducing Water Consumption of Concentrating Solar Power Electricity Generation.â€? Report to Congress. Vazquez, J., N. Castañeda, D. Castañeda. 2009. “First Commercial Application of Sener- tough Collector: High Performance at Reduced Cost.â€? In SolarPACES 2009, 15th International Symposium on Concentrating Solar Power and Chemical Energy Technologies, Berlin. Western Governors Association. 2009. “Western Renewable Energy Zones—Phase 1 Report.â€? h p://www.westgov.org/index.php?option=com_content&view=article &id=219&Itemid=81. 146 A World Bank Study World Bank. 2010. “CSP Scale Up in MENA: Likely Resulting Water Needs and Options for Meeting Them.â€? Washington, DC: World Bank. World Coal Institute. 2009. “The Coal Resource: A Comprehensive Overview of Coal.â€? Chapter 6 Bibliography The chapter was based on the following reports: Ernst & Young and Fraunhofer. 2010. “MENA Assessment of Local Manufacturing Potential for Concentrated Solar Power (CSP) Projects.â€? Report prepared for the World Bank. Fichtner. 2011. “Assessment of Local Manufacturing Capacities for Solar Power Technol- ogies in South Africa.â€? Fichtner Report 6543P09/FICHT-7130099 for World Bank Project P118730. World Bank, Washington, DC. Bibliography Abengoa. 2010. Homepage; pictures that show a solar tower power plant. Date of last site visit September 13. h p://www.abengoasolar.com/corp/web/en/our_projects/ solucar/ps10/index.html. AGC. 2010. “AGC Glass Europe.â€? Company technology information brochure. Date of last site visit September 13. h p://www.agcglass.eu/English/Homepage/Products/ Float-glass-technology/page.aspx/958. Almeco. 2010. “Almeco/Xeliox: Vegaflex the Mirror 100% Aluminum.â€? Technological information brochure. Date of last site visit September 13. h p://www.almecogroup. com/en/products/solar/vegaflex. Altenburg, T. 2009. “Industrial Policy for Low- and Lower-Middle-Income Countries.â€? Preliminary draft. German Development Institute (DIE). Archimede. 2008. Archimede company publication. Date of last site visit September 13, 2010. h p://www.archimedesolarenergy.it/receiver_tube.htm. — — —. 2010. Receiver information. Date of last site visit September 13. h p://www. archimedesolarenergy.com/receiver_tube.htm. Bacon, Robert, and Masami Kojima. 2011. “Issues in Estimating the Employment Gener- ated by Energy Sector Activities.â€? Washington, DC: World Bank. Bha acharya, R., and H. Wolde. 2010. “Constraints on Trade in the MENA Region.â€? IMF Working Paper No. 10/13. Washington, DC: IMF. Bha acharya, S. C., and Chinmoy Jana. 2009. “Renewable Energy in India: Historical Developments and Prospects.â€? Energy 34(2009):981–91. Bilï¬?ngerBerger. 2010. Technological internet brochure, picture from the homepage. Date of last site visit September 13. h p://www.bilï¬?ngerberger.de/C125710E004ABFC5/ CurrentBaseLink/W284PKJ6429WEBTE. BMU. 2010. German Ministry of Environment (BMU). 4781.php#abb1. Date of last visit June 20. h p://www.bmu.de/forschung/doc/. Bosch. 2008. “Auf den Punkt fokussiert. Automatisierungstechnik konzentrier- ende Solarkraftwerke.â€? Date of last site visit September 13, 2010. h p://www. boschrexroth.com/corporate/sub_websites/industries/solar/de/solaranlagen/ referenzen/index.jsp;jsessionid=cbaBKAaUpuhVT9SyQvoSs. Brakmann, George Fathy Ameen Mohammad, Miroslav Dolejsi and Mathias Wiemann. 2006. “Construction of the ISCC Kuraymat (Integrated Solar Combined Cycle Power Plant in Morocco)â€?. Paper FA4-S6 presented at 13th International Symposium Concentrating Solar Power in Developing Countries 147 on Concentrating Solar Power and Chemical Energy Technologies, SolarPaces, June 20–23, 2006, Seville, Spain. Brost, R., A. Gray, F. Burkholder, T. Wendelin, and D. White. 2009. “Skytrough Optical Evaluations Using Vshot Measurement.â€? Proceedings of 15th International Solar- PACES Symposium, September 14–18, 2009, Berlin, Germany. Buck, R. 2008. “Solare Turmtechnologie—Stand und Potenzial.â€? Paper presented at Köl- ner Sonnenkolloquium 2008. Burkholder, F., and C. Kutscher. 2008. “Heat Loss Testing of Scho ’s 2008 PTR70 Para- bolic Trough Receiver.â€? (h p://www.nrel.gov/csp/publications.html). Date of last site visit September 13, 2010. NREL. Casteneda, N., J. Vazquez, M. Domingo, A. Fernandez, and J. Leon. 2006. “Sener Para- bolic Trough Design and Testing.â€? Proceedings of 13th International SolarPACES Symposium, June 20–23, 2006, Seville, Spain. Chen, Y., and T. Pu itanun. 2005. “Intellectual Property Rights and Innovation in Devel- oping Countries.â€? Journal of Development Economics 78(2):474–93. Cohen, G., D. Kearney, and G. Kolb. 1999. “Final Report on the Operation and Maintenance Improvement Program for Concentrating Solar Power Plants.â€? SAND99-1290. Sandia National Laboratories, June. Contractor, F. J., and P. Lorange. 2002. “Why Should Firms Cooperate? The Strategy and Economics. Basis for Cooperative Ventures.â€? In F. J. Contractor, and P. Lorange (eds.). 2002. Cooperative Strategies in International Business. Oxford: Elsevier Sci- ences, pp. 1–26. CSP Today. 2010. “The Concentrated Solar Power Markets Report/Summary.â€? Date of last site visit September 13. h p://www.csptoday.com/cspmarkets/index. shtml?utm_source=CSP%2Bmain%2Bwebsite&utm_medium=Banner&utm_cam paign=CSP%2BMarkets%2Breport%2Bmain%2Bbanner. Dersch, J., G. Morin, M. Eck, A. Häberle. 2009. “Comparison of Linear Fresnel and Para- bolic Trough Collector Systems—System Analysis to Determine Break Even Costs of Linear Fresnel Collectors.â€? Proceedings of 15th International SolarPACES Sym- posium, September 14–18, Berlin, Germany. Dewey & LeBoeuf LLP. 2010. “China’s Promotion of the Renewable Electric Power Equipment Industry. Hydro,Wind, Solar, Biomass.â€? DLR. 2010. Deutsches Zentrum für Luft- und Raumfahrt/German Aerospace Center. Date of last visit March 30. www.dlr.de/rd/en/Portaldata/1/Resources/portal_news/ newsarchiv2008_5/. Doening. 2010. “Concentrating Solar Power Industry Projects, Program Area Thermal Storage.â€? U.S. Department of Energy. Ecostar. 2005. “European Concentrated Solar Thermal Road-Mapping.â€? ECOSTAR. Eichhammer, W., and R. Walz. 2009. “Indicators to Measure the Contribution of Energy Efficiency and Renewables to the Lisbon Targets—Monitoring of Energy Efficiency in EU 27, Norway and Croatia.â€? ODYSSEE-MURE. Fraunhofer ISI. Emerging Energy Research. 2010. “Global Concentrated Solar Power Markets and Strat- egies: 2010–2025.â€? Report April. Erasolar. 2010. Erasolar, revista tecnica de energía solar. Date of last visit June 16. h p:// www.erasolar.es/WEB-146/RESUMEN146.htm. ERC (Energy Research Center). 2010. “Energy Use.â€? Chapter II of report, ERC, June. — — —. 2010. “Past and Ongoing EE Activities and Programs by Donors and GOE.â€? 148 A World Bank Study Estela. 2010. “Solar Thermal Electricity 2025. Clean Electricity on Demand: A ractive STE Cost Stabilize Energy Production.â€? Date of last site visit September 13. h p:// www.estelasolar.eu/index.php?id=22. Evers, H. 2010. “Cost Based Engineering and Production of Steel Constructions.â€? Flabeg. 2010. Technological explanations on company web portal. Date of last site visit September 13. h p://www.flabeg.com/de/solar_produkte_parabolic_de.php Gil, A., M. Medrano, I. Martorell, A. Lázaro, P. Dolado, B. Zalba, and K. Cabeza. 2010. “State of the Art on High Temperature Thermal Energy Storage for Power Genera- tion. Part 1—Concepts, Materials and Modellization.â€? Renewable and Sustainable Energy Reviews 14(2010):31–55. Published by Elsevier. Glaeser, Hans Joachim. 2001. Large Area Glass Coating. ISBN: 3-00-004953-3. Glasstech. 2010. Internet research. The different glass bending processes and machines are described on this website. Date of last site visit September 13. h p://www. glasstech.com/312Solar.aspx. Glaston. 2010. “Automatic Bending Furnace, Product Brochure.â€? Date of last site visit September 13. h p://www.glaston.net/includes/ï¬?le_download.asp?deptid=5272 &ï¬?leid=3499&ï¬?le=ESU_EcoPower_EN.pdf&pdf=1. Graham, E. M. 1982. “The Terms of Transfer of Technology to the Developing Nations: A Survey of the Major Issues.â€? OECD, North/South Technology Transfer, OECD, Paris. Hermann, U., F. Graeter, P. Nava. 2004. “Performance of the SKAL-ET Collector Loop at KJC Operating Company.â€? 12th SolarPACES Symposium, October 6–8, 2004, Oaxaca, Mexico. Hildebrandt, C. 2009. “Hochtemperaturstabile Absorberschichten für linear konzen- trierende solarthermische Kraftwerke.â€? PhD thesis at Universität Stu gart and Fraunhofer ISE. Hydro. 2010. “Hydro Aluminium White Paper CSP.â€? h p://www.hydro.com/en/ Subsites/North-America/39409. Indian Wind Energy Outlook. 2009. International SolarPACES Symposium, September 14–18, 2009, Berlin, Germany. International Workshop on Connections in Steel Structures. 2000. Roanoke. Date of last site visit September 13. h p://www.aisc.org/content.aspx?id=3626. ISE. 2010. Fraunhofer Institut für solare Energiesysteme ISE. Date of last site visit September 13. h p://www.ise.fraunhofer.de/geschaeftsfelder-undmarktbereiche/ solarthermie. Kaefer Insulation. 2010. Expert interview by Fraunhofer ISE. Kearney, A. T. 2010. Solar Thermal Electricity 2025, Clean Electricity on Demand: A rac- tive STE Cost Stabilize Energy Production. Kearney, David. 2007. “Parabolic Trough Collector Overview.â€? Presentation at the Para- bolic Trough Workshop 2007at NREL, Golden, CO, March 2007. Kennedy, C. E. 2002. “Review of Mid to High Temperature Solar Selective Absorber Materi- als.â€? Golden, Colorado: NREL. h p://www.nrel.gov/csp/troughnet/pdfs/31267.pdf Kennedy, C. E. 2005. “CSP FY 2005 Milestone Report.â€? Date of last site visit September 13. Golden, Colorado: NREL. h p://www.nrel.gov/csp/publications.html. Khalil, A., A. Mubarak, and S. Kaseb. 2010. “Road Map for Renewable Energy Research and Development in Egypt.â€? Journal of Advanced Research University of Cairo 1(2010):29–89. Kistner, R., T. Keitel, B. Felten, and T. Rzepczyk. 2009. “Analysis of the Potential for Cost Decrease and Competitiveness of Parabolic Trough Plants.â€? Concentrating Solar Power in Developing Countries 149 Laing, D., W. Schiel, and P. Heller. 2002. “Dish-Stirling-Systeme—Eine Technologie zur dezentralen Stromerzeugung.â€? FVS Themen 2002. LeBlanc, Rick, and SkyFuel. 2010. “Advanced Parabolic Trough Concentrators: Com- mercial Deployments and Future Opportunities.â€? Presentation at Intersolar North America, July 9, 2010. Lewis, J. 2007. “Technology Acquisition and Innovation in the Developing World: Wind Turbine Development in China and India.â€? St Comp Int Dev 42(2007):208–32. Lewis, J., and R. Wiser. 2007. “Fostering a Renewable Energy Technology Industry: An International Comparison of Wind Industry Policy Support Mechanisms.â€? Energy Policy 35(2007):1844–57. Lund, P. 2008. “Effects of Energy Policies on Industry Expansion in Renewable Energy.â€? Renewable Energy 34:53–64. Mazzoleni, R., and R. Nelson. 2007. “Public Research Institutions and Economic Catch- Up.â€? Research Policy 36(2007):1512–28. Medrano, M., A. Gil, I. Martorell, X. Potau, and L. Cabeza. 2010. “State of the Art on High-Temperature Thermal Energy Storage for Power Generation. Part 2—Case Studies.â€? Renewable and Sustainable Energy Reviews 14(2010):56–72, published by Elsevier. Mertins, Max. 2009. “Dissertation technische und wirtschaftliche Analyse von hori- zontalen Fresnel-Kollektoren.â€? Fakultät für Maschinenbau der Universität Karl- sruhe (TH). Morin, G. 2010. 2010. “Design Optimization of Solar Thermal Power Plants.â€? Prelimi- nary version of Ph.D. thesis at University Braunschweig and Fraunhofer ISE, as of June 21, 2010. NEEDS. 2009. “NEEDS New Energy Externalities Developments for Sustainability.â€? Cost development—an analysis based on experience curves. Novatec. 2010. Online technology information brochure. Date of last site visit September 13. h p://www.novatec-biosol.com/index.php?article_id=11&clang=4. NREL (National Renewable Energy Laboratory). 2010. Date of last visit February 19, 2010. www.nrel.gov/csp/troughnet/solar_ï¬?eld.html. Parker. 2008. “Parker actuators optimise productivity of world’s third largest solarpower plant.â€? Date of last site visit September 13, 2010. h p://www.parker.com/portal/ site/PARKER/menuitem.6a1e641def5c26f9f8500f199420d1ca/?vgnextoid=3f086f1e 77aee010VgnVCM10000032a71dacRCRD&vgnextfmt=EN#. Pilkington. 2003. “Pilkington and the Flat Glass Industry, Overview of the Float Glass Production.â€? Date of last site visit September 13, 2010. h p://www.pilkington.com/ pilkington-information/about+pilkington/education/float+process/default.htm. Relloso, S., and E. Delgado. 2009. “Experience with Molten Salt Thermal Storage in a Commercial Parabolic Trough Plant: Andasol-1 Commissioning and Operation.â€? Proceedings of 15th International SolarPACES Symposium, September 14–18. Riffelmann, K. J., J. Kö er, P. Nava, F. Meuser, G. Weinrebe, W. Schiel, G. Kuhlmann, A. Wohlfahrt, A. Nady, and R. Dracker. 2009. “Heliotrough—a New Collector Generation for Parabolic Trough Power Plants.â€? Proceedings of 15th International SolarPACES Symposium, September 14–18. Scho . 2009. “SCHOTT PTR®70 Receiver: The Next Generation.â€? Company bro- chure. Date of last site visit September 13, 2010. h p://www.scho solar.com/de/ produkte/solarstromkraftwerke/scho -ptr-70-receiver/. 150 A World Bank Study Sener. 2007. “Senertrough. The Collector for Extresol-1. 600 meters loop test in Andasol-1 and Test Unit Description.â€? Available at sciencetoday.com. Date of last site visit September 13, 2010. Shioshansi/NREL. 2010. “The Value of Concentrating Solar Power and Thermal Ener- gy Storage.â€? Date of last site visit September 13, 2010. h p://www.nrel.gov/csp/ publications.html. Siemens. 2010. “Solar Receiver UVAC 2010.â€? Product brochure. Date of last visit September 14, 2010. h p://www.energy.siemens.com/us/en/power-generation/ renewables/solar-power/concentrated-solarpower/receiver.htm. Solar Millennium AG. 2008. “Die Parabolrinnen-Kraftwerke Andasol 1 bis 3. Die größten Solarkraftwerke der Welt; Premiere der Technologie in Europa.â€? Information brochure. Erlangen, Germany. Date of last site visit September 13, 2010. www. solarmillennium.de/upload/Download/Technologie/Andasol1-3deutsch.pdf. Solarel. 2010. E-Mail communication with Solarel (manufacturer of EuroTroughparts), and online technology information. Date of last site visit September 13, 2010. h p://www.solarelenergy.com/csp/information-experience. SolarPACES. 2010. Date of last visit September 14, 2010. h p://www.solarpaces.org/ News/Projects/projects.htm. Soutar, Andrew, Bart Fokkink, Zeng Xianting, Tan Su Nee, and Linda Wu. 2001. “Sol-gel Anti-reflective Coatings.â€? SIMTech Technical Report (PT/01/002/ST). SQM. 2010. Online technology information. Date of last site visit September 13, 2010. h p://www.sqm.com/aspx/Chemicals/Specialmoltemsalts.aspx?VarF=1 Sun & Wind Energy. 2010. “CSP Market Overview.â€? Sun & Wind Energy 06/2010. Taggart, Stewart. 2008. “CSP: Dish Projects Inch Forward.â€? Renewable Energy Focus July/ August 2008. Part IV: In the fourth article in a series of articles looking at the dif- ferent aspects of concentrating solar power CSP technology, we turn our a ention to solar dishes. Trieb, Franz, H. Müller-Steinhagen, and J. Kern. 2010. “Financing Concentrating Solar Power in the Middle East and North Africa—Subsidy or Investment?â€? Article sub- mi ed to the Journal Energy Policy by the publisher Elsevier, June 6, 2010. Trieb, Franz, Marlene O’Sullivan, Thomas Pregger, Christoph Schillings, and Wolfram Krewi . 2009. “Characterisation of Solar (2009), Electricity Import Corridors from MENA to Europe—Potential, Infrastructure and Cost.â€? Report prepared in the frame of the EU project, Risk of Energy Availability: Common Corridors for Europe Supply Security (REACCESS), carried out under the 7th Framework Programme (FP7) of the European Commission. UNIDO (United Nations Industrial Development Organization). 2003a. “Methodologi- cal Guide: Restructuring, Upgrading and Industrial Competitiveness.â€? Vienna. — — —. 2003b. “Lao PDR: Medium-Term Strategy and Action Plan for Industrial Develop- ment.â€? Final Report. Vienna. Vote Solar Initiative. 2009a. “Solar Power Plants.â€? March. Date of last site visit Septem- ber 13, 2010. www.votesolar.org. — — —. 2009b. “The Sun Rises on Nevada: Economic and Environmental Impacts of Developing 2,000 MW of Large-Scale Solar Power Plants.â€? www.votesolar.org. Walz, R. N.d. “Integration of Sustainability Innovations within Catching-up Processes (ISI-CUP).â€? Country case studies for wind energy. Unpublished. Concentrating Solar Power in Developing Countries 151 Walz, R., K. Ostertag, W. Eichhammer, N. Glienke, A. Jappe-Heinze, W. Mannsbart, and J. Peuckert. 2008. “Research and Technology Competence for a Sustainable Devel- opment in the BRICS Countries.â€? Fraunhofer Institute Systems and Innovation Research. Fraunhofer IRB Verlag. World Bank. 2008. “Tunisia’s Global Integration: Second Generation of Reforms to Boost Growth and Employment.â€? World Bank Country Studies, Washington, D.C., World Bank. — — —. 2010. “Social and Economic Development Group of the World Bank, Middle East and North Africa Region: Republic of Tunisia, Development Policy Review— Towards Innovation Driven Growth.â€? Report No. 50847-TN, January. h p:// www.windpowerindia.com/. Zelesnik, Olaf. 2002. “Herstellung temperaturstabilertransparenter Oxidschichten.â€? Dissertation. Institut fürMetallphysik und Nukleare Festkörperphysik der Tech- nischen Universität Carolo Wilhelmina zu Braunschweig. Chapter 7 Bibliography The chapter was based on the following report: NOVI Energy. 2011. “Regulatory and Financial Incentives for Scaling up Concentrating Solar Power in Developing Countries—Procurement Practices Analysis.â€? Report prepared for World Bank Project P118730. Bibliography Abener. 2008. “Solar Concentration Workshop Presentation.â€? November. h p://site resources.worldbank.org/INTMENA/Resources/Abener_Solar_WorldBank.pdf. Astrad, Kerstin. 2006. “Examining Influences of EU Policy on Instrument Choice: The Selection of a Green Certiï¬?cate Trading Scheme in Sweden.â€? Bill Brown Climate Solutions. 2009. “Our New Energy Economy.â€? January. h p://bill brownclimatesolutions.blogspot.com/2009_01_25_archive.html. Boothby, Francesca. 2011. “US CSP Project Update: Tough Year Ahead.â€? CSP Today. February.h p://social.csptoday.com/industry-insight/us-csp-project-update-tough- year-ahead. Bureau of Labor Statistics. “Employment Cost Trends.â€? United States Department of Labor. h p://www.bls.gov/ncs/ect/. — — —. 2006. “Escalation Guide for Contracting Parties.â€? United States Department of Labor, Bureau of Labor Statistics. July 27. h p://www.bls.gov/ppi/ppiescalation. htm. — — —. 2010. “Consumer Price Index.â€? United States Department of Labor. h p://www. bls.gov/cpi/. — — —. 2010. “Producer Price Indexes.â€? United States Department of Labor. Busgen, Durrschmidt. 2008. “The Expansion of Electricity Generation from Renewable Energies in Germany. A Review Based on the Renewable Energy Sources Act Progress Report 2007 and the New Feed-in Legislations.â€? Berlin: Federal Ministry of the Environment. CERC Notiï¬?cation. 2010. No. L-1/12/2010-CERC. January 14. h p://cercind.gov.in/ Regulations/CERC_Regulation_on_Renewable_Energy_Certiï¬?cates_REC.pdf. CERC. 2009. Notiï¬?cation No. L-7/186(201)/2009-CERC. September 16. h p://cercind.gov. in/Regulations/Amend_Renewable_Energy_tariff.pdf. 152 A World Bank Study CSP Today. 2009. “IN-DEPTH: In Search of a Good FiT.â€? July. h p://social.csptoday. com/news/depth-search-good-ï¬?t. del Rio and Gual. 2006. “An Integrated Assessment of the Feed-in Tariff System in Spain.â€? Energy Policy 35:994–1012. Fichtner. 2010. “Technology Assessment of Concentrated Solar Power Technologies for a Site Speciï¬?c Project in South Africa.â€? Final report prepared for the World Bank under Project Number P118730. Fouquet, D., and T. B. Johansson. 2008. “European Renewable Energy Policy at Crossroads—Focus on Electricity Support Mechanisms.â€? Energy Policy 36:4079–92. Gan, L., and others. 2007. “Green Electricity Market Development: Lessons from Europe and the US.â€? Energy Policy 35:144–55. Grama, Sorin, Elizabeth Wayman, and Travis Bradford. 2008. Concentrating Solar Power—Technology, Cost, and Markets. Cambridge, MA: Prometheus Institute. www.greentechmedia.com. GreenPeace, SolarPACES, ESTELA. 2009. “Concentrating Solar Power Global Outlook 2009—Why Renewable Energy Is Hot.â€? h p://www.estelasolar.eu/ï¬?leadmin/ ESTELAdocs/documents/Greenpeace_Concentrating_Solar_Power_2009.pdf. Lorenzoni, A. 2003. “The Italian Green Certiï¬?cates Market between Uncertainty and Opportunities.â€? Energy Policy 31:33–42. Lund, Morten, William H. Holmes, Stephen Hall, Jennifer Martin. 2009. “Lex Helius: The Law of Solar Energy.â€? Portland, Oregon: Stoel Rives LLP. h p://www.stoel.com/ webï¬?les/lawofsolarenergy.pdf. Ministry of Statistics and Programme Implementation. “India Statistics. Index Num- ber of Wholesale Prices.â€? h p://mospi.nic.in/Mospi_New/upload/india_statistis/ spb_pg238_index_wp.pdf. NERSA. 2009. “Renewable Energy Feed-in Tariffs Guidelines.â€? h p://www.ameu.co.za/ library/industry-documents/nersa/REFIT%20Reasons%20for%20Decision%20 2%20310309.pdf. Nielsen, L., and T. Jeppesen. 2003. “Tradable Green Certiï¬?cates in Selected European Countries—Overview and Assessment.â€? Energy Policy 31:3–14. Nilsson and Sundqvist. 2006. “Using the Market at a Cost: How the Introduction of Green Certiï¬?cates in Sweden Led to Market Inefficiencies.â€? Utilities Policy 15:49–59. Nishikawa, Warren, Steve Horne, and Jane Melia. 2008. “LCOE for Concentrating Pho- tovoltaics (CPV), Solfocus Inc.â€? Presented at the International Conference on Solar Concentrators for the Generation of Electricity, November 16–19, Palm Desert, CA. NREL (National Renewable Energy Laboratory). “Concentrating Solar Power Projects with Operational Plants.â€? h p://www.nrel.gov/csp/solarpaces/operational.cfm. NVVN. 2010. “Draft Standard Power Purchase Agreement for JNNSM.â€? August 18. h p:// www.nvvn.co.in/Annexure%202%20-%20Power%20Purchase%20Agreement%20 (PPA)%20for%20NEW%20PROJECTS.pdf. Office National des Statistiques—Algeria. Indexes—Prices for Industrial Production. h p://www.ons.dz/-Prix-a-la-Production-Industrielle-.html. Orrick. 2010. “California PPA Curtailment Issues.â€? December. h p://www.orrick.com/ ï¬?leupload/3191.pdf. PG&E (Paciï¬?c Gas & Electric). 2009 “Wholesale Electric Power Procurement.â€? h p:// www.pge.com/b2b/energysupply/wholesaleelectricsuppliersolicitation/. Concentrating Solar Power in Developing Countries 153 — — —. 2009. “Renewables—A achment H: Form of Power Purchase Agreement.â€? July. http://www.pge.com/b2b/energysupply/wholesaleelectricsuppliersolicitation/ renewables2009/index.shtml. Philibert, Cedric. 2004. “International Energy Technology Collaboration and Climate Change Mitigation Case Study 1: Concentrating Solar Power Technologies.â€? h p://www.oecd.org/DATAOECD/25/9/34008620.PDF. Producer Prices. “Principal Global Indicators.â€? h p://www.principalglobalindicators. org/default.aspx. Quaschning, Volker. 2003. “Solar Thermal Power Plants—Technology Fundamentals.â€? Erneuerbare-Energien-und-Klimaschu .de. h p://www.volker-quaschning.de/ articles/fundamentals2/index.php. Radiant & Hydronics. 2006. “AZ Solar Station to Harnesses Sun’s Heat for Power.â€? http://www.radiantandhydronics.com/Archives/BNP_GUID_9-5-2006_A_ 10000000000000267639. Rowlands, Ian. 2004. “Envisaging Feed-in Tariffs for Solar Photovoltaic Electricity: European Lessons for Canada.â€? Department of Environment and Resource Stud- ies, Faculty of Environmental Studies, University of Waterloo, Waterloo. SCE (Southern California Edison). Renewable & Alternative Power—Request for Electric Energy Proposals. h p://www.sce.com/EnergyProcurement/renewables/2010- request-for-proposal.htm. SEPA (Solar Electric Power Association). 2009. “Utility Procurement Study: Solar Electricity in the Utility Market.â€? S.l.: SEPA. Solar Thermal Magazine. h p://www.solarthermalmagazine.com/2010/11/28/innovative- approach-to-concentrating-and-collecting-solar-energy-wins-industry-award/#. Statistics South Africa. Producer Price Index (PPI). h p://www.statssa.gov.za/key indicators/ppi.asp. U.S. Department of Energy. N.d. “Linear Concentrator Systems for Concentrating Solar Power.â€? h p://www.eere.energy.gov/basics/renewable_energy/linear_ concentrator.html. — — —. 2011. Database of State Incentives for Renewables & Efficiency (DSIRE). Washing- ton, D.C.: U.S. DOE. h p://www.dsireusa.org/.US. Wang, Andrew. 2011. “Development Director—SolarReserve.â€? Informational Interview. March. Wiser and others. 2005. “Evaluating Experience with Renewable Portfolio Standards in the United States.â€? Berkeley National Laboratory. World Bank. N.d. World Development Indicators (WDI) & Global Development Finance (GDF). World Bank Databank. h p://databank.worldbank.org/. — — —. 2011. A Review of Cost Elements and Cost Drivers for Concentrating Solar Power. Washington, D.C.: World Bank. YES/NIXUS/CENER. 2010. “LCOEs for CST Technologies.â€? Final report prepared for the World Bank under Project Number P119536. ECO-AUDIT Environmental Beneï¬?ts Statement The World Bank is commi ed to preserving In 2010, the printing of endangered forests and natural resources. this book on recycled paper The Office of the Publisher has chosen to saved the following: print World Bank Studies and Working • 11 trees* Papers on recycled paper with 30 percent • 3 million Btu of total postconsumer ï¬?ber in accordance with the energy recommended standards for paper usage • 1,045 lb. of net greenhouse set by the Green Press Initiative, a non- gases proï¬?t program supporting publishers in • 5,035 gal. of waste water using ï¬?ber that is not sourced from endan- • 306 lb. of solid waste gered forests. For more information, visit www.greenpressinitiative.org. * 40 feet in height and 6–8 inches in diameter C oncentrating Solar Power in Developing Countries: Regulatory and Financial Incentives for Scaling Up is part of the World Bank Studies series. These papers are published to commu- nicate the results of the Bank’s ongoing research and to stimulate public discussion. At present, different concentrating solar thermal (CST) technologies have reached varying degrees of commercial availability. This emerging nature of CST technology means that there are market and technical impediments to accelerating its acceptance, including cost competitiveness, intermittency, and an understanding of technology capability, limitations, and the beneï¬?ts of electricity storage. Many developed and some developing countries are currently working to address these barriers to scale up CST-based power generation. Given the considerable growth of CST technology development in several World Bank Group partner countries, there is a need to assess the recent experience of developed countries in designing and implementing regulatory frameworks to draw lessons that may facilitate the deployment of CST technologies in developing countries. Merely replicating developed countries’ schemes in the context of a developing country may not generate the desired outcomes. Against this background, this report (a) analyzes and draws lessons from the efforts of some developed countries and adapts them to the characteristics of developing economies; (b) assesses the cost reduction potential and economic and ï¬?nancial affordability of various CST technologies in emerging markets; (c) evaluates the potential for cost reduction and associated economic beneï¬?ts derived from local manufacturing; and (d) suggests ways to tailor bidding models and practices, bid selection criteria, and structures for power purchase agreements for CST projects in developing market conditions. World Bank Studies are available individually or on standing order. This World Bank Studies series is also available online through the World Bank e-library (www.worldbank.org/elibrary). ISBN 978-0-8213-9607-0 Energy and Mining 90000 Sector Board 9 780821 396070 SKU 19607