## 8. Government Acquisitions of Equities and Fiscal Transparency

## Source details

**Canonical URL:** [8. Government Acquisitions of Equities and Fiscal Transparency](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1509.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1509.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1509.pdf.json)

---

### Introduction
- Data and scope:
  - Integrated public finance data from IMF’s Government Finance Statistics Yearbook (GFSY) for an unbalanced panel of 25 countries over the 1995–2012 period.
- Definitions and decomposition:
  - Stock-flow residuals defined in line with Government Finance Statistics Manual 2014 (GFSM 2014) as financial transactions and other economic flows; decomposition isolates components governments could use strategically.
  - Realized and unrealized returns on financial instruments (mainly equity and investment fund shares) interpreted as profitability indicators of government investments.
- Empirical summary:
  - Governments tend to increase investment in equities when:
    - generating fiscal surplus balances;
    - increasing gross debt to finance investments;
    - softening the impact of exogenous shocks to other sectors of the economy;
    - in response to expected realized and unrealized returns on equity portfolios.
  - Governments with consistent surpluses or large sovereign wealth funds can generate higher returns from equity portfolios than from debt securities.
- Terminology note:
  - Unrealized returns are defined as those whose prices are derived from secondary markets and are either held by government at the end of an accounting period or sold by government during that period at market price.

### Overview of the literature
- Stock-flow residuals and fiscal gimmickry:
  - Literature focuses on inconsistencies (stock-flow residuals) and fiscal gimmickry where governments adjust accounts to meet visible indicators (deficits and gross debt) while hiding liabilities.
  - Buti et al. (2007) decompose stock-flow adjustments into three components and derive two measures of “hidden deficits”: timing tool (cash vs accrual timing manipulations) and separation of ‘safe’ from potentially ‘unsafe’ assets (safe: securities and equity investments of social security funds; unsafe: loans and equity investments outside social security subsector).
  - Empirical finding: governments subject to fiscal rules (Maastricht) more likely to use sale of financial assets to finance deficits or decrease gross debt (sample: 25 EU countries, 1994–2004).
- Balance-sheet approach argument:
  - Fiscal rules based solely on gross debt/deficits create incentives for “hidden deficits”; a balance-sheet approach emphasizing changes in net worth (or net debt) can reduce loopholes.
- Gap addressed:
  - Prior work did not consider returns on government portfolios; this paper incorporates realized and unrealized returns on government equity portfolios as profitability indicators.

### Decomposing the stock-flow residual
- Conventional partial measure:
  - Difference between changes in gross debt and deficits is incomplete because it omits financial asset transactions and other economic flows.
- Empirical summary (general government, 1995–2012):
  - Average change in debt: 4.0 percent GDP (sd=6.5).
  - Average deficit: 1.8 percent GDP (sd=4.8).
- Missing components reconciling changes in gross debt and fiscal deficits (consistent with GFSM 2014):
  - transactions in financial assets (∑ x̃_i,t^FA);
  - other economic flows (∑ holding gains/losses and volume changes for instruments).
- Figures referenced:
  - Figure 1: relationship between surplus/deficit and changes in gross debt (35 countries, 1995–2012).
  - Figure 2: relationship between fiscal flows ((1c)+(1d)) and changes in government net financial worth (1a) — shows substantial realignment vs Figure 1 confirming non-trivial magnitude of missing components.

### Transactions in financial assets (selected exact figures)
- Table 1 averages (Percent of GDP) for three periods (Pre-2008, 2008, Post-2008):
  - Net Acquisition of Financial Assets: Pre-2008 = 0.9 (2.3); 2008 = 4.0 (6.4); Post-2008 = 1.6 (3.1).
  - Currency and Deposits: Pre-2008 = 0.3 (1.5); 2008 = 1.0 (3.2); Post-2008 = 0.1 (2.1).
  - Debt Securities: Pre-2008 = 0.3 (1.3); 2008 = 0.4 (0.7); Post-2008 = 0.23 (1.5).
  - Loans: Pre-2008 = 0.02 (0.8); 2008 = 1.3 (3.5); Post-2008 = 0.2 (1.0).
  - Equity and Investment Fund Shares: Pre-2008 = -0.01 (1.4); 2008 = 0.9 (2.0); Post-2008 = 0.6 (1.6).
  - Other Accounts Receivable: Pre-2008 = 0.4 (0.7); 2008 = 0.4 (0.7); Post-2008 = 0.4 (0.8).
  - Financial Derivatives and Employee Stock Options: Pre-2008 = -0.02 (0.1); 2008 = -0.001 (0.06); Post-2008 = -0.02 (0.1).
- Cross-country patterns:
  - Advanced economies: more dispersed portfolios with shares and equities more prominent, especially during 2008 bailouts; greater emphasis on loans and equity purchases of temporarily insolvent firms.
  - Emerging/developing economies: higher reliance on currency and deposits.

### Other economic flows (volume and valuation changes)
- Decomposition:
  - Holding gains/losses (valuation changes) — reflect unrealized profits/losses and exchange rate effects on current stock of financial investments.
  - Volume changes — events neither transactions nor valuation changes (natural disasters, reclassifications, unilateral debt write-offs, discovery of assets/liabilities).
- Empirical characteristics:
  - Volume changes: density centered around zero with thin tails range -2 to 2 percent of GDP.
- Table 2 averages (Percent of GDP) for three periods (Pre-2008, 2008, Post-2008) — selected exact figures:
  - Net Acquisition of Financial Assets: Pre-2008 = 1.26 (3.75); 2008 = -1.94 (5.61); Post-2008 = 1.16 (2.81).
  - Equity and Investment Fund Shares: Pre-2008 = 1.30 (3.57); 2008 = -2.86 (6.96); Post-2008 = 1.30 (2.44).
  - Debt Securities: Pre-2008 = -0.03 (0.42); 2008 = 0.49 (2.59); Post-2008 = -0.24 (1.19).
  - Loans: Pre-2008 = -0.03 (0.40); 2008 = 0.60 (1.94); Post-2008 = -0.03 (0.28).
- Interpretation:
  - Transactions in financial assets and other economic flows are significant for understanding fiscal health and transparency.
  - Acquisition of financial assets can:
    - provide liquidity to offset short-term gross debt;
    - generate profits for surplus-generating governments;
    - serve to absorb exogenous shocks to other sectors.
  - Other economic flows supply valuation change information from secondary markets and exchange rates.

### Revisiting stock-flow residuals and fiscal transparency
- Government portfolio objectives differ from private sector objectives (social welfare, economic stability, political considerations).
- Focus on equities (shares and investment fund shares) because:
  - Equities show large transaction magnitudes and variable holding gains;
  - Holding gains on loans are generally zero.
- Returns decomposition and measurement:
  - Effective returns on asset q: r_q = r_q^D + r_q^U where r_q^D = realized (interest/dividend revenue), r_q^U = holding gains (unrealized).
  - Realized returns measured as DR_t / x̅_t (DR = total dividend revenue over period t; x̅_t = stock of government equities at end of period t).
  - Unrealized returns measured as Δ holding gains / x̅_t.
- Figure 5 findings (averages, general government net of social security relative to S&P 500):
  - Governments are relatively conservative vs private sector: lower volatility in unrealized returns.
  - Social security subsector performs on average quite poorly compared with secondary market movements in equity portfolio of remainder of general government.
  - Governments tend to favor acquisition of equities during market downturns — consistent with countercyclical actions to support institutional units.

### Estimation strategy
- Model specification:
  - Two equations estimated: (1) general government transactions in equities; (2) unrealized returns on equities.
  - Key regressors: net lending/borrowing (surplus/deficit) τ_jt; change in gross debt ΔB_jt; banking crisis dummy; realized returns (dividend receipts); unrealized returns; private market returns (year-on-year changes in S&P 500 index); fiscal transparency (IMF ROSC database); exchange rate changes; standard macro controls (GDP, growth).
  - Estimation method: random intercept model (GLS); Hausman test used to validate relative consistency versus fixed effects.

### Empirical results — Unrealized returns (selected coefficients from Table 3)
- Sample: unbalanced panel of 19 countries (bootstrapped standard errors in parentheses).
- Coefficient on market returns (Δ S&P 500):
  - Column 1 (General Government): 0.20*** (0.07).
  - Column 2 (General Government net of Social Security): 0.16** (0.64).
  - Column 3: 0.12*** (0.03).
  - Column 4: 0.12*** (0.03).
- Coefficient on fiscal transparency (σ):
  - Column 1: 6.63* (3.63).
  - Column 2: 9.73* (5.85).
  - Column 3: 9.76 (6.73).
  - Column 4: 6.72 (10.42).
- Observations and fit:
  - Countries: 19, 14, 10, 9 across columns.
  - Observations: 240, 160, 94, 84.
  - Rsq (within): 0.09, 0.06, 0.04, 0.04.
- Interpretation:
  - Governments’ unrealized returns significantly correlated with private market returns: a 1 percent increase in S&P 500 associated with about 0.2 percent higher government returns in Column 1.
  - Fiscal transparency positively associated with unrealized returns: relatively transparent governments hold more profitable equity portfolios by between 6 to 8 percent relative to less transparent governments (see Figure 6).

### Empirical results — Net acquisition of equities (selected coefficients from Table 4)
- Sample: unbalanced panel of 25 countries (bootstrapped standard errors in parentheses).
- Coefficient on net lending/borrowing (τ):
  - General Government (Domestic and Foreign): 0.27*** (0.08).
  - General Government net of Social Security: 0.14*** (0.05).
  - Domestic only columns: 0.12* (0.07); 0.14** (0.07).
- Coefficient on changes in gross debt (ΔB):
  - 0.13*** (0.03); 0.13*** (0.04); 0.08** (0.04); 0.10*** (0.03).
- Coefficient on banking crisis (dummy):
  - 0.57* (.32); 0.52** (0.26); 0.90* (0.53); 0.56* (0.32).
- Coefficient on realized returns (dividends):
  - -0.07 (0.05); -0.06 (0.04); -0.05 (0.07); -0.01 (0.05).
- Coefficient on unrealized returns:
  - -0.04* (0.03); -0.3** (0.13); -0.03* (0.02); -0.02 (0.02).
- Observations and fit:
  - Countries: 25, 18, 18, 15 across columns.
  - Observations: 274, 200, 180, 148.
  - Rsq (within): 0.21, 0.37, 0.21, 0.20.
  - Rsq (overall): 0.41, 0.36, 0.21, 0.23.
- Interpretation:
  - Key drivers of government net acquisitions of equities:
    - positive relation with fiscal balances (surplus countries more likely to acquire equities);
    - positive relation with increases in gross debt (use of debt to finance acquisitions);
    - positive relation with exogenous shocks (banking crisis);
    - realized returns (dividends) play minimal role;
    - negative coefficient on unrealized returns suggests governments increase equity purchases during market downturns (willingness to invest countercyclically).

### Predicted second-order effects and counterfactuals
- Predictions using unrealized returns model (3b) imputed into acquisition model (3a):
  - Figure 7: Governments partially favor equities during market downturns; governments are highly active in equity injections during domestic banking crises (especially in domestic equities).
  - Figure 8: Predicted effect of fiscal transparency on net acquisitions:
    - Less transparent governments expected to acquire relatively larger equity portfolios during normal times.
    - Combined with lower returns, this pattern suggests the potential presence of ‘policy lending’ or ‘unsafe assets’.
- Specification note:
  - Specification treats fiscal transparency and market volatility effects as additive and independent; multiplicative effects explored in alternative specification with promising results but requiring larger sample.

### Discussion and conclusion
- Principal conclusions:
  - Government acquisition of equities can serve multiple roles: generate returns for surplus governments; cushion exogenous shocks to domestic economy; or hide expected losses from policy lending.
  - Government acquisition of equities alone is not definitive evidence of fiscal gimmickry: average returns over time for transparent governments tend to be relatively profitable.
  - The magnitude of unrealized returns on government equity portfolios is a more promising indicator of fiscal transparency.
- Data and transparency limitations:
  - Comprehensive fiscal surveillance based on complete financial statistics (equation (2b)) could reduce incentives for fiscal gimmickry.
  - Data availability constraint: of 140 countries reporting government finance statistics in GFSY 2014, only about 18 percent reported sufficient data to compute equation (2b).
  - Implication: results based on a truncated sample of relatively transparent countries are preliminary and require larger sample sizes for validation.

### Appendix I. Country-Year Coverage (country list, selected entries)
- Australia 2004–2012
- Austria 1997–2012
- Hong Kong 2006–2012
- Colombia 2009–2012
- Cyprus 2001–2012
- Denmark 1996–2012
- Estonia 1996–2012
- Finland 1996–2012
- France 1996–2012
- Greece 1996–2012
- Hungary 1996–2012
- Iceland 2002–2008
- Italy 1996–2012
- Japan 2001-2012
- Lithuania 2008–2012
- Luxembourg 2002–2012
- Malta 2004–2012
- Netherlands 1996–2012
- Norway 2000–2012
- Portugal 1998–2012
- Russian Federation 2005-2012
- Slovak Republic 2003–2012
- Spain 1996–2012
- Sweden 1996-2012
- Turkey 2008-2012
- United Kingdom 1996–2012

### Appendix II. Selected summary statistics (exact figures preserved)
- General Government change in Gross Debt: 5.35 (15.75) Min. -22.46 Max. 30.09 — Source: IMF Government Finance Statistics Yearbook (GFSY)
- General Government Net lending/borrowing: -1.71 (5.08) Min. -15.63 Max. 18.79
- General Government realized returns on equities other than shares (% Stocks): 3.65 (2.89) Min. 0.02 Max. 16.53
- General Government unrealized returns on equities other than shares (% Stocks): 4.18 (12.09) Min. -50.34 Max. 50.53
- General Government Transactions in Financial Assets (% GDP): 1.73 (3.77) Min. -5.35 Max. 34.47
  - Currency and Deposits: 0.51 (2.03) Min. -10.88 Max. 14.65
  - Loans: 0.22 (1.66) Min. -20.09 Max. 14.38
  - Debt Securities: 0.36 (1.86) Min. -14.00 Max. 24.45
  - Equity and Investment Fund Shares: 0.37 (2.16) Min. -6.48 Max. 21.83
  - Derivatives and Employee Stock Options: -0.04 (0.24) Min. -2.64 Max. 0.26
  - Other Accounts Receivable: 0.37 (0.89) Min. -4.66 Max. 6.44
- General Government Other Economic Flows in Financial Assets: 1.00 (3.67) Min. -22.25 Max. 32.29
  - Currency and Deposits: 0.04 (0.33) Min. -0.63 Max. 4.93
  - Loans: -0.01 (0.85) Min. -8.38 Max. 6.91
  - Debt Securities: 0.02 (0.85) Min. -7.00 Max. 11.65
  - Equity and Investment Stock Options: 0.90 (3.80) Min. -25.27 Max. 31.12
  - Derivatives and Employee Stock Options: 0.09 (0.41) Min. -0.79 Max. 3.02
  - Other Accounts Receivable: -0.002 (0.71) Min. -5.79 Max. 7.16
- Gross Domestic Product (ln): 27.38 (2.26) Min. 22.23 Max. 34.13 — Source: IMF International Financial Statistics Yearbook
- Changes in Foreign Exchange Rates (annual %): 0.19 (9.35) Min. -16.95 Max. 40.58
- Banking Crisis: 0.22 Min. 0 Max. 1 — Source: Weber (2012)
- Fiscal Transparency: 0.71 (0.13) Min. 0.38 Max. 0.90 — Source: IMF Fiscal Transparency Report on Observance of Standards and Codes

*Source: IMF Working Paper chapter “8. Government Acquisitions of Equities and Fiscal Transparency” (extracted from the supplied PDF content).*

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

### _wp1509 - References

### Tables
- 1. Pre- and Post-Crisis Net Transactions in Financial Assets by Instrument...........................10
- 2. Pre- and Post-Crisis Other Economic Flows in Financial Assets by Instrument .........................12
- 3. Unrealized Returns on Shares and Equity ...........................................................................17
- 4. Government Net Acquisition of Shares and Equity .............................................................18

### Figures
- 1. General Government Fiscal Balances and Changes in Gross Debt .......................................7
- 2. General Government Flows and Changes in Net Financial Worth ........................................9
- 3. Average Transactions in Select Financial Assets ................................................................10
- 4. General Government Financial Asset Volume Change Densities .......................................12
- 5. Average Returns on Equity Investments, General Government, and Social Security .........14
- 6. Predicted Unrealized Returns on Government Equity Portfolios and Fiscal 
  Transparency ......................................................................................................................17
- 7. Government Acquisitions of Equities—Market Downturns and Bank Crises ....................19

*Source: _wp1509 - References*

### 8. Government Acquisitions of Equities and Fiscal Transparency ..........................................20

### 8. Government Acquisitions of Equities and Fiscal Transparency

### Introduction
- Uses integrated public finance data from the IMF’s Government Finance Statistics Yearbook (GFSY) for an unbalanced panel of 25 countries over the 1995–2012 period.
- Defines stock-flow residuals in line with Government Finance Statistics Manual 2014 (GFSM 2014) as financial transactions and other economic flows, and decomposes them to identify components governments could use strategically.
- Interprets realized and unrealized returns on financial instruments (mainly equity and investment fund shares) as profitability indicators of government investments.
- Empirical findings summary:
  - Governments tend to increase investment in equities when:
    - generating fiscal surplus balances;
    - increasing gross debt to finance investments;
    - softening the impact of exogenous shocks to other sectors of the economy;
    - in response to expected realized and unrealized returns on equity portfolios.
  - Governments with consistent surpluses or large sovereign wealth funds can generate higher returns from equity portfolios than from debt securities.
- Note on terminology: unrealized returns are defined as those whose prices are derived from secondary markets and are, either held by government at the end of an accounting period or sold by government during that period at market price.

### Overview of the literature
- Stock-flow residuals literature:
  - Focus on inconsistencies (stock-flow residuals) and fiscal gimmickry where governments adjust accounts to meet visible indicators (deficits and gross debt) while hiding liabilities.
  - Buti et al. (2007) decompose stock-flow adjustments into three components and derive two measures of “hidden deficits”:
    - timing tool (cash vs accrual timing manipulations);
    - separating ‘safe’ from potentially ‘unsafe’ assets (safe: securities and equity investments of social security funds; unsafe: loans and equity investments outside social security subsector).
  - Empirical finding: governments subject to fiscal rules (Maastricht) more likely to use sale of financial assets to finance deficits or decrease gross debt (sample: 25 EU countries, 1994–2004).
- Argument for balance-sheet approach:
  - Fiscal rules based solely on gross debt/deficits create incentives for “hidden deficits”; balance-sheet approach emphasizing changes in net worth (or net debt) can reduce loopholes.
- Gap in literature addressed by this paper:
  - Prior work did not consider returns on government portfolios; this paper incorporates realized and unrealized returns on government equity portfolios as profitability indicators.

### Decomposing the stock-flow residual
- Conventional partial measure: difference between changes in gross debt and deficits; incomplete as it omits financial asset transactions and other economic flows.
- Empirical summary (sample: general government, 1995–2012):
  - Average change in debt: 4.0 percent GDP (sd=6.5).
  - Average deficit: 1.8 percent GDP (sd=4.8).
- Accounting identities (consistent with GFSM 2014) clarify missing components:
  - Missing components reconciling changes in gross debt and fiscal deficits:
    - transactions in financial assets (∑ x̃_i,t^FA);
    - other economic flows (∑ (holding gains/losses and volume changes) for instruments).
- Figures referenced:
  - Figure 1: relationship between surplus/deficit and changes in gross debt (35 countries, 1995–2012).
  - Figure 2: relationship between fiscal flows ((1c)+(1d)) and changes in government net financial worth (1a) — shows substantial realignment vs Figure 1 confirming non-trivial magnitude of missing components.

### Transactions in financial assets
- Table 1: Net transactions in financial assets (Percent of GDP), averages for three periods (Pre-2008, 2008, Post-2008). Selected exact figures:
  - Net Acquisition of Financial Assets: Pre-2008 = 0.9 (2.3); 2008 = 4.0 (6.4); Post-2008 = 1.6 (3.1).
  - Currency and Deposits: Pre-2008 = 0.3 (1.5); 2008 = 1.0 (3.2); Post-2008 = 0.1 (2.1).
  - Debt Securities: Pre-2008 = 0.3 (1.3); 2008 = 0.4 (0.7); Post-2008 = 0.23 (1.5).
  - Loans: Pre-2008 = 0.02 (0.8); 2008 = 1.3 (3.5); Post-2008 = 0.2 (1.0).
  - Equity and Investment Fund Shares: Pre-2008 = -0.01 (1.4); 2008 = 0.9 (2.0); Post-2008 = 0.6 (1.6).
  - Other Accounts Receivable: Pre-2008 = 0.4 (0.7); 2008 = 0.4 (0.7); Post-2008 = 0.4 (0.8).
  - Financial Derivatives and Employee Stock Options: Pre-2008 = -0.02 (0.1); 2008 = -0.001 (0.06); Post-2008 = -0.02 (0.1).
- Cross-country patterns:
  - Advanced economies: more dispersed portfolios with shares and equities more prominent, especially during 2008 bailouts; greater emphasis on loans and equity purchases of temporarily insolvent firms.
  - Emerging/developing economies: higher reliance on currency and deposits.

### Other economic flows (volume and valuation changes)
- Decomposition:
  - Holding gains/losses (valuation changes) — reflect unrealized profits/losses and exchange rate effects on current stock of financial investments.
  - Volume changes — events neither transactions nor valuation changes (natural disasters, reclassifications, unilateral debt write-offs, discovery of assets/liabilities).
- Empirical characteristics:
  - Volume changes: density centered around zero with thin tails range -2 to 2 percent of GDP.
  - Table 2: Other Economic Flows in Financial Assets (Percent of GDP), averages Pre-2008 / 2008 / Post-2008 (selected exact figures):
    - Net Acquisition of Financial Assets: Pre-2008 = 1.26 (3.75); 2008 = -1.94 (5.61); Post-2008 = 1.16 (2.81).
    - Equity and Investment Fund Shares: Pre-2008 = 1.30 (3.57); 2008 = -2.86 (6.96); Post-2008 = 1.30 (2.44).
    - Debt Securities: Pre-2008 = -0.03 (0.42); 2008 = 0.49 (2.59); Post-2008 = -0.24 (1.19).
    - Loans: Pre-2008 = -0.03 (0.40); 2008 = 0.60 (1.94); Post-2008 = -0.03 (0.28).
- Interpretation:
  - These two missing links (transactions in financial assets and other economic flows) are significant for understanding fiscal health and transparency.
  - Acquisition of financial assets:
    - important for liquidity to offset short-term gross debt;
    - a source of profits for surplus-generating governments;
    - a mechanism to absorb exogenous shocks to other sectors.
  - Other economic flows provide valuation change information from secondary markets and exchange rates.

### Revisiting stock-flow residuals and fiscal transparency
- Government portfolio objectives differ from private sector (social welfare, economic stability, political considerations).
- Focus narrowed to equities (shares and investment fund shares) because:
  - equities show large transaction magnitudes and variable holding gains;
  - holding gains on loans generally zero.
- Returns decomposition and measurement:
  - Effective returns on asset q: r_q = r_q^D + r_q^U where r_q^D = realized (interest/dividend revenue), r_q^U = holding gains (unrealized).
  - Measured as:
    - realized returns: DR_t / x̅_t (DR = total dividend revenue over period t; x̅_t = stock of government equities at end of period t).
    - unrealized returns: Δ holding gains / x̅_t.
- Figure 5 findings (averages, general government net of social security relative to S&P 500):
  - Governments relatively conservative vs private sector: lower volatility in unrealized returns.
  - Social security subsector performs on average quite poorly compared with secondary market movements in equity portfolio of remainder of general government.
  - Governments tend to favor acquisition of equities during market downturns — consistent with countercyclical actions to support institutional units.

### Estimation and results
- Model specification:
  - Two equations estimated for general government transactions in equities and for unrealized returns on equities; key regressors include:
    - net lending/borrowing (surplus/deficit) τ_jt;
    - change in gross debt ΔB_jt;
    - banking crisis dummy;
    - realized returns (dividend receipts);
    - unrealized returns;
    - private market returns (year-on-year changes in S&P 500 index);
    - fiscal transparency (IMF ROSC database, see Weber 2012);
    - exchange rate changes;
    - standard macro controls (GDP, growth).
  - Estimation via random intercept model (GLS); Hausman test used to validate relative consistency versus fixed effects.
- Unrealized returns regression (Table 3) — sample: unbalanced panel of 19 countries (selected coefficient estimates, bootstrapped standard errors in parentheses):
  - Coefficient on market returns (Δ S&P 500):
    - Column 1 (General Government): 0.20*** (0.07).
    - Column 2 (General Government net of Social Security): 0.16** (0.64).
    - Column 3: 0.12*** (0.03).
    - Column 4: 0.12*** (0.03).
  - Coefficient on fiscal transparency (σ):
    - Column 1: 6.63* (3.63).
    - Column 2: 9.73* (5.85).
    - Column 3: 9.76 (6.73).
    - Column 4: 6.72 (10.42).
  - Observations and fit:
    - Countries: 19, 14, 10, 9 across columns.
    - Observations: 240, 160, 94, 84.
    - Rsq (within): 0.09, 0.06, 0.04, 0.04.
  - Interpretation:
    - Governments’ unrealized returns significantly correlated with private market returns: a 1 percent increase in S&P 500 associated with about 0.2 percent higher government returns in Column 1.
    - Fiscal transparency positively associated with unrealized returns: relatively transparent governments hold more profitable equity portfolios by between 6 to 8 percent relative to less transparent governments (see Figure 6).
- Net acquisition of equities regression (Table 4) — sample: unbalanced panel of 25 countries (selected coefficient estimates, bootstrapped standard errors in parentheses):
  - Coefficient on net lending/borrowing (τ):
    - General Government (Domestic and Foreign): 0.27*** (0.08).
    - General Government net of Social Security: 0.14*** (0.05).
    - Domestic only columns: 0.12* (0.07); 0.14** (0.07).
  - Coefficient on changes in gross debt (ΔB):
    - 0.13*** (0.03); 0.13*** (0.04); 0.08** (0.04); 0.10*** (0.03).
  - Coefficient on banking crisis (dummy):
    - 0.57* (.32); 0.52** (0.26); 0.90* (0.53); 0.56* (0.32).
  - Coefficient on realized returns (dividends):
    - -0.07 (0.05); -0.06 (0.04); -0.05 (0.07); -0.01 (0.05).
  - Coefficient on unrealized returns:
    - -0.04* (0.03); -0.3** (0.13); -0.03* (0.02); -0.02 (0.02).
  - Observations and fit:
    - Countries: 25, 18, 18, 15 across columns.
    - Observations: 274, 200, 180, 148.
    - Rsq (within): 0.21, 0.37, 0.21, 0.20.
    - Rsq (overall): 0.41, 0.36, 0.21, 0.23.
  - Interpretation:
    - Key drivers of government net acquisitions of equities:
      - positive relation with fiscal balances (surplus countries more likely to acquire equities);
      - positive relation with increases in gross debt (use of debt to finance acquisitions);
      - positive relation with exogenous shocks (banking crisis);
      - realized returns (dividends) play minimal role;
      - negative coefficient on unrealized returns suggests governments increase equity purchases during market downturns (willingness to invest countercyclically).
- Predicted second-order effects:
  - Using predictions from unrealized returns model (3b) imputed into acquisition model (3a) yields:
    - Figure 7: Governments partially favor equities during market downturns; governments are highly active in equity injections during domestic banking crises (especially in domestic equities).
    - Figure 8: Predicted effect of fiscal transparency on net acquisitions: less transparent governments expected to acquire relatively larger equity portfolios during normal times; combined with lower returns this suggests potential presence of ‘policy lending’ or ‘unsafe assets’.
  - Note on independence: specification treats fiscal transparency and market volatility effects as additive and independent (multiplicative effects explored in alternative specification with promising results but requiring larger sample).

### Discussion and conclusion
- Main conclusions:
  - Government acquisition of equities can serve multiple roles: generate returns for surplus governments; cushion exogenous shocks to domestic economy; or hide expected losses from policy lending.
  - Government acquisition of equities alone is not definitive evidence of fiscal gimmickry: average returns over time for transparent governments tend to be relatively profitable.
  - The magnitude of unrealized returns on government equity portfolios is a more promising indicator of fiscal transparency.
- Data and transparency limitations:
  - Comprehensive fiscal surveillance based on complete financial statistics (equation (2b)) could reduce incentives for fiscal gimmickry.
  - Data availability: of 140 countries reporting government finance statistics in GFSY 2014, only about 18 percent reported sufficient data to compute equation (2b).
  - Implication: results based on a truncated sample of relatively transparent countries are preliminary and require larger sample sizes for validation.

*Source: IMF Working Paper chapter “8. Government Acquisitions of Equities and Fiscal Transparency” (extracted from the supplied PDF content).*

### Appendix I. Country-Year Coverage

### Appendix I. Country-Year Coverage

### Country-Year list
- Australia 2004–2012
- Austria 1997–2012
- Hong Kong 2006–2012
- Colombia 2009–2012
- Cyprus 2001–2012
- Denmark 1996–2012
- Estonia 1996–2012
- Finland 1996–2012
- France 1996–2012
- Greece 1996–2012
- Hungary 1996–2012
- Iceland 2002–2008
- Italy 1996–2012
- Japan 2001-2012
- Lithuania 2008–2012
- Luxembourg 2002–2012
- Malta 2004–2012
- Netherlands 1996–2012
- Norway 2000–2012
- Portugal 1998–2012
- Russian Federation 2005-2012
- Slovak Republic 2003–2012
- Spain 1996–2012
- Sweden 1996-2012
- Turkey 2008-2012
- United Kingdom 1996–2012

### Coverage note
- Country coverage is based on the reporting of transactions and other economic flows in financial assets.

### Source
- IMF GFSY (various years)

---

### Appendix II. Summary Statistics (selected variables)
- General Government change in Gross Debt: 5.35 (15.75) Min. -22.46 Max. 30.09 — Source: IMF Government Finance Statistics Yearbook (GFSY)
- General Government Net lending/borrowing: -1.71 (5.08) Min. -15.63 Max. 18.79
- General Government realized returns on equities other than shares (% Stocks): 3.65 (2.89) Min. 0.02 Max. 16.53
- General Government unrealized returns on equities other than shares (% Stocks): 4.18 (12.09) Min. -50.34 Max. 50.53
- General Government Transactions in Financial Assets (% GDP): 1.73 (3.77) Min. -5.35 Max. 34.47
  - Currency and Deposits: 0.51 (2.03) Min. -10.88 Max. 14.65
  - Loans: 0.22 (1.66) Min. -20.09 Max. 14.38
  - Debt Securities: 0.36 (1.86) Min. -14.00 Max. 24.45
  - Equity and Investment Fund Shares: 0.37 (2.16) Min. -6.48 Max. 21.83
  - Derivatives and Employee Stock Options: -0.04 (0.24) Min. -2.64 Max. 0.26
  - Other Accounts Receivable: 0.37 (0.89) Min. -4.66 Max. 6.44
- General Government Other Economic Flows in Financial Assets: 1.00 (3.67) Min. -22.25 Max. 32.29
  - Currency and Deposits: 0.04 (0.33) Min. -0.63 Max. 4.93
  - Loans: -0.01 (0.85) Min. -8.38 Max. 6.91
  - Debt Securities: 0.02 (0.85) Min. -7.00 Max. 11.65
  - Equity and Investment Stock Options: 0.90 (3.80) Min. -25.27 Max. 31.12
  - Derivatives and Employee Stock Options: 0.09 (0.41) Min. -0.79 Max. 3.02
  - Other Accounts Receivable: -0.002 (0.71) Min. -5.79 Max. 7.16
- Gross Domestic Product (ln): 27.38 (2.26) Min. 22.23 Max. 34.13 — Source: IMF International Financial Statistics Yearbook
- Changes in Foreign Exchange Rates (annual %): 0.19 (9.35) Min. -16.95 Max. 40.58
- Banking Crisis: 0.22 (s.d. not listed) Min. 0 Max. 1 — Source: Weber (2012)
- Fiscal Transparency: 0.71 (0.13) Min. 0.38 Max. 0.90 — Source: IMF Fiscal Transparency Report on Observance of Standards and Codes

*Source: IMF GFSY (various years)*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2015/_wp1509.pdf_