## 1.    Exchange Rate and Nominal Anchor for Major OCD in 2005........................................ 13

## Source details

**Canonical URL:** [1.    Exchange Rate and Nominal Anchor for Major OCD in 2005........................................ 13](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08142.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08142.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08142.pdf.json)

---

### Overview
- Scope:
  - Interaction of exchange rate regimes and nominal anchors with oil price movements for oil-exporting countries (OEC), with focused analysis of Venezuela.
  - References to empirical figures and appendices that provide indicators, fiscal and monetary series, and cross-country classifications.
- Figures referenced (selected): Figure 1 through Figure 8 (see source list within unit for exact titles).
- Appendices informing analysis: Appendix I (Oil Exporting Countries-Selected Indicators), Appendix II (Management of Oil Resources: Oil Funds and International Reserves), Appendix III (Descriptive Statistics and Pair-Wise Correlations).

### Conceptual framework: Oil dominance (OD) and fiscal dominance (FD)
- Monetary base identity and central bank accounting (notation preserved in source):
  - MB = NIR + NDA  (presented as MBNIRNDAΔΔΔ=+ in source tables)
- Mechanisms in OEC:
  - Oil exports can increase Net International Reserves (NIR) and increase government deposits at the central bank (decline in NDA), producing a temporary net zero effect on MB until deposits are spent.
  - Spending of oil receipts increases NDA and expands MB; when spending and NIR changes occur in the same period, central bank accounts may obscure the fiscal origin of MB expansion.
- Net worth / Net Operating Balance approach:
  - Oil extraction represents depletion of a non-renewable government asset; oil-related receipts are a change of one asset for another and should not be treated as pure fiscal revenue in net worth accounting.
  - Declining government net worth (NW) or an unsustainable Net Operating Balance (NOBg) signals fiscal dominance even when stock-of-debt indicators appear benign.
- Central bank budget constraint and change-in-MB identity:
  - Central bank equations (equations (1'), (2'), (2") in source) express MBΔ in terms of changes in net foreign assets, net domestic credit and central bank net worth.
- Consolidated public sector perspective:
  - Consolidated balance and net worth equations (equations (4) and (5) in source) show that using oil receipts for consumption (negative NOB) lowers NW and can be offset by MB increases and/or declines in foreign assets, constraining central bank resistance.

### Practical indicator of OD/FD
- Domestic fiscal balance defined:
  - OBd = revenues from domestic sources minus domestic expenditures.
- Ratio indicator (equation (6) in source):
  - γ = (G - Rno) / Ro
    - Ro = oil-related net receipts
    - Rno = non-oil revenue
- Interpretation:
  - γ > 0 implies domestic deficit financed in part by oil receipts, indicating an initial monetary expansion of oil origin; larger γ implies higher possibility of OD/FD.
  - Indicator complements net worth framework and captures monetary impact of fiscal policy in OEC.

### Transmission mechanisms (qualitative)
- Under fixed exchange rate or surrender requirements:
  - Increase in oil prices → government spending financed by oil receipts → nominal MB expansion → higher demand for tradables and higher nontradable prices → higher domestic inflation.
- Under fully flexible exchange rate:
  - Oil price increase → nominal exchange rate appreciation → smaller MB-driven shock to domestic inflation, but government spending can still raise nontradable prices and domestic inflation.
- Policy implication from framework:
  - Isolating fiscal policy from oil price fluctuations (e.g., via oil funds or rules) reduces transmission of oil price volatility to output, inflation and the current account.

### Cross-country stylized facts (sample: 30 OEC, annual data 1979–2005)
- Oil dominance metrics used:
  - Xo/X (oil exports to total exports), Xo/GDP (oil exports to GDP), Ro/RT (oil-related fiscal receipts to total fiscal revenue).
- Key indicators analyzed: GCG/GDP, pb_GG, ∆%RM (growth of nominal reserve money), ∆% NEER, ∆% REER, CPI change, ∆RM/GDP.
- Empirical patterns (as reported in source):
  - Major OEC with highest oil dominance tended to adopt the exchange rate as nominal anchor; several maintain pegged regimes.
  - Average nominal growth of MB for OEC is closely related to changes in oil prices (Figure 1).
  - Government expenditure correlates with oil dominance (Xo/GDP); consumption spending implies NW reduction.
  - Trend since 1999: gradual decline in government expenditure overall and improved primary balance alongside increased oil receipts (Figure 2).
  - Inflation shows correlation with oil price changes but with considerable cross-country dispersion (Figures 3 and 4).
  - Exceptions: Angola, Nigeria and Venezuela exhibited high inflation and managed floats or exchange controls despite high oil dominance.

### Venezuela: Oil dominance and macroeconomic overview (1960–2005)
- Oil sector importance (exact figures from source):
  - Oil exports represent about 86 percent of total exports on average for 1960–2005.
  - Oil revenue share of total fiscal revenue fluctuated between 86 percent (1974) and 38 percent (1998).
  - Oil GDP share averaged around 22 percent for 1960–2005 (36 percent in 1974, 12 percent in 1986).
- Macroeconomic volatility:
  - Large volatility in inflation, real GDP growth, nominal NEER and monetary base (see Table 2 and Figure 7 in source).
- Exchange rate regimes (table 3 in source; verbatim periods):
  - 1964–1983 Fixed Single Rate (Pegged to the US dollar)
  - 1983–1989 Multiple Exchange Rates-Exchange Controls
  - 1989–1992 Floating
  - 1992–1994 Crawl
  - 1994–1996 Multiple Exchange Rates-Exchange Controls
  - 1996–2002 Bands
  - 2002–2003 Floating
  - 2003–  Fixed Single Rate-Exchange Controls
- Central bank instruments and costs:
  - Since 1990 the central bank placed its own instruments (CBIMC) to control liquidity; stock of AI fluctuated significantly during 1990–2005, sometimes exceeding 100 percent of MB.
  - Financial costs of placing CBIMC often exceeded returns on central bank foreign assets — an indicator of OD/FD (Figure 8).
- Oil stabilization funds and transfers:
  - FIEM created in 1998; by end-2001 FIEM net worth was 50 percent of central bank foreign assets. FIEM replaced by FEM in 2003; only 11 percent of 2001 net worth preserved. No accumulation of oil-related resources since then despite oil price increases (footnote in source).

### Simple correlations for Venezuela (1960–2005) — descriptive evidence
- Variables (logs): oil prices (Venezuelan basket), oil exports (US$), fiscal revenue total and oil-related (local currency), primary fiscal expenditure, monetary base, non-oil nominal and real GDP, non-oil GDP deflator (proxy for domestic inflation).
- Contemporaneous correlations (rates of growth):
  - Changes in nominal oil exports mostly determined by oil price movements; volume changes have very little effect.
  - Changes in fiscal oil revenue closely related to changes in oil exports; fiscal oil revenue highly correlated with total fiscal revenue and primary fiscal expenditure.
  - Changes in fiscal primary expenditure highly correlated with nominal non-oil GDP, non-oil GDP deflator (inflation), and nominal MB.
  - Correlation between growth of fiscal primary expenditure and non-oil real GDP is very small and negative.
- Volatility (five-year moving variances):
  - Volatility of oil price growth and nominal oil export growth are very similar.
  - Volatility of fiscal oil revenue growth highly correlated with volatility of nominal oil exports.
  - Volatility of fiscal primary expenditure growth highly correlated with volatility of oil-related variables.
  - Variances of oil prices and MB converge more closely after 1985.
- Domestic vs overall deficit:
  - Domestic primary deficit substantially larger than overall deficit; five-period dynamic correlation between domestic and overall deficits (as % of GDP) is very low.
- Inference:
  - Correlation indicators do not contradict presence of oil/fiscal dominance in Venezuela: fiscal variables strongly related to oil variables and to MB and inflation measures.

### Empirical testing: VAR / VECM approach for OD/FD (general setup)
- Objectives:
  - Test OD/FD hypothesis: existence of close link between oil prices, domestic primary fiscal deficit and monetary base.
  - Test relationship between monetary base and prices.
- Methodology:
  - VAR and VECM estimated on annual data 1960–2005; oil price treated as exogenous.
  - Variable ordering in VAR: 1) fiscal; 2) monetary; 3) external sector.
  - Lags chosen using Akaike and Schwarz criteria and VAR stability checks.
- Empirical expectations under OD/FD:
  - I-Rs and VDs should show positive response of MB to shocks to domestic primary deficit and no response of domestic primary deficit to shocks to MB.
  - One-way Granger-causality from domestic primary deficit to MB.
  - A cointegrating vector with positive relationship between domestic primary deficit and MB.

### Model 1 (nominal variables, logs) — key empirical findings (1960–2005)
- Variables: LPON (nominal oil prices, exogenous); LDDPN (nominal domestic primary deficit); LNEXEP (nominal exchange rate, Bs./US$); LR (nominal net international reserves, US$); LMB (nominal monetary base).
- Johansen cointegration (LMB and LDDPN):
  - Cointegrating vector (normalized): COINT1A = LMB - 1.075578*LDDPN  (LDDPN coefficient statistically significant, negative and close to one).
- VECM short-run I-R and VD (complete sample):
  - One standard deviation shock to DLDDPN → positive effect on DLMB, explaining around 17 percent of DLMB forecast error variance in a twenty-period horizon.
  - Response of DLDDPN to shocks to DLMB is positive but very close to zero per VD.
- Granger/block causality:
  - Block Granger-causality indicates one-way Granger-causality from LDDPN to LMB.
  - Within VECM, COINT1A(-1) coefficient is significantly different from zero and negative in the DLMB equation; COINT1A(-1) positive but not statistically significant in DLDDPN equation.
- Role of oil prices:
  - DLPON(-1) coefficient positive and statistically different from zero in DLDDPN equation (1 percent level) and in DLR equation (10 percent level) — oil price shocks affect fiscal and reserve behavior.

### Subsample (soft-peg 1960–1982 vs variable-peg 1983–2005) — Model 1 contrasts
- Common result: shocks to DLDDPN positively affect DLMB in both sub-periods.
- Quantitative differences:
  - Soft-peg: shocks to DLDDPN explain approximately 60 percent of DLMB forecast error variance.
  - Variable-peg: shocks to DLDDPN explain around 20 percent of DLMB forecast error variance.
- Responses of DLDDPN to DLMB:
  - Soft-peg: DLDDPN responds positively but mildly to DLMB; shocks to DLMB explain close to 5 percent of DLDDPN forecast error variance.
  - Variable-peg: response of DLDDPN to DLMB negative but close to zero per VD.
- Granger-causality puzzle (soft-peg):
  - Evidence of Granger-causality from monetary base to domestic primary deficit (COINT1A(-1) positive and statistically different from zero in DLDDPN equation), but not the reverse; COINT1A(-1) not statistically significant in DLMB equation.
  - DLPON(-1) not statistically significant in DLDDPN and DLR equations for soft-peg.
- Variable-peg aligns with full-sample findings:
  - Granger-causality from domestic primary deficit to monetary base (COINT1A(-1) negative and significant in DLMB equation).
  - DLPON(-1) positive and significant in DLDDPN equation but not in DLR.

### Model 2 (Hodrick-Prescott gaps — variables as deviations from trend) — summary
- Variables: LPONG (oil prices gap, exogenous); LDDPNG (domestic primary deficit gap); LNEXEPG; LRG; LMBG.
- Unit-root tests: variables are I(0).
- I-R and VD (ordering LDDPNG, LMBG, LRG, LNEXPG):
  - Shock to LMBG → negative impact on LDDPNG.
  - VD: LMBG explains ~1.4 percent of LDDPNG forecast error variance in 20-period horizon.
  - LMBG responds positively to shocks to LDDPNG.
  - VD: LDDPNG explains ~8.7 percent of LMBG forecast error variance in 20-period horizon.
  - No Granger causality between LDDPNG and LMBG.
  - LPONG coefficient positive and statistically significant in equations for LDDPNG, LRG and LMBG.

### Model 3 (percent of GDP / real terms) — summary
- Variables: DLPOR (change in log real oil price, exogenous); VDDPYN (change in nominal domestic primary deficit as % of GDP); DLREXEP; DLRREAL; SM (seigniorage: change in nominal MB as % of GDP).
- Unit-root tests: variables are I(0).
- I-R and VD (ordering VDDPYN, SM, DLRREAL, DLREXEP):
  - VDDPYN response to SM shocks positive but very small; VD: SM explains close to zero of VDDPYN forecast error variance.
  - SM responds positively to VDDPYN shocks; VD: VDDPYN explains ~9.4 percent of SM forecast error variance in 20-period horizon.
  - Block Granger-causality: SM does not Granger-cause VDDPYN, but VDDPYN Granger-causes SM (P-value = 0.072).
  - DLPOR coefficient positive and statistically significant in equations for VDDPYN, DLRREAL and SM.

### Testing monetary base — prices relationship (1960–2005)
- Variables: LMB (log monetary base) and LCP (log CPI).
- Unit-root tests: LMB and LCP are I(1).
- Johansen test: one cointegrating equation but residuals appear non-stationary on visual inspection.
- Gregory and Hansen (G&H, 1996) tests allowing for structural break:
  - G&H rejects null of no cointegration at 1 percent level with structural break in 1968.
  - Level shift model (model C in G&H) produces cointegrating equation (t-statistics in parenthesis):
    - LCP = -1.199748 + 0.972699*LMB - 1.058871*DU68
      - (-10.72852) (47.26916) (-7.12508)
    - DU68 dummy = 1 from 1969 on.
  - Coefficient of LMB is close to one; conclusion: with structural break in 1968, evidence of long-run equilibrium between MB and price level.
- VECM (five lags) for DLMB and DLCP with COINN2(-1) as exogenous (residuals from G&H cointegrating vector):
  - I-R and VD (1960–2005):
    - Shock to DLMB → positive response in DLCP, explaining ≈ 20 percent of DLCP forecast error variance.
    - Shock to DLCP → positive response in DLMB, explaining ≈ 33 percent of DLMB forecast error variance.
  - Block Granger-causality: indicates two-way causality between DLCP and DLMB.
  - COINN2(-1) statistically significant only in DLMB equation → Granger-causality from DLCP to DLMB.
  - Interpretation: evidence consistent with accommodative monetary policy, possibly resulting from OD/FD constraints on monetary policy independence.

### Subsample results for money–prices link (soft-peg vs variable-peg)
- Shocks DLCP → DLMB:
  - Variable-peg: DLCP shocks explain up to 40 percent of DLMB forecast error variance, declining to 31 percent in a ten-period horizon.
  - Soft-peg: DLCP shocks explain up to 21 percent in a gradual increase.
- Shocks DLMB → DLCP:
  - Variable-peg: DLMB shocks explain up to 29 percent of DLCP forecast error variance.
  - Soft-peg: DLMB shocks explain up to 15 percent of DLCP forecast error variance.
- Block Granger-causality:
  - Soft-peg: block tests indicate absence of causality either way; COINN2(-1) positive and significant at 10 percent in DLMB equation → Granger-causality from DLCP to DLMB per VECM residual.
  - Variable-peg: bidirectional causality per block tests (stable VAR estimated for this sub-period where stable VECM including COINN2(-1) could not be obtained).

### Main empirical conclusions (as reported)
- Evidence supports the OD/FD hypothesis in Venezuela for 1960–2005:
  - Nominal-variables-in-logs specification (Model 1) provides strongest support and clear VECM identification of short- and long-run effects.
  - Close long-run relationship between nominal domestic primary deficit and nominal monetary base, driven more by variable-peg period (1983–2005); soft-peg period yields mixed signals.
- Money–prices link:
  - Close relationship between monetary base and prices in both long-run and short-run.
  - Short-run link is bidirectional, interpreted as accommodative monetary policy that may result from OD/FD constraints on central bank independence.
  - Relationship more robust in variable-peg period.
- Oil prices (exogenous) are statistically important in explaining domestic primary deficit across models, reinforcing OD/FD interpretation.

### Appendices and data highlights
- Appendix I (Oil Exporting Countries-Selected Indicators):
  - Classification by Xo/GDP ranges and country listings for snapshots 1965, 1980, 1995, 2005.
  - Definitions: Xo/XT, Xo/GDP, Ro/RT, ∆%RM, ∆%NEER, ∆%REER, GCG/GDP, pb_GG etc.
  - Selected cross-country averages 1999–2005 include exact country-level values (e.g., Venezuela: Xo/XT 82.1; Xo/GDP 24.8; Ro/RT 46.3; ∆% MB 30.3; Inflation 20.5; GCG/GDP 25.4; pb_GG (%GDP) 3.1; Real GDP growth 1.9).
- Appendix II (Oil funds and international reserves — Venezuela):
  - FIEM created in 1998; initial contribution on December 29, 1999; large transfers to public sector in 2003 (~76 percent of FIEM net worth transferred); FIEM became FEM in December 2003.
  - FEM threshold: three-year moving average of oil-related receipts; no contributions after 2003.
  - FONDEN created July 2005 with resources transferred from Central Bank’s international reserves; transfers recorded as domestic asset.
  - Central Bank balance sheet excerpts (selected dates and exact figures in source) show Transfers to Fonden equal 6,079 (30-Sep-05), 10,309 (31-Oct-05), 12,453 (30-Nov-05 onward), Monetary base entries and Adjusted Net Worth series including negative values (e.g., 31-Mar-06 Adjusted Net Worth -191; 31-Dec-06 -4,282).
- Appendix III (Descriptive statistics and pair-wise correlations):
  - Sample 1960–2005 (46 observations for many series).
  - Exact descriptive statistics for series such as RXPXT (Mean 86.99; Std. Dev. 7.72; Observations 46).
  - Pair-wise correlations (rates of change and five-year moving variances) reported with exact numeric entries (examples: DLPON with DLXP 0.94; DLIFP with DLIFT 0.96; VDLPON with VDLXP 0.97; VDLIFP with VDLIFT 1.00).
  - Figures A.III.1–A.III.4 illustrate changes and moving variances; Figure A.III.4 reports lead-lag correlations for overall fiscal balance and domestic primary deficit with exact values for lags/leads i = 0 through i = 5.

### Policy-relevant implications (as discussed in text)
- Central bank independence is necessary for monetary policy to achieve price stability; fiscal dominance implies loss of that independence.
- In OEC, standard fiscal and monetary diagnostics may mask fiscal-origin monetary expansions because oil receipts alter central bank foreign assets and government deposits; net worth and operational indicators (γ ratio, adjusted net worth, oil fund rules) are important complements.
- Institutional arrangements for oil revenue management (FIEM, FEM, FONDEN) and operational rules matter for insulating fiscal policy from oil price volatility; observed practices in Venezuela (large withdrawals, transfers to FONDEN recorded as domestic asset) limited accumulation of stabilization buffers.
- Exchange rate regime matters for transmission of oil shocks to domestic inflation and MB; flexible exchange rates can attenuate some nominal effects but fiscal responses still transmit to nontradable prices.
- Empirical evidence for Venezuela suggests accommodative monetary policy and limited central bank resistance to fiscal impulses driven by oil receipts, highlighting the need for stronger fiscal frameworks and credible institutional rules to restore monetary policy independence.

*Source: _wp08142 - 1.    Exchange Rate and Nominal Anchor for Major OCD in 2005 (PDF chapter/section).*

### 1.    Exchange Rate and Nominal Anchor for Major OCD in 2005........................................ 13

### 1.    Exchange Rate and Nominal Anchor for Major OCD in 2005........................................ 13

### Overview
- Section title: "1.    Exchange Rate and Nominal Anchor for Major OCD in 2005........................................ 13"
- Related country coverage: Venezuela (selected indicators and exchange rate system references appear in the adjacent entries).
- This content unit references multiple figures and appendices that provide empirical indicators, fiscal and monetary series, and cross-country classifications relevant to oil-exporting countries (OEC) and Venezuela.

### Figures referenced within or directly tied to this unit
- Figure 1.    OEC: Average Rate of Nominal Growth of Monetary Base and Changes in Oil Prices  14
- Figure 2.    OEC: Primary Balance of the General Government........................................................ 15
- Figure 3.    Change in the CPI-OEC and Oil Prices, 1979–2005 ....................................................... 15
- Figure 4.    Selected OEC: Average ∆% Change in the CPI, 1999–2005 Ranked by X o /X T ............. 16
- Figure 5.    Oil Prices ($/B- Venezuelan Basket), and Ratio X o /X T, 960–2005 ................................. 16
- Figure 6.    Venezuela: Oil in the Economy ....................................................................................... 17
- Figure 7.    Venezuela: Rates of Growth of Real GDP, CPI, and Oil Prices ..................................... 19
- Figure 8.    Venezuela: Ratios of Central Bank Instruments of Monetary Control (CBIMC) ........... 19

### Appendices and appendix figures relevant to this unit
- Appendices (by title and page):
  - Appendix I.      Oil Exporting Countries-Selected Indicators.................................................................. 34
  - Appendix II.    Management of Oil Resources: Oil Funds and International Reserves .......................... 39
  - Appendix III.   Descriptive Statistics and Pair-Wise Correlations for Selected Economic Indicators.... 41
- Appendix Tables (titles and pages referenced):
  - A.I.1.  Classification and Evluation of Oil Exporting Countries, based on the Ratio of Oil Exports to GDP .................................................................................................. 34
  - A.I.2.  Oil Exporting Countries: Selected Indicators .............................................................. 36
  - A.I.3.  Oil Exporting Countries: Selected Indicators  ............................................................. 37
  - A.I.4.  Oil Exporting Countries: Selected Indicators .............................................................. 38
  - A.II.1. Venezuela: Macroeconomic Stabilization Fund (FEM and Investment Fund for Macroeconomic Stabilization (FIEM) .................................................................... 39
  - A.II.2. Central Bank of Venezuela-Balance Sheet ................................................................. 40
- Appendix Figures (titles and pages referenced):
  - A.III.1. Venezuela: Changes in Oil Prices, Oil Exports, Fiscal Revenue, and Primary Expenditures, 1960–2005 ..................................................................................... 42
  - A.III.2. Venezuela: Change in Primary Fiscal Expenditure, Non-Oil GDP Inflation, and Monetary Base, 1960–2005 .................................................................................. 43
  - A.III.3. Venezuela: Five-Year Moving Variences of Rates of Growth of Selected Indicators, 1965–2005 ............................................................................ 44
  - A.III.4. Venezuela: Overall Fiscal Balance and Domestic Primary Deficit (Percentage of GDP) ............................................................................................. 45

### Key thematic pointers (as indicated by headings and figure titles)
- Interaction of exchange rate regimes and nominal anchors with oil price movements for oil-exporting countries (OEC).
- Monetary aggregates (monetary base growth) and their relation to oil price changes.
- Fiscal positions: primary balances of general government in OEC and Venezuela.
- Inflation dynamics: CPI changes across OEC and Venezuela, including decomposition and rankings (1999–2005).
- Venezuela-focused analysis: role of oil in the economy, real GDP growth, CPI, oil price series, and central bank instruments of monetary control.
- Institutional arrangements for oil revenue management: Macroeconomic Stabilization Fund (FEM/FIEM) and management of international reserves.

*Source: _wp08142 - 1.    Exchange Rate and Nominal Anchor for Major OCD in 2005 (PDF chapter/section).*

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

### _wp08142 - References..............................................................................................................

### I. INTRODUCTION
- Purpose:
  - Present a simple framework to analyze fiscal dominance in oil exporting countries (OEC).
  - Test the oil dominance/fiscal dominance (OD/FD) hypothesis for Venezuela.
- Methods:
  - Vector Autoregression (VAR) models and Vector Error-Correction (VECM) models for empirical analysis.
- Key conclusion:
  - Evidence, both short-run and long-run, supports the validity of the OD/FD hypothesis during the period 1960–2005.

### II. FISCAL DOMINANCE IN OIL ECONOMIES — Conceptual points
- Central bank independence:
  - Independence is necessary for monetary policy to achieve price stability; fiscal dominance implies loss of that independence.
- Standard empirical approaches and limitations for OEC:
  - Approach 1: Analyze central bank accounts — extent to which changes in the monetary base are dominated by changes in central bank net credit to the government.
  - Approach 2: Fiscal sustainability via government’s intertemporal budget constraint (debt accumulation and primary fiscal deficits).
  - Limitation: In OEC, these approaches may fail because oil receipts alter central bank foreign assets and government deposits, masking fiscal-origin monetary expansions. Fiscal dominance may exist even with fiscal surpluses and low public debt.

### A. Oil dominance and fiscal dominance: Analytical framework
- Monetary base identity used (notation preserved in source):
  - MB = NIR + NDA (MBNIRNDAΔΔΔ=+ presented in source)
- Mechanisms in OEC:
  - Oil exports increase NIR when state-owned oil companies sell foreign exchange to the central bank and simultaneously increase government deposits at the central bank (decline in NDA) — net monetary base effect may be zero until deposits are spent.
  - Spending of oil receipts increases NDA and expands MB; if spending and NIR changes occur in same period, central bank accounts may show MB expansion due to NIR increase, obscuring fiscal origin.
- Net worth / net operating balance approach:
  - Government net worth (NW) and Net Operating Balance (NOBg) formulation (equations retained in source).
  - Oil extraction depletes a non-renewable government asset: oil-related receipts are a change of one asset for another and should not be treated as fiscal revenue in net worth framework.
  - Declining NW or an unsustainable path for NOB indicates fiscal dominance even when stock-of-debt indicators appear benign.
- Central bank accounting:
  - Central bank budget constraint and net worth shown (equations (1') and (2') in source).
  - Accounting identity for change in MB (equation (2") in source) expresses MBΔ in terms of changes in net foreign assets and net domestic credit and central bank net worth.
- Consolidated public sector:
  - Consolidated balance and net worth equations presented (equations (4) and (5) in source).
  - If government uses oil receipts to finance consumption (negative NOB), NW falls and this can be offset by MB increase and/or decline in foreign assets — central bank has limited ability to resist without incurring losses.

### Alternative indicator of OD/FD (simple practical measure)
- Define domestic fiscal balance (OBd) = revenues from domestic sources minus domestic expenditures.
- Ratio indicator:
  - γ = (G - Rno) / Ro  (equation (6) in source)
  - Where Ro is oil-related net receipts and Rno is non-oil revenue.
- Interpretation:
  - γ > 0 implies domestic deficit financed in part by oil receipts, indicating an initial monetary expansion of oil origin; larger γ implies higher possibility of OD/FD.
  - Indicator captures monetary impact of fiscal policy in OEC and complements net worth framework.

### B. Transmission mechanism (qualitative pathways)
- Under a fixed exchange rate or surrender requirements:
  - Increase in oil prices → government spending financed by oil receipts → nominal MB expansion → higher demand for tradables and higher nontradable prices → higher domestic inflation (real MB falls).
- Under fully flexible exchange rate:
  - Oil price increase → nominal exchange rate appreciation → smaller shock to domestic inflation via MB, but government spending can still raise nontradable prices and domestic inflation.
- Policy implication:
  - Volatility in oil prices transmits to output, inflation and current account depending on fiscal response and exchange rate regime; isolating fiscal policy from oil price fluctuations reduces transmission.

### III. OIL ECONOMIES: STYLIZED FACTS — cross-country observations (sample and indicators)
- Oil dominance metrics used:
  - Xo/X (oil exports to total exports), Xo/GDP (oil exports to GDP), Ro/RT (oil-related fiscal receipts to total fiscal revenue).
- Indicators analyzed:
  - Government total expenditure and net lending (GCG/GDP), primary balance (pbGG), rate of growth of nominal monetary base (%MB), %NEER, %REER, CPI change (inflation), ΔMB/GDP.
- Data:
  - Sample: 30 OEC, annual data 1979–2005, with sub-periods 1986–98 and 1999–2005.
  - Monetary base figures correspond to “reserve money” in IFS.
- Key empirical patterns reported:
  - Major OEC with highest oil dominance tended to adopt the exchange rate as nominal anchor; several maintain some form of pegged regime (Table 1 in source).
  - Average nominal growth of MB for OEC is closely related to changes in oil prices (Figure 1 in source).
  - Government expenditure correlates with degree of oil dominance (Xo/GDP); consumption spending implies NW reduction.
  - Trend: gradual decline in government expenditure overall and improved primary balance since 1999 alongside increased oil receipts (Figure 2).
  - Inflation shows some correlation with oil price changes, with considerable cross-country dispersion (Figures 3 and 4).
  - Some highly oil-dominant countries had lower inflation, except Angola, Nigeria and Venezuela which had high inflation and managed floats or exchange controls.

### IV. OIL DOMINANCE IN VENEZUELA — descriptive evidence (1960–2005)
- External and fiscal dominance:
  - Oil exports represent about 86 percent of total exports on average for 1960–2005.
  - Oil revenue share of total fiscal revenue fluctuated between 86 percent (1974) and 38 percent (1998).
  - Oil GDP share averaged around 22 percent for 1960–2005 (36 percent in 1974, 12 percent in 1986).
- Macroeconomic volatility:
  - Large volatility in inflation, real GDP growth, nominal NEER and monetary base (table 2 and figure 7 in source).
- Exchange rate history (table 3 in source):
  - 1964–1983 Fixed Single Rate (Pegged to the US dollar)
  - 1983–1989 Multiple Exchange Rates-Exchange Controls
  - 1989–1992 Floating
  - 1992–1994 Crawl
  - 1994–1996 Multiple Exchange Rates-Exchange Controls
  - 1996–2002 Bands
  - 2002–2003 Floating
  - 2003–  Fixed Single Rate-Exchange Controls
- Central bank instruments and costs:
  - Since 1990 the central bank placed its own instruments (CBIMC) to control liquidity; stock of AI fluctuated significantly during 1990–2005, sometimes exceeding 100 percent of MB.
  - Financial costs of placing CBIMC often exceeded returns on central bank foreign assets, an indicator of OD/FD (Figure 8 in source).
- Oil stabilization funds:
  - FIEM created in 1998; by end-2001 FIEM net worth was 50 percent of central bank foreign assets. FIEM replaced by FEM in 2003; only 11 percent of 2001 net worth preserved. No accumulation of oil-related resources since then despite oil price increases (footnote in source).

### A. Oil dominance / fiscal dominance: simple correlations for Venezuela (1960–2005)
- Variables used (in logs): oil prices (Venezuelan basket), oil exports (US$), fiscal revenue total and oil-related (local currency), primary fiscal expenditure, monetary base, non-oil nominal and real GDP, non-oil GDP deflator (proxy for domestic inflation).
- Contemporaneous correlations (rates of growth):
  - i) Changes in nominal oil exports mostly determined by oil price movements; volume changes have very little effect.
  - ii) Changes in fiscal oil revenue closely related to changes in oil exports; fiscal oil revenue highly correlated with total fiscal revenue and primary fiscal expenditure.
  - iii) Changes in fiscal primary expenditure highly correlated with nominal non-oil GDP, non-oil GDP deflator (inflation), and nominal MB.
  - iv) Correlation between growth of fiscal primary expenditure and non-oil real GDP is very small and negative.
- Volatility (five-year moving variance correlations):
  - i) Volatility of oil price growth and nominal oil export growth are very similar (volume unimportant).
  - ii) Volatility of fiscal oil revenue growth is highly correlated with volatility of nominal oil exports.
  - iii) Volatility of fiscal primary expenditure growth is highly correlated with volatility of oil-related variables (oil prices, oil exports, oil fiscal revenue).
  - iv) Variances of oil prices and MB converge more closely after 1985.
- Domestic vs overall deficit:
  - Domestic primary deficit substantially larger than overall deficit; five-period dynamic correlation between domestic and overall deficits (as % of GDP) is very low.
- Overall inference from correlations:
  - Correlation indicators do not contradict the presence of oil/fiscal dominance in Venezuela: fiscal variables are strongly related to oil variables and to MB and inflation measures.

*Source: _wp08142 - References............................................................................................................................... (content as provided).*

### Appendix III presents the matrices of pair-wise correlations for the rates of growth and the

### _wp08142 - Appendix III presents the matrices of pair-wise correlations for the rates of growth and the

### Summary observations from Appendix III
- Appendix III presents the matrices of pair-wise correlations for the rates of growth and the variances, as well as some illustrative charts.
- Key simple correlations reported:
  - A correlation with oil prices is noted for some variables.
  - The monetary base and domestic prices show a strong correlation with fiscal primary expenditures.
  - The simple correlation between non-oil real GDP and fiscal variables seems very weak.

### B. Oil Dominance/Fiscal Dominance Hypothesis for Venezuela — objectives and approach
- Objective: Test the validity of the oil dominance/fiscal dominance (OD/FD) hypothesis for the Venezuelan economy by assessing the existence of a close link between oil prices, the domestic primary fiscal deficit and the monetary base; also examine potential macroeconomic effects, particularly the connection between the monetary base and prices.
- Empirical analysis divided in two parts:
  1. Testing the oil dominance/fiscal dominance hypothesis.
  2. Testing the hypothesis of a relationship between the monetary base and prices.
- Methodology:
  - VAR and Vector Error-Correction (VECM) models are estimated.
  - Annual data for the period 1960–2005 are used for the relevant variables.
  - The VAR methodology is used because it allows simultaneous and dynamic estimation treating all relevant variables as potentially endogenous.
  - Olivo (2001) applied a similar VAR approach using the overall fiscal deficit and different monetary variables; that study generally provided weak support for fiscal dominance.

### The general VAR and VECM representations used
- General VAR representation (equation (7)):
  - ttptptt
    BxyAyAyε++++=
    −−
    ....
    11
  - Where t y and t x are vectors of endogenous and exogenous variables, p A A,....,1 and B are coefficient matrices, and t ε is the vector of innovations.
- VECM representation when variables are I(1) and cointegrated (equation (8)):
  - t
    t
    t
    p
    i
    itit
    eBxyyεα+++ΔΓ=Δ
    −
    ∧
    −
    =
    −
    ∑
    1
    1
    1
  - Equation (8) represents the Vector Error Correction (VECM), where 1−
    ∧
    t
    e embodies the deviation from long-run equilibrium in period (t-1).

### Model specification and identification details
- Endogenous variable vector includes fiscal, monetary and external sector variables. External variables included to control for effects of different exchange rate regimes implemented in Venezuela during the period.
- Oil price is treated as the only exogenous variable in the model.
- Variable ordering in the VAR (based on arguments in Sections II and III): 1) fiscal; 2) monetary; and 3) external sector variables.
- Lags chosen using the Akaike and Schwarz criteria and taking into account VAR system stability.

### Empirical expectations if the OD/FD hypothesis holds
- If the oil dominance/fiscal dominance hypothesis holds, empirical tests should detect a connection between oil prices, the primary domestic fiscal deficit and the monetary base.
- Specific implications to be observed:
  - i) Impulse-response (I-R) functions and variance decomposition (VD) consistent with a positive response of the monetary base to shocks to the domestic primary deficit and no response of the domestic primary deficit to shocks to the monetary base.
  - ii) One-way Granger-causality running from the domestic primary deficit to the monetary base.
  - iii) A cointegrating vector with a positive relationship between the domestic primary deficit and the monetary base.
- Note: nominal variables have been used in some specifications, both log-levels (model 1) and as deviations from trend (model 2), as well as real variables and ratios to GDP (model 3).

### Sample periods and exchange rate regime considerations
- Estimations obtained for:
  - Overall sample period: 1960–2005.
  - Two sub-periods: 1960–1982 and 1983–2005.
- Rationale for sub-periods:
  - 1960–1982 labeled the soft-peg period: the nominal exchange rate was kept fixed with very few modifications.
  - 1983–2005 labeled the variable-peg period: characterized by several variations of fixed and quasi-fixed exchange rate schemes, frequently combined with exchange rate controls.

### Notes on prior literature and availability of full results
- Olivo (2001) used overall fiscal deficit instead of domestic primary deficit and different monetary measures; VAR results provided weak support for fiscal dominance in that work (summarized findings: with annual data 1950–1998, overall fiscal deficit does not Granger-cause changes in net domestic credit or monetary base measured as ratios to trend output; with quarterly data 1983.I–1998.IV, overall fiscal deficit Granger-causes net domestic credit to trend output ratio, but neither overall fiscal deficit nor net domestic credit to trend output ratio Granger-cause inflation).
- The paper notes: "In the interest of shortening the paper, the results have not been presented here, but are available upon demand."

*Source: _wp08142 — Appendix III content as provided.*

### 2.  Testing the hypothesis of a relationship between the monetary base and prices.

### 2.  Testing the hypothesis of a relationship between the monetary base and prices.

### Overview and methodology
- Objective: test for a close relationship between prices and money, both in the long-run and short-run (1960–2005), using VAR-VECM, cointegration (Johansen), impulse-response (I-R) functions, variance decompositions (VD), and Granger-causality.
- Key variables (as used across models): nominal oil prices, domestic primary deficit, exchange rate, net international reserves, monetary base, CPI (price level).
- Unit-root testing: ADF and PP tests indicate many series are I(1) (log-levels) or I(0) depending on model specification; cointegration tests applied where appropriate.

### B.1 Testing the oil dominance—fiscal dominance hypothesis (relevant model details)
- Model 1 (nominal variables, logs):
  - Variables: LPON (nominal oil prices) exogenous; LDDPN (nominal domestic primary deficit); LNEXEP (nominal exchange rate, Bs./US$; end of period); LR (stock of nominal net international reserves, in U.S. dollars); LMB (nominal monetary base).
  - Johansen cointegration result (LMB and LDDPN): presence of a cointegrating vector (normalized) with LDDPN coefficient statistically significant, negative and close to one:
    - COINT1A = LMB - 1.075578*LDDPN
  - Johansen on LMB, LDDPN, LNEXEP, LR: one cointegrating equation but coefficient of LDDPN positive (opposite expected sign).
  - VECM specifications: first differences DLDDPN, DLMB, DLR, DLNEXEP; exogenous COINT1A(-1) and DLPON(-1); Cholesky ordering DLDDPN, DLMB, DLR, DLNEXEP.
  - Short-run I-R and VD (complete sample, 1960–2005):
    - A one standard deviation shock to DLDDPN has a positive effect on DLMB, explaining around 17 percent of DLMB forecast error variance in a twenty-period horizon.
    - The response of DLDDPN to shocks to DLMB is positive but very close to zero per VD analysis.
  - Granger/block causality:
    - Block Granger-causality indicates one-way Granger-causality from LDDPN to LMB.
    - Within the VECM there is no causality either way between DLDDPN and DLMB, but:
      - COINT1A(-1) coefficient is significantly different from zero and negative in the DLMB equation (expected sign).
      - COINT1A(-1) coefficient is positive but not statistically significant (at a 10 percent level) in the DLDDPN equation.
  - Role of oil prices:
    - DLPON(-1) coefficient is positive and statistically different from zero in the DLDDPN equation (at a 1 percent level) and in the DLR equation (at a 10 percent level), indicating oil price shocks affect fiscal and reserve behavior.

- Subsample analysis (soft-peg 1960–1982 vs variable-peg 1983–2005):
  - Common: shocks to DLDDPN positively affect DLMB in both sub-periods.
  - Quantitative differences:
    - Soft-peg: shocks to DLDDPN explain approximately 60 percent of DLMB forecast error variance.
    - Variable-peg: shocks to DLDDPN explain around 20 percent of DLMB forecast error variance.
  - Responses of DLDDPN to DLMB:
    - Soft-peg: DLDDPN responds positively but mildly to DLMB; shocks to DLMB explain close to 5 percent of DLDDPN forecast error variance.
    - Variable-peg: response of DLDDPN to DLMB is negative but close to zero per VD.
  - Granger-causality puzzle in soft-peg:
    - Evidence of Granger-causality from monetary base to domestic primary deficit (COINT1A(-1) positive and statistically different from zero in DLDDPN equation), but not the reverse; COINT1A(-1) not statistically significant in DLMB equation.
    - DLPON(-1) not statistically significant in DLDDPN and DLR equations for the soft-peg period.
  - Variable-peg aligns with full-sample findings:
    - Granger-causality from domestic primary deficit to monetary base (COINT1A(-1) negative and significantly different from zero in DLMB equation).
    - COINT1A(-1) positive but not statistically significant in DLDDPN.
    - DLPON(-1) remains positive and statistically different from zero in DLDDPN but not in DLR.
  - Block Granger-causality tests (both periods): do not reject null of no causality in either direction between DLDDPN and DLMB at standard significance levels.

- Model 2 (variables as deviations from Hodrick-Prescott trend; H-P gaps):
  - Variables: LPONG (nominal oil prices gap) exogenous; LDDPNG (nominal domestic primary deficit gap); LNEXEPG (nominal exchange rate gap); LRG (nominal net international reserves gap); LMBG (nominal monetary base gap).
  - Unit-root tests suggest all variables are I(0).
  - I-R and VD (ordering LDDPNG, LMBG, LRG, LNEXPG) give weak support for OD/FD compared to Model 1:
    - A shock to LMBG has a negative impact on LDDPNG.
    - VD: LMBG explains approximately 1.4 percent of LDDPNG forecast error variance in a 20-period horizon.
    - LMBG responds positively to shocks to LDDPNG.
    - VD: LDDPNG explains approximately 8.7 percent of LMBG forecast error variance in a 20-period horizon.
    - No Granger causality in either direction between LDDPNG and LMBG.
    - LPONG coefficient positive and statistically significant in equations for LDDPNG, LRG and LMBG.

- Model 3 (variables as percentage of GDP or in real terms):
  - Variables: DLPOR (first difference of log real oil price) exogenous; VDDPYN (change in nominal domestic primary deficit as % of GDP); DLREXEP (first difference of log real exchange rate); DLRREAL (first difference of log real net international reserves); SM (seigniorage: change in nominal monetary base as % of GDP).
  - Unit-root tests suggest all variables are I(0).
  - I-R and VD (ordering VDDPYN, SM, DLRREAL, DLREXEP) provide relatively weak support (weaker than Model 1 but more robust than Model 2):
    - VDDPYN response to shocks to SM is positive but very small.
    - VD: percentage of VDDPYN forecast error variance explained by SM is close to zero.
    - SM exhibits a positive reaction to shocks to VDDPYN.
    - VD: VDDPYN explains approximately 9.4 percent of SM forecast error variance in a 20-period horizon.
    - Block Granger-causality: SM does not Granger-cause VDDPYN, but VDDPYN Granger-causes SM (P-value = 0.072).
    - DLPOR coefficient positive and statistically significant in equations for VDDPYN, DLRREAL and SM.

### B.2 Testing the monetary base—prices relationship (main results)
- Cointegration testing between logarithm of monetary base (LMB) and price level (LCP, CPI):
  - ADF and PP tests: LMB and LCP are I(1).
  - Johansen test indicates one cointegrating equation, but residuals appear non-stationary on visual inspection.
  - Gregory and Hansen (G&H, 1996) cointegration tests (allowing for structural break) applied:
    - G&H tests reject null of no cointegration at the 1 percent level with a structural break in 1968.
    - Level shift model (model C in G&H) yields the most robust results and produces cointegrating equation (t-statistics in parenthesis):
      - LCP = -1.199748 + 0.972699*LMB - 1.058871*DU68
        - (-10.72852) (47.26916) (-7.12508)
      - DU68 dummy = 1 from 1969 on.
    - Coefficient of LMB is close to one.
    - Conclusion: taking structural break in 1968 into account, there is evidence of a long-run equilibrium relationship between monetary base and price level.

- VECM (five lags) for DLMB and DLCP with COINN2(-1) as exogenous (residuals from G&H cointegrating vector lagged one period):
  - I-R and VD results (1960–2005):
    - A one standard deviation shock to DLMB produces a positive response in DLCP, explaining approximately 20 percent of DLCP forecast error variance.
    - A one standard deviation shock to DLCP affects DLMB positively, explaining around 33 percent of DLMB forecast error variance.
  - Block Granger-causality tests: indicate two-way causality between DLCP and DLMB.
  - COINN2(-1) coefficient:
    - Has expected sign in both VECM equations.
    - Statistically significant only in the DLMB equation, implying Granger-causality from DLCP to DLMB.
  - Interpretation: evidence consistent with accommodative monetary policy, which may be a consequence of the OD/FD problem limiting independent monetary policy in Venezuela.

- Subsample (soft-peg vs variable-peg) VECM/VAR results for money-prices link:
  - Shocks to DLCP → DLMB:
    - Both periods: positive response of DLMB to DLCP shocks.
    - Variable-peg: DLCP shocks explain up to 40 percent of DLMB forecast error variance and decline to 31 percent in a ten-period horizon.
    - Soft-peg: DLCP shocks explain up to 21 percent in a gradual increase.
  - Shocks to DLMB → DLCP:
    - Both periods: positive impact on DLCP, stronger in variable-peg.
    - Soft-peg: DLMB shocks explain up to 15 percent of DLCP forecast error variance.
    - Variable-peg: DLMB shocks explain up to 29 percent of DLCP forecast error variance.
  - Block Granger-causality:
    - Soft-peg: absence of causality in either direction per block tests; however COINN2(-1) positive and statistically different from zero (at a 10 percent level) in the DLMB equation → Granger-causality from DLCP to DLMB. In DLCP equation COINN2(-1) negative but not statistically different from zero → no Granger-causality from DLMB to DLCP.
    - Variable-peg: bidirectional causality per block tests.
    - Note: for the variable-peg period it was not possible to estimate a stable VECM including COINN2(-1); a simple 5-lag VAR was estimated instead.

### Main conclusions (from the section)
- There is a close relationship between prices and money, both in the long-run and short-run, for the 1960–2005 period; results align with Olivo and Miller (2000) and Olivo (2002).
- Evidence supports the oil dominance/fiscal dominance (OD/FD) hypothesis in Venezuela during 1960–2005, with the nominal-variables-in-logs specification (Model 1) providing the strongest support and enabling clear VECM identification of short- and long-run effects.
- Key finding: a close long-run relationship between nominal domestic primary deficit and nominal monetary base—this relationship appears driven more by the variable-peg period (1983–2005); soft-peg period (1960–1982) yields mixed signals.
- Money-prices link:
  - A close connection between monetary base and prices exists in both short-run and long-run.
  - Short-run link is bidirectional; interpreted as evidence of accommodative monetary policy that may result from OD/FD constraints on monetary policy independence.
  - Relationship between money and prices is more robust in the variable-peg period.
- Oil prices (treated as exogenous) are statistically important in explaining fiscal domestic primary deficit across all models, reinforcing the OD/FD hypothesis.

*Source: _wp08142 - 2.  Testing the hypothesis of a relationship between the monetary base and prices.*

### APPENDIX I. OIL EXPORTING COUNTRIES-SELECTED INDICATORS

### APPENDIX I. OIL EXPORTING COUNTRIES-SELECTED INDICATORS

### Classification and Evolution of Oil Exporting Countries (Table A.I.1)
- Classification based on the ratio of oil exports to GDP with ranges: 0-10, 11-20, 21-30, 31-40, > 40.
- Country lists by year snapshots:
  - 1965: Algeria; Bahrain, Kingdom of; Saudi Arabia; Kuwait; Indonesia; Canada; Iran, I.R. of; Trinidad and Tobago; Libya; Qatar; Colombia; Venezuela, Rep. Bol.; Egypt; Malaysia; Mexico; Nigeria; Norway; Russia; Syrian Arab Republic; United Kingdom; Vietnam.
  - 1980: Cameroon; Ecuador; Angola; Algeria; Gabon; Canada; Egypt; Congo, Republic of; Bahrain, Kingdom of; Kuwait; Colombia; Indonesia; Venezuela, Rep. Bol.; Nigeria; Libya; Kazakhstan; Iran, I.R. of; Trinidad and Tobago; Oman; Mexico; Malaysia; Qatar; Norway; Syrian Arab Republic; Saudi Arabia; Russia; United Arab Emirates; United Kingdom; Vietnam.
  - 1995: Cameroon; Iran, I.R. of; Algeria; Oman; Angola; Canada; Norway; Libya; Saudi Arabia; Bahrain, Kingdom of; Colombia; Syrian Arab Republic; Qatar; Brunei Darussalam; Ecuador; Trinidad and Tobago; United Arab Emirates; Congo, Republic of; Egypt; Venezuela, Rep. Bol.; Nigeria; Gabon; Indonesia; Kuwait; Kazakhstan; Malaysia; Mexico; Russia; United Kingdom; Vietnam.
  - 2005: Cameroon; Ecuador; Iran, I.R. of; Kazakhstan; Algeria; Canada (IT); Norway (IT); Qatar; Angola; Colombia (IT); Russia; United Arab Emirates; Bahrain, Kingdom of; Egypt; Syrian Arab Republic; Venezuela, Rep. Bol.; Brunei Darussalam; Indonesia; Vietnam; Congo, Republic of; Malaysia; Gabon; Mexico (IT); Kuwait; United Kingdom (IT); Libya; Nigeria; Oman; Saudi Arabia; Trinidad and Tobago.
- Source note: selection of OEC based on list of countries presented by OPEC as oil exporting countries.

### Definition of Variables (Tables A.I.2–A.I.4)
- Xo/XT: Ratio of oil exports to total exports of goods (both in U.S. dollars).
- Xo/GDP: Ratio of oil exports to GDP (both in US dollars).
- Ro/RT: Ratio of oil-related revenue to total revenue of the public sector.
- ∆%RM: Growth rate of nominal reserve money.
- Inflation: Percentage change in the consumer price index.
- ∆RM/GDP: Ratio of change in reserve money to GDP.
- ∆%NEER: Percentage change in the nominal effective exchange rate, defined as foreign currency per unit of local currency (- depreciation).
- ∆%REER: Percentage change in the real effective exchange rate, defined as foreign currency per unit of local currency (- depreciation).
- GCG/GDP: Ratio of central government total expenditure and net lending to GDP.
- pb_GG: Ratio of general government primary balance to GDP.

### Selected Indicators — Cross-Country Averages and Standard Deviations (Table A.I.2; All figures in %)
- Average 1999-2005 and Standard deviation 1999-2005 provided for multiple indicators per country. Selected exact values (Average 1999-2005):
  - Angola: Xo/XT 89.6; Xo/GDP 67.0; ∆% MB 190.9; Inflation 143.6; GCG/GDP 51.7; pb_GG (%GDP) -2.5; Real GDP growth 8.4.
  - Bahrain, Kingdom of: Xo/XT 73.5; Xo/GDP 51.9; ∆% MB 19.7; Inflation 0.4; ∆% NEER 1.8; ∆% REER -1.0; GCG/GDP 28.4; Real GDP growth 5.7.
  - Brunei Darussalam: Xo/XT 89.1; Xo/GDP 56.9; Inflation 0.2; GCG/GDP 54.5; Real GDP growth 2.3.
  - Canada: Xo/XT 4.8; Xo/GDP 1.7; ∆% MB 5.2; Inflation 2.3; ∆% NEER 0.2; ∆% REER 2.6; GCG/GDP 16.2; pb_GG (%GDP) 3.6; Real GDP growth 3.4.
  - Colombia: Xo/XT 27.3; Xo/GDP 4.4; Ro/RT 9.9; ∆% MB 18.4; Inflation 7.5; ∆% NEER 1.1; ∆% REER -3.3; GCG/GDP 19.9; pb_GG (%GDP) 1.4; Real GDP growth 2.3.
  - Congo, Republic of: Xo/XT 88.4; Xo/GDP 65.8; ∆% MB 28.8; Inflation 2.4; ∆% NEER 1.6; ∆% REER 1.6; GCG/GDP 29.5; pb_GG (%GDP) 8.2; Real GDP growth 3.6.
  - Ecuador: Xo/XT 44.9; Xo/GDP 11.4; Ro/RT 26.2; ∆% MB -0.5; Inflation 30.2; ∆% REER -14.1; GCG/GDP 3.2; pb_GG (%GDP) 8.3; Real GDP growth 3.3.
  - Kazakhstan: Xo/XT 50.2; Xo/GDP 22.5; ∆% MB 36.8; Inflation 8.1; ∆% NEER 1.8; GCG/GDP 13.5; pb_GG (%GDP) 2.3; Real GDP growth 9.2.
  - Mexico: Xo/XT 10.5; Xo/GDP 2.8; Ro/RT 33.6; ∆% MB 15.0; Inflation 7.2; ∆% NEER 0.6; GCG/GDP 17.5; pb_GG (%GDP) 0.8; Real GDP growth 2.8.
  - Norway: Xo/XT 42.8; Xo/GDP 14.1; Inflation 2.0; ∆% REER 1.0; GCG/GDP 34.7; pb_GG (%GDP) 9.5; Real GDP growth 2.3.
  - Russia: Xo/XT 37.4; Xo/GDP 12.6; Ro/RT 19.5; ∆% MB 41.9; Inflation 25.9; ∆% NEER 4.6; ∆% REER -8.7; GCG/GDP 15.1; pb_GG (%GDP) 5.3; Real GDP growth 6.7.
  - Saudi Arabia: Xo/XT 88.8; Xo/GDP 38.0; ∆% MB 7.4; Inflation -0.2; ∆% NEER 0.6; ∆% REER -1.0; GCG/GDP 33.2; pb_GG (%GDP) n.a.; Real GDP growth 3.5.
  - United Kingdom: Xo/XT 7.8; Xo/GDP 1.4; ∆% MB 6.8; Inflation 2.4; ∆% NEER 0.2; ∆% REER 0.0; GCG/GDP 37.1; pb_GG (%GDP) 0.3; Real GDP growth 2.7.
  - Venezuela, Rep. Bol.: Xo/XT 82.1; Xo/GDP 24.8; Ro/RT 46.3; ∆% MB 30.3; Inflation 20.5; ∆% NEER 1.8; ∆% REER -15.4; GCG/GDP 25.4; pb_GG (%GDP) 3.1; Real GDP growth 1.9.
- Standard deviations 1999-2005 are provided alongside averages in the table for each country and indicator (not exhaustively repeated here).

### Historical Averages: 1986–98 and 1979–2005 (Tables A.I.3 and A.I.4; All figures in %)
- Tables present Average 1986-98 and Average 1979-2005 with corresponding standard deviations for the same indicator set.
- Selected exact values (Average 1986-98):
  - Algeria: Xo/XT 89.3; Xo/GDP 19.9; ∆% MB 13.7; Inflation 16.8; ∆% NEER 2.5; ∆% REER -16.4; GCG/GDP 30.6; pb_GG (%GDP) n.a.; Real GDP growth 1.1.
  - Nigeria: Xo/XT 94.4; Xo/GDP 35.4; Ro/RT 64.8; ∆% MB 32.7; Inflation 32.4; ∆% NEER 3.6; ∆% REER -19.4; GCG/GDP 15.7; pb_GG (%GDP) n.a.; Real GDP growth 3.0.
- Selected exact values (Average 1979-2005):
  - Algeria: Xo/XT 90.1; Xo/GDP 24.8; ∆% MB 16.6; Inflation 11.1; ∆% NEER 10.8; GCG/GDP n.a.; pb_GG (%GDP) 2.2; Real GDP growth 10.7.
  - Nigeria: Xo/XT 94.0; Xo/GDP 35.9; Ro/RT 64.2; ∆% MB 24.9; Inflation 22.7; ∆% NEER 2.6; ∆% REER 12.2; GCG/GDP 19.7; pb_GG (%GDP) n.a.; Real GDP growth 3.0.
- Note: Numerous "n.a." entries throughout the tables indicate unavailable data for specific indicators/countries.

### Source Note for Tables A.I.2–A.I.4
- Source: The data is obtained from the IMF data base. The selection of OEC is based on the list of countries presented by OPEC as oil exporting countries, since it provides a wide range for degree of oil dominance.

---

### APPENDIX II. MANAGEMENT OF OIL RESOURCES: OIL FUNDS AND INTERNATIONAL RESERVES

### Venezuela: FIEM, FEM, and FONDEN — institutional timeline and mechanics
- FIEM (Investment Fund for Macroeconomic Stabilization)
  - Created in 1998, initial contribution in December 1999.
  - Main objective: avoid impact of fluctuations in oil-related receipts on fiscal equilibrium, foreign exchange, and money markets.
  - Original threshold: 5-year moving average of oil-related receipts.
  - Later rule: transfer of 6 percent of oil-related receipts (intended implementation 2003–2007).
  - Resources made available to cover budgetary needs in 2002 and 2003; about 76 percent of FIEM’s net worth was transferred to the public sector in 2003 despite increasing oil prices.
  - Remaining net worth transferred to Macroeconomic Stabilization Fund (FEM) created in December 2003.
- FEM (Macroeconomic Stabilization Fund)
  - Threshold: three-year moving average of oil-related receipts.
  - No contributions made after 2003.
- FONDEN (Development Fund)
  - Created in July 2005 with resources transferred from the Central Bank’s international reserves.
  - Transfers recorded in the Central Bank’s balance sheet as a domestic asset.

### Venezuela: FIEM and FEM flows and net worth (Table A.II.1; In billions of US dollars)
- Contributions (by year): 1999 2; 2000 154,246? (table formatting issues in source); 2001 62,300?; 2002 419,718?; 2003 0? (source table presents concatenated numbers; net worth series follows).
- Return on financial investments (by year): 1999 131; 2000 235; 2001 79; 2002 1410; 2003 2236; 2004 527 (as shown).
- Withdrawals (by year): 1999 8; 2000 943,870?; 2001 2,169; 2002 6,933.
- Expenses: 2; 21; 16 (as shown).
- Net worth (by year): 1999 2; 2000 154,590?; 2001 6,229?; 2002 2,856; 2003 700; 2004 710; 2005 732; 2006 768; 2006 (duplicate) 768.
- Notes in table:
  - 1 Initial contribution on December 29, 1999.
  - 2 As of December 23, 2003, when it stop operations.
  - 3 The FIEM became FEM in December 2003.
- Important: Table A.II.1 in the source PDF contains formatting and alignment issues; numbers are reported exactly as they appear in the table cells.

### Central Bank of Venezuela — Balance Sheet excerpts and transfers to FONDEN (Table A.II.2; In billions of Bolivars)
- Selected dates with exact figures (ASSETS):
  - 31-Aug-05 ASSETS 77,278
  - 30-Sep-05 ASSETS 80,972
  - 31-Oct-05 ASSETS 85,338
  - 30-Nov-05 ASSETS 85,424
  - 31-Dec-05 ASSETS 86,990
  - 31-Mar-06 ASSETS 90,955
  - 31-Dec-06 ASSETS 105,352
- Foreign Assets (selected): 31-Aug-05 95,945; 31-Dec-06 91,566; 31-Mar-06 75,726; 30-Sep-05 72,678.
- Domestic Assets (selected): 31-Aug-05 1,333; 30-Sep-05 8,295; 31-Oct-05 12,930; 30-Nov-05 15,056; 31-Dec-05 15,134; 31-Mar-06 15,229; 31-Dec-06 13,786.
- Transfers to Fonden (as recorded under Other Domestic Assets / Public Sector Bonds / Transfers to Fonden):
  - 31-Aug-05 Transfers to Fonden 0
  - 30-Sep-05 Transfers to Fonden 6,079
  - 31-Oct-05 Transfers to Fonden 10,309
  - 30-Nov-05 Transfers to Fonden 12,453
  - 31-Dec-05 Transfers to Fonden 12,453
  - 31-Mar-06 Transfers to Fonden 12,453
  - 31-Dec-06 Transfers to Fonden 12,453
- LIABILITIES AND NET WORTH (selected):
  - LIABILITIES: 31-Aug-05 62,963; 30-Sep-05 67,267; 31-Dec-06 97,180.
  - Foreign Liabilities: 30-Sep-05 9,858; 31-Dec-06 14,948.
  - Domestic Liabilities: 31-Aug-05 52,539; 30-Sep-05 57,409.
  - Monetary base: 31-Aug-05 18,783; 30-Sep-05 19,522; 31-Dec-05 23,087; 31-Dec-06 44,795? (table shows "44795" formatting).
  - Other domestic liabilities (of which Securities and Public Sector Deposits) listed with exact figures across dates.
  - NET WORTH: 31-Aug-05 14,315; 30-Sep-05 13,706; 31-Dec-06 8,171.
  - Adjusted Net Worth (defined as net worth minus transfers to Fonden): 30-Sep-05 7,627; 31-Oct-05 3,684; 30-Nov-05 1,868; 31-Dec-05 2,267; 31-Mar-06 -191; 31-Dec-06 -4,282; 31-Aug-05 14,315.
- Notes:
  - 1 Includes cumulative balance in operations.
  - 2 According to the Partial Reform of the Central Bank Charter of July 20, 2005, the Central Bank should transfer to a public fund the excess international reserves over what has been determined as the "adequate level."
  - 3 Adjusted Net Worth defined as net worth minus transfers to Fonden, which are recorded as an asset.
- Source: Central Bank of Venezuela.

---

### APPENDIX III. DESCRIPTIVE STATISTICS AND PAIR-WISE CORRELATIONS FOR SELECTED ECONOMIC INDICATORS

### Descriptive statistics (sample size and summary)
- Sample: 1960–2005; Included observations: 46 (for many series).
- Selected series descriptive statistics (exact values as reported):
  - RXPXT: Mean 86.99; Median 91.11; Maximum 95.49; Minimum 68.78; Std. Dev. 7.72; Skewness -0.65; Kurtosis 2.12; Jarque-Bera 4.67; Probability 0.10; Sum 4,001.55; Sum Sq. Dev. 2,678.81; Observations 46.
  - RIFPIFTRIFPGFPYP_YN: Mean 61.31; Median 62.23; Maximum 86.25; Minimum 37.78; Std. Dev. 11.27; Skewness 0.08; Kurtosis 2.53; Jarque-Bera 0.48; Probability 0.79; Sum 2,820.27; Sum Sq. Dev. 5,713.75; Observations 46.
  - Additional combined series reported with exact numeric summaries: Mean 63.47; Mean 21.78; Median 61.05; Median 21.75; Maximum 97.65; Maximum 36.19; Minimum 35.93; Minimum 11.95; Std. Dev. 15.10; Std. Dev. 5.30; Skewness 0.44; Skewness 0.38; Kurtosis 2.60; Kurtosis 2.98; Jarque-Bera 1.81; Jarque-Bera 1.09; Probability 0.40; Probability 0.58; Sum 2,919.76; Sum 1,001.88; Sum Sq. Dev. 10,256.16; Sum Sq. Dev. 1,265.85; Observations 46.

### Pair-wise correlations of rates of change (exact reported matrix excerpts)
- Correlations among first differences (DL...) — selected exact entries:
  - DLPON with DLXP: 0.94; with DLIFP: 0.70; with DLIFT: 0.63; with DLGFP: 0.44; with DLMB: 0.35.
  - DLXP with DLIFP: 0.78; with DLIFT: 0.72; with DLGFP: 0.50; with DLMB: 0.41.
  - DLIFP with DLIFT: 0.96; with DLGFP: 0.77; with DLYNP: 0.51; with DLDYNP: 0.50; with DLMB: 0.62.
  - DLGFP with DLYNP: 0.64; with DLDYNP: 0.62; with DLMB: 0.63.
  - DLYNPR with many series: correlations small/near zero (e.g., DLPON 0.00; DLXP 0.06).
- Correlations listed exactly in matrix form in the source for all pairs.

### Pair-wise correlations of five-year moving variances of growth rates (exact reported matrix excerpts)
- VDLPON with VDLXP: 0.97; with VDLIFP: 0.72; with VDLGFP: 0.69; with VDLMB: 0.16.
- VDLXP with VDLIFP: 0.71; with VDLIFT: 0.75; with VDLGFP: 0.73; with VDLMB: 0.11.
- VDLIFP with VDLIFT: 1.00; with VDLGFP: 0.74; with VDLYNP: 0.68; with VDLDYNP: 0.56; with VDLMB: 0.32.
- VDLGFP with VDLIFT: 0.90; with VDLYNP: 0.20; with VDLDYNP: 0.03.
- VDLYNP with VDLDYNP: 0.87; with VDLMB: 0.44.
- Full matrix provided in source.

### Graphical figures and interpretations included in source (Figures A.III.1–A.III.4)
- Figure A.III.1: Venezuela — Changes in Oil Prices (DLPON), Oil Exports (DLXP), Oil Fiscal Revenue (DLIFP), and Primary Expenditures (DLGFP), 1960–2005 (scatter/time-change panels shown).
- Figure A.III.2: Venezuela — Change in Primary Fiscal Expenditure (DLGFP) versus Non-Oil GDP (DLYNP), Non-Oil GDP Deflator (DLDYNP), Monetary Base (DLMB), and Non-Oil Real GDP (DLYNPR), 1960–2005.
- Figure A.III.3: Venezuela — Five-Year Moving Variances of Rates of Growth of Selected Indicators, 1965–2005 (VDLPON, VDLXP, VDLIFP, VDLGFP, VDLMB panels).
- Figure A.III.4: Venezuela — Overall Fiscal Balance and Domestic Primary Deficit (Percentage of GDP), sample 1960–2005; reported lead-lag correlations between DDP_AYN and DFF_AYN across lags/leads with exact values:
  - Lag/Lead i = 0: 0.0320 and 0.0320
  - i = 1: -0.2478 and 0.1958
  - i = 2: -0.2273 and 0.1858
  - i = 3: -0.2433 and 0.0489
  - i = 4: -0.1490 and 0.0625
  - i = 5: -0.1523 and -0.0353
- Note in figure: Correlations are asymptotically consistent approximations; sample 1960–2005; included observations 46.

*Source: _wp08142 - APPENDIX I. OIL EXPORTING COUNTRIES-SELECTED INDICATORS (data and tables reproduced exactly as presented in the source PDF).*

### REFERENCES

### _wp08142 - REFERENCES

### Monetary policy, inflation targeting, and rules
- Agenor, Pierre-Richard, 2002, “Monetary Policy under Flexible Exchange rates: An Introduction to Inflation Targeting.” (Washington: World Bank).
- Ball, Laurence, 1999, “Policy Rules for Open Economies,” in Monetary Policy Ruled, edited by John Taylor (University of Chicago Press: Chicago).
- Ball, Christopher P., and Javier Reyes, 2004, “Inflation Targeting or Fear of Floating in Disguise: The Case of Mexico in ” International Journal of Finance and Economics,” Vol. 9, pp. 49–69.
- Carare, Alina, Schaechter Andrea, Stone Mark, and Mark Zelmer, 2002, “Establishing Initial Conditions in Support of Inflation Targeting,” IMF Working Paper 02/102 (Washington: International Monetary Fund).
- McCallum, Bennett, and Edward Nelson, 2004, “Targeting vs. Instrument Rules for Monetary Policy,” NBER Working Paper No. 10612 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Soikkeli, Jarkko, 2002, “The Inflation Targeting Framework in Norway,” IMF Working Paper No. 02/184 (Washington: International Monetary Fund).
- Svensson, Lars E.O., 2001, “Inflation Targeting: Should it be Modeled as an Instrument Rule or a Targeting Rule?” Short version of a paper presented at the EEA 2001 Annual Congress (Lausanne).
- Svensson, Lars E.O., 1997, “Inflation Forecast Targeting: Implementing and Monitoring Inflation Targets,” European Economic Review Vol. 41, pp. 1111–46.
- Taylor, John, 1993, “Discretion versus Policy Rules in Practice.” Conference Series on Public Policy, Vol. 39, pp. 195–214 (Carnegie-Rochester).
- Woodford, Michael, 2001, “Fiscal Requirements for Price Stability,” Journal of Money, Credit, and Banking, Vol. 33(3), pp. 669–728.

### Exchange rates, fear of floating, and country studies
- Calvo, Guillermo, and Carmen Reinhart, 2000, “Fear of Floating.” NBER Working Paper No. 7993 (Cambridge, Massachusetts: National Bureau of Economic Research).
- Reinhart, Carmen, 2000, “The mirage of floating exchange rates,” American Economic Review, Vol. 90(2), pp. 65–70.
- Ball, Christopher P., and Javier Reyes, 2004, “Inflation Targeting or Fear of Floating in Disguise: The Case of Mexico in ” International Journal of Finance and Economics,” Vol. 9, pp. 49–69.
- Olivo, Víctor R., “La relación de largo plazo entre la base monetaria y el nivel de precios en Venezuela: 1950–2002,” Serie Documentos de Trabajo No. 57 (Venezuela: Central Bank of Venezuela).
- Olivo, Víctor R., 2001, “Fiscal Dominance in Open Economies with Managed Exchange Rate Regimes: A Theoretican and Empirical Analysis,” Paper presented at a seminar at the Central Bank of Venezuela (Unpublished).
- Olivo, Victor, and Stephen Miller, 2000, “An Analysis of the Long-Run Relationship between Money, Nominal GDP, and the Price Level in Venezuela: 1950–96,” (Connecticut: University of Connecticut).

### Fiscal policy, oil-producing countries, and fiscal stability
- Barnett, Steven, and Rolando Ossowski, 2002, “Operational Aspects of Fiscal Policy in Oil-Producing Countries,” IMF Working Paper 02/177 (Washington: International Monetary Fund).
- Da Costa, Mercedes, and Hugo Juan-Ramon, 2006, “The Net Worth Approach to Fiscal Analysis: Dynamics and Rules,” IMF Working Paper 06/17 (Washington: International Monetary Fund).
- Sargent, Thomas J., and Neil Wallace, 1985, “Some Unpleasant Monetarist Arithmetic,” Federal Reserve Bank of Minneapolis Quarterly Review, Vol. 9(1), pp. 15–31.
- Shabsigh, Ghiath, and Nadeem Ilahi, 2007, “Looking Beyond the Fiscal: Do Oil Funds Bring Macroeconomic Stability?” IMF Working Paper 07/96 (Washington: International Monetary Fund).
- Woodford, Michael, 2001, “Fiscal Requirements for Price Stability,” Journal of Money, Credit, and Banking, Vol. 33(3), pp. 669–728.
- Benigno, Pierpaolo, and Michael Woodford, 2003, “Optimal Monetary Policy and Fiscal Policy: A Linear-Quadratic Approach,” NBER Working Paper No. 9905 (Cambridge, Massachusetts: National Bureau of Economic Research).

### Econometric methods and cointegration tests
- Gregory, Allan, and Bruce Hansen, 1996, “Residual Base Test for Cointegration in Models with Regime Shifts,” in Journal of Econometrics, No. 70.
- McCallum, Bennett, and Edward Nelson, 2004, “Targeting vs. Instrument Rules for Monetary Policy,” NBER Working Paper No. 10612 (Cambridge, Massachusetts: National Bureau of Economic Research).

* _wp08142 - REFERENCES_*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2008/_wp08142.pdf_