FEER for the CFA franc

Abstract

We investigate the behaviour of the real effective exchange rates (REER) of the two CFA franc zone monetary unions – CEMAC and WAEMU – vis-à-vis their long-run equilibrium paths. A reduced form of the Edwards’ (1989) fundamentals equilibrium exchange rate (FEER) model is estimated using the Johansen's (1995) cointegration methodology, and equilibrium paths and associated misalignments are derived for the period 1970 to 2005. Our results suggest that, for both CEMAC and WAEMU, the fundamentals account for most of the exchange rates’ fluctuation: increases in the terms-of-trade, government consumption and productivity tend to appreciate the exchange rate, while increases in investment and openness tend to depreciate it. At end of 2005, we find no evidence that either the CEMAC or WAEMU REERs were significantly over-valued, which suggests that no exchange rate action is currently needed. Our analysis also reveals significant differences in the fundamentals’ marginal impact, and speed of reversion to equilibrium following a shock, which may raise questions about the desirability of maintaining the same parity for both monetary unions.

Acknowledgements

We thank the editor, Mark Taylor, Ralph Chami, Anne-Marie Gulde-Wolf, David Hendry, Lucio Sarno, and seminar participants in IMF seminar presentations for helpful comments and suggestions. The views expressed in this paper are those of the authors and do not necessarily represent those of the IMF or IMF policy.

Notes

¹ See Mussa (1974), Edwards and van Wijnbergen (1986), Obstfeld and Rogoff (1996) and Acemoglu et al. (2003).

² See Dornbusch (1982), Harberger (1986) and Hinkle and Montiel (1999).

³ CEMAC includes Cameroon, the Central African Republic, Chad, Republic of the Congo, Equatorial Guinea and Gabon. WAEMU includes Benin, Burkina Faso, Côte d’Ivoire, Guinea-Bissau, Mali, Niger, Senegal and Togo. See Hadjimichael and Galy (1997) and Masson and Pattillo (2004) for more details on the CFA franc zone.

⁴ Strictly speaking, there are two different currencies called CFAF, the Central African CFA franc and the West African CFA franc, which are distinguished by the meaning of the abbreviation CFAF. Both CFAFs are pegged to the euro at the same rate (655.96 CFAF per euro).

⁵ For evidence on the benefits of CFA membership see, for example, Ebaldawi and Madj (1996), Stasavage (1997), Fouda and Stasavage (2000) and Masson and Pattillo (2001, 2004).

⁶ See, for example, Williamson (1994), Faruqee et al. (1999), MacDonald and Stein (1999) and Wren-Lewis (2003). Also, Driver and Westaway (2004) provide a complete taxonomy of the different empirical approaches on the equilibrium exchange rate estimation in the literature.

⁷ The model is discussed in detail in Edwards (1989, 1994) and Williamson (1994). Cerra and Saxena (2002), Doroodian et al. (2002), Mathisen (2003) and Goh and Kim (2006) are applications of Edwards’ methodology.

⁸ Following the IMF convention, an increase in the REER is defined as an appreciation.

⁹ See Montiel (1999) for a discussion on this. It is noted, however, that most empirical studies using this framework tend to find a positive relationship between the REER and government consumption.

¹⁰ Openness, as a measure of trade restrictions is used by Montiel. Edwards uses two alternative measures (import tariffs as ratio of tariff revenues and the spread between the parallel and official rates), which he acknowledges to have important limitations.

¹¹ If the coefficient is zero in a particular equation, that variable is considered to be weakly exogenous and the VAR can be conditioned on that variable.

¹² As discussed in Quah (1992) and Maravall (1993), a unique decomposition between permanent and transitory components does not exist.

¹³ Moreover, as also discussed in Cerra and Saxena (2002), if simple smoothing processes were sufficient to arrive at the equilibrium values for the fundamental series, then presumably, the same smoothing process could be employed to arrive at the equilibrium real exchange rate series, without the need for the VECM estimation.

¹⁴ Applications of the GG decomposition are also Alberola et al. (1999), Cerra and Saxena (2002) and Mathisen (2003).

¹⁵ This approach has been used by Osbat et al. (2003) and Engels et al. (2005). Another methodology to construct asymptotic SEs for the GG is discussed and applied in Alberola et al. (1999).

¹⁶ The ADF test results in Table B1 of Appendix B are based on a specification with a constant term included. We experimented with specifications that include both a constant and a deterministic time trend. The results are virtually unchanged from those reported in Table B1 in Appendix B.

¹⁷ See also the discussion in Cerra and Saxena (2002).

¹⁸ The estimated BFDIR coefficients for WAEMU (CEMAC) are 0.012 (0.018) with t-statistics 1.17 (1.42). The rest of the fundamentals are virtually unchanged with or without BFDIR.

¹⁹ For example, the 1994 impulse dummies take the value of one at year 1994 and zero otherwise.

²⁰ Cameron started oil production in 1976, which reached its peak in 1985. Since then, oil production steadily declined through the mid-1990s.

²¹ For more details on the test statistics, see Doornik and Hendry (2001).

²² The diagnostic tests mentioned in the text are tests performed on each equation of the VAR separately. Vector/system tests performed on the entire system yield the same results as those for the single-equation tests.

²³ See Doornik and Hendry (2001) for more details on the various Chow tests.

²⁴ As discussed, the choice of the degree of smoothing is arbitrary with larger (smaller) factors generating smoother (less smooth) equilibrium real exchange rate paths. As a robustness check, the equilibrium real exchange rates in Figs 1 and 2 are derived by applying to the explanatory variables an HP filter based on the average of five smoothing factors (10, 30, 50, 100 and 300).

²⁵ For a review of the use of the purchasing power parity theory in understanding and modelling real exchange rate behaviour, see Sarno and Taylor (2004).

²⁶ The base year represents the equilibrium and the misalignment is estimated as the deviation of the REER in the final year of the analysis from the equilibrium base year.

²⁷ This analysis is based on estimated elasticities and cumulative variable changes estimates, similar to the ‘sources of growth’ accounting.

²⁸ Using a variety of estimation methods and samples, Cashin et al. (2002) estimates an average adjustment speed with half-life of 3 months and 4 months for several WAEMU and CEMAC countries, respectively; Mathisen (2003) estimates a half-life of less than a year for Malawi; MacDonald and Ricci (2003) estimate a half-life of 2–2.5 years for South Africa; Rogoff (1996) estimates the longest half-life of 3–5 years for a variety of countries.

²⁹ Generalized impulses are used. As described by Pesaran and Shin (1998), this method constructs an orthogonal set of innovations that does not depend on the VAR ordering.

³⁰ Results in the literature point to the fact that only significant misalignments that are sustained for protracted periods could lead to currency crises. See, for example, J. P. Morgan's ‘Emerging Markets Real Exchange Rate Model’ (2000) and Sarno and Taylor (2002).