## _wp0408

## Source details

**Canonical URL:** [_wp0408](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0408.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0408.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0408.pdf.json)

---

### I. Introduction: fiscal policy surveillance in the CEMAC zone
- CEMAC members: Cameroon, the Central African Republic, Chad, the Republic of Congo, Equatorial Guinea, Gabon.
- Past exchange-rate arrangements:
  - Fixed CFA franc / French franc peg until 1999; peg to the euro from 1999.
  - French treasury guarantees the currency peg in principle; BEAC and the CEMAC’s Executive Secretariat pursue policies to sustain the peg.
- Post-2002 surveillance objectives (evaluated annually by the CEMAC’s Executive Secretariat):
  - A nonnegative basic fiscal balance (overall balance excluding grants and foreign financed investment).
  - Domestic and foreign debt ≤ 70 percent of GDP.
  - Non-accumulation of domestic and/or external arrears.
  - Annual inflation rate ≤ 3 percent.
- Two issues for oil-exporting members (Cameroon, Republic of Congo, Equatorial Guinea, Gabon, and since end-2003 Chad):
  - Oil receipt volatility undermines the headline fiscal balance as an indicator of underlying fiscal stance.
  - Finite oil reserves raise intertemporal efficiency and equity questions; some current oil receipts may need to be saved.

### Key empirical context and indicators (selected figures preserved)
- Oil GDP as percent of total nominal GDP (1998–2002):
  - Cameroon: 7.2, 10.3, 14.9, 13.2, 10.9
  - Republic of Congo: 36.5, 53.4, 65.5, 56.7, 53.6
  - Equatorial Guinea: 75.3, 74.2, 85.3, 88.0, 89.9
  - Gabon: 28.1, 38.5, 48.7, 42.1, 41.8
- Oil receipts as percent of fiscal revenue (1998–2002):
  - Cameroon: 20.0, 23.7, 31.5, 29.7, 25.2
  - Republic of Congo: 53.1, 71.7, 76.6, 68.2, 69.1
  - Equatorial Guinea: 61.2, 76.9, 83.8, 82.2, 91.7
  - Gabon: 54.6, 45.3, 67.5, 64.1, 55.8
- CEMAC basic fiscal balances, 1994–2002 (percent of GDP):
  - Cameroon: -5.3, -2.1, -0.9, -0.2, -0.6, 0.6, 2.7, 2.6, 2.1
  - Republic of Congo: -12.4, -7.1, -6.9, -5.9, -15.9, -0.6, 7.2, 8.9, -0.8
  - Equatorial Guinea: -2.1, -5.1, -5.3, 3.0, -0.5, 3.7, 9.0, 14.6, 14.6
  - Gabon: 1.8, 6.5, 6.2, 11.1, -1.3, 4.3, 13.9, 7.7, 7.0
  - WEO oil price (U.S. dollars per barrel): 15.9, 17.2, 20.4, 19.3, 13.1, 18.0, 28.2, 24.3, 25.0
- Observed pattern examples:
  - 2000 (record oil prices): no country recorded a negative basic balance.
  - 1998 (low petroleum prices): all countries recorded negative basic balances.

### II. Analytical framework: formalization and empirical testing
- Surveillance target (general form):
  - x ≡ R − E, with the constraint B_t ≡ R_t − E_t ≥ x_t.
  - CEMAC practice: x = zero; B is the basic balance (revenue excluding grants minus expenditure excluding foreign-financed investment).
- Revenue decomposition and oil revenue model:
  - R_t = N_t + O_t.
  - O_t = φ_t O_t (oil revenue equals nominal oil GDP times collection ratio φ).
- Non-oil based surveillance target:
  - y ≡ R_N − E, with the constraint B_N,t ≡ R_{N,t} − E_t ≥ y_t.
  - Two options for y:
    - (a) annual flow of oil receipts using a medium-term oil price projection (Section III);
    - (b) an annuity from the net present value of future fiscal oil receipts accounting for finite reserves and production path (Section IV).
- Conditions and empirical testing:
  - Non-oil balance usable if non-oil balance does not vary systematically with oil receipts for exogenous reasons.
  - Non-oil revenue: R_{N,t} = γ_t N_t, where γ is non-oil revenue collection ratio.
  - Relationship of non-oil GDP to oil GDP modeled (first-difference transformed for estimation):
    - ∆N_it = α_1 + θ_0 ∆O_it + θ_1 ∆O_{i,t−1} + u_it.
  - Estimation sample and method:
    - Panel: Cameroon, Gabon, Republic of Congo, Equatorial Guinea.
    - Period: 1994–2001 (2002 excluded).
    - Random-effects estimator preferred by Hausman test.
- Empirical findings (Table 3, preferred specification: First differences, Random Effects, One Lag):
  - Estimated coefficients:
    - θ_0 = -.051 (p-value .402)
    - θ_1 = .181 (p-value .007)
  - Test on ∑θ = 0: statistic 1.71 (p-value .190).
  - R-squared = .118; Observations = 24.
  - Interpretation:
    - Positive and significant θ_1 suggests a positive lagged trickle-down effect of oil GDP on non-oil GDP.
    - Joint effect (θ_0 + θ_1) significant only at the 20 percent level; adding another lag removes significance of θ_1. Treat joint effect as an upper band, not a dependable point estimate.
- Effect on non-oil revenue collection ratio γ (equation 6 decomposition):
  - γ_t = γ̃_t + γ^O_t, where γ̃ is collection ratio on “pure” non-oil GDP and γ^O on oil-induced non-oil GDP.
  - Point estimates (unweighted means across countries, first differences, OLS):
    - γ̃ = 0.233 (standard error: 0.042)
    - γ^O = 0.205 (standard error: 0.111)
  - An F-test fails to reject equality of γ̃ and γ^O at conventional significance levels — no mechanical effect of oil GDP on non-oil revenue collection ratio found.
  - Robustness caveat: γ^O not very robust to sampling period; small sample size and low statistical power.
- Quantitative magnitude illustration (Table 4, basic non-oil balance in 2001, CFA francs billions):
  - Actual basic non-oil balance (2001):
    - Cameroon: 169
    - Gabon: 266
    - Republic of Congo: 183
    - Equatorial Guinea: 192
  - If oil GDP in 2000 had been 10 percent higher:
    - Cameroon: 171
    - Gabon: 269
    - Republic of Congo: 186
    - Equatorial Guinea: 199
  - Conclusion: effects are marginal; basing surveillance on the non-oil balance seems warranted.

### III. Controlling for petroleum price volatility: non-oil-target construction
- Objective: construct a non-oil basic-balance fiscal target that controls for oil price volatility and removes spurious effects of price swings.
- Domestic CFA-franc oil price fluctuations arise from:
  - Swings in the U.S. dollar oil price.
  - Changes in CFA franc / U.S. dollar exchange rate.
- Formal target linking non-oil deficit to an “equilibrium” CFA franc-denominated oil price:
  - y = R_O(p,e, Q, φ̂) − ... with oil revenue R_O = φ̂ Q e p, where:
    - φ̂: projected oil revenue collection ratio;
    - p: medium-term U.S. dollar price per barrel of domestic oil;
    - e: medium-term CFA franc/U.S. dollar exchange rate;
    - Q: annual petroleum production (barrels).
- Practical substitution (Brent used where domestic price projections absent):
  - R_{O,t} = φ_{Brent,t} Q_t e_t p_{Brent,t}, where φ_{Brent,t} = φ_t (p_t / p_{Brent,t}).
- “Equilibrium” oil price considerations:
  - Statistical evidence on petroleum price mean-reversion is mixed; mean-reversion is slow.
  - Futures prices mean-revert over two to three years (Daniel (2002)); market participants use past price averages as forecasts.
  - Interpret “equilibrium” price as an average expectation rather than strict long-run mean.

### Oil price and exchange-rate projection approach (medium-term benchmarks)
- WEO July 2003 projection: Brent price of U.S. dollar 21.5 per barrel used as a medium-term benchmark.
- Alternative time-varying projection: six-year moving average of past oil prices; for 1994–2001 yields projections around 18 to 20 U.S. dollars per barrel; only in 2002 does it approach the WEO July 2003 forecast.
- Medium-term CFA franc/U.S. dollar exchange rate: approximated by a backwards-looking six-year moving average (with correction for 1993/94 devaluation).
- Example impact: in 2000 the actual CFA-franc value of a barrel of Brent was almost twice the synthetic “average” value based on six-year moving averages for oil prices and exchange rate.

### Oil revenue collection ratio φ: measurement and projection
- φ is predetermined to avoid volatility and prevent governments from manipulating actual collection to accommodate higher non-oil deficits.
- Historical series for Brent-based and actual oil revenue collection ratios from 1994; projections produced for 2003–2010.
- Estimation: simple linear trends fitted with structural breaks between historic and projection periods (breaks noted visually, e.g., Chad in 2009, Equatorial Guinea in 2007).
- Cross-country observations:
  - Mature producers (Cameroon, Gabon, Congo) have substantially higher φ than Equatorial Guinea and Chad.
  - Equatorial Guinea: φ increased markedly between 2000 and 2002; projected to reach 50 percent of Brent-based oil GDP (or 60 percent of actual oil GDP) in 2007.
  - Chad: government share expected to increase from 2009 and attain 40-50 percent by 2014.
  - Within mature producers: Cameroon fiscal receipts > 40 percent of oil GDP; Republic of Congo around 25 percent.
- Table 5: country-specific piecewise linear φ specifications (coefficients and standard errors preserved in source table).

### Petroleum production projections and fiscal implications
- Production projections differ markedly:
  - Equatorial Guinea: production expected to increase continuously to a maximum in 2010.
  - Congo: expects to stabilize at a high level.
  - Chad: anticipates a short-lived boom between 2004 and 2009, declining thereafter.
  - Cameroon and Gabon: have passed the heydays; reserves rapidly dwindling.
- Surveillance objective: avoid inflationary pressure from overly expansionary fiscal policies that could threaten the peg and external competitiveness; large public-demand swings from temporary oil-price spikes should be avoided.
- Integration of production projections into surveillance depends on focus: medium-term macro stability versus long-term sustainability.

### Defining and computing sustainable oil-wealth consumption
- Oil wealth defined as net present value of current and future fiscal oil receipts (NPV_Oi,t expressed as in source, preserving equation structure).
- Sustainable oil consumption per capita Ot~R given by equation (11) (structure and variables preserved): depends on socially optimal real discount rate ∂, long-term inflation π, and population N.
- Interpretation:
  - Expression (11) gives sustainable oil consumption per capita that keeps real oil wealth per capita constant; it does not ensure broader fiscal sustainability (requires full government net asset-liability analysis).
- Assumptions and caveats:
  - Future production and prices are hard to project; projections diverge across countries.
  - The socially optimal discount rate is problematic to quantify.
  - Production treated as exogenous (contractual and investment inertia justify this in short-to-medium run; Hotelling (1931) noted).
- Deficit concept choice:
  - Basic fiscal balance includes net interest (“BO-effect”); constant oil-wealth consumption implies a constant real primary, not basic balance.
  - Paper retains basic balance to stay close to CEMAC practice and because switching to primary balance introduces other assets/liabilities not related to oil wealth.
  - For CEMAC, BO-effect likely small: foreign-exchange regulations prevent investing oil receipts outside the zone; BEAC’s “Fund for future generations” remuneration very low (close to zero in real terms).
  - Calculations assume governments start adhering to rule (11) in the year of observation; in the first year BO-effect does not apply.

### IV. Evaluating fiscal surveillance: alternative targets and scenarios (1994–2002 comparisons)
- Four alternative non-oil balance targets compared (Table 6):
  - A. CEMAC current target: ceiling on the non-oil deficit equals current-year oil revenue.
  - B. Medium-term stability criterion using:
    - Table 5 Brent-based φ specifications;
    - medium-term Brent price U.S. dollar 21.5;
    - backward-looking 6-year moving average CFA franc/U.S. dollar exchange rate.
  - C. Medium-term stability as in B but replacing WEO Brent price with backward-looking 6-year moving average.
  - D. Long-term sustainability criterion using equation (11) with:
    - real discount rate ∂ = 5 percent;
    - long-term inflation π = 2 percent;
    - future petroleum receipts and population as projected by national governments.
- Comparative properties and timing sensitivity:
  - Medium-term targets permit larger non-oil deficits in low-price years (e.g., 1994, 1998) than CEMAC target, and tighter stance in high-price years (2000–2002).
  - Long-term sustainability target recommends much smaller non-oil deficits for mature producers (Cameroon, Congo, Gabon).
  - Timing example: Gabon could have afforded a deficit ≈ 6 percent of GDP if stabilizing oil wealth in 1994; eight years later it could afford less than half that size.
- Sensitivity to parameters:
  - Medium-term targets linear in projected oil price and φ: downward revision of either by 10 percent tightens recommended non-oil deficit by 10 percent.
  - Long-run target linear in projected production and prices but not linear in discount rate.
  - Republic of Congo long-term target examples (Table 7):
    - Baseline ∂ = 5 percent: -12.3 (1994), -11.9 (1995), -10.1 (1996), -10.0 (1997), -11.9 (1998), -9.9 (1999), -6.4 (2000), -7.1 (2001), -6.9 (2002).
    - ∂ = 3 percent: -4.3 (1994) ... -2.3 (2002).
    - ∂ = 8 percent: -19.5 (1994) ... -11.7 (2002).
- Performance assessment (summary findings, Table 8):
  - Relative to the CEMAC target:
    - Fiscal discipline strengthened from about 1999; large excess deficits gave way to strong fiscal positions; slippages mainly in Congo and non-oil producing Chad.
  - Relative to medium-term targets:
    - Gabon’s fiscal stance deteriorated for most of 1995–2001; both medium-term targets violated in 2001 (WEO-price based target violated for the first time).
    - Cameroon failed to meet the moving-average based target until 2000 despite appearing compliant under the CEMAC target since 1999.
  - Relative to long-term sustainability:
    - Mature oil economies (Cameroon, Congo, Gabon) failed to save sufficient petroleum receipts in every year observed (exceptions: Equatorial Guinea and Chad between 1996 and 1999).
    - Example: Gabon in 2001 — met CEMAC target easily but was more than 11 percentage points of GDP below long-run sustainability target; CEMAC criterion indicated overperformance by 8 percentage points of GDP while fiscal balance was 11 percent of GDP below sustainable oil-wealth management level.
- WEO oil price series preserved (U.S. dollars per barrel): 15.9 (1994), 17.2 (1995), 20.4 (1996), 19.3 (1997), 13.1 (1998), 18.0 (1999), 28.2 (2000), 24.3 (2001), 25.0 (2002).

### Key findings and policy implications (preserved conclusions)
- Central issues:
  - Oil price volatility can obscure underlying fiscal stance when current-year oil revenue is the surveillance benchmark.
  - Finite oil resources argue for saving some current receipts to finance future consumption; surveillance targets can incorporate medium-term stability and long-term sustainability.
- Empirical evaluation (1994–2002) conclusions:
  - For 2000–2002 (high oil-price period), CEMAC surveillance gives an excessively optimistic picture of fiscal performance.
  - Long-term oil-wealth management perspective indicates CEMAC governments generally spending too much and saving too little, except Equatorial Guinea.
  - Most dramatic: Gabon in 2001 — CEMAC’s criterion suggests overperformance by 8 percentage points of GDP while actual fiscal balance was 11 percentage points of GDP below the sustainable level.
- Practical recommendation:
  - Adopt medium-term stability criteria (moving-average oil prices and exchange rates) and long-term sustainability checks (NPV-based per-capita oil-wealth consumption with explicit discount rate and inflation assumptions) as complements to the current CEMAC target to better capture fiscal stance and intergenerational equity concerns.

*Source: IMF staff paper content unit _wp0408 (excerpts and selected tables and figures preserved).*

### References................................................................................................25

### _wp0408 - References................................................................................................25

### I. Introduction: fiscal policy surveillance in the CEMAC zone
- CEMAC consists of six countries: Cameroon, the Central African Republic, Chad, the Republic of Congo, Equatorial Guinea, and Gabon.
- Prior exchange rate arrangements:
  - Fixed CFA franc / French franc peg until 1999; replaced by a peg to the euro in 1999.
  - The French treasury guarantees the currency peg in principle; BEAC and the CEMAC’s Executive Secretariat pursue policies to sustain the peg.
- Macroeconomic surveillance objectives (post-2002 reform) evaluated annually by the CEMAC’s Executive Secretariat:
  - A nonnegative basic fiscal balance (overall balance excluding grants and foreign financed investment);
  - A level of domestic and foreign debt of no more than 70 percent of GDP;
  - Non-accumulation of domestic and/or external arrears;
  - An annual inflation rate of no more than 3 percent.
- Two issues for oil-exporting members (Cameroon, Republic of Congo, Equatorial Guinea, Gabon, and since end-2003 Chad):
  - Volatility of oil receipts due to petroleum price fluctuations undermines the headline fiscal balance as an indicator of underlying fiscal stance.
  - Finite oil reserves imply intertemporal efficiency and equity questions; part of current oil receipts may need to be saved for future consumption.

### Key empirical context and indicators
- Table 1 (Oil importance, selected CEMAC countries, 1998–2002):
  - Oil GDP as percentage of total nominal GDP (1998–2002, by country):
    - Cameroon: 7.2, 10.3, 14.9, 13.2, 10.9
    - Republic of Congo: 36.5, 53.4, 65.5, 56.7, 53.6
    - Equatorial Guinea: 75.3, 74.2, 85.3, 88.0, 89.9
    - Gabon: 28.1, 38.5, 48.7, 42.1, 41.8
  - Oil receipts as percentage of fiscal revenue (1998–2002, by country):
    - Cameroon: 20.0, 23.7, 31.5, 29.7, 25.2
    - Republic of Congo: 53.1, 71.7, 76.6, 68.2, 69.1
    - Equatorial Guinea: 61.2, 76.9, 83.8, 82.2, 91.7
    - Gabon: 54.6, 45.3, 67.5, 64.1, 55.8
  - Source note: “Prov.” indicates provisional figures. Cameroon fiscal year runs July–June; calendar-year approximations constructed by averaging adjacent fiscal years.
- Table 2 (CEMAC basic fiscal balances, 1994–2002, in percent of GDP):
  - Cameroon (1994–2002): -5.3, -2.1, -0.9, -0.2, -0.6, 0.6, 2.7, 2.6, 2.1
  - Republic of Congo (1994–2002): -12.4, -7.1, -6.9, -5.9, -15.9, -0.6, 7.2, 8.9, -0.8
  - Equatorial Guinea (1994–2002): -2.1, -5.1, -5.3, 3.0, -0.5, 3.7, 9.0, 14.6, 14.6
  - Gabon (1994–2002): 1.8, 6.5, 6.2, 11.1, -1.3, 4.3, 13.9, 7.7, 7.0
  - WEO oil price (U.S. dollars per barrel, 1994–2002): 15.9, 17.2, 20.4, 19.3, 13.1, 18.0, 28.2, 24.3, 25.0
  - Note: Years in which the CEMAC’s fiscal surveillance target was not met are in italics in the source table.
- Observed pattern: 2000 (record oil prices) — no country recorded a negative basic balance; 1998 (low petroleum prices) — all countries recorded negative basic balances.

### II. Analytical framework (formalization and testing)
- Surveillance target general form:
  - xERB
    ttt
    ≥−≡,        (1)
    - R: revenue concept; E: expenditure concept; x: numerical floor; t: period.
  - CEMAC practice: x = zero, B is the basic balance (revenue excluding grants minus expenditure excluding foreign-financed investment).
- Revenue decomposition:
  - NtOtt
    RRR ++= (oil and non-oil revenue components).
  - Oil revenue model:
    - ttOt
      ORφ=,         (2)
      - O: nominal oil GDP; φ: share of oil GDP received by the government (oil revenue collection ratio).
- Non-oil based surveillance target:
  - yERB
    tNtNt
    ≥−≡.        (3)
    - y reflects the level of oil receipts the government can expect “on average” or “in the medium to long term”.
  - Two options for y discussed in subsequent sections:
    - (a) annual flow of oil receipts using a medium-term oil price projection (Section III);
    - (b) an annuity from the net present value of future fiscal oil receipts accounting for finite reserves and production path (Section IV).
- Conditions for basing surveillance on non-oil balance: non-oil balance must not vary systematically with oil receipts for exogenous reasons.
- Empirical testing framework:
  - Non-oil revenue representation:
    - ttNt
      NRγ=,         (4)
      - N: nominal non-oil GDP; γ: non-oil revenue collection ratio.
  - Relationship of non-oil GDP to oil GDP:
    - )(
      ~
      ONNN
      ttt
      &&&
      +=.        (5)
      - N
        ~ : “pure” non-oil economy; )(ON
        t
        &&&
        captures oil-induced fluctuations (trickle-down or Dutch disease).
    - Transformed for estimation:
      - itititiit
        uOON+∆+∆+=∆
        −110
        θα,        (5a)
        - ∆ denotes first differences (detrending).
  - Estimation sample and method:
    - Panel of current CEMAC oil-producing countries: Cameroon, Gabon, Republic of Congo, Equatorial Guinea.
    - Period: 1994–2001 (2002 excluded because GDP figures were preliminary).
    - Estimators considered: random-effects and fixed-effects; Hausman test favors random effects (country-specific effects uncorrelated with regressors).
- Key empirical findings from Table 3 (Correlation between oil and non-oil GDP):
  - Preferred specification: First differences, Random Effects, One Lag (column 1).
  - Estimated coefficients (column 1):
    - 0θ = -.051 (p-value .402)
    - 1θ = .181 (p-value .007)
  - Test on ∑θ = 0: statistic 1.71 (p-value .190).
  - R-squared = .118; Observations = 24.
  - Interpretations:
    - 1θ positive and significant at the 1-percent level in column 1 suggests a positive lagged trickle-down effect of oil GDP on non-oil GDP.
    - Joint effect (0θ + 1θ) significant only at the 20 percent level; adding another lag (column 3) removes significance of 1θ. Hence joint effect should be treated as an upper band, not a dependable point estimate.
- Effect on non-oil revenue collection ratio γ:
  - Decomposition (equation 6):
    - t
      t
      o
      t
      t
      t
      N
      ON
      N
      N)(
      ~
      &&
      &
      γγγ+=,       (6)
      - γ
        ~ : revenue collection ratio on “pure” non-oil GDP; Oγ : collection ratio on “oil-induced” non-oil GDP.
  - Estimation approach uses predicted components from (7) based on Table 3 column 1 estimates.
  - Point estimates (unweighted means across countries, first differences, OLS):
    - γ
      ~ = 0.233 (standard error: 0.042)
    - Oγ = 0.205 (standard error: 0.111)
    - An F-test fails to reject equality of γ
      ~ and Oγ at conventional significance levels — i.e., no mechanical effect of oil GDP on non-oil revenue collection ratio found.
  - Robustness caveat: Oγ estimate not very robust to sampling period (small sample size, low statistical power), but γ
    ~ and Oγ difference remains insignificant across samples.
- Quantitative illustration of magnitude (Table 4):
  - Actual basic non-oil balance in 2001 (in billions of CFA francs):
    - Cameroon: 169
    - Gabon: 266
    - Republic of Congo: 183
    - Equatorial Guinea: 192
  - Balance if oil GDP in 2000 had been 10 percent higher (same units):
    - Cameroon: 171
    - Gabon: 269
    - Republic of Congo: 186
    - Equatorial Guinea: 199
  - Conclusion: effects are marginal; basing surveillance on the non-oil balance seems warranted.

### III. Controlling for volatility in petroleum prices (proposed non-oil-target construction)
- Objective: construct a fiscal target for the non-oil basic balance that controls for oil price volatility and removes spurious effects of price swings.
- Domestic oil price fluctuations arise from:
  - Swings in the oil price denominated in U.S. dollars (Figure 1 in source).
  - Changes in the CFA franc / U.S. dollar exchange rate (Figure 2 in source).
- Formal target linking non-oil deficit to an “equilibrium” CFA franc-denominated oil price:
  - )(pRy
    O
    −=,                        with                                                                                    
    tttttO
    QeppOpRφφ
    ˆ
    )(
    ˆ
    )(==,                                                                                    (8)
    - φ̂ : projected oil revenue collection ratio;
    - p: “medium-term” U.S. dollar price per barrel of domestic oil;
    - e: “medium-term” CFA franc/U.S. dollar exchange rate;
    - Q: annual petroleum production (in barrels).
  - In practice, due to lack of medium-term projections for Cameroon-, Congo-, and Gabon-specific oil, substitute domestic price by Brent price:
    - tt
      Brent
      ttO
      QeppR
      ,
      ˆ
      )(
      φ=,                                                                                                (9)
      - φ, the Brent-based oil revenue collection ratio, is )/(
        Brenttt
        ppφφ=.
- Remarks on “equilibrium” oil price:
  - Statistical evidence on petroleum price properties is mixed; mean-reversion is slow.
  - Futures prices for oil mean-revert and do so over relatively short periods of two to three years (citing Daniel (2002)); market participants use past price averages as forecasts.
  - The “equilibrium” price should be interpreted as an average expectation rather than a strict long-run mean.

*Source: IMF staff paper content unit _wp0408 - References (selected sections: Introduction; Tables 1–4; Sections II–III excerpts).*

### 21.5 per barrel. The WEO projection suggests itself, as it is a widely consulted source for

### _wp0408 - 21.5 per barrel. The WEO projection suggests itself, as it is a widely consulted source for

### Oil price and exchange-rate projection approach
- The WEO July 2003 projection uses a Brent price of U.S. dollar 21.5 per barrel as a medium-term benchmark; this projection exceeds earlier WEO projections because high oil prices between 2000 and 2002 pushed expectations upwards.
- An alternative, time-varying oil price projection is used: a six-year moving average of past oil prices, which for the period 1994–2001 yields projections that hover around 18 to 20 U.S. dollars per barrel; only in 2002 does the moving-average projection approach the WEO July 2003 forecast.
- No fixed numerical WEO projection exists for the CFA franc/U.S. dollar exchange rate; the medium-term exchange rate is approximated by a backwards-looking six-year moving average (with a correction for the devaluation in 1993/94).
- Example of impact: in 2000, the actual CFA-franc value of a barrel of Brent was almost twice as high as a synthetic “average” value based on six-year moving averages for both oil prices and the exchange rate.

### Oil revenue collection ratio: measurement and parametric projection
- The oil revenue collection ratio φ (phi) is predetermined in Equations (8) and (9) to avoid introducing noise from wide year-to-year fluctuations and to prevent governments from manipulating actual collection to accommodate higher non-oil deficits.
- Historical series for Brent-based and actual oil revenue collection ratios are shown from 1994; projections are produced for 2003–2010.
- A simple linear trend was fitted to all countries’ oil revenue collection ratios, allowing structural breaks between the historic and projection periods (visual-inspection breaks noted, e.g., Chad in 2009 and Equatorial Guinea in 2007); specifications are reported in Table 5.
- Key cross-country observations:
  - Cameroon, Gabon, and Congo (mature producers) have substantially higher oil revenue collection ratios than Equatorial Guinea and Chad (younger or prospective producers).
  - Equatorial Guinea’s oil revenue collection ratio increased markedly between 2000 and 2002 and is projected to reach 50 percent of Brent-based oil GDP (or 60 percent of actual oil GDP) in 2007.
  - Chad’s government share in oil GDP is expected to increase from 2009 onwards and to attain 40-50 percent by 2014.
  - Within mature producers, large differences reflect contractual arrangements: Cameroon receives more than 40 percent of oil GDP as fiscal receipts; Republic of Congo around 25 percent.
- Table 5 (specifications for Brent-based oil revenue collection ratio φ) reports country-specific piecewise linear specifications with coefficient estimates and standard errors in parentheses (examples preserved exactly as presented for Cameroon, Gabon, Republic of Congo, Equatorial Guinea, and Chad).

### Petroleum production projections and implications for fiscal policy
- Projections of future oil production differ markedly across CEMAC countries (Figure 4 summary):
  - Equatorial Guinea: oil boom has just begun; production expected to increase continuously until reaching a maximum in 2010.
  - Congo: expects to stabilize oil production at a high level.
  - Chad: anticipates a relatively short-lived oil boom between 2004 and 2009, declining thereafter.
  - Cameroon and Gabon: have already seen the heydays; need to manage rapidly dwindling reserves.
- Surveillance exercise objective: thwart inflationary pressure from overly expansionary fiscal policies to protect the currency arrangement and external competitiveness; large changes in public demand triggered by temporary swings in oil prices should be avoided.
- Whether production projections should be integrated into surveillance depends on whether the exercise focuses on medium-term macro stability only or also on long-term sustainability.

### Defining and computing sustainable oil wealth consumption
- Oil wealth is defined as the net present value of current and future fiscal oil receipts:
  - NPV_Oi,t = ∑_{i=−t}^{∞} [R_{Oi,ti} / (1+∂)^{i}(1+π)]
    (equation (10) structure preserved as in source).
- Distributing oil wealth equitably per capita yields the sustainable oil consumption Ot~R (equation (11) preserved in structure and variables):
  - Ot~R = expression derived from equation (11) where ∂ is the socially optimal real discount rate, π is long-term inflation, and N is population size.
- Interpretation:
  - Expression (11) gives the sustainable oil consumption per capita that holds real oil wealth per capita constant; it does not guarantee broader fiscal sustainability (would require analysis of the government’s entire net asset-liability position).
- Assumptions and caveats:
  - Future petroleum production and prices are hard to project; projected booms differ (e.g., Equatorial Guinea projected through 2010 vs. Chad short-lived).
  - The socially optimal discount rate is problematic and difficult to quantify.
  - Equation (11) treats production as exogenous (justified by contractual and investment inertia in short-to-medium term; Hotelling (1931) theoretical note retained).
- Deficit concept:
  - The basic fiscal balance is not ideal for tracking sustainable oil-wealth management because it includes net interest payments (the “BO-effect”); consuming oil wealth at a constant rate implies a constant real primary, not basic balance.
  - The paper sticks to basic balance to stay close to CEMAC practice and because switching to primary balance would include interest on all financial assets/liabilities, not only oil-wealth related ones.
  - For CEMAC, the BO-effect is likely small: foreign-exchange regulations prevent investing oil receipts outside the zone; remuneration on the BEAC’s “Fund for future generations” is very low (close to zero in real terms).
  - Calculations assume governments start adhering to rule (11) in the very year of observation; in the first year the BO-effect does not apply.

### Evaluating fiscal surveillance: alternative targets and scenarios
- Four alternative targets/floors for the non-oil balance (1994–2002) are compared (Table 6):
  A. CEMAC’s current fiscal surveillance target: ceiling on the non-oil deficit equals current-year oil revenue.
  B. Medium-term stability criterion (equation (9)) using:
     - Table 5 specifications for Brent-based oil revenue collection ratio;
     - medium-term Brent price of U.S. dollar 21.5;
     - backward-looking 6-year moving average for CFA franc/U.S. dollar exchange rate.
  C. Medium-term stability criterion as in [B.] but replacing the WEO July 2003 Brent price projection with a backward-looking 6-year moving average.
  D. Long-term sustainability criterion (equation (11)) using:
     - real discount rate ∂ = 5 percent;
     - long-term inflation π = 2 percent (consistent with the peg of the CFA franc to the euro);
     - future petroleum receipts and population sizes as projected by national governments.
- Comparative properties:
  - Medium-term targets allow larger non-oil deficits in years with low oil prices (e.g., 1994, 1998) than the CEMAC target, and recommend tighter stance in high-price years (2000–2002).
  - Long-term sustainability target recommends much smaller non-oil deficits for mature producers (Cameroon, Congo, Gabon).
  - Timing matters: e.g., Gabon could have afforded a deficit of about 6 percent of GDP had it begun stabilizing oil wealth in 1994; eight years later it could afford less than half that size.
- Sensitivity:
  - Medium-term targets are linear in projected oil price and oil revenue collection ratio: a downward revision of either parameter by 10 percent tightens recommended non-oil deficit by 10 percent.
  - Long-run target is linear in projected petroleum production and prices but not linear in the discount rate. Table 7 illustrates for Republic of Congo:
    - Baseline (5 percent real discount rate) long-term target examples: -12.3 (1994), -11.9 (1995), -10.1 (1996), -10.0 (1997), -11.9 (1998), -9.9 (1999), -6.4 (2000), -7.1 (2001), -6.9 (2002 prelim).
    - With real discount rate 3 percent: -4.3 (1994) ... -2.3 (2002).
    - With real discount rate 8 percent: -19.5 (1994) ... -11.7 (2002).
- Performance assessment (Table 8 summary and preserved findings):
  - Relative to the CEMAC target, fiscal discipline strengthened from about 1999; large persistent excess deficits gave way to strong fiscal positions; occasional slippages mainly in Congo and non-oil producing Chad.
  - Relative to medium-term targets, Gabon’s fiscal stance deteriorated for most of 1995–2001; both medium-term targets were violated in 2001 (WEO-price based target violated for the first time). Cameroon failed to meet the moving-average based target until 2000 despite appearing compliant under the CEMAC target since 1999.
  - Long-term sustainability: mature oil economies Cameroon, Congo, Gabon failed to save sufficient petroleum receipts in every year of observation (exception: Equatorial Guinea and Chad between 1996 and 1999).
  - Dramatic illustration: Gabon and Congo in 2000/01 met the CEMAC target easily (high oil prices and strong U.S. dollar) but failed medium-term and long-run sustainability targets by wide margins (Gabon in 2001 fell short by more than 11 percentage points of GDP versus sustainability target).
- Numerical examples preserved from Table 6 (selected WEO oil price series):
  - WEO oil price: 1994 = 15.9; 1995 = 17.2; 1996 = 20.4; 1997 = 19.3; 1998 = 13.1; 1999 = 18.0; 2000 = 28.2; 2001 = 24.3; 2002 = 25.0 (values preserved exactly as presented).

### Key findings and policy implications
- Two central issues:
  - Volatility of oil prices can obscure a country’s underlying fiscal policy stance when using current-year oil revenue as the surveillance benchmark.
  - Finite oil resources imply desirability of saving some current receipts to finance future consumption; surveillance targets can be adjusted to reflect medium-term stability and long-term sustainability.
- The paper’s empirical evaluation (1994–2002) shows:
  - For 2000–2002 (high oil-price period), the CEMAC surveillance exercise gives an excessively optimistic picture of fiscal performance.
  - Including long-term oil-wealth management considerations indicates governments in the CEMAC zone are generally spending too much and saving too little, except Equatorial Guinea.
  - Most dramatic case: Gabon in 2001 — CEMAC’s criterion indicates overperformance by 8 percentage points of GDP while fiscal balance was 11 percent of GDP below the level necessary for sustainable oil-wealth management.
- Practical recommendation:
  - Consider adopting medium-term stability criteria (using moving-average oil prices and exchange rates) and long-term sustainability checks (NPV-based per-capita oil wealth consumption using explicit discount rate and inflation assumptions) as complements to the current CEMAC target to better capture fiscal stance and intergenerational equity concerns.

*Source: IMF working paper content (excerpts from _wp0408 PDF provided).*

### REFERENCES

### _wp0408 - REFERENCES

### References cited
- Barnett, Steven, and Rolando Ossowski, 2002, “Operational Aspects of Fiscal Policy in Oil-Producing Countries,” IMF Working Paper 02/177 (Washington; International Monetary Fund)
- Chalk, Nigel, 1998, “Fiscal Sustainability with Non-Renewable Resources,” IMF Working Paper 98/26 (Washington; International Monetary Fund)
- Cuc, Milan, 2002a ,“Fiscal Sustainability,” Republic of Congo – Selected Issues and Statistical Appendix (Washington; International Monetary Fund), pp. 27-32
- Cuc, Milan, 2002b, “Collaterized Debt,” Republic of Congo – Selected Issues and Statistical Appendix (Washington; International Monetary Fund), pp. 32-39
- Daniel, James, 2002, “Hedging Government Oil Price Risk,” Journal of Energy and Development, Volume 27, No. 2, pp. 167-78.
- Davis Jeffrey M., Rolando Ossowski, James Daniel, and Steven Barnett, 2001, Stabilization and Savings Funds for Renewable Resources – Experience and Fiscal Policy Implications, IMF Occasional Paper No. 205 (Washington: International Monetary Fund)
- Engel, Eduardo, and Rodrigo Valdés, 2000, “Optimal Fiscal Strategy for Oil-Exporting Countries,” IMF Working Paper 00/112 (Washington; International Monetary Fund)
- Hotelling, Harold, 1931, “The Economics of Exhaustible Resources,” Journal of Political Economy, Volume 39, No. 2, pp. 137-75.
- Liusksila, Claire, Alejandro García, and Sheila Basset, 1994, “Fiscal Policy Sustainability in Oil-Producing Countries,” IMF Working Paper 94/137 (Washington; International Monetary Fund)
- Ntamatungiro, Joseph, 2002, “Analysis of Fiscal and Debt Sustainability,” Gabon – Selected Issues and Statistical Appendix (Washington; International Monetary Fund), pp. 16-35
- Pindyck, Robert, 1999, “The Long-Run Evolution of Energy Prices,” Energy Journal, Volume 20, No. 2, pp. 1-27

*Source: _wp0408 - REFERENCES*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2004/_wp0408.pdf_