## 1. Determinants of Non-oil Current Account

## Source details

**Canonical URL:** [1. Determinants of Non-oil Current Account](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09248.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09248.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09248.pdf.json)

---

### Introduction and headline findings
- Purpose: apply the permanent income model (PIM) to the non-oil current accounts of major oil exporters to assess the extent national consumption decisions are based on permanent versus current oil income and wealth.
- Sample: Kuwait, United Arab Emirates, Qatar, Saudi Arabia, Oman, Iran, Libya, Syria, Algeria, Sudan, Yemen, Venezuela, Trinidad and Tobago, Azerbaijan, Kazakhstan, Russian Federation, Angola, Congo, Equatorial Guinea, Gabon, Nigeria.
- Key headline result:
  - For oil-exporting countries outside Africa, around half of the fluctuations in the private sector non-oil balance are driven by considerations of changes in permanent wealth rather than current income.
  - For the public sector and for African countries the return on oil wealth has no effect—only current income matters.

### Theoretical considerations and empirical formulation
- Framework: canonical forward-looking consumption models and the PIM; countries treated as consumers deciding whether to save or spend a path of income.
- Implications under PIM:
  - temporary changes in income → propensity to consume ≈ real interest rate;
  - permanent changes in income → propensity to consume = one.
- Two deviations that raise the propensity to consume out of current income:
  - liquidity-constrained consumers who consume all income;
  - consumers who discount the future more than the risk-free real interest rate (real interest rate wedge).
- Empirical proxies:
  - current oil cash flow proxied by the current oil surplus (oil balance);
  - permanent income proxied by the present discounted value of oil reserves (oil wealth).
- Testable hypothesis: the sum of coefficients on current oil income and return on oil wealth should equal one (i.e., oil revenues fully reflected in changes in the non-oil current account), operationalized in sign conventions as H0: c + p + (c + p) = -1.

### Data construction and assumptions
- Oil reserves source: British Petroleum.
- Oil price: average annual oil price for Brent, Dubai and West Texas blends used to convert reserves to U.S. dollars.
- Extraction rate: constant, based on average ratio of production to reserves over the 1980–2007 period.
- Discount rate for present value of oil wealth: real rate of return of 4 percent.
- Domestic consumption of oil proxied by one minus the ratio of exports to production; wealth net of domestic consumption multiplied by the ratio of exports to production (averages over 1980–2007 used, except Angola, Russia, Azerbaijan, Kazakhstan use averages since 2000; Syria uses 2004–07 average).
- Population growth: historical averages over 1980–2007, except Qatar and UAE assumed at 2½ percent per annum.
- National consumption decisions represented by movements in the non-oil current account (movements dominated by imports).
- Additional control: net foreign asset term interacted with domestic absorption to capture the income component of NFA.

### Estimation — Non-oil current account (panel time-series) and representative coefficients
- Sample strategy: majority of regressions exclude Africa because African non-oil balance profile differs since early 1990s; time dummies included.
- Representative coefficients (across columns/specifications as reported):
  - Return on oil wealth (uncorrected): -0.15 (weak and insignificant).
  - Return on oil wealth (netting out domestic consumption): -0.39 ***.
  - Return on oil reserves corrected for population (but not netting domestic consumption): -0.17 and -0.24 (in some specifications).
  - Oil balance (current oil cash flow): -0.74 ***, -0.8 ***, -0.76 ***, -0.81 ***, -1.03 ***, -1.04 ***, -1.02 *** (across specifications).
  - Return on oil reserves × GCC dummy: 0.26 * and 0.19 (Qatar drives positive GCC sensitivity; excluding Qatar removes GCC difference).
  - Net foreign assets interacted with domestic absorption t-1: -0.03 ***, 0.04 ***, 0.05 ***, 0.04 ***, 0.08, 0.08, 0.08 (reported across columns).
  - Output growth t-1: -0.08, -0.07, -0.04, -0.08, 0.14, 0.18, 0.14 (insignificant in many specifications).
  - Change in population: 0.1, -0.16, -0.15, -0.13, 5.8, 5.72, 5.83 (varies by specification).
  - Old-age dependency ratio (initial conditions): 0.5, 0.53, 0.6, 0.16, 12.6, 11.55, 12.2.
  - GDP per capita (relative to U.S.): -0.06 ***, -0.04 ***, -0.04 ***, -0.04 ***, 0.34, 0.31, 0.33.
  - NFA (initial): -0.02 *, -0.03 ***, -0.04 ***, -0.03 ***, -0.39, -0.35, -0.38.
- Diagnostics (examples across regressions):
  - Number of observations: 382, 365, 365, 365, 106, 106, 106.
  - Number of countries: 15, 14, 14, 14, 4, 4, 4.
  - R squared: 0.71, 0.75, 0.75, 0.75, 0.84, 0.84, 0.84.
- Interpretation:
  - When domestic consumption is netted out of the wealth term and population growth is corrected for, return on oil wealth becomes significant (around -0.39 ***) and the sum of oil-related coefficients aligns with consumption-smoothing (sum near unity with sign convention).
  - For African countries (excluding Equatorial Guinea in some results), only the current oil balance matters empirically (coefficient empirically equals unity), consistent with severe liquidity constraints; return on oil wealth insignificantly different from zero.
  - Qatar strongly influences GCC sensitivity; excluding Qatar removes GCC-specific effect.
  - Correcting oil-wealth returns for population growth is crucial; without it sums of coefficients are significantly less than unity.

### Key empirical implications (private vs public sectors)
- Private sector (non-oil current account) behavior:
  - Outside Sub-Saharan Africa, around half of fluctuations in the private non-oil balance are linked to changes in permanent oil wealth when wealth measures net out domestic consumption and account for population growth.
  - For African oil exporters in the sample, current oil cash flow drives non-oil current account movements; evidence consistent with liquidity constraints.
- Public sector behavior:
  - Return on oil wealth has no detectable effect on the public sector non-oil fiscal balance in many specifications; current income dominates.
- Policy-relevant points:
  - Adjusting wealth measures for domestic consumption and population growth is crucial for testing PIM-like consumption smoothing in oil exporters.
  - Country heterogeneity matters—aggregating across regions masks important differences (e.g., Qatar, African oil exporters).
  - Where return on oil wealth matters, oil windfalls are at least partly treated as permanent, supporting policies that smooth consumption over time; where current income dominates, policies should recognize limited private/public smoothing capacity.

---

### Section II — Joint modeling (SUR) and expanded regression findings

### Modeling approach and scope
- Technique: seemingly unrelated regressions (SUR) jointly model:
  - oil trade balance → non-oil fiscal balance; and
  - oil trade balance → non-oil current account (aggregate).
- Purpose of the oil trade balance → non-oil fiscal balance regression:
  - captures how much export revenue is transferred into taxation receipts;
  - captures how much of taxation receipts are consumed.
- Non-oil fiscal deficits are stationary (supported by regional non-oil fiscal deficit averages), supporting OLS estimation.
- Regional pattern: differences between African and other Middle East countries in the non-oil fiscal deficit are much smaller than for the non-oil current account.

### Key regression findings — Non-oil fiscal balance
- Return on oil wealth: largely unrelated to the non-oil fiscal balance in all regions; preferred specification (excludes Africa and adjusts return on oil wealth for domestic consumption and population changes) yields an incorrectly signed coefficient.
- Example coefficient: return on reserves (excludes Africa, nets domestic consumption but no population adjustment): -0.11 (weakly significant).
- Oil balance coefficient: stable across unrestricted specifications at -0.25, implying "a 1 percent of GDP increase [in the oil balance] would lead to an increase of -0.25 percent in the non-oil fiscal deficit."
- Imposing nonlinear constraint (weighted sum of oil variables in non-oil equation sum to minus unity):
  - coefficient on oil balance in non-oil fiscal balance rises to -0.35 for countries outside Africa;
  - rises to -0.41 for African countries.
- Private vs public forward-looking behavior:
  - Wealth term in non-oil current account regression: -0.39 (specification adjusting for domestic consumption and population changes).
  - Wealth term in non-oil fiscal balance regression: 0.
  - For African countries: wealth terms have no significant bearing on non-oil balances in either regression.
- Other covariate effects on non-oil fiscal balance:
  - Change in population: negative effect (example: -0.62 ***).
  - Net foreign assets t-1: negative effect (examples: -0.02 **, -0.04 ***).
  - Output growth: stronger output growth improves fiscal balance (examples: 0.17 **, 0.14 *).
  - GDP per capita (relative to U.S.): negative association with fiscal deficits (example: -0.03 ***).
  - Initial NFA: countries with high initial asset stocks have larger fiscal deficits (NFA entries: -0.1 ***, -0.11 ***, -0.13 ***).

### Public and private sectors combined — Non-oil current account (SUR results)
- Introducing non-oil fiscal balance into determinants of non-oil current account allows nonlinear test of oil-related flows.
- Estimated effect of non-oil fiscal balance on non-oil current account (excluding Africa): coefficient = 0.15 **; a 1 percentage point improvement in the non-oil budget balance improves the non-oil current account by 0.15 percent.
- Because the non-oil budget balance is affected by return on wealth and oil balance, the test assesses whether the weighted sum of oil-related influences (private and public channels) on the non-oil current account sums to minus unity.
- Test outcomes:
  - Null that oil terms sum to unity is accepted for specifications allowing a population correction to the wealth term.
  - Under that constraint, return on wealth coefficient rises to -0.42 (versus -0.2 to -0.26 in unrestricted regressions).
  - Without population-growth correction: oil returns are underspent, with the sum of oil-wealth coefficients estimated to be considerably less than unity (reported as -0.8+0.03+0.12*(-0.11–0.22) in the text).
  - For African countries: only the oil balance matters; its coefficient is insignificantly different from unity.
- Representative coefficients from combined regressions:
  - Oil balance: -0.77 ***, -0.76 ***, -0.46 ***, -0.8 ***, -1.27 ***, -1.07 *** (varies by specification).
  - Return on oil reserves (corrected for population): -0.2 ** and -0.26 ** (examples).
  - Net foreign assets interacted with domestic absorption t-1: 0.07 *** (example).
- Diagnostics (examples): Number of observations: 278; Number of countries: 14; R squared: 0.61, 0.51, 0.90 (varies by specification).

### Conclusions and policy-relevant implications (Section II synthesis)
- Intertemporal considerations matter:
  - For oil-exporting countries outside Africa, changes in the return on oil wealth have significant explanatory power for the non-oil current account.
  - When correcting return on wealth for population changes and imposing the sum-to-minus-unity constraint, return on wealth = -0.42 and oil balance = -0.46, implying private-sector response: "slightly less than half" of private non-oil balance fluctuations are driven by permanent-wealth changes rather than current income.
  - The -0.42 return-on-wealth is comparable to estimates for Canadian consumers in Bayoumi and Klein (1997) and within the range of Asdrubali and Kim (2007).
- Contrast with African countries:
  - For African countries, movements in the oil balance have a one-to-one impact on the non-oil current account; return on oil wealth has no effect—only current income matters.
- Fiscal behavior and public-sector myopia:
  - Oil trade balance has significant negative effects on the non-oil fiscal balance with fairly stable coefficients between -0.25 and -0.41.
  - Effects of changes in oil wealth on the non-oil fiscal deficit are largely absent.
  - Implication: outside sub-Saharan Africa, private sector is considerably more forward looking than the public sector.
- Policy takeaway:
  - Analyses of macroeconomic aggregates for major oil-exporting countries should account for intertemporal decisions (permanent vs. current income) to accurately capture private-sector and public-sector responses to oil rents.
  - Where return on oil wealth matters, policies that smooth consumption over time are supported; where current income dominates, policies should recognize limited smoothing capacity and potential liquidity constraints.

*Source — _wp09248 - 1. Determinants of Non-oil Current Account; Section II modeling results (IMF PDF chapter)*

### 1. Determinants of Non-oil Current Account ..........................................................................10

### 1. Determinants of Non-oil Current Account

### Introduction
- Recent sharp swings in the oil price raise the question whether consumption decisions in oil-exporting countries are based on temporary or permanent movements in oil income and wealth.
- The paper applies the permanent income model (PIM) to the non-oil current accounts of major oil exporters to assess the extent to which national consumption decisions are based on permanent versus current income.
- Sample composition: a mix of GCC countries and Sub-Saharan African oil exporters; country list includes Kuwait, United Arab Emirates, Qatar, Saudi Arabia, Oman, Iran, Libya, Syria, Algeria, Sudan, Yemen, Venezuela, Trinidad and Tobago, Azerbaijan, Kazakhstan, Russian Federation, Angola, Congo, Equatorial Guinea, Gabon, Nigeria.
- Key headline result: for oil-exporting countries outside Africa, around half of the fluctuations in the private sector non-oil balance are driven by considerations of changes in permanent wealth rather than current income. By contrast, for the public sector and for African countries the return on oil wealth has no effect—only current income matters.

### Theoretical considerations
- Framework: Treat countries as consumers deciding whether to save or spend a path of income; use canonical forward-looking consumption models and the PIM.
- Implication of PIM: temporary changes in income should have very little impact on consumption (propensity to consume ≈ real interest rate), while permanent changes have a propensity to consume equal to one.
- Two deviations from canonical model to explain observed higher propensity to consume out of current income:
  - Liquidity-constrained consumers who consume all income (Campbell and Mankiw, 1990).
  - Consumers who discount the future at a higher rate than the risk-free real interest rate due to a real interest rate wedge (modeled as e.g., probability of death).
- Empirical formulation for oil exporters uses observable proxies:
  - Current oil cash flow proxied by the current oil surplus (oil balance).
  - Permanent income proxied by present discounted value of oil reserves (oil wealth).
- Testable hypothesis: the sum of coefficients on current oil income and return on oil wealth should equal one (i.e., oil revenues fully reflected in changes in the non-oil current account).
- Specification for public and private sectors:
  - Non-oil fiscal balance equation includes oil balance and oil wealth terms.
  - Non-oil current account (private sector) equation includes oil balance, oil wealth, and the non-oil fiscal balance.
- Null hypothesis tested: H0: c + p + (c + p) = -1 i.e., the oil-related coefficients sum to minus unity when sign conventions are as in the paper.

### Data construction and assumptions
- Oil reserves data source: British Petroleum.
- Oil price: average annual oil price for Brent, Dubai and West Texas blends used to convert reserves to U.S. dollars.
- Extraction rate assumption: constant, based on average ratio of production to reserves over the 1980–2007 period.
- Discount rate for present value of oil wealth: real rate of return of 4 percent.
- Domestic consumption of oil proxied by one minus the ratio of exports to production; wealth net of domestic consumption multiplied by the ratio of exports to production (averages over 1980–2007 used, with exceptions: Angola, Russia, Azerbaijan, Kazakhstan use averages since 2000; Syria uses 2004–07 average).
- Population growth: long-run population growth estimates obtained as historical averages over 1980–2007, except:
  - Qatar and UAE: population growth rates assumed to be 2½ percent per annum.
- National consumption decisions represented by movements in the non-oil current account (non-oil export sector very small; movements dominated by imports).
- Additional control: net foreign asset term interacted with domestic absorption to capture the income component of NFA.

### Estimation — Non-oil current account (panel time-series)
- Sample split: majority of regressions exclude Africa because African non-oil balance profile differs since early 1990s; time dummies included.
- Key empirical findings (selected coefficients and diagnostics as reported):
  - Return on oil wealth (uncorrected for domestic consumption; corrected for population in some specifications):
    - coefficient reported as -0.15 (weak and insignificant) in an initial regression (no correction for domestic consumption).
    - coefficient increases to around -0.39 *** when netting out domestic consumption from the wealth term (column 3).
    - when return on oil reserves is corrected for population but domestic consumption is not netted out, the coefficient reported is -0.17 and -0.24 in some specifications.
  - Oil balance (current oil cash flow) coefficients:
    - -0.74 ***; -0.8 ***; -0.76 ***; -0.81 ***; -1.03 ***; -1.04 ***; -1.02 *** (across various columns/specifications).
  - Interaction and regional effects:
    - Return on oil reserves interacted with GCC dummy: 0.26 * and 0.19 (Qatar drives positive GCC sensitivity; excluding Qatar removes GCC difference).
  - Other variables:
    - Net foreign assets interacted with domestic absorption t-1: -0.03 ***, 0.04 ***, 0.05 ***, 0.04 ***, 0.08, 0.08, 0.08 (reported across columns).
    - Output growth t-1: -0.08, -0.07, -0.04, -0.08, 0.14, 0.18, 0.14 (insignificant in many specifications).
    - Change in population: 0.1, -0.16, -0.15, -0.13, 5.8, 5.72, 5.83 (noted but interpretation depends on specification).
  - Initial conditions:
    - Old-age dependency ratio: 0.5, 0.53, 0.6, 0.16, 12.6, 11.55, 12.2.
    - GDP per capita (relative to U.S.): -0.06 ***, -0.04 ***, -0.04 ***, -0.04 ***, 0.34, 0.31, 0.33.
    - NFA: -0.02 *, -0.03 ***, -0.04 ***, -0.03 ***, -0.39, -0.35, -0.38.
  - Diagnostics (reported across columns):
    - Number of observations: 382, 365, 365, 365, 106, 106, 106.
    - Number of countries: 15, 14, 14, 14, 4, 4, 4.
    - R squared: 0.71, 0.75, 0.75, 0.75, 0.84, 0.84, 0.84.
    - Note: significance notation in table — *** signifies significance at the 99 percent level; ** signifies significance at the 95 percent level; * signifies significance at the 90 percent level.
- Interpretation of regression results:
  - When domestic consumption is netted out of the wealth term and population growth is corrected for, the return on oil wealth becomes a significant determinant of the non-oil balance (coefficient around -0.39 ***) and the sum of the return on oil wealth and oil balance is consistent with the consumption-smoothing hypothesis (sum near unity with appropriate sign convention).
  - For African countries (excluding Equatorial Guinea in some results), the only variable that matters is the current oil balance which empirically has a unit coefficient — consistent with severe liquidity constraints; the return on oil wealth coefficient is insignificantly different from zero.
  - Qatar strongly influences GCC sensitivity to return on oil wealth; excluding Qatar removes the apparent GCC-specific effect.
  - Correcting oil-wealth returns for population growth supports the consumption-smoothing interpretation; failing to correct for population growth leads to sums of coefficients significantly less than unity.

### Key empirical implications
- Private sector (non-oil current account) behavior:
  - Outside Sub-Saharan Africa, around half of fluctuations in the private non-oil balance are linked to changes in permanent oil wealth (when wealth measures net out domestic consumption and account for population growth).
  - For African oil exporters in the sample, current oil cash flow drives non-oil current account movements; evidence consistent with liquidity constraints.
- Public sector behavior:
  - The return on oil wealth has no detectable effect on the public sector non-oil fiscal balance in many specifications; current income dominates.
- Policy-relevant points:
  - Adjusting wealth measures for domestic consumption and population growth is crucial for testing PIM-like consumption smoothing in oil exporters.
  - Country heterogeneity matters — aggregating across regions masks important differences (e.g., Qatar, African oil exporters).
  - Where return on oil wealth matters, oil windfalls are at least partly treated as permanent, supporting policies that smooth consumption over time; where current income dominates, policies should recognize limited private/public smoothing capacity.

*Italic: Source — _wp09248 - 1. Determinants of Non-oil Current Account (IMF PDF chapter)*

### Section II of the paper has emphasized the importance of modeling the relationship between

### _wp09248 - Section II of the paper has emphasized the importance of modeling the relationship between

### Modeling approach and scope
- The analysis models two regression relationships jointly using the seemingly unrelated regressions (SUR) technique:  
  - the oil trade balance on the non-oil fiscal balance; and  
  - the oil trade balance on the non-oil current account (aggregate relationship).  
- Purpose of the oil trade balance → non-oil fiscal balance regression:  
  - captures how much export revenue gets transferred into taxation receipts; and  
  - captures how much of the taxation receipts are consumed.  
- Non-oil fiscal deficits are stationary (supported by regional non-oil fiscal deficit averages), supporting OLS estimation.
- Notable regional pattern: differences between African and other Middle East countries in the non-oil fiscal deficit are much smaller than for the non-oil current account.

### Key regression findings — Non-oil fiscal balance
- Return on oil wealth: largely unrelated to the non-oil fiscal balance in all regions, suggesting fairly myopic fiscal behavior.
  - Preferred specification (excludes Africa and adjusts return on oil wealth for domestic consumption and population changes): coefficient incorrectly signed.
  - Specification that excludes Africa and nets out domestic consumption but does not adjust for population growth: coefficient on return on reserves is -0.11 (weakly significant).
- Oil balance coefficient: stable across unrestricted specifications at -0.25, implying:
  - "a 1 percent of GDP increase [in the oil balance] would lead to an increase of -0.25 percent in the non-oil fiscal deficit."
- When imposing the nonlinear constraint that the weighted sum of the oil variables in the non-oil equation sum to minus unity:
  - coefficient on the oil balance in the non-oil fiscal balance equation rises to -0.35 for countries outside of Africa; and
  - rises to -0.41 for African countries.
- Private vs. public forward-looking behavior:
  - Coefficient on the wealth term in the non-oil current account regression: -0.39 (specification that adjusts return on oil wealth for domestic consumption and population changes).
  - Coefficient on the wealth term in the non-oil fiscal balance regression: 0.
  - For African countries: wealth terms have no significant bearing on non-oil balances in either regression.
- Other covariate effects on non-oil fiscal balance:
  - Change in population: negative effect (examples from table: -0.62 *** in several specifications).
  - Net foreign assets t-1: negative effect (examples: -0.02 **, -0.04 ***).
  - Output growth: stronger output growth improves fiscal balance (example coefficient entries: 0.17 **, 0.14 *).
  - GDP per capita (relative to U.S.): negative association with fiscal deficits (example: -0.03 ***).
  - Initial NFA: countries with high initial asset stocks have larger fiscal deficits (NFA entries: -0.1 ***,-0.11 ***,-0.13 ***).

### Public and private sectors combined — Non-oil current account
- Introducing non-oil fiscal balance into determinants of non-oil current account allows nonlinear test of oil-related flows.
- Estimated effect of non-oil fiscal balance on non-oil current account (excluding Africa):
  - coefficient = 0.15 **, i.e., a 1 percentage point improvement in the non-oil budget balance improves the non-oil current account by 0.15 percent.
  - Interpretation: some oil revenue is spent on domestic goods.
- Interaction and identification:
  - Because the non-oil budget balance is affected by return on wealth and oil balance, the test assesses whether the weighted sum of oil-related influences (private and public sector channels) on the non-oil current account sums to minus unity.
- Test outcomes:
  - Null hypothesis that oil terms sum to unity is accepted for specifications allowing a population correction to the wealth term.
  - When that constraint is imposed, coefficient on the return on wealth rises to -0.42 (versus -0.2 to -0.26 in unrestricted regressions).
  - Without population-growth correction: oil returns are underspent, with the sum of oil-wealth coefficients estimated to be considerably less than unity (reported as -0.8+0.03+0.12*(-0.11–0.22) in the text).
  - For African countries: only the oil balance matters; its coefficient is insignificantly different from unity.
- Representative coefficients from combined regressions:
  - Oil balance: examples include -0.77 ***, -0.76 ***, -0.46 ***, -0.8 ***, -1.27 ***, -1.07 *** (varies by specification and sample).
  - Return on oil reserves (corrected for population): examples include -0.2 ** and -0.26 **.
  - Net foreign assets interacted with domestic absorption t-1: 0.07 *** (example).
  - Diagnostics: sample sizes and fit examples — Number of observations: 278, Number of countries: 14, R squared examples: 0.61, 0.51, 0.90 (varies by specification).

### Conclusions and policy-relevant implications
- Intertemporal considerations matter:
  - For oil-exporting countries outside Africa, changes in the return on oil wealth have significant explanatory power for the non-oil current account.
  - When correcting return on wealth for population changes and imposing the sum-to-minus-unity constraint, the return on wealth coefficient is -0.42 and the oil balance coefficient is -0.46, implying private-sector response: "slightly less than half" of private non-oil balance fluctuations are driven by permanent-wealth changes rather than current income.
  - This -0.42 return-on-wealth is comparable to estimates for Canadian consumers in Bayoumi and Klein (1997) and within the range of Asdrubali and Kim (2007) results.
- Contrast with African countries:
  - For African countries, movements in the oil balance have a one-to-one impact on the non-oil current account; return on oil wealth has no effect—only current income matters.
- Fiscal behavior and public sector myopia:
  - Oil trade balance has significant negative effects on the non-oil fiscal balance with fairly stable coefficients across countries between -0.25 and -0.41.
  - Effects of changes in oil wealth on the non-oil fiscal deficit are largely absent.
  - Implication: outside sub-Saharan Africa, private sector is considerably more forward looking than the public sector.
- Policy takeaway:
  - Analyses of macroeconomic aggregates for major oil-exporting countries should account for intertemporal decisions (permanent vs. current income) to accurately capture private-sector and public-sector responses to oil rents.

*Source: _wp09248 - Section II of the paper has emphasized the importance of modeling the relationship between*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2009/_wp09248.pdf_