## 1. Dissolution of Currency Union: World Sample

## Source details

**Canonical URL:** [1. Dissolution of Currency Union: World Sample](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0631.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0631.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0631.pdf.json)

---

### Introduction and motivation
- Gravity models (Pöyhönen, 1963; Tinbergen, 1962) explain more than half of the variation in trade and have been micro-founded in subsequent work.
- Recent empirical work uses gravity models to analyze the impact of currency arrangements on bilateral trade (Rose, 2000; Frankel and Rose, 2002; Rose and van Wincoop, 2001; Glick and Rose, 2002).
- Frankel and Rose (2002) conclude that two countries that share the same currency trade "roughly three times as much" as they would if they used different currencies.
- Paper focuses on Africa because of rich experience with currency unions despite limited integration in world trade.

### Theoretical channels through which currency unions affect trade
- Currency unions reduce transaction costs between trading partners.
- They create larger markets enabling potential gains from economies of scale and production efficiency (given factor mobility and flexible wages).
- They bring macroeconomic stability by signaling a commitment to reduce inflation and enhance monetary credibility.
- They reduce uncertainty and eliminate exchange rate risk, encouraging investment and facilitating capital mobility.
- Potential losses arise from asymmetric shocks and insufficient shock absorbers (factor mobility and transfers).

### Empirical approach and scope
- Augments the gravity model of Glick and Rose (2002) to investigate:
  - Impact of currency unions on bilateral trade (world and Africa samples).
  - Possible trade diversion effects.
  - Effect of the duration of currency union membership on bilateral trade.
  - Impact of currency unions on price and output co-movements and trade stability.
- Baseline augmented gravity specification includes CU_ij, ln(Y_i Y_j), ln(Pop_i Pop_j), ln(D_ij), Cont_ij, Lang_ij, FTA_ij, ComNat_ij, ComCol_ij, Col_ij, ln(Area_i Area_j).
- Trade diversion indicator: dummy = 1 if the two countries are not in the same currency union but (at least) one is in a currency union with another country.
- Membership duration: include "years" — number of years a pair shares a common currency; include quadratic term to capture nonlinearity.
- Co-movements measured from residuals of second-order autoregressions on ln(P_it), ln(Y_it); generate ij_VP and ij_VY statistics.
- Trade stability measured as coefficient of variation of log real trade over two periods (1950-1976 and 1976-2003); robustness checks include maximal absolute deviation, mean absolute deviation, and standard deviation of residuals from gravity on exports in levels.
- Zero-trade observations treated using log(constant + X_ij) and Tobit; Santos Silva and Tenreyro (2004) pseudo-maximum-likelihood noted as robustness alternative.

### Data and sample
- World sample: 217 countries and political units over 1948-2002.
- Africa sample: 49 countries where at least one of the ij pair is African; intra-Africa sample (both i and j African); Africa-ROW sample (exactly one country African).
- Bilateral trade: IMF’s Direction of Trade Statistics (real U.S. dollars).
- GDP: University of Pennsylvania World Tables 6.1 and World Bank’s World Development Report.
- Colonial past, distance, language: CIA World Factbook 2004.
- Currency unions (1998-2002): IMF’s Exchange Arrangements and Exchange Restrictions Annual Report.
- Prices (PPP of GDP): Penn World Tables 6.1.
- Terms of trade: IMF’s World Economic Outlook 2004.
- Sample coverage extended relative to Glick and Rose (2002): 217 units, period 1948-2002, inclusion of FTAs operating in Africa.

### Key empirical findings (robust across estimation techniques)
- Magnitude of trade effect:
  - For both the world and Africa samples, countries belonging to the same currency union trade "about two times more" with each other than comparable countries that do not share a currency.
  - Cross-section replication (world sample) finds sharing a common currency trades more than "two and a half" times compared with non-union pairs.
  - Allowing FTA dummy to include more agreements (including African FTAs) yields currency union coefficient "2.3".
  - Panel fixed-effects (country-pair intercepts) with country effects yields estimated trade-boosting factors: World: "1.8"; Africa: "1.7".
  - Baseline specification with intra-Africa and Africa-ROW dummies: currency union effect becomes "2.2" times greater; intra-Africa trade-generating effect is "1.8" (panel).
  - Overall currency union trade-generating effect remains about "2.1" and "2.0" for the world and Africa, respectively, in some specifications.
- Trade composition and stability:
  - Evidence that currency unions are associated with trade creation rather than trade diversion; trade diversion dummy shows positive coefficients in fixed-effects specifications for both world and Africa.
  - Intra-Africa trade creation effect is about five times higher than the Africa-ROW trade creation effect.
  - Currency unions make trade more stable in initial OLS specifications (currency union coefficient on trade variability: -0.05 ** and -0.06 ** in OLS); coefficient falls and becomes statistically insignificant once country-pair fixed-effects are included.
  - Mean stability: Africa sample = 0.22; world = 0.18.
- Price and output co-movement:
  - Currency unions associated with increased co-movement of prices for both samples.
  - Currency unions not systematically associated with increased co-movement of outputs.
  - Marginal effect on price co-movements higher for Africa: coefficient of currency union for Africa in price co-movement regressions = 0.04 with country fixed-effects; mean co-movements of prices for the Africa subsample = –0.15.
- Duration of membership:
  - The longer a country participates in a currency union, the greater the benefit to bilateral trade; relationship is nonlinear with diminishing returns.
  - One additional year of membership increases trade by 1.4 and 2.0 percent in the world and Africa samples, respectively (Table 5).
  - World sample: after 10 years, impact on trade is a 15 percent increase [(1 + 0.014)^10]; doubling trade would take about 67 years.
  - Africa sample: after 10 years trade increases by an estimated 22 percent; trade estimated to take 45 years to double.
  - Exits from a currency union are associated with declines in bilateral trade appearing only after extended periods: significant negative effects appear in periods of 16-27 years (world) and 16-21 years (Africa) after leaving; after ten years of currency union exit, bilateral trade declines by about 35 percent for Africa and 55 percent for the world.
- Sensitivity to zero-trade observations:
  - Not eliminating zero observations increases the world and Africa samples by about 45 and 31 percent, respectively.
  - Estimated γ (currency union trade-generating effect) from fixed-effects estimates:
    - World sample: lies in the narrow range of about (1.17, 1.89).
    - Africa sample: lies in the narrow range of about (1.03, 1.85).
    - When including only statistically significant estimates and excluding corrections for zero-trade observations, ranges tighten to (1.59, 1.89) for the world and (1.68, 1.85) for Africa.

### Benchmark coefficient highlights (selected exact values from tables)
- Currency union coefficient examples (coefficient (robust t-statistic)):
  - Table 1 (benchmark): Currency union: 1.00 ** (8.55); 0.84 ** (7.75); 0.99 ** (6.57); 0.54 ** (13.37); 0.58 ** (14.35); 0.55 ** (8.18).
  - Table 2 (Africa): Currency union: 0.79 ** (7.03); 0.53 (1.18); 0.55 ** (2.69); 0.58 ** (14.35); 0.62 ** (4.56); 0.53 ** (6.35).
  - Table 3 (trade diversion tests): Currency union: 0.97 ** (8.77); 1.05 ** (6.81); 0.61 (1.38); 0.59 ** (2.58); 0.73 ** (17.74); 0.70 ** (10.29).
  - Table 4 (sensitivity): OLS γ = 0.84 ** (7.75) → Effect 2.3; Fixed Effects γ = 0.58 ** (14.35) → Effect 1.8; Tobit γ = 0.79 ** (20.30) → Effect 2.2.
  - Table 5 (duration): Years in currency union coefficients include 0.03 ** (7.47); 0.07 ** (7.49); quadratic term (Years)^2: -0.001 ** (4.86).
  - Table 6 (co-movements): Currency union on prices: 0.06 ** (9.07); 0.07 ** (9.35); with fixed-effects 0.03 ** (11.19); 0.04 ** (11.19). Currency union on outputs: not significant (e.g., -0.002 (0.61)).
  - Table 7 (stability): Currency union: -0.05 ** (4.36); -0.06 ** (4.40) in OLS; becomes -0.02 (0.95) and -0.02 (0.64) with fixed-effects.

### Robustness and identification issues
- Omitted-variable bias addressed by estimating OLS cross-section, fixed-effects (country-pair intercepts), random effects, and robust fixed-effects "within" estimator; inclusion of year effects and clustering of errors within country pairs.
- Endogeneity of currency union membership acknowledged; authors treat currency unions as exogenous citing prior work suggesting reverse causality unlikely to fully explain results.
- Model misspecification addressed via:
  - Aggregation checks (sum of logs vs. log of sums).
  - Time dummies to correct deflation issues.
  - Non-linear terms for output and output per capita.
  - Treatments for zero-trade observations (log(constant + X_ij), Tobit).
- Sensitivity analyses show main conclusions on trade generation, duration effects, price co-movements, and limited impact on output co-movements are robust to methodological variants, with caveat that inclusion/exclusion of zero-trade observations affects estimated magnitudes.

### Policy-relevant implications and conclusions
- Currency union membership substantially increases bilateral trade: members trade about twice as much with each other compared with similar nonmembers.
- Trade increase from currency unions does not appear to come with trade diversion from nonmembers; evidence points to trade creation.
- Duration matters: longer membership yields larger trade benefits, but with diminishing returns.
- Currency unions reduce trade volatility (in OLS) and increase price co-movements, but have no significant effect on output co-movements.
- African countries stand to benefit as much from currency union membership as countries in the rest of the world; marginal effects and mechanisms of transmission may vary across regions.
- Given the low trade base in Africa and the protracted period necessary after establishment of a currency union to increase trade significantly, currency union membership is unlikely to be a primary tool to achieve high levels of trade in Africa.
- Methodology does not resolve causal channels linking currency unions, trade, growth, and poverty reduction; omitted political and institutional factors may influence both union membership and trade outcomes and warrant further investigation.

*Source: Excerpt from IMF Working Paper _wp0631 (selected sections on currency unions, trade effects, sensitivity analyses, duration, co-movements, and trade stability).*

### 1. Dissolution of Currency Union: World Sample..............................................................18

### 1. Dissolution of Currency Union: World Sample

### Introduction and motivation
- Gravity models (Pöyhönen, 1963; Tinbergen, 1962) explain more than half of the variation in trade and have been micro-founded in subsequent work.
- Recent empirical work (Rose, 2000; Frankel and Rose, 2002; Rose and van Wincoop, 2001; Glick and Rose, 2002) uses gravity models to analyze the impact of currency arrangements on bilateral trade.
- Frankel and Rose (2002) conclude that two countries that share the same currency trade "roughly three times as much" as they would if they used different currencies.
- The paper focuses on Africa because of its rich experience with currency unions, despite limited integration in world trade.

### Theoretical channels through which currency unions affect trade
- Currency unions reduce transaction costs between trading partners.
- They create larger markets enabling potential gains from economies of scale and production efficiency (given factor mobility and flexible wages).
- They bring macroeconomic stability by signaling a commitment to reduce inflation and enhance monetary credibility.
- They reduce uncertainty and eliminate exchange rate risk, encouraging investment and facilitating capital mobility.
- Potential losses arise from asymmetric shocks and insufficient shock absorbers (factor mobility and transfers).

### Empirical approach and scope
- Augments the gravity model of Glick and Rose (2002) to investigate:
  - Impact of currency unions on bilateral trade (world and Africa samples).
  - Possible trade diversion effects.
  - Effect of the duration of currency union membership on bilateral trade.
  - Impact of currency unions on price and output co-movements and trade stability.

### Key empirical findings (robust across estimation techniques)
- Magnitude of trade effect:
  - For both the world and Africa samples, countries belonging to the same currency union trade "about two times more" with each other than comparable countries that do not share a currency.
- Trade composition and stability:
  - There is evidence that currency unions are associated with trade creation.
  - Currency unions are associated with trade stability.
- Price and output co-movement:
  - Currency unions are associated with increased co-movement of prices among members.
  - Currency unions are not associated with increased co-movement of output among members.
- Duration of membership:
  - The longer a country participates in a currency union, the greater the benefit to bilateral trade.
  - The relationship between duration and trade benefits is not linear; there are diminishing returns to membership.
- Regional comparison:
  - African countries stand to benefit as much from currency union membership as countries in the rest of the world; currency union benefits are not region specific.
  - Marginal effects and mechanisms of transmission may vary across regions.

### Policy-relevant implications
- Currency unions can be an effective tool for stimulating trade.
- Given the low trade base in Africa and the protracted period necessary after establishment of a currency union to increase trade significantly, currency union membership is unlikely to be a primary tool to achieve high levels of trade in Africa.

*Source: 1. Dissolution of Currency Union: World Sample (excerpt).*

### Section V provides concluding remarks.

### _wp0631 - Section V provides concluding remarks.

### Literature review: currency unions and trade
- Rose (2000): trade between countries in the same currency union is "over three times greater" than trade between comparable countries not sharing the same currency.
- Rose finds the effect of decreased exchange rate volatility is positive but "much smaller" than the currency union coefficient; "the increase in trade associated with a currency union is over an order of magnitude larger than that associated with decreasing exchange rate volatility to zero."
- Rose’s explanations for the large currency-union effect include: financial integration induced by a common currency, and signaling of government commitment to long-term integration.
- Rose and van Wincoop (2001), using a structural model, reduce the currency union effect to "about two and a half."
- Glick and Rose (2002), using fixed-effects estimators, find joining a currency union "almost doubles bilateral trade."
- Rose (2004) meta-analysis (34 studies) estimates a range of "30 to 90 percent" for the currency union effect.
- Consensus in the literature: there is a common-currency effect on trade, but "probably substantially smaller than a tripling" (Frankel, 2005).
- Channel literature highlights determinants of beneficial currency-union membership: co-movement of prices, co-movement of outputs, inflation level, and volatility of inflation (Alesina and Barro, 2002).
- Tenreyro (2001): using a probit model, finds higher co-movements of prices increase propensity to form currency unions; co-movement of outputs has no effect.
- African context:
  - Explanations for Africa’s limited participation in world trade include slow economic growth, unfavorable geography, poor infrastructure and transport policies, poor trade policies, ineffective currency unions due to intraregional barriers and constraints to factor mobility, lesser diversification, asymmetric terms of trade shocks, poor fiscal transfer systems, and limited banking and financial sector development.
  - Debrun, Masson, and Pattillo (2005): gains from common currency depend on correlation of terms of trade shocks and interest-group effects (differences in government spending propensities).
  - Coe and Hoffmaister (1999): Africa trades more with the North than other developing regions on average.
  - Subramanian and Tamirisa (2003): importance of distinguishing Anglophone and Francophone Africa.
  - Masson and Pattillo (2004): using Glick and Rose (2002) specification, estimate currency unions increase trade by a factor of three for both Africa and the world.

### Contributions of the paper (as stated)
- Separate empirical analysis for two samples: the world and Africa, to test whether Africa’s experience differs.
- Modifications to standard gravity model estimations: include variables for free trade area, trade diversion, membership duration; extend sample coverage and period.
- Investigate and quantify stylized facts: trade-generating impact of currency union membership; effect on trade stability; price and output co-movements.

### Methodology: augmented gravity framework and extensions
- Baseline augmented gravity model (equation (1)): log(X_ij) = γ CU_ij + Σ_k β_k Z_kij + ... where:
  - X_ij: value of bilateral trade between i and j.
  - CU_ij: binary variable equal to unity if i and j share the same currency.
  - Z_k: vector including ln(Y_i Y_j), ln(Pop_i Pop_j), ln(D_ij), Cont_ij, Lang_ij, FTA_ij, ComNat_ij, ComCol_ij, Col_ij, ln(Area_i Area_j).
- Trade diversion indicator: dummy = 1 if the two countries are not in the same currency union but (at least) one is in a currency union with another country; negative and significant coefficient would indicate trade diversion.
- Membership duration (equation (2)): include "years" variable — number of years a pair shares a common currency — in the gravity specification.
- Co-movements measures:
  - Price co-movement: residuals from second-order autoregression on ln(P_it) and ln(P_jt) (equation (3)); ij_VP computed as sum of squared residuals scaled by T (equation (4)); higher ij_VP = greater synchronization.
  - Output co-movement: analogous process using ln(Y_it) and ln(Y_jt) (equations (5) and (6)), generating ij_VY.
- Trade stability assessment: dependent variable = coefficient of variation of log of real trade (two periods per pair: 1950-1976 and 1976-2003); robustness checks include (i) maximal absolute deviation of log real trade from sample average scaled by sample average, (ii) mean absolute deviation of exports from sample average scaled by average exports, and (iii) standard deviation of residual from conventional gravity on exports in levels.
- Treatment of zero trade observations: use log(constant + X_ij) to include zeros and Tobit estimation to account for censored dependent variable; pseudo maximum likelihood approach of Santos Silva and Tenreyro (2004) noted as a robustness alternative.

### Estimation issues and identification choices
- Omitted-variable bias ("gold medal mistake"): addressed by estimating OLS cross-section, fixed-effects (country-pair specific intercepts), random effects, and robust fixed-effects "within" estimator; inclusion of year effects and clustering of errors within country pairs.
- Endogeneity of currency union membership: acknowledged difficulty of finding valid instruments for a dummy; prior work (Alesina, Barro, Tenreyro, and Rose and van Wincoop) suggests reverse causality unlikely to explain findings; political motives important (especially in Africa); authors treat currency unions as exogenous for analysis.
- Model misspecification concerns:
  - Aggregation of exports and imports: check sensitivity using sum of logs and log of sums.
  - Inappropriate deflation by U.S. aggregate price index: include time dummies to correct global inflation trends.
  - Non-linear effects: include quadratic terms for output and output per capita.
  - Zero-trade observations: addressed as above (log constant, Tobit).

### Data
- Sample: 217 countries and political units over 1948-2002 constitutes the world sample.
- Africa sample: 49 countries where at least one of the ij pair is African; intra-Africa sample (both i and j from Africa); Africa-ROW sample (exactly one country from Africa).
- Data sources:
  - Bilateral trade: IMF’s Direction of Trade Statistics (real U.S. dollars).
  - GDP: University of Pennsylvania World Tables 6.1 and World Bank’s World Development Report.
  - Colonial past, distance, language: CIA World Factbook 2004.
  - Currency unions (1998-2002): IMF’s Exchange Arrangements and Exchange Restrictions Annual Report.
  - Free trade agreements and currency unions: Glick and Rose.
  - Prices (PPP of GDP): Penn World Tables 6.1.
  - Terms of trade: IMF’s World Economic Outlook 2004.
- Extensions relative to Glick and Rose (2002): include 217 units, extend period to 1948-2002, add FTAs operating in Africa omitted in prior work.

### Results: trade-generating effects (benchmarks and Africa)
- Cross-section replication (world sample): countries sharing a common currency trade more than "two and a half" times compared with non-union pairs (replicates Glick and Rose (2002) closely).
- Allowing FTA dummy to include more agreements (including African FTAs): currency union coefficient decreases slightly to "2.3".
- Africa sample (cross-section): currency union trade-generating effect is larger than in the world sample in cross-section estimates; marginal impacts of other determinants differ (common land border more important, common language less important, "ever colony" has greater impact, reversal of signs for "number of islands").
- Panel fixed-effects (country-pair intercepts) results:
  - Accounting for country fixed-effects reduces currency union coefficients significantly.
  - With country effects, estimated trade-boosting factors:
    - World: "1.8"
    - Africa: "1.7"
- Introducing intra-Africa and Africa-ROW dummies (baseline specification):
  - Currency union trade-generating effect becomes "2.2" times greater than for countries not in a currency union (in this specification).
  - Africa-ROW and intra-Africa trade dummies are "highly significant."
  - Sample restricted estimations with country-specific effects confirm beneficial effects of currency union participation in Africa:
    - Trade-generating effect is "1.8" for intra-Africa (panel results reported up to this point).

*Source: _wp0631 - Section V provides concluding remarks.*

### 1.7 for Africa-ROW trade. Finally, following Subramanian and Tamirisa (2003), we also

### _wp0631 - 1.7 for Africa-ROW trade. Finally, following Subramanian and Tamirisa (2003), we also

### Trade-generating effects of currency unions
- Trade-generating effect remains high for both Francophone and Anglophone Africa; no evidence of intra-Francophone or intra-Anglophone undertrading.
- Evidence of increasing integration of both Francophone and Anglophone Africa with the rest of the world: coefficients of global overall trade with the ROW are positive and significant, suggesting that overall trade was about 22 and 25 percent more than the average, respectively.
- Adding a trade diversion dummy shows positive coefficients in specifications with country-pair fixed-effects for both the world and Africa samples, indicating significant trade creation rather than trade diversion.
- Intra-Africa trade creation effect is about five times higher than the Africa-ROW trade creation effect.
- Overall currency union trade-generating effect remains about 2.1 and 2.0 for the world and Africa, respectively.

### Sensitivity analysis of trade-generating estimates
- Methodological variations tested:
  - Dependent variable as average of the logarithm of exports and imports (rather than logarithm of the average).
  - Adding quadratic terms for output and output per capita.
  - Adding time effects.
  - Modifying dependent variable to log(constant + Xij) to handle zero-trade observations (constant set equal to 100 for comparability with Alesina, Barro, and Tenreyro (2002)).
  - Tobit estimation to account for censored dependent variable.
- Impact of zero-trade observations:
  - Not eliminating zero observations increases the world and Africa samples by about 45 and 31 percent, respectively.
  - Excluding zeros may eliminate observations with low currency union elasticity and drive elasticity upwards; including zeros may drive elasticity downward.
- Estimated γ (currency union trade-generating effect) from fixed-effects estimates:
  - World sample: lies in the narrow range of about (1.17, 1.89).
  - Africa sample: lies in the narrow range of about (1.03, 1.85).
  - When including only statistically significant estimates and excluding corrections for zero-trade observations, ranges tighten to (1.59, 1.89) for the world and (1.68, 1.85) for Africa.
- Tobit estimates are sufficiently close to cross-section benchmark results but may suffer from omitted-variable bias if country-pair fixed-effects are not properly accounted for.

### Years of membership, co-movements, and trade stability
- Effect of leaving a currency union (lags analysis following Glick and Rose (2002)):
  - Exits from a currency union are associated with a decline in bilateral trade, with statistically significant negative effects appearing only after an extended period.
  - For both samples, negative effects are not statistically significant immediately after dissolution; significant coefficients appear in periods of 16-27 years (world) and 16-21 years (Africa) after leaving.
  - Some negative effects in absolute terms can be identified only after 30 years.
  - After ten years of currency union exit, bilateral trade declines by about 35 percent for Africa and 55 percent for the world.
  - Cumulative decline after 10 years is greater in Africa than in the world sample.
  - Nonlinear impact motivates inclusion of a quadratic term in duration analysis.
- Impact of duration of membership (Table 5 results):
  - One additional year of membership increases trade by 1.4 and 2.0 percent in the world and Africa samples, respectively.
  - In the world sample, after 10 years, impact on trade is a 15 percent increase [(1 + 0.014)^10]; to double trade would take about 67 years.
  - In the Africa sample, after 10 years trade increases by an estimated 22 percent; trade is estimated to take 45 years to double.
  - Negative and significant quadratic term indicates diminishing returns to membership duration.
- Co-movements of prices and outputs (Table 6 results):
  - Currency unions tend to increase co-movement of prices for both world and Africa samples.
  - Currency unions are not systematically related to co-movement of outputs.
  - Marginal effect of currency unions on price co-movements is higher for Africa.
  - Coefficient of currency union for Africa in price co-movement regressions is 0.04 with country fixed-effects; mean co-movements of prices for the Africa subsample equal –0.15.
  - Lack of significant effects on output co-movements may reflect ambiguous theoretical links and dependence on intra- vs. inter-industry trade.
- Trade stability (Table 7 results):
  - Initial specifications (columns 1 and 2) show currency union makes trade more stable, marginal impact slightly higher in Africa than the world.
  - Once country-pair fixed-effects are included (columns 3 and 4), the currency union coefficient falls and becomes statistically insignificant for both samples.
  - Mean stability: Africa sample is 0.22 (compared to 0.18 for the world).
  - Alternative stability measures and methods (maximal absolute value, mean absolute value, standard deviation of residuals from gravity equation) yield reasonably robust results, except specifications including zero-trade observations over extended periods bias the stability measure.

### Robustness of years-membership, co-movements, and stability findings
- Sensitivity checks applied: dependent variable as average of logarithms; adding quadratic terms for output and output per capita; adding time effects.
- Results robustness:
  - Table 5 findings robust: after 10 years of membership, trade increases by 14-22 percent (world) and 20-29 percent (Africa); diminishing returns always significant.
  - Tables 6 and 7 conclusions insensitive to perturbations, with caveat that inclusion of zero-trade observations can bias stability measures.

### Conclusions (key substantive findings)
- Currency union membership substantially increases bilateral trade: members trade about twice as much with each other compared with similar nonmembers.
- Trade increase from currency unions does not come with trade diversion from nonmembers.
- Duration matters: longer membership yields larger trade benefits, but with diminishing returns.
- Currency unions reduce trade volatility and increase price co-movements, but have no significant effect on output co-movements.
- Results hold for both world and Africa samples, though marginal effects and transmission mechanisms differ across samples.
- Methodology does not resolve causal channels linking currency unions, trade, growth, and poverty reduction; omitted political and institutional factors may influence both union membership and trade outcomes and warrant further investigation.

*Source: Excerpt from IMF Working Paper _wp0631 (selected sections on currency unions, trade effects, sensitivity analyses, duration, co-movements, and trade stability).*

### 2. The corresponding lag of the dummy variable associated with CU dissolution is

### _wp0631 - 2. The corresponding lag of the dummy variable associated with CU dissolution is

### Benchmark Results (Table 1)
- Dependent variable: log of real trade, log(Xij).
- Sample/Estimations summarized by specification (World/Africa; Fixed effects/OLS). Key coefficient estimates (coefficient (robust t-statistic)):
  - Currency union: 1.00 ** (8.55); 0.84 ** (7.75); 0.99 ** (6.57); 0.54 ** (13.37); 0.58 ** (14.35); 0.55 ** (8.18)
  - Log distance: -1.08 ** (52.68); -1.05 ** (52.08); -1.06 ** (19.77)
  - Log product real GDP: 0.92 ** (104.20); 0.93 ** (105.38); 1.01 ** (61.62); 0.16 ** (24.28); 0.15 ** (22.52); -0.10 ** (8.66)
  - Log product real GDP/capita: 0.42 ** (32.28); 0.44 ** (33.80); 0.34 ** (14.77); 0.55 ** (51.62); 0.57 ** (53.47); 0.66 ** (33.23)
  - Common language: 0.36 ** (9.25); 0.35 ** (9.31); 0.29 ** (4.66)
  - Common land border: 0.52 ** (4.71); 0.40 ** (3.85); 1.22 ** (7.30)
  - FTA: 1.14 ** (10.40); 0.64 ** (18.75)
  - FTA (with Africa): 1.19 ** (12.08); 0.92 ** (4.61); 0.30 **; 0.20 **; -14.32; -7.46 (note: table shows mixed cells)
  - Number landlocked in the pair: -0.23 ** (8.17); -0.25 ** (8.82); -0.35 ** (8.86)
  - Number islands in the pair: 0.03 (0.92); 0.03 (0.73); -0.31 ** (5.00)
  - Log product of areas: -0.08 ** (10.75); -0.08 ** (11.26); -0.18 ** (14.70)
  - Common colonizer: 0.55 ** (8.78); 0.51 ** (8.27); 0.30 ** (3.48)
  - Current colony: 0.95 ** (3.91); 0.99 ** (4.02); -0.39 (0.86); 0.43 (5.42); 0.41 ** (5.11); -0.07 (0.49)
  - Ever colony: 1.31 ** (10.75); 1.32 ** (10.97); 2.14 ** (14.42)
  - Same nation: -0.20 (0.20); -0.23 (0.22); 1.63 ** (3.30)
- Observations: 265,262; 265,262; 100,597; 265,262; 265,262; 100,597
- R-squared (within): 0.13; 0.13; 0.03
- R-squared (between): 0.41; 0.40; 0.01
- R-squared (overall): 0.68; 0.68; 0.52; 0.38; 0.36; 0.01
- Hausman test (p-value): 0.00; 0.00; 0.00
- Notes: Robust t-statistics in parentheses. One asterisk (*) denotes significance at the 5 percent level; two asterisks (**) denote significance at the 1 percent level.

### Africa Trade Details (Table 2)
- Dependent variable: log of real trade, log(Xij).
- Key coefficient estimates:
  - Currency union: 0.79 ** (7.03); 0.53 (1.18); 0.55 ** (2.69); 0.58 ** (14.35); 0.62 ** (4.56); 0.53 ** (6.35)
  - Africa-ROW: 0.27 ** (7.90)
  - Intra Africa: 0.42 ** (4.58)
  - Log distance: -1.06 ** (52.00); -0.96 ** (15.64); -1.29 ** (9.92)
  - Log product real GDP: 0.94 ** (104.26); 1.06 ** (64.47); 0.62 ** (9.83); 0.15 ** (22.52); -0.13 ** (10.37); -0.07 * (2.56)
  - Log product real GDP/capita: 0.48 ** (35.94); 0.33 ** (13.95); 0.50 ** (5.80); 0.57 ** (53.47); 0.75 ** (32.71); 0.37 ** (8.34)
  - Common language: 0.33 ** (8.76); 0.27 ** (4.26); 0.40 * (2.01)
  - Common land border: 0.43 ** (4.17); 1.10 ** (3.21); 0.98 ** (4.27)
  - FTA (with Africa): 1.15 ** (13.08); 0.78 ** (5.24); 0.30 ** (12.08); 0.18 ** (3.39)
  - Number landlocked in the pair: -0.28 ** (9.97); -0.37 ** (8.82); -0.33 ** (3.07)
  - Number islands in the pair: 0.04 (1.09); -0.39 ** (6.07); 0.17 (0.83)
  - Log product of areas: -0.09 ** (12.35); -0.20 ** (16.14); -0.01 (0.15)
  - Common colonizer: 0.49 ** (7.91); 0.32 ** (3.46); 0.25 (1.09)
  - Current colony: 0.98 ** (3.92); -0.03 (0.06); 0.41 ** (5.11); -0.10 (0.65)
  - Ever colony: 1.30 ** (11.25); 2.12 ** (14.08)
  - Same nation: -0.31 (0.31); 1.47 ** (2.62)
- Observations: 265,262; 85,759; 14,838; 265,262; 85,759; 14,838
- R-squared (within): 0.13; 0.03; 0.01
- R-squared (between): 0.40; 0.00; 0.01
- R-squared (overall): 0.68; 0.54; 0.40; 0.36; 0.00; 0.01
- Hausman test (p-value): 0.00; 0.00; 0.00

### Trade Creation or Diversion? (Table 3)
- Dependent variable: log of real trade, log(Xij).
- Key coefficient estimates:
  - Currency union: 0.97 ** (8.77); 1.05 ** (6.81); 0.61 (1.38); 0.59 ** (2.58); 0.73 ** (17.74); 0.70 ** (10.29); 0.73 ** (5.35); 1.03 ** (10.80)
  - Trade diversion: 0.23 ** (8.66); 0.10 * (2.35); 0.09 (1.92); 0.09 (0.55); 0.21 ** (21.18); 0.18 ** (10.24); 0.12 ** (6.71); 0.58 ** (10.75)
  - Log distance: -1.05 ** (52.02); -1.05 ** (19.62); -0.96 ** (15.61); -1.29 ** (9.86)
  - Log product real GDP: 0.93 ** (105.28); 1.01 ** (61.82); 1.06 ** (64.45); 0.63 ** (9.89); 0.18 ** (26.35); -0.06 ** (5.35); -0.11 ** (8.01); 0.00 (0.09)
  - Log product real GDP/capita: 0.44 ** (33.56); 0.33 ** (14.46); 0.32 ** (13.64); 0.50 ** (5.61); 0.56 ** (52.34); 0.64 ** (31.95); 0.73 ** (31.74); 0.31 ** (6.94)
  - Common language: 0.33 ** (8.80); 0.28 ** (4.57); 0.27 ** (4.20); 0.40 * (2.00)
  - Common land border: 0.42 ** (4.04); 1.22 ** (7.29); 1.11 ** (3.19); 0.98 ** (4.26)
  - FTA (with Africa): 1.17 ** (14.03); 0.93 ** (7.55); 0.80 ** (5.13); 0.32 ** (12.74); 0.22 ** (5.19); 0.31 ** (5.61)
  - Number landlocked in the pair: -0.24 ** (8.38); -0.35 ** (8.71); -0.36 ** (8.70); -0.33 ** (3.03)
  - Number islands in the pair: 0.01 (0.24); -0.31 ** (5.01); -0.39 ** (6.07); 0.17 (0.86)
  - Log product of areas: -0.08 ** (11.48); -0.18 ** (14.82); -0.20 ** (16.19); -0.01 (0.20)
  - Common colonizer: 0.52 ** (8.42); 0.31 ** (3.60); 0.33 ** (3.51); 0.26 (1.15)
  - Current colony: 1.03 ** (4.17); -0.35 (0.77); -0.02 (0.03); 0.44 ** (5.52); -0.08 (0.53); -0.11 (0.67)
  - Ever colony: 1.25 ** (10.57); 2.10 ** (13.94); 2.09 ** (13.65)
  - Same nation: -0.25 (0.24); 1.60 ** (3.22); 1.46 ** (2.59)
- Observations: 265,262; 100,597; 85,759; 14,838; 265,262; 100,597; 85,759; 14,838
- R-squared (within): 0.13; 0.03; 0.03; 0.02
- R-squared (between): 0.13; 0.02; 0.03; 0.02
- R-squared (overall): 0.68; 0.52; 0.54; 0.40; 0.43; 0.03; 0.00; 0.04
- Hausman test (p-value): 0.39; 0.00; 0.00; 0.00; 0.04

### Sensitivity Analysis of Currency Union Effect (Table 4)
- Dependent variable: log(100+Xij) or variations; multiple estimation methods.
- Key gamma (γ) estimates and effects (coefficient (t-statistic)):
  - Ordinary Least squares: 0.84 ** (7.75) → Effect 2.3; 0.99 ** (6.57) → Effect 2.7
  - Dependent variable log(100+Xij): 0.57 ** (4.85) → Effect 1.8; 0.58 ** (4.05) → Effect 1.8
  - Fixed Effects: 0.58 ** (14.35) → Effect 1.8; 0.55 ** (8.18) → Effect 1.7
  - Fixed effects AND dependent variable calculated as average of logs: 0.64 ** (15.33) → Effect 1.9; 0.62 ** (9.15) → Effect 1.9
  - Fixed effects AND non-linearities added: 0.46 ** (11.46) → Effect 1.6; 0.52 ** (7.82) → Effect 1.7
  - Fixed effects AND time effects added: 0.60 ** (14.84) → Effect 1.8; 0.54 ** (8.13) → Effect 1.7
  - Fixed effects AND dependent variable is log(100+Xij): 0.16 ** (4.55) → Effect 1.2; 0.03 (0.59) → Effect 1.0
  - Tobit estimation: 0.79 ** (20.30) → Effect 2.2; 0.86 ** (16.09) → Effect 2.4
  - County-fixed effects: 0.85 ** (7.86) → Effect 2.4; 1.14 ** (7.82) → Effect 3.1
- Notes: Controls not reported are those listed in the other tables. One asterisk (*) denotes significance at the 5 percent level; two asterisks (**) denote significance at the 1 percent level.

### Duration of Currency Union Membership (Table 5)
- Dependent variable: log of real trade, log(Xij).
- Key coefficients and statistics (coefficient (t-statistic)):
  - Years in currency union: 0.03 ** (7.47); 0.07 ** (7.49); 0.03 ** (5.99); 0.06 ** (5.58); 0.01 ** (8.35); 0.05 ** (13.67); 0.02 ** (8.71); 0.04 ** (8.05)
  - (Years in currency union)2: -0.001 ** (4.86); -0.001 ** (3.26); -0.001 ** (11.00); -0.001 ** (4.58)
  - Log distance: -1.05 ** (52.02); -1.05 ** (52.00); -1.06 ** (19.88); -1.06 ** (19.82)
  - Log product real GDP: 0.93 ** (105.60); 0.93 ** (105.64); 1.01 ** (61.60); 1.01 ** (61.70); 0.15 ** (21.23); 0.15 ** (22.31); -0.11 ** (9.91); -0.10 ** (9.01)
  - Log product real GDP/capita: 0.44 ** (33.88); 0.44 ** (33.87); 0.34 ** (14.77); 0.34 ** (14.72); 0.58 ** (54.57); 0.57 ** (53.23); 0.68 ** (34.18); 0.67 ** (33.07)
  - Common language: 0.35 ** (9.26); 0.35 ** (9.22); 0.29 ** (4.74); 0.29 ** (4.67)
  - Common land border: 0.41 ** (3.94); 0.40 ** (3.89); 1.23 ** (7.29); 1.23 ** (7.29)
  - FTA (with Africa): 1.19 ** (14.35); 1.19 ** (14.41); 0.91 ** (7.36); 0.92 ** (7.42); 0.30 ** (12.13); 0.32 ** (12.66); 0.17 ** (4.00); 0.18 ** (4.18)
  - Number landlocked in the pair: -0.25 ** (8.82); -0.25 ** (8.82); -0.36 ** (8.90); -0.36 ** (8.88)
  - Number islands in the pair: 0.02 (0.67); 0.02 (0.68); -0.31 ** (5.00); -0.31 ** (5.02)
  - Log product of areas: -0.08 ** (11.30); -0.08 ** (11.38); -0.18 ** (14.67); -0.18 ** (14.74)
  - Common colonizer: 0.51 ** (8.24); 0.50 ** (8.06); 0.31 ** (3.53); 0.30 ** (3.41)
  - Current colony: 1.17 ** (4.80); 1.06 ** (4.49); -0.04 (0.09); -0.21 (0.47); 0.58 ** (7.46); 0.48 ** (6.10); 0.11 (0.81); 0.00 (0.01)
  - Ever colony: 1.34 ** (11.10); 1.33 ** (11.04); 2.15 ** (14.69); 2.14 ** (14.60)
  - Same nation: -0.24 (0.21); -0.22 (0.19); 1.62 ** (3.23); 1.73 ** (3.60)
- Observations: 265,262; 265,262; 100,597; 100,597; 265,262; 265,262; 100,597; 100,597
- R-squared (within): 0.13; 0.13; 0.03; 0.03
- R-squared (between): 0.38; 0.40; 0.00; 0.01
- R-squared (overall): 0.68; 0.68; 0.52; 0.52; 0.35; 0.36; 0.01; 0.01
- Hausman test (p-value): 0.00; 0.00; 0.00; 0.00

### Currency Union Impact on Comovements of Outputs and Prices (Table 6)
- Dependent variables: comovement of outputs (Vyij) for (1)-(4); and comovement of prices (Vpij) for (5)-(8).
- Sample size and model details:
  - Observations: 6,992; 3,793; 6,992; 3,793; 7,000; 3,796; 7,000; 3,796 (depending on column)
  - R-squared ranges reported (0.25, 0.15, 0.92, 0.92, 0.12, 0.13, 0.87, 0.91) across columns and estimations.
- Key coefficient estimates (coefficient (t-statistic)):
  - Currency union: -0.002 (0.61); 0.002 (0.55); 0.000 (0.12); 0.000 (0.25); 0.06 ** (9.07); 0.07 ** (9.35); 0.03 ** (11.19); 0.04 ** (11.19)
  - Log distance: 0.002 ** (3.56); 0.008 ** (7.58); -0.001 ** (3.56); -0.001 (0.96); -0.003 (1.79); 0.001 (0.26); -0.004 ** (5.33); -0.003 * (2.23)
  - Log product real GDP: 0.005 ** (22.10); 0.005 ** (14.19); 0.003 ** (5.21); 0.000 (0.30); 0.002 ** (4.18); -0.003 ** (3.96); 0.007 ** (3.34); -0.008 ** (3.57)
  - Log product real GDP/capita: 0.005 ** (16.98); 0.001 (1.63); -0.004 ** (4.39); 0.001 (0.56); 0.012 ** (14.76); 0.016 ** (16.61); -0.004 (1.17); 0.011 ** (3.47)
  - Common language: 0.001 (0.65); -0.002 (1.54); 0.001 * (2.05); 0.000 (0.13); 0.001 (0.42); 0.000 (0.09); 0.002 (1.43); 0.001 (1.15)
  - Common land border: 0.006 * (2.52); 0.005 (1.44); 0.003 ** (3.65); 0.001 (0.82); -0.005 (0.66); -0.007 (0.77); 0.004 (1.46); 0.004 (1.24)
  - FTA '(with Africa): 0.012 ** (4.59); 0.010 ** (2.65); 0.003 * (2.35); -0.001 (0.60); 0.018 ** (2.84); 0.015 (1.89); 0.007 * (2.50); -0.007 * (2.21)
  - Number landlocked in the pair: 0.005 ** (8.19); 0.007 ** (8.65); 0.007 ** (3.88); 0.014 ** (7.30)
  - Number islands in the pair: 0.005 ** (7.14); 0.010 ** (8.08); 0.012 ** (6.77); 0.012 ** (5.45)
  - Log product of areas: 0.000 ** (3.07); 0.000 (0.66); -0.004 ** (9.62); -0.001 (1.94)
  - Common colonizer: 0.004 ** (3.14); 0.009 ** (5.45); 0.002 ** (3.49); 0.001 (0.98); 0.010 ** (3.35); 0.002 (0.50); 0.005 ** (4.06); 0.006 ** (4.48)
  - Current colony: -0.018 (1.70); -0.044 (1.47); -0.007 * (2.03); -0.002 (0.23); 0.049 (1.90); -0.040 (0.68); 0.025 * (2.15); -0.020 (0.70)
  - Ever colony: 0.004 (1.70); 0.018 * (2.01); 0.000 (0.03); 0.001 (0.55); 0.007 (0.72); 0.034 (1.77); -0.006 (1.40); 0.005 (0.70)
  - Same nation: Country fixed effects listed; mixed significance.
- Notes: Robust t-statistics in parentheses. One asterisk (*) denotes significance at the 5 percent level; two asterisks (**) denote significance at the 1 percent level.

### Currency Union Impact on Trade Stability (Table 7)
- Dependent variable: trade variability, the coefficient of variation of log(Xij).
- Sample and estimation:
  - Observations: 18,156 (World); 7,845 (Africa)
  - Models: OLS and Panel fixed effects.
- Key coefficient estimates (coefficient (t-statistic)):
  - Currency union: -0.05 ** (4.36); -0.06 ** (4.40); -0.02 (0.95); -0.02 (0.64)
  - Log distance: 0.05 ** (22.53); 0.07 ** (13.09)
  - Log product real GDP: -0.03 ** (30.40); -0.05 ** (19.77); -0.01 * (2.00); 0.03 ** (2.63)
  - Log product real GDP/capita: -0.02 ** (9.86); -0.02 ** (4.08); -0.04 ** (6.11); -0.08 ** (6.85)
  - Common language: -0.02 ** (5.46); -0.03 ** (3.55)
  - Common land border: 0.02 * (2.01); -0.03 * (2.41)
  - FTA (with Africa): -0.06 ** (6.30); -0.07 ** (4.97); -0.02 (1.46); -0.04 * (2.31)
  - Number landlocked in the pair: 0.01 (1.83); 0.02 * (2.46)
  - Number islands in the pair: 0.01 * (2.51); 0.04 ** (4.90)
  - Log product of areas: 0.00 ** (2.77); 0.01 ** (5.58)
  - Common colonizer: -0.02 * (2.15); 0.00 (0.21)
  - Current colony: -0.01 (0.47); 0.11 ** (3.44); 0.04 (1.29); 0.06 (1.24)
  - Ever colony: -0.02 * (2.55); -0.04 ** (2.65)
  - Same nation: -0.03 (1.00); -0.12 ** (4.10)
- Observations: 18,156; 7,845
- R-squared (within): 0.05; 0.10
- R-squared (between): 0.06; 0.11
- R-squared (overall): 0.18; 0.17; 0.11; 0.01
- Hausman test (p-value): 0.11; 0.00

### Appendices — Main African Regional Arrangements (Appendix A)
- Lists of regional/sub-regional arrangements and member countries (as of August 2005), acronyms provided:
  - CEMAC, ECCAS, COMESA, EAC, WAEMU, CFA Franc Zone, WAMZ, ECOWAS, IOC, SACU, SADC, etc.
- Examples of listed countries and regions: São Tomé and Príncipe; Benin; Cameroon; Burundi; Burkina Faso; Chad; Rwanda; Djibouti; Côte d'Ivoire; Congo; Egypt; Guinea-Bissau; Central African Republic; Ethiopia; Mali; Equatorial Guinea; Angola; Eritrea; Niger; Gabon; Dem. Rep. of Congo; Sudan; Senegal; Togo; Kenya; Uganda; Cape Verde; Liberia; Tanzania; Gambia, The; Malawi; Mauritius; Ghana; Zambia; Madagascar; Guinea; Zimbabwe; Comoros; Sierra Leone; Namibia; Seychelles; Swaziland; Botswana; Reunion; Lesotho; South Africa; Mozambique; etc.
- Acronyms and expansions provided exactly as in the source.

### Sample Data: Variable Definitions and Sources (Appendix B)
- Dependent variable:
  - Xijt: The average value of real bilateral trade between i and j at time t, IMF's Direction of Trade (DoT); Average of available values for export from a to b export from b to a, import into a from b, imports into b from a. Deflated by U.S. Consumer Price Index.
- Explanatory variables and sources (preserved exactly):
  - Yi: Real GDP — World Bank's World Development Indicators
  - Popi: Population — World Bank's World Development Indicators, Penn World Tables 6.1; Glick and Rose (2002) data used for 1948-1950
  - Dij: The distance between i and j — CIA’s World Factbook; Great Circle Method used to calculate distance
  - Langij: Binary variable unity if i and j have a common language — CIA’s World Factbook
  - Contij: Binary variable unity if i and j share a land border — CIA’s World Factbook
  - Landl: Number of landlocked countries in the country-pair (0, 1, or 2) — CIA’s World Factbook
  - Island: Number of island nations in the country-pair (0, 1, or 2) — CIA’s World Factbook
  - Area: Land area of country n, n=i,j — CIA’s World Factbook
  - ComColij: Binary variable unity if i and j were colonies after 1945 with the same colonizer — CIA’s World Factbook
  - CurColij: Binary variable unity if i and j are colonies at time t — CIA’s World Factbook
  - Colonyij: Binary variable unity if i colonized j or vice versa — CIA’s World Factbook
  - ComNatij: Binary variable unity if i and j remained part of the same nation during the sample — CIA’s World Factbook
  - FTAij: Binary variable unity if i and j belong to the same regional trade agreement — WTO publication [http://www.wto.org/english/tratop_e/region_e/region_e.htm]
  - CUij: Binary variable unity if i and j use the same currency at time t — Glick and Rose (2002). For 1998-2002: IMF's Annual Report on Exchange Arrangements and Exchange Restrictions
  - Pi: PPP of GDP — Penn World Tables 6.1; price level of GDP (P) is the PPP over GDP divided by the exchange rate and multiplied by 100
  - Toti: Terms of trade — IMF's World Economic Outlook 2004

### Summary Statistics (Appendix C, Table C1)
- Table C1: Summary Statistics (World sample: 265,262 observations; Africa sample: 100,597 observations)
- Selected means and standard deviations (Mean Std. Deviation for World; Mean Std. Deviation for Africa):
  - Log of real trade: 10.020 3.246; 8.813 2.951
  - Currency union: 0.014 0.119; 0.027 0.161
  - Log distance: 8.167 0.809; 8.175 0.661
  - Log product real GDP: 47.974 2.662; 47.075 2.193
  - Log product real GDP/capita: 16.106 1.470; 15.180 1.259
  - Common language: 0.216 0.411; 0.266 0.442
  - Common land border: 0.030 0.170; 0.029 0.169
  - FTA (with Africa): 0.034 0.180; 0.045 0.207
  - Number landlocked in the pair: 0.262 0.479; 0.402 0.553
  - Number islands in the pair: 0.338 0.537; 0.250 0.470
  - Log product of areas: 24.139 3.299; 24.670 2.904
  - Common colonizer: 0.099 0.298; 0.155 0.362
  - Current colony: 0.002 0.041; 0.002 0.043
  - Ever colony: 0.020 0.139; 0.017 0.129
  - Same nation: 0.000 0.016; 0.000 0.017
  - Years in currency union: 0.359 3.481; 0.705 4.840
  - Comovements of outputs: -0.075 0.030; -0.087 0.031
  - Comovements of prices: -0.155 0.080; -0.163 0.070
  - Trade volatility: 0.181 0.258; 0.224 0.294
- Notes:
  - One value for country pair (6992/7000 observations for the world for co-movements of outputs/prices and 3,793/3,796 for Africa samples, respectively).
  - Two values for country pair (18,156 observations for the world and 7,845 for the Africa sample).

*Italicized source attribution provided separately by the pipeline.*

### Appendix C

### Appendix C

### Regional agreements — Africa
- ECOWAS: Economic Community of West African States
  - Benin (1975), Ghana (1975), Niger (1975), Burkina Faso (1975), Guinea (1975), Nigeria (1975), Cape Verde (1975), Liberia (1975), Senegal (1975), Côte d'Ivoire (1975), Mali (1975), Sierra Leone (1975), Gambia, The (1975), Mauritania (1975), Togo (1975), Guinea-Bissau (1975)
- COMESA: Common Market for Eastern and Southern Africa
  - Angola (1981), Malawi (1981), Tanzania (1981), Burundi (1981), Mauritius (1981), Uganda (1981), Comoros (1981), Mozambique (1981), Zambia (1981), Djibouti (1981), Rwanda (1981), Eritrea (1993), Namibia (1993), Zimbabwe (1981), Ethiopia (1981), Somalia (1981), Madagascar (1993), Kenya (1981), Sudan (1981), Seychelles (1993), Lesotho (1981), Swaziland (1981)
- SADC: Southern African Development Community
  - Angola (1980), Lesotho (1980), Zimbabwe (1980), Botswana (1980), Swaziland (1980), Namibia (1990), Malawi (1980), Tanzania (1980), Mauritius (1995), Mozambique (1980), Zambia (1980), South Africa (1995)
- CEMAC: Central African Economic and Monetary Community
  - Congo, Republic of the (1962), Central African Republic (1962), Gabon (1962), Chad (1962-1968, 1984), Cameroon (1962), Equatorial Guinea (1983)
- WAEMU: West African Economic and Monetary Union
  - Benin (1984), Burkina Faso (1962), Côte d'Ivoire (1962), Mali (1962), Niger (1962), Senegal (1962), Mauritania (1962 - 1994), Togo (1994)

### Regional agreements — Rest of the World
- ASEAN
  - Brunei (1984), Cambodia (1999), Indonesia (1967), Lao PDR (1997), Malaysia (1967), Myanmar (1997), Philippines (1967), Singapore (1967), Thailand (1967), Vietnam (1995)
- EU
  - Austria (1995), Belgium (1958), Denmark (1973), Finland (1995), France (1958), Germany (1958), Greece (1981), Luxembourg (1958), Ireland (1973), Italy (1958), Netherlands (1958), Portugal (1986), Spain (1986), Sweden (1995), United Kingdom (1973)
- US-Israel
  - United States (1985), Israel (1985)
- NAFTA
  - Canada (1989), United States (1989), Mexico (1994)
- CARICOM
  - Antigua and Barbuda (1974), Bahamas (1983), Barbados (1973), Belize (1974), Dominica (1974), Guyana (1973), Grenada (1974), Jamaica (1973), Montserrat (1974), Saint Kitts and Nevis (1974), Saint Lucia (1974), Saint Vincent and the Grenadines (1974), Suriname (1995), Trinidad and Tobago (1973), Haiti (2002)
- PATCRA
  - Australia (1977), Papua New Guinea (1977)
- ANZCERTA
  - Australia (1983), New Zealand (1983)
- CACM
  - Costa Rica (1963), El Salvador (1961), Guatemala (1961), Honduras (1961), Nicaragua (1961)
- SPARTECA
  - Australia (1981), New Zealand (1981), Cook Islands (1981), Micronesia, Federated States of (1986), Fiji (1981), Kiribati (1981), Marshall Islands (1981), Nauru (1981), Niue (1981), Papua New Guinea (1981), Solomon Islands (1981), Tonga (1981), Tuvalu (1981), Vanuatu (1981), Samoa (1981)
- MERCOSUR
  - Argentina (1991), Brazil (1991), Paraguay (1991), Uruguay (1991), Chile (1996), Bolivia (1996)

### Table C2: Free Trade Agreements in the Sample
- (Table title present in source: "Table C2: Free Trade Agreements in the Sample")

### Currency unions in the sample (Table C3)
- Table title present in source: "Table C3: Currency Unions in the Sample"
- Selected entries (currency union members with End dates exactly as shown):
  - Antigua and Barbuda — Barbados 1975
  - Antigua and Barbuda — Dominica ongoing
  - Antigua and Barbuda — Grenada ongoing
  - Antigua and Barbuda — Guyana 1971
  - Antigua and Barbuda — Montserrat ongoing
  - Antigua and Barbuda — Saint Kitts and Nevis ongoing
  - Antigua and Barbuda — Saint Lucia ongoing
  - Antigua and Barbuda — Saint Vincent and the Grenadines ongoing
  - Antigua and Barbuda — Trinidad and Tobago 1976
  - Aruba — Netherlands Antilles ongoing
  - Aruba — Suriname 1994
  - Australia — Kiribati ongoing
  - Australia — Nauru ongoing
  - Australia — Solomon Islands 1979
  - Australia — Tonga 1991
  - Australia — Tuvalu ongoing
  - Bangladesh — India 1974
  - Barbados — Dominica 1975
  - Barbados — Grenada 1975
  - Barbados — Guyana 1971
  - Barbados — Montserrat 1975
  - Barbados — Saint Kitts and Nevis 1975
  - Barbados — Saint Lucia 1975
  - Barbados — Saint Vincent and the Grenadines 1975
  - Barbados — Trinidad and Tobago 1975
  - Belgium — Burundi 1964
  - Belgium — Congo, Democratic Republic of the 1961
  - Belgium — Rwanda 1966
  - Belgium-Luxembourg — Burundi 1964
  - Belgium-Luxembourg — Congo, Democratic Republic of the 1961
  - Belgium-Luxembourg — Rwanda 1966
  - Benin — Burkina Faso ongoing
  - Benin — Cote d'Ivoire ongoing
  - Benin — Equatorial Guinea ongoing
  - Benin — Gabon ongoing
  - Benin — Guinea 1969
  - Benin — Guinea-Bissau ongoing
  - Benin — Madagascar 1982
  - Benin — Mali ongoing
  - Benin — Mauritania 1974
  - Benin — Niger ongoing
  - Benin — Reunion 1976
  - Benin — Senegal ongoing
  - Benin — Togo ongoing
  - Bhutan — India ongoing
  - Bhutan — Pakistan 1966
  - Botswana — Lesotho 1977
  - Botswana — Swaziland 1977
  - Brunei — Malaysia 1971
  - Brunei — Singapore ongoing
  - Myanmar — India 1966
  - Myanmar — Pakistan 1971
  - Cameroon — Benin ongoing
  - Cameroon — Burkina Faso ongoing
  - Cameroon — Central African Republic ongoing
  - Cameroon — Chad ongoing
  - Cameroon — Comoros 1994
  - Cameroon — Congo, Republic of the ongoing
  - Cameroon — Cote d'Ivoire ongoing
  - Cameroon — Equatorial Guinea ongoing
  - Cameroon — Gabon ongoing
  - Cameroon — Guinea 1969
  - Cameroon — Guinea-Bissau ongoing
  - Cameroon — Madagascar 1982
  - Cameroon — Mali ongoing
  - Cameroon — Mauritania 1974
  - Cameroon — Niger ongoing
  - Cameroon — Reunion 1976
  - Cameroon — Senegal ongoing
  - Cameroon — Togo ongoing
  - Central African Republic — Benin ongoing
  - Central African Republic — Burkina Faso ongoing
  - Central African Republic — Chad ongoing
  - Central African Republic — Comoros 1994
  - Central African Republic — Congo, Republic of the ongoing
  - Central African Republic — Cote d'Ivoire ongoing
  - Central African Republic — Equatorial Guinea ongoing
  - Central African Republic — Gabon ongoing
  - Central African Republic — Guinea 1969
  - Central African Republic — Guinea-Bissau ongoing
  - Central African Republic — Madagascar 1982
  - Central African Republic — Mali ongoing
  - Central African Republic — Mauritania 1974
  - Central African Republic — Niger ongoing
  - Central African Republic — Reunion 1976
  - Central African Republic — Senegal ongoing
  - Central African Republic — Togo ongoing
  - Chad — Benin ongoing
  - Chad — Burkina Faso ongoing
  - Chad — Comoros 1994
  - Chad — Congo, Republic of the ongoing
  - Chad — Côte d'Ivoire ongoing
  - Chad — Equatorial Guinea ongoing
  - Chad — Gabon ongoing
  - Chad — Guinea 1969
  - Chad — Guinea-Bissau ongoing
  - Chad — Madagascar 1982
  - Chad — Mali ongoing
  - Chad — Mauritania 1974
  - Chad — Niger ongoing
  - Chad — Reunion 1976
  - Chad — Senegal ongoing
  - Chad — Togo ongoing
  - Comoros — Benin 1994
  - Comoros — Burkina Faso 1994
  - Comoros — Congo, Republic of the 1994
  - Comoros — Côte d'Ivoire 1994
  - Comoros — Equatorial Guinea 1994
  - Comoros — Gabon 1994
  - Comoros — Guinea 1969
  - Comoros — Madagascar 1982
  - Comoros — Mali 1994
  - Comoros — Mauritania 1974
  - Comoros — Niger 1994
  - Comoros — Reunion 1976
  - Comoros — Senegal 1994
  - Comoros — Togo 1994
  - Congo, Republic of the — Benin ongoing
  - Congo, Republic of the — Burkina Faso ongoing
  - Congo, Republic of the — Côte d'Ivoire ongoing
  - Congo, Republic of the — Equatorial Guinea ongoing
  - Congo, Republic of the — Gabon ongoing
  - Congo, Republic of the — Guinea 1969
  - Congo, Republic of the — Guinea-Bissau ongoing
  - Congo, Republic of the — Madagascar 1982
  - Congo, Republic of the — Mali ongoing
  - Congo, Republic of the — Mauritania 1974
  - Congo, Republic of the — Niger ongoing
  - Congo, Republic of the — Reunion 1976
  - Congo, Republic of the — Senegal ongoing
  - Congo, Republic of the — Togo ongoing
  - Côte d'Ivoire — Burkina Faso ongoing
  - Côte d'Ivoire — Madagascar 1982
  - Côte d'Ivoire — Mali ongoing
  - Côte d'Ivoire — Mauritania 1974
  - Côte d'Ivoire — Niger ongoing
  - Côte d'Ivoire — Reunion 1976
  - Côte d'Ivoire — Senegal ongoing
  - Côte d'Ivoire — Togo ongoing
  - Denmark — Faroe Islands ongoing
  - Denmark — Greenland ongoing
  - Djibouti — Benin 1949
  - Djibouti — Burkina Faso 1949
  - Djibouti — Cameroon 1949
  - Djibouti — Central African Republic 1949
  - Djibouti — Chad 1949
  - Djibouti — Comoros 1949
  - Djibouti — Congo, Republic of the 1949
  - Djibouti — Côte d'Ivoire 1949
  - Djibouti — Gabon 1949
  - Djibouti — Guinea 1949
  - Djibouti — Madagascar 1949
  - Djibouti — Mali 1949
  - Djibouti — Mauritania 1949
  - Djibouti — Niger 1949
  - Djibouti — Reunion 1949
  - Djibouti — Senegal 1949
  - Djibouti — Togo 1949
  - Djibouti — Côte d'Ivoire 1949 (entry repeated in source list)
  - Guyana — Trinidad and Tobago 1971
  - India — Maldives 1966
  - India — Mauritius 1966
  - India — Pakistan 1966
  - India — Seychelles 1966
  - Kenya — Somalia 1971
  - Kenya — Tanzania 1978
  - Kenya — Uganda 1978
  - Lesotho — Swaziland ongoing
  - Madagascar — Burkina Faso 1982
  - Madagascar — Mali 1960
  - Madagascar — Mauritania 1974
  - Madagascar — Niger 1982
  - Madagascar — Reunion 1976
  - Madagascar — Senegal 1982
  - Madagascar — Togo 1982
  - Malawi — Zambia 1967
  - Malawi — Zimbabwe 1967
  - Malaysia — Singapore 1971
  - Maldives — Mauritius 1967
  - Maldives — Pakistan 1971
  - Mali — Burkina Faso ongoing
  - Mali — Mauritania 1974
  - Mali — Niger ongoing
  - Mali — Reunion 1976
  - Mali — Senegal ongoing
  - Mali — Togo ongoing
  - Mauritania — Burkina Faso 1974
  - Mauritania — Niger 1974
  - Mauritania — Reunion 1974
  - Mauritania — Senegal 1974
  - Mauritania — Togo 1974
  - Mauritius — Seychelles 1976
  - Montserrat — Saint Kitts and Nevis ongoing
  - Montserrat — Saint Lucia ongoing
  - Montserrat — Saint Vincent and the Grenadines ongoing
  - Montserrat — Trinidad and Tobago 1976
  - Netherlands Antilles — Suriname 1994
  - New Caledonia — French Polynesia ongoing
  - New Caledonia — Vanuatu 1971
  - New Caledonia — Wallis and Futuna ongoing
  - New Zealand — Samoa 1967
  - Niger — Burkina Faso ongoing
  - Niger — Reunion 1976
  - Niger — Senegal ongoing
  - Niger — Togo ongoing
  - Nigeria — Sierra Leone 1965
  - Oman — India 1970
  - Pakistan — Mauritius 1967
  - Pakistan — Seychelles 1967
  - Portugal — Angola 1976
  - Portugal — Cape Verde 1977
  - Portugal — Guinea-Bissau 1977
  - Portugal — Mozambique 1977
  - Portugal — São Tomé and Príncipe 1977
  - Qatar — India 1966
  - Qatar — United Arab Emirates ongoing
  - Reunion — Burkina Faso 1976
  - Reunion — Senegal 1976
  - Reunion — Togo 1976
  - Senegal — Burkina Faso ongoing
  - Senegal — Togo ongoing
  - Somalia — Tanzania 1971
  - Somalia — Uganda 1971
  - South Africa — Botswana 1977
  - South Africa — Lesotho ongoing
  - South Africa — Swaziland ongoing
  - Spain — Equatorial Guinea 1969
  - Sri Lanka — India 1966
  - Sri Lanka — Pakistan 1967
  - Saint Kitts and Nevis — Saint Lucia ongoing
  - Saint Kitts and Nevis — Saint Vincent and the Grenadines ongoing
  - Saint Kitts and Nevis — Trinidad and Tobago 1976
  - Saint Pierre and Miquelon — Benin 1976
  - Saint Pierre and Miquelon — Burkina Faso 1976
  - Saint Pierre and Miquelon — Cameroon 1976
  - Saint Pierre and Miquelon — Central African Republic 1976
  - Saint Pierre and Miquelon — Chad 1976
  - Saint Pierre and Miquelon — Comoros 1976
  - Saint Pierre and Miquelon — Congo, Republic of the 1976
  - Saint Pierre and Miquelon — Côte d'Ivoire 1976
  - Saint Pierre and Miquelon — Djibouti 1949
  - Saint Pierre and Miquelon — Gabon 1976
  - Saint Pierre and Miquelon — Guinea 1969
  - Saint Pierre and Miquelon — Madagascar 1976
  - Saint Pierre and Miquelon — Mali 1960
  - Saint Pierre and Miquelon — Mauritania 1974
  - Saint Pierre and Miquelon — Niger 1976
  - Saint Pierre and Miquelon — Reunion 1976
  - Saint Pierre and Miquelon — Senegal 1976
  - Saint Pierre and Miquelon — Togo 1976
  - Saint Lucia — Saint Vincent and the Grenadines ongoing
  - Saint Lucia — Trinidad and Tobago 1976
  - Saint Vincent and the Grenadines — Trinidad and Tobago 1976
  - Tanzania — Uganda 1978
  - Togo — Burkina Faso ongoing
  - United Kingdom — Bahamas 1966
  - United Kingdom — Bermuda 1970
  - United Kingdom — Cyprus 1972
  - United Kingdom — Falkland Islands ongoing
  - United Kingdom — Gambia, The 1971
  - United Kingdom — Ghana 1965
  - United Kingdom — Gibraltar ongoing
  - United Kingdom — Iraq 1967
  - United Kingdom — Ireland 1979
  - United Kingdom — Israel 1954
  - United Kingdom — Jamaica 1969
  - United Kingdom — Jordan 1967
  - United Kingdom — Kenya 1967
  - United Kingdom — Kuwait 1967
  - United Kingdom — Libya 1967
  - United Kingdom — Malawi 1971
  - United Kingdom — Malta 1971
  - United Kingdom — New Zealand 1967
  - United Kingdom — Nigeria 1967
  - United Kingdom — Oman 1971
  - United Kingdom — Samoa 1967
  - United Kingdom — Sierra Leone 1965
  - United Kingdom — Somalia 1967
  - United Kingdom — South Africa 1961
  - United Kingdom — Saint Helena ongoing
  - United Kingdom — Tanzania 1967
  - United Kingdom — Uganda 1967
  - United Kingdom — Yemen, P.D. Rep. 1972
  - United Kingdom — Yemen 1972
  - United Kingdom — Zambia 1967
  - United Kingdom — Zimbabwe 1967
  - United States — American Samoa ongoing
  - United States — Bahamas, The ongoing
  - United States — Belize 1949
  - United States — Bermuda ongoing
  - United States — Dominican Republic 1985
  - United States — Guam ongoing
  - United States — Guatemala 1986
  - United States — Liberia ongoing
  - United States — Panama ongoing
  - Vanuatu — French Polynesia 1971
  - Vanuatu — Wallis and Futuna 1971
  - Wallis and Futuna — French Polynesia ongoing
  - Yemen, P.D. Rep. — India 1951
  - Yemen, P.D. Rep. — Kenya 1972
  - Yemen, P.D. Rep. — Somalia 1971
  - Yemen, P.D. Rep. — Tanzania 1972
  - Yemen, P.D. Rep. — Uganda 1972
  - Yemen — India 1951
  - Yemen — Kenya 1972
  - Yemen — Somalia 1971
  - Table C3 (concluded) — "Currency unions not covered in Glick and Rose (2002) sample" stated in source

### References (selected list as provided)
- Acemoglu, D., S. Johnson, and J. Robinson, 2001, “The Colonial Origins of Comparative Development: An Empirical Investigation,” American Economic Review, Vol. 91, pp. 1369-1401.
- Alesina, A., and R.J. Barro, 2002, “Currency Unions,” Quarterly Journal of Economics, Vol. 117, pp. 409-36.
- Alesina, A., R.J. Barro, and S. Tenreyro, 2002, “Optimal Currency Areas,” NBER Working Paper No. 9072 (Cambridge, MA: National Bureau of Economic Research).
- Amemiya, T., 1984, “Tobit Models: A survey,” Journal of Econometrics, Vol. 24, pp. 3-61.
- Baldwin, R., 2005, “The Euro’s Trade Effects,” Graduate Institute of International Studies, Geneva. Presented at the June 2005 European Central Bank Conference “What Effects is EMU Having on the Euro Area and its Member Countries?”
- Anderson, J.E., and E. van Wincoop, 2003, “Gravity with Gravitas: A Solution to the Border Puzzle,” American Economic Review, Vol. 93, pp. 170-92.
- Carrere, C., 2004, “African Regional Agreements: Impact on Trade With or Without CU’s,” Journal of African Economies, Vol. 13, pp. 199-239.
- Coe, D.T., and A. Hoffmaister, 1999, “North-South Trade: Is Africa Unusual?” Journal of African Economies, Vol. 8, pp. 228-56.
- Collier P., 1995, “The Marginalization of Africa,” International Labor Review, 134 (4-5), pp. 541-57.
- Collier P., and J.W. Gunning, 1999, “Explaining African Economic Performance,” Journal of Economic Literature, Vol. 37, pp. 64-111.
- Debrun, X., P. Masson, and C. Pattillo, 2005, “Monetary Union in West Africa: Who Might Gain, Who Might Lose, and Why?” Canadian Journal of Economics, Vol. 38, pp. 454-81.
- Fatás, A., and A.K. Rose, 2001, “Do Monetary Handcuffs Restrain Leviathan? Fiscal Policy in Extreme Exchange Rate Regimes,” IMF Staff Papers, Vol. 47, pp. 40-61.
- Feenstra, R.C., J.R. Markusen, and A.K. Rose, 2001, “Using the Gravity Equation to Differentiate Among Alternative Theories of Trade,” Canadian Journal of Economics, Vol. 34, pp. 430-47.
- Fielding, D., and K. Shields, 2001, “Modeling Macroeconomic Shocks in the CFA Franc Zone,” Journal of Development Economics, Vol. 66, pp. 199-223.
- Frankel, J.A., 2005, “Comments on Richard Baldwin’s ‘The Euro’s Trade Effects.’” Presented at the June 2005 European Central Bank Conference “What Effects is EMU Having on the Euro Area and its Member Countries?”
- Frankel, J.A., and A.K. Rose, 1997, “Is EMU More Justifiable Ex Post than Ex Ante?” European Economic Review Vol. 41, pp. 753-60.
- Frankel, J.A., and A.K. Rose, 1998, “The Endogeneity of Optimum Currency Area Criteria,” Economic Journal, Vol. 108, pp. 1009-25.
- Frankel, J.A., and A.K. Rose, 2002, “An Estimate of the Effect of Common Currencies on Trade and Income,” Quarterly Journal of Economics, Vol. 117, pp. 437-66.
- Glick, R., and A.K. Rose, 2002, “Does a CU Affect Trade? The Time Series Evidence,” European Economic Review.
- Honohan, P., and P. Lane, 2000, “Will the Euro Trigger More Monetary Unions in Africa?” Policy Research Working Paper No. 2393 (Washington, DC: World Bank).
- Limao N., and A.J. Venables, 2001, “Infrastructure, Geographical Disadvantage, and Transport Costs,” World Bank Economic Review, Vol. 15, pp. 471-79.
- Masson, P., and C. Pattillo, 2002, “Monetary Union in West Africa: An Agency of Restraint for Fiscal Policies?” Journal of African Economies, Vol. 11, pp. 387-412.
- Masson, P., and C. Pattillo, 2004, The Monetary Geography of Africa (Washington, DC: Brookings Institution).
- Persson T., 2001, “CU’s and Trade: How Large Is the Treatment Effect?” Economic Policy, Vol. 33: pp. 433-48.
- Pöyhönen, P., 1963, “A Tentative Model for the Volume of Trade Between Countries,” Weltwirtschaftliches Archiv, Vol. 90, pp. 93-99.
- Rodrik, D., 1998, “Trade Policy and Economic Performance in Sub-Saharan Africa,” NBER Working Paper No. 6562 (Cambridge, MA: National Bureau of Economic Research).
- Rodrick, D., A. Subramanian, and F. Trebbi, 2002, “Institutions Rule: The Primacy of Institutions over Geography and Integration in Economic Development,” NBER Working Paper No. 9305 (Cambridge, MA: National Bureau of Economic Research).
- Rose, A.K., 2000, “One Money One Market: Estimating the Effect of Common Currencies on Trade,” Economic Policy, Vol. 15, pp. 7-46.
- Rose, A.K., 2001, “CU’s and Trade: The Effect Is Large,” Economic Policy, Vol. 33, pp. 449-57.
- Rose, A.K., 2004, “A Meta-Analysis of the Effect of Common Currencies on International Trade,” NBER Working Paper No. 10373 (Cambridge, MA: National Bureau of Economic Research).
- Rose, A.K., 2005, “Does WTO Make Trade More Stable?” Open Economies Review, Vol. 16, pp.7-22.
- Rose, A.K., and E. van Wincoop, 2001, “National Money as a Barrier to International Trade: The Real Case for CU,” The American Economic Review: Papers and Proceedings, Vol. 91, pp. 386-90.
- Santos Silva, J.M.C., and Tenreyro S., 2004, “The Log of Gravity,” unpublished.
- Smith, C., 2002, “CU’s and Gravity Models Revisited,” Reserve Bank of New Zealand Working Paper No. 7.
- Subramanian, A., and N.T. Tamirisa, 2003, “Is Africa Integrated in the Global Economy?” IMF Staff Papers, Vol. 50, pp. 352-72.
- Tenreyro, S., 2001, “On the Causes and Consequences of Currency Unions,” Harvard University, (unpublished).
- Tenreyro, S., and R.J. Barro, 2003, “Economic Effects of CU’s,” NBER Working Paper No. 9435 (Cambridge, MA: National Bureau of Economic Research).
- Tinbergen, J., 1962, Shaping the World Economy: Suggestions for an International Economic Policy (New York: The Twentieth Century Fund).
- Yeats A., 1998, “What Can Be Expected from African Regional Trade Arrangements? Some Empirical Evidence,” Policy Research Working Paper No. 2004 (Washington, DC: World Bank).

*Appendix C, Tables C2–C3 and References as provided in the source PDF.*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0631.pdf_