May 2016 · Number 14 Assessing policy Trade-Offs with Easy-to-Use Models in PERs – Examples from the Energy Sector Wael Mansour, Shireen Mahdi, expenditure invested in reducing transmission losses; or the impact of different public and Simon Davies and Chadi Bou Habib1 private financing mixes on fiscal balances. Developing these tools allows for flexibility Introduction in answering different policy questions. Each In this note, we summarize three energy tool can be adapted or augmented to focus on a sector models used in Lebanon, Kosovo and broad set of questions as they become relevant. the Comoros, three countries that had The tools presented in this note focused on significant energy-sector challenges and fiscal impacts; consumer tariff impacts; and limited fiscal space available to finance economic impacts. In the Comoros case - what solutions. Public Expenditure Reviews (PERs) are the subsidy needs and will ongoing reforms can help governments to learn from past narrow the utility’s financing gap? In the experiences to identify future risks and Kosovo case – how do investment and subsidy improve future resource allocation. Sometimes choices impact required power this can involve taking difficult decisions. imports/exports and consumer tariffs? In the Developing easy-to-use “tools” that highlight Lebanon case - what is the expected economic the impacts (pros and cons) of different policy loss (in terms of growth) resulting from the options can bring more clarity to difficult deficiency of the electricity sector? Other decision-making. This note presents the potential questions may include how different experiences of three PERs that developed income groups are impacted by tariffs or which models to show the impact of energy policy productive sectors suffer from energy options. Each model was designed to address deficiencies, among others. the specific challenges faced by each country to The tools are based on easy-to-use Excel- ensure relevance and facilitate policy dialogue. based models that can be extended to a wide The models showed policy-makers the links number of sectors. Recent examples include between sectoral performance, financial tax policy and health insurance. While they are performance of state owned enterprises not likely to show one specific policy (SOEs), the budget, and economic recommendation, they can help policy-makers performance. In each case, the models were assess trade-offs. constructed with a similar ‘bottom up’ Comoros 2 : Estimating the size and path for framework based on financial data from SOEs, reducing subsidies energy sector data, combined with budget data. Users are able to model policies changes Comoros Context: Comoros suffers from a or performance parameters to estimate the high cost (diesel based) and low efficiency likely impacts of choices. For example, the electricity sector. The financial position of the impact of subsidies on the budget and main utility (MA-MWE) has deteriorated consumer tariffs; the impact of public sharply in recent years and its subsidy needs 1This Knowledge Note was cleared by Mark Roland 2Comoros Public Expenditure and Fiscal Management Thomas, Practice Manager (GMFDR). Review, 2015. have been growing, amounting to Estimates of the utility’s subsidy needs approximately 1.5 percent of GDP in 2013. These difficulties contributed to the onset of the ongoing electricity crisis. Policy questions: A model was developed to estimate the fiscal and financial recovery implications of reform scenarios for the electricity sector. The model sought to answer the following questions: (i) What are the drivers of the utility’s financing gap, will it need a Estimates of the utility’s operational balance (lines) & financial subsidy going forward, and for position (bars) how long? (ii) What is the expected fiscal impact of the planned reforms and investments in the sector? (iii) What will it take for utility’s long- term financial recovery? An excel based model was constructed to answer these questions based on a three step procedure: ONE: CONSTRUCT THE UTILITY’S FINANCIAL FLOWS Build MAMWE’s income and expense statement, and The policy impact was high. The analysis the budgetary flows between government and the contributed to a shift from ad-hoc funding utility. practices to a costed and budgeted subsidy to TWO: BUILD THE SCENARIOS MA-MWE in 2016 for the first time. The Develop scenarios based on the relevant policy government also used the model to support a questions, potential sector reforms and investments. decision to reduce the administered price of diesel to MA-MWE by 25 percent. This decision THREE: CREATE THE SIMULATIONS represents the only pass-through of the recent Forecast performance under various reform global fuel price reductions to the economy, assumptions and apply them to the financial and has significantly reduced the operating framework to estimate the utility’s financing gap, costs of MA-MWE. operational balance and net financial position. Kosovo 3 : Modeling the impact of capital The results: The analysis delivered the investment on performance and tariffs required estimates of the utility’s financing gap and its financial posiition. It helped to delive Kosovo context: Kosovo was reliant on two three important messages to the authorities. aging coal-fired power plants that would soon First, subsidies are essential to the success of need to close for 97 percent of its energy the recovery plan and should be budgetd for generation. At the same time, the energy sector the next five years. Second, that investing in benefited from explicit and implicit subsidies production capacity without improving that, combined with the amortized generation commercial performance will widen the facilities, kept energy tariffs the lowest in financing gap. Third, that diversifying the fuel Europe. Between 2009 and 2012, explicit mix away from diesel is the most feasible path subsidies including those for imports cost to easing out subsidies and acheiving sustained around 1 percent of GDP. Over the same commercial recovery. period, the government was providing capital investment and loans, which went un-repaid. 3Kosovo Public Finance Review available from: https://openknowledge.worldbank.org/handle/10986/2 0756 June 2016 · Number 14 · 2 The latter were worth 4 percent of GDP by more than domestically-produced energy. 2012. With subsidies keeping tariffs low (a This would also have large Balance of bonanza from which the wealthy benefited Payments impacts. more than the poor), Europe’s lowest energy  Building a new power plant would prices (bar Serbia) would be unable to cover the maintain energy supply long into the costs of much-needed investments. Fiscal space future but would also cost around 20 was limited (partially due to an impending percent of GDP. Either tariffs would need government decision to construct a new to increase or resources would need to be highway rather than address energy issues), found from the public sector. public debt reasonably low (though with market access issues), and social protection for The policy impact: In Kosovo, these results the poor shrinking. There were difficult choices were used to present various scenarios in the to be made to finance the costs of investment in PER. Unfortunately, progress in expanding the sector, which amounted to around 20 Kosovo’s energy provision and issues with one percent of GDP of the two aging power plants has even led to energy supply cuts. Sensitive discussions regarding options for power supply are still Policy questions: Doing nothing wasn’t an ongoing and the Bank team continues to use option in an energy-starved region with this model for various simulations, particularly volatile relations between the countries. But regarding tariff impact of the options. who and how should the investments into the Sample model results: The cost of doing sector be paid for? The tool, based on data from nothing in Kosovo the Energy Regulator and augmented with fiscal and public debt data, shows the impact of various choices on: electricity tariffs; required public resources; public debt and interest costs. The results: The Kosovo tool allows the user to modify several parameters and see the impacts. Parameters that can be modified include: (i) decommissioning, new construction or renovation of plants; (ii) how the work is financed (private, public shares and if public whether from budget or debt); (iii) new subsidy policies; (iv) distribution losses; (v) reactivity of import prices to increased demand; (vi) speed with which renewables come online; (vii) retail margins permitted; (viii) coal royalty policies; (ix) export prices; and (x) growth in demand. By modifying these parameters, the user controls (a) the supply and the demand, with the former controlled through investment choices and parameters, and (b) who pays for these choices. By allowing all costs to be passed to the consumer, the user can see the impact on average energy tariffs. By bringing all costs to the government, the user can see the impact on Lebanon: Modeling economic loss from an public expenditures and public debt. underperforming electricity sector Intermediate options are also available (e.g. by modifying subsidy policy). Lebanon Context: Electricity cuts are widespread across Lebanon reaching a daily Examples of results in Kosovo include: average of 9 hours. These cuts are due to the: (i)  Doing nothing resulted in an energy the insufficient capacity of generation, (ii) supply gap and increasing tariffs (or tariffs set more than 50% below cost recovery growing subsidies) due to imports costing levels, iii) the rapid depreciation of the sector’s June 2016 · Number 14 · 3 physical assets, and iv) the weak governance of and time), operational performance indicators the sector with high operational, financial and (like non-technical losses and bill collection technical losses. Government subsidies to the rate), and tariffs. Using those parameters, the monopoly electricity utility (EdL), covering report discusses 5 scenarios: a no reforms almost exclusively fuel purchase, have reached option used as a baseline, 3 scenarios building 4.2% of GDP in 2011; while the burden on on the emergency investment plan coupled household and businesses was estimated at with 2 sets of tariff adjustment, and a 5 th full 11.4% of GDP. reform program that includes additional investment in generation capacity. Policy Question: Investing in electricity can also be considered as an investment in fiscal Results: Scenario analysis revealed that the consolidation, in household welfare and in the GoL cannot reduce subsidies in a substantial performance of businesses. As such, the report way without increasing electricity tariffs, and examines the fiscal and social returns from that these subsidies will have to be maintained implementing the electricity sector investment for at least 6 years under the most optimistic program approved by the government of scenario. The tool estimates that that the Lebanon (GoL) in 2010. To do so, the report accumulated subsidies will range between developed an Excel based simulation tool with US$9.7 and 11.8 bn. On the welfare impact, the three pillars. report reveals that households are penalized by Pillar 1 uses cash flow analysis to determine the 2 electricity bills averaging US$1292 yearly per net cash position of the utility and sector, and household. This is equivalent to 22% of the computes the Net Present Value (NPV) of the minimum wage and 13% of GDP per capita. accumulated net cash results. While losses and Scenarios also indicate that any reforms that consequently subsidies are estimated at a increase supply would generate savings sectoral level; the NPV can be viewed as the reaching up to $629 per household per year opportunity cost arising from accumulated even if tariffs are increased. The overall welfare losses that could have been re-invested into the gains could therefore go up to US$2.9 bn. per economy. Data used under this pillar are year. Finally the analysis shows that business primarily EdL’s financial statements, planned losses due to outages could reach up to US$7.5 capital investments and debt service contracted bn. (9.8% of GDP) by 2021 under the status quo. as a result. Reforms could save firms up to a yearly US$2.9 bn. Moreover, suggested tariff reforms would Pillar 2 uses micro simulations to compute the add 0.2 to 0.3 percentage point (ppt) to inflation household electricity bill from both public and in year 1 of implementation, followed by 0.1 private sources. To do so it recurs to data on ppt hereafter till 2019. household electricity consumption collected as part of the 2009 nationally representative PSIA household survey, and to data on electricity The Policy Impact: This report came as part of tariffs and on fees for private generators a larger ESW entitled “Lebanon-Boosting services. Growth and Social Welfare” with 3 policy notes Pillar 3 builds on the Input-Output tables of the focusing on electricity reforms, telecom Lebanese National Accounts data to determine reforms, and the economic implications of the overall growth returns from private sector public sector wage increase. The report was savings, and the inflationary impact due to formulated as an on-time support for the tariff adjustment. In addition to national economic unit of the prime minister’s office accounts data, the model uses data on private and with strong collaboration with the sector losses from the ICA survey, along with ministries of economy and of energy. Closed inflation indices. policy discussions were conducted with senior officials, including ministerial level, and The tool allows for scenario building as most of cabinet was informed of the outcome of the the parameters remain flexible to input across simulations. Unfortunately no decision has the projection timeline. Such approach enables been taken and reforms were delayed due to a policy discussion around the sequencing the political deadlock that continue to aspect of reforms. The main simulation paralyses Lebanon. parameters are the investment program (size June 2016 · Number 14 · 4 Household bill according to different scenarios (in US$) 2000 About the authors: No Reform 1800 Wael Mansour, Economist, World Bank’s 1600 Emergency Plan - No Tarrif Increase 1400 Macroeconomics & Fiscal Management Global 1200 Emergency Plan - Tarrif Practice (GMFDR) Increase A Email: wmansour@worldbank.org 1000 Emergency Plan - Tarrif 800 Increase B Shireen Mahdi, Senior Economist, World Bank’s Macroeconomics & Fiscal Management Global Practice (GMFDR) 600 Full Program - Tarrif 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 increase B Email: smahdi@worldbank.org Subsidy to EdL - different scenarios (% of GDP) -5.0% -4.5% Simon Davies, Senior Economist, World -4.0% No Reform Bank’s Macroeconomics & Fiscal Management -3.5% Global Practice (GMFDR) -3.0% Emergency Plan - No Tariff Increase Email: sdavies@worldbank.org -2.5% -2.0% -1.5% Emergency Plan - Tariff Chadi Bou Habib, Senior Country -1.0% Increase A Economist, World Bank’s Macroeconomics & -0.5% 0.0% Fiscal Management Global Practice (GMFDR) Email: cbouhabib@worldbank.org Emergency Plan - Tariff 0.5% Increase B 1.0% 1.5% Full Program - Tariff 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 increase B June 2016 · Number 14 · 5