## _wp0658

## Source details

**Canonical URL:** [_wp0658](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0658.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0658.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0658.pdf.json)

---

### Introduction and motivation
- Central banks’ exclusive right to issue currency gives them privileged access to seigniorage, effectively converting them into unregulated monopolies.
- Historical concern: siphoning of seigniorage to governments through quasi-fiscal expenditures, profit transfers, or subsidized financing raised fiscal dominance and inflationary finance issues.
- Recent reforms: greater central bank independence, prohibitions on direct financing to governments, elimination of quasi-fiscal expenditures.
- New governance concern: retaining seigniorage inside central banks (to build assets, cover carrying costs, or cover operating expenses) raises questions: “Is seigniorage spent ‘wisely’?” and “Do central banks have sufficient governance standards and safeguards?”

### Data, definitions, and methodology
- Empirical base: recent central bank financial accounts and interest rate data for a sample of 101 countries; Appendix I sample used in analyses: 87 countries total; subsamples of 60 “strong” central banks and 27 “weak” central banks (definition: strong = positive structural pre-transfer profits; weak = negative structural profits).
- Monetary magnitudes expressed in U.S. dollars; domestic deflation uses each country’s consumer price index.
- Structural income (SI) = Net Interest Margin + Net Other Structural Income. Transitory net income aggregates valuation changes and transitory revenues/expenditures (realized and unrealized gains/losses on foreign exchange, securities, fixed assets; net transfers to/from reserves; provisions/write-offs; extraordinary gains/losses).
- Reserve currency assumed U.S. dollar for exchange-rate valuation adjustments; currency-composition weights: All other countries: Dollar 70, Euro 25, Yen 5; Euro countries: Euro 70, Yen 30; U.S.: Dollar 50, Euro 50.
- Methodological stance: relies on published financial accounts and broad accounting concepts (enables broad cross-country coverage at the cost of detailed efficiency measures).

### Key observed trends and high-level results
- Seigniorage trend: clear downward trend reflecting both declining inflation and declining demand for currency (1985–2003).
- Carrying cost of foreign reserves has tended to rise, reflecting expanded reserve accumulation (1985–2004).
- Many central banks exhibit sustained losses and negative capital, prompting debate about recapitalization needs.

### Income statement sample means and dispersion (selected figures)
- Structural income (SI) (as percent of currency): Full Sample 6.32; Weak Central Banks -3.26; Strong Central Banks 10.64.
  - Net Interest Margin: Full Sample 5.80 (13.95); Weak -3.35 (10.42); Strong 9.92 (13.43).
  - Other structural income: Full Sample 0.51 (2.39); Weak 0.09 (2.83); Strong 0.72 (2.16).
- Transitory Income: Full Sample 0.25; Weak 5.92; Strong -2.30.
  - Valuation changes: Full Sample -0.14 (17.07); Weak 4.02 (13.27); Strong -2.0 (18.32).
  - Other transitory: Full Sample 0.39 (4.99); Weak 1.91 (6.39); Strong -0.29 (4.10).
- Operating Expenditures: Full Sample 4.85 (6.04); Weak 6.63 (6.26); Strong 4.71 (2.16).
- Transfers: Full Sample 2.53 (5.38); Weak 2.26 (8.48); Strong 2.65 (3.23).
- Retained Profits: Full Sample -0.81 (18.35); Weak -6.23 (17.22); Strong 1.63 (18.45).

Key observation:
- Structural profitability (SI minus operating expenditures) is low: about 1.5 percent of currency (aggregate implication).
- Operating expenditures nearly exhaust the interest margin on average; retained profits are, on average, negative.

### Variance decomposition of retained profits (selected shares)
- Full Sample shares: Net Interest Margin 0.25; Net Other Structural Income 0.03; Valuation Gains/Losses 0.65; Other Transitory Net Income 0.01; Operating Expenditure -0.02; Transfers 0.08.
- Weak central banks: Net Interest Margin 0.15; Net Other Structural Income 0.07; Valuation Gains/Losses 0.45; Other Transitory Net Income 0.05; Operating Expenditure -0.02; Transfers 0.30.
- Strong central banks: Net Interest Margin 0.21; Net Other Structural Income 0.01; Valuation Gains/Losses 0.81; Other Transitory Net Income 0.01; Operating Expenditure -0.04; Transfers 0.01.
- Implication: valuation gains/losses account for the largest share of retained-profits variance, especially for strong central banks (0.81).

### Balance sheet patterns and “shadow capital” (selected figures)
- Full Sample net claims: 154.6 (percent of currency).
  - On foreigners: Assets 277.1; Assets total 354.9; Liabilities -77.9.
  - On Government: Assets 134.1; Liabilities -83.4; Net claims on government 50.7.
  - On Banks: Assets 103.2; Liabilities -276.3; Net claims on banks -173.1.
- Shadow Capital (percent of currency): Full Sample 154.7; Weak 104.2; Strong 177.4.
  - Currency = 100.0 for all samples.
  - Capital: Full Sample 54.7; Weak 4.2; Strong 77.4.
- Differences: Weak central banks have much higher foreign liabilities, higher net claims on government (On Government net 92.2 vs Strong 32.0), lower claims on banks, and much less capital (Capital 4.2 vs Strong 77.4).
- Policy implication: asset accumulation patterns in weak central banks (high foreign liabilities, high government claims, low capital) can undermine credibility and lead to progressive decapitalizations and debt sustainability problems.

### Bivariate statistical findings (selected relationships)
- Retained profits strongly related to transitory net income for both weak and strong banks (suggests avoidance of transferring valuation gains/windfalls).
- Strong negative association between retained profits and transfers for weak central banks (weak banks often forced to transfer profits they do not have).
- Operating expenditures positively related to net interest margin for both groups; link stronger for strong central banks.
- Very strong positive link between claims on government and foreign liabilities across both groups (central banks borrow abroad and on-lend to governments).
- Structural profits negatively related to claims on government and government deposits in weak central banks; opposite in strong central banks.
- Selected significance: t-statistics and 90 percent/95 percent significance marked in source tables (90 percent significance level: *; 95 percent significance level: **).

### Interest rates, risk premia, and valuation adjustments (sample statistics)
- Nominal Rate: Full Sample 9.31; Weak Central Banks 13.05; Strong Central Banks 7.73.
- Inflation: Full Sample 5.16; Weak 9.22; Strong 3.43.
- Nominal Premium: Full Sample 7.21; Weak 10.95; Strong 5.62.
- Real Rate: Full Sample 3.99; Weak 3.72; Strong 4.10.
- Real Premium: Full Sample 4.15; Weak 3.88; Strong 4.26.
- Finding: In real terms there is no significant difference between strong and weak central banks despite higher nominal rates and inflation in weak central bank countries.
- Valuation adjustments computed as the average of two polar methods described in the source (method 1 and method 2), with domestic inflation via each country’s CPI and foreign inflation via reserve countries’ CPIs weighted per Table 9.

### Seigniorage decomposition and uses (selected figures, percent of currency)
- Total seigniorage: Full Sample 19.97; Weak Central Banks 14.66; Strong Central Banks 22.36.
- Components:
  - Rc (seigniorage from currency): Full Sample 8.64; Weak 12.09; Strong 7.09.
  - Rk (seigniorage from capital): Full Sample 2.30; Weak -1.19; Strong 3.87.
  - Rd (seigniorage from nonremunerated deposits): Full Sample 9.03; Weak 3.76; Strong 11.40.
- Uses of seigniorage:
  - Balance sheet costs: Full Sample 9.40; Weak 10.74; Strong 8.79.
    - φexF (carrying cost of international reserves): Full Sample 8.16; Weak 10.47; Strong 7.11.
    - πk (user cost of capital): Full Sample 1.24; Weak 0.27; Strong 1.68.
  - Expenditure flows: Full Sample 6.99; Weak 6.98; Strong 7.00.
    - O (Operating expenditures): Full Sample 4.85; Weak 6.63; Strong 4.06.
    - Z (Other exceptional expenditures): Full Sample -0.39; Weak -1.91; Strong 0.29.
    - T (Transfers): Full Sample 2.53; Weak 2.26; Strong 2.65.
  - Capital Accumulation: Full Sample 3.58; Weak -3.05; Strong 6.57.
    - e∆xF (change in foreign assets, real): Full Sample 20.58; Weak 26.84; Strong 17.77.
    - ∆xG (change in net claims on government): Full Sample 4.28; Weak -13.21; Strong 12.15.
    - ∆xB (change in net claims on banks): Full Sample -14.96; Weak -10.11; Strong -17.14.
    - ∆c (change in currency): Full Sample -6.32; Weak -6.57; Strong -6.21.

Key seigniorage insights:
- Unremunerated deposits (Rd = 9.03 for Full Sample) produce more seigniorage than currency (Rc = 8.64 for Full Sample).
- Balance sheet costs, mainly the carrying cost of international reserves, dominate seigniorage use (~half of total use: Balance sheet costs 9.40 out of Seigniorage 19.97 for Full Sample).
- Operating expenditures account for a substantial share of seigniorage use (O = 4.85, about 35 percent of uses).
- Net asset accumulation dominated by international reserve accumulation (e∆xF constitutes the bulk of capital accumulation flows).
- Weak central banks generate lower total seigniorage (14.66) than strong central banks (22.36) despite higher Rc because weak banks have much lower Rk and Rd.
- Weak central banks: capital accumulation negative (-3.05) vs strong central banks positive (6.57); weak central banks lose real capital.

Interpretations regarding Rd for weak banks:
- Up to 84 percent of deposits at weak central banks could be remunerated (sample-average-based interpretation), versus 32 percent for strong central banks; alternatively, assuming no remunerated deposits implies only 10 percent of weak central banks’ assets on government and banks are remunerated compared with 66 percent for strong central banks.

### Determinants of operating expenditures — demand vs supply effects (empirical results)
- Two service categories: monetary management; financial system stability and development.
- Expected drivers:
  - Soft revenue constraint: demand-side factors dominate (population, per capita GDP, size of financial sector, scope of responsibilities).
  - Hard revenue constraint: supply-side factors dominate (availability of seigniorage, inflation, currency in circulation, wage levels, governance).
- Key empirical findings:
  - Demand model: a 10 percent increase in country size → 7.5 percent increase in operating expenditures; a 10 percent increase in per capita income → 5.8 percent increase in operating expenditures.
  - Supply model: a 10 percent increase in currency issued → 7 percent increase in operating expenditures.
  - Provision of supervisory services increases operating expenditures by less than 2 percent.
  - Improvement in the KKM index from the worst to the best country reduces operating expenditures by slightly over 1 percent.
- Subsample differences:
  - Financial depth (M2/GDP) significantly impacts operating expenditures of weak central banks but not of strong central banks.
- Model selection: Davidson and MacKinnon’s J-statistics indicate the demand model dominates for the aggregate sample — consistent with numerical dominance of strong central banks.
- Conclusion: overall, central banks’ operating expenditures are driven by a soft budget constraint.

### Reduced-form performance index and correlations
- Relative performance index p_i derived from reduced-form residuals: 100(1), i i pe δ =− where i δ is the residual.
- Distribution: best performer spends over three times less than group average; worst performer spends over five times more than group average.
- Weak central banks tend to have lower (worse) expenditure performance indices and cluster on the high-spending side.
- Correlation: central banks that overspend tend to be least profitable — indicating a link between overspending and weak governance.

### Expenditures and macro-financial outcomes (selected regression results, Table 13)
- LOG(Inflation): Coefficient 0.41, Std. Err. 0.24, t-Statistic 1.71*, Prob. 0.0941
- LOG(GDP Volatility): Coefficient 0.13, Std. Err. 0.41, t-Statistic 0.32, Prob. 0.7482
- LOG(Inflation Volatility): Coefficient -0.25, Std. Err. 0.21, t-Statistic -1.21, Prob. 0.2332
- LOG(M2/GDP Volatility): Coefficient -0.11, Std. Err. 0.07, t-Statistic -1.62, Prob. 0.1127
- Constant C: 2.76, Std. Err. 0.56, t-Statistic 4.93, Prob. 0.0000
- R-squared 0.127798; Adjusted R-squared 0.040578; S.E. of regression 1.265758; Sum squared resid 64.08570; Log likelihood -71.80730; F-statistic 1.465231; Prob(F-statistic) 0.230834; Durbin-Watson stat 2.157605
- Interpretation: no clear robust link identified between central banks’ operating expenditures and macro-financial outcomes; higher expenditures associated with lower inflation volatility and financial volatility but higher inflation; significance levels are low.

### Main policy implications and recommendations
- Central bank independence remains necessary; more independent central banks tend to have better governance and do not necessarily increase uncontrolled expenditures.
- Independence must be paired with accountability and transparency:
  - Income statements should distinguish clearly between transitory and permanent (structural) components.
  - Central banks not publishing income statements should do so.
- Allowing central banks to manage budgets and set wage scales is essential for modernizing towards small, highly skilled staffs.
- Transparency and standardized reporting are urgent: inconsistent accounting and volatile valuation gains/losses impede cross-country comparisons and could be exacerbated by switching to international accounting standards without clearer reporting rules.
- Governments should periodically review central bank expenditures and the cost of reserve accumulation; recapitalizations should be opportunities for broader policy debate.
  - Example practice cited: the New Zealand arrangement where seigniorage income belongs to the State but a negotiated share is retained by the central bank for funding.
- Pricing and user charges:
  - Central banks should ensure services are adequately priced and users pay for them.
  - Carrying international reserves is costly; this can justify not paying full market interest on deposits or using unremunerated reserve requirements or fee-based mechanisms to internalize liquidity costs.
- Broader policy question: defining an appropriate level of services provided (e.g., appropriate level of international reserves) remains difficult.

### Caveats, limitations, and directions for further research
- Evidence is cross-sectional from a single year — causality is difficult to establish; panel analysis over longer periods is needed.
- Some conclusions based on simple averages; large standard deviations and heterogeneity may mean results are influenced by outliers.
- Measuring central bank output and efficiency remains inherently difficult; this paper does not measure output in the traditional sense (output per input) nor the quality/quantity of services.
- Performance index should be interpreted with caution; further analysis required to firm up conclusions.

*Source: _wp0658 (excerpts and tables from the IMF working paper PDF).*

### 1.  Sample Countries ...................................................................................................

### 1. Sample Countries

### Introduction: motivation and context
- Central banks’ exclusive right to issue currency gives them privileged access to seigniorage, effectively converting them into unregulated monopolies.
- Historical policy concern: siphoning off of seigniorage to governments (through quasi-fiscal expenditures, profit transfers, or subsidized financing) raised issues of fiscal dominance and inflationary finance.
- Recent reforms: greater central bank independence, prohibitions on direct financing to governments, and elimination of quasi-fiscal expenditures.
- New governance concern: retaining seigniorage inside central banks (to build assets, cover carrying costs, or cover operating expenses) raises questions of efficiency and governance: “Is seigniorage spent ‘wisely’?” and “Do central banks have sufficient governance standards and safeguards?”

### Key observed trends noted in the source
- Seigniorage has followed a clear downward trend reflecting both declining inflation and a declining demand for currency (see Figure 1: World Inflation and Currency Demand, 1985–2003).
- The carrying cost of foreign reserves has tended to rise, reflecting an expanded accumulation of reserves (see Figure 2: International Reserves, 1985–2004).
- These trends have led to sustained losses and negative capital in many central banks, prompting debates about the need for central bank recapitalizations.

### Main research questions addressed
- How much seigniorage is retained versus transferred? If retained, is it used to:
  - build up the central bank’s balance sheet, and, if so, which type of asset; or
  - cover operating expenses?
- How can “poor” financial performance be explained?
  - Is it mainly a reflection of central banks’ balance sheets or their operating expenditures?
  - Does it mainly reflect structural factors (against which central banks can do little), or does it also reflect poor governance?
- Do central banks that spend more deliver “better quality” services (i.e., better monetary or financial stability)?

### Data breadth and methodological stance
- Empirical base: recent central bank financial accounts and interest rate data for a sample of 101 countries.
- Methodological choice: relies on published financial accounts and broad accounting concepts rather than specific management parameters (e.g., number or quality of staff).
- Advantage claimed: allows more comprehensive coverage in terms of both countries and scope of analysis and enables a global economic analysis of central bank behavior based on broad trends in the data.
- Trade-off: this broader approach limits the scope in terms of defining and measuring central bank efficiency.

### Relationship to prior literature
- Fry, Goodhart, and Almeida (1996): pioneering study limited to a sample of 42 central banks and specific governance/management aspects.
- McKinley and Banaian (2005): used a sample of 32 central banks to analyze efficiency of inputs (labor and capital) producing outputs (monetary policy quality index and some financial stability indices).
- The paper builds on a framework by Ize (2005) that identifies “structural profits” as a key concept to assess sustainability of central banks’ financial performance and need for capital.
- References flagged for related debates: Stella (2005) and Ize (2005).

### Structure of the paper (as presented)
- Section II: describes the accounting data used, identifies key linkages, and contrasts accounts of central banks with strong versus weak financial positions.
- Section III: examines how seigniorage is allocated, using an accounting breakdown of its sources and uses.
- Section IV: examines whether central bank operating expenditures are driven by demand for central bank services or the availability of seigniorage, and whether there is a measurable link with overall performance measured by broad macro-financial indicators.
- Section V: concludes with main findings, key policy implications, and limitations of the paper.

*Source: _wp0658 - 1.  Sample Countries (excerpts from the IMF working paper PDF).*

### Appendix I describes the data set used in this study.

### _wp0658 - Appendix I describes the data set used in this study.

### Data, definitions, and measurement
- Sample: 87 countries total; subsamples of 60 “strong” central banks and 27 “weak” central banks (definition: strong = positive structural pre-transfer profits; weak = negative structural profits).
- All monetary magnitudes expressed in U.S. dollars; domestic deflation uses each country’s consumer price index.
- Structural income (SI) = Net Interest Margin + Net Other Structural Income (commissions, fees, other operating income). Transitory net income aggregates valuation changes and transitory revenues/expenditures (realized and unrealized gains/losses on foreign exchange, securities, fixed assets; net transfers to/from reserves; provisions/write-offs; extraordinary gains/losses).
- Currency demand is voluntary (contrast to bank deposits and government deposits).
- Reserve currency assumed U.S. dollar for exchange-rate valuation adjustments; currency-composition weights for foreign reserves: All other countries: Dollar 70, Euro 25, Yen 5; Euro countries: Euro 70, Yen 30; U.S.: Dollar 50, Euro 50.

### Income statement—sample means and dispersion (Table 2 highlights)
- Full Sample structural income (SI): 6.32 (as percent of currency); Weak Central Banks: -3.26; Strong Central Banks: 10.64.
  - Net Interest Margin: Full Sample 5.80 (13.95); Weak -3.35 (10.42); Strong 9.92 (13.43). (Standard deviations in parentheses.)
  - Other structural income: Full Sample 0.51 (2.39); Weak 0.09 (2.83); Strong 0.72 (2.16).
- Transitory Income: Full Sample 0.25; Weak 5.92; Strong -2.30.
  - Valuation changes: Full Sample -0.14 (17.07); Weak 4.02 (13.27); Strong -2.0 (18.32).
  - Other transitory: Full Sample 0.39 (4.99); Weak 1.91 (6.39); Strong -0.29 (4.10).
- Operating Expenditures: Full Sample 4.85 (6.04); Weak 6.63 (6.26); Strong 4.71 (2.16).
- Transfers: Full Sample 2.53 (5.38); Weak 2.26 (8.48); Strong 2.65 (3.23).
- Retained Profits: Full Sample -0.81 (18.35); Weak -6.23 (17.22); Strong 1.63 (18.45).

Key observations:
- Mean transitory net income is close to zero but with very large standard deviation (valuation gains/losses drive volatility).
- Structural profitability (SI minus operating expenditures) is low: about 1.5 percent of currency (aggregate implication).
- Operating expenditures nearly exhaust the interest margin on average; retained profits are, on average, negative for the sample (base year losses).

### Variance decomposition of retained profits (Table 3)
- Full Sample shares: Net Interest Margin 0.25; Net Other Structural Income 0.03; Valuation Gains/Losses 0.65; Other Transitory Net Income 0.01; Operating Expenditure -0.02; Transfers 0.08.
- Weak central banks: Net Interest Margin 0.15; Net Other Structural Income 0.07; Valuation Gains/Losses 0.45; Other Transitory Net Income 0.05; Operating Expenditure -0.02; Transfers 0.30.
- Strong central banks: Net Interest Margin 0.21; Net Other Structural Income 0.01; Valuation Gains/Losses 0.81; Other Transitory Net Income 0.01; Operating Expenditure -0.04; Transfers 0.01.

Implication:
- Valuation gains/losses account for the largest share of retained-profits variance, especially for strong central banks (0.81).

### Balance sheet patterns (Table 5 highlights)
- Full Sample net claims: 154.6 (percent of currency).
  - On foreigners: Assets 277.1; Assets total 354.9; Liabilities -77.9.
  - On Government: Assets 134.1; Liabilities -83.4; Net claims on government 50.7.
  - On Banks: Assets 103.2; Liabilities -276.3; Net claims on banks -173.1.
- Shadow Capital (percent of currency): Full Sample 154.7; Weak 104.2; Strong 177.4.
  - Currency = 100.0 for all samples.
  - Capital: Full Sample 54.7; Weak 4.2; Strong 77.4.

Differences between subsamples:
- Weak central banks: much higher foreign liabilities; higher net claims on government (On Government net 92.2 vs Strong 32.0); lower claims on banks and lower liabilities to banks; much less capital (Capital 4.2 vs Strong 77.4).
- Weak central banks hold nearly as many international reserves as strong central banks but provide more financing to governments and finance higher claims through more foreign debt.

Policy/credibility implication:
- Asset accumulation pattern in weak central banks (high foreign liabilities, high government claims, low capital) can undermine credibility of monetary policy and lead to progressive decapitalizations and debt sustainability problems.

### Bivariate statistical findings (income and balance sheet T-tests; Tables 6a/6b and 7a/7b)
Income statement tests:
- Retained profits strongly related to transitory net income for both weak and strong banks (suggests avoidance of transferring valuation gains/windfalls).
- Strong negative association between retained profits and transfers for weak central banks (weak banks often forced to transfer profits they do not have).
- Operating expenditures positively related to net interest margin for both groups; link stronger for strong central banks.
- No strong link between capital and structural profitability (mild t-statistic between interest margin and capital for strong banks).

Balance sheet tests:
- Both weak and strong central banks: very strong positive link between claims on government and foreign liabilities (central banks borrow abroad and on-lend to governments).
- Stronger link for strong banks: central banks with more government deposits invest more abroad.
- Strong banks: extremely strong relation between bank deposits and claims on banks (central banks intermediate funds for banks domestically); structural profits unrelated to bank deposits (suggests monetary motive rather than budgetary).
- Structural profits negatively related to claims on government and government deposits in weak central banks; opposite in strong central banks.
- More capitalized central banks have higher foreign assets, but only for strong central banks.

Selected t-statistics and significance levels are reported in the source tables (90 percent significance level: *; 95 percent significance level: **).

### Interest rates, risk premia, and valuation adjustments
- Interest rate measurement: yields on treasury bonds used when available; where only treasury bill rates available, bond rate obtained by adding average spread between bond and bill observed in countries with both.
- Table 8 sample statistics:
  - Nominal Rate: Full Sample 9.31; Weak Central Banks 13.05; Strong Central Banks 7.73.
  - Inflation: Full Sample 5.16; Weak 9.22; Strong 3.43.
  - Nominal Premium: Full Sample 7.21; Weak 10.95; Strong 5.62.
  - Real Rate: Full Sample 3.99; Weak 3.72; Strong 4.10.
  - Real Premium: Full Sample 4.15; Weak 3.88; Strong 4.26.
- Finding: In real terms there is no significant difference between strong and weak central banks despite higher nominal rates and inflation in weak central bank countries.
- Valuation adjustments computed using two polar methods (average of both): method 1 uses beginning-of-year balances for price adjustments and end-of-year for volume changes; method 2 reverses this. Domestic inflation via each country’s CPI; foreign inflation via reserve countries’ CPIs weighted per Table 9.

### Seigniorage decomposition and uses (Table 10)
- Total seigniorage (percent of currency):
  - Full Sample: 19.97; Weak Central Banks: 14.66; Strong Central Banks: 22.36.
- Decomposition of seigniorage (components in Table 10):
  - Rc (seigniorage from currency): Full Sample 8.64; Weak 12.09; Strong 7.09.
  - Rk (seigniorage from capital): Full Sample 2.30; Weak -1.19; Strong 3.87.
  - Rd (seigniorage from nonremunerated deposits): Full Sample 9.03; Weak 3.76; Strong 11.40.
- Uses of seigniorage:
  - Balance sheet costs: Full Sample 9.40; Weak 10.74; Strong 8.79.
    - φexF (carrying cost of international reserves): Full Sample 8.16; Weak 10.47; Strong 7.11.
    - πk (user cost of capital): Full Sample 1.24; Weak 0.27; Strong 1.68.
  - Expenditure flows: Full Sample 6.99; Weak 6.98; Strong 7.00.
    - O (Operating expenditures): Full Sample 4.85; Weak 6.63; Strong 4.06.
    - Z (Other exceptional expenditures): Full Sample -0.39; Weak -1.91; Strong 0.29.
    - T (Transfers): Full Sample 2.53; Weak 2.26; Strong 2.65.
  - Capital Accumulation: Full Sample 3.58; Weak -3.05; Strong 6.57.
    - e∆xF (change in foreign assets, real): Full Sample 20.58; Weak 26.84; Strong 17.77.
    - ∆xG (change in net claims on government): Full Sample 4.28; Weak -13.21; Strong 12.15.
    - ∆xB (change in net claims on banks): Full Sample -14.96; Weak -10.11; Strong -17.14.
    - ∆c (change in currency): Full Sample -6.32; Weak -6.57; Strong -6.21.

Key seigniorage insights:
- Unremunerated deposits (Rd = 9.03 for Full Sample) produce more seigniorage than currency (Rc = 8.64 for Full Sample).
- Balance sheet costs, mainly carrying cost of international reserves, dominate seigniorage use (~half of total use: Balance sheet costs 9.40 out of Seigniorage 19.97 for Full Sample).
- Operating expenditures account for a substantial share of seigniorage use (~35 percent of uses; O = 4.85).
- Net asset accumulation dominated by international reserve accumulation (e∆xF constitutes the bulk of capital accumulation flows).
- Weak central banks generate lower total seigniorage (14.66) than strong central banks (22.36) despite higher Rc (currency seigniorage) because weak banks have much lower Rk and Rd.
- Weak central banks: capital accumulation negative (-3.05) vs strong central banks positive (6.57); weak central banks lose real capital.
- Possible interpretations for low Rd in weak central banks:
  - Up to 84 percent of deposits at weak central banks could be remunerated (sample-average-based interpretation), versus 32 percent for strong central banks; alternatively, assuming no remunerated deposits implies only 10 percent of weak central banks’ assets on government and banks are remunerated compared with 66 percent for strong central banks.
- Weak central banks face higher carrying costs of foreign liquidity than strong central banks despite equivalent real risk premia and lower net foreign reserves; for weak banks risk premia and net foreign assets are positively correlated (weak countries may raise reserves to buy stability).

### Synthesis of main findings and policy-relevant implications
- Central banks’ financial accounts are highly volatile and opaque: valuation gains/losses create large year-to-year dispersion; lack of uniform reporting formats impedes cross-country comparison.
- Structural profitability is low (about 1.5 percent of currency) and operating expenditures often nearly exhaust net interest margins, producing negative retained profits on average.
- Weak central banks differ systematically: poorer and smaller on average; lower governance index (KKM index: Full Sample 0.39; Weak -0.01; Strong 0.54); negative structural profits; much lower capital (4.2 vs 77.4 percent of currency); higher operating expenditures (6.63 vs 4.71).
- Weak central banks may lack independence: transfers explain a sizable fraction of variance of retained profits for weak banks; weak banks often forced to transfer non-existent profits.
- Accumulation patterns of weak central banks (high foreign liabilities, higher on-lending to government, lower capital) can worsen credibility and lead to decapitalization and debt sustainability concerns unless addressed by policy changes or recapitalization.
- Transparency and standardized reporting are urgent needs: the switch to international accounting standards (charging all valuation changes against profits) could exacerbate volatility and opacity unless reporting is clarified and standardized.

*Source: _wp0658 - Appendix I describes the data set used in this study.*

### conclusion only holds, however, under a strict definition of capital. Under a broader

### _wp0658 - conclusion only holds, however, under a strict definition of capital. Under a broader

### Capital, seigniorage, and weak vs. strong central banks
- Under a strict definition of capital, the paper's main conclusion holds; under a broader definition (that includes currency), weak central banks, as a group, still accumulate some capital (or at least did so in the year this study is based on).
- Accumulation by weak central banks is much lower than by strong central banks but might be sufficient to stave off a downward spiral into debt unsustainability.
- Weak central banks do not seem to face substantial penalties regarding interest rate premia.
- Weak central banks are able to accumulate international reserves despite weak financial condition due to a high accumulation of liabilities with government.16
- Governments end up benefiting from seigniorage proceeds only when their central banks are strong.
- Strong central banks have nearly insignificant exceptional expenditures; weak central banks report significant exceptional incomes — possibly circumstantial but potentially reflecting “creative accounting.”

### Determinants of central banks’ operating expenditures — demand vs supply drivers
- Two broad categories of central bank services: 
  - monetary management (price stability and money market development), and 
  - financial system stability and development (supervision and payments, including provision of currency).
- Expected drivers:
  - Under a soft revenue constraint (strong central banks): demand-side factors dominate — population, per capita GDP, size of financial sector, scope of responsibilities (e.g., banking supervision).
  - Under a hard revenue constraint: supply-side factors dominate — availability of seigniorage (inflation, currency in circulation, currency growth), cost of central bank services (wage level), and governance indices.
- Empirical indicator: “The very high (82 percent) correlation between currency issued and central banks’ operating expenditures” suggests supply-side effects might dominate at first sight.

### Empirical model comparison and main quantitative findings
- Models compared: Demand-side regressors (log population, log per capita GDP, log M2/GDP, banking supervision dummy) vs Supply-side regressors (log currency in circulation, KKM governance index, average inflation rate over past 10 years).
- Key empirical results (summary of elasticities and effects reported):
  - In the demand model, clear size and wealth effects with increasing returns: a 10 percent increase in country size translates into a 7.5 percent increase in operating expenditures; a 10 percent increase in per capita income translates into a 5.8 percent increase in operating expenditures.
  - In the supply model, increasing returns present: a 10 percent increase in currency issued translates into a 7 percent increase in operating expenditures.
  - Provision of supervisory services increases operating expenditures by less than 2 percent.
  - Improvement in the KKM index from the worst to the best country reduces operating expenditures by slightly over 1 percent.
- Subsample differences:
  - Financial depth (M2/GDP) significantly impacts operating expenditures of weak central banks but not of strong central banks; other effects are very similar across subsamples.
- Model selection:
  - Using Davidson and MacKinnon’s J-statistics for non-nested models, the demand model dominates the supply model for the aggregate sample — consistent with numerical dominance of strong central banks.
  - Conclusion: Overall, central banks’ operating expenditures are driven by a soft budget constraint.

### Reduced-form equilibrium model and performance index
- An equilibrium reduced-form model combines significant demand and supply factors; residuals of this reduced-form are used to derive a “relative performance” index, p_i, for central bank i:
  - 100(1), i i pe δ =−
  - where i δ is the residual of the reduced-form regression. Thus, p_i is the percentage over- or underspending of central bank i compared to its peers.
- Distribution of performance:
  - The best performer spends over three times less than the group average.
  - The worst performer spends over five times more than the group average.
  - Weak central banks tend to have lower (worse) expenditure performance indices and are grouped on the right-hand side of the performance chart.
- Correlation with profitability:
  - Scatter analysis between expenditure performance index and ratio of structural profitability to currency indicates central banks that overspend tend to be least profitable — pointing toward weak governance.

### Relationship between expenditures and macro-financial outcomes
- Simple cross-country regression of the expenditure performance index against indicators of macro-financial stability (average and standard deviation over past 20 years of yearly inflation, standard deviation of output growth, standard deviation of M2/GDP — all detrended).
- Results summarized (Table 13):
  - LOG(Inflation): Coefficient 0.41, Std. Err. 0.24, t-Statistic 1.71*, Prob. 0.0941
  - LOG(GDP Volatility): Coefficient 0.13, Std. Err. 0.41, t-Statistic 0.32, Prob. 0.7482
  - LOG(Inflation Volatility): Coefficient -0.25, Std. Err. 0.21, t-Statistic -1.21, Prob. 0.2332
  - LOG(M2/GDP Volatility): Coefficient -0.11, Std. Err. 0.07, t-Statistic -1.62, Prob. 0.1127
  - Constant C: 2.76, Std. Err. 0.56, t-Statistic 4.93, Prob. 0.0000
  - R-squared 0.127798; Adjusted R-squared 0.040578; S.E. of regression 1.265758; Sum squared resid 64.08570; Log likelihood -71.80730; F-statistic 1.465231; Prob(F-statistic) 0.230834; Durbin-Watson stat 2.157605
- Interpretation: No clear link identified between central banks’ operating expenditures and macro-financial outcomes; higher expenditures associated with lower inflation volatility and financial volatility but higher inflation; significance levels are low.

### Policy implications and recommendations
- Central bank independence remains necessary; central banks with more governance problems also appear to be the least independent.
- More independent central banks tend to have better governance, alleviating concerns that independence would lead to uncontrolled increases in central bank expenditures.
- Allowing central banks to manage their budgets and set wage scales is essential for evolving toward a modern central bank with few, highly skilled staff.
- Governance concerns are not limited to financially weak central banks:
  - Operating expenses are strongly related to profitability and vary widely across similar environments, suggesting many central banks do not strive to minimize costs on behalf of their shareholder.
  - However, operating expenses are more closely linked to demand factors than supply factors — supporting a benign interpretation where central banks aim to fulfill their stability mandate without weighing marginal public-sector trade-offs.
- Independence must be paired with accountability and transparency:
  - Income statements should distinguish clearly between transitory and permanent (structural) components.
  - Central banks not publishing income statements should do so.
- Transparency enables systematic peer comparisons and benchmarking.
- Governments should periodically review central bank expenditures, including the cost of reserve accumulation; recapitalizations may be opportunities for broader policy debates.
  - Example practice: the New Zealand arrangement where seigniorage income belongs to the State but a negotiated share is retained by the central bank for funding.
- Pricing of services and user charges:
  - Central banks should ensure services are adequately priced and paid by users.
  - Carrying international reserves is costly; this can justify not paying full market interest (i.e., not paying the risk premium) on deposits.
  - Unremunerated reserve requirements or fee-based mechanisms may be justified to internalize liquidity costs.
  - The remaining difficult issue is defining a proper level of services provided, e.g., an appropriate level of international reserves.

### Caveats, limitations, and directions for further research
- Study limitations:
  - Cross-sectional evidence from a single year — raises causality issues and difficulty distinguishing stable patterns from circumstantial occurrences.
  - Some conclusions based on simple averages; large standard deviations across subsamples indicate considerable heterogeneity; risk that results may be dominated by large deviations in a few countries.
  - Overcoming these issues requires panel analysis over a sufficiently long period (and a larger data-preparation investment).
- Assessing central bank output is inherently difficult:
  - Lack of counterfactual evidence of economic performance without a central bank.
  - The study does not measure central bank efficiency in the traditional sense (output for a given input) nor the quality or quantity of services.
  - Measuring output would require a well-defined output variable and decomposition of operating expenditures into price (staff salaries) and volume (employment, staff education) components.
- The performance index should be interpreted with caution; more analysis is needed to firm up conclusions.

_Italic: Source — content as provided from the PDF chapter/section._

### REFERENCES

### _wp0658 - REFERENCES

### Cited works

- Davidson, R., and J. MacKinnon, 1981, Several Tests for Model Specification in the Presence of Alternative Hypotheses, Econometrica, Vol. 49, No. 3, pp. 781–93.
- Eichengreen, B. and D. J. Mathieson, 2000, “The Currency Composition of Foreign Exchange Reserves: Retrospect and Prospect,” IMF Working Paper 00/131 (Washington: International Monetary Fund)
- Fry, M., C. Goodhart, and A. Almeida, 1996, Central Banking in Developing Countries (New York: Routledge).
- Kaufmann, D., A. Kraay, and M. Mastruzzi, 2003, “Governance Matters III: Governance Indicators for 1996–2002,” World Bank Policy Research Department Working Paper.
- Ize, A., 2005, “Capitalizing Central Banks: A Net Worth Approach,” Staff Papers, International Monetary Fund, Vol. 52, No. 2, pp. 289–310.
- McKinley, V. and K. Banaian, 2005, Central Bank Operational Efficiency: Meaning and Measurement,” in Central Bank Modernization, ed. by Peter Nicholl and Neil Counts, (London: Central Banking Publications).
- Stella, P., 1997, “Do Central Banks Need Capital?” IMF Working Paper 97/83 (Washington: International Monetary Fund).
- Stella, P., 2005, “Central Bank Financial Strength, Transparency, and Policy Credibility,” Staff Papers, International Monetary Fund, Vol. 52, No. 2, pp. 335–65.

*Source: _wp0658 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0658.pdf_