What Can We Learn from the CFA Franc Zone? David Fielding Lambert Bamba Simeon Coleman Akira Nishiyama Anja Shortland Jean-Paul Azam Mike Bleaney Kevin Lee Kalvinder Shields David Stasavage OECD, Paris 21.02.2006 1. Benin; 2. Burkina; 3. Cote d Ivoire; 4. Guinea-Bissau; 5; Niger; 6. Mali; 7. Senegal; 8; Togo; 9. Cameroon; 10. C.A.R.; 11. Chad; 12. Congo; 13. Gabon; 14. Eq. Guinea 1
The CFA Franc Zone 14 countries: 12 former French colonies + 2 late additions. Two monetary unions: UEMOA + CEMAC. Two central banks: BCEAO + BEAC. Two currencies (both called Franc CFA). Both currencies pegged to the Euro (formerly the French Franc). The peg is maintained by the French Treasury; this frees up the African central banks monetary policy. Recent History Persistent Balance of Payments deficits in some CFA countries in the 1980s. High public and private borrowing from the central bank in some countries. 1994 devaluation. Reform of rules governing central bank lending. Now ΔM corresponds to ΔNFA in the medium term. Similar to a currency board, but with much greater flexibility. 2
Four Questions Does the Franc Zone promote regional integration? 50 years of data versus 5 in the Euro Zone. Has there been any substantial economic convergence? Have the monetary authorities made good use of the flexibility given to them? What is the impact of the system on the poorest households? Regional Integration (Economica 72: 683-704, 2005) Controlling for distance & language, what factors affect the volume of bilateral trade? And the degree of business cycle correlation? Does a fixed exchange rate matter? Does a common currency matter? What about the wider policy environment? 3
Regional Integration In the 1980s, being a member of the Franc Zone had an enormous impact on participation in regional trade. The effects in the 1990s were substantial but much smaller. In neither case does membership of the same currency area matter. Perhaps the 1990s more closely reflect an exchange rate stability effect, rather than a policy distortion effect. Marginal Effects of Common CFA Membership on Bilateral Trade 1980s Both CFA / 4.37 different currency 1990s Both CFA / 1.14 different currency Both CFA / 3.65 same currency Both CFA / 1.79 same currency Common 1.56 border Common 1.59 border 4
There is no evidence that CFA membership has any impact on business cycle correlations. More trade demand shocks transmitted across borders BCC. More trade more specialization more asymmetry in supply shocks BCC. Looking More Deeply at Economic Convergence We look at both nominal and real convergence indicators. Has there been any nominal convergence in the UEMOA since the Convergence Pact of 1999? How much real asymmetry remains in the UEMOA and CEMAC? 5
Nominal Convergence Indicators Inflation External deficit / GDP Budget balance / GDP Tax revenue / GDP Public wages / tax revenue Capital spending / tax revenue Nominal Convergence Indicators: (i) Inflation 0.8 0.6 0.4 0.2 0.0-0.2 82 84 86 88 90 92 94 96 98 00 Burkina Cote d Ivoire Mali Niger Senegal Togo norm 6
Nominal Convergence Indicators: (ii) Capital Spending 0.30 0.25 0.20 0.15 0.10 0.05 0.00 86 88 90 92 94 96 98 00 Burkina Cote d Ivoire Mali Niger Senegal Togo Benin norm Nominal Convergence Indicators: (iii) Tax Revenue 0.25 0.20 0.15 0.10 0.05 86 88 90 92 94 96 98 00 Burkina Cote d Ivoire Mali Niger Senegal Togo Benin norm 7
Nominal Convergence Performance Inflation External deficit / GDP Budget balance / GDP Tax revenue / GDP Public wages / tax revenue Capital spending / tax revenue Nominal Convergence Performance Inflation External deficit / GDP Budget balance / GDP Tax revenue / GDP Public wages / tax revenue Capital spending / tax revenue 8
Nominal Convergence Performance Inflation External deficit / GDP Budget balance / GDP Tax revenue / GDP Public wages / tax revenue Capital spending / tax revenue Real Convergence Measures We have 40+ years of data on price and output movements in the Franc Zone. Do the different economies face a similar macroeconomic environment? Are asymmetries smoothed out over time? How fast? Look at a typical shock causing prices (or output) to rise in the region. 9
Price Dynamics Output Dynamics 10
Real Convergence Measures Substantial heterogeneity in price and output shocks. Some patterns emerge: Gabon/Congo versus the rest. Price asymmetries are smoothed out (but not very quickly). Output asymmetries persist indefinitely. So there is no single monetary policy suitable for all countries. Monetary Policy The Franc Zone central banks are free to pursue an independent short-term monetary policy. But there is substantial macroeconomic heterogeneity across the member states: no single policy is ever best for all. So how active are the central banks? We look at the BCEAO. 11
What Drives the BCEAO Interest Rate? In the long-run, the BCEAO discount rate is constrained by the French rate (plus a risk premium that we estimate at 2.5%). In the short run, it has the ability to adjust the rate in response to aggregate UEMOA business cycle fluctuations. Does it make use of this ability? BCEAO discount rate France CB refinancing rate 10 2.5 60 132 period 12
What Drives the BCEAO Interest Rate? Probability of an Interest Rate Cut Variable 1 s.d. 2 s.d. 3 s.d. UEMOA inflation 1% 13% 62% Output gap 2% 24% 83% Govt debt / GDP 0% 2% 6% Foreign assets / GDP 3% 5% 93% What Drives the BCEAO Interest Rate? The BCEAO does respond in a systematic way to aggregate economic conditions in the UEMOA. But it is very cautious. Typical movements in prices and output have almost no impact on the likelihood of a change in the discount rate. Only extreme changes prompt action. 13
Country-Level Monetary Policy The BCEAO does still have some countryspecific monetary instruments. E.g. country-specific reserve requirements. There are some variations in these credit instruments. But they the do not vary systematically with economic activity. The CFA has delivered substantial benefits (price stability, trade integration). But there are potential costs for members of a monetary union with heterogeneous macroeconomic characteristics. There is very little macroeconomic convergence, especially in those areas beyond the direct control of the central banks. So monetary policy is extremely conservative. 14