61096 TRANSITION TO A LOW-EMISSIONS ECONOMY IN POLAND The World Bank Poverty Reduction and Economic Management Unit Europe and Central Asia Region February 2011 ©2011 THE INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT / THE WORLD BANK 1818 H STREET NW WASHINGTON DC 20433 TELEPHONE: 202-473-1000 INTERNET: WWW.WORLDBANK.ORG Disclaimer This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this paper do not necessarily reflect the views of the Executive Directors of The World Bank 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. Rights and permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally grant permission to reproduce portions of the work promptly. All queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. Find this report and related materials at: www.worldbank.org/pl/lowemissionseconomy TRANSITION TO A LOW-EMISSIONS ECONOMY IN POLAND THE WORLD BANK POVERTY REDUCTION AND ECONOMIC MANAGEMENT UNIT EUROPE AND CENTRAL ASIA REGION FEBRUARY 2011 VICE PRESIDENT: PHILIPPE LE HOUEROU SECTOR DIRECTOR: LUCA BARBONE (FORMER) YVONNE TSIKATA SECTOR MANAGER: BERNARD FUNCK (FORMER) SATU KAHKONEN TASK TEAM LEADER: ERIKA JORGENSEN LESZEK KASEK ACRONYMS AND ABBREVIATIONS BAU Business-as-usual JI Joint Implementation CCS Carbon capture and storage KASHUE-KOBiZE National Administration of the Emissions CDM Clean Development Mechanism Trading Scheme-National Center for Emission CES Constant elasticity of substitution Balancing and Management CGE Computable General Equilibrium KLEMS EU database on capital (K), labor (L), CO2 Carbon dioxide energy (E), materials (M) and service CO2e Carbon dioxide equivalent inputs (S) productivity DSGE Dynamic Stochastic General Equilibrium LULUCF Land use, land-use change and forestry EAs Emission Allowances MAC Marginal abatement cost EC European Commission MacroAC Macroeconomic abatement cost EE Energy efficiency MacroMAC Macroeconomic marginal abatement cost EERP European Economic Recovery Package MicroMAC Microeconomic marginal abatement cost EIA US Energy Information Administration MEMO model Macroeconomic Mitigation Options model EITE sectors Energy-intensive and trade-exposed sectors MIND module Microeconomic Investment Decisions module EOR Enhanced Oil Recovery MtCO2e Millions of metric tons of CO2 equivalent EU European Union MWh Megawatt hours EU10 Bulgaria, Czech Republic, Estonia, Hungary, NPV Net present value Latvia, Lithuania, Poland, Romania, Slovakia, Non-ETS Sectors not covered by the ETS system and Slovenia OECD Organization for Economic Co-operation EUROSTAT the statistical office of the European Union and Development ETS Emissions Trading Scheme of the European PIT Personal income tax Union (also, the sectors included PL Poland (abbreviation used in figures) in the trading scheme) PPS Purchasing power standard GDP Gross domestic product ppm Parts per million GHGs Greenhouse gases R&D Research and development GTAP Global Trade Analysis Project database ROCA model Regional Options of Carbon Abatement GUS Główny Urząd Statystyczny, or Poland’s model Central Statistical Office Solar PV Solar photovoltaic power GW Gigawatt (1000 MW) TREMOVE EU traffic and emissions motor vehicle GWh Gigawatt hour model (approximate) HEV Hicksian equivalent variation tCO2e Metric tons of CO2 equivalent HVAC Heating, ventilation, and air conditioning toe Tons of oil equivalent IBS Instytut Badan Strukturalnych, or Institute UNFCCC United Nations Framework Convention for Structural Research, Warsaw, Poland on Climate Change IGCC Integrated gasification combined cycle VA Value-added IPCC Intergovernmental Panel on Climate Change VAT Value-added tax CONTENTS Acknowledgements 6 Introduction 23 a. Poland’s greenhouse gas emissions 29 b. Carbon abatement targets and policy challenges for Poland 35 c. A suite of models to assess emissions abatement 45 d. Business-as-usual scenarios for Poland 35 e. The Microeconomic Marginal Abatement Cost (MicroMAC) curve and Poland’s abatement options 51 f. The Macroeconomic Mitigation Options (MEMO) model and the macroeconomic impact of the abatement package 61 g. The Regional Options for Carbon Abatement (ROCA) model and implementing EU climate policy 67 h. Energy sector options and their macroeconomic impact 81 i. Energy efficiency options and their macroeconomic impact: a first look 99 j. Transport: an alternative engineering approach to mitigation options 109 Conclusions and additional issues 125 References 191 ACKNOWLEDGEMENTS This report was prepared by a core team led by Erika Jorgensen and including Leszek Kasek, Ryszard Malarski, Ewa Korczyc, John Allen Rogers, and Gary Stuggins. The report draws heavily from several background notes and papers including those prepared by Maciej Bukowski and IBS Warsaw (macroeconomic modeling), Christoph Böhringer and Loch Alpine Economics (macroeconomic modeling), McKinsey & Company Poland (engineering modeling), and ECORYS Rotterdam (transport modeling). This report was undertaken under the guidance of Luca Barbone (Sector Director), Bernard Funck (former Sector Manager), and Roumeen Islam (acting Sector Manager). Peer review was provided by Dominique Van Der Mensbrugghe and Kirk Hamilton. The team has received valuable comments, suggestions, and contributions from numerous other colleagues including the European Commission, Marcel Ionescu-Herioiu, Christine Kessides, Kseniya Lvovsky, Vikram Cuttaree, Thomas Laursen, Roumeen Islam, Jane Ebinger, Govinda Timilsina, David Tarr, Marianne Fay, and Rosalinda Quintanilla. Special thanks go to Orsalia Kalantzopoulos, former Country Director for Central/South Europe and the Baltics, for initiating this work. This report also benefits from a series of consultations and workshops with counterparts in Poland on the methodological approach of the report, the long-term development scenarios that underlie the analysis, the specific scenarios selected for the energy sector, and preliminary results of the analysis. Senior officials from the Ministry of Economy and the Ministry of Finance and their technical staff have been primary counterparts during the preparation of this report; and the team has benefited more generally from technical discussions and comments from the representatives of many institutions in Poland, including: the Chancellery of the Prime Minister, the Ministry of Environment, the Ministry of Infrastructure, the National Bank of Poland, the Energy Regulatory Office, the National Administration of the Emissions Trading Scheme, the Public Board of the National Program for Reduction of Emissions, and various representatives from academia, businesses, and consulting firms. The report received generous financial support on the development of methodology from the UK Department for International Development via the World Bank’s Energy Sector Management Assistance Program (ESMAP) as part of its program of Low Carbon Growth Country Studies. INTRODUCTION page 24 INTRODUCTION Against the backdrop of agreement that global coordinated action is needed to prevent dangerous climate change,1 individual countries are thinking through the implications of climate action for their economies and people. Some countries are already observing the impact of global warming on local weather and water supply. Others wish to position themselves as leaders in the ongoing international negotiations. With the expectation that a global price for carbon2 will eventually be established, some countries may wish to push to the front on emerging clean technology industries and avoid ‘stranded assets’—expensive long-lived infrastructure such as dirty coal-burning generators. Some simply want to inform policymaking on a key issue. Given that Poland ratified the Kyoto Protocol and hosted the December 2008 round of international climate negotiations,3 ‘carbon’ mitigation is not a new issue for Poland. But with its obligations as a mem- ber of the European Union making that commitment more concrete, it is an opportune time to assess more thoroughly the complex economic impact of emissions mitigation by Poland, in particular the expected tradeoffs between reducing greenhouse gases4 (GHGs) and sustaining economic growth and employment. There is a broad consensus that the world is warming and that human activity is primarily to blame. Average global temperatures and sea levels are rising while the extent of Arctic sea ice, mountain glaciers, and snow cover is declining. The Intergovernmental Panel on Climate Change (IPCC) has concluded that warming of the Earth’s climate system is un- equivocal and that anthropogenic (human-made) greenhouse gas emissions, generated mostly by the burning of fossil fuels and deforestation and changes in land use, are to blame. The level of carbon dioxide (the most important GHG) in the atmosphere is already the highest concentration in the last 650,000 years (at 379 parts per million (ppm) in 2005 as compared with 280 ppm in the preindustrial era). Via the ‘greenhouse effect’, these high and rising levels of GHGs are projected to raise average global temperatures over the next 100 years by 1 to 6°C. 5 (See Box 1). 1 Climate change is defined as changes in the mean or variability of weather (generally, temperature, precipitation and wind) over a multi-year period, generally 20 or 30 years (following usage by the Intergovernmental Panel on Climate Change). 2 Note that throughout this report, the words ‘carbon’ and ‘emissions’ are used interchangeably as a shorthand for greenhouse gas emissions, usu- ally measured in carbon dioxide equivalent (CO2e) units. 3 Poland is a signatory to the 1992 United Nations Framework Convention on Climate Change (UNFCCC) and has ratified the 1997 Kyoto Protocol. 4 Greenhouse gases trap heat within the atmosphere, creating the greenhouse effect (warming of the atmosphere which would otherwise have a temperature of -19°C). In this report, GHGs refer to the anthropogenic greenhouse gases covered by the UNFCCC: carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O); and the F-gases (or halocarbons) covered by the Kyoto Protocol: hydrofluorocarbons, perfluorocarbons, and sul- phurhexafluoride. 5 IPCC (2007), Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Inter- governmental Panel on Climate Change [Core Writing Team, R.K. Pachauri and A. Reisinger (eds.)]. Intergovernmental Panel on Climate Change, Geneva, Switzerland, 104 pp. page 25 Box 1. The Intergovernmental Panel on Climate Change’s (IPCC) Fourth Assessment Report The Intergovernmental Panel on Climate Change (IPCC), established by the United Nations in 1988, assesses scientific information and environmental and economic consequences of climate change, in support of the United Nations Framework Convention on Climate Change (UNFCCC). This box summarizes key conclusions of their 2007 Assess- ment Report, which today appears conservative in its conclusions. According to its most recent Report, representing a consensus view among more than 2000 scientists worldwide on climate change, warming of the climate system is unequivocal, based on evidence from increases in global average air and ocean temperatures (especially over the last 50 years), widespread melting of snow, Arctic ice, and mountain glaciers (over the last 30 years or so), and rising global average sea level (over the last 50 years or so). This global warming is being driven by rising atmospheric concentrations of greenhouse gases, in particular carbon dioxide. Human activities result in emissions of four long-lived GHGs: carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O) and halocarbons (a group of gases containing fluorine, chlorine and bromine that can destroy strato- spheric ozone). Atmospheric concentrations of GHGs increase when emissions are larger than removal processes. The IPCC report notes that “global atmospheric concentrations of CO2, CH4 and N2O have increased markedly as a result of human activities since 1750 and now far exceed pre-industrial values”. From a pre-industrial level of 280 ppm, CO2 concentrations were at 390 ppm in mid-2010. 1 There is little doubt that human activity is the cause of higher GHG levels and, therefore, of climate change. “Global increases in CO2 concentrations are primarily due to fossil fuel use, with land-use change providing another significant but smaller contribution. It is very likely [with a confidence level greater than 90 percent] that the observed increase in CH4 concentration is predominantly due to agriculture and fossil fuel use. The increase in N2O concentration is primar- ily due to agriculture.”2 Overall, human-made emissions rose 70 percent between 1970 and 2004 (see Figure 1), driven primarily by the energy supply sector. Fi Figure gure Figu re 1 1. Gl Glob Global obal al a annnual nual nnu al emissions e emi miss ions ssio of greenhouse ns of gr gree eenh nhou ouse g gas ases es se gases Note: N ote: Not ( (a) te: (a a) Gl Global obbal a annual nnua l ann ual emi l em emissions issi is ions sion fa s of anthropogenic nt nth throp hrop poggen eniic GHG ic GHGs G HGs HG from 197 s from 1 1970 970 97 to 200 0 to 2004 2 004 00 4 (b ( (b) Sh Shar are ) Sharee of d f dif different if iff fere ferenttaanthropogenic nt nth hrop throp pog gen eniic G GHGs HG HGs ic GHG s in total in total tot le emissions mi miss issi ions in ions 2004 in 20 04 (c) ) Share (c) Sh Share hare of different diff of di ff ffer eren ent sectors sect t sector s in ors total t in tototal anthropogenic anth l ant hrop hroppoggeneniic GHG e ic GHG emissions miissi miss ions ions in 2004 in 20 (Forestry ( Fore 04 (Fo rest stry ryyiincludes ncl inc lud des ludes deforestation.) de fore fo defo rest re st stat atio at ion io n.)) Gt GtCO CO2-e eq is gig -eq gig gigatonnes igat at aton onne on ne ness (b (bil (billions il illi li lion on ons s of mmetric metetri et ri ric c to tons tons) ns) ns c ) of car carbon ar arbo bo bonn didiox dioxide ox oxid id ide equi equivalent, e eq ui uiva va vale le lent nt comm common nt, a co mmon mm m on met metric et etri ric ba ric based basese sedd on the th differing diff e di ff ffer erin er ing in warming g wa warm rmin rm in ing influences infl g in flue fluenc ue nc nces es of of ea ch GHG each GHG. G HG HG. So Source: urce ur Sour ce ce: : IP CC (20 IPCC (2007), ( 2007 20 07) 07 p. 5 ), p 5.. In the absence of additional climate mitigation policies, global GHG emissions are projected to increase by 25 to 90 percent between 2000 and 2030. Using a range of scenarios, world temperatures are projected to rise by between 1.1 and 6.4°C compared with 1980-99 (with a confidence level greater than 66 percent) while sea levels will rise by 18 to 59 cm during the 21st century (but with a high degree of uncertainty). Further, with a confidence level greater than 90 percent, there will be more frequent warm spells, heat waves and heavy rainfall; and with confidence level greater than 66 percent, there will be an increase in droughts, tropical cyclones and extreme high tides. Abrupt or irreversible impacts are possible, such as partial melting of polar ice sheets (which would cause meters of sea level rise); changes in ocean circulation such as the Gulf Stream; and, if global average temperature increase exceeds about 3.5°C, extinction of 40 to 70 percent of terrestrial species and widespread coral mortality in marine ecosystems. Source: IPCC (2007). 1 Data from US National Aeronautics and Space Administration’s Jet Propulsion Laboratory. 2 IPCC (2007), p. 5 page 26 INTRODUCTION Unfettered climate change will impose enormous costs unevenly distributed across countries, with developing countries faring the worst. As the World Bank’s World Development Report 2010 has stressed, the projected rise in tem- peratures will create “a vastly different world from today, with more extreme weather events, most ecosystems stressed and changing, many species doomed to extinction, and whole island nations threatened by inundation”.6 A 2°C warming above preindustrial levels will cause more frequent and stronger extreme weather events, including heat waves, drought, flooding, and hurricanes; increased water stress in many world regions and especially in Africa and Asia; declining food production in many tropical regions as cereals become no longer cultivable in low latitudes; coastal erosion and aquifer salinization; and damaged ecosystems and biodiversity loss, including widespread dying off of coral reefs and shifting ranges for pests and diseases. These consequences will fall disproportionately on developing countries, with estimates of a 4 to 5 percent permanent reduction in annual income per capita in Africa and South Asia and a global average GDP loss of about 1 percent.7 While other parts of the globe will face the greatest harm, the countries of Central and Eastern Europe and Cen- tral Asia8 face considerable threats from climate change. Rising temperatures and shifting precipitation patterns will aggravate winter floods and summer droughts and heat waves. Precipitation intensity is expected to increase across the region while water availability is projected to decrease everywhere but Russia. The rapid melting of the region’s glaciers will reduce summer water availability, with severe impacts in irrigation-dependent Central Asia. Changes in sea level will affect the four major basins—the Baltic Sea, the East Adriatic and Turkey’s Mediterranean coast, the Black Sea, and the Caspian—as well as the Russian Arctic Ocean, threatening low-lying areas such as, for example, Poland’s heavily populated coast. Increased temperatures and changing hydrology are expected to generate substantial tree loss and degradation, the northward migration of pests, and the return of malaria to Europe.9 The international community has been negotiating a coordinated response to the threat of climate change for some time. Under the Kyoto Protocol to the United Nations Framework Convention on Climate Change (UNFCCC), indus- trialized countries and economies in transition (or ‘Annex 1’ countries under the UNFCCC) committed in 1997 to reduce greenhouse gas emissions by about 5.2 percent during 2008-12 compared to 1990. The UNFCCC climate summits in Copenhagen in December 2009 (Conference of the Parties or COP-15) and in Cancun in December 2010 (COP-16) aimed to make progress on post-2012 emission targets and their allocation. A 2 to 2.5°C increase in global temperatures above preindustrial levels by 2050 has been accepted as a target because it is considered achievable while also likely to prevent some of the most catastrophic potential effects of climate change, such as major increases in global sea level and disrup- tion of agriculture and natural ecosystems. The stabilization of greenhouse gases at 450 ppm CO2e (or carbon dioxide- equivalent),10 which would provide a 40 to 50 percent chance of limiting the temperature rise to 2°C, requires emissions to be reduced by at least 50 to 85 percent in 2050 compared to 2000 levels and global emissions need to peak prior to 2020, according to the IPCC. Intermediate targets for 2020 have also been suggested, including an indicative range of 25 to 40 percent reductions compared to 1990 for developed and transition countries.11 The European Union has taken a proactive stance through its ‘climate and energy package,’ setting ambitious miti- gation targets for its members for 2020 in advance of an international agreement. Following the European Council’s decision for unilateral emissions reductions of 20 percent by 2020 at its March 2007 summit, the package of measures referred to as the ‘20-20-20 targets’ was approved by the European Parliament in December 2008 and became law in June 2009.12 By 2020, EU emissions are to be cut by 20 percent (or 30 percent if a global deal is reached); energy efficiency 6 World Bank (2009), World Development Report 2010: Development and Climate Change, p. 1. 7 World Bank (2009), p. 5. 8 Central and Eastern Europe and Central Asia includes: Albania, Armenia, Azerbaijan, Belarus, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Estonia, Georgia, Hungary, Kazakhstan, Kosovo, Kyrgyzstan, Latvia, Lithuania FYR Macedonia, Moldova, Montenegro, Poland, Romania, Serbia, Slovakia, Slovenia, Tajikistan, Turkey, Turkmenistan, Ukraine, and Uzbekistan. 9 Fay, Marianne, Rachel I. Block, and Jane Ebinger, eds. (2010), Adapting to Climate Change in Eastern Europe and Central Asia (World Bank). 10 GHGs differ in their warming influence (radiative forcing) on the global climate system due to their different radiative properties and lifetimes in the atmosphere. These warming influences may be expressed through a common metric based on the radiative forcing of CO2. CO2-equivalent emission is the amount of CO2 emission that would cause the same time-integrated radiative forcing, over a given time horizon, as an emitted amount of a long-lived GHG or a mixture of GHGs. 11 IPCC (2007). 12 European Union (2008), The Climate Action and Renewable Energy Package: Europe’s Climate Change Opportunity. page 27 is to be increased by 20 percent; and 20 percent of energy used is to come from renewables.13 Higher emission sectors are included in an EU-wide cap-and-trade system (the Emissions Trading Scheme) while other sectors face national targets only. Thus, EU members such as Poland already face specific obligations for climate action. Poland faces a particular challenge in CO2 mitigation because of its reliance on abundant domestic coal. 85 percent of Poland’s GHG emissions come from the energy sector, and more than 90 percent of electricity comes from coal-fired power plants (which emit the highest levels of CO2 per unit of electricity of any power generation technology, and roughly two to three times as much as equivalent gas-fired plants). Despite progress over the last two decades, Poland’s economy remains twice as energy intensive as the EU average. Also, while emissions overall have fallen by near 30 percent since Poland’s transition to a market economy began, those from the transport sector have grown by almost three-quarters (al- though they still constitute just over 10 percent of total emissions). There is understandable concern in Poland that a move towards a lower carbon economy will boost electricity prices, already amongst the highest in the region, which in turn will undermine welfare and profitability, with devastating effect on employment at home and competitiveness abroad. How costly will it be for Poland to move to a lower carbon path? What combination of energy efficiency, shifts in fuel for power generation, and other measures is most desirable? How steep is the tradeoff between carbon abatement and growth? Through the Low Carbon Growth Country Studies Program, the World Bank has been supporting selected coun- tries’ work on lower carbon development paths. In 2007, the donor community asked the Bank to build on its ex- perience in developing country-specific marginal abatement cost curves which aggregate the incremental costs of GHG mitigation measures relative to a business-as-usual scenario and the associated financing needs. The Bank was asked to assist in preparation of low carbon country case studies for Brazil, China, India, Indonesia, Mexico, and South Africa. These studies aim to integrate carbon abatement targets with objectives for economic growth and poverty alleviation. The Bank has been careful to ensure that these studies are client led, to help ensure the transition to implementation. As a result, each study has taken a different approach, appropriate to the client country and building on experience. This report on Poland draws on that ongoing experience but aims to go further in addressing the macroeconomic impact of a low emis- sions growth strategy by integrating ‘bottom-up’ engineering analysis with ‘top-down’ economy-wide modeling. The rest of the report is organized along the following lines. The next section provides background on Poland’s green- house gas emissions. Then section B sets out Poland’s existing carbon abatement targets and key policy challenges related to GHG mitigation. The next section summarizes the innovative methodological approach used by the report. Section D discusses the methods and implications of constructing business-as-usual or reference scenarios. Section E provides the major findings from the first model, the engineering approach, on the costs of measures aimed at GHG mitigation for Po- land. Section F explains how these findings are expanded and revised by incorporation into the first macroeconomic mod- el. Section G provides an analysis of the economic impact through 2020 of mitigation measures within the constraints of EU policy arrangements. Section H examines the energy sector and how Section E’s findings are enhanced by optimization of the structure of the energy sector. Section I takes a first look at the challenges of energy efficiency. Section J provides additional analysis of the transport sector. The last section provides some notes on additional issues and further work. 13 Renewable energy (or renewables) is energy which comes from natural resources such as sunlight, wind, rain, tides, and geothermal heat, which are renewable (naturally replenished). a. POLAND’S GREENHOUSE GAS EMISSIONS page 30 POLAND’S GREENHOUSE GAS EMISSIONS Poland is not among the largest emitters of greenhouse gases globally, but its economy is among the least carbon- efficient in the EU. Poland’s global share in GHG emissions is just 1 percent; and its per capita emissions are about the average for the EU. Poland cut its emissions considerably as a side effect of the restructuring of transition to a market economy, but the link between growth and emissions has re-emerged in recent years. A critical difference in the make-up of Poland’s emissions is the dominance of the power sector and its extraordinary dependence on coal. Apart from energy sector, Poland’s transport sector has experienced very high rates of emission growth, and energy efficiency, although im- proving, remains below EU averages. Poland contributes marginally to the global carbon footprint, with a share in global GHG emissions equal to about 1 percent. The EU as a whole is responsible for about 13 percent of global emissions, while China and the US, the larg- est emitters, are responsible for almost 40 percent of global emissions between them. (Figure 2). On a per capita basis, Poland emits about 10 metric tons of CO2e (tC02e) each year, which is the average across the EU (with most countries at between 7 and 15 tC02e per capita). On average, Europeans emit less than half the greenhouse gases of North American or Australian citizens. Nonetheless, this level remains well above the global average of 7 tC02e as well as the benchmark of 2, the average global per capita emissions consistent with a 2°C rise in temperature.14 Figu Fi gure re 2 Figure 2. Wo Worl World’s rld’ d s largest rge lar st g gest gre greenhouse reen enho hous use gas e ga emit emitters, s emitte ters rs, 20 2005 2005, per percent 05, in p cent ercent Other, 28.9 EU15, 11.0 Czech Rep, 0.4 Mexico, 1.7 Estonia, 0.1 Canada, 2.0 Hungary, 0.2 Latvia, 0.0 Brazil, 2.7 Lithuania, 0.1 Japan, 3.6 EU10, 2.4 Poland, 1.0 India, 5.0 Slovakia, 0.1 Slovenia, 0.1 Bulgaria, 0.2 Russia, 5.2 Romania, 0.3 China, 19.2 US, 18.4 S Sour So Source: our urce ce ce: W : Wo World orl rld ld Re R Resources eso Reso sour ur urce ces In ces I Institute, nst stit titut ute te, W World or e, Wor orld ld Ban B Bank ank ank st staff taf aff ff ca calculations. lcul lc cal lat ati tio ions ions . ns. Despite unremarkable overall emissions levels, Poland’s economy remains among the least carbon-efficient in the EU. In 2007, around 1.3 metric tons of CO2e were required to produce €1 million in GDP, while the EU average was less than 0.5 tC02e. This high emissions-intensity of the economy is due partly to high amounts of CO2 generated by the en- ergy consumed but also to the high energy intensity of production in Poland. While in the EU on average, consumption of energy equal to one ton of oil equivalent15 generates 2.5 metric tons of CO2, in Poland the same ratio is around 3.4 (Figure 3), despite the downward trend of carbon intensity in Poland over the last two decades. At the same time, energy used per million euros of GDP, at 400 tons of oil equivalent, greatly exceeds the EU-wide average of 169 (Figure 11) and stands at about the world average (Figure 4). Among transition economies, Poland’s performance appears better: its carbon intensity on a per capita basis is situated in about the middle of the countries of Eastern and Central Europe and Central Asia (see Figure 5). 14 The Contraction and Convergence model developed by the Global Commons Institute estimates that to contain global warming to 2°C increase, which is typically associated in climate models with a CO2e concentration of 400-500 ppm, emissions per capita must come down to 2 tC02e per capita by 2050. The Institute has advocated for an egalitarian sharing of emissions abatement costs under which every country brings emissions per capita to the same level. 15 Toe (ton of oil equivalent) is the amount of energy released by burning one ton of crude oil, approximately 42 GJ or 11.63 MWh (according to the IEA and OECD). page 31 Fi Figure gure Figu re 3 3. CO2 i int intensi nten sit ensi y of e ty ene energy nerg rgy use y us Pol e in P Poland olan and and d an EU27 d EU 27 Fi Figu Figure gure re 4 Ener Energy 4. En gy i ergy int inten nten ens sit ity y across sity ac acro ross ss countries, cou c ntri ountries es, 2007 2007 (toe/M€) ( (to toe/ e/M€ M€) ) 2,500 Poland CO2 intensity EU 27 CO2 intensity 4.0 2,000 3.5 1,500 toe/M€ tCO2/toe 3.0 1,000 2.5 500 2.0 0 LU LT AT SE World DE HU EE LV CAN ES PT US KOR RO CY FI SI BE RU IE BRA IT CZ FR MT CHN NL BG MEX GR DK PL SK JP UK IND EU 27 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Note Note: Note: int : CO2 intensity inten ensi sity y is ty meas is me measure asur ure m metric etri e in met c to ric tons ns of per ton of CO2 per tons t onss of oil oil equ e equivalent q iv qu alen ivalent t co cons consumed. nsum ed. En umed Ener Energy ergy gy int i intensity nten ensi ty y is sity is th the e ra ratio tio ratio of gross g gros oss inland s inland inla consumption c nd con sump onsu mp pti tion of en on of energy enerergy (in gy ( toe, t oe, in toe tons , to ns of of oi oil equivalent) qui l eq vale uiva lent nt) ) to GDP GDP ( (in million m in mil illi lion on eur euros e os at uros 2000 at 20 prices). 00 p rice rices) ). s) Source: So Sour Sour u ce European Euro e: Eu ope pean an CCommission, om o mi miss iss ssiion, Wor ion io World W or Bank B o ld Ban ank staff s k st aff ff ca taf calculations. cal lcul lc ati tio lations ionss. Fi Figu Figure gure re 5 5. Ca Carbon rbon Carb on i int intens nten ens ity sit Cen Central y in C tral entral a and nd E Eas Eastern tern aste rn E Eur Europe urop ope and e an d Ce Cent Central ntra ral Asia Asia, l As 2005 ia, 2005 20 Slovenia 18 16 GDP per capita, , US$ ‘000 14 Czech Republic Slovak Republic 12 Hungary Estonia 10 Croatia Lithuania 8 Poland Latvia Turkey Russian Federation 6 Romania Bulgaria Turkmenistan 4Macedonia, FYR Kazakhstan Serbia Bosnia and Armenia Albania Herzegovina 2 Georgia Azerbaijan Belarus Ukraine Kyrgyz Republic Uzbekistan 0 Moldova Tajikistan 0 2 4 6 8 10 12 14 16 18 20 GHG emissions per capita, tCO2 e Note No Note: te te: : Si ze of Size of ci circ circle rc rcle le ind indicates i nd ndic ic icat at ates t es tot total ot otal CO2e em al CO emis emissions is issi si sion on ons s fo for for each cou each c country. ou ount nt ntry y. ry Source: S Sour our So ce ce: urce World : Wo W rld orl Bank B ld Baank staff s k st taff c taff ta calculations. al alcu lcu cullation ion lationss. page 32 POLAND’S GREENHOUSE GAS EMISSIONS Poland’s transition to a market economy had a co-benefit of sharply reduced carbon emissions. From 564 million metric tons of CO2e in 1988, greenhouse gas emissions collapsed along with output through 1990 (declining 20 percent), as inefficient, often highly energy-intensive plants shut down during the early years of transition. The period of 1996 to 2002 witnessed another 17 percent decline in emissions but while GDP expanded. Overall, although Poland’s GDP near doubled during 1988 to 2008, its GHG emissions were reduced by about 30 percent. Nevertheless, during the last half decade or so, a more traditional positive correlation between GDP growth and GHG emissions has re-established itself. (See Table 1 and Figure 6).16 Table 1. Poland’s greenhouse gas emissions, 1988, 2000, and 200 Emissions, in MtCO2e 1988 2000 2008 GHG emissions (without LULUCF) 564.0 390.2 395.6 Net emissions/removals by LULUCF -28.7 -24.5 -39.2 GHG net emissions with LULUCF 535.3 365.7 356.4 GHG emissions (without LULUCF) 1988 to 2000 2000 to 2008 1988 to 2008 Changes in emissions, % -30.8 1.4 -29.9 Average annual growth rates, % -3.0 0.2 -1.8 Notes: LULUCF is land use, land use change, and forestry. Source: Fourth National Communication under the UNFCCC. Figure 6 Fi Economic 6. Economi ic gro growth wth th a nd anddGGHG HG e emissions i ions i missi Poland, in Pol land, d 19 1988 1988-2008 2008 88-200 8 GDP index, 1988 = 100 CO2 emissions index, 1988 = 100 200 150 100 50 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 S Sour So our urce ce ce: Source: : Wo W World rld ld Re orl Reso Resources R eso sour ur urce ces ce I Institute, s In stit titut nst ute te, UNF U UNFCCC, NFCC NF CC CCC C Central en C, Cen ent tr tral lSStatistical tati ta tatist St tic sti ical l Off O Office, ffi ffic ice, Wor ice W World or orld ld Ban Bank B ank st ank taf aff staff ff ca cal calculations. lc lcul ati lat ions ions tions. Poland’s types and sources of greenhouse gas emissions resemble those for the rest of the EU except for the elec- tricity sector. The breakdown of Poland’s greenhouse gas emissions by type of gas show that its emissions are predomi- nantly CO2 (with a more than 80 percent share), with the EU overall at about the same level. Compared with the rest of the world, emissions from agriculture are less important in the EU and in Poland. One point of departure from the EU and even from the EU1017 is Poland’s greater emissions from the electricity and heat sector (Figure 7 and Figure 8). 16 Net emissions removals by land use, land use change, and forestry (LULUCF) are shown in Table 1. Because they are not a central issue for Poland and because consistent cross-country measurement of LULUCF remains under discussion, the remainder of this report considers emissions without LULUCF. 17 The EU10 consists of Bulgaria, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Romania, Slovakia, and Slovenia. page 33 Fi Figure gure Figu 7. GH re 7 G em GHG emis emissions issi sion ons gas s by ga gas, s, 2 200 20077 007 Figu Fi Figure gure 8. GH re 8 GHG emis emissions G emissi sion ons se s by s ector, ctor ect 2007 or, 20 07 Electricity & Heat Manufacturing & Construction CH4 CO2 Fluorinated gases N2O Transportation Other Fuel Combustion 100% Fugitive Emissions Industrial Processes 90% Agriculture Waste 100 80% 9.2 9.2 8.8 16.1 70% 80 60% 13.4 12.7 9.9 10.3 50% 72.9 80.0 82.3 83.0 60 11.8 9.7 14.2 19.5 40% 8.8 12.0 30% 40 13.8 12.8 20% 20 39.4 46.1 10% 32.6 32.0 0% 0 World* EU27 EU10 Poland World* EU27 EU10 Poland Note No Note: te te: : *w wor orld *world ld dat d data ata at from a fr 2 2005. om 200005 00 5. Source: Sour S So our urce ce: ce World W : Wo rld orl Resources R Reso ld Reeso sour ur urce ces ce Institute, I Inst s Innst sti itut it itut ute European E urop ur e, Eur op pea ean Commission, Comm C n Co mmi ommis issi sion ion on, World W , Wo rld orl Bank ld Ba B ank staff k sta s taff ta calculations. ff c al alcu lcula lati cul ion ons tions. Poland’s energy mix is dominated by coal to such an extent that it is an outlier in both Europe and globally. In con- trast to the EU overall or even to the EU10, in Poland solid fuels (coal and lignite) constitute 57 percent of gross inland energy consumption (Figure 9). The share of natural gas (13 percent) and renewable energy (5 percent) are significantly below the EU15 and EU10. Also, Poland is one of 11 countries in the EU and one of 3 countries in the EU10 with no en- ergy generated by nuclear power plants. Poland’s dependence on domestically available coal is one of the highest in the world. Over 90 percent of electricity in Poland is generated from coal and lignite (Figure 10), which is the highest share in the EU. Figure 9 Fi Energy 9. E consumption tion b nergy consumpti byyf fuel fuel, 200 2007 l, 2 7 007 Fi Figure 1 10 10. El Electricity 0. E tri lect ity gener icit generat ati tion b fuel, by f l 20 uel, 07 2007 Coal Oil Natural gas Nuclear Renewables Other Coal Natural Gas Oil Nuclear Renewables Other 100% 100% 3 0 8 7 5 9 3 12 16 9 13 6 13 80% 80% 20 21 21 26 28 24 10 60% 60% 25 91 34 40% 23 40% 36 57 59 20% 39 20% 26 29 18 0% 0% World EU27 EU10 Poland EU27 EU10 Poland No Note: te Note te: : En Ener Energy ergy er gy con c consumption onsu onsump sumppti tion on is g is gr gross os osss in inla inland la land nd con consumption con onsu su sump mp tion pti on of of en ener energy. er ergy gy gy. Source: Sour Sour Source ce: ce: Eu European E Euro uro rope pe Commission, C p an Com ommi iss missssiio ion, ion n, Wor World W orld or Bank ld Ban B ank an staff k st aff ff ca taf calculations. cal lcul lc lat ati tio ions ions ns.. Poland has made considerable advances in energy efficiency in the past 20 years; yet further efforts are required to bring it to Western European standards. Per unit of GDP, Poland’s economy is still more than twice as energy intensive as the EU average.18 Advances in energy efficiency, which were dramatic during 1988 to 2000, have slowed during the most recent decade (see Figure 11). Consumption of energy per € of GDP has fallen by half during 1990 to 2007, from 781 tons of oil equivalent required for every hundred million euros of output to 400. From a level of energy intensity 3.4 times higher than the EU average, Poland as of 2007 stands 2.4 times above the EU. 18 Alternative statistics, using GDP adjusted for purchasing power parity, as reported by the IEA, suggest a smaller gap between Poland’s and Western European energy intensity of about 30 percent. Fi Figure gure Figu 11 re 1 11. 1. E Ene Energy rgy nerg inte y in intensity tens nsit ity EU27 a y in EU27 nd P and Pol Poland, olan and in d, i toe/ toe/M€ n to M€ e/M€ Poland energy intensity EU27 energy intensity Poland EU27 850 781 750 650 550 toe/M€ 450 400 350 250 233 150 169 50 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 No Note Note: te te: Ener Energy : Ener ergy gy int i intensity nt nten ensi en sity sity y is is th thee ra rati ratio ti tio gross o of g ro ross ss inl i inland nlan nlan and cons consumption d co ns nsum umpt um p io ption n of e ene energy ne nerg rg gy (i (in toe toe, n toe, e, ton t tons ons of oil ons oil equ e equivalent) q iv quival al alen en ent) to GD t) to GDP P (i (in mill millions n millio ll ions ions fe of euros ur u os at 2000 t 20000 prices). i es) price ). Source: Sour So ur urce ce: ce European Euro : Euro rope pe pean Commission, C ommi an Com miss iss ssiion, Wor ion io World W orld or Bank ld Ban B an ank staff k st af ff ca aff calculations. cal lcul lc ulat lat ati ions io ions ns. While the energy sector currently dominates Poland’s emissions profile, emissions from the transport sector have been growing at a high rate. Energy sector emissions have fallen by one-third since 1988, although the sector still pro- duces near half of the country’s greenhouse gases. Transport, on the other hand, while constituting about 10 percent of overall GHG emissions has grown by almost three-quarters since transition. Moreover, Poland still has relatively low rates of motorization, which argues that the growth of road transport will likely be high going forward. Further complicating the picture is the very high share of used vehicles, which tend to be much more fuel inefficient and polluting (see Figure 12). Figure 1 Fi 12. 12 Ch 2. C Change hange i in GHG nG emissio HG emissi i ions by by k key sect tor, 1 sector, 1988 988 198 8tto o22006, 006 200 in percent 6, i t Energy sector -30.1 Mfg. & construction -20.5 73.5 Transport -51.4 Other sectors Industrial processes -17.1 -60 -40 -20 0 20 40 60 80 Note No Note: te te: Indu Industrial : In du dust st stri ri rial processes al p roce ro cess cess sses es emi e emissions miss mi ss ssio io ions ns con consist c on onsi si sist of by st of by-product y-p pro rodu duct duct or or fu fugi fugitive gi g ti ve emi tive e emissions mi miss ss ssio io ions ns of of gr greenhouse g ee eenh nhou nhou ouse se g gases, ases ases es, excl excluding , excl ud udin clud in ing g em emis emissions issi is sion si ons ons from fuel f from fueuel combustion. l co mb mbus ti tion busti on.. Source: Sour So urce urce ce: UNFCCC, : UN UNFCFC FCCC CC, Gr CC Greenhouse Greeeenh ee nhou nhou ouse Gas Inv se Gas Inventory, I nv nven en ento tory to ry, 2006. ry 2006 20 06. 06 The level and structure of Poland’s greenhouse emissions will be important as the next sections lay out the chal- lenges of moving towards a lower carbon growth path. Poland’s overall carbon intensity of GDP, the sectoral composi- tion of emissions, its dependence on coal, and its progress to date will all be important factors in assessing the economic costs of abatement. The combination of large energy and carbon efficiency gaps in Poland and huge investment require- ments in energy, infrastructure, and housing suggests there is a substantial scope for climate-smart policy choices that would likely yield benefits regardless of climate developments. b. CARBON ABATEMENT TARGETS AND POLICY CHALLENGES FOR POLAND CARBON ABATEMENT TARGETS page 36 AND POLICY CHALLENGES FOR POLAND The international agreement on climate change that will eventually supersede the Kyoto Protocol and, more im- mediately, compliance with EU policies on climate change, pose policy challenges for Poland. Poland’s greenhouse gas emission levels and its achievements to date seem to argue that further movement towards a low emissions economy might simply be a matter of accelerating existing momentum. On the other hand, Poland’s heavy dependence on coal would seem to make such a transition highly challenging. While Poland has been actively involved in global discussions on climate change policy and has easily met the Kyoto Protocol’s requirements for emissions reduction, the country now faces a complex set of regulations under the EU’s climate and energy package and will likely face ambitious mitigation targets as part of an eventual global agreement. Poland has been an active participant in international negotiations on climate change. Poland participated in the Earth Summit in Rio de Janeiro in 1992 and adopted the United Nations Framework Convention on Climate Change. In mid-1998, Poland ratified the Kyoto Protocol (Box 2) and committed to a modest reduction of GHG by 2012 (see Box 2). The country hosted the 14th Conference of the Parties (COP-UNFCCC) to review the Convention’s progress and discuss its successor at Poznan, Poland in December 2008; and with Indonesia (the chair of the preceding COP in December 2007), and Denmark (the chair for December 2009), Poland formed a ‘troika’ of countries aiming at a new treaty agreement to replace the Kyoto Protocol which lapses in 201 Box 2. The Kyoto Protocol of the UNFCCC The Kyoto Protocol is an international agreement linked to the United Nations Framework Convention on Climate Change. While the UNFCCC encouraged industrialized countries to reduce GHG emissions, the Kyoto Protocol set bind- ing targets for Annex 1 Parties--37 industrialized countries, including all members of the European Union--by an aver- age of 5 percent against 1990 levels over the five-year period 2008-2012. Some countries with economies in transition negotiated a pre-transition base year: Bulgaria (1988), Hungary (average of 1985 to 1987), Poland (1988), Roma- nia (1989) and Slovenia (1986). The Protocol was adopted in Kyoto, Japan, in December 1997 and entered into force in 2005. Among industrialized countries, only the United States failed to ratify the Protocol. Countries must meet their targets primarily through national measures, but the Protocol offers three alternative (or flex- ibility) mechanisms: • International emissions trading (“the carbon market”). Annex 1 countries that have emissions permitted but not “used” are allowed to sell excess capacity to other Annex 1 countries. Because carbon dioxide is the principal green- house gas, the market is called the “carbon market.” At present, the European Union Emissions Trading Scheme (EU ETS), established in 2005, is the largest in operation. • Clean development mechanism (CDM). Annex 1 countries can meet part of their caps using credits generated by CDM emission reduction projects in developing countries. • Joint implementation (JI). Annex 1 countries can invest in JI emission reduction projects in any other Annex 1 coun- try as an alternative to reducing emissions domestically. Source: UNFCCC website. As noted earlier, Poland’s transition to a market economy had a co-benefit of sharply reduced carbon emissions, causing it to outperform against its Kyoto commitments. The country continues to exceed its Kyoto targets by a large margin, with its 2007 level of emissions 29 percent below the base year (against a target of 6 percent).19 For the EU over- all, it has been the large shifts in energy intensity by transition countries who joined the EU in 2004 and 2006 that have allowed the EU as a whole to meet its aggregated Kyoto commitments. The 15 members of the European Union from before 2004 have reduced greenhouse emissions by 4 percent by 2007 compared to the base year, while the EU10 have 35 percent lower emissions. Together, the 27 members of the EU have cut emissions by about 12 percent (exceeding their combined Kyoto target of 8 percent).20 Energy security and climate action have been included as key priorities in recent government strategy documents. The Poland 2030 Report was prepared by the Board of Strategic Advisers to the Prime Minister and presented in May 19 Poland can sell its surplus emission reductions to deficit countries under the Kyoto Protocol, which sets specific rules and arrangements (Box 2). The first transaction was with Spain in November 2009, followed by two transactions with Japan and one with Ireland. By end-May 2010, the total value of transactions amounted to over €80 million. 20 But note that Kyoto commitments are for the EU15 and separately for another 10 member states. page 37 2009.21 One of its ten key development priorities is a harmonization of climate change and energy challenges to ensure adequate energy supplies while meeting environmental targets, including climate protection. Among Poland’s objectives for 2030 are to achieve economic growth without additional demand for primary energy and to reduce the energy inten- sity of Polish economy to the EU-15 level. The Government’s commitment to low-emissions growth was confirmed in the Energy Policy of Poland until 2030, adopted by the Council of Ministers in November 2009.22 In line with the above, the primary aims of Polish energy policy are to: • improve energy efficiency; • enhance the security of fuel and energy supplies; • diversify electricity generation by introducing nuclear energy; • develop the use of renewable energy sources, including biofuels; • develop competitive fuel and energy markets; and, • reduce the environmental impact of the power industry. As an EU member state, Poland is subject to EU policies on climate change mitigation. In particular, Poland must com- ply with the climate and energy package referred to as “the 20-20-20 targets”, which was approved in December 2008 (see Annex 1 for more details). This package requires comprehensive action by EU members on overall emissions reduction across all sectors in the economy. With the EU aiming to lead the global action against climate change, it has set ambitious targets for 2020 for itself: a 20 percent reduction in greenhouse gas emissions compared to 1990 levels (a 14 percent reduction compared to 2005); a 20 percent renewable energy target as a percent of gross final energy consumption, including a 10 percent share of biofuels in the transport fuel market; and a 20 percent reduction in primary energy use compared to projected levels under a business-as-usual scenario, to be achieved through energy efficiency improvements. Table 2. Breakdown of EU 20-20-20 regulations by sector groups ETS s(that must use the EU Emissions Trading Scheme for CO2 permits) Non-ETS (with national targets) Power Non-power Power stations and Oil refineries, coke ovens, iron and steel, cement, Transport, construction, services, smaller industrial other large fuel com- glass, lime, bricks, ceramics, and pulp, paper and and energy installations, agriculture, and waste. bustion installations board, petrochemicals, ammonia, aluminum, acid production, and aviation (possibly covered from 2011 or 2012). Source: World Bank staff based on EU publications. 21 Board of Strategic Advisers to the Prime Minister, (2009), Poland 2030: Development Challenges, . 22 Ministry of Economy (2009), Energy Policy of Poland until 2030, Appendix to Resolution no. 202/2009 of the Council of Ministers, Government of Poland (10 November). CARBON ABATEMENT TARGETS page 38 AND POLICY CHALLENGES FOR POLAND Figure 13. GHG emissions in Poland and EU, by ETS and non-ETS sectors, %, 2005 100% households other services 80% public services non ETS finance 60% construction trade 40% transportation ETS light industry 20% agriculture fuels 0% energy Poland EU27 heavy industry Note: N Not te: Th ote: Thee brea b breakdown reakd kd kdowown n is a is app approximate, ppro xima roxiimat te base te, based b asedd on sec s sector ecttor dat tor data d ata ta a ava available vail abl ilab le in ble the in th e na nati national ti tion onal a accounts. ccou l acc ount nts ts. Bec Because B ause ecau se tht the he he EU regu regula- EU re la- la- gul tions tion ti ons apply appl s ap ply y by ins installation, i tall nsta atio llat ion n, the the ETS share s ETS sha re for hare Poland for Pol P and olan overestimated o vere d is ove rest stim imat ateded as includes i as it inc nclu lude des small smal s sm all energy ener l en gy ins ergy installations, i tall nsta atio llat ions ns, fo for r ex example. ampl exam e. ple Accordingly, Accordingl g y, the non-ETS share is underestimated. Based on the 2005 data rep reported ported by y KASHUE-KOBiZE, , ETS emissions in 2005 20 05 were were wer e about ab abou out t 17 MtCO MtCO MtC O2e lolowe lower wer than r th an the the res r results esul ults f ts fro from romm ththe brea breakdown e br eakdkdowown n by secsector s tor ecto r (2 (203 03 i instead nste ins ad of tead of 22 220 MtCO 0 Mt e). CO2e) Source: Sour So ce: urce World Worl : Wo rld d Ba Bank staff s nk sta ff cal taff calculation c alcu lati cula on bas tion based b ased ed on on EU KLE KLEMS K LEMS database MS dat d abas atab ase e (o (on productivity prod n pr oducucti vity tivi by in ty by industry dust indu stry ry for for E EU member memb U me mber states s tate er sta tes with s wi th ab breakdown kdown i reakd into nt contributions to cont ibuti trib tions from f capital ca ital pit (K), (K) l (K labor ), l abor b (L)(L), ( L), energy (E)(E), ( materials E), mat teri l (M ials (M) ) and service d servi ice iinputs nput (S)). ( ts (S)S))). The 20-20-20 package segments sectors into two groups as well as setting multiple targets. The targets for over- all reduction and for renewable energy have been translated into legally binding commitments, while the third target (on energy efficiency) has been left as indicative only. Large installations in energy-intensive sectors are covered by the EU-wide Emissions Trading Scheme (EU-ETS), a cap-and-trade arrangement (see Table 2). These sectors—energy, heavy industry, and fuels—are referred to as ’ETS’ sectors, while everything else is categorized as ‘non-ETS’ sectors. In Poland, approximately 60 percent of CO2 emissions in 2005 were generated in the ETS sectors (compared with about 40 percent in the EU as a whole) (see Figure 13). For the non-ETS sectors, the package requires a reduction in emissions by 10 percent compared to 2005 in the EU27. That EU-wide target was translated into a national target for Poland of an increase in its non-ETS emissions by 14 percent (see Figure 14). Poland also committed to a 15 percent share of renewable energy in gross final energy consumption by 2020 (up from 7.5 percent in 2006), including a 10 percent share of biofuels in the transport fuel market (overall fuel consumption).23 23 The third “20” of the 20-20-20 package—a 20 percent reduction of primary energy consumption compared to the business-as-usual level for 2020 (that is, an increase in energy efficiency by 20 percent)—is not legally binding. page 39 Figure 14. GHG emission targets in the EU 20-20-20 package Note Notes: s: Emi Notes: E Emission miss ssio ion cred crediting n cr itin edit ingg fr from om p roje roje j ct cts projects s un unde undertaken dert aken rtak in ot en in othe other her r co countries untr coun ies tries wa was s se set p und t up under u er the nder the Kyo K Kyoto y to Protocol yo Proto Pro col toco l (s ( (see Box 2) ee Box 2). 2 ). CDM CDM is Clean Cle Cle lean an Dev Development D ev evel elop el op ment pme Mechanism; M nt Mec echa echa hani ni nism sm; sm ; JI is Joint is Jo in int Join Implementation. Impl t Im pl p em enta en emen ta tati ti tion on on. Source: Sour S ce: ource regulations; r : EU reg eggul ati tions lations; World W ; Wo orl rld ld Ba Bank B k st ank staff s taff calculations. taff c alcu allcullati lati tion onss. The EU targets are more ambitious than Kyoto targets and, therefore, likely to require more efforts, sectoral ad- justments, and resources from EU members to achieve. In contrast to Kyoto, there are no overall country targets. The national targets are only for non-ETS sectors, while the reduction target for ETS sectors is EU-wide. In the most important ETS sector—power—auctions will be phased in gradually from 2013, and full auctioning of ETS permits is to be in place by 2020. Aviation may be included into the EU-ETS as early as 2011. Other industrial ETS sectors will step up to full auction- ing by 2020, while sectors particularly vulnerable to competition from producers in countries without comparable carbon constraints (carbon leakage24) will have until 2027 to be phased in. In addition, auctions will be open. Thus, any EU opera- tor will be able to buy allowances in any member state. The EU ETS phasing-in process is presented in detail in Annex 1. The EU 20-20-20 package contains both an EU-wide cap-and-trade approach and possible national carbon taxes. The EU Emissions Trading Scheme for energy-intensive large installations is a cap-and-trade mechanism—policymakers set quantities and the market determines the price. The abatement target for the ETS sectors is EU-wide, and emissions in the EU in 2020 will have to be 21 percent lower than in 2005 (Figure 15). For smaller installations and those in less energy-intensive sectors, each member state may specify additional domestic abatement policies to comply with their country-specific targets, and many may consider introducing carbon taxes in these sectors—that is, setting prices instead of quantities. As discussed in Annex 2, policymakers can choose between controlling price and controlling quantity, taking 24 Carbon leakage occurs when emissions reductions are offset by increases in other countries. For example, if the emissions policy of the first country raises local costs, a trading advantage may be created for the other country with lower standards, and production may move offshore. CARBON ABATEMENT TARGETS page 40 AND POLICY CHALLENGES FOR POLAND into consideration aspects such as transparency, operating (or transaction) costs, public acceptability, dynamic efficiency, revenue and distributional issues, and international harmonization. Figure 1 Fi 15. 155. E EU EU-wide U-wid ide and dP and oland’ Poland’s l d’s 20 2020 20 targets, ETS a ETS nd andd non non-ETS -ETS ETS sectors, M MtCO CO2e and tCO 2005 d % vs. 20 05 EU27 Poland 3000 2500 -10 % 2000 MtCO2e 1500 -21 % 1000 500 +14 % 0 2005 2020 2005 2020 ETS non ETS So Source: ur Sour urce ce: ce UNFCCC, : UNFCCC CCC FC CC, , Eu Euro European rope rope p an Com C Commission, ommi miss iss ssiion io ion, n, Wor World W or orld ld Ban B Bank ank ank st staff aff af ff ca cal calculations. lcul lc lat ulat ati ions io ions ns. . Poland is likely to receive significant fiscal revenues in the future from ETS auctions. Revenues from the ETS will add to member states’ revenues in proportion to the emissions traded through their national systems. The allocation of the revenues will be a national policy choice; however, countries are encouraged by the EU to use at least half of them on ‘green’ initiatives. The EU also encourages that part of the revenues go towards helping developing countries adapt to climate change. Future fiscal revenues from ETS auctions in Poland, as elsewhere, will be inversely related to the number of allowances allocated for free. The potential derogations for the modernization of electricity generation25 and granting emission allowances for free will proportionally decrease revenues from auctions. A very rough estimate of total potential allocation over 2013-2020 is 1.8 billion metric tons of CO2. Assuming, for example, that 50 percent of allowances in the EU will be distributed for free during 2013-2020, the remaining 900 million will be auctioned. At a price of €15, which is close to the market price of carbon as of late May/June 2010, total revenues during 2013-20 would reach more than €13 billion; if the price were higher, e.g., €25, revenues would exceed €22 billion. According to some preliminary estimates for 2013 applying EU regulations26, Poland may have 155 MtCO2 of allowances available for auctions (see Table 20 in Annex 1), which would be worth over €2 billion at a carbon price of €15 per metric ton, or near €4 billion at a price of €25. These amounts are equivalent to 0.8 and 1.3 percent of projected GDP in 2013, but any derogations or free allocations would reduce the auction receipts available for the budget. There are costs, however, in this dual and segmented approach. In principle, the initial allocation of the EU-wide emis- sion cap between ETS and non-ETS sectors at the EU level and the subsequent split of the non-ETS budget across Member States need not have adverse implications for cost-effectiveness as long as comprehensive emissions trading across all segments of the economy is assured. However, current EU legislation does not foresee such trading. Diverging marginal abatement costs across emission sources are a likely consequence of this segmentation, causing emission abatement in the EU overall to become much more expensive compared to a comprehensive EU-wide cap-and-trade system and caus- 25 A derogation is a provision in EU legislation that permits greater flexibility in the application of the law to take into account special circumstances. In this case, under EU rules, transitional free allocations will be available for any new power plants for which construction began by end-2008, as part of support for modernization of electricity generation. 26 E. Smol (2010), Metodyka wraz z Przykładowym Obliczeniem „Limitu” Krajowej Emisji Gazów Cieplarnianych dla Polski na lata 2013-2020 (Dyrek- tywa EU ETS i Decyzja NON-ETS) [Methodology and calculation of the country’s GHG emissions limit for Poland 2013-2020] , KASHUE-KOBiZE. page 41 ing substantial burden shifting between ETS and non-ETS segments (Box 3). While EU legislation does allow for some flexibility, with limited abatement beyond EU borders through crediting from emission-saving projects undertaken in third countries via Joint Implementation (JI) or the Clean Development Mechanism (CDM), only part of EU member reductions can be covered through extra-EU abatement, in order to comply with principles of additionality and supplementarity.27 Although the rules on access to CDM credits are complex, in summary, non-ETS sectors are allowed to purchase about one-third of their emission reduction requirement from outside the EU, and ETS sectors can offset up to one-fifth of their ETS requirement (Figure 14 and Annex 1). Another potential source of excess costs can be traced back to the use of multiple instruments in EU climate policy. Setting overall emissions targets, binding goals for renewable energy production, and proposals for energy efficiency im- provements is liable to generate excess costs due to overlapping and counterproductive regulation. If a target for energy efficiency becomes binding in addition to those for renewable energy and overall emissions, the outcome will move even further from the cost-effective solution generated by comprehensive emission trading and likely create additional costs. In an economically efficient setting, the relative contribution of renewables and energy efficiency should be determined by markets. 27 These are two important principles of the flexibility mechanisms under the Kyoto Protocol, in particular the CDM: that there is additionality of any emissions-reducing project (to avoid giving credits to projects that would have happened anyway) and that supplementarity holds, i.e., that internal abatement of emissions should take precedent before external participation in flexible mechanisms. CARBON ABATEMENT TARGETS page 42 AND POLICY CHALLENGES FOR POLAND Box 3. Excess costs of emission market segmentation Figure 16 illustrates the pitfall of EU emission market segmentation, based on (estimated) aggregate marginal abate- ment cost curves for the ETS and non-ETS sectors in the year 2020. Total emission abatement in 2020 equals the dif- ference between EU-wide baseline emissions and the targeted emission ceiling (86 percent of the 2005 EU emission level or a 14 percent reduction). Comprehensive emissions trading leads to a uniform EU-wide emission price τ* at the intersection of marginal abatement cost curves C’ETS and C’nonETS. The efficient allocation of abatement burden between ETS and non-ETS sectors (A*ETS and A*nonETS) will be endogenously determined through the uniform emission price τ*. If instead ETS and non-ETS markets are not linked through emissions trading, then the administrative partitioning of abatement requirements between ETS and non-ETS sectors by setting emissions ceilings for each segment must exactly equal the efficient split to achieve cost-effectiveness. However, to do so, the EU planning authority would require per- fect information on the future effective abatement requirement as well as the future marginal abatement cost curves for ETS and non-ETS sectors. If the estimated marginal abatement cost curves in Figure 16 are reasonably accurate, in particular that the slope of the curve for non-ETS sectors is much steeper, then the prescribed EU partitioning (requiring about 60 percent of total abatement from ETS sectors and about 40 percent from non-ETS sectors) is rather inefficient, shown by ĀETS and ĀnonETS. The deadweight loss with differential emission pricing (with the marginal cost of abatement from non-ETS sectors, τnonETS, far above that for ETS sectors, τETS) is shown by the shaded area. Furthermore, a policy with at least twenty-eight CO2 prices (one for ETS, one for the non-ETS sector in each EU Member State) will further boost excess costs. Figure 16. Deadweight loss in emission markets A*nonETS A*ETS ĀnonETS ĀETS S Sour ource ce: Source:: Bö Böh Böhringer, hringe inge hri C , A. g r, C., A Löschel, Lö Lösc sch hel, U. hel Mosl slen lener U Moslener, er, , and and T.F. F. Rutherford T.F Ruth Ruthe herf rfor ford d (2 (2009), ( 009) 00 9) ), EU Climate Cli C li lima te P mate Policy ol oli licy icy Up pt o 20 to 2020: 20: 20 2020 : An Economic E Eco cono nomi mic ic Impact Impact pa Impa Assessment, A ct Ass ss sses essm es sm ent en smen Energy E t, Ene ne nerg rgy rg Economics y Econom on Econ omi om ics 31 ics ic 31, 29 295–305. 295 305 5–30 5. At first glance, it seems Poland is not far from meeting the EU’s 2020 overall targets of 20 percent reduction com- pared to 1990. Because there is no overall national reduction target for Poland, measuring the distance to the 2020 tar- get is hypothetical. It assumes that Poland has to reduce its emissions in line with the 20 percent EU-wide target relative to 1990. In 2007, Poland’s GHG emissions were already 12 percent below the 1990 level (Figure 17). To achieve the 20 page 43 percent 2020 reduction target, Poland would need to reduce its emissions by an additional 35 MtCO2e, which is 9 percent below the 2007 level. However, given that Poland’s impressive achievements during the 1990s were driven by economy- wide restructuring, it is a more complicated question going forward as to whether ongoing efficiency gains and sectoral evolution will outweigh the rising demand for energy generated by economic growth. Figu Fi gure re 1 Figure 17. 177. P Pol Poland’s olan and’ d s hi historical stor hist ical oric al GHG GHG e emi emissions ssio miss ns an ions andd EU EU-wide -wid wide e 2020 t 2020 target arge tar t get 600 500 -20% vs 1988 -12% vs 1990 400 -20% vs 1990 MtCO2e 300 564 454 200 399 363 100 0 1988 Kyoto Base Year 1990 2007 2020 in line with the EU targets S Sour So our Source: urce ce ce: UNFC UNFCCC, : UNFC FCCC CC CC, E Euro European uro , Europe pe pean C Commission, ommi an Com miss iss ssi ion, ion io W World or n, Wor orld ld Ban B Bank an ank k st staff aff af ff ca calculations lc lcul cal ulat lat ati io ions ions Poland’s target for renewable energy appears more challenging. According to the renewable energy sources directive, while the EU overall has committed to raise the share of renewables in final energy demand from 8.7 percent in 2005 to 20 percent in 2020, Poland has to double its share from 7.5 percent in 2006 to 15 percent in 2020. Compared to other EU members, this target does not look overambitious (Figure 18). Yet, progress in recent years in Poland has been relatively slow, with the renewable energy source share in final energy consumption growing from 2.3 percent in 1992 to 6.5 per- cent by 2000, but only to 7.5 percent in 2006. In addition, Poland’s renewables are not diversified. Biomass dominates, while sources like hydro, wind, solar, and geothermal energy have not been developed (Figure 19). Figure Figu re 1 Figure 18. 188. S Sha Share hare of re o rene f re renewable newa wabl ble ener energy e en gy s ergy sou sources ourc rces fin final es in f al inal Figu Fi Figure re 1 gure 19. 199. S Sha Share hare re o of rene renewable f re wabl newa ble e en ener energy gy s ergy sources ourc sou rces es i in gro gross ng ss ross gy consump energy ption consumption inland energy energ consumption, gy consumpt , 2007 p ion, 49 2006 Gap to RES target 2020 Biomass Hydro Wind Solar Geothermal 50 8 45 42 38 7 40 34 1.5 1.1 35 31 30 6 30 5 24 25 25 23 23 0.2 % 25 20 20 4 % 18 18 17 20 16 15 16 14 13 13 14 13 13 15 15 11 10 3 5.6 5.4 4.8 10 2 5 1 0 LT LU AT SE DE HU 0 EE LV PT ES RO FI SI CY BE EL CZ IT IE FR BG MT NL SK DK PL UK EU 27 EU27 EU10 Poland Note: N ote: Not G Gross te: Gros ross s inl i inland land nla e energy nerg d ene rg gy co cons consumption umpt nsum ion is ption e is ene energy nerg rg gy co consumption cons umpt nsum ion by ption by th t the he ene he energy e rg nerggy se sector ct ctor tor it itself, i ts elf tsellf, f, distribution di dist istrib uti tribut ion an tion and d tran t transformation ransf sfor forma mati ti tion on losses, loss loss sses es, an es and final fina d fina nal energy l ener ergy ener consumption c gy con onsu onsumpti mp sump tion by tion by en end users user d us er ers (final s (final in (fin energy e al ene nerg nergy rgy co consumption). ns nsum cons um umpt pt ptio ion) ion) n). Source: So urce Sour ce: European Euro : Eu ropepe Commission, C p an Comommi issi miss ion World ion, Wor W ld Ban orld Bank B ank staff k st aff af ff ca calculations. callcul lc lati ulations. ions Energy efficiency is often seen as the easy place to start in considering GHG mitigation, but exploiting the energy efficiency agenda is not easy. It is often seen as a ‘win-win’ option, with benefits realized relatively quickly and lower upfront costs. Yet much of energy efficiency potential remains untapped because of the many obstacles to investments in energy efficiency: inadequate domestic energy prices and lack of payment discipline, insufficient information on suitable technologies, too few contractors and service companies, and financing constraints. Effective energy efficiency interven- tions combine market-based approaches (which send correct price signals) with regulations (which support changes in practices and behaviors of economic agents). The two components operate coherently only in tandem––regulations will not bring results without adequate energy pricing policy. A transition towards a low-emissions economy may also present opportunities to Poland. As more regions and countries adopt abatement targets, the demand for products and processes with lower greenhouse gas emissions will accelerate. Innovation will be critical in this growing market for clean technology—the expertise and equipment related to new developments in areas such as renewable energy (in particular, wind power, solar power, biomass, hydropower, and biofuels), electric motors and low emission transportation, energy efficient lighting and appliances, and green buildings. The energy sector, the dominant source of today’s emissions, is also the focus of much clean technology—clean energy. Given the well-established fact that the private sector acting alone will tend to underinvest in research and development (R&D), governments who are moving early towards abatement, such as Poland’s, need to consider whether active support to clean technology R&D is an important complementary policy measure. This description of Poland’s mitigation targets and some of the complexities facing policymakers helps set the stage for a more comprehensive assessment of Poland’s possible transition to a low emissions growth path. Three aspects, in particular, stand out and will be the focus of the next sections of this report: • The negotiation of a base year for Kyoto obligations that preceded Poland’s transition to a market economy further eased the already modest targets for 2012. Selection of base years affects the strictness of agreed targets (often de- fined as a percent reduction), but even more critical to understanding how much adjustment will be needed to hit a policy target for GHG mitigation, how is the economy likely to develop in the absence of the climate change target, under ‘business-as-usual’? • Despite slow progress in UNFCCC negotiations, it has been relevant for some time for Polish policymakers to consider in detail, how challenging would more ambitious overall mitigation targets be for Poland? • As the EU 20-20-20 package progresses in implementation, with its complex set of overlapping regulations, what impacts will compliance have on Poland’s economy? c. A SUITE OF MODELS TO ASSESS EMISSIONS ABATEMENT A SUITE OF MODELS page 46 TO ASSESS EMISSIONS ABATEMENT To address the issues set out in the last section, engineering and sectoral analyses are integrated into macroeco- nomic modeling via a suite of models, to allow improved analysis of the feasibility of emissions mitigation, includ- ing its impact on growth, sectoral output and employment. With the objective of assessing the macroeconomic and fiscal implications of greenhouse gas mitigation policies for Poland, a suite of innovative analytic tools were developed to analyze abatement prospects not only from the usual bottom-up engineering perspective but also with economy-wide models that explicitly link to the technological options assessed in the engineering approach. This work builds on existing analyses of low carbon growth undertaken for other countries, in particular, the other six low carbon growth country stud- ies supported by the World Bank (Brazil, China, Mexico, India, Indonesia, and South Africa).28 First, a bottom-up engineer- ing model with an intensive analysis of the power sector helped identify cost-effective abatement measures. Then a large scale, multi-sector dynamic stochastic general equilibrium model translated the engineering model’s technical options into economic impacts. A multi-sector, multi-country computable general equilibrium model which incorporated a hybrid bottom-up and top-down representation of the power sector analyzed the economic impact on Poland of EU climate policy implementation. Lastly, an alternative engineering approach to the transport sector was developed to examine this key sector in more detail. Together, this suite of models yields a series of insights on how Poland might best move towards a lower carbon future. Other low carbon growth country studies have generally depended on detailed bottom-up sectoral work, often supplemented by separate top-down macroeconomic modeling; thus, it seemed that the next methodological step would be full integration of approaches into a single model. Sectoral work can provide country-specific recom- mendations for action at the sector or subsector level while macroeconomic modeling ensures the basic consistency of projected sectoral growth rates, energy demand, and other key variables. The recent World Bank study on Mexico took this approach, with in-depth sectoral and subsectoral studies (of electric power, oil and gas, energy end-use, transport, and agriculture and forestry) and relatively simple macroeconomic modeling focused on energy demand.29 The report by the UK Committee on Climate Change (2008) is a good example of applying these two complementary approaches with more sophisticated macroeconomic modeling.30 When this study began, its objective was to develop a model that inte- grated bottom-up analysis with top-down analysis. However, it became clear that the more comprehensive a model, the more complex it needs to be, and the more likely it will become a ‘black box’ intelligible only to its designers. Given the long horizon of this modeling—10 and 20 years—degrees of uncertainty are amplified. Thus, this study moved towards the more robust approach of developing a suite of models shaped to the policy scenarios and sectoral questions to be assessed and the availability of data. By taking this diversified approach, it is hoped policymakers will focus not just on bottom line conclusions but also keep an eye on the assumptions and structures that generated those results. In other words, modeling should be for insights, not for numbers. Three (and a half) complementary and interlinked models for Poland were developed to quantify the economic im- pact of CO2 mitigation, taking advantage of available data and leveraging existing models. The most familiar of these models is likely the widely-used microeconomic Marginal Abatement Cost (MAC) curve which provides a simple first-order ranking of technical options for GHG mitigation by sector based on the net present value of costs and savings per metric ton of CO2 equivalent avoided. Then, two different economy-wide models were developed, a dynamic stochastic general equilibrium model and a computable general equilibrium model, both of which are standard tools for economic impact assessment.31 The Macroeconomic Mitigation Options (MEMO) model, a DSGE model of Poland revised to include energy and emissions, assesses the macroeconomic impact of the options costed in the MicroMAC curve. It is linked to the Mi- croMAC curve via a Microeconomic Investment Decisions (MIND) module which grouped the technology levers into seven packages, including an optimized package of options for the energy sector. The Regional Options of Carbon Abatement 28 See the documents on the ESMAP (Energy Sector Management Assistance Program) website at http://www.esmap.org/esmap/node/69. 29 Johnson, Todd M., Feng Liu, Claudio Alatorre, and Zayre Romo, eds., (2008), Mexico Low-Carbon Study, World Bank (December). 30 Committee on Climate Change (2008), Building a Low-Carbon Economy - the UK’s Contribution to Tackling Climate Change, The First Report of the Committee on Climate Change, UK, December. 31 Among top-down models, there is often an exaggerated divide between econometric demand-driven Keynesian models and CGE models. Popular but unjustified arguments against the informational value of CGE models include that these models must be calibrated (and thus, lack empirical evidence) and can neither reflect disequilibria (such as unemployment or under-utilization of production capacities) nor transitional dynamics. In turn, econometric Keynesian models are often accused of a lack of micro-foundations. These claims ignore substantial developments during the last two decades to overcome such policy-relevant shortcomings. When it comes to providing a sound and flexible backbone tool for economy-wide climate policy analysis, a strong case can be made for applying CGE or DSGE models, which have become a standard tool for economic impact assessment. page 47 (ROCA) model, a country-level CGE model for energy and GHG mitigation policy assessment adapted to Poland, analyzes implementation of the EU 20-20-20 policy in the context of global policy scenarios, with an emphasis on spillover and feedback effects from international markets.32 The last “half” model is a detailed sectoral approach for road transport, the sector with the fastest growing emissions and central to Poland’s commitments under EU 20-20-20 (as a non-ETS sector). It makes use of the EU transport and environmental model, TREMOVE,33 updated with the latest information and policy intentions, here denoted as the TREMOVE Plus model. All three (and a half) used very similar “business-as-usual” reference scenarios (within the limitations of data) against which to measure policy changes (discussed in detail in the next section). Figure 20 below summarizes the modeling approach, which is described in more detail in the rest of this section. Model Figure 20. M del od l suite f for l low-emissions growth for Poland h assessment f l d Poland •Multi-region CGE (international spillovers, pre- existing market distortions, hybrid power sector) •Peer-reviewed model applied to Poland •Impact of EU 20-20-20 package •8 sectors plus 5 energy subsectors •Poland, rest of EU, other industrialized, developing world MacroMAC ROCA Model curve •TREMOVE model MEMO (road transport) •2 scenarios •Dynamic stochastic GE Model •Passenger, freight (endogenous growth, business cycles) •Poland model + energy and climate redesign MicroMAC TREMOVE •Macro impact of options •11 sectors Curve Plus •Poland; rest of EU Model •Marginal abatement cost (NPV cost per tCO 2 e) •~120 technology options •10 sectors Sou Sou Source: ourc rce: rc W World orld or e: Wor ld Ban B Bank an ank staff. taf k st ff. f. aff 32 The MEMO model allows financing for energy sector investments via the private or public sector and through a choice of policy closures: reduction in social transfers or public consumption, higher taxes such as VAT, or an indirect carbon tax. The ROCA model assumes public consumption remains constant and taxes adjust. The models’ critical exogenous variables include carbon and fossil fuel prices into the future, which have an impact both on the energy mix and on final macroeconomic results. The two macro models reflect Poland as a small, open economy with access to external offsets (both CDM and JI as allowed under EU policy). 33 TREMOVE is an EU transport and environmental policies assessment model. The acronym is derived from earlier EU transport models—approxi- mately, traffic and emissions motor vehicle model. A SUITE OF MODELS page 48 TO ASSESS EMISSIONS ABATEMENT The Microeconomic Marginal Abatement Cost (MicroMAC) curve represents a ranking by net cost of about 120 emission reduction ‘levers’. The costs and abatement potential of more than 200 technical options or ‘levers’ for GHG abatement across the 10 largest sectors of the economy were analyzed, pared down to a set of 120 most relevant for Poland. They include measures such as a shift from coal to nuclear power and energy efficiency standards for new residen- tial buildings. A business-as-usual case was constructed to serve as the baseline for future emissions reductions. Detailed bottom-up estimates for each intervention or lever were constructed. Possibilities and constraints in the power sector were studied with particular care. Levers that would require significant consumer lifestyle changes were not considered. Costs did not take into account transactions costs, taxes, subsidies, feed-in tariffs, and other governmental measures. A risk-free financing rate of 4 percent was used to generate net present values. Third, these levers were ranked according to their costs and presented in a summary graphic, a marginal abatement cost (MAC) curve. This visual presentation provides a wealth of information to policymakers and transforms the high-level objective of emissions abatement into detailed and specific sectoral choices. However, MicroMAC curves need to be read with a degree of caution--despite their apparent simplicity, they are heavily dependent on underlying assumptions, including the business-as-usual scenario, costs and abatement potential of each technology, and appropriate discount rates. The Macroeconomic Mitigation Options (MEMO) model is a large scale DSGE model of Poland that includes energy and emissions. The earlier version of the model, without climate change features, is known in Poland among economists and macroeconomic policymakers, having been applied to issues such as the impact of joining the common currency area.34 This dynamic stochastic model is large scale, with over 2000 variables (compared to the typical DSGE with fewer than 200 variables). The version redesigned for GHG abatement drew data from GUS (Poland’s statistical office), EURO- STAT, and the EU KLEMS databases,35 used 2006 as a base year, and has 11 sectors (agriculture and food; light industry; heavy industry; mining and fuels; energy; construction; commerce; transport; financial services; public services; and other services). It models an open economy, trading goods with the foreign sector (the rest of the EU). Special care was devoted to the real side of the economy. Labor markets feature imperfect competition where both seeking a job and finding em- ployees is costly, wages are negotiated, and unemployment exists. The tax structure is detailed (corporate and personal income taxes, VAT and other taxes such as on property are included). Public expenditures include public consumption, investment, and transfers to households. Production includes a full input-output table for capital, labor, energy and mate- rials. Emissions are generated as a byproduct, based on the amount of energy used and the energy intensity of the sector. (See Annex 3 for more details.) The MEMO model ran simulations to calculate the macroeconomic impact of MAC curve mitigation options. The model was calibrated on the most recent available data for Poland, and only a small number of parameters are exog- enous, based on outside empirical studies. The model is capable of mimicking the cyclical properties of the data with just four shocks. Four financing methods, or model closures, were considered (adjustment of public consumption, of social transfers, of VAT, or of personal income tax). The impact of abatement measures on a large variety of macroeconomic variables by sector, such as output, employment, emissions, household welfare, and fiscal revenues and expenditures, can be estimated by the model, which provides 5-year snapshots through 2030. This structure allows a dynamic assessment of the macroeconomic impact of the options costed in the MicroMAC curve, including a new visual presentation—a mac- roeconomic version of the MicroMAC curve. While this simplified graphic of results helps communicate the main findings of the MEMO model, and while this model remains both highly flexible and heavily detailed, it is a very large and complex model not easily accessible even to the professional economist. A key innovation of the MEMO model is the design of a method to link to the MicroMAC curve, via a Microeco- nomic Investment Decisions (MIND) module. The MIND module transforms the MicroMAC curve levers so they can be analyzed in the MEMO model. Each lever is described by two 20-year time series (2010-2030), reflecting the expected capital and operating expenditures/revenues from the given measure. Those numbers include technological assumptions, e.g., the scope of investments in the GHG abatement technologies and the resulting operating expenditures or savings. While derived from the engineering analysis, the dataset was supplemented and updated in accordance with macroeco- nomic data from EUROSTAT. In particular, for each lever from the MicroMAC package, new estimates were calculated of projected GHG abatement to be achieved if the lever is implemented. The MIND module was applied to find those abate- 34 IBS (2008) “Assessing Effects of Joining Common Currency Area with Large-Scale DSGE model: A Case of Poland”, IBS Working Paper #3/2008, Institute for Structural Research, Warsaw, available at http://ibs.org.pl/publikacja/Effects_of_Joining_Common_Currency. 35 GUS is Główny Urząd Statystyczny, or Central Statistical Office, the national statistics office in Poland. EU KLEMS is the EU database on capital (K), labor (L), energy (E), materials (M) and service inputs (S) productivity. page 49 ment opportunities which are relatively cheap, offer considerable carbon abatement potential and are technically feasible via a multi-criterion optimization. The MIND module creates seven packages of levers to be analyzed, including an optimized package of options for the energy sector. It assigns each lever from the MicroMAC package to one of seven categories: (1) agriculture interven- tions, (2) industry carbon capture and storage (CCS)36 and distribution maintenance, (3) chemical processes, (4) energy efficiency, (5) fuel efficiency, (6) mixed energy/fuel efficiency, and (7) low-carbon energy supply (via energy sector invest- ments). While the first six intervention groups were selected in the sectoral bottom-up analysis, the composition of levers in the last and most important sector (energy) was determined endogenously by the MIND module. The optimization was carried out by, first, computing the NPVs of new power plants of each type. Then, the government subsidy necessary to equalize its NPV with that of a traditional coal plant is calculated, within the constraint of the overall GHG reduction target in the energy sector (which according to the bottom-up sectoral data is about 50 percent relative to the BAU scenario). Finally, the cheapest feasible energy-mix package is determined, taking into account any technological constraints (such as the maximum availability of a given technology), energy production constraints (i.e., the BAU level of energy consump- tion), and the GHG reduction target (desired abatement). This optimal energy package is incorporated into the overall mitigation policy-options package that forms the basis for the MEMO model simulations. The Regional Options of Carbon Abatement (ROCA) model is a country-level CGE model for energy and GHG mitiga- tion policy assessment adapted to Poland. It starts from a static multi-sector, multi-region CGE model framework that has been used repeatedly for analysis of country and regional CO2 mitigation, with open source code, use of a well-known algorithm, and extensive peer reviewing. Aggregating from the GTAP database (version 7),37 the ROCA model of Poland contains 8 sectors (chemicals, aviation, other transport, non-metallic minerals, iron and steel, non-ferrous metals, paper- pulp-print, and other), emphasizing those that are more energy intensive, as well as 5 energy subsectors (coal, crude oil, natural gas, refined oil products, and electricity38). To capture key features of the economy, the model includes important market distortions (taxes, unemployment) and a simple government sector (one good). The energy sector gets innovative treatment, discussed below. As compared with the MEMO model, the ROCA model, since it is focused on implementation of the EU 20-20-20 package, produces results for 2020. As explained in Annex 4 , due to the limited availability of projec- tions of non-CO2 gases, the model tracks CO2 only.39 The ROCA model is designed to analyze implementation of the EU 20-20-20 policy package in the context of global policy scenarios, with an emphasis on spillover and feedback effects from international markets. With this objective in mind, key determinants of economic adjustment to CO2 emission constraints are incorporated, in particular: • A hybrid bottom-up/top-down representation of power sector production possibilities with a detailed activity analysis representation of discrete power generation options while production technologies in other sectors are described in a conventional top-down aggregate manner through continuous functional forms trading off alternative input and output choices. This more complex formulation in the most important emissions sector prevents sudden abandon- ment of existing power generators when prices shift as well as allowing detailed analysis of power sector behavior. • Global coverage of international trade and energy use across 4 countries/ regions (Poland, other EU, other industri- alized, developing countries) to allow analysis of international spillovers and feedback from climate policies in each country, especially those large enough to influence international prices. • The incorporation of initial energy/trade taxes and labor market rigidities to reflect the interaction of climate policy regulation with pre-existing market distortions. With this formulation, no-regrets options for abatement policies are possible, whereby direct or indirect benefits large enough to offset their implementation costs are generated. 36 Carbon capture and storage (or sequestration) encompasses a number of technologies that can be used to capture CO2 from point sources, such as power plants and other industrial facilities; compress it; transport it mainly by pipeline to suitable locations; and inject it into deep subsurface geological formations for indefinite isolation from the atmosphere. 37 The Global Trade Analysis Project (GTAP) database includes information on trade, production, consumption, and intermediate use of commodities and services, as well as GHG emissions and land use. It covers many sectors and all parts of the world. It is housed at Purdue University. 38 Data limitations in the GTAP database prevented a decomposition that included district heating, which is important in Poland and largely fueled by coal. 39 To allow comparisons with the MAC curve and MEMO models, it is assumed that the adjustment in non-CO2 emissions will be proportional, so the percentage change in overall GHG emissions is assumed to be the same as the percentage change in CO2 emissions. This is consistent with the ap- proach used in the MEMO model, where changes in overall GHG emissions are explained through interventions associated with carbon emissions from the combustion of fossil fuels. • The appropriate representation of institutional settings and policy instruments for climate policy implementation, in- cluding the complex rules for the ETS and non-ETS sectors, and revenue recycling possibilities (e.g., lump-sum versus labor subsidies) from carbon pricing. The ROCA model, as a modern CGE model, benefits from strong microeconomic foundations and the incorporation of market imperfections as well as more complex power production technologies. It can analyze complex and overlapping policies such as the EU climate package. However, it provides only an assessment for 2020 and only considers CO2 emis- sions, and it is limited to 2004 as a base year because of data availability. The last piece of modeling in this study is an alternative engineering approach to transport sector mitigation op- tions with the TREMOVE Plus model. The EU transport and environmental model, TREMOVE (v. 2.9-2009), is an EU-wide policy assessment model, designed to study the effects of different transport and environmental policies on the emissions of the transport sector. Calibrated for 31 countries, the model is used to estimate the impact of policies such as road pric- ing, public transport pricing, emission standards, and others. To become TREMOVE Plus, the model was updated with new projections of transport activity and the latest disaggregated data on vehicle stocks from a wide range of sources includ- ing vehicle sales and car import data and from interviews with government officials. While the MicroMAC curve analysis assessed options in road transport, it did not produce separate business-as-usual projections for the sector. Working from a slightly different set of assumptions, the TREMOVE Plus model considers explicitly the characteristics of Poland’s road transport sector and assesses the impact of existing policy commitments and possible mitigation options. This quick overview of the models to be applied to Poland’s potential for low emissions growth should be able to provide a sense of the scope of analysis to come. Each model has trade-offs and simplifications. The validity of the set of assumption that underpin each model will depend to a great extent on what question is posed for analysis. What will be important as simulations and results are reported is to link those findings to the underlying model design and the data that drives it. Model-based analysis can put decision-making on a more informed footing. Yet any model, no matter the complexity, remains a crude approximation of the real world, so numerical results will always need to be interpreted with caution. “Modeling for insights, not for numbers, is the real challenge.”40 40 Loch Alpine technical paper, p. 37. d. BUSINESS- -AS-USUAL SCENARIOS FOR POLAND page 52 BUSINESS-AS-USUAL SCENARIOS FOR POLAND A business-as-usual scenario is fundamental to the calculation of costs of carbon abatement, but they can be gen- erated by differing methodologies using separate datasets. It is difficult to project the path of an economy over a 15 or 25 year period, and it is not surprising that sectoral details will differ significantly across models. The Marginal Abate- ment Cost (MicroMAC) curve model constructs a relatively simply reference scenario, which matches official projections for growth and energy demand and presumes a rising level of efficiency. While more or less matching overall emissions projections, the Macroeconomic Mitigation Options (MEMO) model forecasts in great detail that the Polish economy will shift relatively quickly towards less carbon intensive sectors (mainly services). The Regional Options for Carbon Abatement (ROCA) model generates similar aggregate emissions levels for business-as-usual, but with a very different development path for sectors, with less sectoral transformation. The detailed sectoral approach of the TREMOVE Plus model of road transport provides a scenario of rapid emissions growth for road transport, a major component of non-ETS emissions. The comparison of reference scenarios generated by the suite of models draws attention to the fact that, since each of the models illuminates important aspects of the economics of GHG mitigation, policymakers will need to be ready to consider multiple model results, rather than a single answer. The development path of the economy in the absence of new low emissions policy measures is the correct com- parator for policymakers considering abatement actions. Mitigation targets are almost always defined against a base year. For example, Poland’s non-ETS national target is defined as: emissions in 2020 will be no more than 14 percent above the level in 2005. But such a definition provides little indication of the degree of challenge involved in meeting the target. What matters is the size of the reduction compared to the expected level of emissions in the target year. This expected level is a matter for projections, determined by assumptions about the growth rate of emissions in the absence of additional policy--the business-as-usual (BAU) emission baseline. Central to these assumptions on future emissions are predictions of GDP growth and accompanying energy demands (Figure 21). Faster expected growth translates to faster rising emissions, and the higher is the future emission level in the absence of climate policy, the more stringent are the effective reduction targets and, thus, the costs of abatement. Fi 21 21. Figure 2 Basi Basic 1. B ic d drivers rivers i f GH of GHGG emi emissions issions i th growth Note Note: Note: All : Al l in ggrowth rowt ro wth rate rates, h ra tess, whe w where re ene here energy e rg nerg gy in inte intensity nsit tens ity y is cha c change hang ng ge in toe toe p er € GDP per GDP; G DP; ; emissions emis em issi sion onss in inte intensity tens nsit ity c change y is cha ng hangge in tCO tCO2e peper p r toe; and toe; emissions e and emi mi missss ssio ions ions ela elasticity e last last icit ic stic ity ity is gro growth g ro rowt wt wth tCO2e pe h of tCO per r € GD GDP P The figure . The fig igur figur uree ma make makes kes ke clea clear s cl ear ear th that at cou c countries ou ount ntri ntries ries wit w with it ith high higher h hi gh er pop gher p population op opul ulat ulatio ion atio n growthth or hihigher gher h income i growth will th wilill lfface faster f ace fast rising ter ri i e ising emissions. missi i ions. Th The emi emissions i issions elasticity l ticit elasti ity of f GD GDP P growth will th wil ill need l ne ed t b d to be e reduced duced red dbby y even more to t offoffsett GDP ffset GDP g growth. rowth th. Source: Sour S ce: ource derived d : deeri rive ived from Kay d from Kaya K ayaa an and Yokobori Y d Yok okob kobo bori (1997). ( i (19 1997 19 97) 97 ). page 53 The global financial crisis temporarily reduced GHG emissions but with little long-term impact. In the case of Poland, the effect of the global financial crisis on GHG emissions into the future is negligible. Poland was the only EU country that avoided recession in 2009, although real GDP growth slowed down from 5.1 percent in 2008 to 1.8 percent in 2009. In 2010, Poland’s economy rebounded to a 3.8 percent growth rate and is expected to continue a gradual recovery in the years beyond (Box 4). Box 4. Poland’s growth projections and the global financial crisis The BAU scenarios used in this report are based on pre-crisis growth trends. The global financial crisis triggered the worst economic recession since World War II, and post-crisis growth around the world could be dampened by continued uncertainty and scarce and more expensive capital. This could well lower global energy demand and GHG emissions go- ing forward. For example, global energy consumption is likely to have declined in 2009 for the first time since 1981. In the EU, verified emissions in the ETS are estimated to have declined by 11.6 percent from 2008 to 2009. Finally, among the EU member states from Central and Eastern Europe, economic performance in 2008 mattered for GHG emissions. Estonia and Latvia, which already saw contractions in GDP in 2008 linked to the global financial crisis, experienced re- ductions in GHG emissions (Figure 22). Fi 22 Figure 2 2. G 22. Gro th i wth Growth in GHG nG HG emissi i ions a emissions dG nd and DP i GDP 2008 in 20 in C 08 i Cent ent tral ral la nd dE and tern E East Eastern Europe 6 Growth in GHG emissions, in % 4 2 0 Poland -6 -4 -2 0 2 4 6 8 -2 -4 -6 -8 -10 GDP growth, in % In Poland, the impact of the global financial crisis on growth projections is likely to remain muted. First, Poland’s econ- omy has shown remarkable resilience to the global financial crisis. In 2010, Poland’s economy performed well, after being the only member state of the European Union to avoid recession in 2009. Growth in Poland accelerated from 1.8 percent in 2009 to 3.8 percent in 2010. Second, the BAU scenario incorporates an annual growth rate of 3.5 percent for Poland going forward. This compares to average growth of 5.1 percent from 2003 to 2008, and to growth forecasts of around 4 percent growth over the medium-term from the IMF, OECD and the Government. Overall, while the global financial crisis has slowed growth in 2009 and might have moderated growth prospects in the coming years, it is unlikely to have derailed the longer-term growth prospects which underlie the BAU scenarios. Source: European Environment Agency; European Commission, communication IP/10/576), May 2010; World Bank staff calculations. Because each model used in this report builds from a distinct dataset, there is no singular business-as-usual sce- nario. In the academic literature, BAU projections are often based on simple extrapolation of historical trends or the ap- plication of a single emissions elasticity to expected GDP growth, since rising incomes push up energy demand (usually the dominant driver of emissions) and, in turn, GHG emissions. A steady-state baseline, in which all physical quantities grow at an exogenous uniform rate while relative prices remain unchanged would have the virtue of providing a transparent reference path for the evaluation of policy interference. However, such a path would be unlikely to match official business- as-usual projections, limiting the interest of the model results to policymakers who need more realistic comparisions. page 54 BUSINESS-AS-USUAL SCENARIOS FOR POLAND Instead, each of the models used here produced its own business-as-usual scenario, using data that allowed for the level of detail needed for that model, but following the broad outlines of official growth projections. The MEMO model’s BAU scenario is also closely matched to the simpler formulation of the MicroMAC curve, and the BAUs of the MEMO and ROCA models have been broadly harmonized through 2020, the end-point of the ROCA model. However, they do have points of difference, which help illuminate some of the underlying assumptions of the projections. Lastly, the TREMOVE Plus transport model takes a very different approach to projecting road transport emissions than the MicroMAC curve method, which demonstrates more starkly the importance for policymakers of understanding how models generate their numbers. The MicroMAC curve business-as-usual baseline for emissions through 2030 was constructed from the bottom up. It was calculated based on future production levels for industry and future activity levels in transport and buildings and assuming natural improvements in technological efficiency as new capital replaces old. For example, the BAU baseline for power was calculated by estimating the required level of electricity production and the probable fuel mix in 2030, assuming no efforts were made to reduce emissions and only accounting for greater efficiency of new power plants. In transport, estimates were based on forecasted traffic growth in Poland, both in terms of increasing numbers of passenger cars and average distances travelled. This baseline scenario projects that Poland’s GHG emissions will grow 30 percent above 2005 levels by 2030, to 503 MtCO2e. This translates to an annual growth rate of 1.1 percent, compared with real GDP growth over the same period projected at 3.4 percent per annum (consistent with government projections). As a result, the carbon intensity of the economy continues to decline, driven by the ongoing expansion of the services sector and other (unidentified) efficiency improvements. In a 10-sector disaggregation, emissions from transport and cement are projected to rise fastest, about doubling by 2030. In transport, that growth is fuelled by expected increases in passenger cars per 1000 inhabitants, and in cement, by continuously strong growth in construction (Figure 23). Figure 2 Fi 23 23. 3. M Mi MicroMAC MAC icroMAC curve BA BAU U scenari io e scenario missi i ions gro emissions wth th growth Annual growth, Absolute growth 2005-2030 2030 vs. 2005 Annual emissions % % MtCO2e 503 1,1 30 Power 0,8 22 178 386 146 Buildings 1,0 29 74 Transport 2,8 100 57 72 25 Chemicals 0,9 26 36 23 Iron and Steel 1,6 50 20 14 Industry 15 Petroleum and Gas 1,4 42 10 16 8 31 Cement 2,7 94 33 8 8 Agriculture -0,3 -7 53 63 0,0 -1 Waste Other 1 0,7 20 2005 2030 Note Notes: Notes: I s: Ind Industry, ndus ustr try bui buildldin ings y, buildings,gs, and and transport tr tran spor ansp t se ort sect sectors ctor ors not i s do not include nclu inc lude de ind i indirect irec ndirectt em emis emissions issi sion s fr ons om ele from e electricity lect ctri ricity con city c onsu sump mpti consumptiontion on and w well- and wel l ell- to-tank t t k emi o-tank emissions issions i for for ffuel; l th uel; accounted these are account teddf for or iin the n th the power a and nd d P&G P&G sect sectors respectively; tors respe cti tively; l b buildings ildi uild sector ings sect includes tor i ludes ncl d emissions emis em issi sion ons from s from heaheat. h t. Oth eat. Other O er inc ther includes: i nclu lude s: min des: mining, m ing, inin light g, lig l ight industry, ht ind i ustr ndus y, food try, food foo d & beverage be beve vera rage industry, i ge ind ustr ndus y, glass try, gla g ss production, lass produ pro duct ion, ctio colored n, col c ored olor ed mmetals, met etals, als, off-road off of road f-ro transport, t ad tra nspo rans p rt po rt, an and other othe d ot her sectors. sect r se ctors. ors Source: McKinsey background y technical backg ground papaper. p p r. pe page 55 For the MEMO model, the BAU scenario through 2030 was estimated econometrically, based on continuation of the trends and convergence processes observed in the EU and Poland in the recent past. In forecasting the develop- ment of an economy over 25 years, convergence is a sensible assumption. The MEMO BAU estimation assumes that Po- land will continue to converge towards the economic structure of the average EU country in line with the path experienced by EU members in the recent past. Using EUROSTAT data for 21 EU members, including Poland, during 1996-2006, panel regressions estimated the pace of convergence across 11 sectors for value-added share, energy intensity, and emission intensity. Long-term growth trends for the 21 countries were estimated based on the same data. Then projections of the key variables for the EU26 and Poland through 2030 were generated based on the growth trends for the EU adjusted by the convergence rates for each sector. Once the convergence process is completed, i.e., the country reaches the average EU level, it continues to grow at the average, trend rate. (See Annex 5 for more details on the estimation procedure). The convergence of Poland’s economy towards EU averages in the MEMO BAU scenario builds in moderation of GHG emissions via the ongoing shift towards less emission-intensive sectors such as services and via improved efficiency in each sector. These developments generate a path for GHG emissions that lies below that for energy con- sumption which, in turn, lies below the path for growth of value-added (Figure 24). As the production structure in Poland converges to the European average over the next 20 years, services (especially financial services) are expected to expand their share in value added, while shares of fuels and agriculture are to diminish (Figure 25). At the same time, all sectors with the exception of households (which consume energy and produce emissions) effectively converge to EU levels of energy intensity (defined as energy per unit of GDP) within the next 10 years. Households, in turn, are gradually closing the gap to be just 1.3 times more intensively emitting in 2030 against 2.8 times in 1996. (See Annex 5 for details on energy intensity forecasts.) Lastly, in the majority of sectors, emission intensities (defined as emissions per unit of GDP) in Poland move closer to average EU values by 2030 without completing convergence. Overall emissions intensity improves by more than 40 percent, allowing projected greenhouse gas emissions to expand by just over 40 percent to 2030 while production rises by a factor of 1.5. Figure 2 Fi 24. 24 MEM MEMO 4. M EMO O BA BAU U projecti j tions f projections or Poland for land Pol d Value added Energy consumption GHG emissions 2.5 2.0 Year 2006=1 1.5 1.0 0.5 0.0 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022 2024 2026 2028 2030 So Sour Source: urce urce: ce IBS : IB S technical tech hni ica cal nica p paper. l pap r. pe page 56 BUSINESS-AS-USUAL SCENARIOS FOR POLAND Fi Figure gure Figu 25 re 2 25. 5. M MEM MEMOO BA EMO U va BAU valu value lue e ad added ded adde sec d by s sector, ecto tor 200 2005-2030, r, 2 5-20 005 30, % 2030 100% other services 90% public services 80% finance 70% construction 60% trade 50% transportation 40% light industry 30% agriculture 20% fuels 10% energy 0% heavy industry 2005 2010 2020 2030 S Sour So Source: our urce ce: ce IBS : IB S technical te technica ica hnical l pa pape paper, pe per r, r, World Wor Wor orld ld Bank Ban ank Ban st taf aff k staff ff ca cal calculations. lc lcul lat ati tio ions ions ns. . The MEMO model BAU scenario for Poland generates a U-shaped path for GHG emissions levels. After shrinking by 12 percent from 1990 to 2007, GHG emissions are projected by the MEMO model BAU scenario to recover to about the same level as in 1990 by 2020. With continued economic growth, however, emissions in 2030 are projected to be almost 20 percent higher than in 1990 (Figure 26). Figure 2 Fi 26. 266. G GHG HG e emissions i ions a missi nd dM and MEM MEMO EMO BAU O BA scenario U scenari io 2 202 2020/2030 020/2030 0/20 30 600 +19.7% vs 1990 500 -20% vs 1988 +0.2% vs 1990 -12% vs 1990 400 MtCO2e 300 564 544 454 455 200 399 100 0 1988 Kyoto Base Year 1990 2007 2020 BAU 2030 BAU S Source: Sour So our urce ce: ce: UN UNFC UNFCCC, FC FCCC CC CC, IBS , IB S technical te tech hni ica cal nical pa p paper, pe p r, r, World Wor orld World Bank Ban Ban ank st k staff taf aff ff ca cal calculations. lc lcul lat ati tio ions ions. ns. The ROCA model, with more emphasis on international interactions, derives a business-as-usual scenario in line with external projections from the US Energy Information Administration and the European Commission. For the ROCA model, the BAU scenario through 2020 (the time horizon of this model) is based on projected energy input de- mands across sectors, GDP levels, and the international price for crude oil, drawn from the EIA and complemented by page 57 more detailed forecasts from the EC on Poland and the rest of the EU,41 as provided by the 2007 PRIMES projections.42 The model adjusts sectoral productivities such that all sectors remain on the benchmark isocost line so that cost and ex- penditure functions are kept as close as possible to the initial static technologies and preferences underlying the base-year calibration (see more details in Annex 4). As noted above, mitigation targets often look quite different when measured against the business-as-usual emis- sions levels: Poland’s non-ETS sector target of no more than 14 percent growth compared to 2005 translates to cuts of 22 percent of emissions in 2020. Table 3 summarizes how EU 20-20-20 emission reduction obligations defined against 2005 levels translate into effective emission reduction requirements when compared to the ROCA BAU emissions levels for 2020. For simplicity, the EU-wide ETS reduction target of a 21 percent reduction in 2020 relative to 2005 is as- sumed to apply to each member state individually. The overall EU target of a 10 percent reduction in non-ETS emissions aggregates from Poland’s commitment to increase emissions by no more than 14 percent by 2020 and the rest of the EU’s promise to cut non-ETS emissions by 12.5 percent (see Section b for a description of ETS and non-ETS sectors and commitments). Because of the underlying growth in emissions through 2020, the effective reduction requirements in the non-ETS sectors for Poland become markedly higher. Table 3. Nominal and effective emission reduction targets for 2020 for Poland and the EU, in % Nominal Effective GHG reduction targets GHG reduction targets (relative to 2005) (relative to ROCA BAU) Poland (total) -4.4 -22.7 ETS -21.0 -23.7 Non-ETS +14.0 -22.0 Other EU (total) -16.6 -18.8 ETS -21.0 -24.3 Non-ETS -12.5 -13.5 Source: Loch Alpine technical paper and World Bank staff calculations. The bottom-up activity-based transport model takes a very different, detailed sector approach to constructing a business-as-usual scenario for passenger and freight road transport in Poland. Like the MicroMAC curve, it was constructed based on future activity levels and considered forecasted traffic growth, in particular, expected high growth in passenger cars and distances travelled. However, the TREMOVE Plus transport model built a BAU scenario for road transport in far greater detail than the MicroMAC curve did, allowing for more of the distinctive character of Poland’s road transport sector. Starting from the EU’s 2009 TREMOVE baseline scenario for Poland, data and assumptions were updated, generating a higher path of baseline road transport emissions through 2030, more closely matching Poland’s official GHG inventory of emissions for 2000-07. Importantly, the BAU calculations consider explicitly which transport and environment policies should be included in the reference scenario, and in this aspect, the transport BAU is quite differ- ent from the other BAU scenarios. (See Section j for discussion). Total GHG emissions from road transport are forecast to increase 93 percent from 2005 levels by 2030 (or 210 percent compared with 1990). The modeling revealed that key characteristics of today’s road transport have a significant influence on the path going forward, in particular: the prepon- derance of imported used cars and the advanced age of the passenger fleet; low motorization rates and very low mileage driven per car compared with the EU15; and a highly competitive road freight sector which has already marginalized rail freight and has been shifting to newer and bigger trucks. Although different methods were applied to generate the business-as-usual scenarios for each economy-wide model, the resulting projections are generally similar, while their points of difference are illuminating. The ROCA model’s BAU projections for emissions in 2020 are very close to the projection for Poland underlying the MicroMAC curve model, and the MEMO model’s BAU 2020 forecast is not far (Figure 27). The change, relative to 1990, ranges from 0.2 to 41 EIA (2009); European Commission (2008). 42 The PRIMES Energy System Model of the European Commission analyzes market-related mechanisms influencing energy demand and supply and technology penetration as well as energy policy, including all EU member states. page 58 BUSINESS-AS-USUAL SCENARIOS FOR POLAND 3.6 percent growth in overall emissions. The projections for 2030 are more divergent: the MEMO BAU for 2030 emissions is 9 percentage points higher than the projection made in the MicroMAC curve model. Both models suggest a significant increase in Poland’s GHG emissions by 2030--by 20 percent and 11 percent relative to 1990, respectively. As noted above, the higher is the future emission level in the absence of climate policy, the more stringent is any reduction target defined against a base year, and, thus, the costs of abatement. Figure 2 Fi 27 27. 7. C Compa Comparing i economy-wid ring economy-wide ide BA BAU scenarios U scenari ios 20 2020 2020/2030 20/2 /203 030 0 Change relative to 1990 2030 MAC 503 +10.7% curve 2030 544 +19.7% MEMO 2020 470 +3.5% ROCA 2020 MAC 466 +2.6% curve 2020 455 +0.2% MEMO 0 100 200 300 400 500 600 MtCO2e Note: N Not No te: Th te: ote The ROCA e RO model CA m del od lp produces ro rod duce duce duces emi miss iss ssiions so ions io s CO2 emissions so eq equivalent qui uiva iva val le lent GHG e t GHG mi emissions miss iss ssiions io w ions wer were er ere estimated. e es ti tima ima mat ted ted. ted. Source: So Sour S u ce our ce: : IB IBSS technical tech te nica ica hnical p r, McKinsey paper, l pa p pe McK cKin inse inse sey technical t y te hni ech ica nica l pa cal p paper, p r, Loch pe Loc och Loc h Alpine pine Alpi e technical tec ech hni hn hnicical paper, lp per ape , World Worl Wo ld Bank rld Ba Bankk staff taff c taff sta calculations. al alcu lculati lati la cul o s. t on The question of which sectors will find the shift towards low emissions more difficult is central to policymakers’ concerns. Although the BAU scenarios for 2020 of overall GHG emissions in the MEMO and ROCA models do not differ significantly, the decomposition between ETS and non-ETS sectors suggests some important differences (Figure 28). The MEMO BAU projections indicate a heavier burden for ETS sectors, while according to ROCA BAU projections, the major challenge will be faced by the non-ETS sectors. While the MEMO BAU scenario projects ETS sectors to expand by 20 per- cent relative to 2005 by 2020, the ROCA BAU scenario predicts constant emissions during the period. The MEMO BAU projections seem to indicate Poland will have little problem in fulfilling the country-specific target for the non-ETS sectors under the EU 20-20-20 package (since the projected 15 percent increase under business-as-usual is very close to the 14 percent increase ceiling). In contrast, the ROCA BAU projections warn of a significant challenge for non-ETS sectors, with emissions increasing by 46 percent between 2005 and 2020. The TREMOVE Plus model’s projections for emissions growth from road transport between 2005 and 2020—68 percent—also suggests that non-ETS sectors may pose the greater challenge. page 59 Figure 28. GHG emissions in Poland, in MtCO2e and %, 2005 and 2020 2020~EU targets 14 non-ETS 2020 ROCA BAU non-ETS 46 2020 MEMO BAU 15 2005 -21 2020~EU targets ETS ETS 0 20 2020 ROCA BAU 2020 MEMO BAU -4 Total Total 22 18 0 100 200 300 400 500 -30 -20 -10 0 10 20 30 40 50 60 MtCO2 e % relative to 2005 Note No Note: te te: : Th The MEMO e ME MO ETS ETS and n non-ETS and nonon-E p projections ETS proro roje ject jectio ct io ions ns are are cor corrected c or orre re rect cted ct ed f for or s sma small mall ma ll ene e energy nerg ne rgy rg inst installations y inst stal alla al lati lation ti ons on e s as exp explained xp xpla la lain ined in in ed i n Fi Figu Figure gu gure re 13’ 13’s 1 note note. 3 s note. te Th The ROCA e RO CA model mod del produces p rod l pro duces duce s CO2 emissions emi miss issiions so ions equivalent so eq ui uiva ival le lent GHG e t GHG mi emissions miss issi ions wer ions were w eree es esti estimated. tima ti mat ted. P ted Poland’s ol olan land’ d s EU ETS d’ t target arge ETS tar gett is is ass a assumed ssum umedd to be the be th same e sa me (as (as a p percentage er cent erce ag ntag change) ge ch ange chan ge the E g ) as the EU-wide EU- U wi wide target. t de tar ge g t. arge Source: Sour So ce: urce : IB IBSS technical techni technica cal paper, p l pap r, Loch pe Loc Loch p ne technical h Alpine Al pi Alpi tec techn ical hnic paper, al p ap per er, , World Wo Worl rld d Bank Bank staff staff sta calculations. c ff cal alcu cula tion lations s. BAU projections are central to the costing of economic adjustment. The suite of models generate broadly consistent paths for the Polish economy through 2030, but some points of divergence will have important implications for the costs of transition to a low emissions growth path, discussed in the sections that follow which present the simulations of each of the models. It will be important to keep in mind some of the underlying assumptions driving the business-as-usual scenarios, and, thereby, the results on abatement costs. For example, the MEMO model’s BAU scenario predicts that the emissions intensity of output will fall by more than 40 percent but note that, because of the mechanics of the model, there is no specification of required policies or behavioral shifts. The ROCA model BAU scenario projects less sectoral transfor- mation, and if it turns out to be the more accurate forecast, Poland is more likely to face a sizeable challenge to contain the growth of carbon emissions in non-ETS sectors. The transport BAU confirms the probability of rapid and challenging non-ETS emissions growth, and, in addition, raises the question of how to determine which abatement measures or policy choices should be considered part of business-as-usual and which remain available as additional abatement levers. These contrasts within the model suite should be kept in mind as policymakers consider the simulation results of each of the models. e. THE MICROECONOMIC MARGINAL ABATEMENT COST (MICROMAC) CURVE AND POLAND’S ABATEMENT OPTIONS THE MICROECONOMIC MARGINAL ABATEMENT COST page 62 (MICROMAC) CURVE AND POLAND’S ABATEMENT OPTIONS The Microeconomic Marginal Abatement Cost (MicroMAC) curve is a bottom-up engineering approach to assessing GHG abatement, with an intensive analysis of the power sector. Using detailed sectoral data, the MAC curve model creates a ranking by net cost of about 125 emission reduction levers and presents the measures via a well-known visual summary tool—the MicroMAC curve. The MicroMAC curve summarizes a large amount of policy-relevant information in an easily understandable format. It shows that Poland can significantly reduce emissions, with total abatement potential of a 31 percent reduction from 2005 levels, or 47 percent below the 2030 level in the MicroMAC curve business-as-usual scenario. However, capturing this full potential will require concerted, targeted actions by government, business, and consumers. According to the analysis of the MicroMAC curve, mitigation measures will take some time to deliver lower emissions, and Poland may have trouble meeting its 2020 EU targets. The MicroMAC curve also identifies that the majority of Poland’s abatement potential is associated with the switch to low-carbon energy supply (via energy sector investments) and with energy efficiency improvements. Abatement measures do not have negative net costs after implementation bar- riers are considered, and the overall cost of implementing the MicroMAC curve levers will rise by at least 50 percent. In order to implement all the low-carbon levers, additional investment of about 0.9 percent of annual GDP will be needed during 2011-2030. Finally, the cost of abatement is sensitive to financing costs, fuel prices, and technology. MicroMAC curves summarize a large amount of policy-relevant information in an easily understandable format. As discussed earlier (see Section c), this model analyzes about 125 technical options for Poland for GHG mitigation across the 10 largest sectors of the economy. First, a business-as-usual case was constructed to serve as the baseline for future emis- sions reductions, based on future production levels for industry and future activity levels in transport and buildings and assuming natural improvements in technological efficiency. Then, detailed bottom-up estimates of the costs and potential abatement volume for each intervention were constructed, with particular attention to the power sector. The options or ‘levers’ are ranked by the net present value of costs and savings per metric ton of CO2 equivalent avoided. Finally, the levers were ordered according to their costs in a summary graphic, a marginal abatement cost (MAC) curve. This visual presen- tation provides a wealth of information to policymakers and transforms the high-level objective of emissions abatement into detailed and specific sectoral choices. The curve can be used to compare the size and cost of opportunities, assess the relative importance of sectors, and estimate the overall size of the emissions reduction opportunity. A consistent approach across technical options required some simplification. Options include only available technolo- gies or those expected to be available before 2030 (so, for example, carbon capture and storage is included but biodiesel from algae is not). Levers that cost more than €80 per tCO2e were excluded since these are of less interest and tend to be early-stage and uncertain techniques. Also, levers that would require significant consumer lifestyle changes (such as switching to public transportation or lowering home temperatures) were not considered. Then, abatement cost is calcu- lated as the sum of incremental capital expenditures in net present value terms and incremental operational expenditures or savings in NPV terms. Costs did not take into account transactions costs, taxes, subsidies, feed-in tariffs, and other governmental measures. A risk-free discount rate of 4 percent was used to generate net present values; and all costs are in 2005 real euros. The MicroMAC curve shows that Poland can significantly reduce emissions but capturing the full potential would be a major challenge. The cost curve identifies potential abatement of 236 MtCo2e by 2030 at a unit price of €80 or less (Figure 29). The weighted average cost is about €10 per tCO2e, ranging from minus €130 (net benefits) to almost €80. The width of each column on the curve represents the emissions reduction potential by 2030 compared to the BAU sce- nario. The height of each column represents the average cost of avoiding one tCO2e by 2030 by replacing the underlying (reference) technology with a low-carbon technology. For example, in the power sector, a coal-fired power plant would be replaced by a gas or nuclear power plant. Altogether, the total abatement potential represents a 31 percent reduction from 2005 levels, or 47 percent below the 2030 level in the MicroMAC curve business-as-usual scenario. However, captur- ing this full potential will require concerted, targeted actions by government, business, and consumers. Significant gains will have to be made in the energy efficiency of buildings and transportation, and the share of low-carbon energy sources will have to rise to over 50 percent of total electricity supply by 2030 (from just 2 percent in 2005). If the total abatement potential is achieved, Poland will succeed in decreasing its GDP emissions intensity by almost 70 percent against current levels (see Section a andFigure 3). page 63 Figure 29. Microeconomic Marginal Abatement Cost (MicroMAC) curve for Poland , 2030 Chemicals CCS, retrofit Emission abatement cost Average cost: Off-shore wind EUR/tCO2e ~10 EUR/tCO2e Iron & Steel CCS, new built Biomass co-firing 80 Retrofit building envelope, commercial 70 Biomass dedicated 60 Diesel LDV effectiveness Coal CCS 50 Gasoline LDV effectiveness On-shore wind 40 Biogas 30 Nuclear 20 New built efficiency package, 10 residential 0 -10 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 190 200 210 220 230 -20 -30 Organic soils Iron & Steel CCS, retrofit -40 restoration CCS in downstream -50 Advanced retrofit building envelope – residential -60 Cogeneration -70 -90 Landfill – gas electricity Abatement potential -100 generation MtCO2e in 2030 -110 -120 Recycling new waste -130 -140 Basic retrofit building -150 envelope, residential Not Note: e: Eac Note: Each Eachh co colu column mn is lumn one is on e of the the 123 a abatement bate 123 aba ment teme nt mea m measures sure easu s (o res (onl (only nly y th the most e mo sig st s significant igni nifi ficant one cant ones o ness ar are e na named). med) name The d). The he heig height ight of th ht of e the columns lumns i col is the t s th he cost tiin n € per ab abated bat d tC ted tCO T O2e. Th The he wid width idth th i the amou is th amount te nt emissions missi i ions can beb red duced reduced. S d. Some measures are sh hown shown e b with net benefits en e ef s (n e its (negative ( egative eg costs). e co s s) osts). The scenario e sc s e ario en assumes o ass ssum u es that 6 GW G o of nuclear uclea f nu power powe e r po be er will b installed s alle e inst 2030, ed by 203 0, p 030, providing ovidi ro d ngg about ab out abou 15% t 15 % of electricity. elect ele rici ctri ty. city Source: Source: M McKinsey Kinsey tech cKi technical t hniicall paper. According to the analysis of the MicroMAC curve, mitigation measures will take some time to deliver lower emis- sions, and Poland may have some trouble meeting its 2020 EU targets. By 2020, emissions could be reduced by 20 percent against the BAU forecast, but this overall reduction translates to just 3 percent below 2005. By contrast, Poland needs to reduce overall emissions by more than 4 percent relative to 2005, assuming that emissions from ETS sectors will be reduced in line with EU-wide abatement (see Table 3 and note that the segmentation of sectors under EU rules creates multiple targets). The pace of abatement would pick up significantly only after 2020, when major projects in the power sector became operational, such as large-scale offshore wind generation, nuclear plants, or carbon capture and storage. The majority of the abatement potential is associated with the switch to low-carbon energy supply and energy effi- ciency improvements. The levers can be usefully grouped into four categories: energy efficiency (including transport); low- carbon energy supply (via energy sector investments); carbon capture and storage (CCS) in power and industry; and other measures (in industry, waste management, and agriculture) (see Figure 30). About 70 percent of total abatement potential is related to the first two categories: efficiency improvements and low-carbon energy. Shifting from coal to low-carbon al- ternatives in the power sector such as wind, nuclear, biomass, or biogas, could abate 100 MtCO2e by 2030, or 42 percent of the total, at an average cost of €21 per tCO2e. The coming years will be critical for the future fuel mix in Poland’s power sector because a sizeable share of the current coal plants are set to retire. Their replacements will have lasting impact on GHG emissions, but choosing an optimal fuel mix is a complex undertaking. Thus, the MicroMAC curve analysis considered five scenarios with different costs and abatement, which are discussed further in Section h on energy. At the same time, the re- duction of energy demand through more energy efficient buildings, vehicles, and industrial equipment could bring abate- THE MICROECONOMIC MARGINAL ABATEMENT COST page 64 (MICROMAC) CURVE AND POLAND’S ABATEMENT OPTIONS ment of 68 MtCO2e, or about 30 percent of total, at a negative cost of €14 per tCO2e, according to the MicroMAC curve. The most important opportunities in this category are in the buildings sector, where strict efficiency controls for new building and better insulating existing ones could abate almost 30 MtCO2e by 2030. More fuel-efficient vehicles could generate abatement of about 10 MtCO2e by 2030. Together these efficiency measures could help reduce growth in elec- tricity demand from 1.5 percent per annum to about 0.9 percent.43 Energy efficiency challenges are taken up in Section i. Abatement measures do not have negative net costs after implementation barriers are considered, and the overall cost of implementing the MicroMAC curve levers will rise by at least 50 percent. Energy efficiency measures con- tribute substantially to the aggregate low average cost of the MicroMAC curve levers, but the idea of significant savings opportunities being ignored by households and firms does not make economic sense. Clearly, there must be some other hurdles preventing these savings-creating measures from being taken up at once. Three groups of barriers to these mea- sures are: high upfront investment costs (for example, for an energy-efficient car), principal-agent problems (such as the owner, operator, occupant, and bill payer of a building being separate entities), and lack of information (about what sav- ings are likely). A fourth, and potentially most difficult obstacle, is the costs of implementation across a high number of small entities (for example, with residential lighting). These barriers have not been costed, but if the simple assumption is made that unrecognized costs will inevitably shift negative costs into positive, then, at a minimum, the weighted average cost across the MicroMAC curve of €10 per tCO2e will rise to €15 per tCO2e (if all costs are set to zero or above). Fi 30 30. Figure 3 Mi 0. M MicroMAC icroMA MAC abatement batement poten C curve: ab ti potential ial lf Poland for P l di oland n2 in 203 030 0b fi by groups of nterventions i interventions Share of total GHG emissions abatement Average cost MtCO2e annually % EUR/tCO2e 525 503 500 Business-as-usual 475 1 Energy efficiency 29% -14 450 425 400 Low-carbon energy 386 2 42% 21 375 supply 350 325 CCS in power 3 15% 38 and industry 300 Emissions after reduction 4 Remaining levers 14% -1 275 250 267 225 Total/average 236 MtCO2e 10 EUR/tCO2e 200 2005 2010 2015 2020 2025 2030 N Not ote: te: Ener Note: Energy E nergygy eff e efficiency ffi fficienc iency ici y in i includes ncl lud udes des mea measures m sure easu s in res in b buildings, ui uild ild ldiings ings g , tran t transport ransp sp por ort exce except t ex pt s cept switch wi wit itch to bi tch to biofuels, b iofu iof fuel els ls, a and d a few nd in ind few in i industry ndu dust stry try y( such h as (such as cogeneration). cog ne ge coge nera rati ion) tion ). Source: Sour So ur urce ce ce: McKinsey McKi : Mc Ki nsey ns Kins technical t ey tec echn echn hnic ical ic al pap paper. p ap aper er. er 43 About half of CCS potential is related to equipping coal power plants. The remaining 16 MtCO2e potential lies in industry, particularly in iron and steel and in chemicals. These estimates are based on assumptions about storage potential in Poland and the speed of technical progress. CCS’s large-scale deployment before 2030 remains uncertain, but it could well have a major impact on emissions after 2030. The last category, other measures, generates half its abatement potential from reduction of nitrous oxide and methane in waste management and agriculture. Measures include recycling, methane capture and use, improving agronomic practices, and reflooding peat lands. page 65 In order to implement all the low-carbon levers, additional investment of about 0.9 percent of annual GDP will be needed during 2011-2030. The total amount of investment needed is estimated at €92 billion. The annual amount would grow with time as more expensive abatement opportunities were implemented, offset by growing operational cost savings. Also, investment needs would be unevenly distributed across sectors. The capital costs would be relatively modest in industry, but considerable for power, buildings, or transport. In the latter two sectors, by 2030 they would be fully offset by substantial savings in operational costs. Across all sectors, businesses and households are expected to save about €30 billion on operational costs, or about 0.3 percent of GDP on average per annum. Figu Fi Figure gure re 3 31 31. Mic 1. M MicroMAC icro roMA MAC curv C cu curve: rve: e: i inv investment estm nves tmen ent and t an d op oper operational erat atio nal iona l co cost st s sav savings, ings avin 2010 2010-2030 gs, 20 2030 10-2030 Required additional investment Annual average in each 5-year period, EUR billion Period 2011-2015 2016-2020 2021-2025 2026-2030 6.9 5.4 3.6 2.5 Share 0.6 0.7 0.9 1.1 of GDP, % Operational cost savings Annual average in each 5-year period, EUR bn -0.4 -1.1 -1.6 -2.9 Share 0.1 0.2 0.3 0.4 of GDP, % S Sour So our Source: urce ce ce: : Mc M McKinsey cKi Kins McKi ns nsey ey y tec technical tech echn hni hnic ical lp ap per er. . paper. The cost of abatement is sensitive to financing costs, fuel prices, and technology. Higher energy prices reduce the net costs of energy efficiency measures. If the price of oil were 50 percent higher than the assumption of $62 per barrel in 2030, the average overall cost of abatement would fall from €10 to €4 per tCO2e. If technologies turned out to be more expensive in terms of capital investment, then abatement costs would rise in proportion. For example if the total capital to install 1 GW of nuclear capacity rises by €500 million, the abatement cost of the nuclear lever would rise by about €4 per tCO2e. Perhaps most importantly, an interest rate higher than the risk-free rate of 4 percent used so far would directly increase the cost of capital-intensive technologies with relatively short lifetimes such as wind turbines and hybrid vehicles. Applying an interest rate of 8 percent, the overall average abatement cost would rise from €10 to €19 per tCO2e. MicroMAC curves communicate complex engineering information in a simple and accessible way; however, they need to be read with a degree of caution. Despite their apparent simplicity, they are heavily dependent on underlying assumptions, including the business-as-usual scenario, costs and abatement potential of each technology, and appropri- ate discount rates. For some levers, especially emerging technologies, uncertainty about volume and estimated costs can be significant. The adoption rate of new technologies depends strongly on energy prices as well as on cost and perfor- mance improvements, neither of which can be predicted with precision. The ease of capture or of implementation has been considered in a simple way, creating an additional dimension to the net present value of each measure that alerts policymakers to the importance of sequencing. Further, given the substantial impact the shift to a low emissions scenario will have on an economy, policymakers rightly have concerns about economic impact beyond discounted investment and operating costs. The next chapter presents a new methodology that allows just such an expanded assessment. f. THE MACROECONOMIC MITIGATION OPTIONS (MEMO) MODEL AND THE MACROECONOMIC IMPACT OF THE ABATEMENT PACKAGE THE MACROECONOMIC MITIGATION OPTIONS (MEMO) page 68 MODEL AND THE MACROECONOMIC IMPACT OF THE ABATEMENT PACKAGE The Macroeconomic Mitigation Options (MEMO) model, a large scale DSGE model of Poland, can provide a dynamic assessment of the macroeconomic impact of GHG options, including a new visual presentation—a macroeconomic version of the MicroMAC curve. The innovative linking of this economy-wide model to the bottom-up engineering ap- proach of MicroMAC curve model allows analysis of the varying macroeconomic and fiscal implications of GHG abatement measures, across four public financing options. For the comprehensive abatement package, the MEMO model simulations finds that GHG emissions will be reduced by 24 percent by 2020 and by 47 percent by 2030, with an economic impact that is generally negative but appears affordable. Not surprisingly, the fall in GDP is driven by recession in emission-inten- sive sectors, which bear the heaviest burden of abatement. At a more disaggregated level, the model finds that it is the switch to low-carbon energy and fuel efficiency measures that provide the bulk of abatement and that the technologies with the largest abatement potential do not necessarily impose the biggest macroeconomic cost. Finally, the MicroMAC curve can be transposed into a Macroeconomic Abatement Cost and Macroeconomic Marginal Abatement Cost curves to examine in detail the impact on growth associated with the implementation of specific abatement measures. The MEMO model is a DSGE model of Poland redesigned to address climate and energy issues. This very large dy- namic stochastic model has a detailed treatment of the real side of the economy and was calibrated on the most recent available data for Poland. It has 2 ‘countries’ (Poland and the rest of the EU) to allow for trade, and 11 sectors with a full input-output table and emissions generated as a byproduct. Both public revenues and expenditures are disaggregated, and different financing methods (or model closures) were applied. Imperfect competition in labor markets allows for un- employment, and exogenous shocks generate cycles. The model provides 5-year snapshots through 2030 of the impact of abatement measures on a large variety of macroeconomic variables by sector, such as output, employment, emissions, household welfare, and fiscal revenues and expenditures.44 (See Sections c and d and Annex 3 for more details.) The MEMO model’s BAU scenario through 2030 was estimated econometrically, based on continuation of the trends and convergence processes observed in the EU and Poland in the recent past, as discussed in more detail in Section d. This approach builds in moderation of GHG emissions via the ongoing shift towards less emission-intensive sectors such as services and via improved efficiency in each sector. As a result, GHG emissions levels follow a U-shaped path, recovering to 1990 levels by 2020, then rising to 20 percent above 1990 levels by 2030. The innovative linking of this economy-wide model to the MicroMAC curve engineering model is achieved by a Microeconomic Investment Decisions (MIND) module. The MIND module transforms the MicroMAC curve levers, in- corporating additional data on the projected capital and operating expenditures, the potential efficiency and emissions gains, and required government subsidies to cover the additional costs of most options compared to the business-as-usual technology. The levers were combined into seven technological clusters or ‘micro-packages’ based on economic similari- ties to simplify analysis. While for the first six categories, potential emissions mitigation for individual levers follows the engineering analysis technological assumptions, the composition of levers in the last and most important sector (energy) was determined endogenously by the MIND module with the constraint that overall abatement by 2030 would reach 47 percent compared to the BAU level, matching the MicroMAC curve overall estimate. The MIND module was applied to find those abatement opportunities which are relatively cheap, offer considerable carbon abatement potential and are techni- cally feasible via a multi-criterion optimization. (See Section h and Annex 3 for more details.) The micro-packages are: • chemical processes, such as catalyst optimization; • industry CCS and distribution maintenance; • agriculture interventions, such as grassland management; • energy efficiency, such as insulation for new residential buildings; • fuel efficiency, such as hybrid passenger vehicles and other transport measures; • mixed energy/fuel efficiency, such as retrofitting heating and air conditioning in commercial buildings; and, • low-carbon energy supply investments, such as gas-powered generation plants and small hydropower facilities. 44 Household welfare is measured as discounted future consumption of goods and leisure. page 69 Box 5. How do the bottom-up abatement opportunities work in the top-down model? Low-carbon energy supply investments require a fuel switch towards lower carbon technologies, such as wind power. Such a measure means investing in an option with a lower NPV compared to a traditional coal-fired power plant, which is the business-as-usual or reference technology in the energy sector. In order to be implemented, that difference in value must be absorbed by someone. It is assumed in the MEMO model, for simplicity, that it is the public sector that covers the additional costs. During the two to five-year construction phase for the new generation plant, investment spending rises in the energy sector. Both the interest rate (the price of capital) and the prices of investment goods are pushed up, crowding out capital accumulation in other sectors. Domestic energy prices rise because of higher priced and expanded private investment in the energy sector, and this more costly energy is detrimental to overall growth. If the public costs are covered by higher taxes, an additional tax distortion is added to the economy. When construction is completed and operations begin, these effects are unwound: energy production becomes cheaper due to reduced fuel costs (since free wind costs less than coal), which leads to lower energy prices. The relative price of investment goods declines, to the benefit of the other sectors which increase their capital accumulation. An energy efficiency measure such as switching to efficient commercial lighting will induce broadly similar effects on investment at the beginning of implementation, balanced by later savings on operational costs resulting from lower en- ergy consumption. As noted above, it will be light rather than heavy industry that benefits early on. If the benefits during operation outweigh the initial costs, growth will be enhanced, together with changes in the structure of firms’ inter- mediate consumption. The sectors that have implemented the efficiency measure then enjoy lower costs of production. Starting from the basic assessment of net costs and investment demands in the MicroMAC curve analysis, the MEMO model determined the macroeconomic and fiscal implications of Poland’s technological options for GHG mitigation. Each of 119 individual mitigation levers identified in the bottom-up analysis was incorporated in the model.45 While all levers reduce emissions either by reducing energy intensity (energy used per unit of output) or emissions intensity (emissions per unit of energy), they vary in sectoral and fiscal impact. For example, an energy sector investment to move towards low-carbon supply such as construction of a wind power facility generates higher demand in the early years for the output of heavy industry (for the necessary capital goods), whereas an energy efficiency measure such as switching to efficient commercial lighting raises demand for light industry goods. Since most of the abatement measures require gov- ernment support, they will have a direct impact on fiscal balances, as well as indirect effects. The government then needs to adjust other government spending or taxes (since the model assumes Ricardian equivalence46). For example, an increase in public subsidies to the energy sector might be used to spur nuclear plant construction and financed by an increase in the value-added tax. (See Box 5 for more details on how the levers were integrated and Box 6 for more on public financing options. An annotated list of all levers can be found in Annex 7.) The MEMO model reports the impact of each lever on output, emissions, employment, household wealth, and fiscal revenues and expenditures. 45 The remaining 4 levers included in the MAC curve were not significant and, lacking sufficient information, were dropped from the MEMO model analysis. 46 Ricardian equivalence suggests that consumers internalize the government’s budget constraint so that it does not matter whether a government finances its spending with debt or a tax increase since the effect on demand will be the same. THE MACROECONOMIC MITIGATION OPTIONS (MEMO) page 70 MODEL AND THE MACROECONOMIC IMPACT OF THE ABATEMENT PACKAGE Box 6. Public financing options or ‘closures’ in the MEMO model The MEMO model assumes that government must, in one way or another, pick up the tab for the excess costs of most of the mitigation measures (that is, for the cost above the reference technology, or equivalently, the mitigation unit costs displayed in the MicroMAC curve). Every lever has an impact on fiscal revenues and expenditures—directly for the majority that require a public subsidy or a carbon tax, and also indirectly as a result of changes in the behaviors of economic agents due to the implementation of the mitigation measure. The model assumes fiscal neutrality; therefore, changes resulting from the levers must be “closed” adequately in the budget accounts to assure that public spending is equal to public revenues and the deficit level set by the government. Given Ricardian equivalence in the model, financing the deficit via debt is equivalent to financing by spending cuts or tax increases. However, the choice of which taxes to increase or which expenditures to cut does affect the model results. Thus, the model considers four closures: • change in public consumption, • change in social transfers, • change in value-added taxes (VAT), and • changes in personal income taxes (PIT). As noted in the text, the simulations closing the model with adjustments to VAT are selected as the central model results. Implementation of the comprehensive abatement package generates a low carbon scenario in which GHG emis- sions are reduced by 24 percent by 2020 and by 47 percent by 2030. (See Figure 32). In the low carbon scenario simulations of the MEMO model, Poland’s GHG emissions in 2020 would fall from about 455 MtCO2e to 346 MtCO2e. In 2030, GHG emissions would stand at about 288 MtCO2e, as compared with about 544 MtCO2e in the BAU scenario. Because the BAU projection for 2020 is very close to actual emissions in 1990, the 24 percent reduction relative to 1990 is about the same as that relative to 2020. The 47 percent reduction relative to the BAU projection for 2030 corresponds to 37 percent abatement relative to 1990. Note that under this scenario, Poland would likely meet the overall EU 20-20-20 target of a 20 percent reduction in total emissions by 2020 as compared to 1990 (or emissions of 363 MtCO2e).47 Figure 3 Fi 32. 322. H Hi Historical istorical i l emiissions i emissions in i nP Poland l d and oland under der ME d und MEMO MO l low carb bon scena carbo rio i scenario 600 +19.7% 500 -20% vs 1988 +0.2% -12% -24% 400 -24% vs BAU -37% MtCO2e -47% vs BAU 300 564 544 454 455 200 399 346 288 100 0 1988 Kyoto Base Year 1990 2007 2020 BAU 2020 Low Carbon 2030 BAU 2030 Low Carbon Note Note: Note: not oth : If not o otherwise ther erwi se ind wise indicated, i ndic icat ated ed, th e pe the percentage p rcen rcenta tage g cha ge changes c ng hangges abo a above ve the bove b bars ars the bar s in indi indicate dica te a cha cate change changng ge re relative lati rela ve to tive to 19 1990 1990. 90. Th The data e da for ta for 2007 20 07 is is the the latest latest late st available ava avail able ilab year. year. le yea r. The low carbon The low carbo car bon scenario n sc enar scen io is ario implementation impl is im plem enta emen tion tati of al on of all l 12 122 abatement abat 2 ab emen atem ent options opti t op ons tions fr from om the Micro- M icro the Mic ro- MAC MA C curve. curv cu e. rve Source: Sour urce Sour ce ce: UNFCCC, : UN FC FCCC UNFC CC CC, IB IBS S technical tech te ch chni nica ni ca cal paper, pape l pa per pe r, McKinsey McK McK cKin in sey se inse y technical tech te chni chnical ca nica paper, pape l pa per per, World Wor orld Wor ld Bank Ban ank Ban k staff staf aff staf calculations. calc f calc lcul ul ulat atio atio ns ns. ions 47 Unsurprisingly, these results are very close to those of the MAC curve model which projects emissions after abatement for 2020 at 373 and for 2030 at 267. Since total abatement in the MEMO model is constrained to match the 47 percent 2030 achievement of the MAC curve package, reductions in quantity terms should also be similar. page 71 The economic impact of implementation of the full abatement package is generally negative but appears afford- able. The effects on emissions, GDP, value added, employment, household welfare, and public expenditures and revenues under four different MEMO model closures (reduction in public consumption, reduction in social transfers, increase in VAT, and increase in PIT) are presented in Table 4. The impact of the entire package on GDP and value-added, relative to BAU, is consistently negative over the twenty-year period of the model, but the losses approach zero by 2030 (depending on the financing assumption). The losses in real GDP range from 1.5 to 2.2 percent for 2015, peaking at 1.8 to 3.1 per- cent for 2020, moderating to 0.3 to 2.4 percent for 2025, and settling at -0.7 (gain) to 0.7 percent for 2030. At the peak point for costs to GDP, around 2020, output losses are approximately comparable to one year’s potential real GDP growth. (Note that the apparent positive influence on GDP under VAT financing is a GDP accounting artifact because the increase in VAT inflates GDP. Value-added, the better measure of impact under this closure, is reduced over the entire period. The VAT closure is the only variant in which the behavior of value-added and GDP diverge significantly.) The exception to the pattern of GDP losses is for simulations using adjustments in social transfers for financing, under which the impact on GDP turns positive by 2030. However, in this variant, the drop in welfare is the most substantial, since households react to shrinking transfers by giving up leisure, which diminishes their well-being. THE MACROECONOMIC MITIGATION OPTIONS (MEMO) page 72 MODEL AND THE MACROECONOMIC IMPACT OF THE ABATEMENT PACKAGE Table 4. Macroeconomic and fiscal impact of GHG abatement package, deviation from BAU, in % Closure Variable 2015 2020 2025 2030 GHG emissions -10.33 -24.01 -39.31 -47.34 in public consumption GDP -2.13 -3.08 -2.42 -0.66 Reduction Value Added -2.20 -3.19 -2.53 -0.74 Employment -2.73 -2.09 -2.76 -2.33 Household welfare -1.03 -1.64 0.01 0.52 Government expenditures -2.64 -3.05 -2.15 -1.06 Government revenues -2.20 -3.15 -2.76 -1.13 Closure Variable 2015 2020 2025 2030 GHG emissions -10.38 -23.86 -39.03 -47.01 in social transfers GDP -1.52 -1.89 -0.28 0.68 Reduction Value Added -1.51 -1.93 -0.35 0.69 Employment -0.67 3.16 6.34 3.34 Household welfare -2.88 -4.49 -3.27 -1.63 Government expenditures -2.86 -1.83 0.94 0.76 Government revenues -1.55 -1.86 -0.49 0.30 Closure Variable 2015 2020 2025 2030 GHG emissions -10.56 -24.14 -39.48 -47.50 Increase in VAT GDP -1.53 -1.79 -0.83 0.16 Value Added -2.88 -3.42 -2.81 -1.63 Employment -2.59 -0.52 -0.20 -0.85 Household welfare -1.85 -2.88 -1.54 -0.66 Government expenditures 1.75 3.23 6.06 5.58 Government revenues 2.19 2.59 4.30 4.77 Closure Variable 2015 2020 2025 2030 GHG emissions -11.16 -24.63 -40.15 -47.92 GDP -2.18 -2.37 -2.07 -0.63 Increase in PIT Value Added -2.17 -2.41 -2.03 -0.56 Employment -6.13 -4.84 -7.35 -6.79 Household welfare -1.82 -2.49 -0.88 -0.09 Government expenditures 1.38 2.90 5.44 5.16 Government revenues 1.74 2.41 3.93 4.38 Note: Household welfare is defined as the sum of discounted consumption flows of goods and leisure. Source: IBS technical paper, MEMO model simulations. Not only is GDP negatively affected, but employment is also reduced by the abatement measures, and fiscal im- pacts mimic the public financing choice. The employment loss, expressed as the deviation from BAU in 2015-2030, ranges from 2.8 to 2.1 percent under the reduced public consumption closure, 2.6 to 0.2 percent if VAT is increased, and 7.4 to 4.8 percent if personal income taxes are raised. In contrast, if social transfers were reduced in order to subsidize the transition to low emissions economy in Poland, the employment level would fall below the BAU scenario by 0.7 percent in 2015 but recover to levels above BAU by 2030 (by 3.3 percent). As noted above for the impact on GDP, restructuring of public spending away from less productive categories (such as social transfers) is supportive of job creation (because page 73 households react to shrinking transfers by giving up leisure). At the same time, the fiscal implications of the introduction of low carbon interventions are directly associated with the financing closure. If reductions in public consumption or social transfers are selected, then government expenditures are adjusted to the falling GDP level and the subsequent decline in public revenues. On the other hand, if the government covers its costs with VAT and PIT tax hikes, tax revenues rise. Since the VAT financing option generates results that lie about on the average of the four model closures, for the remainder of this section, only simulations using VAT financing will be discussed. Table 5. Decomposition of the macroeconomic impact of GHG abatement package, deviation from BAU, in % Variable 2015 2020 2025 2030 Value Added in ETS -4.06 -5.76 -7.51 -9.06 Value Added in non-ETS -2.72 -3.10 -2.16 -0.60 Employment in ETS -3.38 -1.19 -2.73 -5.15 Employment in non-ETS -2.29 -0.52 0.17 0.39 GHG emission in ETS -11.60 -27.38 -44.20 -52.46 GHG emission in non-ETS -8.7 -18.6 -30.6 -37.5 Emission intensity of Value Added -7.91 -21.45 -37.73 -46.63 Energy intensity of Value Added -0.53 -7.86 -11.02 -11.57 Note: Energy, heavy industry, and fuels are ETS sectors (see Table 2 note on approximation of ETS and non-ETS sectors based on national accounts data). The model closure for this simulation is an increase in VAT. Source: IBS technical paper, MEMO model simulations. The fall in GDP is driven by recession in emission-intensive sectors, which bear the heaviest burden of the entire abatement cost. Value-added in ETS sectors is projected to shrink by more than 9 percent by 2030, with employment falling by more than 5 percent. The shift away from ETS sectors reduces the overall energy intensity of value-added by about 12 percent by 2030 and emission intensity by almost half (see Table 5). Table 6. Decomposition of abatement by micro-package, reduction relative to BAU, in % Closure Group of abatement levers 2015 2020 2025 2030 agriculture interventions 0.77 1.27 1.69 1.89 industry CCS and distribution mainte- 0.01 0.41 3.65 3.79 Increase in VAT nance chemical processes 0.45 0.71 0.90 1.00 energy efficiency 2.20 3.49 4.32 4.87 fuel efficiency 3.39 7.14 11.90 14.84 mixed energy/fuel efficiency 0.27 0.64 0.98 1.17 low-carbon energy supply 3.46 10.48 16.03 19.95 Total impact on emissions 10.56 24.14 39.48 47.50 Note: Model closure is increase in VAT. Source: IBS technical paper, MEMO model simulations. Analyzed at the level of micro-packages, the switch to low-carbon energy and fuel efficiency measures provide the bulk of GHG abatement. Energy efficiency measures are most important in the early years, contributing 20 percent of mitigation in 2015. From 2020 onward, about 40 percent of potential mitigation derives from low-carbon energy supply measures (see Table 6 and Figure 33). The remaining interventions are dominated by fuel efficiency measures that concen- trate mostly in the transport and waste management sectors (contributing about one-third of abatement). THE MACROECONOMIC MITIGATION OPTIONS (MEMO) page 74 MODEL AND THE MACROECONOMIC IMPACT OF THE ABATEMENT PACKAGE Figu Fi gure re 3 Figure 33 33. 3. D Dec Decomposition ompo ecom siti posi tion on o of abat abatement f ab emen atementt by m mic micro-package icro packa ro-pac ge kage 600 agriculture 544 interventions 550 industry CCS 500 chemical processes 454 450 energy efficiency MtCO2e 400 fuel efficiency 350 mixed energy/fuel efficiency 300 low-carbon energy 288 supply 250 BAU 200 Low Carbon Path 1990 1995 2000 2005 2010 2015 2020 2025 2030 Note: N otte: M Model oddel l l closure is i s increase i in i V n VAT VAT. AT. Categories Cat tego g ries i are mi micro-packages icro-p pack kageg s( mit iti (mitigation igati tion op pti tions gr options g grouped oupe by p db economic y econo i charac- mic h teristics). teri te ist sti tic ics)) Ener s). Energy Energy sector s gy sec tor ect investments tor inve i stme nvesttment nts ts ar e low are low-carbon l ow-ca rbon carbbon eneenergy e nerg rgy supply supp y supplly mea ly measures m easu sure res such s su ch ha s ga as gas-fired gas fire fi s-fired power powe d po r pl wer plants, plan lantts on ts, onshore sh hore shor wind ind e wi d power ge generation, g i neration, and d IG IGCC GCC coal l l plants. Fuel Fuel efficiency l eff ff ffi iciiency y measures are mosmostly ly i tly in he transport sector. Energy the n th E gy effi Energ efficiency i fficiencyy mea- sures sure su res are s ar mostly e mo most ly in stly buildings. buil in bu ding ildi ngs s. Mix Mixed M ixed energy/fuel e ed enenergrgy/ y/fu el eff fuel efficiency e icie fficienc ncy measures meas y me ures asur building b es bui uild ldin ing measures meas g me ures asur that tha es that also t al have have so have fu fuel el imp impact. i act mpac t. Source: S ource: IB IBSS technical tech i l paper, World hnical W ld Bank B k stafftaff st calculations. ff callcul tions. lati The technological micro-packages with the largest abatement potential do not necessarily impose the biggest mac- roeconomic cost. Is there a trade-off between growth and abatement that arises from the MEMO model simulations, at least in the projection window up to 2030? The switch to low-emissions energy supply provides about 40 percent of abatement through 2030 and also imposes the biggest negative impact on GDP through 2030 (of about one percent each year). Fuel efficiency measures, on the other hand, while contributing 30 percent of overall abatement, begin to enhance GDP significantly by 2025 and provide a net boost to growth overall. Energy efficiency measures are most important in the early years, contributing 20 percent of mitigation in 2015, while costing over one percent in GDP losses in 2015 but switching to mildly growth-enhancing by 2025. In contrast, industry CCS (carbon capture and storage) contributes only marginally to emissions abatement while costing about one-half percent per year in lower GDP after 2020, leaving it the second most expensive micro-package in terms of growth. While GHG abatement is projected to rise steadily over time as more levers become operational, the impact on macroeconomic performance demonstrates a U-shaped pattern. Losses to real GDP are highest in 2020 and then gradually reverse to a modest positive impact by 2030. (See Table 7 and Figure 34). Table 7. Decomposition of GDP impact by micro-package, deviation of real GDP from BAU, in % Closure Group of abatement levers 2015 2020 2025 2030 agriculture interventions 0.00 0.00 -0.01 -0.01 industry CCS and distribution maintenance 0.00 -0.08 -0.79 -0.34 Increase in VAT chemical processes -0.04 -0.05 -0.05 -0.06 energy efficiency -1.28 -0.57 0.11 0.18 fuel efficiency -0.06 0.00 0.58 0.97 mixed energy/fuel efficiency -0.02 0.03 0.12 0.15 low-carbon energy supply -0.13 -1.13 -0.79 -0.73 Total impact on GDP -1.53 -1.79 -0.83 0.16 Note: Model closure is increase in VAT. Source: IBS technical paper, MEMO model simulations. page 75 Fi Figu Figure gure re 3 34 34. Dec 4. D Decomposition ecom ompo posi tion siti of on o f GD GDP impa impact P impact ct o of low f lo w ca carb carbon rbon on p package acka pac kage ge b by micr micro-pa y mi cro o-pa pacckage kage ckage 1.5 mixed energy/fuel efficiency GDP deviation from BAU, in % 1.0 fuel efficiency 0.5 energy efficiency 0.2 0.0 chemical processes -0.5 industry CCS and -0.8 distribution maintainance -1.0 agriculture interventions -1.5 -1.5 low-carbon energy supply -1.8 -2.0 GDP deviation from BAU 2015 2020 2025 2030 Note: N ote: Not M te: Mod Model odel del cl clos closure losur ure is inc e is i increase ncre reas ase in VAT e in VAT. V AT. Change AT Ch Chan angege i n re in al real GDP is l GDP is me meas measured ured asur a against d agagai in inst tbbusiness-as-usual bususiiness- ines as-u s-as -usu sual sce scenario. ls nari cena io. C rio Categories at ateg ori tegor ies ar ies e are micro-packages micr mi cro- o pa pack ckag es ( ages (mitigation (mi miti gati tiga on o tion options opt ptio ions ns ggrouped roup gro ed b uped byy economic ec econ omic onom ic characteristics). cha chararact cter isti eris cs) tics ). Energy Ene Energrgyy sector se ctor sect or investments inv i nves estm tmen ents are ts are llow-carbon ow-c low car arbo bon energy ener n en gy ergy supply supp su ly measures pply measu mea res sure s su such ch as gas-fired gas- as ga s fi fire red power powe d po wer plants, plan r pl ts, ants onshore , on shor onsh ore e wi wind power p nd pow ower generation, gener er gen erat atio ion, and IGC n, and IGCC I GCC coal C co plants. pla al p lant s. Fue nts. Fuel Fuell efficiency ef effi cien fici cy ency measures me meas ures asur es are mostly are mos m tly ostl the tra y in the transport t rans nspop rt sec po sector. s tor ecto r. Energy Ene Energ gy ef rg efficiency effi cien fici cy ency measures m y mea easu sure res are s ar mostly most e mo ly stlyyiin n bu buildings. ildi buil ng gs. Mix ding Mixed M ed ene ixed energy/fuel e rg nerggy/ y fu fuel el eff efficien- e icie ffic n- ien- y measures b cy building uil ildi dingg measures th that t al at also lso hhave f li ave fuel impact. p ct. mpa t Source: IBS technical paper, World Bank staff calculations. The MicroMAC curve can be transposed into a macroeconomic abatement cost curve (or MacroAC curve) which shows the economic effects associated with the implementation of specific abatement measures. Like the Micro- MAC curve, the appeal of the MacroAC curve is the visual presentation (see Figure 35).48 The horizontal axis depicts CO2e abatement relative to BAU levels in percent,49 while the vertical axis shows real output deviation from BAU levels in per- cent. Growth enhancing measures are located above the X axis, while those which hamper growth are below the X axis. The colors of the bars indicate to which micro-package each lever belongs. It seems logical that a policymaker would be interested in these economic impacts of low carbon measures, as well as the net present value available from MicroMAC curve analysis.50 (Note that, unlike the MicroMAC curve, the area under the MacroAC curve (the size of rectangles) has no economic interpretation.) This disaggregation by individual measure allows the diversity within micro-packages to become evident, and both important similarities and contrasts to the MicroMAC curve appear. Table 7 above presented results by micro-package, showing that agriculture and chemical processes interventions have rather negligible macroeconomic impact, while indus- try CCS and distribution maintenance translate into negative effects on real GDP, second only to the cost of low-carbon energy supply. In contrast, the MacroAC curve illustrates that many individual energy efficiency measures are moderately supportive of economic growth, such as energy efficiency improvements in residential and commercial buildings (lighting, HVAC); also growth-enhancing are small hydro power plants, direct use of landfill gas, recycling and composting new waste, and on-shore wind. The messages of the MicroMAC and MacroAC curves are consistent on several levers: some of the most expensive measures in financial terms, such as CCS, also have a negative impact on real GDP. On others, the MacroAC analysis reorders the MicroMAC curve ranking: for example, on-shore wind power and small hydro power plants are superior to many energy efficiency measures by the metric of GDP growth. Nuclear power shifts rightward, to become among the most expensive measure (because even a 2030 horizon compares poorly with nuclear plants’ 10-year construc- tion phase but 60-year lifespan). Fuel efficiency measures still fare well (but note that the MicroMAC curve classifies these measures as energy efficiency). 48 The presentation derives from simulations using the VAT financing closure, but the results for the alternative closures are similar. 49 To translate the horizontal axis between the MacroAC curve and the MAC curve, it should be noted that one percent GHG abatement relative to BAU 2020 corresponds to about 4.6 tCO2e, while one percent in 2030 corresponds to about 5.4 tCO2e. 50 The units are somewhat different between the two models. For the MAC curve, costs are reported in € per tCO2e abated, while for the MacroAC curve, costs are reported as a percent GDP deviation from BAU for implementation of each lever up to its abatement potential. THE MACROECONOMIC MITIGATION OPTIONS (MEMO) page 76 MODEL AND THE MACROECONOMIC IMPACT OF THE ABATEMENT PACKAGE Figu Fi gure Figure re 3 35 35. MEM MEMO 5. M EMO O mo mode model: del: Mac Macroeconomic l: M acro roec econ onom ic A omic Aba Abatement bate teme ment Cos Cost nt Costt (M (Mac (MacroAC) acro AC) roAC ) cu curv curve, rve e, 2 2020 020 202 0aand nd 2 203 20300 030 Emission abatement impact, GDP % change Abatement potential, % change in 2030 N Note ote: Note: : A po positive si iti tive ive v al alue lue on value on ththe he ve vert vertical ic ical rti laaxis xis xi is me mean meansans that s thhat an an ab abat abatement batem emen t me ent measure meas ure asur e incr i increases ease ncrea s GD ses GDPP. Curve Cur Curve ve is for the is for he wh th whole hol ole le ab abate- bate abat e- me ment package. nt pac p acka ac kage ka ge. ge . Model Mode Mo del de l closure clos cl os osur ur uree is increase inc inc ncre re reas ase in VAT. ase VAT VAT AT. The area . The area under und nder und er the the curve cur urve cur ve (the (th the (th e size size of of rectangles) rect ctan rectangl an gl gles es es) has no economic ) has eco eco cono no nomi mi micc interpre- inte in terp rpre terp re- re tation. tati tati on. 1% GHG on tion GHG abatement aba aba bate teme ment teme potential p nt potot oten enti en tial ti in 20 al in 2020 corresponds c 20 coror orre re resp sp onds on spon to ab ds to about abou ou outt44.66 tC .6 tCOO2ee, , while whi hile whi 1% in 2030 le 1% 203 203030 0 corresponds co corr rres rres espo po pond nds nd s to about abo abo bout 5.4 ut 5.44 tCO tCO2e . e. Source: So urce ur Sour ce ce: IBS : IB S technical tech te chni ch nical ca nica paper, pape l pa per pe r, MEMO MEM MEMEMO O model mode mo del de simulations. simu l si mu mulalati lation ons tion s. page 77 The impact on GDP of various components of the abatement package shifts over time and becomes more positive, as investments are completed and operations begin for each lever. Comparing the 2020 and 2030 MacroAC curves, it can be seen that the GDP cost of many levers diminishes, and the shape of the curve flattens. Those with the most expensive and protracted investment phase will take longest to have a positive impact on GDP, in particular, some mea- sures related to energy supply, such as nuclear installations with construction periods that extend over more than half of the projection horizon. By contrast, investments in on-shore wind energy farms contribute positively to growth by 2020, since they can start operations much faster and require much smaller capital expenditures. However, with an even longer projection horizon beyond 2030, the shift from GDP growth-hampering to growth-enhancing will certainly materialize for levers such as nuclear plants, which have the longest lifespan of any available energy sector technology. The ranking of coal IGCC installations also improves significantly over time, while the GDP impact of conventional gas, geothermal, and solar PV deteriorates between 2020 and 2030. The gains from energy efficiency improvements in retrofitted buildings increase over time. A number of other energy efficiency measures which undercut GDP in 2020 support growth by 2030, including lighting controls and replacements of bulbs in commercial buildings, HVAC enhancements and replacements of bulbs in residential buildings, higher energy efficiency standards in buildings, energy efficiency advances in transport vehicles, and landfill-gas-electricity generation. An alternative presentation of the macroeconomic costs of the low carbon scenario is the Macroeconomic Marginal Abatement Cost (MacroMAC) curve, with unit costs of GDP change compared to abatement potential. When the impact on GDP compared to the business-as-usual levels is scaled to cost per percent of GHG abatement relative to BAU, it is easier to see which measures are ‘cheaper’ (see Figure 36). In other words, the Y values on the MacroMAC curve are elasticities of real GDP relative to carbon abatement. It follows that the area of each of the rectangles on the MacroMAC curve equals the height of the same rectangle on the MacroAC curve. While the area under the Macro AC curve (the size of the rectangles) has no economic interpretation, the area under the MacroMAC curve is of interest: the size of each rect- angle equals the impact on GDP of carbon abatement via that lever. The area under the MacroMAC curve in total defines the overall impact of the entire abatement package on real GDP. This interpretation is similar to that of the bottom-up MAC curve (in which the area under the curve equals the financial cost of the abatement package)51 (see Figure 36). 51 Note that the units of measurement differ between the MAC curve and the MacroAC and MacroMAC curves. See footnote 61. THE MACROECONOMIC MITIGATION OPTIONS (MEMO) page 78 MODEL AND THE MACROECONOMIC IMPACT OF THE ABATEMENT PACKAGE Figure 36. Macroeconomic Marginal Abatement Cost (MacroMAC) curve, 2020 and 2030 Note: Note Note: A popositive p si siti tive e val alue value on th e on the e vert vertical ertic al a ical xis axis is me mean means ans that s th an ab at an abat abatement atem emen ent meas measure t measure re in incr increases crea ease ses GDP s GDP. Curve Currve is for e is the for the whol whole hole e ab abate- ate abat e- ment me nttppackage. ackkageg .MModel oddell closure l is i i s increase in i VAT. VAT. The n VAT Th area of f any rectangle y re t gl ctan equals qual gle eq the the GDP ls th effect ffect GDP eff (loss or g t (l gain) aiin) ) of carbon f ca b rbon abatement abat ab atemen em atem ent en via t vi any a an specific spec y sp ecific if ecif lever. l ic levever er. ever Source: Source Sour : IB ce: IBSS technical tech te nica hniical paper, pape l pa per r, MEMO MEM MEMEMO O model mod mo del simulations. simu del siimullati tion lationss. page 79 The MacroAC and MacroMAC curves are fully consistent; their differing formulations suit them to different policy applications. The curves may appear to show different magnitudes of carbon abatement levers’ impact on real GDP; however, the GDP effect of a specific lever always has the same sign on the two curves, indicating either a GDP gain or loss, but the ranking of unit costs in terms of GDP change is not identical to the MacroAC curve’s total GDP cost per measure. Two examples of a shift in ranking are nuclear plants, which no longer rank as the most expensive option on a per unit basis, and wind energy, which ranks further to the right (more costly) once abatement potential is considered. The MacroAC curve serves well to inform policymakers who have committed to a package of abatement measures (such as those recommended by the MicroMAC curve analysis) as to the impact of that package on growth. The MacroMAC curve is the preferable tool if policymakers are considering what the content of a package should be and impact on GDP is a factor to be used in that decision. With the extension of the time horizon to 2030, some levers change their position on the MacroMAC curve and generate different macroeconomic effects. Between 2020 and 2030, the same flattening of the cost curve observed for the MacroAC curve holds for the MacroMAC curve, reflecting the unsurprising finding that in 2030 there are more abatement options which may enhance GDP growth. By 2030, additional measures related to fuel and energy efficiency improvements in transport and industry have positive impact. The biggest improvements from 2020 to 2030 in unit macro costs of carbon abatement appear in geothermal energy, introduction of hybrid cars, CCS installations (which remain GDP- reducing), and enhancements in lighting and HVAC in buildings. Numerous, often diminutive energy efficiency measures deliver relatively little GHG reduction individually but a package of the most GDP-enhancing measures could achieve critical mass (and get policymakers attention). Rather than focusing only on the interventions capable of significant GDP impact on their own, a package of small but effective levers could raise growth to a greater extent and at lower macro cost. Thus, an abatement policy oriented to a broad range of energy efficiency measures could be more effective in the long term in stimulating economic growth than a policy focused solely on the largest interventions. The MEMO model simulations provide a more complex assessment of possible abatement measures than the Mi- croMAC curve. This additional information changes the ranking of many technological abatement measures, although the new simulations also illustrate that there will be no single metric against which to judge the preferability of abatement choices. The creation of a new visual presentation helps summarize a good portion of these new findings. However, while this simplified graphic of results helps communicate the main findings of the MEMO model, and while this model remains both highly flexible and heavily detailed, the challenge of working with very large and complex models remains--they are not fully intelligible to anyone but their maker. Also, as noted in Section d, the convergence of Poland’s economy towards EU averages, while clearly a technically superior approach to constructing a BAU scenario, builds in moderation of GHG emissions via the ongoing shift towards less emission-intensive sectors such as services and via improved efficiency in each sector. This presumption of flexibility of factors of production between sectors is critical also to the relatively modest long- term costs of a low carbon scenario. It is the self-evident appeal to policymakers of this path-breaking model and its deep integration of bottom-up and top-down approaches that makes it particularly important that underlying assumptions be kept in view as policymakers consider the policy actions that would constitute a low emissions growth path for Poland. Further, Poland also faces immediate and specific obligations under EU legislation, which tie policymakers’ hands through predetermined targets and policies. The next chapter takes up this question of the cost of compliance to the EU 20-20-20 climate package. g. THE REGIONAL OPTIONS FOR CARBON ABATEMENT (ROCA) MODEL AND IMPLEMENTING EU CLIMATE POLICY THE REGIONAL OPTIONS FOR CARBON ABATEMENT (ROCA) page 82 MODEL AND IMPLEMENTING EU CLIMATE POLICY The Regional Options of Carbon Abatement (ROCA) model is a country-level CGE model for energy and GHG miti- gation policy assessment adapted to Poland to analyze implementation of the EU 20-20-20 policy package. The model considers some key variations on climate policy design that meet the same emission reduction targets and some alternative model assumptions that further illuminate the impact on Poland’s economy in 2020 through the analysis of 11 simulations. The ‘Main’ scenario, defining a central set of assumptions, finds that Poland bears a higher economic burden than the average EU country because of the predominance of coal in power generation. The market segmentation created by the EU’s division of economic sectors into ETS and non-ETS categories greatly elevates the marginal cost of abatement for non-ETS industries, and removing that segmentation reduces overall compliance costs for Poland. Similarly, relaxing the restriction on use of ‘where-flexibility’ (allowing emission reductions in the least-cost location) dramatically reduces compliance costs and the need for adjustment, as most abatement is off-shored. Then, an additional aspect of EU policy is incorporated into the ROCA model—overlapping regulation in the form of an EU target for renewable energy sources—to determine conditions in which it may be (counter-intuitively) welfare-improving. The model considers vari- ous policy choices under the control of the Polish government. First, alternative revenue recycling via wage subsidies is analyzed, which generates a weak ‘double dividend’ (reducing emissions while easing distortions in the labor market) and lower unemployment. Then, the loosening of restrictions on the scope of nuclear power is found to cut compliance costs for Poland by about one-third (although the engineering feasibility of installation of nuclear power by 2030 is generally agreed to be about 6 GW, about half the capacity necessary to generate the 35 percent of electricity projected under this scenario). Lastly, the granting of free emission allowances to energy-intensive and trade-exposed sectors, which might be vulnerable to carbon leakage, preserves sector output but generates overall losses in GDP. The ROCA model is a multi-sector, multi-country CGE model incorporating a hybrid bottom-up and top-down representation of the power sector. Starting from a much-reviewed CGE model framework and drawing data from the GTAP database, the ROCA model has eight sectors (with the disaggregation focused on energy intensive sectors) as well as five energy subsectors to allow for detailed analysis. Some key market distortions such as energy and trade taxes and unemployment are included. The model emphasizes spillover and feedback effects from international markets, with four countries or regions (Poland, other EU, other industrialized countries, and developing countries). The energy sector gets innovative treatment, with a hybrid bottom-up and top-down representation of power sector production possibilities. The model’s horizon stretches to 2020, the deadline for the EU 20-20-20 package obligations, and institutional settings and policy instruments for climate policy implementation are included, including the complex rules for the ETS and non-ETS sectors, and revenue recycling possibilities from carbon pricing. For data limitation reasons, it models only CO2 emissions, and it derives a business-as-usual scenario in line with external projections that foresees less energy-saving sectoral trans- formation (than the MEMO model baseline) (see Sections c and d and Annex 4 and Annex 6 for more details). The ROCA model is applied to assess compliance costs of the main features of the EU 20-20-20 package. The central constraint on Poland’s economy in 2020 in this modeling exercise is the need to meet the targets for emissions set out in the December 2008 EU climate change and energy package for emission-intensive (or ETS) sectors and for other sectors (non-ETS) such that the EU overall can reduce emissions by 20 percent in 2020 compared with 1990. Across all scenarios, these central provisions of EU climate policy legislation hold. Differential emission reduction targets are imposed for the ETS and non-ETS segments of the respective economies. For Poland, these targets are taken to be a 21 percent reduction in ETS sectors compared to 2005 (which is the EU-wide target) and a 14 percent increase in non-ETS sectors (the agreed national target). For ETS sectors, EU-wide carbon trading ensures equal prices of emissions abatement across the EU in all scenarios. For non-ETS sectors, the model assumes that each EU country imposes a domestic CO2 tax which equalizes marginal abatement costs only across each country’s domestic non-ETS emission sources. The ROCA model is a multi- region model, designed to analyze international feedbacks (both the impact of EU policy choices on global markets and international spillovers triggered by emission abatement policies of other major industrialized regions). Other industrial- ized countries (the third region in the model) are assumed to face a 2020 target of 4.8 percent abatement compared to 2005, roughly corresponding to pre-Copenhagen official pledges. They do not participate in carbon trading or offsets but rather set a uniform domestic carbon tax. The fourth region, developing countries, faces no target and supplies CDM projects. (See Table 8). The model considers some key variations on climate policy design that meet the same emission reduction targets and some alternative model assumptions that further illuminate the impact on Poland’s economy in 2020. In addi- tion to setting multiple abatement targets, EU policy segments sectors into two groups, creates additional requirements on renewable energy and sets indicative targets on energy efficiency. The agreed legal and regulatory structure for ETS page 83 sector emissions is likely to generate significant fiscal revenues from auctions for governments, conditional on the propor- tion allocated for free to assist affected industries; and governments will need to decide how to spend these revenues. Further, countries are free to design their own domestic policies to achieve non-ETS targets, including a domestic carbon tax.. To better understand how climate regulations generate economic costs, the ROCA model analyzes scenarios on the excess costs of emission market segmentation and of ceilings on foreign offsets (CDM limits) and on the efficiency losses of overlapping regulations. The importance of the method selected for revenue recycling for the economy-wide costs of emission abatement is addressed. Then some central features of the model are explored, through scenarios that vary the role of technology and technological policy constraints in power generation and the implications of international market spillovers (terms-of-trade) for the costs and effectiveness of sub-global (e.g., EU) climate policy action. The scenarios con- sidered by the ROCA model are summarized in Table 8. Table 8. Summary of scenario characteristics simulated in the ROCA model Scenario Basic assumptions • Emission reduction targets for 2020 are set as in EU climate package and Copenha- gen pledges: - Poland must reduce emissions in ETS sectors by 21 percent compared to 2005 and may increase emissions in non-ETS sectors by no more than 14 percent; the EU-26 (for the ‘rest of the EU’) faces 21 and 12.5 percent reduction targets for ETS and non-ETS sectors respectively (see Table 3). - Other industrialized countries must achieve 4.8 percent reduction as compared to 2005 levels, roughly corresponding to pre-Copenhagen official pledges. This region represents key other Kyoto Protocol Annex 1 countries: Canada, USA, Australia, New Zealand, Japan, and Russia. - The remaining region, developing countries, has no emissions target. All scenarios • Emissions trading focuses on the existing EU carbon market: - There is EU-wide emissions trading for energy-intensive industries (ETS sectors). - Access for EU non-ETS sectors varies according to scenario. - The other regions do not participate in international emissions trading. • Flexibility in the form of CDM offsets is included: - For the EU, varying access according to scenario. - The other industrialized region does not have access to CDM offsets. - The developing countries region is the supplier of CDM offsets. • Domestic carbon taxes are used: - For EU non-ETS sectors, each EU country imposes a domestic tax. - The other industrialized region sets a uniform domestic carbon price. - Revenue recycling varies by scenario. • No access for EU non-ETS sectors to carbon market. • Limits to CDM offsets for EU as prescribed by the EU climate package: non-ETS sectors are allowed to offset up to 33 percent of emission reductions, ETS sectors, 20 percent. Main • Lump-sum recycling to households of revenues from carbon tax and auctioning of allowances in carbon market. • Bottom-up activity analysis characterization of power supply technologies in the EU. • Restricted use of nuclear power at BAU capacity level. Flexible emissions trading Like Main but with access for EU non-ETS sectors to carbon market. Flexible trading & offsets Like Main but with EU non-ETS sector access to carbon market and no CDM limits. Renewables target Like Main but with target quota for renewable power generation in EU. Wage subsidy Like Main but with revenue recycling via wage subsidies. Unrestricted nuclear Like Main but without nuclear expansion ceiling in Poland. Restricted gas Like Main but with ceiling on gas use in Poland’s power generation at BAU level. Like Main but with output-based allowance allocation to energy-intensive and trade-exposed Free 30 % allowances (EITE) sectors in EU (free allocation of 30 percent of EITE sectors’ 2005 emission level). Like Main but with output-based allowance allocation to EITE sectors in EU (free alloca- Free 70% allowances tion of 70 percent of 2005 emission level). Top-down power sector Like Main but with top-down characterization of power production. Like Main but without international terms-of-trade effects (Poland is treated as a small Small open economy open economy). Source: Loch Alpine technical paper. THE REGIONAL OPTIONS FOR CARBON ABATEMENT (ROCA) page 84 MODEL AND IMPLEMENTING EU CLIMATE POLICY A summary of results of the 11 simulations, that explore aspects of meeting the EU targets, identifies which com- ponents of regulations are more costly, and demonstrates the interaction with other policy choices and model assumptions, is below. (See Table 9). Changes in real GDP in 2020, in unemployment, and in output of energy-intensive and trade-exposed sectors are compared to the business-as-usual scenario for 2020, as variables of central concern to economic policymakers. The structure of the power sector is determined for each simulation (except the case in which the power sector is simplified). Marginal costs of abatement are indicated as the price per metric ton of CO2 (tCO2). (More detailed results are available in Annex 6, in particular for the other industrialized countries and developing countries). Differences in marginal abatement costs across regions and sectors (e.g., ETS versus non-ETS) in the table reveal scope for direct overall cost savings through increased “where-flexibility” (the degree to which emission reductions are allowed to take place at the least-cost geographic location, regardless of nation-state boundaries).52 Abatement achievements before the use of flexibility mechanisms for each scenario illustrate how the off-shoring of abatement varies. While where- flexibility enhances efficiency, nevertheless, it is possible that an individual country such as Poland can be worse off from (more comprehensive) emissions trading due to adverse terms-of-trade effects for energy goods, energy-intensive com- modities or CO2 emission allowances. The ‘Main’ scenario sets out the central set of assumptions that will be altered in other scenarios. No trading is allowed between ETS and non-ETS emissions. Current regulations’ limits on use of CDM offsets hold (see Annex 1 for details). There is no target for renewables in power supply53. There is no free allocation of emission rights—all are auc- tioned54; and the revenues from auctioned ETS allowances and domestic taxation of non-ETS emissions are recycled lump- sum to households. The use of nuclear power–both in Poland as well as the rest of the EU–is limited to the BAU level, reflecting public concerns on the operation of nuclear power plants and the unresolved issue of long-term nuclear waste management. However, despite concerns about energy security, there is no ceiling on gas-powered electricity generation. Two of the innovative aspects of the ROCA model are applied here: this scenario includes detailed technology in the power sector; and the modeling of the interplay of four countries/ regions allows analysis of international terms of trade effects. 52 Greater flexibility lowers implementation costs. The Kyoto Protocol includes several mechanisms that allow for ‘where-flexibility’ (see Box 2). 53 The targets for renewables in the power sector were analyzed in one of the simulations, but note that because of data limitations, the ROCA model is not able to capture these targets exactly as formulated in EU regulations. The simulations in this report refer to the share of renewable in power supply, while the EU targets refer to the share of renewa bles in final energy demand. 54 The potential derogations in the energy sector for the modernization of electricity generation were not modeled in the ‘Main’ scenario both be- cause of limitations of the model and also the evolving interpretations of these regulations. In order to capture derogations in further work, some approximation of the projected adjustment path from 30 percent of allowances for the energy sector auctioned in 2013 to full auctioning in 2020 will be required. page 85 Table 9. ROCA model: economic impacts of alternative emission mitigation scenarios Flexible emissions trad- Flexible trading & off- Top-down power sector Free 30% allowances Free 70% allowances Small open economy Unrestricted nuclear Renewables target Outcome indicator Main scenario Restricted gas Wage subsidy BAU sets ing Real GDP (% change from BAU) Poland -1.40 -1.16 -0.28 -1.02 -0.98 -1.12 -1.02 -1.46 -1.40 -1.40 -1.74 EU26 -0.55 -0.41 -0.08 -0.37 -0.42 -0.45 -0.54 -0.56 -0.55 -0.54 Other industrialized -0.28 -0.28 -0.25 -0.27 -0.2 -0.28 -0.28 -0.28 -0.28 -0.28 Developing countries -0.11 -0.09 -0.08 -0.09 -0.09 -0.1 -0.1 -0.11 -0.11 -0.11 Unemployment (change in percentage points, relative to BAU unemployment rate) Poland 0.53 0.41 0.10 0.35 -0.39 0.37 0.55 0.53 0.52 0.44 0.49 EU26 0.17 0.12 0.03 0.04 -0.07 0.16 0.17 0.17 0.17 0.14 Output of energy-intensive and trade-exposed sectors (% from BAU) Poland -2.66 -2.82 -0.29 -1.13 -2.08 -1.86 -2.85 -2.40 -2.05 -1.94 -4.42 EU26 -0.73 -0.78 0.20 0.14 -0.55 -0.66 -0.74 -0.64 -0.51 -0.37 Technology shares of power sector (in % from total) Poland coal 84.1 73.5 70.2 81.9 75.6 73.3 50.7 78.9 73.4 73.4 73.5 gas 5.1 11.8 14.0 6.4 6.2 12.0 6.9 5.8 11.8 11.9 11.5 oil 0.9 1.1 1.1 1.0 1.0 1.1 1.0 1.1 1.1 1.1 1.1 nuclear 5.2 5.9 6.0 5.4 5.5 5.8 35.5 5.9 5.9 5.9 6.0 renewable 4.5 7.8 8.7 5.3 11.7 7.8 5.9 8.2 7.8 7.8 7.9 CO2 values (US$ per tCO2) ETS 29.7 36.4 7.9 10.7 30.1 26.9 30.1 29.8 29.8 22.0 29.7 Poland 87.2 36.4 7.9 86.6 91.3 88.3 87.4 87.3 87.3 87.8 67.6 Non-ETS EU26 81.9 36.4 7.9 79.6 84.0 81.8 81.9 82.0 82.1 81.0 CO2 reduction (inland, % from BAU) Poland -20.1 -19.7 -4.9 -15.8 -19.9 -24.2 -23.3 -18.4 -20.0 -20.0 -21.2 EU26 -14.7 -14.7 -2.8 -15.1 -14.7 -14.3 -14.4 -14.8 -14.7 -14.7 Other industrialized -16.5 -16.5 -16.5 -16.5 -16.5 -16.5 -16.5 -16.5 -16.5 -16.5 Developing countries -0.8 -0.8 -3.3 -0.8 -0.8 -0.8 -0.8 -0.8 -0.8 -0.8 Note: Simulation based on CO2 , not GHG, reduction. EU26 is the EU excluding Poland. Inland CO2 reduction excludes credits from CDM offsets. See Table 8 for definitions and description of scenarios. Source: Loch Alpine technical paper; ROCA model simulations; World Bank staff calculations. The findings of the ‘Main’ scenario illustrate that Poland bears a higher economic burden than the average EU country because of its relatively high abatement targets for non-ETS sectors with strong baseline emissions growth. Setting a non-zero price of carbon generates a negative shock to emissions-intensive sectors. Since power generation in Poland is predominantly coal-based, it will be hard hit. CO2 reduction from the sector takes place through rising electricity prices (by about 20 percent, much more than in the rest of the EU), a decline in output by about 10 percent, the expan- sion of CO2-free renewable power production, and, to a more limited extent, fuel shifting to gas (since nuclear power is assumed restricted to BAU levels) (see Figure 49). The higher costs of production for those sectors in which (fossil fuel) THE REGIONAL OPTIONS FOR CARBON ABATEMENT (ROCA) page 86 MODEL AND IMPLEMENTING EU CLIMATE POLICY energy inputs represent a significant share of direct and indirect costs leads to a loss in competitiveness, depressing production. In the new equilibrium, real wages are lower, and unemployment rises (although by only half a percentage point). The effects on real GDP are modest but more than twice as high for Poland as for the rest of the EU (with a loss of 1.4 percent of GDP).55 Ceteris paribus, differences in economic adjustment costs between countries and regions can be traced back to differences in the effective emission reduction targets: Poland faces the highest reduction requirement vis-à-vis the BAU situation in 2020 for non-ETS sectors (of about 22 percent, see following paragraph and Table 3), driven by expected strong baseline emission growth in its non-ETS sectors. Another important cost determinant is the ease of carbon substitution which is embodied implicitly in the sector-specific production technologies and consumer preferences. A third determinant is international feedback and spillover effects, when countries’ terms of trade and global prices are affected by abatement policies.56 (See Table 9 and Figure 37, Figure 38, and Figure 39). Energy-intensive and trade-exposed industries are not devastated by carbon abatement, but market segmentation drives the marginal cost of abatement in non-ETS sectors to almost three times the level in ETS sectors. The common EU price for ETS emissions amounts to roughly US$30 per mt of CO2. Energy-intensive and trade-exposed (EITE) industries (a subset of ETS sectors) are worried about ‘carbon leakage’57 and negative repercussions of emission constraints on pro- duction and employment.58 However, the simulations suggest that policy concerns in Poland as well as the rest of the EU about drastic adjustment effects in EITE sectors are unwarranted—these sectors are harmed more than the average (with 2.7 percent loss in output in 2020 compared to GDP losses of 1.4 percent), but the contraction is still moderate in size. A key reason is that, in the ‘Main’ scenario, other industrialized countries also undertake abatement, and CDM offsets put a scarcity rent on carbon emissions for developing countries such that the shift in comparative advantage is dampened. (However, it must be acknowledged that the impact on EITE sectors is reported as an average; and, at a more disaggre- gated level, production and employment shocks for specific sectors can be markedly higher.) At the same time, marginal abatement costs in the non-ETS sectors for both Poland and the rest of the EU are much higher than the ETS value (at a shadow price of US$87 per tCO2 for Poland, and US$82 for other EU), revealing less potential for cheap emission abate- ment in the non-ETS sectors given that the effective relative reduction requirements in the non-ETS sectors are similar (in the case of Poland: 21 percent effective reduction in ETS, 22 in non-ETS) or even lower (in case of the EU: 24 ETS, 14 non- ETS). The differences between ETS and non-ETS prices drive the direct excess costs of EU emission market segmentation, which are alleviated to some degree through limited low-cost CDM imports (at a price of only US$1 per tCO2). 55 The welfare cost to Poland is close to 1 percent, more than three times the cost to the rest of the EU. The cost in GDP and welfare to other indus- trialized countries is about half to two-thirds that of the EU (see Annex 6). 56 The impact of spillover effects dominates in the case of developing countries. Although they have no emission reduction pledges and should benefit through CDM, they face non-negligible welfare losses because of deterioration in their terms of trade (see Annex 6). 57 In this context, carbon leakage refers to the possibility that countries with more relaxed emissions policy may gain a trading advantage and produc- tion may move offshore to the cheaper country with lower standards. 58 According to EU criteria, a sector is considered at risk of emission leakage if the direct and indirect additional production from emissions controls costs exceed 5 percent of gross value added and the total value of its exports and imports exceeds 10 percent of the total value of its turnover and imports. page 87 Figure 37. ‘Main’ scenario : carbon emissions, % change vs. 2005 Inland Offsets Change vs 2005 in percent EU ETS -4.3 -16.7 -21 Poland ETS -2.8 -18.2 -21 EU non-ETS -4.1 -8.4 -12.5 Poland non-ETS -4.6 18.6 +14 -25 -20 -15 -10 -5 0 5 10 15 20 in % Fi Figure gure Figu re 3 38 38. 8. ‘Main Ma in’ sc ‘Main’ scen scenario enar ario io 2 2020: 020: 202 0: c car carbon arbo bon pric prices n pr es ices Figu Fi Figure re 3 gure 39. 399. ‘ ‘Main’ Ma Main scen scenario in’ sc ario enar io 2 202 2020: 020: 0: m macroeconomic acro mac econ roec onom omic ic v vari- ari var i- ables in Poland and the EU 100 1.0 87 Poland EU 26 0.5 90 82 0.5 Poland EU 26 0.2 80 0.0 % vs. 2020 BAU 70 US$ per tCO2 60 -0.5 -0.6 50 -1.0 -0.7 40 30 30 -1.5 30 -1.4 -2.0 20 10 -2.5 0 -3.0 -2.7 ETS carbon price Non-ETS carbon price GDP EITE output Unemployment Note Note: Note: 2004 : 20 04 isis the th he base y year base yea earr in in th t the he he m od del model la and nd d can can be trea treated be tr ted eate d as an an apapproximation prox appr imat imati oxi ion ion for for 202005, 05, wh 2005 which whi hich ich is is th t the he he bas base b asee ye year ar for for th the t he he 20-20-20 2 EU 20- 20-2 0 20 20 regulations. re regu gu g la lati ons tion EU is rest s. EU rest res t of the the EU excluding excl EU ex udin cluding Poland. g Po land Pola Inland nd. In land Inla carbon c nd car arbo bon emissions emis n em sion issi onss ar are from e fr om witwithin w ithi hin the n th region. regi e re g on gi Offsets Offs on. Of ets fsets are ar carbon carb e ca rbon emissions on emi e miss ssio ions ns ach achieved achieieve vedd by the region regio the reg ion through n th roug thro ughh CD CDM. EITE E M. EIT ITE value-added value E is val ue-aadd dded ed of energy- ener of en ergy gy- and and trade-exposed trade- trad expo e ex pose sed sectors sect d se ors ctor s (non-metallic (n on-m (non -metetal alli lic minerals, mine c mi rals nera ls, ir iron on and and ste steel s el p teel products, ro rodu duct s, non cts non-ferrous n -fer on-ferro rous us met metals, m als etal s, ppaper–pulp–print, per–p ap pul ulp p–pr int p in and t, a refined r nd ref efin ed oil ined oil pproducts). ro rodu duct s). UR is cts) the is th e h g in change the unem i th unemployment ploy pl yment rate trate t i in nppercentage ercentag t ge popoints. p intts. Th carbon The ca b p rbon price rice i in non-ETS i non-ET ETS sectors S se ctors t is i shadow s a shhaddow pr price. p ice. Source: Loch Alpine technical paper; ROCA model simulations. Equalizing the carbon price between ETS and non-ETS sectors in the ‘Flexible emissions trading’ scenario raises costs to ETS sectors but reduces overall compliance costs for Poland by about one-sixth of lost GDP and for the EU by about one-quarter. In this alternative, Poland and the rest of the EU are assumed to establish a comprehensive emis- sion trading market so there is only one EU-wide CO2 price. In 2020, the ROCA model finds that price to be about US$36 per ton (against US$30 per ton for ETS sectors in the ‘Main’ scenario). The integrated EU carbon market generates greater pressure on the electricity sector and other ETS sectors. In this scenario, the ETS sectors must undertake more emission abatement in order to reduce the expensive abatement burden of the non-ETS sectors. As a consequence, electricity prices rise further, power production declines more sharply, and the structural shifts in power generation across technologies become more accentuated. Likewise, the negative repercussions on EITE sectors are slightly more pronounced. Equaliza- THE REGIONAL OPTIONS FOR CARBON ABATEMENT (ROCA) page 88 MODEL AND IMPLEMENTING EU CLIMATE POLICY tion of marginal abatement costs across all EU emission sources reduces compliance costs for the EU compared to the cost identified in the ‘Main’ scenario by roughly 10 percent in welfare terms, or by 17 of lost GDP for Poland and 25 percent for other EU. At first glance, the excess costs of CO2 segmentation seem rather modest, but they might easily be significantly higher if, instead of the assumptions of the ‘Main’ scenario, the EU contained 27 differentiated non-ETS markets, cost- effective non-ETS regulation via a uniform carbon tax were replaced by a more realistic patchwork of overlapping domestic regulations, or if CDM access were more restrictive (since the EU regulations are complex and open both to interpretation and revision). (See Table 9 and Figure 40, Figure 41, and Figure 42). Relaxing the restriction on use of ‘where-flexibility’, in the ‘Flexible trading and offsets’ scenario, dramatically reduces compliance costs and the need for adjustment, as most abatement is off-shored. The ‘Flexible trading and offsets’ scenario assumes a comprehensive EU carbon trading regime and, in addition, relaxes the supplementarity constraints on CDM. Poland and the rest of the EU are now allowed to import up to their nominal emission reduction requirement with respect to 2005 levels (which still leaves obligations for domestic abatement from the substantial gap to the higher effective reduction requirement with respect to 2020). Access to more CDM abatement drastically decreases overall economic costs for Poland and the rest of the EU: CO2 prices drop to around US$8 per ton. Poland and the rest of the EU shift the bulk of abatement to the developing world, paying only about US$2 per tCO2 in CDM credits. While other EU emission reduction in the ‘Main’ and ‘Flexible emissions trading’ scenarios amounts to 15 percent compared to BAU levels, it is only about 3 percent in this scenario. The difference between the costs of US$2 per tCO2 for a CDM credit and the EU-internal CO2 price of US$8 is captured by the shadow price on the CDM quota, with quota revenues accruing to EU governments. The required economic adjustment in the power sector and the rest of the economy as compared to BAU is greatly reduced (see Figure 49). This scenario stands in stark contrast to the ‘Main’ scenario, and its assumptions are illustrative rather than highly realistic. For example, if the EU were to allow substantial use of foreign offsets, other industrialized countries would also likely be competing for CDM imports, pushing up their price. (See Table 9 and Figure 40, Figure 41, and Figure 42). Figure 4 Fi 40. 400. ‘ ‘Wh ‘Where-flexibility’ Where-fl flexib ibil iliity’ ’ scena scenarios: rios: i carbon ca b e rbon emissions, missi i ions, % change h 200 2005 vs. 2 5 005 Inland Offsets Change vs 2005 in percent Flex cdm -16.5 -0.1 EU Flex trade -4.2 -12.4 -16.6 Main -4.2 -12.4 Flex cdm -21.2 11.7 Poland Flex trade -3.8 -5.7 -9.5 Main -3.4 -6.1 -25 -20 -15 -10 -5 0 5 10 15 20 in % page 89 Figure 4 Fi 41 41. 1. ‘ ‘Wh ‘Where-flexibility’ flexib Where-fl ibil ilit ity’ scenarios: ’ scena i rios: carbon ca b p rbon prices rices i Fi Figure 4 42 42. 2. ‘ ‘Where-flexibility’ Where-fl ‘Wh flexib ibil ilit ’ scena ity’ scenarios: rios: i macroeconomic macroecono i mic variables, in % vs. BAU Main Flexible emissions trading Flexible trading & offsets 100.0 87.2 81.9 80.0 US$ per tCO2 60.0 40.0 36.4 36.4 36.4 29.7 20.0 7.9 7.9 7.9 0.0 ETS Poland EU non-ETS non-ETS Note: N Not No te: te: Se ote See See th the e ex pl plan lanat ati tio explanationion ion un und under der de der Fi Figure gu gure Figu 37 to re 37 to Fi F Figure igur ig igur ure 39. e 39 . Source: Sour So Sour u ce ce: Loch L : Looch ch Al Alpine A lp lpin ine ine te technical chni ch nica ica cal l pa paper; p p r; pe ROCA R ; ROC OC OCA model A mo del ode simulations. del si simu lati ula o s. tion Fi Figure Fig re 4 gure 43. 433. ‘ ‘Re ‘Renewables Rene newabl ablees t target’ arge tar get’ t’ s scenario: cena enari rio carbon arbo o: car n em bon emi emissions, ission ions, % cha hang nge change 2005 e vs. 20 05 Inland Offsets Change vs 2005 in percent Renew EU ETS -3.6 -17.4 Main -21 -4.3 -16.7 Poland Renew -10.2 -10.8 ETS -21 Main -2.8 -18.2 non-ETS non-ETS Renew -4.1 -8.4 EU -12.5 Main -4.1 -8.4 Poland Renew -4.6 18.6 +14 Main -4.6 18.6 -25 -20 -15 -10 -5 0 5 10 15 20 in % THE REGIONAL OPTIONS FOR CARBON ABATEMENT (ROCA) page 90 MODEL AND IMPLEMENTING EU CLIMATE POLICY Fi Figure 44 44. 4 ‘R 4. ‘ ‘Renewables Renewabl bles targe target’ ’ scena t’ rio: i carb scenario: bon carbon Fi Figure 45 45. 4 ‘R ‘Renewables 5. ‘ bles targe Renewabl target’ t’ ’ scena rio: i macroeconomi scenario: ic vari macroeconomic i- vari- p pr ices prices ables, in % vs. BAU Main Renewables target 100.0 87.2 86.6 81.9 79.6 80.0 US$ per tCO2 60.0 40.0 29.7 20.0 10.7 0.0 ETS Poland EU non-ETS non-ETS Note: N Not No te: ote te: Se See th See the explanation he ex plan pllanat ati tio ion ion ununder undde der Fi der Figure gu Figu g re 37 37 to to Fi Figure F igur ig e 39 ure 39. Source: Sour So Sour urce ce ce: Loch L : Looch Alpine h Al A lp ine te ine lpin technical ch nica ica hnical paper; p l pape p r; ROCA R ; ROC OC OCA model mod A mo del si del de simulations. simu imu mullation ion lati s. ons Overlapping regulation in the form of an EU target for renewable energy sources, modeled in the ‘Renewables tar- get’ scenario, can improve welfare rather than imposing additional costs, because of initial distortions and market imperfections. The use of multiple overlapping instruments in climate policy would seem to pose the risk of additional costs, e.g., the EU’s 20-20-20 policy package target of a 20 percent penetration of renewable energy by 2020 in primary energy consumption, with specific targets for each Member State.59 To impose this green quota on top of the explicit emission reduction targets for ETS and non-ETS sectors, the ‘Renewables target’ scenario starts from the ‘Main’ scenario and adds subsidies for renewable power generation, financed lump-sum by EU governments, sufficient to increase the renewables share in the power sector by 50 percent above the share in the ‘Main‘ scenario.60 However, this policy has a presumably unintended side effect: the increased share of renewables reduces pressure on the ETS emission ceiling, reduc- ing the ETS carbon price.61 As a consequence, both renewable power producers and the most emission-intensive power producers—coal—benefit (see Figure 49). Surprisingly, the simulations find that the GDP impact of an additional green quota for Poland and the rest of the EU is positive. While counterintuitive at first glance, it is the ROCA model’s inclusion of initial distortions (taxes, subsidies, and unemployment) which allows the possibility of welfare-improving second-best effects of additional policy constraints. Since Poland and the EU have relatively high taxes on energy use, imposition of subsidies to renewables moves regulation toward a more uniform and lower tax on carbon (with efficiency gains). Then, in the presence of initial unemployment, subsidies to renewables reduce the downward pressure on the real wage and thereby alleviate the increase in unemployment. Note that if all initial taxes and subsidies are set to zero in the model and then a green quota is imposed on top of an overall carbon emissions cap, the expected result emerges that the additional green constraint generates excess costs. (See Table 9 and Figure 43, Figure 44, and Figure 45). Revenue recycling via wage subsidies reveals a weak double dividend in which unemployment is reduced. The next alternative scenario reflects upon the scope of a double dividend from environmental regulation.62 Instead of the lump- sum transfer of rents from CO2 regulation, the ‘Wage subsidy’ scenario assumes revenue-neutral subsidies to labor. If pric- ing of carbon joint with wage subsidies effects an increase in real wages, unemployment will be lower, reducing the costs of emission mitigation (the ‘weak’ double dividend hypothesis). In the presence of initial tax distortions and labor market 59 The EU climate package contains explicit promotion of renewable energy production both because of its importance to emissions mitigation but also the possibility of technology spillovers and concerns about energy security. 60 Note that this target falls short of Poland’s commitment to a 15 percent share of renewable energy in gross final energy consumption by 2020. 61 Although in theory, the price of electricity could either increase or decrease as a consequence of renewable targets, in this scenario, it is markedly lower than in the ‘Main’ scenario (together with higher electricity production and demand). 62 The double dividend hypothesis postulates that increased taxes on polluting activities can provide two kinds of benefits: an improvement in the en- vironment; and, an improvement in economic efficiency from the use of environmental tax revenues to reduce other taxes that distort labor supply and saving decisions. A weak double dividend claim is that returning tax revenues through cuts in distortionary taxes leads to cost savings relative to the case where revenues are returned lump sum. page 91 imperfections, deliberate revenue recycling may be able to ameliorate the negative impacts of emission regulation. The ‘Wage subsidy’ scenario supports this weak double dividend hypothesis when revenues from emission regulation are not returned lump-sum but used to subsidize labor costs. In this case, the downward pressure of emission pricing on wages can be more than offset through wage subsidies such that real wages increase and unemployment falls. Revenue recycling through labor cost cuts rather than lump-sum transfers does not change the marginal costs of abatement but generates substantial infra-marginal cost savings through the implicit relaxation of labor market rigidities. As a consequence, the changes in the power generation mix and the sectoral structural change compared to the ‘Main’ scenario are relatively small with labor-intensive industries performing slightly better. (See Table 9 and Figure 46, Figure 47, and Figure 48). Fi Figure gure Figu re 4 46 46. ‘Wage 6. ‘Wa Wage ge s sub subsidy’ sidy ubsi scen scenario: dy’ sc ario enar io:: ca carb carbon on e rbon emissions, miss emi ions ssio ns, % ch chan change ange ge v vs. vs 200 2005 s. 2 5 005 Inland Offsets Change vs 2005 in percent W sub EU ETS -4.3 -16.7 -21 Main -4.3 -16.7 Poland W sub -3.2 -17.8 ETS -21 Main -2.8 -18.2 non-ETS non-ETS W sub -4.1 -8.4 EU -12.5 Main -4.1 -8.4 Poland W sub -4.6 18.6 +14 Main -4.6 18.6 -25 -20 -15 -10 -5 0 5 10 15 20 in % Figu Figure re 4 Figure 47. 477. ‘ ‘Wage Wa Wage ge s subsidy’ ubsi sub sidy dy’ sc scen scenario: ario enar io:: ca carb carbon rbon on p prices, rice pri s, ces Figure Figure Figu 48 48. re 4 ‘Wage 8. ‘Wa Wage ge s sub subsidy’ sidy ubsi scen dy’ sc scenario: enar ario io: macr macroeconomic : ma oeco croe cono nomi mic c va variables, riab vari les ables, in US$ per tCO2 in % vs. BAU Main Wage subsidy 100.0 91.3 87.2 81.9 84.0 80.0 US$ per tCO2 60.0 40.0 29.7 30.1 20.0 0.0 ETS Poland EU non-ETS non-ETS No Note Note: te te: : See th See the e ex expl explanation pl plan an at atio anat ion io n un unde under de der Figu Figure r Fi gure gu re 37 37 to Fig F Figure igur igure ur 39. e 39 . Source: Sour So Sour urce ce: ce: Lo Loch L och Alpine h Al A lp ine te ine lpin technical tech nica hni cal ica paper; p l pa p r; pe ROCA R ; ROC OCA OC model mod A mo del si del de simulations. imu simu mul lati lati lation ons ons. THE REGIONAL OPTIONS FOR CARBON ABATEMENT (ROCA) page 92 MODEL AND IMPLEMENTING EU CLIMATE POLICY Technology-specific policy constraints in power generation for Poland, in particular safety concerns related to nuclear power and energy security concerns related to imported gas, are exogenous to this modeling; however, re- moving any restriction on the scope of nuclear power cuts compliance costs for Poland by about one-third. The role of technology-specific policy constraints in power generation for Poland is investigated by two additional scenarios. The ‘Unrestricted nuclear’ scenario allows for the expansion of nuclear power in Poland beyond BAU levels, and the ‘Restricted gas’ scenario limits the use of gas in Poland’s power system to the BAU level. Apart from the costs and potentials of al- ternative renewable sources for power production, the ease of substituting away from coal in Poland’s electricity system hinges to a large extent on policy constraints and long lead times for nuclear expansion and the willingness to increase dependency on foreign gas imports. Both the nuclear and the gas option are subject to political economy concerns; yet economic analysis can at least provide a price tag to these options. Under the ‘Unrestricted nuclear’ scenario, the increase of electricity prices and, in turn, the decline of electricity output is roughly halved as the share in nuclear power generation goes up from about 6 percent in the ‘Main’ scenario to more than 35 percent in this scenario (see Figure 49) (although installation of so much nuclear capacity is unlikely to be feasible by 2020).63 Overall compliance costs are cut by one-third for Poland. By comparison, the ‘Restricted gas’ scenario keeps the nuclear ceiling and, in addition, restricts the use of gas- fired power plants in Poland to the BAU level (see Figure 49). The additional costs of constrained fuel switching from coal to gas in Poland’s power production are relatively modest. (See Table 9 and Figure 50, Figure 51, and Figure 52). Figu Fi gure re 4 Figure 49 49. 9. R ROC ROCAA mo OCA model: del: mode l: e ele electricity lect ctri rici city ty g generation ener gen atio erationn mi mix mix, in x, i n% Note: N Not ote No te: See te: Se See Ta Tabl T Table abl ble e 8 for fo defi d definitions for de niti efiniiti tion ons on and s an desc d description escri d de ipt pti tio ion ion of s scenarios. cena ce f sce nari ios os. rios. Source: Sour Sour urce ce: ce Loch : Lo Alpine A ch Alp lp pin ine technical tech e tech chni ni nica cal ca paper; p l pa p r; pe ROCA R ; ROC OC OCA model mode A mo del si del simulations. simu mula mu la lati ti tion on onss. 63 The MEMO model’s detailed optimization modeling of the power sector concludes that only about 11 percent of electricity supply can be nuclear by 2020. page 93 Figure 50. ‘Technology constraint’ scenarios: carbon emissions, % change vs. 2005 Inland Offsets Change vs 2005 Nuclear -6.2 -14.8 in percent EU ETS No gas -4.0 -17 -21 Main -4.3 -16.7 Nuclear -36.2 15.2 Poland ETS No gas -5.7 -15.3 -21 Main -2.8 -18.2 Nuclear -4.1 -8.4 Poland EU non- ETS No gas -4.1 -8.4 -12.5 Main -4.1 -8.4 Nuclear -4.6 18.6 non-ETS No gas -4.6 18.6 +14 Main -4.6 18.6 -35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 in% Figu Figure re 5 Figure 51. 511. ‘ ‘Technology Te Tech chno nolo logy gy c con constraint’ onst stra int’ rain t sscenarios: cena scenari rios os: carb carbon : ca on rbon Fi Figu Figure gure re 5 52. 52 ‘Technology 2. ‘Tech Techno nolo logy c con onst stra rain gy constraint’ int’ t sscenarios: cena sce rios nari os: : ma macr macroeconomic croe cono oeco mic nomi c prices variables, in % vs. BAU Main Unrestricted nuclear Restricted gas 100.0 88.3 87.2 87.4 81.9 81.8 80.0 81.9 US$ per tCO2 60.0 40.0 29.726.930.1 20.0 0.0 ETS Poland EU non-ETS non-ETS No te: te Note Note:: Se See e th the he ex plan pllan at anat atio explanationion io und n un der de der Fi under Figu gu gure Figure 3 re 377. 37. Source: Sour Sour Source ce: ce: Lo Loch L och Alpine h Al A lpin lp ine technical ine te ch hniica cal nica paper; p l pap r; pe ROCA R ; ROC OC OCA model mod A mo del si del de simulations. imu simu mul la lati tion ion s. ons THE REGIONAL OPTIONS FOR CARBON ABATEMENT (ROCA) page 94 MODEL AND IMPLEMENTING EU CLIMATE POLICY Figure 53. ‘Competitiveness risk’ scenarios: carbon emissions, % change vs. 2005 Inland Offsets Change vs 2005 Free 30 -4.3 -16.7 in percent EU ETS Free 70 -4.3 -16.7 -21 Main -4.3 -16.7 Free 30 -2.9 -18.1 Poland ETS Free 70 -2.9 -18.1 -21 Main -2.8 -18.2 Free 30 -4.1 -8.4 Poland EU non- ETS Free 70 -4.1 -8.4 -12.5 Main -4.1 -8.4 Free 30 -4.6 18.6 non-ETS Free 70 -4.6 18.6 +14 Main -4.6 18.6 -25 -20 -15 -10 -5 0 5 10 15 20 in % Figu Figure re 5 Figure 54. 544. ‘ ‘Competitiveness Co Comp mpet etit itiv iven enes ess risk risk’ s ri scen sk’ sc scenarios: enar ario ios: car carbon s: c bon arbo n Figu Figure Figure re 5 55. 55 ‘Competitiveness 5. ‘Co Comp mpet etit iven itiv ess enes s ri risk risk’ sk’ sc scen scenarios: ario enar s: m ios: mac macroeconomic acro roec onom econ ic omic prices, in US$ per tCO2 variables, in % vs. BAU Main Free 30% allowances Free 70% allowances 100.0 87.2 87.3 87.3 81.9 82.0 82.1 80.0 US$ per tCO2 60.0 40.0 29.7 29.8 29.8 20.0 0.0 ETS Poland EU non-ETS non-ETS No Note: te te: Note : Se See e thhe ex the plan pl explanation anat at atio io ion und under n un de der der Fi Figu Figure g re 37 gu 37. 3 7. Source: Sour Sour Source ce ce: Loch L : Looch Alpine h Al A lp ine te ine lpin technical tech nica hni cal ica paper; p l pape ROCA R p r; ROCOC OCA model mod A mo del si del de simulations. simu imu mul la lati tion lati ons ons. s. page 95 Granting free allowances to energy-intensive and trade-exposed sectors, which might be vulnerable to carbon leakage, is equivalent to a production subsidy that preserves sector output while generating overall losses in GDP. Unilateral emission caps with rising CO2 prices raise concerns on the competitiveness of energy-intensive and trade- exposed (EITE) industries and the potential for carbon leakage, particularly to emerging economies that lack comparable regulation.64 As a response, the EU 20-20-20 package allows for free allocation of emission allowances to EITE sectors.65 The ‘Free 30 percent allowance’ and ‘Free 70 percent allowance’ scenarios include partial free allocation of emission rights to EITE sectors in Poland and the rest of the EU. In the ‘Free 30 percent allowance’ scenario, 30 percent of the EITE sec- tors’ 2005 emissions are handed out for free; for the ‘Free 70 percent allowance’ scenario, the share is 70 percent. The allowance allocation is provided conditional on firms’ production decisions (via dynamic allocation) so as to provide an implicit subsidy to firms’ output.66 While the emissions price ensures producers still have economic incentives to reduce their emissions intensity, the subsidy discourages them from reducing emissions by decreasing production. While eligible sectors may benefit and leakage may be reduced, to meet the domestic cap those foregone domestic reductions must be made up elsewhere, driving up the emissions price and abatement costs. Since EITE sectors do not dominate overall emissions and production activity, real GDP and welfare losses of production subsides under the ‘Free 30 percent allow- ance’ and ‘Free 70 percent allowance’ scenarios are small. However, these costs can go up substantially if the share of EITE sectors increase or the scope of smart revenue recycling decreases. From a global efficiency perspective, the distortionary effects of output subsidies may be justified to a certain extent as a second-best policy for reducing counterproductive carbon leakage. A more subtle, selfish motive for output-based allocation from the perspective of a unilaterally abating region could be the strategic manipulation of the terms-of-trade. (See Table 9 and Figure 53, Figure 47, and Figure 55). Simulations run with a top-down representation of production possibilities in power generation generates a flat- ter marginal abatement cost curve and significantly lower marginal abatement costs in the ETS sector, illustrating again that model assumptions must remain in the forefront as results are presented. The ‘Top-down power sector’ scenario serves as a sensitivity analysis on the characterization of production possibilities in power generation, replacing the activity analysis representation of electricity generation in Poland and the rest of the EU with a conventional top-down approach. The direct costs of emission abatement in production are determined through the ease of fuel switching and energy efficiency improvements. A bottom-up modeling approach captures these costs through the activity analysis rep- resentation of discrete alternative technologies with different input intensities. The merit ordering of technologies with respect to abatement costs then yields an explicit marginal abatement cost curve where the area below the step func- tion equals total abatement cost. The bottom-up approach describes current and prospective technologies in detail and, therefore, is well suited to the analysis of technology-oriented regulation. However, there are practical shortcomings to the bottom-up approach when it comes to economy-wide analysis.67 In contrast, a top-down representation of produc- tion technologies captures substitution possibilities through constant elasticities of substitution which can be estimated from aggregate market data on prices and quantities.68 Estimates for marginal abatement costs in the ETS sector are sig- nificantly lower for the ‘Top-down power sector’ scenario which translates into distinctly lower overall compliance costs compared to the ‘Main’ scenario. The reason is the flatter marginal abatement cost curve associated with the specific es- timates for the top-down cross-price substitution elasticities–a difference between bottom-up and top-down approaches which indicates the need for further sensitivity analysis on technology characterization in key industries. (See Table 9 and Figure 56, Figure 57, and Figure 58). 64 There is by definition no leakage in the ROCA model, since CDM offsets are allowed (as in the ‘Main’ scenario) which requires the definition of an emission ceiling for developing regions (set at the BAU level); and other industrialized countries also have emission reduction pledges. 65 These simulations provide interesting insights on a related issue: national requests for EU permission to provide some portion of ETS allowances for free to the power sector starting in 2013 (or “derogations”). 66 If instead, the free allocation were tied to criteria exogenous to firms (static allocation), then the free allowances would be equivalent to a lump-sum transfer of scarcity rents to firms without affecting potential relocation or shut-down decisions. 67 The activity analysis representation of multiple production sectors becomes quickly intractable. In particular, there is the challenge to mimic rea- sonable supply price responses through the deliberate choice of lower (decline) bounds and upper (expansion) bounds for technologies as well as the calibration of technologies to specific supply elasticities. The substitution and transformation possibilities of aggregate production then are implicitly provided by the weighted input and output choices across all technologies. 68 As described in Section c, the multi-sector, multi-region CGE model developed for Poland combines a bottom-up characterization of the power sector with a top-down representation of all other industries. While the former approach reflects the paramount role of technology choices and technology regulations in the power sector for overall CO2 abatement, the latter can be based on empirical estimates for cross-price elasticities between capital, labor, material and energy inputs. THE REGIONAL OPTIONS FOR CARBON ABATEMENT (ROCA) page 96 MODEL AND IMPLEMENTING EU CLIMATE POLICY Figure 56. ‘Power options and trade effects’ scenarios: carbon emissions, % change vs. 2005 Inland Offsets Change vs 2005 Topdown -5.1 -15.9 in percent EU ETS Sm open -25.3 4.3 -21 Main -4.3 -16.7 Topdown -25.4 4.4 Poland ETS Sm open -0.9 -20.1 -21 Main -2.8 -18.2 Topdown -4.1 -8.4 Poland EU non- ETS Sm open -13.6 1.1 -12.5 Main -4.1 -8.4 Topdown -4.6 18.6 non-ETS Sm open -4.6 18.6 +14 Main -4.6 18.6 -35 -30 -25 -20 -15 -10 -5 0 5 10 15 20 in % Figu Figure re 5 Figure 57. 577. ‘ ‘Power Po Powe r op wer options tion opti onss an and trad trade d trade effe e ef effects’ fect cts’ s s scenarios: cena sce rios nari : os: Figu Figure Figure 58 58. re 5 ‘Power 8. ‘Po Powe wer r op options tion opti s an ons andd tr trad trade ade effe effects’ e ef cts’ fect s s sce scenarios: cena rios nari os:: carbon prices, in US$ per tCO2tCO2 macroeconomic variables, in % vs. BAU Main Top-down power sector Small open economy 100.0 87.287.8 81.981.0 80.0 67.6 US$ per tCO2 60.0 40.0 29.7 29.7 22.0 20.0 0.0 ETS Poland EU non-ETS non-ETS Note: N Not No te: ote te: Se See th See the explanation e ex planat pl ati tio ion ion ununder undder de der Fi Figu Figure gu g re 37 37. 3 7. Source: S So our Sour urce ce ce: Loch L : Looch Alpine h Al A lpin lp ine te ine technical tech nica ica hnical paper; p l pape p r; ROCA R ; ROC OCA OC model mod A mo del si del de simulations. imu simu mul lati lati la on ons tion s. s. Secondary terms-of-trade effects on the costs of emission abatement in Poland are found to be critical for Poland: In the ‘Small open economy’ scenario, the costs of compliance are roughly 50 percent higher than in the ‘Main’ scenario. The final scenario highlights the importance of secondary terms-of-trade effects for the costs of emission abate- ment in Poland. The ‘Small open economy’ scenario treats Poland as a small open economy, keeping international prices at the BAU level (suppressing terms-of-trade changes from international spillover effects).69 This scenario eliminates the feedback effects from international markets (by assuming that international prices are fully exogenous) and excludes international spillover effects from policy actions of trading partners (which for Poland is clearly omitting an important dimension of economic interactions since its economy is integrated in the EU common market and trades predominantly with EU partners). The strong mutual interdependence suggests that Poland is not only affected by the partners’ reac- tion to their own domestic policy actions but in turn responds to policy actions of trading partners. It would be expected that in any economic impact analysis of CO2 mitigation strategies in Poland, international spillover effects from emission regulation in the rest of the EU (and beyond) are likely to play an important role. The bulk part of Poland’s emission is regulated through the EU ETS where the marginal abatement costs emerge from supply and demand responses across all energy-intensive industries in the EU. The impact on the competitiveness of Poland’s industry primarily hinges on recipro- 69 The scenario adopts the international CDM prices and the EU ETS price generated by the ‘Main’ scenario. page 97 cal action by the rest of the EU. Finally, international fuel market responses are driven by abatement of all major world economies. (All these spillover effects are incorporated in a consistent manner into the ROCA model.) In the small open economy setting, the costs of compliance are roughly 50 percent higher as compared to the ‘Main’ scenario. However, as noted above, this scenario is useful as a comparator, not as an accurate representation of reality: the shift in comparative advantage at the expense of Poland’s industries is overstated because comparable emission regulation in the rest of the EU (and other industrialized regions) is omitted; and similarly, important terms-of-trade effects from global energy markets are missing. (See Table 9 and Figure 56, Figure 57, and Figure 58). The ROCA model’s analysis of the cost of compliance with the EU 20-20-20 climate policy for Poland’s economy in 2020 provides an informative counterpoint to the MEMO model’s assessment of the macroeconomic impact of an ambitious abatement package. This section has called attention to the important principle that the design of policy mat- ters for its effectiveness and efficiency. Simulations demonstrated that reducing the market segmentation designed into EU climate policy and allowing more ‘where-flexibility’ should foster overall cost-efficiency of emission abatement. These simulations move forward from the simpler policy world of the MEMO model, where a public subsidy induces implemen- tation of abatement measures and a tax increase (or one of three other public financing options) balances the budget. Nevertheless, the ROCA model also assumes optimal policy responses within the constraints of the EU regulations. That is, ETS sectors trade emissions rights in an EU-wide carbon market, allowing for reasonably cost-efficient abatement through a decentralized market mechanism. In non-ETS sectors, the model assumes a unified domestic CO2 tax. In practice, however, various EU members have shown an inclination to adopt a myriad of command-and-control measures (such as standards for tire pressure or mandatory tests for efficient driving), which will drive up the real-world costs of abatement compared to the ROCA model results. Policymakers should conclude that complex rules and regulations usually impose extra costs at the macroeconomic level, even if they seem well-tailored to the sector or issue. To explore this question of re- sponses at the sectoral level, and to complement the last three sections’ exploration of economy-wide modeling, the next sections turn to details in three sectors critical to a low emissions growth path: energy, energy efficiency, and transport. h. ENERGY SECTOR OPTIONS AND THEIR MACROECONOMIC IMPACT ENERGY SECTOR OPTIONS page 100 AND THEIR MACROECONOMIC IMPACT Although the energy sector has been at the center of all the analysis so far, this section sets out a more detailed examination of some key aspects of energy sector options, and, in particular, the careful optimization of the fu- ture structure of the electricity sector carried out for the MEMO model. Low-carbon energy supply options present significant opportunities for abatement, but they are usually quite expensive up front, take a long time to build, and then become long-lived assets with low operating costs. The power sector requires long lead times and, for some technolo- gies, has very long technical lifespans. The combination of technologies chosen or new investments will depend not only on capital costs, operational savings, and carbon abatement potential, but also energy security, domestic sourcing, and a raft of other issues. It is not surprising that the models applied here forecast the structure of the power sector to shift only slowly. The ROCA model’s results argue that a strong shift towards nuclear power is the option most likely to reduce emissions without harming the economy. The MEMO model, which takes the most sophisticated approach to selecting the structure of the sector, uses an optimization model to determine the cheapest feasible energy-mix package within multiple constraints. The additional fact that the average age of Poland’s existing energy infrastructure is high and, there- fore, ready for replacement, provides an opportunity for the country’s energy sector agenda to largely coincide with the low-emissions agenda. Choices about the energy fuel mix today will have lasting implications for Poland’s power sector emissions, but at the same time, given the lumpiness and complexity of low-carbon energy sector investments, maximizing potential abatement requires decisiveness and swift action. The power sector both produces the most emissions (at 38 percent of total) and holds the most potential for mitigation, including both demand reductions through energy efficiency, and fuel mix decisions. Figure 59 has extracted energy sector options from the MicroMAC curve and shows the full technical potential of abatement from the energy sector at 120 MtCO2 or 60 percent below 2005 levels. Five abatement scenarios for the power sector were constructed as part of the MicroMAC curve analysis to illustrate the impact of different tech- nologies on abatement potential. The maximum abatement potential is delivered by a ‘low emissions’ scenario in which coal power blocks are allowed to undergo natural retirement, with remaining power demand met by wind and nuclear energy. If the remaining power demand is instead met by gas plants, then the abatement potential falls from 120 MtCO2 to just 68. Implementation challenges for these scenarios are not only technological but also involve complicated tradeoffs related to energy security, nuclear waste risk, the economic importance and employment impact of local coal mining, and upfront investment costs versus future savings on operating costs. Figure 5 Fi 59. 59 Mi 9. M Microeconomic icroeconomi ic M Mar Marginal i l Ab ginal atement t tC Abatement Cost ost t((MicroMAC) MicroMAC (Mi MAC) for ) curve f or l low-carbon ow-carbon b energy sect tor i ector nvestme t nts t investments Abatement cost Weighted average EUR/tCO2e cost 22 EUR/tCO2e 80 Solar PV 70 Off-shore wind 60 Coal CCS 50 Demand reduction On-shore wind due to energy Biogas 40 efficiency improvements in 30 other sectors (~29 TWh) 20 10 0 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 105 110 115 120 125 130 135 140 145 -10 Biomass dedicated Co-generation Biomass co-firing Nuclear Small hydro Abatement potential MtCO2e in 2030 Note: Note Note: Each : Ea hc ch column ol olum lumn one of n is one the of th he 10 ab a abatement bateme bate nt mea ment measures. m easusure res T The s. Thhe he hei h height eigh ght of t the he he c f th columns ol olum lumns is th ns is he co the cost st in in € peper p abat abated r abbatedd tCO tCO2e . Th e. The T he he width idth widt hiis the t s thhe amount emissions t emi i issions be can b reduced. er d ed. educ d S Some ome measures are shown h ith n with net t ben et benefits b fits ( efit (negative neg gati costs). tive co st ts). ) Th scenario The scenari io assumes assu as su sume me mes that s th at 6 GW GW of nucnuclear nucle uc le lear power p ar pow ow ower will w il ill er wil installed i l be ins nstall ta nstalled ll by 20 ed by 2030, 30 30, pr 2030 providing prov ovid ov id idin ing ab ing about abou ou out 19% t 19 electricity. ele % of ele lect ct ctri rici ri city ci ty ty. Source: So urrce ce: urce McKinsey : Mc McKiKi Kins nse ns ey tec ey technical techn ec hn hnicical pap ical paper p aper ap er page 101 The power sector, clearly central to any switch to a low emissions economy, requires long lead times. Most of the energy sector capital stock will need to be replaced, which not only involves the substantial time required to construct new energy facilities, but also for a good portion of today’s capital stock to pass through a normal lifespan before replace- ment. New investments today will be required to be able to provide the energy supplies needed to satisfy higher energy demand in the future. For example, construction of an integrated gasification combined cycle (IGCC) coal plant takes 3 years while around 5 years are needed for a natural gas power plant. The first nuclear energy blocks of 1.5 GW might be able to start operating in Poland in 2020 at the earliest.70 There are also differences in the technical lifespan of power plant operations. For example, a gas-fired power turbine can operate for about 25 years and a conventional coal power plan for 45 years while nuclear power plants can operate for up to 60 years (see Table 10).71 Benefits from investments in energy sector generation and infrastructure pay off in the long term. However, in order to secure benefits in the future, e.g., after 2020, modernization efforts have to be launched now. Given the enormous investment needs in energy infra- structure and housing, it is also critical that Poland does not lock into unsustainable development patterns for long-lived infrastructure (see Table 10). Table 10. Key features of energy sector technologies available in Poland until 2030 Technology Life span Max Installed Base Production Uptime Investment time years 2005 (GW) 2030 (GW) 2005 (%) 2030 (%) Construction (years) Coal CCS retrofit 40 0.0 5.8 38% 64% 4-6 Coal CCS new built 40 0.0 3.5 79% 87% 5-8 Coal IGCC 25 0.0 5.8 90% 90% 4-6 Gas CCS retrofit 40 0.0 2.8 68% 68% 4-6 Gas CCS new built 25 0.0 0.7 29% 35% 4-6 Biomass dedicated 40 0.0 0.9 80% 80% 3-5 Biomass CCS new built 40 0.0 5.8 80% 80% 4-6 Nuclear 60 0.0 6.0 90% 90% 10-12 On shore wind 20 0.1 10.0 24% 24% 2-4 Off shore wind 20 0.0 6.0 32% 34% 3-5 Solar PV 25 0.1 1.7 10% 10% 1-2 Solar conc. 20 0.0 1.4 91% 91% 5-7 Geothermal 30 0.0 0.7 80% 90% 4-6 Small hydro 25 0.9 1.7 35% 35% 3-7 Coal conventional 45 30.8 32.0 54% 74% 5-7 Gas conventional 25 0.7 3.6 75% 65% 4-6 The structure of the power sector will shift only slowly, even with government commitment to a low emissions scenario. Figure 60 summarizes the energy sources for electricity generation in Poland today, under the ROCA model’s BAU 2020 structure of electricity generation, two low carbon scenarios for 2020, and the MEMO model’s 2030 projec- tion. In the ROCA BAU, power production in Poland is projected to be heavily coal-based (84 percent) while gas-fired power generation (5 percent) and electricity from renewables (5 percent) play a smaller role. Nuclear power, which is not 70 Already factoring in that the first decisions on nuclear power have already been made by the Government in early 2009 when the Council of Minis- ters adopted a resolution on the development of the nuclear energy program in Poland (January 13, 2009) and designated the Minister of Treasury to cooperate with the state-owned company Polish Energy Group (PGE), which will have a leading role in the program’s implementation. Also, on May 12, 2009, the Council of Ministers established the Government’s Plenipotentiary to the Polish Nuclear Power. 71 Production uptime also varies between technologies: for example, 90 percent for nuclear versus only 10 percent for solar PV. Uptime is expected to increase significantly by 2030 as compared to 2005 due to technological progress. For example, conventional coal-fired power generation is projected to be able to operate for 74 percent of the time by 2030, as compared with 54 percent in 2005. The predicted decline in production uptime for conventional gas (from 75 percent in 2005 to 65 percent in 2030 is driven by the assumption that these plants will be back-up options for renewable energy sources such as wind turbines. ENERGY SECTOR OPTIONS page 102 AND THEIR MACROECONOMIC IMPACT operating in Poland today, will be phased in by 2020 with a projected share of around 5 percent in the BAU scenario. This outcome contrasts with the projected structure of power generation in the rest of the EU in 2020 under BAU, which is much more balanced across coal (20 percent), gas (30 percent), nuclear (26 percent) and renewables (21 percent). Twenty years from now, under an ambitious program of abatement, coal may diminish to fueling just half of Poland’s power. By 2020, however, even if meeting the EU 20-20-20 targets, coal is likely to remain either three-quarters (under the ROCA model projections) or two-thirds (under the MEMO model forecast) of the power sector. Fi 60 60. Figure 6 Curre 0. C Current nt ta nd d proj and ject ted d el projected electricity lect tri icit mix ix i ity mi Poland, in P land, ol d 20 2020 20 a nd andd22030 030 203 0 Note: Note Note: : In tht the he MEM he MEMO M EMO mod model, O mo del del, Co Coalal is Co l is Coalal l con c conventional, onve vent ntiional iona C l, Coa Coal oal IGCC IGCC, l IG Coal CC, Co al l CCS CCS new new b ui uililt, built, and lt an d Co Coalal CCS lC new b CS new built ui uil ilt lt wi with with en ith en- ha nced nc hanc hanced oil rec ed oil r over ec ecov ov ery er recovery (EOR y (E OR); OR (EOR); Gas ); Ga is Ga s is Gas s co conv nven nv en enti ti on tion conventional, onal al al, Ga Gas CCS s CC new S ne w bu buil ilt il built, and Gas t, and CCS new Gas CCS b new buiui uilt lt wit built wit ith withh EO EOR; R; Re Rene ne newa wa wabl Renewablebl ble is On e is sh hore shor shoree wi ind nd, wind, d Sm Smalall Smallll hyd hydro dro, Geo hydro, G th eoth thererma Geothermal, mal l, a nd and B d Biiomass ioma Biomass ss d ed icat edi dic ted dedicated. d ed. Sour Sourcece: Source: L : Lo och Loch hA lpine lpi Al ine te Alpine tech ical hni nica technical pape l pa per, paper, ROC R r, ROCA OCA OC mo mod A modeldel simu siimul del simulations, lati lati tion ons, IBS s, I BS technical t tec ech hni hnicical lp aper pap er, paper, MEMO model , MEMO mod dell simulations, s si im imullat ati tions ions, W , Wo B orl rld ld Ba World ank Bank k staff staf aff calculations. ff ca lcul calc atio ulat ns. ions The ROCA model simulations provide some further insights on possibilities for transformation in the power sector. As noted above, with implementation of a ‘best guess’ set of policies to meet the EU 2020 targets (the ‘Main’ scenario), coal remains 74 percent of the sector. The outcome is even more biased towards coal if offshore offsets (CDM transac- tions) are unrestricted because then the country can use cheap coal and cheap abatement (bought from developing countries). The only stronger shift away from coal by 2020 comes if there is no ceiling on nuclear power (as a policy mat- ter, not as a technological constraint, although engineering feasibility likely limits nuclear in 2020 well below this ROCA simulation). In that case, coal falls to just over half of power generation, and nuclear power picks up one-third of genera- tion (see Figure 61). page 103 Figure 61. ROCA model: electricity generation mix, in % No Note: te: te Note See : See Ta Tabl Table ble ble 8 for for de fo d definitions efini defi iti tion ion ons and s an d de desc d description ript escriipt p io ion ion of s f sce scenarios. ce cena nari ios os. rios Source: Source ur Sour ce ce: : Loch Alp Loch Alpine A lp pin ine technical tech e tech chni nica nical ca paper; p l pa pe ; ROC p r; ROCA R OCA OC model mode A mo del si del simulations. simu mu la mula lati ti tion ons on s. An optimal energy mix scenario was constructed to be used in the MEMO model’s macroeconomic simulations. The Microeconomic Investment Decisions (MIND) module of the MEMO model is applied to the power sector, as described in section c, to determine the cheapest feasible energy-mix package, taking into account technological constraints such as the maximum availability of onshore wind power, the necessary electricity production to meet projected demand, the emissions reduction target, and the goal of minimizing the public subsidies necessary for all power options more expen- sive than coal power plants. The MIND optimization module finds the combination of energy sources which provides the biggest CO2 abatement at the lowest cost. The model suggests that the share of conventional coal in the overall energy mix should decline over the next two decades and be replaced by coal IGCC, nuclear, and onshore wind. There are only two types of energy plants that have positive net present values: geothermal and conventional coal. The disadvantage of the first is low capacity, and of the latter, high CO2 emissions. The alternative power plants which can effectively help to mitigate CO2 emissions in the energy sector all have negative NPVs. Among them, the cheapest are onshore wind and nuclear power plants, and the most expensive are dedicated biomass and new built biomass with CCS (see Table 11 and Figure 62). ENERGY SECTOR OPTIONS page 104 AND THEIR MACROECONOMIC IMPACT Table 11. Basic economic features of individual energy investment levers CO2 emitted per Potential capacity Energy price sub- CO2 price subject NPV GWh in 2030 in GW ject to NPV=0 to NPV=0, PLN Coal CCS, new built -0.45 103.3 3.5 0.53 202 Coal CCS, new built with EOR -0.45 145.7 0.5 0.53 205 Coal IGCC -0.46 69.8 5.8 0.52 181 Gas CCS, new built -0.93 47.2 0.7 0.58 253 Gas CCS, new built with EOR -0.09 134.6 0.5 0.43 61 Biomass, dedicated -1.64 558.1 0.9 0.63 921 Biomass, co-firing -0.28 714.8 0.5 0.45 401 Biomass, CCS new built -1.41 80.2 5.8 0.92 740 Nuclear -0.30 0.0 6.0 0.52 167 Onshore wind -0.12 0.0 10.0 0.45 78 Offshore wind -0.49 0.0 6.0 0.60 270 Solar PV -0.63 0.0 1.7 1.14 946 Solar concentrated -0.93 0.0 1.4 1.15 962 Geothermal 0.38 0.0 0.7 0.35 -44 Small hydropower -0.07 0.0 1.7 0.43 56 Coal conventional 0.01 796.8 38.0 0.42 N/A Gas conventional -0.69 386.1 3.6 0.47 161 Source: IBS technical paper, MIND module simulations. page 105 Figure 62. MEMO model: optimal energy sector scenario The energy mix 100% in this scenario Coal conventional shows a sharp decrease in the Coal CCS new built 90% utilization of conventional Coal IGCC coal plants and 80% a strong uptick Coal CCS new built with EOR in the share of nuclear and wind Gas conventional plants as well as 70% IGCC coal. The Gas CCS new built increase in share of gas plants can 60% Gas CCS new built with EOR be attributed to relatively low gas Nuclear prices. Thus, coal 50% remains the main Small hydro source of energy despite the in- 40% geothermal creases in alterna- tive sources. Solar conc. 30% Solar PV Off shore wind 20% On shore wind 10% Biomass CCS new built Biomass.co-firing 0% 2015 2020 2025 2030 Biomass dedicated S Source: Sour So our u ce: IB ce: IBSS technical te tech chni nica ica cal l pa p paper, pe p r, MIND MIN IND MIN mo odu D module dule dul le s simulations. imul im ati tio ulat s. ions The power sector scenarios deliver different patterns of emissions abatement, cost varying amounts of capital, and require differing levels of government subsidy to break even. Figure 63 presents basic features of 13 alternative energy mixes (see Annex 9 and Annex 10 for definitions). To enable direct comparisons, they were ‘standardized’ to achieve simi- lar reduction in GHG emissions, except the Delayed Action (which cannot reach the same abatement level) and Gas (with relatively high GHG intensity) scenarios. The Wind + Solar and Ministry of Economy scenarios are the most expensive op- tions, due to the relatively high reliance on expensive solar plants, and require the highest government subsidies. Through 2030, the share of the public contribution to capital expenditures exceeds the contribution by the private sector, whereas in the Optimal and other more balanced scenarios, public contribution is about 20 percent. ENERGY SECTOR OPTIONS page 106 AND THEIR MACROECONOMIC IMPACT Figure 63. Comparison of alternative energy investment scenarios GHG emissions 170 Delayed action 160 Gas Wind+biomass 150 Wind + Solar CCS 140 Mt CO2e Nuclear 130 20% shift to Wind + Solar 20% shift to CCS 120 20% shift to Nuclear 20% shift to wind+biomass 110 Reference 100 High gas price (opitmal) 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Ministry of Economy Total capital expenditures 40 35 Wind + Solar Ministry of Economy 30 Delayed action PLN billions (constant prices) Wind + biomass 25 Gas 20 CCS Nuclear 15 20% shift to Wind + Solar 20% shift to CCS 10 20% shift to Nuclear 20% shift to wind+biomass 5 Reference 0 High gas price (opitmal) 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Government subsidy in total capital expenditures 20 Wind + Solar 18 Wind+biomass 16 Delayed action PLN billion (constant prices) CCS 14 Ministry of Economy 12 Nuclear 10 20% shift to Wind + Solar 8 20% shift to CCS 6 20% shift to Nuclear 4 20% shift to wind+biomass 2 High gas price (opitmal) 0 Reference 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Gas Share of government subsidy in total capital expenditures 100% Wind+biomass 90% CCS 80% Wind + Solar 70% Delayed action Ministry of Economy 60% Gas 50% Nuclear 40% 20% shift to Wind + Solar 30% 20% shift to CCS 20% 20% shift to Nuclear 10% 20% shift to wind+biomass 0% High gas price (opitmal) 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 Reference page 107 The macroeconomic impact of alternative energy packages is inversely related to the average annual required capi- tal expenditures. (See Figure 65). Only the cheapest packages (which includes the optimal scenario) keep the deviation of GDP close to zero over the entire 20 year period. Scenarios that include sharp investment peaks (including the Ministry of Economy scenario) impose a higher cost on GDP, because they are not more equally balanced over time (see Figure 65 and Figure 66). Below is a summary table on macroeconomic impact of the optimal scenario and its sensitivity analysis (see Table 12). Figu Figure re 6 Figure 65 65. 5. R Rel Relation atio elat n be ion betw between twee een inve investment n in stme vest nt ment Fi Figu Figure re 6 gure 66. 666. E Emi Emission miss ion ssio abat abatement n ab atem emen ent the t in t ene energy he e rgy nerg y se sect sector or a ctor nd t and the he r req required equired uired cost and GDP elasticity of energy packages public subsidy in total capital expenditures (CAPEX) 1.4 4 2030 Annual public subsidy in CAPEX (PLN billion) 1.2 3.5 Average annual investment cost, 1.0 3 reference = 1.0 0.8 2.5 trendline 0.6 2 0.4 1.5 0.2 1 0.0 0.5 -0.20 -0.15 -0.10 -0.05 0.00 -0.2 0 GHG emissions abatement impact 2030, 0% 10% 20% 30% 40% 50% 60% 70% GDP elasticity -0.4 GHG emissions abatement in the energy sector by 2030 (max = 100%) S Sour So Source: our urce ce ce: IBS : IB S technical tech te hni nica ica cal l pa paper, p pe p r, r, MIND MIN MIN IND D module modul du dule mod s le si simulations. im imulati tio lations ns. ions. Energy sector modernization is important and can facilitate a shift to a lower emissions economy. Because energy assets in Poland, including energy infrastructure, are already far along in their lifespan, widespread rehabilitation or retire- ment will be necessary in order to assure undisrupted energy supplies and safety. The concurrence of new obligations for carbon abatement requiring substantial new investment with, at the same time, assets nearing their replacement point could greatly improve outcomes for Poland. The country should be well-placed to avoid stranded assets such as large new coal power plants that become too expensive once carbon is taxed (or restricted administratively). It seems that Poland’s energy sector agenda largely coincides with the low-carbon agenda. A well-managed modernization agenda of the en- ergy intensive sectors (covered by the EU-ETS), in particular the power sector, may allow Poland to meet the goal of emis- sions abatement while providing needed infrastructure at an affordable price. Table 12. Macroeconomic impact of the optimal energy mix scenario: sensitivity analysis Closure Scenario GHG abatement (in % vs BAU) GDP change (in % vs BAU) GDP elasticity vs GHG abatement 2015 2020 2025 2030 2015 2020 2025 2030 2015 2020 2025 2030 Reference (low gas 3.39 10.40 16.03 19.91 -0.15 -1.68 -1.31 -0.95 -0.04 -0.16 -0.08 -0.05 price) Public consumption 20% shift to Wind 3.61 10.54 16.21 20.16 -0.22 -1.56 -1.34 -0.88 -0.06 -0.15 -0.08 -0.04 + Solar 20% shift to CCS 3.40 10.38 16.01 19.95 -0.19 -1.81 -1.24 -0.99 -0.06 -0.17 -0.08 -0.05 20% shift to 3.40 10.64 16.30 20.29 -0.15 -2.00 -1.46 -1.15 -0.04 -0.19 -0.09 -0.06 Nuclear 20% shift to Wind 3.57 10.50 16.12 20.06 -0.15 -1.59 -1.24 -0.80 -0.04 -0.15 -0.08 -0.04 + Biomass Closure Scenario 2015 2020 2025 2030 2015 2020 2025 2030 2015 2020 2025 2030 Reference (low gas 3.41 10.40 15.98 19.90 -0.06 -1.23 -0.89 -0.71 -0.02 -0.12 -0.06 -0.04 price) 20% shift to Wind Social transfers 3.62 10.54 16.16 20.15 -0.10 -1.11 -0.95 -0.65 -0.03 -0.11 -0.06 -0.03 + Solar 20% shift to CCS 3.42 10.38 15.95 19.94 -0.08 -1.35 -0.83 -0.76 -0.02 -0.13 -0.05 -0.04 20% shift to 3.41 10.65 16.24 20.28 -0.04 -1.52 -1.00 -0.90 -0.01 -0.14 -0.06 -0.04 Nuclear 20% shift to Wind 3.58 10.51 16.07 20.05 -0.04 -1.18 -0.82 -0.59 -0.01 -0.11 -0.05 -0.03 + Biomass Closure Scenario 2015 2020 2025 2030 2015 2020 2025 2030 2015 2020 2025 2030 Reference (low gas 3.46 10.48 16.03 19.95 -0.14 -1.13 -0.78 -0.73 -0.04 -0.11 -0.05 -0.04 price) 20% shift to Wind 3.68 10.61 16.22 20.20 -0.17 -1.02 -0.83 -0.66 -0.05 -0.10 -0.05 -0.03 + Solar VAT 20% shift to CCS 3.47 10.47 16.00 20.00 -0.16 -1.24 -0.72 -0.77 -0.04 -0.12 -0.05 -0.04 20% shift to 3.47 10.74 16.30 20.34 -0.13 -1.39 -0.88 -0.91 -0.04 -0.13 -0.05 -0.04 Nuclear 20% shift to Wind 3.63 10.58 16.12 20.10 -0.12 -1.07 -0.69 -0.61 -0.03 -0.10 -0.04 -0.03 + Biomass Closure Scenario 2015 2020 2025 2030 2015 2020 2025 2030 2015 2020 2025 2030 Reference (low 3.55 10.71 16.10 20.06 -0.16 -1.33 -0.92 -0.89 -0.05 -0.12 -0.06 -0.04 gas price) 20% shift to Wind 3.78 10.83 16.29 20.31 -0.20 -1.21 -0.96 -0.83 -0.05 -0.11 -0.06 -0.04 + Solar PIT 20% shift to CCS 3.58 10.72 16.06 20.11 -0.19 -1.47 -0.86 -0.93 -0.05 -0.14 -0.05 -0.05 20% shift to 3.57 11.00 16.38 20.46 -0.16 -1.62 -1.06 -1.09 -0.05 -0.15 -0.06 -0.05 Nuclear 20% shift to Wind 3.72 10.80 16.18 20.20 -0.15 -1.26 -0.85 -0.76 -0.04 -0.12 -0.05 -0.04 + Biomass Source: IBS technical paper, MIND module simulations. i. ENERGY EFFICIENCY OPTIONS AND THEIR MACROECONOMIC IMPACT: A FIRST LOOK ENERGY EFFICIENCY OPTIONS page 110 AND THEIR MACROECONOMIC IMPACT: A FIRST LOOK Energy efficiency measures hold out the promise of relatively low cost abatement that works directly to delink car- bon from growth, the essence of a low-carbon economy. While energy sector abatement options are generally about reducing emissions intensity (CO2e created by energy use), energy efficiency focuses on reducing energy intensity (energy required for each euro of output). Both of these factors will reduce Poland’s emissions elasticity (with respect to growth) (see Figure 21). Poland’s economy is still more than twice as energy intensive as the EU average, suggesting that potential improvements should be easy to find. Indeed, energy efficiency measures are essential to the MicroMAC curve analysis, and many are growth-enhancing by 2030 in the MEMO model analysis. Deeper detailed analysis of energy efficiency options in Poland is needed to be able to provide more specific recommendations on how to overcome implementation obstacles that are preventing households and businesses from realizing the financial savings embedded in many of these measures. Energy efficiency measures play a central role in the MicroMAC curve analysis because of their substantial potential, apparent low price, and impact on growth. Together energy efficiency levels generate about one-third of the MicroMAC curve’s potential abatement for Poland. Many of these measures are assessed to have negative financial costs, generating net savings through reduced energy costs after an initial investment. Lastly, energy efficiency measures in general reduce the energy intensity of the economy, beginning the necessary delinking of economic growth from CO2 emissions. Of the 60 energy efficiency measures analyzed by the MicroMAC curve model, about two-thirds of the abatement potential is in the buildings sector (at an average savings of €14 per tCO2e), including better insulation, more energy-efficient appli- ances, water heaters, and lighting. About 20 percent of savings come from energy efficiency measures in the transport sector (saving €8 per tCO2e), from more fuel-efficient vehicles.72 The remaining 15 percent is in industry (saving €6 per tCO2e), and such measures as improving motor systems in chemical plants and implementing energy efficiency projects in petroleum and gas (see Figure 67). Abatement measures do not in reality have negative net costs after implementation barriers are considered. House- holds and businesses are not ignoring significant savings opportunities from implementing these measures. Instead, it is accepted that various implementation barriers are discouraging action. As mentioned in section e, the types of barriers likely to be preventing up-take are: high upfront investment costs (for example, for an energy-efficient car), principal- agent problems (such as the owner, operator, occupant, and bill payer of a building being separate entities), and lack of information (about what savings are likely). A fourth, and potentially most difficult obstacle, is the costs of implementa- tion across a high number of small entities (for example, with residential lighting). In the absence of analysis of these barriers, a simple assumption is that no NPVs for abatement levers can drop below zero. If so, then the weighted average cost across the MicroMAC curve of €10 per tCO2e will rise to €15 per tCO2e, and the overall cost of implementing the MicroMAC curve levers will rise by at least 50 percent. 72 The MicroMAC curve analysis includes transport energy efficiency measures in the broad category of energy efficiency. These levers, combined with other transport measures, are considered together by the TREMOVE Plus model of road transport in section j. page 111 Figure 67. Microeconomic Marginal Abatement Cost (MicroMAC) curve for energy efficiency levers Cost Buildings Transport Industry EUR/tCO2e DHP Behavioral/procedural changes Water heating – replacement of electric, residential Improved maintenance & process control Behavioral changes and improved maintenance & process Energy efficiency projects requiring CAPEX at overall plant level (co-generation) control in upstream oil and gas production Energy efficiency II (general) Retrofit HVAC, commercial Energy efficiency GHP Lighting new build controls, commercial projects requiring Retrofit HVAC maintenance, residential Waste heat recovery CAPEX at process Aggregated new build efficiency package, commercial Energy efficiency (general) unit level 80 Motor Systems, new build DHF Lighting – Lighting retrofit controls, commercial 70 Appliances, residential 60 switch CFLs Lighting – T12 to T8/T5, commercial to LEDs, Retrofit building envelope, residential 50 CHP, retrofit residential 40 Retrofit building envelope, package 2, residential More energy 30 efficient new Co-generation, CHP, new build 20 builds in new build GHF 10 upstream 0 -10 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 68 -20 -30 Aggregated new build Lighting – switch CFLs to -40 Diesel LDV energy effic. efficiency package, Abatement potential LEDs, commercial Gasoline LDV energy efficiency residential MtCO2 in 2030 -50 Electronics, consumer, residential -60 Motor Systems, retrofit Lighting – switch incandescents to LEDs, Co-generation, retrofit -70 commercial -80 Lighting – switch incandescents to LEDs, Gasoline LDV energy efficiency -90 residential Appliances – refrigerators, -100 Retrofit HVAC controls, commercial commercial -110 Retrofit building envelope, commercial Total abatement potential: 68 MtCO2e Electronics – office, -120 Energy efficiency projects requiring CAPEX at process unit commercial Average cost: -14 EUR/tCO2e -130 level in upstream oil and gas -140 Retrofit HVAC – electric resistance heating to electric heat pump, residential Retrofit HVAC – gas/oil heating, residential Note Note: Note: Ener Energy : En ergy gy eff e efficiency ff ffic ienc icie ncyy me meas measures ures asur i es inc include nclu lude de tra transport t nspo rans port rt s sec sector. ecto tor Eac Each r. Eachh column colu co mn is lumn one of is one the 60 f the 60 abatement ab abat atem ent emen meas t me measures asur ures es (on (only ( ly the only the mo st sig most significant signi nifi ficant one cant ones o nes are s ar named). name e na med) . Th d). The height heig e he ight ht of the of th columns colu e co lumn mns the s is t cost c he cosost per aba t in € per abated a ted bate d tC tCO O2e. The w width idth The wid th is is th the amou amount e am nt ount emissions emissi is emi sion ions can s ca be r n be reduced. educ edduced ed. Some mea d Some measures m easu sure res are s ar shown e sh hown show with ith n n wi net et benefits ben tb enef efit fit its s (neg (negative ( gat neg tive ati costs). ive co ts) st ) s). Source: Source Sour ce: McKinsey : McKi M Kins cKi ey tec nsey technical techhni hnic l pap ical paper. p er. aper Initial analysis of the macroeconomic impact of energy efficiency measures in the MEMO model found that al- though most energy efficiency measures individually have little potential, if they could be grouped together for implementation, they could be an important carbon abatement tool. Figure 67 presents the MacroMAC curve for just energy efficiency interventions. Among energy efficiency measures, the waste management levers are the most promising for abatement potential and also for their impact on economic growth. As for low-carbon energy supply options, energy efficiency measures are expected to switch from growth hampering towards growth enhancing as soon as the investment phase is finished, and the curve flattens between 2020 and 2030. For example, residential efficiency and envelope shifts drastically from far on the right, with the most costly measures in terms of growth, in 2020, to the far left and the most growth-enhancing. ENERGY EFFICIENCY OPTIONS page 112 AND THEIR MACROECONOMIC IMPACT: A FIRST LOOK Figure 68. Macroeconomic Marginal Abatement cost (MacroMAC) curve for energy and fuel efficiency micro i packages k micro-packages, 20 2020 20 a andd2 nd 030 030 2030 Emission abatement impact, GDP elasticity Abatement potential, % change in 2020 Emission abatement impact, GDP elasticity Abatement potential, % change in 2030 Note No te te: Note:: Mo Mode Model del de clos closure l cl os osur ur ure i increase ncre nc e is inc re reas as ase V VAT. AT AT. e in VAT Source: Sour S So our urce ce: ce IBS : IB S technical te tech nica hni cal ica paper, p l pa pe r, MEMO p r, MEM EMO MEM O model mod mo de del simulations. del si imu simu mul lati la lati tion ons ons. s. page 113 Exploiting the energy efficiency agenda is not easy, but it is often seen as a ‘win-win’ option, with benefits realized relatively quickly and lower upfront costs. Much of energy efficiency potential remains untapped because of the many obstacles to investments: inadequate domestic energy prices and lack of payment discipline, insufficient information on suitable technologies, too few contractors and service companies, and financing constraints. The government needs to address these issues in a coordinated manner. Effective energy efficiency interventions combine critical market-based approaches (which send correct price signals) with irreplaceable regulatory activity (which supports changes in practices and behaviors of economic agents). Numerous but limited energy efficiency measures deliver little abatement individually, but a package of the most growth-enhancing measures could achieve critical mass (and warrant policymakers attention). Rather than focusing only on the interventions capable of significant GDP impact on their own, a package of small but ef- fective levers could raise growth to a greater extent and at lower macroeconomic cost. Thus, an abatement policy oriented to a broad range of energy efficiency measures could be more effective in the long term in stimulating economic growth than a policy focused solely on the largest interventions. j. TRANSPORT: AN ALTERNATIVE ENGINEERING APPROACH TO MITIGATION OPTIONS TRANSPORT: AN ALTERNATIVE ENGINEERING APPROACH page 116 TO MITIGATION OPTIONS While the energy sector, as the dominant source of today’s emissions, necessarily receives a much attention and analysis in any study of low carbon policies, and while energy efficiency, with well-known potential for ‘no-regrets’ actions, is rightfully high on policymakers’ agendas, Poland also needs to consider how to address the sector with the fastest growing emissions—transport. This section presents the findings of an alternative engineering approach to transport sector mitigation options, applying the TREMOVE Plus model to the road transport sector in Poland. Road transport GHG emissions in Poland are converging from a low historic base towards EU averages. While contributing about 10 percent of overall emissions, road transport constitutes about 30 percent of non-ETS emissions. The objective of sustainability and greening of the transport sector is not new for the EU, but the EU 20-20-20 climate package is now the centerpiece. A business-as-usual scenario through 2030 was developed for passenger and freight road transport in Poland, using the TREMOVE Plus model. This forecast incorporated key characteristics of Poland’s transport sector, in par- ticular, a high number of imported used cars, low motorization rates and low mileage, and a highly competitive freight sector that has been shifting to newer and bigger trucks. Emissions from road transport are expected to almost double between 2005 and 2030. Because steady technological improvements are already incorporated into the BAU projections, the two low carbon scenarios developed by the TREMOVE Plus model include only modest technological improvements and concentrate on other emissions-reducing policy measures. The results of the scenarios present a more worrying vision than previously established for the road transport sector in Poland, with abatement unlikely to hold emissions growth below 35 percent through 2020. The TREMOVE Plus model is a bottom-up activity-based transport model is based on the familiar EU transport and environmental model. The EU-wide model was updated for Poland with new projections of transport activity and the latest disaggregated data. Working from a slightly different set of assumptions than the MicroMAC curve analysis, the TREMOVE Plus model considers explicitly the characteristics of Poland’s road transport sector and assesses the impact of existing policy commitments and possible mitigation options. The model takes a very different, detailed sector approach to constructing a business-as-usual scenario for passenger and freight road transport in Poland, allowing for more of the distinctive character of Poland’s road transport sector, in particular, expected high growth in passenger cars and dis- tances travelled. Importantly, the BAU calculations consider explicitly which transport and environment policies should be included in the reference scenario, and in this aspect, the transport BAU is quite different from the other BAU scenarios. The preponderance of imported used cars in Poland and the advanced age of the passenger fleet; low motorization rates and very low mileage driven per car compared with the EU15; and a road freight sector already highly competitive are found to be decisive factors for a projection of GHG emissions from road transport that shows almost doubling between 2005 and 2030. Emissions from the transport sector have been growing at a high rate since accession to the European Union. Transport constitutes a modest 10 percent of overall GHG emissions in Poland but grew by 74 percent between 1988 and 2006. Within Polish transport, road transport is particularly dominant, generating 92 percent of sectoral GHG (see Figure 69).73 Sharply rising trade volumes, increasing at an annual rate of more than 20 percent, and quickly expanding GDP, at an annual rate of 6.7 percent between 2000 and 2007, have been directly stimulating the demand for road freight trans- port. EU accession has also increased household incomes which, together with the sudden availability of affordable sec- ondhand cars from other EU member states—principally Germany, has led to a dramatic increase in private motorization (measured as passenger cars per thousand people). Although the motorization rate in Poland is still quite low compared to the EU15 countries, it is fast catching up, fuelled by the import of second hand vehicles. At the national level in 2007 there were 383 cars per 1000 inhabitants, whereas only four years earlier, the average figure stood at 294 vehicles per 1000 inhabitants. (In Western Europe, the current average is about 500-600 cars per 1000 inhabitants.) Increasing car ownership and use is translating into higher GHG emissions, particularly given the age structure of the Polish vehicle fleet. Out of 13.4 million passenger cars registered in 2006, every fourth car was 6 to 10 years old, and two-thirds were over 10 years old. Only one in eight passenger car is less than 5 years old. As a result, the Polish passenger car fleet is relatively fuel inefficient and polluting. 73 Although energy, followed by agriculture and forestry, are the major sources of global GHG emissions, transport—emitting 13 percent of global CO2e—is the fastest growing sector and the one most closely linked to the consumption of fossil fuels. In the EU, transport is the only sector where the emissions of greenhouse gases increased between 1990 and 2006, rising 28 percent while emissions overall shrunk by 3 percent. Road trans- port is the largest contributor to EU transport emissions, accounting for 71 percent of all EU GHG emissions in 2006. page 117 Figure 69. Transport CO2 emissions in Poland by mode, 2006 S Sour So our urce ce ce: Source: : Po P Pol Poland’s ola land nd’ d’s Gre Greenhouse G reen reen enh hous ho hous use Gas In Gas e Ga I Inventory nvent Inve ntor tor ory y Re Repo Report. R p rt epo t. rt. The EU has numerous policies, regulations, and laws aimed at sustainability and greening of the transport sector, of which the EU 20-20-20 climate package is now the centerpiece. As discussed in Section b and elsewhere in this report, the climate package sets national ceilings for emissions from non-ETS sectors, which includes transport (contribut- ing about 30 percent of non-ETS emissions). Poland is limited to 14 percent growth in non-ETS emissions during 2005 to 2020 (which translates to a reduction compared to the business-as-usual level projected for 2020). In addition, the pack- age sets a target for renewable energy at 20 percent of gross final energy consumption, including a 10 percent share of biofuels in the transport fuel market. Various transport policy measures have been put in place (see Table 13), and the Eu- ropean Commission expects that this package will contribute about one-third of the reductions required from the non-ETS sectors. At the same time, Poland has national transport policies, aimed at the high-level goal of developing an efficient and modern transport system but also including numerous measures to improve air quality, reduce pollution, and reduce GHG emissions. For example, in order to meet EU requirements, in 2007 Poland adopted a long-term plan on the promo- tion of biofuels and other renewable fuels for 2008-2014. Outside of emissions mitigation issues, as a member of the EU, Poland is required to (i) ensure the development of a competitive internal market for transport through market opening and liberalization, (ii) facilitate investment in prioritized transport infrastructure and (iii) reform infrastructure pricing and taxation to encourage more efficient use of transport infrastructure. Over the coming two decades, these transport poli- cies are expected to work at somewhat cross-purposes with climate policies, encouraging significant additional private motorization and increasing mobility of the population. Table 13. EU sustainable transport policy measures Policies and laws Year Description Greening transport package 2008 EC initiatives through 2009 on transport sustainability Marco Polo II 2007-13 Funding for projects to achieve modal shift for freight transport Directive promoting the use of cleaner vehicles 2009 Public authorities and operators to take into account energy through public procurement consumption, CO2 emissions and local air pollutants when pur- chasing road transport vehicles Directive promoting biofuels in road transport 2009 Sets mandatory national targets for renewable energy share in transport Passenger car and light duty CO2 emission stand- 2009 Targets for new passenger cars to reach 130g/km by 2015 ards Rules on vehicle labeling to promote more energy- 1999 Fuel economy and CO2 information available for consumers. efficient vehicles Source: Transport technical paper, World Bank 2010. TRANSPORT: AN ALTERNATIVE ENGINEERING APPROACH page 118 TO MITIGATION OPTIONS A business-as-usual scenario through 2030 was developed for passenger and freight road transport in Poland, us- ing the TREMOVE Plus model. A plausible development scenario for the road transport sector in the absence of new policy measures, to provide a reference for comparing the future effects of policy measures and combinations of policy measures, was developed by detailed analysis of: (i) the demand for road transport; (ii) vehicle ownership and the impact of secondhand car imports; (iii) size and composition of the vehicle fleet; (iv) driving conditions (urban, rural and high- ways); and, (v) emission factors (i.e., the diffusion of technology). This analysis was undertaken using the TREMOVE Plus model, building on the EU transport and environmental model TREMOVE (v2.9-2009) for Poland which was taken as a starting point and updated with projected development of the transport activity and vehicle stock (by type, technology, fuel use, age and GHG emissions factors for each class of vehicle), from a wide range of sources including, vehicles sales and car imports data, and as a result of interviews with stakeholders in Polish governmental organizations. Table 14 gives an overview of the resulting business-as-usual vehicle stock and mobility indicators, taking into account the expected growth in demand for passenger and freight transport based on GDP, population, motorization and the improvement in quality and extension of road infrastructure. The pattern of overall road transport emissions through 2030 can be seen in Figure 70. Table 14. Overview of the business-as-usual vehicle stock and mobility indicators 2008 2010 2015 2020 2025 2030 Population (million) 38.0 37.9 37.6 37.3 37.0 36.6 Motorization (per 1,000 inhab.) 422 451 523 562 590 605 Vehicle kilometers (billion VKM) 105.0 118.0 158.1 184.7 213.7 246.6 Passenger car VKM per capita 2,762 3,113 4,203 4,951 5,772 6,737 Passenger cars (million) 15.3 16.4 18.4 20.2 21.6 22.3 VKM per passenger car 6,882 7,218 8,599 9,153 9,913 11,058 Source: Transport technical paper, World Bank 2010. A high number of imported used cars has a direct impact on the age and technology structure—and hence, the emissions performance—of the vehicle fleet. Secondhand cars flooded the Polish market after EU accession in 2004, and sales of new cars stood at about half the level of imported used car sales. Between 2006 and 2010, about 75 percent of cars registered for the first time in Poland were secondhand imports. In 2004, 73 percent of imported cars were over 10 years old. These cheaper cars have fostered rapid growth in vehicle ownership so that in 2008, Poland had 422 cars per 1000 inhabitants (compared to about 600 cars in Western Europe). In recent years, the average age of secondhand car imports has been dropping, and it is projected to continue to do so in the BAU projection, with the net effect of pushing down the average age of new registrations from 5.6 years in 2010 to 2.7 years by 2030. The annual mileage driven per passenger car is expected to increase dramatically from its current level of about 2800 kilometers to typical EU15 levels of about 6700 kilometers by 2030. (See Table 14). Spurred by tighter integra- tion of Poland within the EU, improved highway system, and higher family income, this increase in mileage per vehicle, when coupled with the expected increase in the motorization rate, is a main source of expected growth in fuel consump- tion and CO2 emissions. Fleet-weighted average emissions factors are then needed to compute overall fuel consumption and emission of local and global pollutants. The TREMOVE Plus model assumes that the present EU long-term targets for new car CO2 emissions will be met; and since these standards are outside of the control of Poland, they should be included in the BAU scenario. Consequently, new car emissions are assumed to fall to 95 gram CO2 per vehicle kilometer by 2020. To achieve these emissions standards, a series of improvements to internal combustion engine vehicles will be necessary. Potential options include a wide range of technologies and measures that constitute 14 bundles for passenger cars and light duty vehicles in the MicroMAC curve, estimated to generate about 8 MtCO2e in GHG mitigation. In this model, these measures are part of the reference scenario because vehicle manufacturers are expected to use these 14 bundles to com- ply with the EU vehicle efficiency regulations. The BAU scenario shows emissions for passenger cars growing from 21.2 MtCO2e in 2010 to 31.2 MtCO2e in 2030. page 119 Freight transport in Poland has witnessed a rapid increase in freight ton kilometers. The Polish road trucking sector is very competitive both nationally and internationally, partially because a large fraction of owner-drivers and small freight companies has provoked cut-throat competition that has kept prices down. Road haulage is considered more reliable, flexible and faster than rail and the trucking sector has grown substantially. Even low value bulk materials like coal are increasingly transported by trucks. In contrast to passenger cars, the age of Poland’s truck fleet has been falling since EU accession, since many Polish trucks operate on international routes within the EU where they are required to comply with current EU standards. At the same time, fuel use has been shifting towards diesel, which by 2007 had an 85 percent share in the truck sector, leading to lower emissions factors for freight. Truck size has also been rising, and larger trucks are sub- stantially more efficient per ton-kilometer than smaller commercial vehicles. The impact of these changes in technology, fuels, and increasing truck size comes together in the BAU scenario as a steady decline in the emissions factor. As for pas- senger cars, part of the projected gains come from an assumption that many of the efficiency improvements in medium and heavy duty trucks analyzed in the MicroMAC curve will occur in the BAU scenario. For trucks, 12 bundles of measures, estimated to create mitigation of approximately 1 MtCO2e by 2030 are part of the reference scenario. The business-as-usual scenario of the TREMOVE Plus model projects total CO2e emissions from road transport to climb by 210 percent compare to 1990 and 93 percent compared to 2005 emissions levels despite the inclusion of technological progress. Emissions are projected to increase from 21.3 MtCO2e in 1990 and 34.2 MtCO2e in 2005 to 65.9 MtCO2e in 2030. A detailed breakdown of emissions by source is provided in Table 15 below. These projected emis- sions push Poland far above its agreed 14 percent growth in emissions for non-ETS sectors. Freight transport appears the bigger challenge, with a higher growth rate of emissions of 124 percent forecast through 2030 and, in consequence, a rising share of overall road transport emissions. The TREMOVE Plus BAU scenario projects faster growth in emissions than some other road transport projections, for example, that from the EU PRIMES model or the EC’s TREMOVE model, but the TREMOVE Plus BAU matches more closely the path of actual emissions (see Figure 70). The overall growth rates match fairly closely the transport scenarios that are part of the BAU projections for the MicroMAC curve model (which estimates 100 percent emissions growth for the transport sector overall during 2005 to 2030) and somewhat higher than the MEMO model (with 64 percent emissions growth forecast but for a more broadly defined transport sector). However, the assumptions of policies included in the BAU scenarios varies, explicitly in the case of the MicroMAC curve analysis as compared to TREMOVE Plus and implicitly in the case of the MEMO model BAU scenario. These assumptions are critical to define the possibilities for mitigation from BAU emissions levels in the low carbon scenario analysis that follows below. Table 15. TREMOVE Plus BAU scenario of Poland road transport CO2 emissions, 1995-2030 Annual change Total growth CO2 emissions BAU (million tonnes) 1995 2000 2010 2020 2030 2010-2030 (%) 2005-2030 (%) HDV (heavy duty vehicle >16t) 4.8 6.4 14.2 19.0 22.0 4.5% 131% MDV (medium duty vehicle 3.5-15t) 1.2 1.6 3.4 4.6 5.4 4.7% 132% LDV (light duty vehicle <3.5t) 0.3 0.4 0.9 1.2 1.4 4.5% 133% Vans 1.1 1.7 2.3 2.8 3.5 4.3% 77% Total commercial truck transport 7.3 10.0 20.8 27.6 32.3 4.5% 124% Motorcycles, moped, buses, tractors 2.5 2.4 2.3 2.3 2.4 0.4% 2% Passenger cars 12.2 15.1 21.2 28.9 31.2 3.9% 74% Total road transport sector 22.0 27.5 44.3 58.8 65.9 4.1% 93% Source: Transport technical paper, World Bank 2010. TRANSPORT: AN ALTERNATIVE ENGINEERING APPROACH page 120 TO MITIGATION OPTIONS Figure 70. TREMOVE Plus BAU carbon emissions scenario for Poland’s road transport sector, 1990-2030 70.0 TREMOVE Plus BAU TREMOVE 2009 60.0 PRIMES 2007 Official emissions inventory 50.0 MtCO2e 40.0 30.0 20.0 10.0 1990 1995 2000 2005 2010 2015 2020 2025 2030 Note Note: Note:: TR TREM TREMOVE EMOV OVEE 20 2009 09 is is EC BAU pro roje j ct jectio ion n. BAU projection. Source: Sour Sour Source ce: ce Transport T Tran : Transp ranspor sp ort or technical t tech hni nica cal ica paper, pape l paper per, World Wor orld Wor ld Bank Ban ank Bank 2010. 2010 20 10 10. Because steady technological improvements are already incorporated into the BAU projections, the two low car- bon scenarios developed by the TREMOVE Plus model include only modest technological improvements and con- centrate on other emissions-reducing policy measures. The BAU scenario already projects relatively high efficiency of medium and heavy duty vehicles, with a pathway determined by increasing truck capacity and load utilization. Passenger car developments, guided by tighter emissions standards, similarly reflect relatively high efficiency along the BAU path. Thus, abatement from technological improvements will be limited across the road transport sector. Instead, the two lower carbon scenarios for the TREMOVE Plus model include few technological improvements and focus on policy measures that require behavioral changes. These scenarios add a set of ‘Precautionary’ and a set of ‘Proactive’ non-technological measures such as road pricing, fuel tax increases, eco-driving, parking policies and the promotion of public mass transport together with greater mode share for walking and cycling. To determine which policies to model, fifteen bundles, or areas, of road transport policies were evaluated for abatement cost, effectiveness, and potential timing. The policy areas with highest potential were then used to create the scenarios (see Table 16). page 121 Table 16. Overview of TREMOVE Plus low carbon scenario policy measures Policy measure Description Reduction in 2030 vs. BAU, in % Road pricing Introduction of electronic tolling on motor and 4.2% expressways; and gradual introduction of con- gestion charging in major cities Fuel tax increase for passenger Gasoline price increase of 10% 5.2% cars Fuel tax linked to CO2 standard Annual gasoline price increase equal to emissions 18% for passenger cars standard tightening3 Fuel price increase for trucks Diesel price increase of 10% 1.8% Eco-driving Introduction of eco-driving course to improve 4.7% fuel efficiency Parking policies Parking fees for entire inner city regions of all cit- 3.5% ies Promotion of non-motorized and Promotion of walking and cycling; and of metro, 2.3% public transport trams, and buses; and park and ride Larger heavier trucks and logis- More use of larger and heavier vehicles with more 25% tics efficiency efficient logistic chains and distribution efficiency Source: Transport technical paper, World Bank 2010. The results of the TREMOVE Plus model’s two low carbon scenarios present a more worrying vision than previously established for the road transport sector in Poland, with abatement unlikely to hold emissions growth below 35 percent. The ‘Precautionary’ and ‘Proactive’ abatement scenarios contain similar measures, but they have been quantified under different levels of effort in each scenario. For example, the ‘Proactive’ scenario includes the fuel tax policy linked to CO2 standards and a 10 percent fuel tax for trucks, while the ‘Precautionary’ scenario includes a fuel price increase of 5 percent for both cars and trucks. The other measures also deliver significantly less abatement under the ‘Precautionary’ scenario, due to less effort. The impact of the ‘Precautionary’ scenario’s policy measures is disaggregated in Table 17, and the ‘Proactive’ scenario in Table 18. Overall, the freight sector has greater potential for emissions cuts; and the single most powerful policy measure is larger heavier trucks plus logistics improvements (with 4.3 percent or 7.2 percent abatement). In the ‘Proactive’ scenario, a near competitor is the fuel tax linked to auto emissions standards, which delivers 5.6 percent abatement. In 2020, the two scenarios still leave emissions 58 percent and 35 percent higher than in 2005 respectively, exceeding the 14 percent growth target for non-ETS sectors by 21 percentage points even in the more stringent ‘Proactive’ scenario. The abatement potential in 2030 is estimated at approximately 12 percent and 27 percent in the ‘Precautionary’ and ‘Proactive’ scenarios with respect to the BAU scenario. The path of GHG emissions in each scenario are shown in Figure 71. Table 17. TREMOVE Plus ‘Precautionary’ scenario emissions reduction by policy intervention, MtCO2e Emissions reduction (MtCO2e) 2020 2030 Passenger cars: 2.6 3.1 Road pricing 0.6 0.7 Fuel tax increase 0.8 0.8 Eco-driving 0.5 0.7 Parking policy 0.5 0.5 Promotion non-motorized (public) transport 0.3 0.4 Freight trucks: 2.1 4.6 Fuel price increase truck transport 0.2 0.3 Larger heavier trucks, logistics efficiency 1.9 4.3 Total reduction in ‘Precautionary’ scenario 4.7 7.7 Source: Transport technical paper, World Bank 2010. TRANSPORT: AN ALTERNATIVE ENGINEERING APPROACH page 122 TO MITIGATION OPTIONS Table 18. TREMOVE Plus ‘Proactive’ scenario emissions reduction by policy intervention, MtCO2e Emissions reduction (MtCO2e) 2020 2030 Passenger cars: 9.0 10.2 Road pricing 1.1 1.3 Fuel tax linked to CO2 standard 5.2 5.6 Eco-driving 1.0 1.5 Parking policy 1.0 1.1 Promotion non-motorized (public) transport 0.7 0.7 Freight trucks: 3.5 7.7 Fuel price increase truck transport 0.4 0.5 Larger heavier trucks, logistics efficiency 3.1 7.2 Total reduction in ‘Proactive’ scenario 12.5 17.9 Source: Transport technical paper, World Bank 2010. Fi 71 71. Figure 7 TRE 1. T TREMOVE REMO MOVE VE P Plus Pl carbon lus ca b e rbon emissions missi i ions projjecti tions f projections dt for road transport ransport tb y scena by rio i scenario BAU scenario Cautionary scenario Proactive scenario Technology scenario Proactive + technology 70 60 50 MtCO2e 40 30 20 2005 2010 2015 2020 2025 2030 Note: Note Note: Prec Precautionary : Pr ecau auti tion onar ary scen scenario y sc enarario io con c contains tain onta s po ins p policy licy li y mea cy measures m sure easu ress su such ch as road as ro pricing, ad p ricing rici ng g, fu fuel tax inc el tax increases, i reas ncre ases es, and , an eco eco-driving. d ec driv o-dr ivin ing g. Pro Proactive P acti roac ve tive scenario scen sc enar io con ario contains conta tain s sa ins me mea same measures m easusure res but s bu with t wi greater g th grereat ater er eff effort. e ort ffor Technology T t. Tec echn hnol ogy olog scenario scen y sc enar ario contains c io cononta tain ins policy poli s po cy mea licy measures m sure easu s fo res for modest r mo dest mode techno- t st tec hno- echn o logical l loogi gica cal improvement i l impr mpr p ov ovem emenent in tru t in trucks t cks ruck (medium ks (m ( edi med dium ium an and heavy h d heav eavy duty d y dutty v uty vehicles). eh ehi hicl icles). les)) Source: S ource Sour ce: Transport : Tran T spor ransp ort t te technical tech hni nica ical paper, pape l pa per r, World Wor ld Bank World Bank Ban k 2010. 2010 20 10 10. Abatement of emissions in the transport sector is likely to require diverse and coordinated action by both local and national government. In transport, the authorities at all levels will need to encourage lower-emission modes of transport and design infrastructure and, especially, public spaces in cities with better access for public and non-motorized transport. To shift passenger traffic away from cars, integrated ticketing and integrated terminals for air, rail, bus, and tram transport will be important as well as expansion of rail and bus rapid transit service.74 Critical for freight traffic will be improvements 74 Bus rapid transit (BRT) is a term applied to a variety of public transportation systems using buses to provide faster, more efficient service than an ordinary bus line. Often this is achieved by making improvements to existing infrastructure, vehicles and scheduling. The goal of these systems is to approach the service quality of rail transit while still enjoying the cost savings and flexibility of bus transit. page 123 in intermodal freight transport75 and other measures to encourage a shift away from roads towards inland water and rail, as well as interoperability of transport modes between countries.76 Urban planning needs to consider carbon abatement. Cities, towns and districts can and should be designed and redesigned to minimize transport needs and with alternative transport in mind. Instead of building infrastructure to accommodate increasing numbers of cars, governments should invest in alternative arrangements that allow and encourage cyclists, pedestrians, and public transport riders (see Box 7). Box 7. The impact on emissions of Poland’s urban transport and urban planning policies A key area of transport not discussed elsewhere in this chapter is urban transport which accounts for a large share of overall emissions from the transport sector. Poland has a relatively high urbanization rate (of 61 percent), and most recent population growth in Poland has happened in and around larger cities. The suburbs of larger metropolitan areas have absorbed much of the population growth and been the site of the majority of new housing developments. While population overall is declining, the suburbs of Warsaw are growing, as are smaller cities’ suburbs. The shift towards suburban sprawl of Polish cities has significant implications for urban transport as well as other urban services. It is accompanied by a shift towards single-family units, likely with greater per capita residential energy require- ments. Importantly, expansion into peri-urban areas has been occurring at great distances from the city center, with haphazard new developments along main arteries driven by low land prices. Residents depend on private cars for com- muting into the city. Meanwhile, public investment has focused on improving and expanding roads rather than public transport. Most metropolitan areas do not have properly integrated public transportation networks: Warsaw’s network is fragmented and focused in the center city. A main result has been that the hard urban cores that defined Polish cities before 1990 have given way to scattered sub- urban developments that make public transportation networks in those areas impractical as well as raising challenges for provision of water, sewage, electricity, and solid waste management services. All of these aspects will contribute to raising GHG emissions in metropolitan areas. The jurisdictions that make up Polish metropolitan areas need to come together around an integrated regional urban planning approach to guide new developments in a more sustainable fashion and fostering opportunities to involve the private sector in regional service provision. The TREMOVE Plus model generates a BAU scenario for road transport against which reducing emissions from transport will be particularly difficult. The transport BAU scenario illuminates the kinds of policies that must be consid- ered part of business-as-usual for convergence to EU averages to happen over the next decades. That is, while the MEMO model abstracts from how Poland’s economy comes to resemble the EU average, this engineering approach must be specific. Similarly, the MicroMAC curve presumes some unidentified efficiency gains, but then counts as an option every possibility for GHG abatement. The transport modeling shows that many of these options are already part of business- as-usual and are not available to create additional mitigation. Within this modeling framework, the analysis finds that Poland still has relatively low rates of motorization, which argues that the growth of road transport will likely be high going forward. Further complicating the picture is the very high share of used vehicles, which tend to be much more fuel- inefficient and polluting. Long distance freight and passenger transport need substantial changes; to achieve a strategic shift towards a lower carbon intensity (per passenger- and tonne-kilometer transported) while promoting long-term devel- opment and contributing to a better quality of life. A paradigm shift is urgently needed to reshape and refocus the urban transport sector toward transport systems that not only get urban inhabitants where they need to go, but get them there sustainably and with little impact on the environment. Transport infrastructure design and development decisions taken over the coming years will directly affect this long term sustainability. Infrastructure investments have a long life; design decisions made centuries ago are still evident in many European towns and cities. If cities develop around the needs of private motorization and other unsustainable aspects of transport, as many cities currently do, they will lock in to a high energy-consuming development trajectory which will be difficult to change at a later date. 75 Intermodal freight transport involves the transportation of freight in an intermodal container or vehicle, using multiple modes of transportation (rail, ship, and truck), without any handling of the freight itself when changing modes. The method reduces cargo handling, and so improves security, reduces damages and losses, and allows freight to be transported faster. 76 The challenge of interoperability across national borders is most clear for railways, which have greater or lesser interoperability depending on whether both countries conform to standards of gauge, couplings, brakes, signaling, communications, loading gauge, and operating rules. CONCLUSIONS AND ADDITIONAL ISSUES Better understanding of technological options, economic ramifications, and policy impact enhance the likelihood that Poland can move quickly towards a low emissions growth path. Such a transition will deliver additional benefits, including added energy security from increased energy efficiency and use of renewable energy sources, human health benefits from transport and other improvements that reduce local air pollutants, and strategic and competitive advan- tages that are more likely to accrue to countries that pursue low emissions development early. While this report provides some complex assessments and new analytic tools for policymakers, its analysis, rather than exhausting the central issues related to a transition to low-emissions growth, has identified a number of additional economic issues for further work. One area for further research is follow-on development of the economywide and engineering models and the links between them. Having developed a suite of models with some new methods of integrating bottom-up with top-down, a direction for further work would be additional integration and harmonization of the models. In particular, remaining dif- ferences in the business-as-usual projections, the approach to modeling the power sector, and transport sector treatment could be resolved, albeit not easily Alternatively, and perhaps more fruitfully, these models can be transferred to government or local ownership to serve as tools for policymakers going forward. Further, the preservation of alternative models, which produce differing results, highlights continually for model users the criticality of model assumptions and simplifications. Supplementary analysis using the existing economy-wide models or off-the-shelf models suited to the topic might fruitfully be applied to a number of issues. Extending the time horizon to 2050 would allow a more balanced treatment of the impact of long-gestation mitigation opportunities such as nuclear power plants in the MEMO model. The inclusion of R&D expenditures and technological progress would allow improved analysis of the long-term gains in the economy from implementation of energy efficiency measures. Distributional and regional impacts would be of interest and could be approached simply using existing household survey data. Sectoral and bottom-up or engineering analysis could also be usefully supplemented. More detailed bottom-up analysis of energy efficiency options in Poland would help clarify the nature of implementation barriers and assign costs to them. This richer database could then be linked to the MEMO model to assess the macroeconomic impacts of a coor- dinated and significant program of energy efficiency. Sector studies of agriculture, land use, and forestry would assist in moving from the financial costing of the MicroMAC curve to understanding how to implement abatement measures in these sectors in Poland. The complexity of EU policy leaves many questions and regulatory aspects still to be analyzed. Public subsidies war- rant further attention, since existing distortions and overlapping regulations mean that the impact of an additional tax or subsidy is hard to predict. In particular, a better understanding is needed of the system of ‘white certificates’ which are intended to encourage investment in energy efficiency measures. The macroeconomic and fiscal implications of deroga- tions (or free allowances ) in the ETS system, of various recycling options of revenues from ETS auctions, of the possible introduction of carbon taxation in non-ETS sectors, and the proposal to re-introduce an excise tax on coal from 2012 also merit analysis. Lastly, better modeling of renewable energy sources and their complex EU regulations would better inform decisions on power sector investments. One last area that this report did not investigate was how to foster the new business opportunities that may arise for those countries that move earlier to a low-emissions economy. For all the same reasons that Poland has thrived following its transition to a market economy and its accession to the EU—high levels of education, conservative macroeco- nomic management, respect for the rule of law, middling infrastructure, and proximity to Europe—it might be expected that in the transition to a low-emissions economy, Poland would find a way to maximize the benefits and minimize losses. This report provides a detailed assessment of many aspects of a low emissions growth strategy for Poland, develop- ing insights via a suite of models that should provide ongoing assistance to policymakers in Poland. Policymakers may find reassuring the report’s main message that Poland’s transition to a low-emissions economy, while not free or simple, is affordable. However, capturing the full package of technologically feasible and economically sensible abate- ment measures requires coordinated and early action by the government. With an ambitious approach, Poland can aim to reduce its GHG emissions by about one-third by 2030 (relative to 1990) with little cost to incomes and employment. Similarly, meeting the EU targets for 2020 appear generally feasible for Poland at modest cost, albeit likely more challeng- ing for less energy-intensive sectors such as transport than for sectors that can access the efficiencies of EU-wide carbon trading. Poland has already weathered one economic transition and emerged with a strong and flexible economy. This next transition—to a low emissions growth path—while requiring an evolution in lifestyles and priorities over the next 20 years, may well turn out to be much easier. REFERENCES page 192 REFERENCES Technical background papers (available upon request) McKinsey technical paper: McKinsey & Company (2009), Assessment of Greenhouse Gas Emissions Abatement Potential in Poland by 2030, Poland (December), available at http://www.mckinsey.com/clientservice/sustainability/pdf/greenhouse_gas_abatement_potential.pdf Loch Alpine technical paper: Boeringer, C. (2010), Economic Impact of CO2 Mitigation Strategies for Poland: CGE Modeling and Capacity Building, Loch Alpine Economics, Ann Arbor, June 13. IBS technical paper: Bukowski, M., and P. Kowal (2010), Modelling the Transition to a Low Carbon Economy with a Large Scale DSGE Model of Poland, IBS [Institute for Structural Research], Warsaw, May. Transport technical paper: Rogers, John Allen, ECORYS Nederland BV, Christine Fallert Kessides, and Marcel Ionescu-Herioiu (2010), Mitigation of Greenhouse Gas Emissions from Poland’s Transport Sector, World Bank, Washington, DC, May. Other references Auerswald, P., and L. Branscomb (2003), “Valleys of Death and Darwinian Seas: Financing the Invention to Innovation Transition in the United States,” Journal of Technology Transfer (28:227). Avato, Patrick, and Jonathan Coony (2008), “Accelerating clean energy technology research, development, and deployment: lessons from non-energy sectors,” World Bank working paper no. 138, Washington DC. Badri, N. G., and T. L. Walmsley, eds. (2008), Global Trade, Assistance and Production: The GTAP 7 Data Base, Center for Global Trade Analysis, Purdue University. Böhringer, C., A. Löschel, U. Moslener, and T.F. Rutherford (2009), “EU Climate Policy Up to 2020: An Economic Impact Assessment,” Energy Economics (31: 295–305). Committee on Climate Change (2008), Building a Low-carbon Economy - the UK’s Contribution to Tackling Climate Change, The First Report of the Committee on Climate Change, December, available at http://www.theccc.org.uk/reports/building-a-low-carbon- economy. Council of European Union (2008), 17215/08, Energy and climate change - Elements of the final compromise, Brussels, 12 December. EIA (2009), International Energy Outlook, Energy Information Administration, United States Department of Energy, available at http:// www.eia.doe.gov/oiaf/ieo/. European Commission (2008), European Energy and Transport: Trends to 2030—2007 update, Directorate General for Energy and Transport, available at http://ec.europa.eu/dgs/energy_transport/figures/trends_2030_update_2007/energy_transport_trends_2030_ update_2007_en.pdf European Commission (2009), Regional challenges in the perspective of 2020: Regional disparities and future challenges, Background paper on climate change, Report to the Directorate-General for Regional Policy, April. European Commission (2010), Unlocking Europe’s potential in clean innovation and growth: analysis of options to move beyond 20%, Brussels. European Union (2008), The Climate Action and Renewable Energy Package: Europe’s Climate Change Opportunity. page 193 Fay, Marianne, Rachel I. Block, and Jane Ebinger, eds. (2010), Adapting to Climate Change in Eastern Europe and Central Asia, Office of the Chief Economist, Europe and Central Asia Region, World Bank, Washington DC. Fischer, C. (2006), How can renewable portfolio standards lower electricity prices? RFF Discussion Paper 06-20-REV, Resources for the Future, Washington DC. Government of Poland (2009), Poland 2030: Development Challenges, Board of Strategic Advisers to the Prime Minister, May. IBS (2008) “Assessing Effects of Joining Common Currency Area with Large-Scale DSGE model: A Case of Poland”, IBS Working Paper #3/2008, Institute for Structural Research, Warsaw, available at http://ibs.org.pl/publikacja/Effects_of_Joining_Common_Currency. IPCC ( 2007), Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, R.K. Pachauri and A. Reisinger (eds.)]. Intergovernmental Panel on Climate Change, Geneva, Switzerland. Jacoby, H.D., and A.D. Ellerman (2002), The Safety Valve and Climate Policy, Report No. 83, MIT Joint Program on the Science and Policy of Global Change. Johnson, Todd M., Feng Liu, Claudio Alatorre, and Zayre Romo, eds., (2008), Mexico Low-Carbon Study, World Bank (December). Kaya, Yoichi, and Keichi Yokobori, eds. (1997), Environment, Energy, and Economy: Strategies for Sustainability (Tokyo: United Na- tions University Press). Ministry of Economy (2009), Energy Policy of Poland until 2030, Appendix to Resolution no. 202/2009 of the Council of Ministers, Government of Poland (10 November). Nordhaus, W. (1997), “Discounting in economics and climate change,” Climatic Change (37, 315–328). Okagawa, A., and K. Ban (2008), “Estimation of Substitution Elasticities for CGE Models,” mimeo, Osaka University, April 2008, avail- able at: http://www.esri.go.jp/jp/workshop/080225/03_report5_Okagawa.pdf Pope, J., and A.D. Owen (2009), “Emission trading schemes: potential revenue effects, compliance costs and overall tax policy issues,” Energy Policy (37, 4595–4603). Porter, Michael E. (1990), The Competitive Advantage of Nations (New York: The Free Press). Republic of Poland (2006), Fourth National Communication under the United Nations Framework Convention on Climate Change (Warsaw). Smol, E. (2010), Metodyka wraz z Przykładowym Obliczeniem „Limitu” Krajowej Emisji Gazów Cieplarnianych dla Polski na lata 2013- 2020 (Dyrektywa EU ETS i Decyzja NON-ETS) [Methodology and calculation of country’s GHG emissions limit for Poland 2013-2020], KASHUE-KOBiZE [National Administration of the Emissions Trading Scheme- National Center for Emission Balancing and Management]. US National Academy of Sciences (2008), Understanding and Responding to Climate Change, Washington D.C., National Academy Press. Weitzman, Martin L., (1974), “Prices vs. Quantities,” Review of Economic Studies (XLI, 477-91, October). World Bank (2009), World Development Report 2010: Development and Climate Change, Washington DC. World Bank (2010), Lights Out? The Outlook for Energy in Eastern Europe and the Former Soviet Union, Europe and Central Asia Region, Washington DC. Figures Figure 1. Global annual emissions of greenhouse gases 25 Figure 2. World’s largest greenhouse gas emitters, 2005, in percent 30 Figure 3. CO2 intensity of energy use in Poland and EU27 31 Figure 4. Energy intensity across countries, 2007 (toe/M€) 31 Figure 5. Carbon intensity in Central and Eastern Europe and Central Asia, 2005 31 Figure 6. Economic growth and GHG emissions in Poland, 1988-2008 32 Figure 7. GHG emissions by gas, 2007 33 Figure 8. GHG emissions by sector, 2007 33 Figure 9. Energy consumption by fuel, 2007 33 Figure 10. Electricity generation by fuel, 2007 33 Figure 11. Energy intensity in EU27 and Poland, in toe/M€ 34 Figure 12. Change in GHG emissions by key sector, 1988 to 2006, in percent 34 Figure 13. GHG emissions in Poland and EU, by ETS and non-ETS sectors, %, 2005 38 Figure 14. GHG emission targets in the EU 20-20-20 package 39 Figure 15. EU-wide and Poland’s 2020 targets, ETS and non-ETS sectors, MtCO2e and % vs. 2005 40 Figure 16. Deadweight loss in emission markets 42 Figure 17. Poland’s historical GHG emissions and EU-wide 2020 target 43 Figure 18. Share of renewable energy sources in final energy consumption 43 Figure 19. Share of renewable energy sources in gross inland energy consumption, 2007 43 Figure 20. Model suite for low-emissions growth assessment for Poland 47 Figure 21. Basic drivers of GHG emissions growth 52 Figure 22. Growth in GHG emissions and GDP in 2008 in Central and Eastern Europe 53 Figure 23. MicroMAC curve BAU scenario emissions growth 54 Figure 24. MEMO BAU projections for Poland 55 Figure 25. MEMO BAU value added by sector, 2005-2030, % 56 Figure 26. GHG emissions and MEMO BAU scenario 2020/2030 56 Figure 27. Comparing economy-wide BAU scenarios 2020/2030 58 Figure 28. GHG emissions in Poland, in MtCO2e and %, 2005 and 2020 59 Figure 29. Microeconomic Marginal Abatement Cost (MicroMAC) curve for Poland , 2030 63 Figure 30. MicroMAC curve: abatement potential for Poland in 2030 by groups of interventions 64 Figure 31. MicroMAC curve: investment and operational cost savings, 2010-2030 65 Figure 32. Historical emissions in Poland and under MEMO low carbon scenario 70 Figure 33. Decomposition of abatement by micro-package 74 Figure 34. Decomposition of GDP impact of low carbon package by micro-package 75 Figure 35. MEMO model: Macroeconomic Abatement Cost (MacroAC) curve, 2020 and 2030 76 Figure 36. Macroeconomic Marginal Abatement Cost (MacroMAC) curve, 2020 and 2030 78 Figure 37. ‘Main’ scenario : carbon emissions, % change vs. 2005 87 Figure 38. ‘Main’ scenario 2020: carbon prices 87 Figure 39. ‘Main’ scenario 2020: macroeconomic variables in Poland and the EU 87 Figure 40. ‘Where-flexibility’ scenarios: carbon emissions, % change vs. 2005 88 Figure 41. ‘Where-flexibility’ scenarios: carbon prices 89 Figure 42. ‘Where-flexibility’ scenarios: macroeconomic variables, in % vs. BAU 89 Figure 43. ‘Renewables target’ scenario: carbon emissions, % change vs. 2005 89 Figure 44. ‘Renewables target’ scenario: carbon prices 90 Figure 45. ‘Renewables target’ scenario: macroeconomic variables, in % vs. BAU 90 Figure 46. ‘Wage subsidy’ scenario: carbon emissions, % change vs. 2005 91 Figure 47. ‘Wage subsidy’ scenario: carbon prices, in US$ per tCO2 91 Figure 48. ‘Wage subsidy’ scenario: macroeconomic variables, in % vs. BAU 91 Figure 49. ROCA model: electricity generation mix, in % 92 Figure 50. ‘Technology constraint’ scenarios: carbon emissions, % change vs. 2005 93 Figure 51. ‘Technology constraint’ scenarios: carbon prices 93 Figure 52. ‘Technology constraint’ scenarios: macroeconomic variables, in % vs. BAU 93 Figure 53. ‘Competitiveness risk’ scenarios: carbon emissions, % change vs. 2005 94 Figure 54. ‘Competitiveness risk’ scenarios: carbon prices, in US$ per tCO2 94 Figure 55. ‘Competitiveness risk’ scenarios: macroeconomic variables, in % vs. BAU 94 Figure 56. ‘Power options and trade effects’ scenarios: carbon emissions, % change vs. 2005 96 Figure 57. ‘Power options and trade effects’ scenarios: carbon prices, in US$ per tCO2 96 Figure 58. ‘Power options and trade effects’ scenarios: macroeconomic variables, in % vs. BAU 96 Figure 59. Microeconomic Marginal Abatement Cost (MicroMAC) curve for low-carbon energy sector investments 100 Figure 60. Current and projected electricity mix in Poland, 2020 and 2030 102 Figure 61. ROCA model: electricity generation mix, in % 103 Figure 62. MEMO model: optimal energy sector scenario 105 Figure 63. Comparison of alternative energy investment scenarios 106 Figure 65. Relation between investment cost and GDP elasticity of energy packages 107 Figure 66. Emission abatement in the energy sector and the required public subsidy in total capital expenditures (CAPEX) 107 Figure 67. Microeconomic Marginal Abatement Cost (MicroMAC) curve for energy efficiency levers 111 Figure 68. Macroeconomic Marginal Abatement cost (MacroMAC) curve for energy and fuel efficiency micro-packages, 2020 and 2030 112 Figure 69. Transport CO2 emissions in Poland by mode, 2006 117 Figure 70. TREMOVE Plus BAU carbon emissions scenario for Poland’s road transport sector, 1990-2030 120 Figure 71. TREMOVE Plus carbon emissions projections for road transport by scenario 122 Figure 72. Illustration of EU’s 2020 climate change and energy targets 128 Figure 73. A myriad of policy instruments in the EU 2020 package 134 Figure 74. Quantity target and price target under certainty 136 Figure 75. Quantity target and price target under uncertai 137 Figure 76. MEMO model block structure 140 Figure 77. MEMO model: production structure 142 Figure 78. Diagrammatic overview of ROCA model framework 146 Figure 79. Calibration to exogenous emission projections 149 Figure 80. MEMO model BAU estimation procedure 152 Figure 81. MEMO BAU projection of energy intensity in Poland 153 Figure 82. MEMO BAU projection of GHG emission intensities in Poland 154 Tables Table 1. Poland’s greenhouse gas emissions, 1988, 2000, and 200 32 Table 2. Breakdown of EU 20-20-20 regulations by sector groups 37 Table 3. Nominal and effective emission reduction targets for 2020 for Poland and the EU, in % 57 Table 4. Macroeconomic and fiscal impact of GHG abatement package, deviation from BAU, in % 72 Table 5. Decomposition of the macroeconomic impact of GHG abatement package, deviation from BAU, in % 73 Table 6. Decomposition of abatement by micro-package, reduction relative to BAU, in % 73 Table 7. Decomposition of GDP impact by micro-package, deviation of real GDP from BAU, in % 74 Table 8. Summary of scenario characteristics simulated in the ROCA model 83 Table 9. ROCA model: economic impacts of alternative emission mitigation scenarios 85 Table 10. Key features of energy sector technologies available in Poland until 2030 101 Table 11. Basic economic features of individual energy investment levers 104 Table 12. Macroeconomic impact of the optimal energy mix scenario: sensitivity analysis 108 Table 13. EU sustainable transport policy measures 117 Table 14. Overview of the business-as-usual vehicle stock and mobility indicators 118 Table 15. TREMOVE Plus BAU scenario of Poland road transport CO2 emissions, 1995-2030 119 Table 16. Overview of TREMOVE Plus low carbon scenario policy measures 121 Table 17. TREMOVE Plus ‘Precautionary’ scenario emissions reduction by policy intervention, MtCO2e 121 Table 18. TREMOVE Plus ‘Proactive’ scenario emissions reduction by policy intervention, MtCO2e 122 Table 19. Comparing the Kyoto Protocol and EU 20-20-20 policy package 129 Table 21. EU ETS Phasing-in Process 134 Table 22. ROCA model sectors and regions 148 Table 23. MEMO BAU estimated convergence rates, EU21 during 1996-2006 153 Table 24. MEMO model: sensitivity analysis of twenty percent deviation from optimal scenario 165 Table 25. Macroeconomic impact of alternative energy packages 167 Boxes Box 1. The Intergovernmental Panel on Climate Change’s (IPCC) Fourth Assessment Report 25 Box 2. The Kyoto Protocol of the UNFCCC 36 Box 3. Excess costs of emission market segmentation 42 Box 4. Poland’s growth projections and the global financial crisis 53 Box 5. How do the bottom-up abatement opportunities work in the top-down model? 69 Box 6. Public financing options or ‘closures’ in the MEMO model 70 Box 7. The impact on emissions of Poland’s urban transport and urban planning policies 123 Pre-press by JURCZYK DESIGN. www.jurczykdesign.com This study poses the question of how Poland can tran- sition to a low emissions economy as successfully as it underwent transition to a market economy in the early 1990s. With the EU policies on climate change and 2020 targets already in place, Poland faces immediate policy challenges. What are the implications for Poland of imple- menting EU policies on energy and climate change? Could the country commit to more ambitious overall greenhouse gas mitigation targets for the longer term—to 2030 and beyond? What technological options are available, and how expensive are they? Would there be high costs in lost growth and employment? The report addresses these questions while advancing the methodological approach of the World Bank’s low carbon studies by integrating ‘bot- tom-up’ engineering analysis with ‘top-down’ economy- wide modeling. The economic impact is presented using a unique macroeconomic version of the well-known mar- ginal abatement cost curve. Find this report and related materials at: www.worldbank.org/pl/lowemissionseconomy