## _wp0643

## Source details

**Canonical URL:** [_wp0643](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0643.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0643.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0643.pdf.json)

---

### Introduction and scope
- Focus: quantifies quasi-fiscal deficits (QFDs) for the electricity and gas sectors in eight CIS countries: Armenia, Azerbaijan, Georgia, Kyrgyz Republic, Moldova, Tajikistan, Ukraine, Uzbekistan.
- Timeframe and analysis period: up to 2003; key comparisons for 2000 and 2002.
- Objective: review Fund conditionality under stabilization and reform programs aimed at eliminating QFDs and improving conditions for economic growth.

### Definition and measurement of quasi-fiscal deficit (QFD)
- QFD definition: the revenue-expenditure gap of a public utility company, computed as the difference between actual revenue charged and collected at regulated prices and the revenue required to fully cover operating costs of production and capital depreciation.
- Measurement approach: "end-product approach" comparing actual cost of energy to domestic end-users with cost-recovery tariffs.
- Formal equations (variables preserved as in source):
  - (1) QFD = Q – R
  - (2) Q = 1/(1-ℓ)*(Ch + Ci + Co) * APC
  - (3) R = (Ch + Ci + Co) * T * Ccash
  - Variable definitions preserved from source: Q, R, ℓ, Ch, Ci, Co, APC, T, Ccash.
- APC: average cost of production per kWh reflecting maintenance and depreciation; excludes future capacity expansion and is below long-run marginal cost (LRMC).
- Notes on limitations: implicit nature of QFDs, cross-country data source differences, possible divergence from other IMF documents.

### Key quantified findings (Table 1: Quasi-Fiscal Deficits in Electricity and Gas Sectors in 2002, in percent of GDP)
- Country breakdown of QFDs in 2002 (Electricity / Gas / Total):
  - Armenia: Electricity 0.6; Gas 1.1; Total -0.5
  - Azerbaijan: Electricity 11.6; Gas 9.8; Total 1.8
  - Georgia: Electricity 5.7; Gas 5.5; Total 0.2
  - Kyrgyz Republic: Electricity 12.2; Gas 12.0; Total 0.2
  - Moldova: Electricity 4.2; Gas 3.3; Total 0.9
  - Tajikistan: Electricity 22.7; Gas 12.4; Total 1.3
  - Ukraine: Electricity 8.4; Gas 5.3; Total 3.1
  - Uzbekistan: Electricity 26.6; Gas 15.9; Total 10.7
- Source note: Authors' estimates.

### Cross-country and temporal context (extended fiscal deficits)
- In 2002, QFDs typically exceeded recorded general government fiscal deficits in all sample countries except Armenia.
- Extended fiscal deficit (general government fiscal balance plus QFD):
  - Armenia: remained below 5 percent of GDP in 2002.
  - Tajikistan and Uzbekistan: extended deficits close to 25 percent of GDP in 2002.
  - Kyrgyz Republic: extended deficit about 17 percent of GDP in 2002.
- Caveat: adding government fiscal deficit and QFD may double count mutual transfers; explicit net transfers typically small due to large implicit subsidy schemes.

### Electricity sector indicators and contributions (Table 2 and Table 4)
- Table 2 indicators (QFD (% of GDP); Actual Tariff (UScts/kWh); Cost Recovery Tariff (UScts/kWh); Cash Collection (% of Bill); Excess Losses (% of Prod.); Cost Recovery (%)):
  - Armenia: QFD 1.1; Actual Tariff 3.7; Cost Recovery Tariff 3.7; Cash Collection 90; Excess Losses 10; Cost Recovery 81.6
  - Azerbaijan: QFD 9.8; Actual Tariff 2.3; Cost Recovery Tariff 3.8; Cash Collection 30; Excess Losses 51; Cost Recovery 7.2
  - Georgia: QFD 5.5; Actual Tariff 4.1; Cost Recovery Tariff 4.1; Cash Collection 35; Excess Losses 53; Cost Recovery 3.3
  - Kyrgyz Republic: QFD 12.0; Actual Tariff 1.2; Cost Recovery Tariff 2.3; Cash Collection 33; Excess Losses 35; Cost Recovery 11.1
  - Moldova: QFD 3.3; Actual Tariff 5.0; Cost Recovery Tariff 5.0; Cash Collection 75; Excess Losses 19; Cost Recovery 61.0
  - Tajikistan: QFD 21.4; Actual Tariff 0.4; Cost Recovery Tariff 2.1; Cash Collection 86; Excess Losses 5; Cost Recovery 13.9
  - Ukraine: QFD 5.3; Actual Tariff 3.1; Cost Recovery Tariff 4.0; Cash Collection 87; Excess Losses 13; Cost Recovery 58.8
  - Uzbekistan: QFD 15.9; Actual Tariff 1.0; Cost Recovery Tariff 3.5; Cash Collection 50; Excess Losses 18; Cost Recovery 12.3
  - Source note: Authors' estimates; cost recovery tariff estimates provided by World Bank staff; excess losses defined as technical and commercial losses above the normative level.
- Table 4 contributions to electricity QFD (Underpricing / Non-payment / Losses, In percent of GDP):
  - Armenia: 1.1 / 0.0 / 0.5 / 0.6
  - Azerbaijan: 9.8 / 4.4 / 4.8 / 0.5
  - Georgia: 5.5 / 0.0 / 5.1 / 0.4
  - Kyrgyz Republic: 12.0 / 4.2 / 3.1 / 4.7
  - Moldova: 3.3 / 0.0 / 1.7 / 1.6
  - Tajikistan: 21.4 / 19.6 / 0.6 / 1.2
  - Ukraine: 5.3 / 2.5 / 1.1 / 1.7
  - Uzbekistan: 15.9 / 10.4 / 2.2 / 3.3
- Observations:
  - Mispricing major contributor in high-deficit countries (Tajikistan, Uzbekistan).
  - Losses exceptionally high in the Kyrgyz Republic.
  - Georgia: tariffs near cost recovery but bill collection low.

### Electricity tariffs by consumer category (Table 3, Actual Tariffs, USct/kWh, 2002)
- Households / Industry / Other / Average:
  - Armenia: 4.3 / 3.1 / 3.6 / 3.7
  - Azerbaijan: 2.0 / 3.3 / 2.3 / 2.3
  - Georgia: 5.5 / 2.2 / 3.2 / 4.1
  - Kyrgyz R.: 0.9 / 1.7 / 1.2 / 1.2
  - Moldova: na / na / na / 5.0
  - Tajikistan: 0.1 / 0.6 / 0.2 / 0.4
  - Ukraine: 2.6 / 3.0 / 3.6 / 3.1
  - Uzbekistan: na / na / na / 1.0
- Source: IMF country desk data.

### Gas sector indicators and contributions (Table 5 and Table 6)
- Table 5 indicators (QFD (% of GDP); Actual Tariff (US$/1000); Cost Recovery Tariff1/ (US$/1000); Gas Import Price (US$/1000); Cash Collection (% of Bill); Excess Losses (% of Prod.); Cost Recovery3/ (%)):
  - Armenia: QFD -0.58; Actual Tariff 0.05; Cost Recovery Tariff 8.0; Gas Import Price 55.0; Cash Collection 91; Excess Losses 6; Cost Recovery 118.8
  - Azerbaijan: QFD 1.8; Actual Tariff 15.0; Cost Recovery Tariff 30.0; Gas Import Price --; Cash Collection 47; Excess Losses 15; Cost Recovery 20.0
  - Georgia: QFD 0.2; Actual Tariff 105.0; Cost Recovery Tariff 70.0; Gas Import Price 65.0; Cash Collection 67; Excess Losses 12; Cost Recovery 89.0
  - Kyrgyz Republic: QFD 0.27; Actual Tariff 7.1; Cost Recovery Tariff 65.5; Gas Import Price 46.3; Cash Collection 98; Excess Losses 26; Cost Recovery 84.5
  - Moldova: QFD 0.96; Actual Tariff 6.8; Cost Recovery Tariff 65.0; Gas Import Price 60.0; Cash Collection 78; Excess Losses 8; Cost Recovery 78.8
  - Tajikistan: QFD 1.35; Actual Tariff 7.9; Cost Recovery Tariff 63.7; Gas Import Price 48.4; Cash Collection 55; Excess Losses 47; Cost Recovery 75.5
  - Ukraine: QFD 3.14; Actual Tariff 4.1; Cost Recovery Tariff 62.5; Gas Import Price 51.6; Cash Collection 89; Excess Losses 55; Cost Recovery 59.6
  - Uzbekistan: QFD 10.7; Actual Tariff 11.1; Cost Recovery Tariff 25.0; Gas Import Price --; Cash Collection 50; Excess Losses 72; Cost Recovery 0.6
- Table 6 contributions to gas QFD (Underpricing / Non-payment / Losses, In percent of GDP):
  - Armenia: -0.5 / -0.9 / 0.3 / 0.1
  - Azerbaijan: 1.8 / 0.9 / 0.5 / 0.3
  - Georgia: 0.2 / -0.7 / 0.7 / 0.2
  - Kyrgyz Republic: 0.2 / -0.2 / 0.0 / 0.4
  - Moldova: 0.9 / -0.1 / 0.9 / 0.1
  - Tajikistan: 1.3 / 0.2 / 1.0 / 0.1
  - Ukraine: 3.1 / 2.1 / 0.6 / 0.4
  - Uzbekistan: 10.7 / 7.0 / 2.8 / 0.9
- Observations:
  - Mispricing smaller problem in gas than in electricity.
  - Several countries reported gas tariffs above cost recovery levels.
  - Cash collections for gas generally higher than for electricity (between 50 and 90 percent).
  - Excess losses in gas highest in the Kyrgyz Republic.

### Macroeconomic relevance and empirical findings
- Negative macroeconomic and structural implications of QFDs:
  - undermine financial discipline and fiscal space;
  - keep nonviable enterprises afloat, delaying restructuring;
  - distort relative prices and reduce incentives to save energy;
  - distort resource allocation, crowding out labor-intensive SMEs;
  - increase external vulnerability in energy-importing countries.
- Financing patterns for energy companies’ imbalances: budgetary transfers, wage arrears, arrears to domestic suppliers, external arrears, and capital depreciation.
- Consequences of failing to eliminate QFDs: gradual decay of capital stock and supply shortages.
- Empirical correlations reported in source:
  - Strong correlation between electricity QFD (% of GDP) and electricity consumption per PPP GDP: R2 = 0.9458 for CIS 2002 sample; R2 = 0.7459 for selected CEE and CIS 2001 sample.
  - Regression for selected CEE and CIS countries: Inefficiency = -0.058x + 0.72; R2 = 0.377; t-stat for coefficient (-3.48).
  - Low electricity tariffs correlate strongly with high QFDs (Figures and regressions reported).
  - Figure 9 regression: QFD in % of GDP regressed on actual tariff as ratio of average production cost: y = -0.20x + 23.01; R2 = 0.93; coefficient t-stat -9.09.
  - Figure 8 regression: y = -3.58x + 19.25 ( -4.50 t-stat), R2 = 0.77.
  - Figure 7 t-statistics and regression specifics reported in source.

### IMF conditionality on the energy sector (1993–2003): scope, types, and intensity
- Aggregate conditionality:
  - 155 performance criteria, structural benchmarks, and prior actions imposed for the eight countries in 1993–2003.
  - Intensity peaked in 1998 and 1999; Ukraine accounted for more than half of conditions during the peak.
  - In 2001–03 the volume of conditions fell to less than half of the peak years.
  - Only Georgia and Tajikistan had more conditionality in 2002–03 than in previous years.
- Composition of measures (counts and shares reported):
  - Tariffs: 40 (26%)
  - Collections: 14 (9%)
  - Privileges: 17 (11%)
  - Arrears: 8 (5%)
  - Financial Balance: 16 (10%)
  - Privatization: 19 (12%)
  - Restructuring: 14 (9%)
  - Other: 18 (12%)
  - Debt: 9 (6%)
- Rationale for changes in intensity: complexity of energy reforms, World Bank leadership on many measures, 2001 Fund guidelines streamlining conditionality.
- Evidence of declining marginal compliance with increasing number of conditions (Figure 5 reported).

### Performance record of conditionality and observed outcomes
- Aggregate compliance:
  - Aggregate compliance rate under Fund conditionality was 61 percent in 1993−2003.
  - Excluding Ukraine, aggregate compliance was 66 percent (Total 155, implemented 94, 61% overall; Total excl. Ukraine 92, implemented 61, 66%).
- Country-level compliance (selected):
  - Azerbaijan: total 13, implemented 9, compliance 69%
  - Armenia: total 12, implemented 6, compliance 50%
  - Georgia: total 16, implemented 10, compliance 63%
  - Kyrgyz Republic: total 15, implemented 10, compliance 67%
  - Tajikistan: total 13, implemented 9, compliance 69%
  - Moldova: total 20, implemented 17, compliance 85%
  - Ukraine: total 63, implemented 33, compliance 52%
  - Uzbekistan: total 3, implemented 0, compliance 0%
  - Overall total: total 155, implemented 94, compliance 61%
- By-type compliance (selected):
  - Tariffs: 40 total, 20 implemented, 50% compliance
  - Bill collection: 14 total, 5 implemented, 36% compliance
  - No arrears: 8 total, 5 implemented, 63% compliance
  - Reduce energy privileges: 17 total, 10 implemented, 59% compliance
  - Financial balance: 16 total, 12 implemented, 75% compliance
  - Privatization: 19 total, 12 implemented, 63% compliance
  - Foreign energy debt: 9 total, 9 implemented, 100% compliance
  - Regulation, restructuring: 14 total, 10 implemented, 71% compliance
  - Other: 18 total, 11 implemented, 61% compliance
- Observations:
  - No significant correlation found between compliance rate and QFD outcomes across sample countries (examples: Moldova high compliance/low QFD; Tajikistan high compliance/large QFD; Ukraine many measures but QFD remained high).
  - Lowest compliance in politically sensitive areas: bill collection and tariffs.

### Specific program experiences and selected measures
- Tariff policies and outcomes:
  - About 25 percent of measures addressed tariffs; only half implemented.
  - By 2002, only Armenia, Georgia, and Moldova had raised electricity tariffs to cost recovery levels.
  - Regression (Figure 9): y = -0.20x + 23.01; R2 = 0.93; coefficient t-stat -9.09.
- Country examples (selected):
  - Armenia: under the 1996 ESAF program, residential and industrial tariffs were raised to cost recovery levels; earlier 1995–96 measures not observed.
  - Georgia: tariff unification and increases implemented as a 1997 benchmark but later deviated; World Bank conditionality later implemented measures.
  - Moldova: tariff increases implemented in 1996 and 1998 via debt surcharge; later increases under World Bank conditionality.
  - Kyrgyz Republic: electricity tariffs increased by 20 percent on April 1, 2002; PRGF programmed electricity QFD to decline from 12 percent of GDP in 2002 to 11 percent in 2003 (achieved).
- Bill collection and losses:
  - Bill collection conditionality had 36% compliance.
  - Examples of improved collection: Armenia (cutting off nonpayers in 1996), Ukraine (improvements in 2000 after cut-off policy).
  - No program included specific conditionality on loss prevention.
- Energy use privileges and social protection:
  - Removal of categorical privileges was politically difficult; mixed program results across countries.
  - Examples: Armenia removed categorical privileges for heating and hot water in 1995; Tajikistan eliminated energy privileges as a prior action in 2003.
- Payment arrears and external debt measures:
  - Few programs had explicit domestic arrears conditionality; some targeted external energy debt restructuring (Georgia, Tajikistan, Moldova).
- Privatization, regulation, and market reforms:
  - Partial privatization success in Georgia, Moldova, and Ukraine; mixed progress and many cases where World Bank led measures.

### Outcomes, lessons learned, and energy efficiency trends (Table 9 and Figures)
- Energy inefficiency changes 1992–2001 (Energy Use per PPP GDP; Electricity consumption per PPP GDP):
  - Armenia: 1992 = 0.71 / 1.03; 2001 = 0.27 / 0.41
  - Azerbaijan: 1992 = 0.76 / 0.55; 2001 = 0.50 / 0.64
  - Georgia: 1992 = 0.85 / 0.95; 2001 = 0.22 / 0.34
  - Kyrgyz Republic: 1992 = 0.69 / 1.11; 2001 = 0.28 / 0.82
  - Moldova: 1992 = 0.82 / 0.98; 2001 = 0.53 / 0.57
  - Tajikistan: 1992 = 1.30 / 2.36; 2001 = 0.53 / 2.34
  - Ukraine: 1992 = 0.68 / 0.58; 2001 = 0.63 / 0.49
  - Uzbekistan: 1992 = 1.57 / 1.40; 2001 = 1.26 / 1.01
  - Simple Average: 1992 = 0.92 / 1.12; 2001 = 0.53 / 0.83
  - Memorandum items: Europe & Central Asia 1992 = 0.59 / 0.58; 2001 = 0.41 / 0.42; Low Income 1992 = 0.35 / 0.18; 2001 = 0.25 / 0.15; Middle Income 1992 = 0.38 / 0.34; 2001 = 0.24 / 0.26; OECD 1992 = 0.20 / 0.37; 2001 data partly truncated in source.
- Key lessons:
  - Energy inefficiencies remained high by 2001 despite overall improvement since 1992; electricity sector progress was less pronounced.
  - Reform inertia due to vested interests, weak regulatory quality, and corruption; Figure 12 regression: y = 3.1435x -0.5938, R2 = 0.4094.

### Affordability and LRMC implications (Table 10 and related)
- Electricity expenditure in percent of household income (actual / at LRMC prices 1/):
  - Armenia 8.4 / 11.1
  - Azerbaijan 3.8 / 9.3
  - Georgia 12.4 / 16.7
  - Kyrgyz Republic 1.9 / 9.1
  - Moldova 5.2 / 7.3
  - Tajikistan 0.2 / 6.6
  - Ukraine 2.9 / 5.8
- Memorandum items:
  - Russia 1.0 / 0.5
  - Baltics 2.3 / 2.6
  - CIS 3.4 / 6.6
  - Central Europe 3.6 / 3.5
- To cover LRMC of production, payments would need to increase to 6.6 percent of household income.
- EBRD LRMC reference: 8 U.S. cents per kilowatt hour—the average retail price in the United States.

### Bill collection, tariff behavior, and operational impediments
- Empirical patterns:
  - Kyrgyz Republic: tariffs increased by 20 percent in April 2002; cash collections declined to 32 percent of billing in 2003 (from 58 percent in 2001).
  - Cross-country regression examples: y = 5.9954x + 56.148, R^2 = 0.1858 (tariffs vs. collection); CIS sample cubic fit: y = -5.0096x^3 + 45.064x^2 - 110.94x + 119.15, R^2 = 0.3487.
  - Suggested U-shaped relationship between tariffs and bill collection.
- Operational impediments: inability to disconnect noncompliers, meter manipulation, inspectors’ collusion.
- Countermeasures cited: relocation of meters in locked boxes, separation of meter reading from billing, mandatory payments through banks, computerization of consumer accounts.

### Policy options and recommended sequencing
- First-best (most ambitious):
  - Include an audit-based QFD of the energy sector as a separate performance criterion in Fund-supported programs.
  - Preconditions: reliable business accounting, application of International Accounting Standards (IAS), development of the auditing profession.
  - Benefits: covers all sector operations, reduces scope for evasion, improves transparency.
- Second-best:
  - Set the energy sector’s QFD as a performance criterion using the end-user approach as a yardstick.
  - Complementary steps: audits by internationally reputable firms and adoption of IAS; include the deficit in annual budget documentation.
- Third option (governance-focused, lower ambition):
  - Make the cost of QFD explicit in the budget; cover budgetary flows related to energy (zero tolerance of arrears; taxation of utility consumption; restructuring of tax arrears and external debts; social transfers for canceling discounted tariffs).
  - Leave tariff reform, cash collections, losses, and management under World Bank conditionality (or Fund if no Bank operation exists).
- Final (crisis-driven) option:
  - Reforms may occur only through severe supply crises (blackouts), raising public support for tariff increases but at higher adjustment costs.
- Social protection and sequencing:
  - Compensate vulnerable populations; link energy cost changes to social safety nets in the budget; discuss and approve tariff and compensation measures simultaneously in parliament.
  - Social compensation for short-term losers should not be viewed as a fiscal stability risk given long-term gains from energy reform.

### Epilogue — selected country developments (post-2002 estimates and program updates)
- Armenia:
  - Separate companies created for financial settlements, dispatch, and high voltage distribution; independent boards appointed.
  - Tariffs covered costs already in 2002; bill collection rates reported at 100 percent; losses declined; electricity QFD virtually eliminated (remaining issue: historical debts).
- Azerbaijan:
  - Natural gas cost recovery ratio reportedly rose to 85 percent by end-2004 from 50 percent in 2002; bill collection rates around 50 percent.
  - Electricity: SOCAR provides underpriced fuel and gas inputs; tariffs below cost recovery and payment collection ineffective.
- Georgia:
  - Bill collection rose from 35 percent in 2002 to over 50 percent in 2004; cost recovery ratio declined to 94 percent as tariff increases lagged cost increases.
  - 2004 PRGF-supported program introduced indicative targets for bill collections.
- Kyrgyz Republic:
  - Electricity QFD declined from 12 percent of GDP in 2002 to below 9 percent in 2004 (estimate).
  - New PRGF-supported program approved in early 2005 continued to include QFD as an indicative program target.
- Moldova:
  - Energy-related foreign supplier debt remains sizeable; collection and pricing losses persist; reforms incomplete.
- Tajikistan:
  - Gas tariffs at cost recovery level; power tariffs did not cover costs despite 2003 hikes.
  - World Bank estimated tariffs would have to increase fourfold in 2004 to achieve cost recovery.
  - Authorities estimate combined electricity and gas QFD may have declined to nearly half 2002 level after some tariff rationalization.
  - 2005 PRGF-supported program introduced a structural benchmark for auditing the TajikGas monopoly.
- Ukraine:
  - Electricity bill collection reportedly 90 percent at end-2004.
  - Better cost recovery due to increased industrial consumption where tariffs are highest.
  - Remaining test: improve cost recovery ratio in gas supplies (current average gas tariff covers around 80% based on reported economic costs in source).
- Uzbekistan:
  - Authorities reportedly increased electricity tariffs significantly in 2003–04, raising cost recovery ratio from 28 percent in 2002 to about 70 percent in 2004; collection rates improved.

### Appendix — measures, implementation, and compliance (selected aggregates)
- Appendix aggregates (selected):
  - Total conditionality: 155 total measures, 94 implemented, 61% compliance.
  - Total excl. Ukraine: 92 total measures, 61 implemented, 66% compliance.
- Measures and implementation by type (selected totals):
  - Tariffs: 40 total, 20 implemented, 50% compliance
  - Bill collection: 14 total, 5 implemented, 36% compliance
  - Reduce privileges: 17 total, 10 implemented, 59% compliance
  - Financial Balance: 16 total, 12 implemented, 75% compliance
  - Privatization: 19 total, 12 implemented, 63% compliance
  - Foreign debt rescheduling: 9 total, 9 implemented, 100% compliance
  - Regulation, restructuring: 14 total, 10 implemented, 71% compliance
- Country selected compliance snapshots:
  - Azerbaijan: 13 total, 9 implemented, 69% compliance
  - Armenia: 12 total, 6 implemented, 50% compliance
  - Georgia: 16 total, 10 implemented, 63% compliance
  - Kyrgyz Republic: 15 total, 10 implemented, 67% compliance
  - Tajikistan: 13 total, 9 implemented, 69% compliance
  - Moldova: 20 total, 17 implemented, 85% compliance
  - Ukraine: 63 total, 33 implemented, 52% compliance
  - Uzbekistan: 3 total, 0 implemented, 0% compliance

*Source: _wp0643 - References (authors' estimates and tables as presented in the source PDF).*

### References..............................................................................................................

### _wp0643 - References

### Introduction and scope
- Focus: quantifies quasi-fiscal deficits (QFDs) for the electricity and gas sectors in eight CIS countries: Armenia, Azerbaijan, Georgia, Kyrgyz Republic, Moldova, Tajikistan, Ukraine, Uzbekistan.
- Timeframe and analysis period: up to 2003; key comparisons for 2000 and 2002.
- Objective: review Fund conditionality under stabilization and reform programs aimed at eliminating QFDs and improving conditions for economic growth.

### Definition and measurement of quasi-fiscal deficit (QFD)
- QFD definition: the revenue-expenditure gap of a public utility company, computed as the difference between actual revenue charged and collected at regulated prices and the revenue required to fully cover operating costs of production and capital depreciation.
- Rationale: state-owned utilities are quasi-fiscal agencies because government sets tariffs and tolerates underpricing, non-collection, and commercial losses.
- Measurement approach: "end-product approach" comparing actual cost of energy to domestic end-users with cost-recovery tariffs.
- Formal equations:
  - (1) QFD = Q – R
  - (2) Q = 1/(1-ℓ)*(Ch + Ci + Co) * APC
  - (3) R = (Ch + Ci + Co) * T * Ccash
  - Variable definitions preserved from source: Q, R, ℓ, Ch, Ci, Co, APC, T, Ccash.
- APC: average cost of production per kWh reflecting maintenance and depreciation; excludes future capacity expansion and is below long-run marginal cost (LRMC).
- Notes on limitations: implicit nature of QFDs, cross-country data source differences, possible divergence from other IMF documents.

### Key quantified findings (Table 1: Quasi-Fiscal Deficits in Electricity and Gas Sectors in 2002, in percent of GDP)
- Armenia: Electricity 0.6; Gas 1.1; Total -0.5
- Azerbaijan: Electricity 11.6; Gas 9.8; Total 1.8
- Georgia: Electricity 5.7; Gas 5.5; Total 0.2
- Kyrgyz Republic: Electricity 12.2; Gas 12.0; Total 0.2
- Moldova: Electricity 4.2; Gas 3.3; Total 0.9
- Tajikistan: Electricity 22.7; Gas 12.4; Total 1.3
- Ukraine: Electricity 8.4; Gas 5.3; Total 3.1
- Uzbekistan: Electricity 26.6; Gas 15.9; Total 10.7
- Source note: Authors' estimates.

### Cross-country and temporal context (Figure 1 summary)
- In 2002, QFDs typically exceeded recorded general government fiscal deficits in all sample countries except Armenia.
- Extended fiscal deficit (general government fiscal balance plus QFD):
  - Armenia: remained below 5 percent of GDP in 2002.
  - Tajikistan and Uzbekistan: extended deficits close to 25 percent of GDP in 2002.
  - Kyrgyz Republic: extended deficit about 17 percent of GDP in 2002.
- Caveat: adding government fiscal deficit and QFD may double count mutual transfers; explicit net transfers typically small due to large implicit subsidy schemes.

### Electricity sector breakdown (Table 2: Quasi-Fiscal Deficit Indicators of the Electricity Sector in 2002)
- Armenia:
  - QFD (% of GDP): 1.1
  - Actual Tariff (UScts/kWh): 3.7
  - Cost Recovery Tariff (UScts/kWh): 3.7 (estimate provided by World Bank staff)
  - Cash Collection (% of Bill): 90
  - Excess Losses (% of Prod.): 10
  - Cost Recovery (%): 81.6
- Azerbaijan:
  - QFD (% of GDP): 9.8
  - Actual Tariff (UScts/kWh): 2.3
  - Cost Recovery Tariff (UScts/kWh): 3.8
  - Cash Collection (% of Bill): 30
  - Excess Losses (% of Prod.): 51
  - Cost Recovery (%): 7.2
- Georgia:
  - QFD (% of GDP): 5.5
  - Actual Tariff (UScts/kWh): 4.1
  - Cost Recovery Tariff (UScts/kWh): 4.1
  - Cash Collection (% of Bill): 35
  - Excess Losses (% of Prod.): 53
  - Cost Recovery (%): 3.3
- Kyrgyz Republic:
  - QFD (% of GDP): 12.0
  - Actual Tariff (UScts/kWh): 1.2
  - Cost Recovery Tariff (UScts/kWh): 2.3
  - Cash Collection (% of Bill): 33
  - Excess Losses (% of Prod.): 35
  - Cost Recovery (%): 11.1
- Moldova:
  - QFD (% of GDP): 3.3
  - Actual Tariff (UScts/kWh): 5.0
  - Cost Recovery Tariff (UScts/kWh): 5.0
  - Cash Collection (% of Bill): 75
  - Excess Losses (% of Prod.): 19
  - Cost Recovery (%): 61.0
- Tajikistan:
  - QFD (% of GDP): 21.4
  - Actual Tariff (UScts/kWh): 0.4
  - Cost Recovery Tariff (UScts/kWh): 2.1
  - Cash Collection (% of Bill): 86
  - Excess Losses (% of Prod.): 5
  - Cost Recovery (%): 13.9
- Ukraine:
  - QFD (% of GDP): 5.3
  - Actual Tariff (UScts/kWh): 3.1
  - Cost Recovery Tariff (UScts/kWh): 4.0
  - Cash Collection (% of Bill): 87
  - Excess Losses (% of Prod.): 13
  - Cost Recovery (%): 58.8
- Uzbekistan:
  - QFD (% of GDP): 15.9
  - Actual Tariff (UScts/kWh): 1.0
  - Cost Recovery Tariff (UScts/kWh): 3.5
  - Cash Collection (% of Bill): 50
  - Excess Losses (% of Prod.): 18
  - Cost Recovery (%): 12.3
- Source note: Authors' estimates; cost recovery tariff estimates provided by World Bank staff; excess losses defined as technical and commercial losses above the normative level.

### Electricity-sector observations and patterns
- Actual tariffs in 2002 varied significantly across countries; lowest in Tajikistan and Kyrgyz Republic where APCs were also lowest.
- Cost-recovery ratios were notably low in Tajikistan, Kyrgyz Republic, Uzbekistan, and Azerbaijan.
- Highest tariffs observed in Moldova, reflecting high production cost and significant share of imported gas as input to power generation.
- Cash collection: as high as 90 percent in Armenia, Tajikistan, and Ukraine; around 33 percent in Azerbaijan, Georgia, and Kyrgyz Republic.
- Excess technical and commercial losses most severe in the Kyrgyz Republic, Moldova, and Uzbekistan.
- Residential tariffs were cross-subsidized by higher industrial tariffs in all sample countries except Armenia and Georgia.

### Macroeconomic relevance and policy implications
- QFDs materially reduce the transparency of fiscal soundness; general government fiscal balance understates total fiscal pressures where large implicit subsidies exist.
- In countries with very large QFDs (e.g., Tajikistan, Uzbekistan), energy sector implicit subsidies represent a major constraint on macroeconomic adjustment and resource allocation.
- Reform emphasis: donor-supported programs focused on rehabilitating capital stock, unbundling vertically integrated monopolies, privatization, sector restructuring, and tariff reforms to shift balance from supply-side bias to demand-side efficiency.
- Implementation weakness: governments often prioritized production and distribution investments while tariff and payment reforms lagged, leaving extensive implicit subsidy schemes in place.

*Source: _wp0643 - References (authors' estimates and tables as presented in the source PDF).*

### 1.7 cents for nuclear power. See http://www.nei.org

### _wp0643 - 1.7 cents for nuclear power. See http://www.nei.org

### Electricity sector: tariffs, cross-subsidization, and contributions to the QFD (2002)
- Table 3 (Actual Tariffs, USct/kWh) — sectors (Households / Industry / Other) and Average:
  - Armenia: 4.3 / 3.1 / 3.6 / 3.7
  - Azerbaijan: 2.0 / 3.3 / 2.3 / 2.3
  - Georgia: 5.5 / 2.2 / 3.2 / 4.1
  - Kyrgyz R.: 0.9 / 1.7 / 1.2 / 1.2
  - Moldova: na / na / na / 5.0
  - Tajikistan: 0.1 / 0.6 / 0.2 / 0.4
  - Ukraine: 2.6 / 3.0 / 3.6 / 3.1
  - Uzbekistan: na / na / na / 1.0
- Source: IMF country desk data.

- Table 4 (Contributions to the Electricity QFD in 2002, based on Table 2) — contributions to Total QFD (In percent of GDP) split by Underpricing / Non-payment / Losses:
  - Armenia: 1.1 / 0.0 / 0.5 / 0.6
  - Azerbaijan: 9.8 / 4.4 / 4.8 / 0.5
  - Georgia: 5.5 / 0.0 / 5.1 / 0.4
  - Kyrgyz Republic: 12.0 / 4.2 / 3.1 / 4.7
  - Moldova: 3.3 / 0.0 / 1.7 / 1.6
  - Tajikistan: 21.4 / 19.6 / 0.6 / 1.2
  - Ukraine: 5.3 / 2.5 / 1.1 / 1.7
  - Uzbekistan: 15.9 / 10.4 / 2.2 / 3.3
- Commentary from source:
  - In high-deficit countries (Tajikistan and Uzbekistan), the main contribution to the electricity QFD was mispricing.
  - In the Kyrgyz Republic, losses were exceptionally high.
  - In Georgia, where tariffs were at cost recovery levels, bill collection remained low.
- Note: Cash collection rate is measured relative to billing; weakness in billing coverage (poor metering, meter tinkering, theft) is reflected in excess loss rates.
- Typical sector cost pattern: cost of delivering power to industry is lower than to households because of economies of scale.

### Gas sector: tariffs, cash collection, excess losses, and QFD (2002)
- Table 5 (Quasi-Fiscal Deficit Indicators of the Gas Sector in 2002) — indicators (QFD (% of GDP); Actual Tariff (US$/1000); Cost Recovery Tariff1/ (US$/1000); Of which: Gas Import Price (US$/1000); Cash Collection (% of Bill); Excess Losses (% of Prod.); Cost Recovery3/ (%))
  - Armenia: QFD -0.58; Actual Tariff 0.05; Cost Recovery Tariff 8.0; Gas Import Price 55.0; Cash Collection 91; Excess Losses 6; Cost Recovery 118.8
  - Azerbaijan: QFD 1.8; Actual Tariff 15.0; Cost Recovery Tariff 30.0; Gas Import Price --; Cash Collection 47; Excess Losses 15; Cost Recovery 20.0
  - Georgia: QFD 0.2; Actual Tariff 105.0; Cost Recovery Tariff 70.0; Gas Import Price 65.0; Cash Collection 67; Excess Losses 12; Cost Recovery 89.0
  - Kyrgyz Republic: QFD 0.27; Actual Tariff 7.1; Cost Recovery Tariff 65.5; Gas Import Price 46.3; Cash Collection 98; Excess Losses 26; Cost Recovery 84.5
  - Moldova: QFD 0.96; Actual Tariff 6.8; Cost Recovery Tariff 65.0; Gas Import Price 60.0; Cash Collection 78; Excess Losses 8; Cost Recovery 78.8
  - Tajikistan: QFD 1.35; Actual Tariff 7.9; Cost Recovery Tariff 63.7; Gas Import Price 48.4; Cash Collection 55; Excess Losses 47; Cost Recovery 75.5
  - Ukraine: QFD 3.14; Actual Tariff 4.1; Cost Recovery Tariff 62.5; Gas Import Price 51.6; Cash Collection 89; Excess Losses 55; Cost Recovery 59.6
  - Uzbekistan: QFD 10.7; Actual Tariff 11.1; Cost Recovery Tariff 25.0; Gas Import Price --; Cash Collection 50; Excess Losses 72; Cost Recovery 0.6
- Table 6 (Contributions to the Gas QFD in 2002, based on Table 5) — Underpricing / Non-payment / Losses (In percent of GDP):
  - Armenia: -0.5 / -0.9 / 0.3 / 0.1
  - Azerbaijan: 1.8 / 0.9 / 0.5 / 0.3
  - Georgia: 0.2 / -0.7 / 0.7 / 0.2
  - Kyrgyz Republic: 0.2 / -0.2 / 0.0 / 0.4
  - Moldova: 0.9 / -0.1 / 0.9 / 0.1
  - Tajikistan: 1.3 / 0.2 / 1.0 / 0.1
  - Ukraine: 3.1 / 2.1 / 0.6 / 0.4
  - Uzbekistan: 10.7 / 7.0 / 2.8 / 0.9
- Source commentary:
  - Mispricing is a smaller problem in the gas sector than in the electricity sector.
  - Several countries reported gas tariffs above cost recovery levels.
  - Payment discipline and loss prevention in gas appear better than in electricity.
  - Cash collections for gas were generally higher than for electricity, between 50 and 90 percent of the billing.
  - Excess losses in gas were highest in the Kyrgyz Republic.

### Macroeconomic relevance of Quasi-Fiscal Deficits (QFDs)
- QFDs have several negative macroeconomic and structural implications:
  - undermine financial discipline in government agencies and households, weakening fiscal policy effectiveness and narrowing fiscal space;
  - keep nonviable enterprises afloat, delaying enterprise restructuring and undermining economy-wide productivity growth;
  - distort relative prices, reducing incentives to save energy;
  - distort resource allocation by locking resources in energy and capital intensive production, crowding out labor intensive small and medium-size enterprises;
  - increase energy-importing countries’ external vulnerability by creating a wedge between domestic and international prices;
  - distort the size of the government in the economy.
- Financing patterns of energy companies’ imbalances typically include:
  - budgetary transfers, wage arrears, arrears to domestic input suppliers, external arrears, and capital depreciation.
- Consequences of failing to eliminate QFDs:
  - gradual decay of the capital stock and increasing supply shortages; effective capacity of power generation may have shrunk below the optimal level despite previously large sector size.
- QFDs are associated with energy inefficiencies:
  - Figures and regressions in the source show strong correlations between electricity QFD in percent of GDP and electricity consumption per PPP GDP (sample regressions with high R2: R2 = 0.9458 for CIS 2002 sample; R2 = 0.7459 for selected CEE and CIS 2001 sample).
  - A regression in Figure 7 for selected CEE and CIS countries: Inefficiency = -0.058x + 0.72; R2 = 0.377; t-statistics reported in source (t-stat for coefficients: (-3.48)).
  - Low electricity tariffs correlate strongly with high QFDs (Figures 8 and 9 in source).
- Footnotes and empirical notes from source:
  - The functional form of regression lines was chosen on basis of explanatory power.
  - Energy-related external debt has been a significant problem in Armenia, Moldova, Georgia, and Ukraine.

### IMF conditionality on the energy sector: scope, types, and intensity (1993–2003)
- Overall conditionality overview:
  - 155 performance criteria, structural benchmarks, and prior actions were imposed for the eight countries in the sample in 1993–2003.
  - Intensity of conditionality peaked in 1998 and 1999; Ukraine accounted for more than half of the number of conditions during the peak period.
  - In 2001–03 the volume of conditions fell to less than half of that in the peak years, mainly due to downscaling in Ukraine.
  - Only Georgia and Tajikistan had more conditionality in 2002–03 than in previous years.
- Rationale for changing conditionality intensity:
  - Complexity of energy sector reforms made comprehensive programs difficult to monitor; many measures fell outside IMF core expertise.
  - World Bank took lead on many energy sector reform measures; Fund conditionality was streamlined after 2001 guidelines.
  - Evidence of declining marginal compliance with increasing number of conditions (Figure 5; noted relationship significant even excluding Ukraine).
- Types of conditionality measures (1993–2003) and counts (Figure 6):
  - Tariffs: 40 (26%)
  - Collections: 14 (9%)
  - Privileges: 17 (11%)
  - Arrears: 8 (5%)
  - Financial Balance: 16 (10%)
  - Privatization: 19 (12%)
  - Restructuring: 14 (9%)
  - Other: 18 (12%)
  - Debt: 9 (6%)
- Focus of IMF conditionality by country:
  - Ukraine: most comprehensive; 60% of measures applied to cost recovery; one-fifth addressed privatization, regulation, sector restructuring.
  - Armenia, Georgia, Tajikistan: narrower conditionality focused on tariffs, bill collection, energy use privileges.
  - Azerbaijan: majority of measures on pricing and transparency.
  - Kyrgyz Republic: earlier extensive conditionality declined after 2000 due to new World Bank program.

### Performance record of conditionality and observed outcomes
- Aggregate compliance:
  - Aggregate compliance rate under Fund conditionality was 61 percent in 1993−2003.
  - Excluding Ukraine, aggregate compliance was 66 percent (Total excl. Ukraine 92 total measures, 61 implemented? Source table: Total 155, implemented 94, 61% overall; Total excl. Ukraine 92, implemented 61, 66%).
- Country-level illustrative compliance and observations (from source tables and discussion):
  - Azerbaijan: total 13, implemented 9, compliance 69%
  - Armenia: total 12, implemented 6, compliance 50%
  - Georgia: total 16, implemented 10, compliance 63%
  - Kyrgyz Republic: total 15, implemented 10, compliance 67%
  - Tajikistan: total 13, implemented 9, compliance 69%
  - Moldova: total 20, implemented 17, compliance 85%
  - Ukraine: total 63, implemented 33, compliance 52%
  - Uzbekistan: total 3, implemented 0, compliance 0%
  - Overall total: total 155, implemented 94, compliance 61%
- By-type compliance (examples from source Table 7/8 summary):
  - Tariffs: 40 total, 20 implemented, 50% compliance
  - Bill collection: 14 total, 5 implemented, 36% compliance
  - No arrears: 8 total, 5 implemented, 63% compliance
  - Reduce energy privileges: 17 total, 10 implemented, 59% compliance
  - Financial balance: 16 total, 12 implemented, 75% compliance
  - Privatization: 19 total, 12 implemented, 63% compliance
  - Foreign energy debt: 9 total, 9 implemented, 100% compliance
  - Regulation, restructuring: 14 total, 10 implemented, 71% compliance
  - Other: 18 total, 11 implemented, 61% compliance
- Observations and implications:
  - No significant correlation found between compliance rate and QFD outcomes across sample countries.
    - Examples: Moldova had high compliance and relatively low QFD; Tajikistan had high compliance but a large QFD in 2002; Armenia had low compliance but low energy sector deficit; Uzbekistan implemented none of measures but had few conditionalities; Ukraine implemented many measures (3–4 times other countries) yet QFD remained high in early 2000s.
  - Compliance was lowest in bill collection and tariffs—the most politically sensitive areas.
  - The lack of significant correlation between compliance and QFD outcomes raises doubts about conditionality design, though limited critical mass of reforms may also explain weak outcomes.

### Specific program experiences and selected measures
- Tariff policies:
  - About 25 percent of all measures addressed tariffs; only half of those were implemented.
  - Residential electricity tariffs in most CIS countries were below other transition countries’ average in 2003; in extreme Central Asia cases residential tariffs would need to be raised tenfold to reach Central Europe average tariff level.
  - By 2002, only Armenia, Georgia, and Moldova had raised electricity tariffs to cost recovery levels.
  - Regression (Figure 9) in source: QFD in % of GDP regressed on actual tariff as ratio of average production cost: y = -0.20x + 23.01; R2 = 0.93; (coefficient t-stat -9.09 reported).
- Country program notes:
  - Armenia: under the 1996 ESAF program, residential and industrial tariffs were raised to cost recovery levels; earlier 1995–96 STF and SBA tariff conditionality not observed.
  - Georgia: increase and unification of household and industrial electricity tariffs implemented as structural benchmark in 1997, later deviated; two subsequent benchmarks aiming at improved tariff structure were delayed then dropped from IMF program and implemented under World Bank conditionality in 2001; a surcharge to reduce energy-related external debt was legislated as a prior action but surcharge later not carried out as a structural benchmark.
  - Moldova: tariff increases implemented in 1996 and 1998 based on adding a debt surcharge to energy costs; later increases under World Bank conditionality.
  - Kyrgyz Republic: prior action on increase of power tariffs observed in early 2002; electricity tariffs increased by 20 percent on April 1, 2002.
  - Kyrgyz Republic PRGF programs: 2001 set monitoring of the QFD under a review clause and scheduled tariff increase in 2002; 2003 program included structural benchmark on semi-annual reporting and an indicative target for the QFD (electricity sector QFD programmed to decline from 12 percent of GDP in 2002 to 11 percent in 2003, which was achieved).
- Financial balance monitoring:
  - Financial balance of energy sector seldom directly addressed in IMF programs.
  - Armenia: energy sector financial balance subject to quarterly ceilings in 1996–2001; difficulty measuring and forecasting deficits led to downgrading of the performance criterion to an indicative target in 2002.
  - Moldova: only a non-quantified commitment to improve financial performance was a program benchmark in 1996.

### Policy-relevant conclusions and implications (from source discussion)
- Tariff policy is the most effective tool to address QFDs despite political sensitivities; empirical evidence in the source shows strong negative correlation between tariffs/cost-recovery ratios and electricity QFD.
- Improving billing coverage, metering, and reducing meter tampering and theft are critical because weak billing coverage increases measured excess loss rates and contributes to QFD.
- Addressing non-payment and technical/commercial losses is essential, particularly in countries where losses (e.g., Kyrgyz Republic) are a dominant contributor to QFD.
- Comprehensive energy sector reform requires coordinated approaches across pricing, collection, metering, privatization, regulation, and restructuring; however, IMF conditionality should be parsimonious and coordinated with World Bank responsibilities for effective monitoring and implementation.

*Source: Authors' estimates and IMF country desk data as presented in _wp0643 - 1.7 cents for nuclear power. See http://www.nei.org (PDF content provided).*

### 2000. Under the pressure from the Fund and the World Bank—but without formal

### _wp0643 - 2000. Under the pressure from the Fund and the World Bank—but without formal

### Tariff reform and quasi-fiscal deficits (QFDs)
- Further tariff increases were implemented in 2002 despite strong political resistance.
- Tariff reform in Ukraine faced opposition, especially in the early program years, and many performance criteria were missed; by 2002 many enterprises had become profitable and paid higher tariffs.
- In Tajikistan, household tariffs were raised in 1998, but the exchange rate depreciation after the 1998 Russian crisis largely eroded the gains in cost recovery.
- In 2000, scheduled electricity tariff increases in Tajikistan were not implemented; in 2002 gas tariffs were increased and a commitment was given to periodically adjust all utility tariffs.
- Figure 8 regression reported: y = -3.58x + 19.25 ( -4.50 t-stat), R2 = 0.77.

### Bill collection, losses, and their relation to QFDs
- Bill collection rates and excess losses do not seem to correlate significantly with QFDs (Figures 10 and 11).
- Tajikistan: reported electricity bill collection rate of 86 percent in 2002 but QFD remained the highest among comparator countries because of very low tariffs.
- Georgia: collection rate was 35 percent but tariffs were high, reflected in only a moderate QFD.
- Low reported losses do not necessarily imply low QFDs as evidenced by Tajikistan, Azerbaijan, and Uzbekistan.
- Under Fund conditionality, measures on bill collection often failed; average compliance rate during 1993–2003 was only 36 percent.
- Georgia: 1995 structural benchmark to raise electricity bill collections to 60 percent of sales was not achieved; reintroduced at 70 percent in 2002 program but again without success.
- Armenia: cutting off nonpaying customers successfully introduced in 1996 contributing to higher collection rates.
- Ukraine: collection rates improved in 2000 after a strong cut-off policy.
- Tajikistan: mid-2002 gas bill collection target observed; in 2003 such a target was dropped and replaced by other energy sector reform measures.
- Kyrgyz Republic and Moldova: programs did not have conditionality on bill collections.
- No program included specific conditionality on loss prevention.
- Figure 11 axes: Cash Collection Rate in % of Bill (0–100); QFD in % of GDP (0–24).

### Energy use privileges and social protection
- Removing tariff discounts was politically difficult; many privileges inherited from the Soviet era were categorical and lacked means testing.
- Armenia removed categorical privileges for heating and hot water in 1995; a benchmark to convert exemptions on electricity payments to explicit income support failed in 1996.
- Kyrgyz Republic: benchmark to replace privileges with direct budgetary transfers to low-income households failed in 1999; a 2002 law reduced the number of privileged users only by 20 percent.
- Moldova: two program conditions in 1998 to eliminate energy privileges met strong political opposition; only in 2000 were energy privileges streamlined as a prior action for the PRGF program, paving the way for further tariff increases under the World Bank program supplemented by direct budget subsidies to the poor.
- Ukraine: attempts to remove untargeted subsidies on energy and abolish tax exemptions to energy companies failed in 1998.
- Tajikistan: program aimed at eliminating energy privileges in 2003 and this was achieved as a prior action.

### Payment arrears and external debt measures
- Few programs had explicit conditionality on domestic energy arrears.
- Armenia 1999 program included a prior action on completion of restructuring of energy sector arrears to domestic creditors; measure only partially implemented.
- Ukraine introduced targets on verification and clearance of arrears of budgetary organizations to energy companies in 1998; full clearance was not achieved; budgetary netting operations and restructuring of local government energy arrears occurred over the next five years.
- Kyrgyz PRGF programs included performance criteria on the stock of budgetary arrears to KyrgyzEnergo since early 2000; these targets were generally observed.
- Since approval of the three-year PRGF arrangement in late 2001, no new arrears of the central government to KyrgyzEnergo piled up under the continuous performance criterion; offsetting tax liabilities against energy bills continued.
- Conditionality applied to energy-related external debt in some highly indebted countries:
  - Georgia: adoption of a strategy for netting out energy debts was a prior action for concluding the second review under the PRGF in 2002; implementation was put on hold.
  - Tajikistan: prior action advising foreign energy suppliers of no government guarantees on energy-related debt.
  - Moldova: external debt owed to Russian Gazprom was in part rescheduled and in part swapped for Gazprom’s majority equity in Moldovagas.

### Privatization, regulation, and market reforms
- Privatization under Fund conditionality was partially successful in Georgia, Moldova, and Ukraine.
- Georgia: 1997 prior action on issuance of a privatization plan for Telasi distribution company was observed; a structural benchmark on a private management contract for transmission was introduced in the second review under the PRGF in July 2002 and observed with some delay in early 2003.
- Moldova: program measures for privatizing Moldovagas, five electricity distribution companies, and power generation in 1999 and 2000; three distribution companies were sold to the Spanish firm Union Fenosa in early 2000; other sales were unsuccessful and conditionality on privatization was dropped as it became part of the World Bank SAL operation.
- Ukraine: 1999 structural benchmark to establish legal framework and complete preparatory process for privatization of several large enterprises, including power companies, was observed; subsequent energy privatization conditionality covered by the Bank program.
- Ukraine (1999–2001): Fund conditionality focused on improving transparency of Naftogas Ukraine including by setting up a new auditing mechanism; target to publish audited data was not observed.
- Kyrgyz Republic: 2000 conditionality on privatization of three energy companies (KyrgyzGasMunaizat, KyrgyzGas, and Munai) including audits and tenders was not implemented; an external debt strategy adopted in July 2001 included privatization of four power distribution companies with proceeds for external debt reduction; action plan approved with minor delay but no progress in privatizing the companies despite World Bank CAS support.
- Other measures: Ukraine EFF targets on conducting gas auctions and liberalizing gas sales/imports; Armenia 1995 structural benchmark on liberalizing oil product and electricity market not observed satisfactorily; Tajikistan prohibition on new barter trade contracts (except for aluminum) issued in 1996 as a prior action but barter continued.

### Outcomes and lessons learned
- By 2001, energy inefficiencies in the sample CIS countries were still high after ten years into reforms (Table 9).
- Overall energy use efficiency improved in all countries between 1992 and 2001; progress in the electricity sector was less significant.
- Some countries backtracked or made virtually no progress in electricity efficiency (Azerbaijan and Tajikistan); Armenia and Georgia achieved large gains.
- Possible explanation: electricity is less tradable than other energy forms (notably oil products); law of one price does not seem to apply to electricity as reflected in different actual tariff levels.
- Compliance rate of Fund conditionality on energy sector reforms was 61 percent, lower than average compliance on Fund structural conditionality.
- Table 9: Energy Inefficiency for Selected CIS Countries (1992 and 2001)
  - Armenia: Energy Use per PPP GDP 1992 = 0.71; Electricity consumption per PPP GDP 1992 = 1.03; Energy Use per PPP GDP 2001 = 0.27; Electricity consumption per PPP GDP 2001 = 0.41
  - Azerbaijan: 1992 = 0.76; 1992 = 0.55; 2001 = 0.50; 2001 = 0.64
  - Georgia: 1992 = 0.85; 1992 = 0.95; 2001 = 0.22; 2001 = 0.34
  - Kyrgyz Republic: 1992 = 0.69; 1992 = 1.11; 2001 = 0.28; 2001 = 0.82
  - Moldova: 1992 = 0.82; 1992 = 0.98; 2001 = 0.53; 2001 = 0.57
  - Tajikistan: 1992 = 1.30; 1992 = 2.36; 2001 = 0.53; 2001 = 2.34
  - Ukraine: 1992 = 0.68; 1992 = 0.58; 2001 = 0.63; 2001 = 0.49
  - Uzbekistan: 1992 = 1.57; 1992 = 1.40; 2001 = 1.26; 2001 = 1.01
  - Simple Average: 1992 = 0.92; 1992 = 1.12; 2001 = 0.53; 2001 = 0.83
  - Memorandum Item: Europe & Central Asia 1992 = 0.59; 1992 = 0.58; 2001 = 0.41; 2001 = 0.42
  - Memorandum Item: Low Income 1992 = 0.35; 1992 = 0.18; 2001 = 0.25; 2001 = 0.15
  - Memorandum Item: Middle Income 1992 = 0.38; 1992 = 0.34; 2001 = 0.24; 2001 = 0.26
  - Memorandum Item: OECD 1992 = 0.20; 1992 = 0.37; 2001 = 0.19; 2001 = ...
- Achieving a breakthrough in eliminating QFDs is difficult because vested interests in the energy sector would lose from reform; weak government, poor regulatory quality, and corruption are often combined with low energy-sector efficiency (Figure 12).
- Figure 12 regression reported: y = 3.1435x -0.5938, R2 = 0.4094.
- Reform inertia often generated by former state-owned enterprises capturing the state; anecdotal evidence of rent-sharing (e.g., electricity meter inspectors accruing unrecorded collections).
- Slow progress in reducing QFDs reflects government sensitivity to social costs; EBRD affordability ratios were referenced (text cut off before completing the affordability statistic).

*Source: Authors' estimates; World Economic Outlook (IMF, 2005) and World Development Indicators (World Bank, 2004); figures and tables as presented in the source text.*

### 3.4 percent of household income in the CIS region in 2000 (Table 10).

### _wp0643 - 3.4 percent of household income in the CIS region in 2000 (Table 10)

### Affordability and cost recovery
- Electricity expenditure in percent of household income (actual / at LRMC prices 1/):
  - Armenia 8.4 / 11.1
  - Azerbaijan 3.8 / 9.3
  - Georgia 12.4 / 16.7
  - Kyrgyz Republic 1.9 / 9.1
  - Moldova 5.2 / 7.3
  - Tajikistan 0.2 / 6.6
  - Ukraine 2.9 / 5.8
- Memorandum items:
  - Russia 1.0 / 0.5
  - Baltics 2.3 / 2.6
  - CIS 3.4 / 6.6
  - Central Europe 3.6 / 3.5
- To cover the Long Run Marginal Cost (LRMC) of production, payments would need to increase to 6.6 percent of household income.
- EBRD LRMC reference: 8 U.S. cents per kilowatt hour—the average retail price in the United States.

### Bill collection, tariffs, and empirical relationships
- Tariff increase and collection example:
  - Kyrgyz Republic: tariffs increased by 20 percent in April 2002; cash collections declined to 32 percent of billing in 2003, down from 58 percent in 2001.
- Cross-country evidence:
  - Scatter analysis reported: apparent positive correlation between tariffs and bill collection (figure regression line y = 5.9954x + 56.148, R^2 = 0.1858).
  - CIS sample fitted cubic: y = -5.0096x^3 + 45.064x^2 - 110.94x + 119.15, R^2 = 0.3487 (Figure 14).
- Suggested behavioral pattern:
  - Possible U-shaped relationship between tariffs and bill collection: at low tariffs collection can be high; as tariffs rise but remain below cost recovery, quality may not improve and incentives to pay decline; once tariffs reach levels sufficient to finance visible quality improvements, incentive to comply may increase.
- Operational impediments to collection:
  - Inability to disconnect noncomplying customers (technical limits and lists of “strategic” customers).
  - Meter manipulation and inspectors’ collusion; countermeasures cited: relocation of meters in locked boxes, separation of technical meter reading from billing, mandatory payments through banks or collection centers, computerization of consumer accounts.

### Privatization, governance, and energy debts
- Mixed outcomes of privatization and governance reforms:
  - Georgia: privatization of local power distribution company reduced blackouts in Tbilisi but investor financing cut off due to high technical/commercial losses.
  - Moldova: three of five distribution companies sold to a Spanish investor in early 2000; supply cuts enforced to nonpayers reduced interruptions in central and southern Moldova; subsequent partial renationalization steps taken.
  - Armenia: sale attempts failed in 2001; strategy shifted to restructuring and management contracts; one distribution company privatized in August 2002.
  - Kyrgyz Republic and Tajikistan: privatizations did not attract strategic investors due to low tariffs and regional cooperation problems.
- Evidence that payment problems can be solved with good corporate governance even when companies remain state-owned (examples: Hungary, Bulgaria, Poland).
- Energy-related debt and restructuring:
  - Armenia restructured 85 percent of energy companies’ domestic debt by end-1999; external energy debt reduced in 2002 via a debt-equity swap with Russia.
  - Georgia: debt-equity swap in 2002 involving a gas trading company associated with Gazprom.
  - Ukraine developed a debt-restructuring plan for energy companies but had not implemented it by end−2003.

### Policy options (ranked by ambition and recommended sequencing)
- First-best (most ambitious):
  - Include an audit-based quasi-fiscal deficit (QFD) of the energy sector as a separate performance criterion in Fund-supported programs.
  - Preconditions: reliable business accounting, application of International Accounting Standards (IAS), development of the auditing profession.
  - Benefits: covers all sector operations, reduces scope for evasion, improves transparency, and increases government flexibility in choosing measures.
- Second-best:
  - Set the energy sector’s QFD as a performance criterion using the end-user approach as a yardstick (suitable where accounting systems are less developed).
  - Complementary steps: energy enterprises audited by internationally reputable firms and adoption of IAS to move toward the financial balance approach.
  - Include the deficit in annual budget documentation to reveal fiscal trade-offs.
- Third option (governance-focused, lower ambition):
  - Reduce influence of vested interests via governance improvements without directly addressing QFD.
  - Fund conditionality: make cost of QFD explicit in the budget; cover budgetary flows related to energy (zero tolerance of arrears on energy bills; taxation of utility consumption; restructuring of tax arrears and external debts; social transfers in exchange for canceling discounted tariffs).
  - Leave tariff reform, cash collections, losses, and management under World Bank conditionality (or under Fund conditionality only if no Bank operation exists).
- Final (crisis-driven) option:
  - If ownership and political will are lacking, reforms may occur only through severe supply crises (blackouts), which can raise public support for tariff increases but at higher adjustment costs and uncertain timing.
- Social protection and sequencing:
  - Vulnerable populations must be compensated for higher energy costs.
  - General wage increases may protect average wage earners if inflation is contained and real wages rise.
  - Link energy cost changes to social safety nets in the budget to increase transparency; discuss and approve both measures simultaneously in parliament.
  - Social compensation for short-term losers should not be viewed as a fiscal stability risk given long-term gains from energy reform.

### Epilogue — selected country developments (post-2002 estimates and program updates)
- Armenia:
  - Separate companies created for financial settlements, electricity dispatch, and high voltage distribution; independent boards appointed.
  - Tariffs covered costs already in 2002; bill collection rates reported at 100 percent; losses declined further; QFD in electricity virtually eliminated (remaining issue: historical debts).
- Azerbaijan:
  - Natural gas cost recovery ratio reportedly rose to 85 percent by end-2004 from 50 percent in 2002; bill collection rates remain at around 50 percent.
  - Electricity: SOCAR provides underpriced fuel and gas inputs to power generation; tariffs below cost recovery and payment collection ineffective.
- Georgia:
  - Bill collection rose from 35 percent in 2002 to over 50 percent in 2004 due to forceful disconnection policy and expanded metering.
  - Cost recovery ratio declined to 94 percent as tariff increases lagged cost increases.
  - 2004 PRGF-supported program introduced indicative targets for bill collections of electricity and gas.
- Kyrgyz Republic:
  - QFD of the electricity sector declined from 12 percent of GDP in 2002 to below 9 percent in 2004 (estimate may not capture increases in cost recovery level from rising world oil prices).
  - New PRGF-supported program approved in early 2005 continues to include QFD as an indicative program target.
- Moldova:
  - Energy-related foreign supplier debt remains sizeable; dependency on natural gas imports leaves the country vulnerable to import price fluctuations.
  - Collection and pricing losses persist; reforms remain incomplete.
- Tajikistan:
  - Gas tariffs at cost recovery level; power tariffs do not cover costs despite 2003 hikes.
  - World Bank estimates suggested tariffs would have had to increase fourfold in 2004 to achieve cost recovery.
  - Authorities estimate combined electricity and gas QFD may have declined to nearly half the level observed in 2002 after some tariff rationalization.
  - 2005 PRGF-supported program introduced a structural benchmark for auditing the TajikGas monopoly.
- Ukraine:
  - Electricity bill collection reportedly achieved 90 percent at end-2004.
  - Better cost recovery achieved due to increased industrial consumption where tariffs are highest.
  - Key remaining test: improve the cost recovery ratio in gas supplies (footnote: based on economic costs including the gas import price approximately $54/tcm, local transport and gas conversion fees, and depreciation cost estimated at about $10/tcm—Table 5—but excluding long-term expansion, the current average gas tariff covers around 80 percent).
- Uzbekistan:
  - Authorities reportedly increased electricity tariffs significantly in 2003–04, raising the cost recovery ratio from 28 percent in 2002 to about 70 percent in 2004; collection rates improved.

*Source: IMF working paper content (section and tables as provided).*

### Appendix Table 1. Measures a

### Appendix Table 1. Measures a

### Measures and Implementation by Country (selected entries and aggregates)
- Azerbaijan
  - 4 total, 3 Impl, 75% Compliance
  - 4 total, 2 Impl, 50% Compliance
  - 1 total, 1 Impl, 100% Compliance
  - 2 total, 1 Impl, 50% Compliance
  - 2 total, 2 Impl, 100% Compliance
- Armenia
  - 2 total, 1 Impl, 50% Compliance
  - 3 total, 1 Impl, 33% Compliance
  - 1 total, 1 Impl, 100% Compliance
  - 2 total, 1 Impl, 50% Compliance
  - 0 total, 0 Impl, ...
  - 3 total, 2 Impl, 67% Compliance
- Georgia
  - 0 total, 0 Impl, ...
  - 5 total, 4 Impl, 80% Compliance
  - 3 total, 1 Impl, 33% Compliance
  - 1 total, 1 Impl, 100% Compliance
  - 1 total, 1 Impl, 100% Compliance
- Kyrgyz Republic
  - 3 total, 3 Impl, 100% Compliance
  - 1 total, 1 Impl, 100% Compliance
  - 3 total, 2 Impl, 67% Compliance
  - 2 total, 1 Impl, 50% Compliance
  - 1 total, 1 Impl, 100% Compliance
- Tajikistan
  - 0 total, 0 Impl, ...
  - 6 total, 4 Impl, 67% Compliance
  - 1 total, 1 Impl, 100% Compliance
  - 4 total, 2 Impl, 50% Compliance
  - 1 total, 1 Impl, 100% Compliance
- Moldova
  - 1 total, 1 Impl, 100% Compliance
  - 3 total, 3 Impl, 100% Compliance
  - 2 total, 2 Impl, 100% Compliance
  - 4 total, 4 Impl, 100% Compliance
  - 10 total, 7 Impl, 70% Compliance
- Ukraine
  - 6 total, 4 Impl, 67% Compliance
  - 15 total, 5 Impl, 33% Compliance
  - 9 total, 2 Impl, 22% Compliance
  - 3 total, 2 Impl, 67% Compliance
  - 4 total, 2 Impl, 50% Compliance
  - 12 total, 10 Impl, 83% Compliance
  - 14 total, 8 Impl, 57% Compliance
- Uzbekistan
  - 0 total, 0 Impl, ...
  - 0 total, 0 Impl, ...
  - 0 total, 0 Impl, ...

Aggregates (selected)
- Total (selected rows)
  - 16 total, 12 Impl, 75% Compliance
  - 40 total, 20 Impl, 50% Compliance
  - 14 total, 5 Impl, 36% Compliance
  - 8 total, 5 Impl, 63% Compliance
- Total excl. Ukraine (selected rows)
  - 10 total, 8 Impl, 80% Compliance
  - 25 total, 15 Impl, 60% Compliance
  - 5 total, 3 Impl, 60% Compliance
  - 5 total, 3 Impl, 60% Compliance
- Another block of totals
  - Total: 17 total, 10 Impl, 59% Compliance
  - Total: 9 total, 9 Impl, 100% Compliance
  - Total: 33 total, 22 Impl, 67% Compliance
  - Total: 18 total, 11 Impl, 61% Compliance
  - Total excl. Ukraine: 13 total, 8 Impl, 62%; 9 total, 9 Impl, 100%; 21 total, 12 Impl, 57%; 4 total, 3 Impl, 75%

### Measures and Implementation by Type (Appendix Table 3)
- Tariffs
  - 20 total, 11 Impl, 55% Compliance
  - 7 total, 4 Impl, 57% Compliance
  - 13 total, 5 Impl, 38% Compliance
  - 40 total, 20 Impl, 50% Compliance
- Bill collection
  - 5 total, 2 Impl, 40% Compliance
  - 8 total, 2 Impl, 25% Compliance
  - 1 total, 1 Impl, 100% Compliance
  - 14 total, 5 Impl, 36% Compliance
- No arrears
  - 3 total, 2 Impl, 67% Compliance
  - 0 total, 0 Impl, ...
- Reduce privileges
  - 8 total, 6 Impl, 75% Compliance
  - 1 total, 1 Impl, 100% Compliance
  - 8 total, 3 Impl, 38% Compliance
  - 17 total, 10 Impl, 59% Compliance
- Financial Balance
  - 7 total, 5 Impl, 71% Compliance
  - 1 total, 1 Impl, 100% Compliance
  - 8 total, 6 Impl, 75% Compliance
  - 16 total, 12 Impl, 75% Compliance
- Privatization
  - 13 total, 8 Impl, 62% Compliance
  - 5 total, 3 Impl, 60% Compliance
  - 1 total, 1 Impl, 100% Compliance
  - 19 total, 12 Impl, 63% Compliance
- Foreign debt rescheduling
  - 1 total, 1 Impl, 100% Compliance
  - 6 total, 6 Impl, 100% Compliance
  - 2 total, 2 Impl, 100% Compliance
  - 9 total, 9 Impl, 100% Compliance
- Regulation, restructuring
  - 4 total, 2 Impl, 50% Compliance
  - 10 total, 8 Impl, 80% Compliance
  - 0 total, 0 Impl, ...
  - 14 total, 10 Impl, 71% Compliance
- Other
  - 4 total, 3 Impl, 75% Compliance
  - 14 total, 8 Impl, 57% Compliance
  - 0 total, 0 Impl, ...
  - 18 total, 11 Impl, 61% Compliance

Type aggregates
- Total by type
  - 65 total, 40 Impl, 62% Compliance
  - 52 total, 33 Impl, 63% Compliance
  - 38 total, 21 Impl, 55% Compliance
  - 155 total, 94 Impl, 61% Compliance
- Total excl. Ukraine by type
  - 52 total, 35 Impl, 67% Compliance
  - 16 total, 12 Impl, 75% Compliance
  - 24 total, 14 Impl, 58% Compliance
  - 92 total, 61 Impl, 66% Compliance

### Measures and Implementation by Energy Subsector (Appendix Table 4 — selected items)
- Tariffs (subsector-level excerpts)
  - Entries include "31......", "10 10", "33%", "80%", "100%", "67%", "50%", "33%", "0%"
- Bill collection
  - Instances showing "1 total, 1 Impl", "100%", and other fragmented entries
- No arrears
  - Instances showing "1 total, 0 Impl", "0%", and "3 total, 2 Impl", "67%"
- Reduce privileges
  - Instances showing "2 total, 1 Impl", "50%", and "1 total, 1 Impl", "100%"
- Financial Balance
  - Instances showing "2 total, 1 Impl", "50%", and "3 total, 3 Impl", "100%"
- Privatization
  - Multiple fragmented entries, including "0 total, 0 Impl", "5 total, 2 Impl", "40%", "67%"
- Foreign debt rescheduling
  - Multiple instances with "0 total, 0 Impl" and "1 total, 1 Impl", "100%"
- Regulation, restructuring
  - Multiple fragmented entries, including "3 total, 1 Impl", "33%" and "0 total, 0 Impl"
- Other
  - Instances indicating "3 total, 2 Impl", "67%" and other fragmented values

Notes within subsector tables
- Text fragments and ellipses (...) indicate incomplete entries in the table as presented.
- Footnote: "1/ Reflects measures reviewed under the Third Review of th e PRGF arrangement although this review was never completed." (as shown in table)

### Conditionality by years (Appendix Table 5 series — selected country summaries)
- Armenia (Appendix Table 5a)
  - Subcategory examples: Tariffs, Bill collection, No arrears, Reduce privileges, Financial Balance, Privatization, Foreign debt rescheduling, Regulation, restructuring, Other
  - Example totals: 13 total, 9 Impl, (69% shown in context)
- Georgia (Appendix Table 5b)
  - Example totals: 20 total, 17 Impl, 85% (selected row)
- Kyrgyz Republic (Appendix Table 5c)
  - Example totals: 63 total, 33 Impl, 52% (selected row)
- Tajikistan (Appendix Table 5d)
  - Example totals: 13 total, 9 Impl, 69% (selected row)
- Moldova (Appendix Table 5e)
  - Example totals: 30 total, 0 Impl, 0% (selected row)
- Ukraine (Appendix Table 5f)
  - Entries include many zeros across categories for the presented year range
- Azerbaijan (Appendix Table 5g)
  - Example totals: 30 total, 0 Impl, 0% (selected row)
- Uzbekistan (Appendix Table 5h)
  - Year columns include 1993 through 2002 with fragmented zeros and ellipses

Year notes and fragments
- Years referenced across tables include 1993, 1994, 1995, 1996, 1997, 1998, 1999, 2000, 2001, 2002, 2003.
- Several cells contain "..." indicating missing or nonreported entries in the provided extract.
- Selected explicit year-note: "The target on submission of a semi-annual report on the electricity sector QFD (1st half 2003) was met. The indicative targe t on the electricity QFD for end-2003 was also met (not included)."

*Source: Appendix tables and extracts as provided in the supplied content unit.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0643.pdf_