The Impact of Electricity Tariff Reforms and Alternative Mitigating Measures David Coady PSIA Group Fiscal Affairs Department International Monetary Fund
Based on background paper for an IMF Selected Issues chapter 2005 The welfare impacts of electricity tariff reforms and alternative compensating mechanisms: Evidence from Tajikistan (with Franziska Gassmann and Irina Klytchnikova)
Background I Tariff subsidies common in transition and developing countries >10% GDP in Kazakhstan, Tajikistan, Uzbekistan 3-4% GDP in Bulgaria, Georgia Adverse implications for financial position of state-owned enterprises and investments in capacity and distribution systems Increase in tariffs viewed as key component of sector reform strategy
Background II Price increases can have substantial adverse impact on low-income households Often viewed as an important political barrier to reform agenda Emphasis put on alternative approaches to mitigating adverse impacts of reform Use of lifeline tariffs Use of targeted safety net programs
Structure of Presentation Policy background in Tajikistan 2004 Tariff structure, energy patterns, reforms Methodology used in the analysis Magnitude and distribution of impact of tariff reforms on households Targeted social protection programs Existing and potential Comparison of cost-effectiveness of lifeline tariffs with mitigation through targeted transfers
Policy Background in Tajikistan 2004 Single state-owned enterprise dominating generation and distribution (Barki Tajik) Universal household access to network but supply problems common Cross-subsidization of residential consumers by industry Large subsidies to privileged groups and nonpayment of bills Average tariffs substantially below cost-recovery tariffs WB estimated subsidy at 19% GDP 2003 Average tariff only 25% of cost-recovery level (6.125dr/kWh)
Methodology Identify existing tariff structure and reforms Use household survey data to analyze electricity expenditures and tariff patterns Tajikistan Living Standards Survey 2003 Energy Household Survey 2004 Apply existing tariff structure to back out electricity quantities Apply new tariff to identify impact on households Impact on low-income groups and alternative mitigating measures
Evolution of Tariff Structure 2002 2003 (pre-june) 2003 (post-june) 2004 Lifeline Limit (household/month) 150 250 250 250 Lifeline Rates - Summer - Winter 0.5 0.5 1.6 1.6 0.8 1.6 0.8 1.6 Above-Lifeline - Summer - Winter 1.1 1.1 2.7 2.7 1.35 2.7 1.35 2.7
Electricity Consumption Patterns 2003 Electricity Consumption (kwh/month) Bottom Quintile 2 nd Quintile 3 rd Quintile 4 th Quintile Top Quintile All Households Proportion of Households >0 Summer 0.854 0.832 0.821 0.790 0.766 0.812 >0 Winter 0.891 0.880 0.882 0.859 0.854 0.873 >50 Summer 0.839 0.822 0.817 0.782 0.764 0.804 >50 Winter 0.876 0.869 0.878 0.853 0.853 0.866 >100 Summer 0.774 0.776 0.757 0.739 0.718 0.753 >100 Winter 0.860 0.866 0.857 0.844 0.843 0.854 >250 Summer 0.562 0.535 0.498 0.460 0.476 0.506 >250 Winter 0.658 0.644 0.620 0.598 0.643 0.632 Monthly Consumption Electricity Budget Share 344 334 328 314 330 330 0.040 0.035 0.032 0.031 0.031 0.034
Alternative Tariff Levels and Structures Baseline Scenario 1 Scenario 2 (Lower lifeline threshold) Scenario 3 (Equalizing summer and winter rates) Scenario 4 (Greater differentiation in rates) Lifeline Limit (household/month) 250 50 50 50 Lifeline Rates - Summer - Winter 0.8 1.6 0.8 1.6 1.6 1.6 0.5 0.5 Above-Lifeline - Summer - Winter 1.35 2.70 1.35 2.70 2.70 2.70 4.51 4.51
Impact of Alternative Tariff Structures Consumption Quintiles Scenario 1 (Existing Structure) Scenario 2 (Lower lifeline thresholds) Scenario 3 (Equalizing seasonal rates) Scenario 4 (Greater rate differentiation) Implicit subsidy as percent of income Bottom 0.086 0.079 0.069 0.041 2 nd quintile 0.076 0.070 0.061 0.037 3 rd quintile 0.070 0.063 0.056 0.034 4 th quintile 0.069 0.062 0.055 0.034 Top 0.067 0.061 0.054 0.032 All 0.073 0.067 0.059 0.035 Average tariff 1.67 2.05 2.56 4.00 (dr/kwh) Ratio to costrecovery tariff 0.273 0.335 0.418 0.653
Distribution of Implicit Subsidy Consumption Quintiles Scenario 1 (Existing Structure) Scenario 2 (Lower lifeline thresholds) Scenario 3 (Equalizing seasonal rates) Scenario 4 (Greater rate differentiation) Share of total implicit subsidy (in percent) Bottom 0.203 0.204 0.202 0.202 2 nd quintile 0.200 0.200 0.200 0.200 3 rd quintile 0.201 0.201 0.201 0.202 4 th quintile 0.194 0.193 0.194 0.195 Top 0.210 0.201 0.202 0.201
Existing Social Protection Programs Social Pensions Paid to those unable to work and not entitled to social insurance pension (pensioners, disabled, deceased earner) Cash Compensation Program (CCP) Reformed program introduced in 2000 targeted to poorest 20% children aged 6-15 years in each school (sharing, national 2002) School committees chose eligible students and deliver transfers without explicit criteria Energy Compensation Mechanism (ECM) Introduced in Jan2003 with tariff increases In principle in-kind means tested but in practice qualitative assessment by village committee Discount for consumption up to 100-150 kwh per month given to electricity company based on village-level records
Alternative Compensation Mechanisms Identify alternatives in household survey data (TLSS2003) Cash Compensation Program Hypothetical Proxy-Means Program Evaluate targeting performance Ability to transfer benefits to large proportion of low-income households (coverage) without high leakage to others Choose poorest 20% based on predicted probability of receiving program
Coverage and Leakage Rates Participation Rates Benefit Shares CCP PMP CCP PMP Bottom 0.250 0.444 0.249 0.433 2 nd Quintile 0.211 0.291 0.209 0.287 3 rd Quintile 0.191 0.148 0.194 0.149 4 th Quintile 0.183 0.083 0.188 0.084 Top 0.148 0.044 0.157 0.045 All 0.200 0.200 1.000 1.000
Alternative Mitigation Strategies Comparing three reform alternatives Instead of moving to cost-recovery tariffs for all households, move to structure of Scenario 4 Move to cost-recovery tariffs and compensate households using budgetary savings relative to Scenario 4 via CCP Move to cost-recovery tariffs and compensate households using budgetary savings relative to Scenario 4 via PMP All alternatives have same fiscal impact Under Scenario 4, government finances subsidies from budget Under transfer programs, government covers transfer bill
Net Impact of Alternative Measures Income Impact of Reform CCP PMP Scenario 4 Bottom -0.011 0.015-0.045 2 nd Quintile -0.034-0.021-0.040 3 rd Quintile -0.035-0.042-0.036 4 th Quintile -0.025-0.053-0.035 Top -0.013-0.057-0.035 All -0.023-0.032-0.038
Policy Implications Electricity tariff subsidies are not likely to be a cost effective approach to protecting low-income households Developing more targeted compensation mechanisms will substantially reduce the fiscal cost of mitigation (for all structural reforms) Targeted transfers can be focused on different populations (infants, children, adults, elderly etc) and conditioned to both generate physical/human capital accumulation and improve targeting Measures should be integrated into a comprehensive social protection strategy to ensure consistent targeting, avoid duplication and decrease administrative costs