## INFLATION DETERMINANTS AND IMPLICATIONS IN BULGARIA

## Source details

**Canonical URL:** [INFLATION DETERMINANTS AND IMPLICATIONS IN BULGARIA](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2015/_cr15120.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2015/_cr15120.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2015/_cr15120.pdf.json)

---

### A. Introduction
- Bulgaria experienced sustained deflation since mid-2013; HICP headline inflation has been negative since August 2013.
- Bulgaria’s deflation is among the deepest in the EU since mid-2013, except for Greece and Cyprus.
- Key proximate factors: inward spillover of low inflationary pressure from the EU, consecutive electricity price cuts in 2013, and fast falling global commodity prices since late 2014.
- Analysis period: January 2012 to February 2015 (regressions use monthly data for 2012–14; sample period 2012m1–2014m12).

### B. Stylized facts
- Four phases of inflation since 2012:
  - 2012: headline inflation close to EU average.
  - Early 2013–early 2014: sharp decline coinciding with administered price reductions.
  - Rest of 2014: broadly stable in negative territory.
  - End-2014 to Feb-2015: deepening deflation then uptick in Feb-2015, in line with EU movements.
- Electricity price cuts in 2013:
  - Cumulative 15 percent reduction for residential households (three cuts).
  - March 5, 2013: between 6.2–7.3 percent cut depending on location.
  - August 1, 2013: daytime tariffs reduced by between 4.2–4.9 percent; night-time by between 3.4–6.8 percent.
  - January 1, 2014: daytime tariffs reduced by 1 percent; night-time rate cut by 10 percent; industrial consumers’ prices reduced by 1.5 percent.
- Administered prices’ role:
  - After the first cut, contribution from administered prices to inflation shrank rapidly.
  - Negative contribution from the second cut offset positive non-administered contributions, pushing headline inflation negative (first time since 2002 administered prices contributed negatively).
  - Administered prices’ deflationary role became negligible after mid-2014; from October 2014 administered prices resumed a positive contribution after electricity prices were raised by around 10 percent.
- Core vs non-core:
  - Deflationary pressure came from both core and non-core components; core components fell below zero slightly ahead of non-core and increased in relative importance since late 2013.
  - Core inflation in Bulgaria tracked closely EU core inflation during the period.

### C. Model and data
- Empirical framework: expectation-augmented Phillips curve using monthly data for 2012–14.
- Regression specification:
  - π = α + β π_{t-1} + δ π^{E} + ρ π^{EU_core} + x θ + ε
  - Variables: π = headline inflation in Bulgaria; π^{E} = inflation expectation (one-year ahead mean from Consensus Economics); π^{EU_core} = EU HICP core inflation; x = vector of external and domestic shocks (world oil and food prices, administered price changes, indicators of domestic demand); lags of dependent variable included.
- Measurement:
  - All variables expressed in 12-month growth rates in percent unless noted.
- Data sources:
  - Eurostat HICP; Consensus Economics expectations; IMF WEO world oil and food price indices; IMF nominal effective exchange rates; domestic indicators (wage growth, retail sales growth in constant prices, employment growth, unemployment rate).

### D. Results (monthly regressions, sample 2012m1–2014m12)
- Main empirical findings:
  - Inflation inertia is modest and fades quickly: one-period lag coefficient statistically significant at 1 percent; second lag very small and insignificant.
  - Fast and sizable transmission from EU core inflation to Bulgaria’s headline inflation: estimated contemporaneous coefficient suggests roughly one-to-one transmission.
  - Commodity prices (world oil and food) are statistically significant determinants, though magnitudes are small.
  - Changes in administered prices have a direct, statistically significant impact on headline inflation (coefficient < 1).
  - Nominal effective exchange rate (NEER) has a modest effect on headline inflation.
  - Domestic demand indicators vary: wage growth significant with small magnitude; retail sales growth not significant; unemployment rate coefficient has unexpected sign; employment growth not significant.
- Selected regression coefficients (S.E. in parentheses; *, **, *** denote 10%, 5%, 1% significance levels):
  - Inflation(-1): 0.421(0.101)*** (other specifications: 0.564(0.114)***, 0.465(0.085)***, 0.542(0.085)***)
  - Inflation expectations (one-year ahead): -0.122(0.218) (other specs: -0.136(0.245), -0.403(0.218)*, -0.097(0.218))
  - Contribution of administered prices: 0.658(0.248)** (other specs: 0.490(0.323), 1.092(0.289)***, 0.484(0.289)*)
  - Wage growth: 0.067(0.026)**
  - Unemployment rate: 0.538(0.156)***
  - EU area core inflation: 1.088(0.456)** (other specs: 1.090(0.528)*, 1.074(0.423)**, 1.220(0.423)**)
  - Nominal effective exchange rate: -0.105(0.057)* (other specs include -0.073(0.082), -0.048(0.054))
  - World oil price inflation: 0.023(0.006)*** (other specs: 0.019(0.006)***, 0.025(0.006)***, 0.021(0.006)***)
  - World food price inflation: 0.032(0.010)*** (other specs: 0.034(0.011)***, 0.017(0.010)*, 0.295(0.010)**)
  - Observations N = 36; R2 = 0.984, 0.980, 0.986, 0.980 across specifications.
- Decomposition of contributions to headline inflation changes:
  - Whole sample (2012m1–2014m12): external factors dominate—combined weight of spillover from low EU core inflation and declining commodity prices exceeds 80 percent.
  - Sub-sample Feb 2013–Jan 2014 (prior to first electricity price cut to after third cut): administered price changes weigh around 35 percent; combined with weak domestic demand yields roughly a 50/50 split between external and domestic factors for that period.
  - Note: contributions excluding lag inflation and inflation expectations are normalized to 100 percent.

### E. Outlook (projection assumptions and sensitivities)
- Baseline projection:
  - Inflation projected to bottom out and start an upward trend in 2015; headline inflation expected to reach positive levels by end-2015.
- Key assumptions:
  - EU core inflation subdued before recovering in late 2015 to 2014Q4 levels.
  - Oil prices stabilize in the first quarter and slightly recover toward end-2015, consistent with the April 2015 IMF World Economic Outlook assumptions.
  - Nominal wage growth kept at 2014 levels.
  - Administered prices kept at end-2014 levels (acknowledging an increase in electricity prices may be required to support cost recovery by the electricity company).
  - Inflation expectations remain stable.
- Sensitivities:
  - A ¼ percentage point reduction in EU core inflation for 2015 will lower Bulgaria’s headline inflation by about 0.3 percentage points.
  - A 2-percent upward adjustment in administered prices starting in 2015Q3 will raise annual inflation in Bulgaria by around 0.2 percentage points (partly reflecting the low base in 2014).

### F. Implications
- Channels through which sustained deflation poses risks:
  - Demand postponement: falling prices may postpone consumption and investment as holding cash yields a positive real return.
  - Real debt increase: deflation increases the real value of debts, worsening debtors’ financial condition and improving creditors’ condition—net effect likely negative on aggregate demand per Fisher (1983).
  - Nominal rigidity and real adjustment: deflation limits room for nominal adjustments, forcing real adjustments; downward nominal wage rigidity could complicate competitiveness adjustments even if nominal wage growth remains at or below levels reflecting inflation and productivity gains.
- Policy-relevant observations:
  - External spillovers (EU core inflation and commodity prices) are the dominant drivers of recent deflationary pressures; policy responses should account for this external exposure.
  - Administered price adjustments (notably electricity tariffs) had large short-run effects; changes in administered prices are an important domestic policy instrument with measurable inflation impact.
  - Monitoring and addressing exchange rate pass-through, commodity price volatility, and domestic wage dynamics are important for near-term inflation management.

*Prepared by Zaijin Zhan; analysis period and regressions based on monthly data for 2012–14 (sample period 2012m1–2014m12).*

### Deflationary risks and constraints under the currency board
- Policy mechanisms to address deflationary pressures are constrained under the currency board framework.
- Key channels through which sustained deflation can harm the economy:
  - Increases real debt burdens and raises real yields on cash holdings, amplifying negative effects of corporate indebtedness on investment decisions.
  - Limits scope for macroeconomic policy to relieve deflationary pressures, making measures to reduce corporate indebtedness more important.

### Impact on private demand, corporate sector, and investment
- Falling prices may have adversely affected private demand for consumption and investment; causal identification is difficult over a short period with multiple shocks.
- Bulgaria’s non-financial corporate (NFC) sector has one of the highest indebtedness levels among NMS.
- Sustained deflation would exacerbate the investment-suppressing effect by increasing real debt burdens.

### Fiscal implications of deflation
- About 1/2 of the fiscal balance deviation in 2014 from the original budget target (or about 1.1 percentage points of GDP) can be attributed to lower inflation than assumed in the original 2014 budget (absent corrective expenditure actions).
- At the 2014 collection level, a one percentage point drop in consumer inflation, other things being equal, leads to a 0.3 percentage point increase in the deficit and, in turn, public debt.
- The sensitivity of fiscal outturns to price assumptions and limited fiscal space implies the need for flexibility to adjust spending if deflationary pressures are more sustained than currently projected.

### Labor market, wages, and competitiveness
- Post-crisis adjustment occurred more on employment than on wages, especially among low-skilled workers, contributing to a sharp increase in unemployment.
- Since early 2013, real wage growth has remained well above productivity growth; nominal wage growth only started to moderate in 2014 but not sufficiently to bring real wages in line with productivity given sustained deflation.
- An appreciation of Bulgaria’s ULC-based REER, including relative to NMS peers, has escalated in recent years; continued appreciation could gradually erode competitiveness.

### Effects of lower commodity prices
- Bulgaria imports about 11–12 percent of GDP in petroleum products, fuel, and gas, of which more than 60 percent is to be re-exported; this and longer-term contracts limit the short-term impact of falling prices.
- Loss-producing state utility companies have kept key administered prices low relative to regional comparators, including through three electricity price reductions during 2013-14.
- Lower commodity prices are expected to give some breathing space to the national electricity company (NEK) to realign its pricing structure more closely with international prices.

### Conclusions — drivers and outlook
- Declining inflation since 2012 and deflation since mid-2013 were due to both external and domestic factors; inward spillover of low inflationary pressure from the EU was an underlying trend, exacerbated by consecutive electricity price cuts in 2013 and fast falling global commodity prices since mid-2014.
- Outlook for 2015:
  - Deflation is expected to lessen gradually in 2015.
  - With assumptions of gradual stabilization of EU core inflation and global commodity prices and administered price changes in line with historical patterns, year-on-year headline inflation is expected to rise gradually, ending the year in positive territory.
  - The outlook is subject to large uncertainties, especially downside risks; the effectiveness of the ECB QE plan in stabilizing core inflation in the EU has a direct impact on Bulgaria’s inflation outlook.
- Observed macroeconomic effects to date:
  - Worsening the fiscal balance.
  - Raising the public and private debt burden.
  - Eroding part of the adjustment in nominal wages.

### Policy recommendations (inflation and deflation response)
- Given limited policy options to influence inflation directly (small open economy, no independent monetary policy, limited fiscal space), priorities include:
  - Clear communication on the sources and temporary nature of current deflationary pressure to help anchor inflation expectations and avoid a deflationary spiral.
  - Clarification of measures to deal with the private sector debt overhang.
  - Tighter expenditure controls to adjust to large uncertainties around the inflation outlook.
  - Pay closer attention to nominal wage levels, tying increases to productivity gains and accounting for the effect of falling prices.
  - Use moderation in energy prices as a window of opportunity to implement reforms in the energy sector, including closer alignment of residential utility tariffs with global prices/cost structures, accompanied by enhanced protection of poor households against tariff adjustments.

### Non-Financial Corporate Debt Overhang — summary diagnostics (as of end-2013)
- Total non-financial sector indebtedness: over 150 percent of GDP on a consolidated basis (third highest among NMS—below only Croatia and Slovenia).
- NFC debt: over 110 percent of GDP in 2013 (highest among NMS and well-above the Euro area average).
- Debt securities: less than 4 percent of GDP of NFC debt; loans dominate.
- Among NFC loans: about 45 percent of GDP consists of loans from Bulgarian banks.
- Foreign direct (largely inter-company) loans: estimated at roughly 40 percent of GDP.

### Liquidity and solvency indicators (consolidated, end-2012)
- Debt-to-income ratio: over 400 percent of income.
- Interest Coverage Ratio (ICR): 5 as of end-2012 (coverage of interest payments by gross operating income); during the global crisis, aggregate ICR fell to as low as 3.
- Implication: sensitivity of corporate balance sheets to macro-financial shocks.

### Solvency measures and leverage indicators (2012 data)
- Gross leverage: about 85 percent of equity.
- Net debt-to-equity (accounting for firms’ assets): about 45 percent.
- Cross-country context: Bulgaria appears less exposed to solvency risk than some NMS such as Croatia, Latvia, and Slovenia.

### Empirical findings on leverage and investment
- NMS panel: higher debt overhang reduces investment with estimated impacts ranging from 1 to 6 percent depending on the debt overhang proxy.
- Bulgaria:
  - ICR significant over 2004–2013: firms’ liquidity risk matters for funding and investment decisions.
  - Debt-to-equity measure not significant for full 2004–2013 sample but becomes significant in the post-crisis period (2011–13).
  - Post-crisis results: consistent negative relationship between leverage and investment across sectors and firm sizes.
- Example NMS dynamic GMM results (selected):
  - Constant: 1.97*** (std. err. 0.44***)
  - IK_it-1: 0.07** (std. err. -0.03***)
  - SK_it-1: -0.04*** (std. err. 0.01***)
  - D_it-1: -0.06*** (std. err. 0.01***)
  - Observations: 1,800,892 (NMS) and 686,110 (Bulgaria) as reported.

### Database and balance-sheet diagnostics (Orbis firm-level micro data, 2004–2013)
- Bulgaria sample in 2013: over 200,000 companies; micro and small enterprises represent 96 percent of total firms; about 85 percent of companies in non-tradable sectors.
- Orbis sample captures about 35 percent of aggregate corporate-sector debt (per Eurostat sectoral accounts) on average for Bulgaria.
- Orbis sample summary statistics (selected):
  - Bulgaria IK: 154,550 observations; mean 1.7932; std. dev. 2.090; 25th perc. 0.020; median 0.140; 75th perc. 0.51
  - Bulgaria SK: 203,050 observations; mean 29.7738; std. dev. 3.231; 25th perc. 1.41; median 3.72; 75th perc. 11.76
  - Bulgaria DE: 209,831 observations; mean 2.235; std. dev. 3.300; 25th perc. 0.20; median 0.60; 75th perc. 1.77
  - Bulgaria ICR: 166,600 observations; mean 14.0736; std. dev. 8.32; 25th perc. 2.47; median 5.42; 75th perc. 13.00
- Data treatment: analysis focuses on solvent companies with positive equity; observations trimmed below/above the 1st/99th percentile for each regression variable.
- Sector and size patterns:
  - Indebtedness higher for smaller firms and for non-tradable sectors (notably utilities).
  - Almost 20 percent of firms (accounting for 32 percent of total NFC assets in the sample) had ICR below the precautionary threshold of 2 in 2013.
  - 10 percent of firms had ICR lower than 1.
  - Debt at risk concentrated in SMEs (19 percent of total firms); a smaller number of large firms with debt at risk account for 21 percent of assets in the sample.
  - By sector: manufacturing accounts for 9 percent of assets in debt-at-risk, followed by utilities and sales at 7 percent each.

### Policy implications and conclusions (corporate deleveraging)
- Key risks and facts:
  - Bulgaria’s corporate sector is among the most leveraged among NMS; corporate debt accounts for over 55 percent of banks’ domestic loans in Bulgaria.
  - Bad and restructured loans in the NFC sector account for 65 percent of the total, as of end-2014.
- Growth and investment drag:
  - High corporate debt overhang is empirically associated with a drag on investment and growth in the NMS sample; Bulgaria shows persistent relationships post-crisis.
- Policy recommendations to support orderly deleveraging and reduce costs:
  - Support an orderly and efficient adjustment process aimed at restoring corporate productivity and growth.
  - Ensure a self-reinforcing institutional framework to prevent continued build-up of imbalances in specific segments of the economy.
  - Encourage an efficient corporate debt restructuring framework.
  - Promote alternative funding sources.
  - Use macro-prudential tools and tax measures to promote firms’ long-term viability.
- Caution:
  - Corporate deleveraging tends to be protracted and, if implemented via a generalized withholding of credit to all firms, can generate significant macro-financial spillovers (heightened bankruptcies and rising NPLs in banks’ balance sheets).

### Annex I — Cross-Country Policy Initiatives to Support an Orderly Deleveraging Process (summary)
- Corporate restructuring framework:
  - Effective insolvency regimes enable re-organization of viable firms and liquidation of non-viable firms; approaches vary from government-sponsored market-based models to government-directed models.
  - Bulgaria: amendments to corporate insolvency legislation (including limiting backdating of insolvencies to three years); voluntary restructuring available but not specifically regulated; insolvency proceedings rose from 390 in 2011 to 1,339 in 2012.
- Supervisory activity and banks’ NPL management:
  - Independent AQRs and stress testing plus timely capital replenishment effective in other countries; Bulgaria’s BNB relies on a capital conservation buffer of 2.5 percent and a systemic risk buffer of 3 percent aimed at ensuring capital adequacy.
- Targeted tax incentives:
  - Time-bound tax incentives have been used elsewhere to accelerate restructuring; Bulgaria applies thin capitalization rules if liabilities exceed three times equity.
- Access to funding sources:
  - Well-targeted government guarantees and diversification of funding sources can support viable firms; Bulgaria: SMEs benefit from EU-financed equity and debt instruments and guarantee schemes (National Guarantee Fund; JERAMIE initiative).
- Macro-prudential tools:
  - Sectoral capital requirements or variable risk weights can target sectoral excesses; Bulgaria currently relies on capital conservation and systemic risk buffers with scope to develop further sectoral measures under Pillar II if warranted.

*Source: IMF staff report text (Selected Issues and Article IV Consultation material on Bulgaria; content unit: _cr15120).*

### REFERENCES ____________________________________________________________________________12

### INFLATION DETERMINANTS AND IMPLICATIONS IN BULGARIA

### A. Introduction
- Bulgaria experienced sustained deflation since mid-2013; HICP headline inflation has been negative since August 2013.
- Bulgaria’s deflation is among the deepest in the EU since mid-2013, except for Greece and Cyprus.
- Key proximate factors discussed: inward spillover of low inflationary pressure from the EU, consecutive electricity price cuts in 2013, and fast falling global commodity prices since late 2014.
- Analysis period: January 2012 to February 2015 (regressions use monthly data for 2012–14; sample period 2012m1–2014m12).

### B. Stylized facts
- Four phases of inflation since 2012:
  - 2012: headline inflation close to EU average.
  - Early 2013–early 2014: sharp decline coinciding with administered price reductions.
  - Rest of 2014: broadly stable in negative territory.
  - End-2014 to Feb-2015: deepening deflation then uptick in Feb-2015, in line with EU movements.
- Electricity price cuts in 2013:
  - Cumulative 15 percent reduction for residential households (three cuts).
  - March 5, 2013: between 6.2–7.3 percent cut depending on location.
  - August 1, 2013: daytime tariffs reduced by between 4.2–4.9 percent; night-time by between 3.4–6.8 percent.
  - January 1, 2014: daytime tariffs reduced by 1 percent; night-time rate cut by 10 percent; industrial consumers’ prices reduced by 1.5 percent.
- Administered prices’ role:
  - After the first cut, contribution from administered prices to inflation shrank rapidly.
  - Negative contribution from the second cut offset positive non-administered contributions, pushing headline inflation negative (first time since 2002 administered prices contributed negatively).
  - Administered prices’ deflationary role became negligible after mid-2014; from October 2014 administered prices resumed a positive contribution after electricity prices were raised by around 10 percent.
- Core vs non-core:
  - Deflationary pressure came from both core and non-core components; core components fell below zero slightly ahead of non-core and increased in relative importance since late 2013.
  - Core inflation in Bulgaria tracked closely EU core inflation during the period.

### C. Model and data
- Empirical framework: expectation-augmented Phillips curve using monthly data for 2012–14.
- Regression specification:
  - π = α + β π_{t-1} + δ π^{E} + ρ π^{EU_core} + x θ + ε
  - π = headline inflation in Bulgaria; π^{E} = inflation expectation (one-year ahead mean from Consensus Economics); π^{EU_core} = EU HICP core inflation; x = vector of external and domestic shocks (world oil and food prices, administered price changes, indicators of domestic demand); lags of dependent variable included.
- All variables expressed in 12-month growth rates in percent unless noted.
- Key data sources: Eurostat HICP; Consensus Economics expectations; IMF WEO world oil and food price indices; IMF nominal effective exchange rates; domestic indicators (wage growth, retail sales growth in constant prices, employment growth, unemployment rate).

### D. Results (monthly regressions, sample 2012m1–2014m12)
- Main empirical findings:
  - Inflation inertia is modest and fades quickly: one-period lag coefficient statistically significant at 1 percent; second lag very small and insignificant.
  - Fast and sizable transmission from EU core inflation to Bulgaria’s headline inflation: estimated contemporaneous coefficient suggests roughly one-to-one transmission.
  - Commodity prices (world oil and food) are statistically significant determinants, though magnitudes are small.
  - Changes in administered prices have a direct, statistically significant impact on headline inflation (coefficient < 1).
  - Nominal effective exchange rate (NEER) has a modest effect on headline inflation.
  - Domestic demand indicators vary: wage growth significant with small magnitude; retail sales growth not significant; unemployment rate coefficient has unexpected sign (likely reflecting simultaneous decline of inflation and unemployment during part of the sample); employment growth not significant.
- Regression coefficients (selected specification outputs from table; S.E. in parentheses; *, **, *** denote 10%, 5%, 1% significance levels):
  - Inflation(-1): 0.421(0.101)*** (other specifications show 0.564(0.114)***, 0.465(0.085)***, 0.542(0.085)***)
  - Inflation expectations (one-year ahead): -0.122(0.218) (other specs: -0.136(0.245), -0.403(0.218)*, -0.097(0.218))
  - Contribution of administered prices: 0.658(0.248)** (other specs: 0.490(0.323), 1.092(0.289)***, 0.484(0.289)*)
  - Wage growth: 0.067(0.026)**
  - Unemployment rate: 0.538(0.156)***
  - EU area core inflation: 1.088(0.456)** (other specs: 1.090(0.528)*, 1.074(0.423)**, 1.220(0.423)**)
  - Nominal effective exchange rate: -0.105(0.057)* (other specs include -0.073(0.082), -0.048(0.054))
  - World oil price inflation: 0.023(0.006)*** (other specs: 0.019(0.006)***, 0.025(0.006)***, 0.021(0.006)***)
  - World food price inflation: 0.032(0.010)*** (other specs: 0.034(0.011)***, 0.017(0.010)*, 0.295(0.010)**)
  - Observations N = 36; R2 = 0.984, 0.980, 0.986, 0.980 across specifications.
- Decomposition of contributions to headline inflation changes:
  - For whole sample (2012m1–2014m12), external factors dominate: combined weight of spillover from low EU core inflation and declining commodity prices exceeds 80 percent.
  - For sub-sample Feb 2013–Jan 2014 (prior to first electricity price cut to after third cut), administered price changes weigh around 35 percent; combined with weak domestic demand yields roughly a 50/50 split between external and domestic factors in depressing headline inflation for that period.
  - Note: contributions excluding lag inflation and inflation expectations are normalized to 100 percent.

### E. Outlook (projection assumptions and sensitivities)
- Baseline projection: inflation is projected to bottom out and start an upward trend in 2015; headline inflation expected to reach positive levels by end-2015.
- Key assumptions underpinning outlook:
  - Core inflation in the EU will remain subdued before recovering in late 2015 to 2014Q4 levels.
  - Oil prices will stabilize in the first quarter and slightly recover toward end-2015, consistent with the April 2015 IMF World Economic Outlook assumptions.
  - Nominal wage growth is kept at 2014 levels.
  - Administered prices are kept at end-2014 levels (acknowledging an increase in electricity prices may be required to support cost recovery by the electricity company).
  - Inflation expectations remain stable.
- Sensitivities:
  - A ¼ percentage point reduction in EU core inflation for 2015 will lower Bulgaria’s headline inflation by about 0.3 percentage points.
  - A 2-percent upward adjustment in administered prices starting in 2015Q3 will raise annual inflation in Bulgaria by around 0.2 percentage points (partly reflecting the low base in 2014).

### F. Implications
- Sustained deflation poses important risks through three channels:
  - Demand postponement: falling prices may postpone consumption and investment as holding cash yields a positive real return.
  - Real debt increase: deflation increases the real value of debts, worsening debtors’ financial condition and improving creditors’ condition—net effect likely negative on aggregate demand per Fisher (1983).
  - Nominal rigidity and real adjustment: deflation limits room for nominal adjustments, forcing real adjustments; downward nominal wage rigidity could complicate competitiveness adjustments even if nominal wage growth remains at or below levels reflecting inflation and productivity gains.
- Policy-relevant observations:
  - External spillovers (EU core inflation and commodity prices) are the dominant drivers of recent deflationary pressures; policy responses should account for this external exposure.
  - Administered price adjustments (notably electricity tariffs) had large short-run effects; changes in administered prices are an important domestic policy instrument with measurable inflation impact.
  - Monitoring and addressing exchange rate pass-through, commodity price volatility, and domestic wage dynamics are important for near-term inflation management.

*Prepared by Zaijin Zhan; analysis period and regressions based on monthly data for 2012–14 (sample period 2012m1–2014m12).*

### 16.      In Bulgaria, these risks are high, and, under the currency board framework, policy

### 16.      In Bulgaria, these risks are high, and, under the currency board framework, policy

### Deflationary risks and constraints under the currency board
- Policy mechanisms to address deflationary pressures are constrained under the currency board framework.
- Key channels through which sustained deflation can harm the economy:
  - Increases real debt burdens and raises real yields on cash holdings, amplifying negative effects of corporate indebtedness on investment decisions.
  - Limits scope for macroeconomic policy to relieve deflationary pressures, making measures to reduce corporate indebtedness more important.

### Impact on private demand, corporate sector, and investment
- Falling prices may have adversely affected private demand for consumption and investment; causal identification is difficult over a short period with multiple shocks.
- Bulgaria’s non-financial corporate (NFC) sector has one of the highest indebtedness levels among NMS; high indebtedness has been shown to negatively affect firms’ investment decisions.
- Sustained deflation would exacerbate the investment-suppressing effect by increasing real debt burdens.

### Fiscal implications of deflation
- About 1/2 of the fiscal balance deviation in 2014 from the original budget target (or about 1.1 percentage points of GDP) can be attributed to lower inflation than assumed in the original 2014 budget (absent corrective expenditure actions).
- At the 2014 collection level, a one percentage point drop in consumer inflation, other things being equal, leads to a 0.3 percentage point increase in the deficit and, in turn, public debt.
- The substantial sensitivity of fiscal outturns to price assumptions and limited fiscal space implies the need for flexibility to adjust spending if deflationary pressures are more sustained than currently projected.

### Labor market, wages, and competitiveness
- After the global financial crisis, labor market adjustment in Bulgaria occurred more on quantity (employment) than on wages, especially among low-skilled workers, contributing to a sharp increase in unemployment.
- Since early 2013, real wage growth has remained well above productivity growth; nominal wage growth only started to moderate in 2014 but not sufficiently to bring real wages in line with productivity given sustained deflation.
- An appreciation of Bulgaria’s ULC-based REER, including relative to NMS peers, has escalated in recent years; continued appreciation could gradually erode competitiveness.

### Effects of lower commodity prices
- Bulgaria imports about 11–12 percent of GDP in petroleum products, fuel, and gas, of which more than 60 percent is to be re-exported; this and longer-term contracts limit the short-term impact of falling prices.
- Loss-producing state utility companies have kept key administered prices low relative to regional comparators, including through three electricity price reductions during 2013-14, leaving little space to pass recent price declines to end-users.
- Lower commodity prices are expected to give some breathing space to the national electricity company (NEK) to realign its pricing structure more closely with international prices.

### Conclusions (Section G) — drivers and outlook
- Declining inflation since 2012 and deflation since mid-2013 were due to both external and domestic factors; inward spillover of low inflationary pressure from the EU was an underlying trend, exacerbated by consecutive electricity price cuts in 2013 and fast falling global commodity prices since mid-2014.
- Outlook for 2015:
  - Deflation is expected to lessen gradually in 2015.
  - With assumptions of gradual stabilization of EU core inflation and global commodity prices and administered price changes in line with historical patterns, year-on-year headline inflation is expected to rise gradually, ending the year in positive territory.
  - The outlook is subject to large uncertainties, especially downside risks; the effectiveness of the ECB QE plan in stabilizing core inflation in the EU has a direct impact on Bulgaria’s inflation outlook.

### Observed macroeconomic effects to date
- Sustained deflation so far has adversely affected macroeconomic performance by:
  - Worsening the fiscal balance.
  - Raising the public and private debt burden.
  - Eroding part of the adjustment in nominal wages.

### Policy recommendations
- Given limited policy options to influence inflation directly (small open economy, no independent monetary policy, limited fiscal space), policy priorities include:
  - Clear communication on the sources and temporary nature of current deflationary pressure to help anchor inflation expectations and avoid a deflationary spiral.
  - Clarification of measures to deal with the private sector debt overhang.
  - Tighter expenditure controls to adjust to large uncertainties around the inflation outlook (illustrated by the overshooting of the original fiscal balance target in 2014).
  - Pay closer attention to nominal wage levels, tying increases to productivity gains and accounting for the effect of falling prices.
  - Use moderation in energy prices as a window of opportunity to implement reforms in the energy sector, including closer alignment of residential utility tariffs with global prices/cost structures, accompanied by enhanced protection of poor households against tariff adjustments.

### Non-Financial Corporate Debt Overhang — summary diagnostics
- As of end-2013:
  - Bulgaria’s total non-financial sector indebtedness was third highest among NMS—below only Croatia and Slovenia—at over 150 percent of GDP on a consolidated basis.
  - NFC debt stood at over 110 percent of GDP in 2013, the highest level among NMS and well-above the Euro area average.
  - Debt securities comprised less than 4 percent of GDP of NFC debt; loans dominate.
  - Among NFC loans, about 45 percent of GDP consists of loans from Bulgarian banks.
  - Foreign direct (largely inter-company) loans are estimated at roughly 40 percent of GDP.
- Historical drivers:
  - Pre-accession foreign capital inflows (FDI and foreign bank-intermediated loans) and credit-fueled domestic demand booms contributed to large current account deficits and net external debt accumulation; credit was increasingly directed to non-tradable sectors.
- Liquidity and solvency indicators (consolidated, end-2012):
  - Debt-to-income ratio remained at significantly high levels (over 400 percent of income), well above peers, despite declines since 2008.
  - Interest Coverage Ratio (ICR) was 5 as of end-2012 (coverage of interest payments by gross operating income), above standard thresholds of 2 or 1 used to identify firms with “debt at risk,” but one of the lowest among NMS after Hungary and Slovenia.
  - During the global crisis, aggregate ICR fell to as low as 3 for the entire system, indicating sensitivity of corporate balance sheets to macro-financial shocks.

*Source: IMF staff report text (Selected Issues and Article IV Consultation material on Bulgaria).*

### 12.      Solvency risk measures the potential inability of firms to keep the value of their assets above

### 12.      Solvency risk measures the potential inability of firms to keep the value of their assets above that of their liabilities

### Solvency measures and leverage indicators
- Leverage commonly measured by debt-to-equity ratios; a net leverage ratio deducts firms’ liquid and financial assets (currency and deposits, securities, and loans) from the debt stock (following Iossifov and Zumer (2014)).
- Bulgaria observations (latest available data for 2012):
  - Gross leverage: about 85 percent of equity.
  - Net debt-to-equity (accounting for firms’ assets): about 45 percent.
- Cross-country context: Bulgaria appears less exposed to solvency risk than some NMS such as Croatia, Latvia, and Slovenia.

### Macro-financial implications of corporate leverage
- Theoretical context:
  - Perfect capital market theory (Modigliani and Miller, 1985) implies investment is independent of financing mix; market frictions (e.g., asymmetric information) can make leverage matter for investment (Meyers, 1984).
  - Excess leverage can raise vulnerabilities and amplify firms’ sensitivity to income and interest shocks (Bernanke and Gertler, 1989).
- Empirical summary (NMS panel and Bulgaria-specific):
  - NMS panel: higher debt overhang reduces investment with estimated impacts ranging from 1 to 6 percent depending on the debt overhang proxy (debt-to-equity or ICR).
  - For Bulgaria:
    - ICR is significant over 2004–2013: firms’ liquidity risk (capacity to repay) matters for funding and investment decisions.
    - Debt-to-equity measure not significant for full 2004–2013 sample but becomes significant in the post-crisis period (2011–13).
    - Post-crisis results: consistent negative relationship between leverage and investment across sectors and firm sizes.

### Empirical literature and methodology
- Literature finds a financial accelerator effect: weak balance sheets amplify adverse shocks on firm investment, especially in downturns and for smaller firms (Fazzari et al. (1988); Bernanke et al. (1999); Vermeulen (2000); Goretti and Souto (2013)).
- Econometric approach in this chapter:
  - Dynamic panel specification: IK_it = α + β IK_it-1 + γ SK_it-1 + δ D_it-1 + ε_it.
  - Dependent variable IK_it: investment-to-capital ratio.
  - Debt overhang D proxied by debt-to-equity and by the Interest Coverage Ratio (ICR = EBITDA / interest payments).
  - Lagged sales-to-capital SK included as control.
  - Estimation via GMM two-step system estimator (Blundell and Bond, 1998); STATA module by Roodman (2003).

### Database and balance-sheet diagnostics
- Data source: Orbis firm-level micro data (Bureau van Dijk) for Bulgaria and nine other NMS; sample period 2004–2013.
- Coverage and representativeness:
  - Bulgaria sample in 2013: over 200,000 companies; micro and small enterprises represent 96 percent of total firms; about 85 percent of companies in non-tradable sectors.
  - Orbis sample captures about 35 percent of aggregate corporate-sector debt (per Eurostat sectoral accounts) on average for Bulgaria.
  - Firms in Orbis tend to show higher leverage than Eurostat sectoral accounts, with large dispersion: leverage ranges from 20 percent (lower quartile) to 170 percent (upper quartile) of equity in the sample.
  - Bulgarian median leverage around 60 percent of equity in the Orbis sample.
- Summary sample statistics (Orbis dataset, regression variables):
  - NMS (obs, mean, std. dev., 25th perc., median, 75th perc.):
    - IK: 2,409,023; 2.5015; 46.000; 0.000; 0.110; 0.42
    - SK: 3,256,278; 35.6216; 40.711; 1.50; 4.93; 16.33
    - DE: 1,482,313; 2.505; 5.980; 0.20; 0.62; 1.86
    - ICR: 822,040; 28.1880; 0.743; 3.00; 8.00; 21.67
  - Bulgaria:
    - IK: 154,550; 1.7932; 2.090; 0.020; 0.140; 0.51
    - SK: 203,050; 29.7738; 3.231; 1.41; 3.72; 11.76
    - DE: 209,831; 2.235; 3.300; 0.20; 0.60; 1.77
    - ICR: 166,600; 14.0736; 8.32; 2.47; 5.42; 13.00
- Data treatment: analysis focuses on solvent companies with positive equity; observations trimmed below/above the 1st/99th percentile for each regression variable to reduce outliers.
- Sector and size patterns (2013, Bulgaria Orbis sample):
  - Indebtedness higher for smaller firms and for non-tradable sectors (notably utilities).
  - Firms’ indebtedness by size (percent of equity, firms' average): micro, small, medium, large pattern shows increasing indebtedness for smaller sizes.
  - Firms’ indebtedness by sector (percent of equity, firms' average): manufacturing, utilities, construction/real estate, wholesale/retail sales, etc., show heterogeneity.

### Liquidity pressures and "debt at risk"
- In 2013 (Bulgaria Orbis sample):
  - Almost 20 percent of firms (accounting for 32 percent of total NFC assets in the sample) had ICR below the precautionary threshold of 2 (classified as "debt at risk").
  - 10 percent of firms had ICR lower than 1 (EBITDA insufficient to cover interest).
  - Debt at risk concentrated in SMEs (19 percent of total firms).
  - A smaller number of large firms with debt at risk account for 21 percent of assets in the sample.
  - By sector: manufacturing accounts for the bulk of debt at risk (9 percent of assets), followed by utilities and sales among non-tradables (7 percent of assets each).

### Econometric results (NMS panel and Bulgaria)
- NMS panel:
  - Evidence of negative sensitivity of investment-to-capital ratio to corporate debt overhang after controls.
  - Estimated coefficients significant and with expected signs; rejecting δ = 0 (perfect capital markets).
  - Impact of debt overhang on investment ranges from 1 to 6 percent depending on proxy.
- Bulgaria-specific:
  - ICR significant over full 2004–2013 period; debt-to-equity significant only in post-crisis period.
  - Robustness:
    - Before 2010: higher corporate leverage negatively affects mainly micro firms; utilities sector particularly affected.
    - Post-crisis: consistent negative relationship between leverage and investment across sectors and sizes.
- Example regression diagnostics and coefficients (NMS sample dynamic GMM two-step system estimator; significance indicators preserved as in source):
  - Constant: 1.97*** (std. err. 0.44***)
  - IK_it-1: 0.07** (std. err. -0.03***)
  - SK_it-1: -0.04*** (std. err. 0.01***)
  - D_it-1: -0.06*** (std. err. 0.01***)
  - AR(1) test: -1.68* (-3.34***)
  - AR(2) test: 0.99 (-1.05)
  - Observations: 1,800,892 (NMS) and 686,110 (Bulgaria) as reported in the source tables.
- Bulgaria-specific table excerpts (selected):
  - Full sample / 2004-10 / 2011-13 (BGR: D = DE or ICR):
    - Example coefficient patterns and test statistics preserved in the source (see original tables for precise values).

### Policy implications and conclusions
- Key risks and facts:
  - Bulgaria’s corporate sector is among the most leveraged among NMS, implying important liquidity risks and potential solvency concerns under severe macro-financial shocks.
  - Corporate debt accounts for over 55 percent of banks’ domestic loans in Bulgaria.
  - Bad and restructured loans in the NFC sector account for 65 percent of the total, as of end-2014.
- Growth and investment drag:
  - High corporate debt overhang is empirically associated with a drag on investment and growth in the NMS sample; Bulgaria shows persistent relationships post-crisis.
- Policy recommendations to support orderly deleveraging and reduce costs:
  - Support an orderly and efficient adjustment process aimed at restoring corporate productivity and growth.
  - Ensure a self-reinforcing institutional framework to prevent continued build-up of imbalances in specific segments of the economy.
  - Policy initiatives that have supported deleveraging in other country experiences (as discussed in the Annex) include:
    - An efficient corporate debt restructuring framework.
    - Promotion of alternative funding sources.
    - Use of macro-prudential tools and tax measures to promote firms’ long-term viability.
- Caution on deleveraging process:
  - Corporate deleveraging tends to be protracted and, if implemented via a generalized withholding of credit to all firms, can generate significant macro-financial spillovers (heightened bankruptcies and rising NPLs in banks’ balance sheets).

*IMF staff summary of chapter content.*

### Annex I. Cross-Country Policy Initiatives to Support an Orderly

### Annex I. Cross-Country Policy Initiatives to Support an Orderly Deleveraging Process

### Corporate restructuring framework
- Effective insolvency regimes are pivotal to:
  - Re-organization of financial and operational structure of distressed but viable firms.
  - Liquidation of non-viable firms.
  - Enforcement and foreclosure processes to enable effective realization of collateral.
- Cross-country approaches vary by severity and context:
  - Government-sponsored market-based models: guidelines for voluntary out-of-court debt workouts (Iceland, Latvia, Portugal, Romania; referenced as the London Approach).
  - Government-directed models: committees with strong powers including binding arbitration (Thailand, Korea) or centralized asset management companies.
- Bulgaria-specific reforms and issues:
  - Amendments to corporate insolvency legislation to address earlier deficiencies, including limiting the backdating of insolvencies (to three years) and clarifying rules for set-off and avoidance of certain transactions.
  - Voluntary restructuring is available to creditors and debtors but is not specifically regulated.
  - Number of insolvency proceedings increased from 390 in 2011 to 1,339 in 2012 (COFACE Bankruptcy Report).
  - Judicial bottlenecks to timely and predictable insolvency proceedings remain a concern.

### Supervisory activity and banks’ NPL management
- Intense supervision promotes banks’ recognition of losses and prompt recourse to debt restructuring.
- Independent AQRs (and stress testing) plus timely actions to replenish capital shortfalls have been effective (example: Spain).
- Complementary supervisory actions can ensure adequate debt recovery and restructuring capacity (examples: Ireland, Portugal), including use of independent workout specialists.
- Use of specific guidelines for NPL management has been applied (example: Romania).
- Bulgaria-specific macro- and micro-prudential measures:
  - Bulgarian National Bank (BNB) relies on two macro-prudential capital instruments:
    - capital conservation buffer of 2.5 percent
    - systemic risk buffer of 3 percent
  - These are intended to ensure credit institutions sustain full coverage of NPLs (net of impairments) with own funds exceeding 8 percent in terms of total capital adequacy ratio.
  - Micro-prudential measures have been introduced at individual banks’ level to address high NPLs for affected banks.

### Targeted tax incentives
- Time-bound tax incentives (over 2–3 years) have been used to accelerate corporate debt restructuring (examples: Thailand; more recently Iceland and Latvia).
- Targeted tax measures can reduce distortions from different tax treatment of debt versus equity:
  - Thin capitalization rules limit interest expenditure deductions for over-leveraged firms while minimizing undesired impact on capital investment.
  - Allowances for new corporate equity (ACE) can enhance tax neutrality and avoid pro-cyclicality (examples: Latvia and Italy).
- Bulgaria-specific tax rule:
  - Thin capitalization rules apply if the company’s liabilities exceed three times its equity.

### Access to funding sources
- Continued access to funding for viable but over-indebted firms is critical to prevent liquidity pressures turning into solvency problems.
- Government-provided guarantees to special bank credit lines have been used (examples: Italy, Portugal) to alleviate high credit risk premia and collateral requirements; especially relevant for SMEs.
- Such programs need to be well-targeted (e.g., firms with strong credit scores) and time-limited to contain fiscal and moral hazard risks (experience: Korea).
- Diversification of funding sources is necessary, including improved transparency and information sharing via public and private credit bureaus to help investors assess credit standing.
- Bulgaria-specific funding support:
  - SMEs benefit from equity and debt financial instruments financed by EU funds under different programs.
  - Guarantee schemes are extended by the National Guarantee Fund to banks for loans towards investment projects under the EU Rural Development Program and OP ”Development of the Fishery sector”.
  - Both equity and debt instruments are financed by the JERAMIE initiative under the EU OP “Competitiveness”.

### Macro-prudential tools
- Macro-prudential measures can help secure firm balance sheets and prevent new imbalances.
- Tools beyond standard Basel III measures include sectoral capital requirements or variable risk weights to target sectors showing signs of exuberance, requiring banks to hold additional capital buffers.
- Past examples: higher risk weights on commercial real estate loans in Australia in 2004; higher risk weights on corporate lending in India in 2005–2006.
- Bulgaria-specific stance:
  - BNB currently relies on the capital conservation and systemic risk buffers cited above.
  - Sectoral limits and measures under Pillar II could be developed, if warranted, going forward.

*Source: Annex I. Cross-Country Policy Initiatives to Support an Orderly Deleveraging Process (content unit: _cr15120).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/scr/2015/_cr15120.pdf_