## _wp0796

## Source details

**Canonical URL:** [_wp0796](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0796.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0796.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0796.pdf.json)

---

### I. Introduction — purpose and approach
- Research question: What benefits do oil funds bring to oil exporting countries beyond fiscal stabilization?
- Key argument: Benefits of self-insurance provided by oil funds extend beyond the fiscal sector to monetary aggregates and prices, potentially improving overall macroeconomic stability.
- Empirical approach:
  - Panel dataset of 15 countries (nine with oil funds and six without) spanning three decades.
  - Controls for unobserved time-invariant heterogeneity using panel fixed effects and random effects estimators.
- Data constraints:
  - Absence of information on oil fund balances.
  - Lack of monthly series for GDP, non-oil GDP, and consumption—dependent variables limited to inflation and volatility of prices, broad money, and the real effective exchange rate (REER).

### II. Rationale for oil funds — channels and literature
- Stated rationales for oil funds:
  - Short-term self-insurance against international oil price volatility.
  - Long-term savings/transfer between generations (future generations fund), which can still provide short-term stability.
- Channels through which oil funds can provide macro stability:
  - Smoothing fiscal expenditures (fiscal cushion).
  - Reducing burden on monetary policy (sterilization substituted by fiscal-side operations).
- Literature summary and critiques:
  - Existing empirical work focuses on fiscal outcomes and finds ambivalent or limited fiscal benefits.
  - Criticisms: persistence in global oil prices may render self-insurance suboptimal; market-based contingent instruments may be preferable when markets are deep.
  - Institutional perspective: oil funds can act as binding fiscal institutions (self-control) and may have real effects on outcomes.

### III. Empirical specification and data
- Identification and estimators:
  - Parameter δ captures effect of presence of an oil fund on volatility; hypothesized δ < 0.
  - Primary estimators: OLS with robust standard errors, panel fixed effects (demeaning), and random effects (one-way error components).
  - Fixed effects used to control for time-invariant, unobserved heterogeneity.
  - Endogeneity caveat: presence of an oil fund may be endogenous to time-varying factors; authors acknowledge inability to find adequate instruments and present results as robust non-causal associations.
  - Practical mitigation: use of lagged oil fund dummy where possible.
- Dependent variables (due to data limits):
  - Inflation (growth rate of CPI) and coefficient of intra-year variation (CV) of: CPI, broad money, REER.
  - CV defined as intra-year standard deviation divided by intra-year mean, based on monthly observations.
  - Log transformation of dependent variables to reduce influence of outliers.
- Key explanatory controls:
  - Real GDP growth rate (annual).
  - Share of oil in total exports.
  - Oil price growth rate (contemporaneous and lagged).
  - Time dummies for years with sharp oil price fluctuations: 1973, 1974, 1979, 1986, 1998, 2000.
  - Broad money to GDP ratio as indicator of financial depth / proxy for exchange rate policy environment.
- Sample and data periods:
  - Oil-fund countries (9): Bahrain, Chile, Kuwait, Mexico, Norway, Oman, Sudan, Trinidad and Tobago, Venezuela.
  - Non-oil-fund countries (6): Egypt, Indonesia, Nigeria, Saudi Arabia, UAE, UK.
  - Annual information on broad money and CPI with monthly observations used for 1973-2003 (Chile missing monthly broad money for 1977 and 1978).
  - REER data available from 1980 onwards.

### IV. Summary statistics (selected)
- Table 1 highlights:
  - Mean CV of broad money, CPI and REER are significantly lower when an oil fund is in place than when it is not (exact mean values presented in the source table).
  - Share of oil in total exports: "62 percent" when an oil fund is in place versus "40 percent" when it is not.
  - International oil price growth rate: "18 percent" when an oil fund is in place versus "7 percent" when it is not.
- Sample note:
  - Figure 1 indicates even spread of fund ages; five funds established in the last five years of sample, three existed for more than 15 years.

### V. Empirical results — main findings
- Estimation diagnostics:
  - F-test: reject null that country-specific effects are jointly zero → supports fixed effects.
  - Hausman tests: null that unobserved effects are uncorrelated with regressors could not be rejected → random effects not rejected; authors note random vs fixed estimates are similar given large T, small n.
- Broad money volatility (Table 2):
  - Lagged oil fund dummy coefficients and t-stats:
    - OLS = -0.30 (t = -3.67)
    - Fixed effects = -0.19 (t = -2.06)
    - Random effects = -0.20 (t = -2.23)
  - Authors' interpretation: existence of a fund lowers the CV of broad money by about "20 percent"; result is statistically significant and robust.
  - Oil price growth rate (lagged) shows positive and significant association with broad money CV (example: lagged coefficient 0.17, t = 3.34 in OLS).
- Inflation and price volatility (Table 3):
  - Inflation (growth rate of CPI, log) — lagged oil fund dummy:
    - OLS = -0.17 (t = -4.6)
    - Fixed effects = -0.26 (t = -3.8)
    - Random effects = -0.23 (t = -3.7)
    - Authors' interpretation: inflation is lower by "one-quarter" in countries with an oil fund than ones without it.
  - CPI volatility (CV of CPI, log) — lagged oil fund dummy:
    - OLS = -0.78 (t = -5.9)
    - Fixed effects = -0.83 (t = -6.1)
    - Random effects = -0.82 (t = -6.0)
    - Authors' interpretation: lower CV of CPI by more than "80 percent" when an oil fund is present.
  - Financial depth (broad money/GDP) associated with lower price growth and volatility (example: financial depth coefficient on CPI CV: OLS = -3.95, t = -7.5).
  - Oil price growth rate associated with lower inflation in some specifications (example: OLS oil price growth rate coefficient = -0.87, t = -2.2).
- REER volatility (Table 4):
  - Lagged oil fund dummy coefficients and t-stats:
    - OLS = -0.25 (t = -2.7)
    - Fixed effects = -0.21 (t = -1.7)
    - Random effects = -0.21 (t = -1.6)
  - Authors: estimated ~"20 percent" reduction in CV of REER but effect is borderline statistically significant and not robust to regressors.
  - Robust findings: drop in annual real GDP growth significantly increases REER volatility (highly significant), and year-to-year falling oil prices induce greater intra-year REER volatility.
- Additional tests:
  - Interaction terms between oil fund dummy and oil price measures show no significant mitigating interaction effect.
  - Figure 2 shows intra-year CV of international oil prices has been increasing since the 1970s—ruling out a decline in oil price intra-year volatility as an explanation for downward trends in macro volatility.

### VI. Conclusions and policy implications
- Main conclusion:
  - Evidence of a robust negative association between presence of an oil fund and inflation and the volatility of broad money and prices in oil exporting countries.
  - Effect on REER volatility is weaker and less robust.
- Interpretation cautions:
  - Fixed effects control for time-invariant unobserved heterogeneity, reducing concern that “prudent” countries self-select into funds; nevertheless, time-varying endogeneity remains unresolved due to lack of instruments—results are non-causal associations.
  - Lack of data on oil fund balances prevents analysis of optimal fund size.
  - Paper does not address whether oil funds mitigate Dutch disease effects.
- Policy implications discussed by authors:
  - Oil funds can serve as macroeconomic self-insurance beyond fiscal stabilization, reducing pressure on monetary policy and contributing to lower inflation and monetary aggregate volatility.
  - Financial depth complements oil funds: greater financial development is associated with lower inflation and price volatility.
  - Oil funds should be viewed as one element of a broader policy toolkit (including monetary policy, exchange rate regime choices, and market-based insurance where available).

*Source: _wp0796 - References*

### 1. Summary Statistics...................................................................................................

### 1. Summary Statistics

### Panel regression estimates: broad structure
- 2. Panel Regression Estimates: The Effects of Existence of Oil Funds on Volatility of Broad Money in countries With and Without Oil Funds — page 13

### Panel regression estimates: inflation and price volatility
- 3. Panel Regression Estimates: The Effect of Existence of Oil Funds on Inflation and Volatility of Prices in Countries With and Without Oil Funds — page 14

### Panel regression estimates: real exchange rate volatility
- 4. Panel Regression Estimates: The Effect of Existence of Oil Funds on the Volatility of the Real Exchange Rate in Countries With and Without Oil Funds — page 15

### Figures listed
- Figure 1. Age Profile of Oil Funds — page 11
- Figure 2. Coefficient of Intra-year Variation of International Oil Prices — page 15

_Provided from: _wp0796 - 1. Summary Statistics (PDF) — canonical URL: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0796.pdf_

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

### _wp0796 - References..............................................................................................................

### I. Introduction — purpose and approach
- Research question: What benefits do oil funds bring to oil exporting countries beyond fiscal stabilization?
- Key argument: Benefits of self-insurance provided by oil funds extend beyond the fiscal sector to monetary aggregates and prices, potentially improving overall macroeconomic stability.
- Empirical approach: Tests on a panel dataset of 15 countries (nine with oil funds and six without) spanning three decades; controls for unobserved time-invariant heterogeneity using panel fixed effects and random effects estimators.
- Data constraints noted: absence of information on oil fund balances and lack of monthly series for GDP, non-oil GDP, and consumption—restricting dependent variables to inflation and volatility of prices, broad money, and the real effective exchange rate (REER).

### II. Rationale for oil funds — channels and literature
- Two stated rationales for oil funds:
  - Short-term self-insurance against international oil price volatility.
  - Long-term savings/transfer between generations (future generations fund), which can still provide short-term stability.
- Two channels through which oil funds can provide macro stability:
  - Smoothing fiscal expenditures (fiscal cushion).
  - Reducing burden on monetary policy (sterilization substituted by fiscal-side operations).
- Literature summary and critiques:
  - Existing empirical work focuses narrowly on fiscal outcomes (e.g., Davis, Ossowski, and Fedelino, 2003; Davis and others, 2001) and finds ambivalent or limited fiscal benefits.
  - Criticisms: persistence in global oil prices may render self-insurance suboptimal; market-based contingent instruments may be preferable when markets are deep (cited: Davis and others, 2001; Devlin and Titman, 2004; Daniel, 2003).
  - Institutional view: oil funds can act as binding fiscal institutions (self-control) and may have real effects on outcomes (cited: Poterba, 1996; Alesina and others, 1999).

### III. Empirical specification and data
- Econometric identification:
  - Key parameter δ captures effect of presence of an oil fund on volatility; hypothesized δ < 0.
  - Primary estimators: OLS with robust standard errors, panel fixed effects (demeaning), and random effects (one-way error components).
  - Rationale for fixed effects: control for time-invariant, unobserved heterogeneity that could correlate with oil fund adoption.
  - Endogeneity caveat: presence of oil fund may be endogenous to time-varying factors; authors acknowledge inability to find adequate instruments and thus present results as robust non-causal associations (citing Acemoglu, 2005).
  - Practical mitigation: use of lagged oil fund dummy where possible.
- Dependent variables (due to data limits):
  - Inflation (growth rate of CPI) and coefficient of intra-year variation (CV) of: CPI, broad money, REER (CV defined as intra-year standard deviation divided by intra-year mean, based on monthly observations).
  - Log transformation of dependent variables to reduce influence of outliers.
- Key explanatory controls:
  - Real GDP growth rate (annual).
  - Share of oil in total exports.
  - Oil price growth rate (contemporaneous and lagged).
  - Time dummies for years with sharp oil price fluctuations (1973, 1974, 1979, 1986, 1998, 2000).
  - Broad money to GDP ratio as an indicator of financial depth / proxy for exchange rate policy environment.
- Sample composition and data periods:
  - Oil-fund countries (9): Bahrain, Chile, Kuwait, Mexico, Norway, Oman, Sudan, Trinidad and Tobago, Venezuela.
  - Non-oil-fund countries (6): Egypt, Indonesia, Nigeria, Saudi Arabia, UAE, UK.
  - Annual information on broad money and CPI with monthly observations used for 1973-2003 (exception: Chile missing monthly broad money for 1977 and 1978).
  - REER data available from 1980 onwards.

### IV. Summary statistics (selected)
- Table 1 (disaggregated by oil fund status) highlights:
  - Mean coefficient of variation (CV) of broad money, CPI and REER are significantly lower when an oil fund is in place than when it is not (exact mean values presented in the source table).
  - Share of oil in total exports: "62 percent" when an oil fund is in place versus "40 percent" when it is not.
  - International oil price growth rate: "18 percent" when an oil fund is in place versus "7 percent" when it is not.
- Sample note: Figure 1 indicates even spread of fund ages; five funds established in the last five years of sample, three existed for more than 15 years.

### V. Empirical results — main findings
- Estimation diagnostics:
  - F-test: reject null that country-specific effects are jointly zero → supports fixed effects.
  - Hausman tests: null that unobserved effects are uncorrelated with regressors could not be rejected → random effects not rejected; but authors note random vs fixed estimates are similar given large T, small n.
- Broad money volatility (Table 2):
  - Lagged oil fund dummy coefficient: OLS = -0.30 (t = -3.67); Fixed effects = -0.19 (t = -2.06); Random effects = -0.20 (t = -2.23).
  - Interpretation reported by authors: existence of a fund lowers the CV of broad money by about "20 percent"; result is statistically significant and robust.
  - Oil price growth rate (lagged) shows positive and significant association with broad money CV (e.g., lagged coefficient 0.17, t = 3.34 in OLS).
- Inflation and price volatility (Table 3):
  - Inflation (growth rate of CPI, log):
    - Lagged oil fund dummy: OLS = -0.17 (t = -4.6); Fixed effects = -0.26 (t = -3.8); Random effects = -0.23 (t = -3.7).
    - Authors' interpretation: inflation is lower by "one-quarter" in countries with an oil fund than ones without it.
  - CPI volatility (CV of CPI, log):
    - Lagged oil fund dummy: OLS = -0.78 (t = -5.9); Fixed effects = -0.83 (t = -6.1); Random effects = -0.82 (t = -6.0).
    - Authors' interpretation: lower CV of CPI by more than "80 percent" when an oil fund is present.
  - Financial depth (broad money/GDP) associated with lower price growth and volatility (e.g., financial depth coefficient on CPI CV: OLS = -3.95, t = -7.5).
  - Oil price growth rate associated with lower inflation in some specifications (e.g., OLS oil price growth rate coefficient = -0.87, t = -2.2).
- REER volatility (Table 4):
  - Lagged oil fund dummy: OLS = -0.25 (t = -2.7); Fixed effects = -0.21 (t = -1.7); Random effects = -0.21 (t = -1.6).
  - Authors: estimated ~"20 percent" reduction in CV of REER but effect is borderline statistically significant and not robust to regressors.
  - Robust findings: drop in annual real GDP growth significantly increases REER volatility (highly significant), and year-to-year falling oil prices induce greater intra-year REER volatility.
- Additional tests:
  - Interaction terms between oil fund dummy and oil price measures show no significant mitigating interaction effect.
  - Figure 2 (oil price intra-year CV) indicates intra-year CV of international oil prices has been increasing since the 1970s—ruling out a decline in oil price intra-year volatility as an explanation for downward trends in macro volatility.

### VI. Conclusions and policy implications
- Main conclusion:
  - Evidence of a robust negative association between presence of an oil fund and inflation and the volatility of broad money and prices in oil exporting countries.
  - Effect on REER volatility is weaker and less robust.
- Interpretation cautions:
  - Results control for time-invariant unobserved heterogeneity (via fixed effects), reducing concern that “prudent” countries self-select into funds; nevertheless, time-varying endogeneity remains unresolved due to lack of instruments—results are therefore non-causal associations.
  - Lack of data on oil fund balances prevents analysis of optimal fund size.
  - Paper does not address whether oil funds mitigate Dutch disease effects.
- Policy implications (as discussed by authors):
  - Oil funds can serve as macroeconomic self-insurance beyond fiscal stabilization, reducing pressure on monetary policy and contributing to lower inflation and monetary aggregate volatility.
  - Financial depth complements oil funds: greater financial development is associated with lower inflation and price volatility.
  - Given limitations, oil funds should be viewed as one element of a broader policy toolkit (including monetary policy, exchange rate regime choices, and market-based insurance where available).

*Italic source: _wp0796 - References..............................................................................................................*

### References

### References

### Fiscal institutions, budget rules, and fiscal performance
- Alesina, A., and R. Perotti, 1999, “Budget Deficits and Budget Institutions,” in Fiscal Institutions and Fiscal Performance, ed. By J. Poterba and J. von Hagen (Chicago: University of Chicago Press).
- Alesina, A., R. Hausmann, R. Hommes, and E. Stein, 1999, “Budget Institutions and Fiscal Performance in Latin America,” Journal of Development Economics, Vol. 59, No. 2, pp. 253–73.
- Fabrizio, S., and A. Mody, 2006, “Can Budget Institutions Counteract Political Indiscipline?” IMF Working Paper, 06/123 (Washington: International Monetary Fund).
- Poterba, J.M., 1996, “Do Budget Rules Work?” NBER Working Paper No. 5550 (Cambridge MA: National Bureau of Economic Research).
- Stein, E., E. Talvi, and A. Grisanti, 1999, “Institutional Arrangements and Fiscal Performance: The Latin American Experience,” in Fiscal Institutions and Fiscal Performance, ed. By J. Poterba and J. von Hagen (Chicago: University of Chicago Press).
- von Hagen, J., and I. Harden, 1995, “Budget Processes and Commitment to Fiscal Discipline,” European Economic Review, Vol. 39 (April), pp.771-9.
- Perotti, R., and Y. Kontopoulos, 2002 “Fragmented Fiscal Policy,” Journal of Public Economics, Vol. 86, pp. 191-222.

### Oil revenue management, stabilization and savings funds, and commodity risk
- Daniel, J.A, 2003, “Hedging Government Oil Price Risk,” in “Stabilization and Savings Funds for Nonrenewable Resources,” ed. By J.M. Davis, R. Ossowski, and A. Fedelino, IMF Occasional Paper No. 205 (Washington: International Monetary Fund).
- Davis, J.M., R. Ossowski, and A. Fedelino, eds., 2003, Fiscal Policy Formulation and Implementation in Oil-Producing Countries, (Washington: International Monetary Fund).
- Davis, J.M., R. Ossowski, James Daniel, and Steven Barnett (2001), “Stabilization and Savings Funds for Nonrenewable Resources,” IMF Occasional Paper No. 205, (Washington: International Monetary Fund).
- Devlin, J., and S. Titman, 2004, “Managing Oil Price Risk in Developing Countries,” World Bank Research Observer, Vol. 19, No. 1, pp. 119-39.
- Fasano, U., 2000, “Review of the Experience with Oil Stabilization and Savings Funds in Selected Countries,” IMF Working Paper, 00/112. (Washington: International Monetary Fund.
- Hausmann, R., A. Powell, and R. Rigobon, 1993 “An Optimal Spending Rule Facing Oil Income Uncertainty (Venezuela),” in External Shocks and Stabilization Mechanisms, ed. By E. Engel and P. Meller (Washington: Inter-American Development Bank).
- Wakeman-Linn, J., P. Mathieu, and B. van Selm, 2003, “Oil Funds in Transition Economies: Azerbaijan and Kazakhstan,” in “Stabilization and Savings Funds for Nonrenewable Resources,” ed. by J.M. Davis, R. Ossowski, and A. Fedelino, IMF Occasional Paper No. 205 (Washington: International Monetary Fund).
- Basch, M, and E. Engel, 1993, “Temporary Shocks and Stabilization Mechanisms: The Chilean Case,” in External Shocks and Stabilization Mechanisms, ed by E. Engel and P. Meller (Washington: Inter-American Development Bank).

### Exchange rate regimes, reserves, and volatility
- Devereux, M.B., and P.R. Lane, 2003, “Understanding Bilateral Exchange Rate Volatility,” Journal of International Economics, Vol. 50 , pp. 109–32.
- Hviding, K., M. Nowak, and L. A. Ricci, 2004, “Can Higher Reserves Help Reduce Exchange Rate Volatility,” IMF Working Paper 04/189 (Washington: International Monetary Fund).
- Reinhart, C.M., and K.S. Rogoff 2002 "The Modern History of Exchange Rate Arrangements: A Reinterpretation," NBER Working Papers No. 8963, (Cambridge MA: National Bureau of Economic Research).

### Social policy, labor, and individual behavior under insecurity
- Ehrlich, I., and G. Becker, 1972, “Market Insurance, Self-Insurance and Self-Protection.” Journal of Political Economy Vol. 80, pp. 623-48.
- Freeman, R., 1984. “Longitudinal Analyses of the Effects of Trade Unions.” Journal of Labor Economics Vol. 2, pp. 1-26.
- Gill, I.S., and N. Ilahi 2000. “Economic Insecurity, Individual Behavior and Social Policy,” Paper prepared for the Regional Study on Managing Economic Insecurity in Latin America and the Caribbean, World Bank.

### Methodology, empirical strategies, and econometrics
- Angrist, J.D, and A. B. Krueger, 1998, “Empirical Strategies in Labor Economics," Working Paper 401, Industrial Relations Section, (Princeton, NJ: Princeton University).
- Wooldridge, J.M. 2002, Econometric Analysis of Cross Section and Panel Data. (Cambridge, MA: MIT Press).

### Regional and thematic outlooks
- International Monetary Fund, 2005 Regional Economic Outlook, Middle East and Central Asia, (Washington: International Monetary Fund).
- Serven, L., and A. Solimano, 1989, “Private Investment and Macroeconomic Adjustment: An Overview,” (World Bank: Policy Research Working Paper WPS 339).

*Source: _wp0796 - References*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2007/_wp0796.pdf_