94187 Armenia Power Sector Policy Note THE WORLD BANK December 2014 Armenia Power Sector Policy Note Artur Kochnakyan Brendan Larkin-Connolly Denzel Hankinson Ani Balabanyan Matteo Morgandi Anne Olivier 2 Armenia Power Sector Policy Note ©2014 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 All rights reserved This report 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 volume 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 report 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. For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright Clearance Centre Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 202-522- 2422; e-mail: pubrights@worldbank.org. 3 Foreword This presentation was prepared as part of the World Bank’s Power Sector Policy Note for Armenia (the Note). The objectives of the Note is to inform the Government’s policy thinking by identifying the principal challenges that the power sector faces and outlining solutions for overcoming them. The Note also discusses some broader energy sector issues related to the gas tariff structure and demand. The Note will also be disseminated to increase understanding and awareness among key stakeholders and the general public on the key challenges facing the power sector and the potential solutions for overcoming them. This will help to promote improved dialogue and collaboration between the Government and the other key stakeholders. The Note was prepared based on the data generated by the relevant energy companies in Armenia, the Public Service Regulatory Commission, the World Bank internal data bases, and discussions with the Ministry of Energy and Natural Resources and Power System Operator. 4 Synopsis of Main Challenges and Solutions Highest-cost and dilapidated Hrazdan TPP will need to run post 2016 to avoid supply gap because new lower cost CCGT cannot be realistically constructed earlier than 2020. Several transmission assets are a threat to supply reliability. Many transmission lines and substations incur high outage rates, which could lead to system-wide failure. Affordability is a growing concern. Climbing energy costs increased the share of household energy expenses to 10%. It will get worse as the much needed new investments are made. Deterioration of governance and financial standing of state power companies. In recent years tariffs were frequently lagging cost-recovery, the financial management decisions were not always prudent, and a key generation asset was sold through a direct negotiated sale. In order to address the above challenges the Government needs to: Add new generation: Immediately focus public and private financial and implementation capacity on a new gas-fired generation unit of around 500 MW. Revise tariff structure: Remove perverse tariff incentives that are accelerating winter electricity consumption and unnecessary gas consumption by public facilities and commercial establishments. Rehabilitate key transmission assets: Rehabilitate key substation and transmission lines critical for system-wide reliability of power supply. Implement financial recovery plan: Design and implement financial recovery plan for state-owned power companies, including consistent application of cost recovery tariffs. Protect the poor: Top up social assistance program to make basic level of consumption affordable 5 Table of Contents • Power Sector Structure • Synopsis of Challenges and Solutions • Principal Challenges and Solutions  Supply Adequacy  Supply Security  Affordability  Governance • Annexes 6 Overview: Power Sector Structure • The sector is fully unbundled • The distribution company is the single buyer • The sector is regulated by independent and • The Ministry of Energy and Natural Resources competent regulator develops and implements energy policy Dispatching National Dispatch Center Commercial Metering Settlement Center Public Services Generation Regulatory Medzamor NPP Thermal Plants Hydro Plants Commission (Tariffs, Service Quality and Transmission High Voltage Grid Licensing) Distribution Electricity Network of Export/Import Armenia (ENA) Flow of power/services Flow of funds End-Users Customers 7 Generation capacity to meet forecast electricity demand Challenges Suboptimal electricity and gas tariff structures Build new capacity Solutions Scale up energy efficiency Improve tariff structure Financing options study, feasibility study and ESIA for a new Combined Cycle Gas Turbine station Develop Loriberd and Shnogh Key Next Steps Study on energy efficiency potential Introduce marginal cost based tariffs Study on gas tariff structure 8 Challenge #1: Supply Adequacy To avoid supply gap, highest-cost Hrazdan TPP will need to run until new capacity is built Obsolete and expensive Hrazdan TPP may  All units of Hrazdan TPP will be past their operating lives by 2016 need to be run until 2020 to avoid supply gap  Supply reliability may be jeopardized given disrepair and obsolescence of Hrazdan TPP  Hrazdan TPP uses 40% more gas per 1 kWH compared to a new CCGT Generation tariffs, VAT inclusive (AMD/kWh) 2014 Hrazdan-5 40.1 Yerevan CCGT 34.9 Hrazdan TPP 60.0 Sevan-Hrazdan 8.6 Vorotan 9.4 ANPP 13.7 Small HPPs 25.3  550 MW Yerevan TPP is retired  Only 50% of 800 MW Hrazdan TPP is available to meet winter peaks Effective from August 1, 2014  Only 45% of hydro capacity is available to meet winter peaks 9 Challenge #1: Supply Adequacy Current electricity tariff structure promotes over-consumption in winter, contributing to high peak Deviations from Marginal Cost Based Pricing Principles: Seasonality. No seasonal tariff, although winter marginal costs are significantly higher than summer. Time of use. Differential between peak and off-peak, or day and night, tariffs does not reflect the difference in the marginal cost of service. Fixed charges. No fixed component in the monthly bill despite significant customer-related (and demand- related) costs of service, which are not driven by kWh consumed. Voltage levels. Allocation of revenue requirement to different customer classes does not reflect differences in the marginal cost of serving different voltage levels. 10 Challenge #1: Supply Adequacy Current gas tariff structure promotes inefficient consumption – Volume-based tariff encourages high consumption  Small volume consumers pay more (156 AMD/m3) than large volume consumers (~ 114 AMD/m3) for all units consumed  Perverse incentive for customers near the edge of the first block (hospitals, schools, SMEs) to over-consume to obtain the low wholesale price. Several public facilities are not interested in the WB/GEF Energy Efficiency Project given such tariff structure. – Single-part tariff discourages utility from measures that reduce consumption (no fixed monthly charge and variable energy tariff) 11 Solutions: Supply Adequacy: Supply Plans Analyzed A mix of nuclear, natural gas, and renewable energy supply plans were analyzed Supply Plan Summary Gas All capacity needs met by new gas plants Gas + RE Gas plants + 141 MW Loriberd and Shnogh HPPs by 2023 + 540 MW in PV, wind, geothermal by 2030 Large NPP Large NPP (VVER-440) by 2026, with any additional capacity needs met by new gas plants Large NPP + RE Large NPP + 141 MW Loriberd and Shnogh HPPs by 2023 + 540 MW in PV, wind, geothermal by 2030  VVER has net capacity of 1000 MW  Renewable energy (RE) capacity is based on GoA targets for penetration of RE  Under each scenario EE savings are assumed to grow by 5 MW per hour each year from 2015 to 2024, reaching 50 MW per hour in savings by 2024  260 MW of existing Small HPP capacity assumed for all scenarios 12 Solutions: Supply Adequacy: Least Cost Plan (LCP) And Demand Sensitivity  “Gas” is the LCP under the Base-case  “Gas” remains the LCP under Low and High power demand scenarios “Large NPP” is more expensive under Base-case “Large NPP” is less expensive under demand given high capital cost and over-capacity High-case demand Economic NPV (in Million US$) Supply Plan Base Low High Demand Demand Demand Gas 1,728 1,374 2,123 Gas + RE 2,500 2,297 2,793 Large NPP 3,341 3,015 3,639 Large NPP + 4,271 4,134 4,674 RE Long-Run Average Incremental Cost (LRAIC) by Demand Scenario Supply plans with “RE” are more expensive given that most of the RE projects considered have high capital costs and low capacity factors Base-case: Base-case power demand, base-case gas prices based on the formula for pricing of Russian gas imports , and current volume of swap with Iran 13 Solutions: Supply Adequacy: Sensitivity Analysis of LCP “Gas” is the LCP even in case of High gas prices LRAIC of Supply Plans under Various Gas Prices LCP is more sensitive to gas price changes given that it is comprised of gas plants only. Other scenarios are less sensitive given RE and Large NPP ∆P(imports) = 0.65 x ∆P(Orenburg price) + 0.35 x ∆US-CPI  Wholesale gas prices in Orenburg region* will have the largest impact on Armenian gas costs for power sector given the import gas pricing for Armenia Real tariff (US$) 2014 2015 2016 2017 2018 2020 2022 2024 2026 2028 2030 Large consumers (Base-case) $/tcm 239 241 249 258 268 287 308 331 353 370 387 Large consumers (High-case) $/tcm 239 241 254 267 282 313 341 372 401 424 447 Large consumers (Low-case) $/tcm 239 240 244 249 253 262 276 292 312 333 356 • Wholesale gas price scenarios are based on “Forecast of Social-Economic Development of Russia until 2030” released by Russia’s Ministry of Economic Development; • Current Gas Purchase Agreement with Russia is assumed to be extended post-2019 with the same pricing formula 14 Solutions: Supply Adequacy: Renewables  Shongh and Loriberd HPPs should be part of LCP  Geothermal may be be part of LCP if >250o resource is found  Loriberd can be developed as peaking plant adding 66 MW of firm capacity  Geothermal is low-cost base load unlike several other RE technologies LEC of Renewable Energy Technology Unrealized Generation potential (MW) (GWh/yr) Distributed solar PV 1,300 1,800 Concentrating solar power (CSP) 1,200 2,400 Utility scale solar PV 830 – 1,200a 1,700 – 2,100a Wind 300 650 Geothermal power* at least 150 at least 1,100 Small hydropower 100 340 Shnogh HPP 75 300 Loriberd HPP 66 205 Biomass 30 230 Biogas 5 30 Landfill gas 2 20 * For 5 sites for which information is available 15 Solutions: Supply Adequacy: Gas Pipeline Capacity  Total pipeline capacity is sufficient to supply all of the gas required for LCP, however,  LCP will be feasible only if Armenia imports more gas from Iran Combined capacity of North- No spare capacity on North-South pipeline in South and Armenia-Iran pipelines 2025 if imports from Iran are not increased Capacity of North-South pipeline Current Imports from Iran and Russia Currently Armenia receives only 0.4 bcm of gas from Iran under the swap deal whereas the capacity of the Iran-Armenia pipeline is 2.3 bcm/year 16 Solutions: Supply Adequacy: Nuclear Portfolios Medium NPP may be a lower cost compared to Large NPP Long-Run Average Incremental Cost by Demand Scenario Medium NPP may have a cost between LCP and Large NPP  Medium NPP may have small or no surplus capacity, thus, lower energy cost compared to Large NPP  Medium NPP may have lower per kW capital cost. Even with the same per kW capital cost as Large NPP, it will still be lower cost due to smaller or no capacity surplus Detailed feasibility study will be required to assess technical and economic viability of Medium NPP if the GoA decides to pursue it 17 Solutions: Supply Adequacy: New Plants Needed  Two new gas plants needed under LCP: 500 MW by 2020 and 500 MW by 2026  One new plant is needed by 2020 under all other supply plans 2nd 500 MW Gas Plant Gas + RE: 345 MW plant by 2020 1st 500 MW Gas Plant Large NPP: 500 MW plant by 2020 Large NPP + RE: 345 MW plant by 2020 Assumptions:  ANPP off-line in 2016 for rehab, on-line again in 2017, Capacity retired in 2026 to meet  Reserve margin (RM) = 300 Peak: MW LCP w/  Capacity factor of HPPs based on actual supply Base during 2013 peak Case Demand 18 Solutions: Supply Adequacy: Construction of New CCGT To commission a new gas-fired CCGT by 2020, the GoA needs to start the project now For the new gas-fired CCGT to be on-line by 2020, the GoA will need to stick to below schedule: Key Steps Duration Schedule * Decide on financing option 6 months February 2015 Select a Consultant for a feasibility study, ESIA and 7 months March 2015 preparation of bidding documents Complete the feasibility study and preparation of 12 months March 2016 bidding documents Complete financial structuring and construction 44 months November 2019 Test and commission the CCGT 2 months January 2020 * The schedule assumes project preparation activities start in September 2014 19 Solutions: Supply Adequacy: Financing New Investments LCP can be financed entirely with public debt, however…  Entirely public financing is not a sustainable strategy given large needs in other sectors  New CCGTs, Loriberd and Shnogh HPPs, and other RE are good candidates for PPPs 60% - Legal limit of debt/previous year GDP Supply Scenario Capital cost (billion $) The public debt assumed for Gas 1.2 “Gas + RE” or Large NPP supply Gas + RE 2.3 plans will breach the limit Large NPP 6.1 Large NPP + RE 7.3 All RE projects are planned as private and not included in calculation of public debt impact 20 Solutions: Supply Adequacy: Energy Efficiency Energy efficiency can help meeting forecast power demand in a least-cost way Key next steps :  Conduct a comprehensive study to estimate economically and financially viable energy efficiency potential in the country. There are no reliable estimates of sectoral energy efficiency potential  Adopt two-part seasonal and improved time-of-day electricity tariff structure  Eliminate structural flaws in the gas tariff, which create a perverse incentive for some end-users to increase gas consumption to lower bills  Mandate energy efficiency in new construction projects and all facility renovations programs financed by the Government and donors.  Provide capital grants to the poor for energy efficiency retrofits Actual energy savings achieved under WB/GEF project Type of Facility Energy Savings Payback period Cost of electricity savings = AMD (%) (years) 19/kWh or 40% of unit cost under Schools 51-52% 6 LCP Hospitals 51% 5 Street Lighting 45-53% 4-5 Other facilities 41% 5 21 Solutions: Supply Adequacy: Improve Tariff Structure Introduce marginal cost based electricity tariff to reduce the need for new supply Two-part seasonal tariff with improved time-of-day bands may reduce peaks and need for new generation Electricity Two part tariff structure reflects the fact that Per kWh charge is differentiated by season and time- many costs are not linked to kWh sold: of-use to better reflect how incremental costs are incurred:  Per kWh tariff reflects energy and capacity costs  Winter Peak: 8:01 PM -12:00 AM, Sep – Feb  Winter Off-peak: 12:01 AM - 8:00 AM, Sep- Feb  Fixed monthly charge reflects customer-related costs.  Summer Peak: 8:01 PM - 12:00 AM, Mar - Aug  Summer Off-peak: 12:01 AM - 8:00 AM, Mar – Aug  Per kWh tariff differentiated by season and time-of-use Gas Conduct a study to determine:  Marginal cost of supply to each category of consumers  Allocate revenue requirement among classes according to marginal costs 22 Regional Trade: Short-Term Limited regional connectivity and short-term trade opportunities  One 220 kV and two 110 kV lines with Georgia allow for trading (up to 400 MW) in island operation. However, trading has decreased almost to zero in 2011-2013  Two 220 kV lines with Iran allow for significant trading (around 400 MW). Trading primarily occurs under the electricity-for-gas swap deal Export potential is limited in the short-term  Excess short-term summer capacity and fully depreciated generation assets may make summer exports attractive, however,  Exports to Turkey are not possible given unresolved political issues  Georgia does not require energy in summer  No relationship with Azerbaijan, which could have imported summer energy depending on its gas price policy for the energy sector  Summer energy surplus in Armenia will significantly reduce as gas-for-electricity swap expands with Iran Short-term average Current export tariffs  In the short-term, winter exports are not feasible costs (US$/kWh)* (US$/kWh)** given: Armenia 0.056 No exports  Expensive gas-fired generation cannot compete Georgia 0.020 0.020-0.030*** with thermal exports from Azerbaijan Azerbaijan 0.030 0.040-0.045 Imports are not realistic given capacity balances in Turkey 0.090 - neighboring countries and political issues Iran 0.050 - * Bank team estimate; ** Platt’s energy and national public sources; *** Such export tariffs are possible because summer HPP generation costs are below $0.01/kWh 23 Regional Trade: Long-Term Limited long-term export opportunities  In the long-term, summer and winter exports are not likely to be competitive because: • No excess capacity at existing hydropower plants due to increase in domestic demand • Absence of large low-cost untapped hydropower potential • New gas or nuclear capacity is not likely to be cost-competitive for exports Ongoing and planned regional interconnection projects: LRAIC  $140 million 400 kV line from Hrazdan TPP to Iran (2017). This 1000 MW line (US$/kWh)* may be used to expand the swap, which is very beneficial for Armenia under Armenia 0.12-0.17 current terms Georgia 0.07-0.08  $150 million 400 kV line to Georgia (2020). The Government needs to carefully Azerbaijan 0.04-0.13 assess the economic rationale for the project because:  No power trade with Georgia even with existing 220 kV line Turkey 0.09-0.13  Limited opportunities for long-term power exports through Georgia: Iran 0.06-0.12 - Georgia has cheap hydropower surplus itself * Bank team estimates - Georgian transmission interconnections are not sufficient to wheel Armenian power  Transit of Georgian power to Iran through 400 kV line to Iran may be considered Notes: The LRAICs will largely depend: Georgia: Large HPPs developed and import prices of gas; Azerbaijan: Price of gas for the power sector; Iran: Price of gas for the power sector; Armenia: Gas or new NPP to meet long-term forecast demand 24 Dilapidated critical power transmission lines Challenges Dilapidated critical substations Prioritize investments Solutions Rehabilitate existing assets Adopt “Risk Index” for prioritization of transmission investments Key Next Steps Secure financing for rehabilitation of critical transmission lines and substations 25 Challenge #2: Supply Security Outages per transmission line 2.5 times higher than the average for well-performing utilities* * An average of a number of US and European utilities 26 Challenge #2: Supply Security  Poor condition of critical power transmission assets jeopardizes supply security  HVEN does not have a methodology for prioritizing transmission investments 1. Poor condition of several critical transmission assets due to age (>45 years) and absence of investments in: - Substations used to evacuate power from large generators (e.g. substation of ANPP) - Critical power transmission lines evacuating power from large generators (e.g. Echmiadsin 110 kV OTLs from ANPP) 2. Economic costs of bottlenecks include: - Suboptimal dispatch: Additional cost incurred due to the failure of a transmission asset that forces generation resources to be dispatched in a sub-optimal way - Emergency repairs: Additional costs incurred by the need for unplanned emergency repairs and replacements when assets unexpectedly fail. 3. Lack of formal methodology for prioritizing transmission investments leads to: (a) ad- hoc decision-making on investments, and (b) difficulty to secure much needed investments 27 Solutions: Supply Reliability: Prioritization of Investments Develop/adopt a methodology for prioritizing transmission investments • Long-term: Full scale risk assessment based on detailed, component-level information on condition of each asset in the system. Major engineering undertaking. • Short-term: Basic risk assessment based on available data. A “Risk Index” can be created based on assessment of:  Consquences of Failure (CoF): a higher score means a greater impact on end-users  Probability of Failure (PoF): a higher score means higher probability of failure, based on asset-specific information about: ‐ The age of the asset relative to its useful life ‐ The number of outages experience by that asset over the past five years, and ‐ The duration of outages experienced by that asset over the past five year 28 Solutions: Supply Reliability: Risk Index Risk Index (RI) can be adapted to available data to prioritize investments The Risk Index is calculated as the estimated product of the Probability of Failure (POF) and Consequences of Failure (CoF): 1. PoF is computed by scoring each asset on the average of three metrics: − % of life remaining − Number of outages in 2008-2013 − Duration of outages in 2008-2013 2. COF is a quantitative measure of the consequences of an asset’s failure for the system CoF x PoF = Risk Index Asset Category Score % of remaining Score # of outages Score Duration of Score Substations at generators 5 life outages 51+ 5 (minutes) Primary lines from 4 <20% 5 1,441+ 5 generators 21-40% 4 31-50 4 61-1,440 4 Secondary lines from 3 41-60% 3 11-30 3 generators 11-60 3 61-80% 2 1-10 2 Other 220 kV lines or 2 1-10 2 >80% 1 0 1 substations 0 1 Other 110 kV lines or 1 substations 29 Solutions: Supply Reliability:110 kV Lines  7 lines have high rehab priority with total investment cost of $16 million  Those lines serve generation units with a failure likely to cause system wide outages Echmiadsin and Line Priority Rank Length (km) Cost Financing Status Shahumyan-2 Sevan 2 10 $1.0 Not secured lines are critical for reliability Echmiadsin 3 24 $3.3 Not secured and safety of Shahumyan-2 5 33 $3.0 Not secured ANPP Karmir-2 6 10 $0.7 Not secured Bjni 7 25 $3.8 Not secured Bjni and Shahumyan-1 Shahumyan-1 7 14 $3.0 Not secured also serve ANPP Karmir-1 12 10 $1.0 Not secured TOTAL 126 $16 30 Solutions: Supply Reliability: Transmission Substations Substation at ANPP has the highest rehab ($20 million) priority given that a failure can cause ANPP safety issues and loss of load Substation Priority kV Financing Substation Priority kV Financing No Rank Status Rank Status rehabilitation at Metsamor Metsamor 1 110/220 Not Zovuny 10 220 PTRP substation secured Marash 10 220 PTRP since 1977 Yerevan 1 110/220 ETNIP Charentsavan 10 220 ESRP AF TPP/CCGT Ararat-2 10 220 PTRP Yerevan TPP Haghtanak 4 220 ESRP AF Ashnak 10 220 PTRP currently Lichk 8 220 PTRP relies on only Agarak-2 13 220 PTRP 110 kV part of Shahumyan-2 8 220 PTRP substation to Vanadzor-1 8 220 ETNIP  Armenia does not have connect to transmission Eghegnadzor 8 220 PTRP operational back-up dispatch grid Shinuhayr 8 220 PTRP center for emergency situations ETNIP - Electricity Transmission Network Improvement Project (WB) ESRP AF - Additional Financing for Electricity Supply Reliability Project (WB) PTRP – Power Transmission Rehabilitation Project (ADB) 31 Increasing electricity and Challenge gas tariffs Provide targeted assistance Solutions through PFBP Decide on power and gas subsidy option Improve targeting through Key Next Steps better proxy-means testing Scale up energy efficiency in residential sector 32 Challenge #3: Affordability  Energy spending as % of total is estimated at 10% – a level considered to be energy poverty  2013 increase of gas and electricity tariffs is estimated to have increased poverty by 3% Energy expense share increase was highest for the poor – 13.6% Without gas subsidy provided by the Government since 2011, the increase of the energy share of expenditure for the poor would have been 15.7% Since April 2011, families registered in PFBP with poverty score above zero paid a reduced tariff of AMD100/m3 for first 300 m3 of consumption *Energy/electricity poverty refer to households spending more than 10% of their budgets on energy or electricity. 33 Challenge #3: Affordability Share of gas-based heating is reducing, while wood-based heating is increasing - Gas based heating may reduce further given significant gas tariff increase in 2013 - Electricity based heating cannot substitute gas given it is 2x more expensive - Increased use of wood for heating has long-term negative consequences for: (a) Public health given higher likelihood of poisonings and accidents with custom-made heaters (a) Forests covering only 9% of Armenia’s territory 34 Challenge #3: Affordability: Results of Qualitative Survey Poor families struggle to ensure adequate heating Results of a qualitative heating survey conducted during 2013/2014 heating seasons suggest that: - FBP recipients reported substituting gas with wood so to avoid consuming >300 m 3 - threshold for FBP eligibility. The costs of energy use for these households may be underestimated if wood costs are not considered. - FBP recipients reported using the total amount of assistance on energy during the heating season. - Poor households reported cutting down on expenses for healthcare and children’s education to cope with energy price increases. Some also report burning old furniture and clothing when they cannot afford wood. - Manure is a last resort source of energy that is increasingly used in rural areas. - Poor households reported that they could no longer reduce their energy use without major impact on their health 35 Challenge #3: Affordability Large tariff increases will be needed:  In 2016, because highest-cost Hrazdan TPP will need to be run to replace ANPP for a year  In 2020, under all supply plans given the cost of new gas CCGT  In 2026, under all supply plans given the cost of new generation Residential tariffs will be the lowest under Least-Cost Plan (LCP) 2013 average tariff Note: All tariffs are in 2013 prices Note: Tariffs were computed using marginal cost based methodology; swap with Iran was included in the forecast of tariffs 36 Challenge #3: Affordability LCP will have the lowest impact on poverty In 2016, tariffs will increase poverty by 2.8% under all investment scenarios given that expensive Hrazdan TPP will need to replace ANPP for a year However, in 2026, LCP (Gas) will increase poverty only by 0.7%, whereas “Large NPP+RE” by 3.2% Note: Tariffs were calculated assuming commercial financing for all investments given that public financing is not realistic from public debt sustainability perspective Note: Impacts of all other scenarios fall in between “Gas” (LCP) and “Large NPP + RE” (highest -cost) 37 Solutions: Affordability: Trade-Offs Involved Coverage, targeting and fiscal trade-offs are involved when selecting subsidy delivery mechanism Advantages Disadvantages Lifeline tariff • Fairly good targeting (53% of • Limited coverage - only 16% of poor and through existing beneficiary households are poor) 10% of population PFBP • Low administrative costs • No incentives for energy efficiency and • Public perception of specific conservation measure to protect the poor • No rationale to deliver the benefit through bill in case of electricity, Additional cash • No additional administrative costs • Limited coverage of poor (16%) benefit to • More efficient than in-kind subsidy • Increased horizontal disparity between recipients of PFBP of electricity covered and non-covered poor by PFBP • May be perceived as a benefit increase, not energy-specific (needs appropriate communication) Cash transfer to  Broader coverage than PFBP (14% • Unknown targeting performance, likely all in database of population in 2012) worse than PFBP with score >0 38 Solutions: Affordability: Overview of Options Affordability for the poor will depend on methodology for quantifying the benefit Objective Description Annual cost per adult- Average annual equivalent benefit Option 1. Mitigation Benefit is calculated to allow 5,880 AMD 19,400 AMD of each tariff increase households to maintain same level of (9,500 elect + 9,900 (2013 used in consumption despite price increase. (2,880 gas gas) analyses) & 3,000 elect) Does not attempt to mitigate existing (5% of mean annual energy deprivation FBP transfer) Option 2. Closing gap Amount needed to 10,000 AMD 32,600 AMD in consumption close the gap between the median (9,900 elec. + 22,700 between Q1 and Q2 spending on energy of 2nd lowest and (5,000 gas gas) the lowest quintile, assuming Q2 is & 3,000 elect) not as energy deprived (9% of mean annual PFBP transfer) Option 3. Meeting Benefit is calculated depending on Requires further research into standardized levels minimum energy number of household residents, size of energy consumption needs of house, region, etc. 39 Solutions: Affordability: Short-term Options & Fiscal Costs With existing targeting mechanism only 16-25% of the poor may be covered at an annual cost of 0.05%-0.08% of annual GDP Option 1 (tariff impact mitigation) & Option 2 (closing gap between Q1 and Q2) would cost the same for electricity, but Option 2 would be more costly for gas. Coverage and expenditure of compensation with Options 1 and 2 30% 0.06% 0.05% 25% 0.05% % of the poor covered 0.04% % of 2012 GDP 20% 0.04% 0.03% 15% 0.03% 0.02% 0.02% 10% 0.02% 0.02% 0.02% 0.01% 5% 0.01% 0% 0.00% Targeted gas Cash to FBP Cash to all Targeted gas Cash to FBP Cash to all Cash to FBP Cash to all lifeline tariff beneficiary reg, score >0 lifeline tariff beneficiary reg, score >0 beneficiary reg, score >0 Gas (option 1) Gas (option 2) Electricity (option 1 or 2) 40 Solutions: Affordability: Medium-term Compensation Options & Fiscal Costs Broadening coverage to more households will require an updated targeting system Improving targeting will require: - A new Proxy Means Testing (PMT) formula that uses information on household composition and characteristics, traceable expenditure and incomes, to derive a probability of households being at specific income threshold - Extensive piloting in welfare centers, and close monitoring during implementation Targeting accuracy of improved PMT is higher Covering 40% of poor with new PMT cost 0.09% of GDP, compared to 0.07% of GDP to cover 21% of the poor under existing targeting mechanism of PFBP 41 Imprudent financial management of state power companies Tariffs frequently lagging cost-recovery Challenges Power sector assets sold through direct negotiations Public communication requires improvement Design and implement a financial recovery plan for state-owned power companies Consistently maintain tariffs at full cost- recovery levels Solutions Sell power sector assets through competitive tenders Improve communication of sector issues to key stakeholders and general public 42 Challenge #4: Deteriorating Governance: Financial Mismanagement The state-owned power companies are in financial distress: AMD24 billions of debts and payables (27% of their total revenue) The financial standing of state-owned power companies deteriorated due to: Low tariffs: The Government tried to mitigate the impact of gas price increases on end-user electricity tariffs by significantly reducing the O&M expenses and by virtually eliminating the profit and depreciation allowed in the tariffs. “Life support” to non-functional chemical plants: The Government used the funds of the state- owned power companies to finance salaries of the Nairit and Vanadzor chemical plants. The total debt of those chemical plants to the power sector is estimated at AMD22 billion. Financing of the gas subsidy for the poor: The state power sector companies were mandated to finance the ADM1.1 billion gas life-line subsidy to the poor. Financing of non-core business assets: The power sector carries on its balance and pays for operation and capital renovation of a large center used for official government receptions. 43 Challenge #4: Deteriorating Governance: Tariffs In recent years, tariffs were frequently lagging cost recovery In 2009-2013, tariffs were lagging cost recovery because of:  0 or reduced depreciation and return on assets for state-owned companies (e.g. HVEN, Vorotan, Yerevan TPP) Tariffs (AMD/kWh), VAT inclusive 1-Apr-09 1-Oct-09 1-Apr-10 1-Apr-11 1-Apr-12 7-Jul-13 1-Aug-14 HVEN 1.1 0.4 0.9 1.0 0.4 1.3 1.6 Vorotan 1.7 1.2 2.2 5.3 5.7 9.5 9.4  Negligible contributions to ANPP decomissioning fund. Around $7 million is accumulated in ANPP decommissioning fund whereas around $350 million* will be required starting from 2026  Reducing margin of the distribution company (Electric Networks of Armenia, ENA) due to:  Significant unexpected increase of costs due to longer-than-usual periodic maintenance of ANPP and the need to buy significantly more expensive replacement power from Hrazdan TPP; and  Unchanged end-user tariffs. Reduction of the margin had negative financial implications for ENA and may jeopardize reliability of supply :  ENA capital investments reduced from AMD 28 billion in 2009 to AMD 8 billion in 2012  ENA net profit margin reduced from 3.2% in 2010 to negative 9.2% in 2012 * USAID estimate, 2008 44 Challenge #4: Deteriorating Governance: Sale of Vorotan and Financial Management  Direct negotiated sale of Vorotan hydropower cascade  Imprudent financial management decisions at some state-owned companies Sale of Vorotan could have been done through competitive process to maximize the benefits:  The GoA received $180 million for 404 MW Vorotan Cascade. This translates into $450,000/MW of installed capacity.  One of the GoA objectives was to attract private investments for much needed rehab. The Government had a Euro 50 million loan from KfW for rehab of Vorotan.  One of the GoA objectives was to attract experienced private operator to improve efficiency. The new private operator (Countour Global) is currently running two hydropower plants in Brazil with installed capacity of 40 MW.* Vorotan has been operated by a competent team without major incidents.  The GoA disclosed very limited information to the public Imprudent financial management decisions at some state-owned companies:  HVEN took commercial loans that were lent to Vorotan Cascade and Yerevan TPP: There were several instances when HVEN assumed short-term commercial debt that was on-lent to Vorotan Cascade and Yerevan TPP because the latter could not borrow due to week balance sheets.  HVEN was used to implement a project that was not part of its core business: HVEN was used as the implementing entity for the construction of new gas pipeline to Iran and was penalized for US$5 million in 2012 for improper tax accounting of the gas pipeline project with significant impact on its financial standing.  Huge non-business related expenses: AMD 400 million payroll of Nairit financed by Vorotan Cascade * Countour Global web-site, http://www.contourglobal.com/portfolio#, accessed on August 10, 2014 45 Challenge #4: Deteriorating Governance: Communication Fragmented public disclosure and discussion of key power sector issues  Limited dialogue or openness to discuss sector specific problems with broader stakeholders. Sufficient clarifications and information were not provided to the key stakeholders and general public on: (a) new gas agreements signed with Russia in December 2013; (b) sale of Vorotan Cascade; and (c) benefits for Armenia from electricity-gas swap deal with Iran.  Inefficient communication of the power tariff review decisions to the public. Recent electricity tariff increases resulted in false public understanding of the reasons for such increase. Limited public support for power sector projects and reforms:  Growing, vocal and strong political and social opposition to justified tariff increases  Public skepticism of ongoing and new investments aimed at improvement of reliability and adequacy of power supply 46 Solutions: Governance: Financial Recovery Plan and Competitive Sales  Design and implement a financial recovery plan for state-owned power companies  Privatize assets through open and transparent competitive tendering Design and Implement a financial recovery plan for state-owned power:  Discontinue financing of chemical plants  Discontinue the practice of using state-owned power companies to finance the gas subsidy  Settle the inter-company debts through off-sets and write-offs  Divest non-core business assets  Ensure adequate revenue for sector companies to operate and invest in capital repair and adequate O&M to ensure reliable power supply Consider open and transparent competitive tenders as the preferred approach for sale of energy assets because:  The legal and regulatory environment in the power sector is conducive for private investments  The GoA would maximize revenue from asset sales  The GoA would attract experienced operator(s)s that can improve management and efficiency 47 Solutions: Governance: Improved Communication Improvement in communication can generate public support … Key Pre-Conditions for Improved Communication:  Development of an action-oriented communication plan with consensus building objectives and goals, core messages and their positioning, key audience, the most appropriate channels in the public domain to ensure smooth implementation  Gradual public communication on the most critical issues of the sector, including long- term challenges of supply adequacy and related large investment needs, which will require tariff increases 48 49 Acronyms and Abbreviations AMD Armenian Dram LCP Least Cost Plan ANPP Armenian Nuclear Power Plant LRAIC Long-Run Average Incremental Cost BCM Billion Cubic Meters MW Megawatt CCGT Combined Cycle Gas Turbine NPP Nuclear Power Plant CNP China Nuclear Power NPV Net Present Value COF Consequence of Failure OTL Overhead Transmission Line CPI Consumer Price Index POF Probability of Failure EE Energy Efficiency PMT Proxy Means Testing ENA Electric Networks of Armenia PV Photovoltaic FBP Family Benefit Program RE Renewable Energy GDP Gross Domestic Product RM Reserve Margin GWh Gigawatt-hour ROR Run-of-River HPP Hydropower plant TCM Thousand Cubic Meters HVEN High Voltage Electric Networks TPP Thermal Power Plant kV Kilovolt WWTP Waste Water Treatment Plant kWh Kilowatt-hour 50 Annex 1: Government Objectives Outlines the GoA’s Armenian Development Energy Sector Development strategic objectives for economic growth, Strategy (ADS): 2014-2025 Strategy (2005) poverty reduction, and national security National Energy Security Concept (2013) Identifies objectives and priorities in the energy sector Objectives in the Energy Sector: i. uninterrupted supply of energy ii. satisfaction of basic needs of consumers: affordable prices, reliable energy supply and energy efficiency iii. minimize economic effects of importing energy iv. safe operation of ANPP until it can be replaced with a new Emphasizes the NPP importance of v. environmentally viable energy supply investing in vi. creation of financially viable energy system infrastructure vii. maintenance and further development of export oriented and economically efficient power system. 51 Annex 1: Government Objectives, Cont. Energy Sector National Program on National Energy Action Plan of the Development Strategy Energy Saving and RE Security Concept MENR (2007) (2005) (2007) (2013) Policy direction & specific measures 1. Development of 2. Efficient use of own 3. Safe operation of 4. Diversification of energy security resources: ANPP & construction of a energy supply & system: • Increase use of RE new NPP: regional integration: • Identify energy resources • Complete ANPP • Gas storage security threats • Promote EE safety enhancement • Develop oil reserves • Improve and maintain safe • Armenia-Iran and • Explore potential of institutional & operation Armenia-Georgia bio fuels and other legal foundations possible fuel • ANPP 400/500 kV • Gradual alternatives decommissioning transmission lines liberalization of Feasibility study, • Integration into CIS • Search for new energy markets design works and energy market, fossil fuel deposits commissioning of regional power • Diversify imports new ANPP unit markets 5. Ensure financial sustainability and economic 6. Ensure energy security in case of efficiency of sector emergency situations or time of war 52 Annex 2: Overview: Power Sector 2% average annual increase in domestic consumption in 2010-2013 Electricity Consumption by Customer, 2008- 2013 Generation (2013) Installed Capacity 3500 (MW) 2600 (Higher summer Available Capacity – capacity is due to larger Summer (MW) availability of hydro) Available Capacity – 1900 Winter (MW) Load Profile (2013) Consumption (2013) Total Consumption (GWh) 5267 Summer Peak Demand (MW) 960 Winter Peak Demand (MW) 1520 * Adjusted for unusually cold winter of 2013 53 Annex 2: Overview: Power and Gas Tariffs  Gas and electricity tariffs are at short-term cost-recovery level  In 2010-2013, electricity and gas tariffs for residential customers in increased by 29% - Gas tariffs increased due to increase of border price of imported gas from Russia - Electricity tariffs increased given that 30% of generation is gas-based - Electricity tariffs are differentiated by voltage levels and time-of-day Electricity tariffs Day Night (AMD/kWh) Residential 38 28 0.4 kV 38 28 6 (10) kV 35 25 35+ kV 29 25 Gas tariffs (AMD or USD/cubic meter) Small consumers (less than AMD 156 - Gas tariffs are differentiated by volume 10,000 m3/month of consumption Large consumers (10,000 m3 per $277 (AMD 114) month and more) 54 Annex 2: Overview: Gas-for-Electricity Swap with Iran Since 2011, Armenia has been exporting electricity Iran in exchange for gas Summary of the Agreement: Benefits for Armenia:  Gas is imported through the Iran-Armenia gas pipeline with annual capacity of 2.3 billion m3 Helps to reduce the cost of domestic supply: (commissioned in 2011).  New efficient Yerevan CCGT and Hrazdan-  Electricity-for-Gas Swap Agreement targets to 5 generate 4.5 kWh with 1m3 of gas increase average annual amount of gas received to - If electricity is supplied during day and 2.3 bcm and electricity supplied – 6.9 billion kWh. peak hours, then Armenia keeps 1.5-4.5 - Currently only 360 million m3 received annually in kWh of power as profit exchange for 1.2 billion kWh - “Profit” power is sold in domestic market - Full contractual quantities can be achieved after without fuel component in the tariff commissioning of 400 kV transmission line Benefits for Iran:  Exchange rates for electricity supplied under the Swap Agreement are differentiated by the time of Helps to reduce unmet demand in Northern the day: regions - 12:01 AM – 8 AM: 6 kWh per 1m3 - 8:01 AM – 8 PM: 3 kWh per 1m3 - 8:01 PM – 12:00 AM: 1.5 kWh per 1m3  Yerevan CCGT is party to the Swap Agreement 55 Annex 2: Overview: Gas Sector  In 2010-2013, domestic consumption recovered to pre-crisis levels  Growth in domestic gas consumption driven by power sector Vertically integrated monopoly gas company 100% owned by the Russian Gazprom - 2,000 km transmission pipeline - 11,000 km distribution pipelines - Installed gas storage capacity of around 190-195 mln m3 with current operational capacity of 127-130 mln m3 Import Pipeline Capacity 2008 2011 2013 (bcm/year) Number of residential North-South (gas from Russia) 3.65 customers 490,000 580,000 626,000 Iranian pipeline (gas from Iran) 2.30 * Both pipelines owned by Gazprom 56 Annex 3: Electricity Demand Forecast Model  Econometric model derived to forecast electricity demand in Armenia  Demand forecasts were produced for both residential and non-residential demand 1) Demand was estimated using a log-log model with the Elasticity Estimate following functional form: = ° + 1 + 2 Residential 2) The models were fit using data for electricity demand (D), real Price -0.44 GDP (Y), and real tariff price (P) for the 1996-2013 Income 0.30 3) Assumptions on GDP and tariffs were fed into the models to produce demand estimates Non-Residential  GDP was based on ADS 2014-2025 Price -0.17  For the first iteration of demand forecasts real prices were kept constant Income 0.47  For the second iteration, tariffs under the least cost plan were used Low income elasticity of residential demand is due to 100% access rate and high saturation with household appliances. Low income elasticity of industrial demand is due structure of Armenian economy with forecast growth driven by knowledge intensive and non energy intensive industries (e.g. machinery, IT, software, diamond processing/polishing, tourism) 57 Annex 3: Electricity Demand Forecast In 2014-2030, electricity demand is forecast to grow 1.6-3.2% per year 10,000 8,000 6,000 GWh 4,000 2,000 0 Low High Base Note: (1) Base GDP growth set using forecasts from the Armenia Development Strategy (2014); low GDP growth is base case minus 2%; high GDP growth is base case plus 2%. (2) Forecast takes into account effects of increased tariffs under least cost plan. 58 Annex 3: Forecast of Gas Import Price Russian gas import prices are expected to be more predictable than before Under the new 2011 agreement, import price of Russian gas is linked to wholesale price in Russia’s Orenburg region and US Consumer Price Index and is adjusted annually Gas Price Forecast assumptions: Gas price forecasts are based on the Orenburg wholesale gas price scenarios from the “Forecast of Social-Economic Development of Russia until 2030” released by Russia’s Ministry of Economic Development and Bank team assumptions on the US Urban Consumer Price Index (CPI-U) 59 Annex 3: Gas Demand Forecast Model Constant-elasticity demand function was assumed to derive gas demand forecast 1) Demand was estimated using the following linear functional form: D= a*Y + b*P 2) D=average rate of growth of demand between successive forecast periods, a=income elasticity; Y=growth of real GDP between successive forecast periods; b=price elasticity of demand; P=change of real gas prices between successive forecast periods 3) Assumptions on GDP and tariffs were fed into the models to produce demand estimates  GDP was based on ADS 2014-2025  Real tariff forecast was derived using forecast domestic price of Russia in Orenburg region and forecast US CPI-U (IMF) Category of Consumer Income Elasticity Price Elasticity Residential 0.70 -0.25 Industrial and commercial 0.80 -0.10 Transport 0.50 -0.25 60 Annex 4: LCP Analyses: Methodology Least-cost planning analyses followed the below methodology 1. A dispatch simulation was run for each supply plan under all demand scenarios to determine the annual kWh output of each plant. 2. Generation and emission costs for each supply plant were calculated:  Capital costs for each supply plan were calculated given the international prices  Variable O&M, fuel, and emission costs were calculated using the kWh output multiplied by plant specific characteristics (e.g. heat rate, non-fuel variable cost of generation).  Fixed O&M was assumed to equal the international benchmarks for each generation technology included in the supply plant  Emission costs 3. Economic NPV was calculated as the sum of the total cost of generation and emission costs for each supply plan over the planning period, discounted at 10% Note: Long-run average incremental costs for each supply plan were computed as the ratio of the present value of incremental total costs and present value of all energy generated during the evaluation period 61 Annex 4: Key Assumptions for LCP Analyses General Assumptions  ANPP out for one year in 2016 for life-extension related work  Hrazdan TPP run until new CCGT is on-line  Vorotan Cascade and Sevan-Hrazdan Cascades will implement required rehab  Actual 2013 hourly generation profile of existing HPPs was used for their forecast dispatch  Iran demand under electricity-for-gas swap deal was modelled assuming actual 2013 hourly exports profile  Renewable energy technologies were assigned “capacity credit” based on their average capacity during top 10% of load hours in a year (solar PV is assumed 0% availability during peaks)  No electricity trade with Georgia  No seasonal electricity swap with Iran  Gas-for-electricity swap with modelled assuming current levels: 1.2 billion kWh of electricity in exchange for 360 million cubic meters of gas  CO2 emission costs factored in economic analyses  Discount rate of 10% 62 Annex 4: Key Assumptions for LCP Analyses Capital Cost Assumptions $/kW Data Source Capital cost of new CCGT 1100 Recent CCGT projects in the region Capital cost of Medium NPP 4250 Ongoing CNP projects Capital cost of Large NPP 5500 Pre-feasibility study for Armenia VVER- 440 reactor, 2010 Loriberd HPP (storage) 2150 Feasibility Study Update, 2009 Shnogh HPP (ROR) 1850 Government and Bank team estimate Flash cycle geothermal 3700 GeoFund 2: Armenia Geothermal Project Utility-scale PV 2400 SREP Investment Plan Project Wind 2200 SREP Investment Plan Project 63 Annex 4: Nuclear GoA should consider other commercially available nuclear technologies • The average expected cost of reactors under construction varies widely from country to country, from a low of US$1,951 per kW (India) to a high of US$6,938 per kW (Finland) • Average expected costs also vary from $1,841 to $8,400 depending on the NPP technology Average Expected Cost per kW by Plant Technology VVER-440 Armenia is considering is one of the highest cost II III III+ IV 64 Annex 4: Nuclear, cont. Average construction time of NPP projects increased • Construction times have steadily increased Recent Experience with Construction since the first nuclear plants were built in the Times* of NPPs Globally 1960s, and in recent years, they have seen wider variation than ever before.1 • Reasons for increased delays in NPP projects include:  Regulation and licensing. Licensing delays have increased following the Fukushima disaster.  Longer public consultations given increased safety concerns.  Scarcity of equipment. Slow global demand has decreased the availability of specialized Source: Bank Team Research construction equipment *Construction times for nuclear units do not include the Owner’s development time or licensing activities. 65 Annex 4: Profile of Loriberd and Shnogh HPPs Loriberd Shnogh Loriberd HPP and Tashir SHPP Two construction options: One construction option: ROR design ROR design Capacity: 54 MW Capacity: 75 MW Production: 208 mln. kWh Production: 300 mln. kWh Cost: $114 million Cost: $137 million Peaking Design Capacity: 66 MW Production: 206 mln. kWh Cost: $128 million Implementation Implementation - 2 years for technical studies and tender - 2 years for technical studies and tender - 5 years for implementation - 5 years for implementation 66 Annex 5: Renewable Energy Targets Capacity installed (MW) Generation (GWh) 2020 2025 2030 2020 2025 2030 Small Hydro 377 397 397 1,049 1,106 1,106 Wind 50 100 150 117 232 349 Geothermal 50 100 150 373 745 1,116 PV 40 80 80 88 176 176 Total 492 677 777 1,627 2,259 2,747 Note: All RE targets are included in the renewable energy scenarios. 67