Documento(s) 14 de 20

Medhora reviews the experience of the West African Monetary Union (UMOA), which became the West African Economic and Monetary Union (UEMOA) in January 1994. A review of the literature shows the economic performance of UMOA countries to have been relatively good in comparison to that of other Sub-Saharan African countries, particularly in terms of the low rates of inflation which have been observed. This may be attributed to various factors, including the discipline imposed by the fixed exchange rate, the existence of a relatively independent, supranational bank, and the French guarantee of convertibility for the CFA franc. However, the author also discusses some of the problems UMOA has had to contend with. He seeks to draw lessons from this experience for other groups of developing countries that are contemplating monetary integration in West Africa or elsewhere. A monetary union is more likely to be successful if the supranational central bank is empowered to override national authorities in major areas of money and finance. In exchange for giving up sovereignty over monetary matters, members gain the advantage of stable, noninflationary monetary policies insulated from political interference. Risk and transactions costs are reduced, and this could help stimulate investment and growth, while encouraging greater economic integration among the member states.

¹ Although I retain sole responsibility for the final contents, I wish to thank Youssouf Dembele, Abdoulaye Diagne, Henri Josserand, Sams Dine Sy, Ousmane Badiane, Jean Coussy, Mohammed Mah’Moud, and Diery Seck for their comments on an earlier version of this paper.

THE WEST AFRICAN Monetary Union (UMOA), which became the West African Economic and Monetary Union (UEMOA) on 10 January 1994 is one of the world’s most far-reaching examples of monetary integration. The Union’s monetary arrangements, which remain unchanged by the expansion of UMOA’s functions into the economic sphere under UEMOA, make it a “complete” monetary union in the sense that its members share a fully convertible common currency issued by a supranational central bank that oversees the operations of an external reserve pool. There are seven member countries: Benin, Burkina Faso, Côte d’Ivoire, Mali, Niger, Senegal, and Togo. The common currency is the CFA franc, issued by the Banque Centrale des États de l’Afrique de stratégies (BCEAO), based in Dakar. Despite statutory and other limitations on action by individual governments, UMOA member countries retained some leeway — too much leeway, perhaps — in fiscal, trade, and monetary matters, and the need for greater harmonization of economic policies was one of the principal reason for the transformation of UMOA into UEMOA.

The Union’s monetary arrangements have obvious appeal. They include the use of a fully convertible currency backed by a G-7 country, risk-free investment within the franc zone, economies of scale resulting from the issue of a common currency, and the existence of an apolitical central bank that can pursue consistent policies. The cost to member countries has been the loss of the exchange rate as an instrument of macroeconomic policy. Although Cobham and Robson (this volume) consider this cost to be of secondary importance, UMOA’s inability even jointly to devalue the CFA franc was the subject of much debate as those countries struggled to adjust without devaluation, through the dreadful 1980s and early 1990s. Having now succeeded in devaluating the CFA franc in January 1994, and having managed the process as well as could be hoped so far, UMOA now appears better equipped than in the past to meet the challenges of adjustment.

The OAU’s “Abuja Declaration” of 1991 and numerous statements by West African institutions seem to suggest a strong desire to enhance monetary integration, and UMOA’s nearly two decades of experience since it last went through major institutional changes provides fertile ground to address such issues. This chapter presents a review and analysis of UMOA’s experience to date. Its future prospects are examined, and lessons for generalizing this model to other countries in the region are considered.

INSTITUTIONAL FRAMEWORK AND ECONOMIC PERFORMANCE

UMOA’s institutional arrangements have received attention elsewhere and will only be discussed here briefly.² The Union is headed by the Conference of the Heads of State, and the BCEAO is headed by a Council of Ministers, comprising the finance minister of each member country. Each member country has input into the BCEAO’s decision-making process, and, nominally, all members including France have one “vote” each. In practice, decisions are made by consensus. However, the era of structural adjustment has brought considerable influence from abroad, and it is apparent that de facto, some members are more equal than others.

² The standard reference for a critical analysis of the economic arrangements of the franc zone remains Guillaumont and Guillaumont (1984). This book takes a decidedly optimistic view of UMOA, which continues to inspire the authors in later work (Guillaumont et al. 1988, for instance). Other works include that of Julienne (1988), which provides a more descriptive and anecdotal account of the early years of the BCEAO (1955–75), and more critical, but still positive overviews (Vinay 1980; Bhatia 1985; Neurrisse 1987; Vallée 1989; Medhora 1992a). Sacerdoti (1991) presents a recent description of events and arrangements in the region. The formal statutes governing the UMOA and BCEAO may be found in UMOA (1962, 1973, 1989).

The BCEAO conducts an annual programming exercise to estimate credit requirements, both Union-wide as well as by member country, before it makes its credit allocation decisions. By statute, government borrowing from the BCEAO is limited to 20% of the government’s previous year’s fiscal receipts. Governments are free to borrow at home and abroad and are required only to inform the central bank of their actions.

In normal circumstances, each member is required to contribute 65% of its external reserves to an operations account run by the BCEAO and maintained with the French Treasury in Paris. Any serious payments imbalance of one member is covered by the external assets of the others. If the account as a whole runs low, the remaining 35% of the reserves may be called in. If this, too, is inadequate, a vaguely defined “crisis management” scheme takes over, but, more importantly, the French Treasury stands ready to augment the account, in an unlimited and unconditional manner. This, in practice, is how the French guarantee of full convertibility of the CFA franc is established. This system allowed the exchange rate to remain unchanged at 50 CFA francs to 1 French franc from 1948 until January 1994, when it was changed, overnight, to 100 CFA francs to 1 French franc.

The advantages and disadvantages of the UMOA system have been the subject of a growing body of literature regarding the impact of UMOA membership on various indicators of economic performance. For a long time, the obvious stability and well-being of UMOA countries was contrasted with the “learning by doing” pratfalls of the nonfrancophone African countries, and a positive association developed between UMOA-type arrangements and development. A permanently fixed and externally guaranteed exchange rate coupled with a supranational central bank was expected to breed low inflation and encourage savings, investment, possibly exports, and, ultimately, a higher rate of growth. However, fixed exchange rates increasingly fell into disrepute in the 1970s and 1980s, specifically in the case of UMOA, as it struggled with problems of structural adjustment and increased loss of competitiveness on world markets.

Devarajan and de Melo (1987) examined the impact of UMOA by comparing growth in GNP in the franc zone with 11 sets of other countries, variously defined, during the period 1960–82. Using a variance-components model, they find that for the entire period, growth in the CFA countries was higher than that in other Sub-Saharan African countries, but lower, or not significantly different, than that of other groups. However, stronger results were obtained for the CFA group relative to comparable groups when the sample period was divided into 1960–73 and 1973–82. With the CFA group of countries improving its performance from one period to the other, the authors concluded that “the discipline imposed by monetary union participation was helpful for adjustment during the period of generalized floating and supply shocks” of the more turbulent latter period (Devarajan and de Melo 1987, p. 493).

This somewhat crude test was supplemented by Guillaumont et al. (1988) who specified a growth equation and ran econometric tests to compare actual growth with “predicted” growth for a sample of CFA and non-CFA countries. Here, the CFA group compared favourably with others. During the subperiod 1970–81, the “unexplained” component of growth attributable to institutional or other causes was superior to that of other Sub-Saharan African countries as well as non-African developing countries. However, the result is conditioned by the strong performance of the larger countries (Côte d’Ivoire and Cameroon). The Central African Republic, Chad, Gabon, Senegal, and Togo fared relatively poorly compared with the non-African group of developing countries. This hints at the possibility that not every member may benefit, or benefit equally, from the arrangements of the franc zone and suggests the need for further research on this topic.

Elbadawi and Majd’s (1992) more recent findings are more pessimistic. A modified-control-group approach, which controls for initial conditions and changes in the internal, external, and policy environment, was used to measure the marginal impact of membership in the franc zone relative to a group of Sub-Saharan African and a group of low-income countries. This study yielded very different results for the 1970s and 1980s. In the 1970s, zone membership had a largely positive effect on all variables (the sole exception was growth compared with the low-income developing countries). However, during the 1980s, zone membership had a negative effect on growth, exports, investment, and savings compared with both comparison groups. These results echo Devarajan and de Melo’s (1990) previous work using a similar method.

The conclusion may be drawn that the arrangements of the franc zone may have “worked” over the long haul, but were unable to prevent or sustain the shocks of the past decade. What was ominous for the long term was that overvaluation of the currency during this latter period was fought with expenditure reductions that fell largely on investment spending, thus boding ill for future productive capacity and growth in the region. One bright spot emerges: no matter what time frame and what comparison group, inflation was unambiguously lower in the franc zone.

Linkage of the CFA franc to the French franc is expected to have a stabilizing influence on the economies of the franc zone for various theoretical and institutional reasons. Whether this transpires empirically depends on the extent to which inflation in the CFA countries is found to track French inflation. It also depends on the stability of the French franc relative to other world currencies. Evidence regarding the correlation between inflation in the franc zone and French inflation suggests an imperfect relationship. De Macedo (1986) found inflation in UMOA to have been largely insulated from that in France between 1958 and 1982. If anything, there was a negative correlation. However, later work, by Honohan (1990a) did find the expected tendency for inflation rates to converge, between 1964 and 1987.

Other studies have directly addressed the issue of real exchange-rate stability in the franc zone relative to other regions. Paraire (1988) and Honohan (1983) found lower or statistically indistinct nominal and real exchange-rate variability in the CFA countries relative to other developing countries. Elbadawi’s (1991) results were similar through to 1988, but he found differences in exchange-rate variability between the CFA countries, as a result of differing rates of inflation from country to country. Exchange rate variability in the franc zone increased in the late 1980s, but remained low by developing country standards.

A few broad conclusions may be derived from this literature. Historically, UMOA countries have enjoyed favourable growth rates compared with other Sub-Saharan African countries. Arguably, given the potential for omitted variables and specific regional circumstances, this is the only valid comparison group, and unfavourable comparisons with Asian and other developing countries should be taken less seriously. However, the stronger growth rate of UMOA countries has been slipping, even with respect to this narrow group. This can, at least partly, be attributed to the inability of the central bank to control inflation in the late 1970s and early 1980s and the overvaluation of the currency

resulting at least in part from such inflation. The poor performance of savings and investment is probably linked to the ongoing financial-sector crisis in the region and capital flight due to the overvalued CFA franc. Overvaluation of the CFA franc also offers an obvious explanation for the region’s weak export performance. Still, the arrangements have produced lower inflation and more stable real exchange rates, thanks to the common central bank and convergence to French indicators.

THE BCEAO

As the institution in charge of monetary policy in seven countries, the BCEAO has an onerous responsibility, for which it is only partly equipped. We consider here a number of features conditioning BCEAO’s ability to deliver sound monetary policy:

• its relatively high degree of independence;
• a de facto bias toward its larger member countries, as revealed by the pattern of seigniorage flows;
• its inability to prevent a financial sector crisis in some countries;
• its inability to manage the level of macroeconomic activity in some countries; and
• its inability to prevent overvaluation of the CFA franc.

CENTRAL BANK INDEPENDENCE

The importance of central bank independence can be appreciated in the context of Kydland and Prescott’s (1977) notion of “time consistent” and “time inconsistent” policies, that so changed the tone of the “rules versus discretion” debate in monetary policy. In that view, an announced “role” will not be credible (i.e., time consistent) if it is merely announced and not underpinned by forces that make it irrevocable. On the other hand, policy announcements may still be time consistent, despite the lack of an explicitly stated rule, if they are always seen to be enforceable. Such arguments have been used to explain why some central banks that tried switching to strict money roles failed (those of the United States and Canada in the early 1980s), whereas others that have no explicit money rules have been more successful in managing inflation (Germany and Japan). The reason is that credibility cannot be acquired by simply announcing a rule. It has to be earned, either through reputation (which presumably explains the success of the central banks in Germany and Japan) or through fundamental statutory changes (as in New Zealand, Chile, and possibly Canada).

Monetary announcements are deemed to be credible if they are made by an authoritative source, i.e., an independent, apolitical, technocratic central bank, preferably with a history or statutory commitment to a single, achievable goal, such as low and stable inflation. Grilli et al. (1991) have composed an index of central banks’ independence in the OECD countries. Applying the index reveals that: central bank independence may be classified as “political,” “economic,” or both and that one usually goes with the other; in the post-war era (1950–89), the more independent central banks are associated with lower rates of inflation; and this result does not come with a real cost in terms of output growth. These results may be seen as the empirical application of the Kydland and Prescott (1977) proposal, and the strongest evidence to date in its favour.³

The application of this analytical framework to the BCEAO shows it to have an enviable degree of independence and, therefore, credibility. The BCEAO lies above the average for OECD central banks, largely because of its freedom from pressure from any single fiscal authority and the statutory limits on government borrowing (Figure 1).⁴ This degree of independence has delivered lower inflation, as indicated earlier, independently of whatever else might be said about the performance of UMOA in the late 1980s and early 1990s.

³ For a full and recent discussion of the design and practice of central banking, see Downes and Vaez-Zadeh (1991).

⁴ See Appendix to this chapter for details on the derivation of the index of central bank independence.

SEIGNIORAGE

Seigniorage is the command’ over real resources that a central bank captures by issuing “high-powered” money. This monetary seigniorage is then used for central bank operating costs, retained, or returned to the public via dividends to the national treasury or via subsidized lending to designated sectors. How the benefits of seigniorage are used is an issue for public and central bank policy.

In UMOA, there is the additional issue of distributing seigniorage rights between the member countries; this is examined in Honohan (1990b) and Medhora (1992b, 1993). Dividends, until they ceased in 1987, were allocated equally among the members of the Union, and thus had a built-in bias toward the smaller (and coincidentally poorer) members. The alternative of distributing seigniorage via subsidized lending in effect means that the larger members (or, at least, those members with large sectors entitled to the preferred credit) benefit most.

Figure 1. Central Bank independence: OECD and BCEAO. Source: derived from Grilli et al (1991); see also Appendix 1.

The BCEAO’s active role in resolving the financial sector crisis alluded to below resulted in a marked shift in the pattern of seigniorage allocation. For example, comparing the boom years (1976–80) with the bust years (1984–89), subsidized lending and operating costs grew as a proportion of total seigniorage, at the expense of dividends (Figure 2).

This had profound implications in the intraregional distribution of seigniorage (Figure 3). Most remarkable is Côte d’Ivoire’s share, which is not proportional to any indicator of its relative size within the region, and has grown at the expense of the smaller members’ shares. The tighter monetary policy of the late 1980s has also had an effect on the availability of dividends for distribution, leading to an increase in the share of seigniorage benefits being consumed by operating costs (Figure 1).

With the elimination of the “taux d’escompte préférentiel,” one source of seigniorage for the members has been removed, but the BCEAO’s role in restructuring insolvent banks will ensure a similar pattern of distribution in the foreseeable future. What should be an issue of macroeconomic and central banking policy and intraregional relations has become caught up in the whirl of crisis management. One could make a valid case for the central bank distributing seigniorage equally among its members, or unequally to favour a target group of countries, but it is more difficult to defend an allocation policy based on compensation for financial mismanagement!

Figure 2. Distribution of seigniorage by use (average annual flows, as%).
Source: Medhora (1992b).

THE FINANCIAL-SECTOR CRISIS

The ongoing financial-sector crisis, primarily in Côte d’Ivoire, Senegal, and Benin, has drained the region of savings and scarce management resources and set back financial development. A combination of a weak external and domestic economic environment, coupled with poor lending practices and inadequate supervision by the authorities are to blame for this situation.⁵

⁵ For an overview of financial-system management — and mismanagement — in developing countries, see World Bank (1990). For some detail of the African situation see Seck and El Nil (1992) and Callier (1991). Brief descriptions of the UMOA situation may be found in Honohan (1990b) and de Zamaroczy (1992), as well as in recent annual reports of the BCEAO.

Figure 3. Distribution of seigniorage by country (country shares in total seigniorage, as %).
Source: Medhora (1992b).

The designation of some sectors as privileged, entitled to funds at the BCEAO’s below-market discount rate led to abuse of the system. Problem institutions were either not identified or not dealt with soon enough. After passively refinancing credit to governments, the agricultural sector, and designated, entities for years, the BCEAO found itself dealing with the aftermath of a crisis among lending institutions at the end of the 1980s. The familiar “too big to fail” logic meant that the central bank was also the chief, perhaps only, institution in the region deemed capable of rectifying the situation with minimal damage. The result was that large amounts of assets, both performing and nonperforming, were rediscounted by the central bank.

A strong case can be made that a central bank — particularly that of a monetary union — should not have to deal with financial-sector regulation. If it must do so, it should be given all the powers normally associated with domestic financial-sector regulation agencies. Central banks in developing countries have traditionally been given a primary role in financial-sector regulation and development because of the obvious links between credit creation, financial intermediation, and economic development.

However, when compulsory subsidized lending is injected into this equation, the risk of abuse increases, and monetary policy becomes “hostage” to such abuse.

A system of market-based interest rates combined with tighter supervision and an internally financed insurance scheme for the financial sector would, in principle, unlink monetary policy from financial-sector events. The move to a more market-based system and the creation of an autonomous Commission Bancaire at the Union level, as spelled out in UMOA (1989), goes a long way toward addressing the problem. It is not entirely clear how the regulatory powers of the new agency and its monitoring capacity will coincide with those of the BCEAO, but, on pure central banking grounds, the BCEAO could make a strong case for having this aspect given over almost completely to the new agency.⁶ In any case, the BCEAO is not rid of the matter entirely, as it will almost certainly continue to be involved in the resolution of financial crises.

⁶ This “narrow” view of central banking operations, although popular in some quarters in discussions on the creation of a European central bank, is not universally accepted. Folkerts-Landau and Garber (992) argue that the separation of responsibilities for price stability and financial-sector supervision could impede the creation of truly regional financial markets, making the central bank no more than a “monetary policy rule.” They address the charge that credit allocation abuse may compromise the functioning of a common central bank by arguing that an independent central bank with its own resources at stake is more likely than an independent monitoring agency to assess accurately the solvency of potential borrowers (p. 30). However it may be useful to differentiate between a liquidity crisis and a solvency crisis. The former is very much the responsibility of a central bank; the latter, in UMOA, has been the result of poor lending policies and practices by the BCEAO. A self-contained insurance and regulatory body whose mandate and accounts are transparent is at least as likely to be rigorous in its operations, and possibly more so, than a central bank, except perhaps one that is exceptionally independent.

DOMESTIC CREDIT CREATION AND EXTERNAL BORROWING

The BCEAO’s role in macroeconomic management of the franc zone depends upon its willingness and ability to manage levels of credit and economic activity in the member countries. The BCEAO’s control over money supply is exercised principally through the 20% rule described earlier. Unfortunately, the 20% rule tends to be viewed more as a right than a ceiling, and credit expansion at such levels may not always be appropriate.

One assumes a desire to control the expansion of domestic credit on the part of the BCEAO, because the BCEAO is largely free from political influence, and its mandate includes defending the historical parity of the CFA franc with the French franc. However, the ability of the BCEAO to control domestic credit is less certain.⁷ Its member countries have small open economies characterized by perfect capital mobility and a fixed exchange rate; and the level of economic activity in individual countries will be conditioned by the amount of external borrowing and its positive effect on the national money supply whenever hard currency is exchanged for CFA francs. Such borrowing is likely to reflect the credit-worthiness of the borrowing countries and the degree of international confidence in the CFA franc. Countries in need of funds may also tap into,the surplus reserves of other member countries, through access to the BCEAO reserve pool, or into the French Treasury which stands ready to replenish the BCEAO’s operations account when necessary. Countries running large fiscal deficits can thus draw upon three sources of credit that are beyond the control of the BCEAO: international financial markets; deficit financing from the BCEAO’s reserve pool; and contributions to the operations account by the French Treasury. All this substantially circumscribes the BCEAO’s ability to manage the level of economic activity in individual countries as well as the Union as a whole.

⁷ Domestic credit creation in UMOA has been the subject of some discussion in Bhatia (1971, 1985), IMF (1963, 1969), and Medhora (1992a).

This situation underscores the importance of the BCEAO’s monitoring role with regard to external borrowing. Unfortunately, even that monitoring role has been handicapped. Although the central bank is supposed to estimate credit demand and supply in each member country annually, the reporting requirements regarding external borrowing by individual governments is weak. It is not at all clear that the BCEAO has been operating with “full information” in this regard, both in terms of the completeness, as well as the speed of reporting. A tightening of the relevant statutes or some definition of a mechanism by which the BCEAO is empowered to oversee credit developments within its region seems in order. Put another way, a supranational central bank cannot behave like a supranational central bank if it does not have the mandate to do so. At the very least, a free and full flow of relevant information seems well within the limits of what such a mandate should entail.

The most obvious reflection of how economic management has varied from country to country is indicated in the different rates of inflation which are observed. Between 1977 and 1990, the consumer price index rose by 130% in Côte d’Ivoire, 118% in Senegal, 100% in Burkina Faso, 74% in Togo, and 61% in Niger (similar data for Benin are not available). These differences highlight not so much a failing in monetary policy in the Union, as a lack of coordination between the BCEAO’s policies and the fiscal policies of individual members.

Such differences will bear, in turn, on the economic competitiveness of these countries, as reflected in the evolution of the real effective exchange rate (Figure 4). Less obvious is the general loss of international competitiveness of the franc zone through the 1980s and early 1990s. Although this competitiveness depends on the relative level of inflation in the zone as reflected by the real effective exchange rate, it also depends upon the evolution of the terms of trade for traditional exports, the burden of debt service the region is obliged to bear, the zone’s ability to develop nontraditional exports, and the exchange rate policy of competing countries (such as Ghana and Nigeria or other developing countries). Evolving trends in all these variables combined to make the parity level of the CFA franc increasingly uncompetitive through the 1980s and early 1990s. The 50% devaluation of the CFA franc in January 1994 was the confirmation of what everyone had come to know about the continued and secular uncompetitiveness of the franc zone at the old parity.

Figure 4. Real effective exchange rates (1978 = 100).
Source: IMF, International Financial Statistics.

ADJUSTMENT WITHOUT DEVALUATION

That the inevitable should have been put off for so long can be explained in many ways, and is not unique to UMOA’s case. Individual countries have hobbled along with overvalued exchange rates before. Yet the UMOA case is special for two reasons. One is the illustration it provides of the special difficulty of exchange rate devaluation in a monetary union. The second is the lesson to be derived from its efforts to adjust “without devaluation.”

What complicates devaluation in a monetary union, is that the different countries have different needs. It was commonly acknowledged that Côte d’Ivoire and Senegal needed to devalue much more than Niger or Togo at the other extreme, and the data on inflation in these countries bore this out. Coming to agreement on an appropriate rate of devaluation under those circumstances was problematic at best, and it is not surprising that Côte d’Ivoire was in the lead of countries calling for devaluation.

There were also some good reasons to resist devaluation: the fear of destroying confidence in the UMOA system, the fear of spiraling inflation, and fears of repeated, and ineffectual devaluations, as have occurred elsewhere.

The special link of UMOA to France delayed the process for two reasons. One was France’s opposition to devaluation, for reasons mentioned above and others of a geopolitical nature. The other was France’s readiness to facilitate any attempt to adjust without devaluation.

The upshot was that the UMOA countries attempted to adjust in the classic manner of the gold standard, with France’s augmentation of the operations account softening the blow of deflation for member countries. Indeed the franc zone as a whole was a net drain on the French Treasury more or less continuously for some years after 1987, according to the calculations of M’Bet and Niamkey (1990).⁸ One may argue that this promise of convertibility — open-ended and unconditional as it has been — only delayed the inevitable. The incentive to remain competitive is weaker in the presence of such a guarantee, and it no doubt introduces a form of “moral hazard,” if not in the actions of the BCEAO, then at least in the actions of some of the governments, some of the time.

⁸ Interestingly, this is a recent phenomenon, encountered previously only in 1983, and then in a minor way.

Of course, it would be quite wrong to characterize the past decade as one where adjustment was merely postponed. It could be argued, as Collier and Gunning (1992) do, that the guarantee of convertibility allows UMOA governments to pursue policies (such as trade liberalization or industrial diversification) that have long-term payoffs but short-term costs, as these countries are not bound, to the usual extent, by immediate foreign-exchange constraints. This is a luxury that most developing countries cannot claim to have. UMOA countries have, with varying degrees of success, initiated adjustment policies.

However, exchange-rate adjustment without devaluation (“artificial” devaluation) requires direct — therefore visible and contentious — pressure on wages, costs, prices, and the public-sector deficit. A nominal devaluation can achieve the same result less transparently and more quickly. It, too, needs to be accompanied by other austerity measures, but presumably these are more severe in the absence of an accompanying nominal devaluation. In a survey of artificial devaluations, Laker (1981) considers them to be second-best solutions to a nominal devaluation. Judging by the theoretical literature as well as UMOA countries’ experience of the last decade or so, countries that attempt this approach to correct a disequilibrium of major proportions should expect one or more of the following:

• some public resistance as wage and cost controls and higher tariffs on imports are announced;
• allocative inefficiencies, because the reach of such a plan would never be as universal as that of a nominal devaluation;
• capital flight due to uncertainty about the success of the operation;
• a proliferation of administrative costs and corruption, as a system of increased import tariffs and export subsidies is put into place;
• an imbalance in the structure of economic incentives due to limitations on the ability to subsidize exports in the face of fiscal pressures; and
• the risk that after all the time and pain, a competitor may achieve the same result by simply devaluing.

UMOA’s experience of the 1980s and early 1990s, culminating in the devaluation of the CFA franc after 10 years of procrastination, suggests the need for a degree of flexibility in the formula for pegging the Union’s common currency. It remains to be seen whether future adjustments in UMOA may be easier now that the barrier of devaluation has been breached.

LESSONS FOR OTHERS

The UMOA experience is replete with lessons regarding the advantages and costs of monetary integration, the limits which should be imposed on the role of the central bank, and exchange rate management.

A “first principles” approach to monetary union must necessarily find its way through the literature on optimum currency areas, initiated by Mundell (1961) and surveyed by Ishiyama (1975). However, using a single criterion to determine the “optimum” area, as some of the early theorists did, is of limited practical use. Ishiyama concluded his survey of the issue by proposing a more flexible cost-benefit approach. To Canzoneri and Rogers (1990), this means exploring the trade-off between the loss of sovereignty over monetary policy (and with it control over the use of seigniorage as a form of tax revenue), and the savings in transaction costs inherent in a single-currency system. The feasibility and desirability of monetary integration also depends on historical, regional, and political considerations. For a small country overwhelmed by regional events, there may well be no choice at all. In all cases, only rough and ready “calculations” can be made about joining, because the counterfactual is itself purely hypothetical. What matters is not the optimality of the solution, but its practicality.

The adjustment problems that UMOA has had to face in the past decade should not blind us to the benefits of membership in a monetary union of this sort, nor lead us to overestimate the benefits of sovereignty in matters of monetary policy. However, we should be aware that at least some of the desirable features of UEMOA (such as the French guarantee of convertibility) will not accrue to other regional groupings, or an expanded “franc zone.”

Furthermore, membership in a monetary union largely involves “buying into a package,” involving substantial changes in the way individual countries conduct their affairs in some areas of policy. Monetary union has strong fiscal overtones. There is no a priori reason why a common currency area should also have uniform intraregional trade and taxation policies; for obvious examples, consider the creative array of trade impediments and tax regimes within the United States, Canada, and the “united” Europe.⁹ However, the creation of a common central bank means that individual governments have little discretion in the use of the inflation tax. On the assumption that the common central bank will choose a policy geared to inflation at the low end of the regional spectrum (or lower still), members with historically high rates of inflation will face a larger adjustment than those with inflation rates close to those sought for the union as a whole.¹⁰

9 There is a large body of literature on the links between monetary union and fiscal coordination. Bhatia (1985) contains a brief survey, and Allen (1976) and Robson (1971) deal with the nuts and bolts of the issue. For recent discussions on the subject in the European and US contexts, see de Cecco and Giovannini (1989) and Eichengreen (990), respectively. Although a monetary union can function, at a certain level, in the absence of fiscal and trade policy harmonization and factor mobility, it will do so more efficiently in a more economically integrated setting. UMOA has an ongoing history of attempts at enhanced integration in nonmonetary areas and will be pursuing such efforts more aggressively under the new UEMOA Treaty.

10 Among the ECOWAS countries during the 1980s, annual inflation ranged from an average of 5–7% in UMOA and Liberia, to 17% in the Gambia, 21% in Nigeria, 48% in Ghana, and 63% in Sierra Leone.

A common central bank that has a few, well defined, goals is far more likely to be successful than one burdened with many tasks. This is arguably true of national central banks, and particularly true of a supranational one, as the BCEAO’s experience demonstrates. Tasks such as financial-sector regulation and deposit insurance are best left to separate agencies.

Regional imbalances within a nation-state are often corrected by the existence of stabilizing tax and transfer payment flows, as Eichengreen (1990) shows for the United States. The mechanics of this form of “fiscal federalism” is at the heart of the debate over monetary integration in Europe. However, it is probably unrealistic to expect intermember equalization payments of the type seen in the United States and Canada and envisaged in Europe in a monetary union of developing countries. The lack of such fiscal federalism will mean the persistence of pockets of both chronically poor and perennially gifted regions, as experience proves all over the world. As Youssouf Dembélé pointed out in his comments to me on this paper, a regional development bank that is adequately capitalized and has an explicit mandate to fund projects with regional overtones may go some way toward complementing a common central bank in the spirit of economic integration. The West African Development Bank (BOAD: Banque ouest-africaine de développement) has not been an outstanding success in this regard, but a more effective regional development bank could clearly playa useful role in promoting intraregional trade and a better balanced process of industrialization.

Care should be taken in any equalization formula to respect certain principles of separation and transparency. Separation implies that central banking decisions be made independent of political pressures, through the separation of central-banking and development-banking functions. Transparency means that the common central bank’s seigniorage allocation policy should be clear, open, and statutorily unalterable. Because this policy would be agreed upon by the members, there is no reason why it might not have an equalizing bias of some sort.

Management of the exchange rate (or management of the average inflation rate union-wide) is a valid task for a common central bank, but there is no reason why the external value of the currency has to be immutably fixed. Indeed, it is highly unlikely that any future monetary union of developing countries would gain the type of guarantee that the CFA franc has historically enjoyed from the French Treasury. The choice of exchange rate regime should be a matter specific to the economic and institutional features of each particular monetary union. The possibility of an expanded or transformed franc zone linked to the ECU is an interesting option, discussed at length in Cobham and Robson (this volume). These authors join Guillaumont and Guillaumont (1989) and Collier (1992) in suggesting that an integrated Europe might extend some form of guarantee of convertibility to a larger African monetary grouping, perhaps through “associate status” in the European monetary union. Even if this were to happen, the guarantee would undoubtedly be far more restrictive than the current one, and the job of the common (African) central bank would include some management of the nominal external value of the common currency.¹¹

The need for a monetary union to manage the nominal external value of a common currency raises new issues in the management of the union’s reserve pool. Reserve pooling is an integral part of any monetary union, and in its various forms has been a successful half-step toward closer monetary coordination in many parts of the world.¹² In UMOA, the system has worked automatically — surplus countries have covered the deficits of the others, and the French Treasury has covered the deficit of the franc zone account as a whole. When surpluses accrue, franc zone deposits in Paris earn a market rate of interest, in favour of the member countries. In a monetary union with no immutably fixed external parity, the coverage of deficits by certain countries is no longer automatic. Rules are needed to define when and how much to devalue and what conditions to impose on deficit countries, if the reserve pool is to operate effectively and without charges of bias.

11 There is, of course, some question about the viability of UMOA itself when the European monetary system is transformed into a full monetary union. In principle, the parity value of the CFA could be reattached to the ECU, and there is no reason why France could not maintain its guarantee of convertibility of the CFA franc, if it wished. Some rethinking of the UMOA arrangement would seem the most likely option, however.

12 See Kaplan and Schleiminger (1989), who detail the ultimately successful efforts to bring Europe from the constrictive bilateralism of the post-war years to full convertibility a decade later. Medhora (1992c) and M’Bet and Niamkey (1990) quantify the gains from the reserve pools in the UMOA and BEAC regions, respectively.

CONCLUSION

This overview of the UMOA experience testifies to a number of advantages of monetary unification. Under the proper conditions, a monetary union will minimize conversion, valuation, and transactions costs; substantially reduce intraregional risk; and provide an independent, apolitical monetary regime that should breed low inflation. There will be some saving of international reserves, and other economies of scale will be realized in areas of high fixed cost. Harmonization of some regulations and laws will further reduce costs, thus helping to promote regional trade.

The UMOA experience is also instructive of some of the conditions for successful monetary integration. In matters of money and finance, such as the degree of external borrowing or control over the exchange rate, members should be subservient to the common central bank, or at the very least face more onerous reporting requirements than has been the case in UMOA. UMOA experience suggests that the 20% rule that governs credit to national governments by the BCEAO is insufficient and should be buttressed by controls over foreign borrowing. A more comprehensive rule is thus needed if the common central bank is to have real control of the macroeconomic situation in the member states.

Over a wide range of “normal” conditions, i.e., from 1963 to the early 1980s, the Union functioned well. Hindsight tells us that it was not equipped to deal with deliberate or accidental overborrowing, overlending, overspending, and fraud in the decade that followed. The Union will not be able to prevent macroeconomic imbalance if it is not allowed to function in the proper spirit and it could constitute an impediment to economic adjustment if it is excessively inflexible, too long, as UMOA seems to have been regarding the parity of the exchange rate. A properly designed monetary union will promote good policy and ensure the materialization of all potential benefits from the union.

APPENDIX I:
MEASURING THE INDEPENDENCE OF CENTRAL BANKS

Grilli et al. (1991) divide central banks’ independence into political independence (defined as “the capacity to choose the final goal of monetary policy”), and economic independence (defined as “the capacity to choose the instruments with which to pursue [the] goals”). The authors construct two indices of central bank independence by awarding one point for each of the following attributes.

POLITICAL INDEPENDENCE

1. Governor not appointed by government.
2. Governor appointed for more than 5 years.
3. All the board not appointed by government.
4. Board appointed for more than 5 years.
5. No mandatory participation of government in the board.
6. No governmental approval of monetary policy formulation required.
7. Statutory requirement that central bank pursues monetary stability among its goals.
8. Legal provisions that strengthen the central bank’s position in conflicts with the government.

ECONOMIC INDEPENDENCE

1. Government access to direct credit facility of the central bank not automatic.
2. Direct credit facility available at market rate of interest.
3. Direct credit facility explicitly temporary.
4. Direct credit facility limits amounts loaned.
5. Central bank does not participate in primary market for public debt.
6. Discount rate set by central bank.
7. Banking supervision not entrusted to central bank (two points) or not entrusted to central bank alone (one point).

For its political independence, the BCEAO gets three points for the first three criteria (because no individual government controls such decisions), and one half point for the fourth (because its board comprises ministers of finance, some of whom have better tenure and credibility than others). No points are given for criteria 5, 6, and 8, but a point is given for 7, as the bank has a statutory obligation to maintain the internal and external value of the CFA franc. The total here is four and a half.

For its economic independence, the rules governing the BCEAO’s direct credit facility are such that points are given for criteria 2, 3, and 4, but not for 1 and 5. The discount rate is set by the bank and, since 1989, a separate agency has been established to oversee the financial system in the union; thus, one point is given for each of criteria 6 and 7. The total here is five.

No doubt, such an index conceals the subtleties of the monetary policymaking process in countries, and there is also the issue of whether the existence of a statute necessarily means that’ its intent is carried out in practice. This would particularly be the case if such an index were constructed for some developing countries. In our case, this framework represents well the position of the BCEAO in such matters — slightly above the OECD norm and probably well above the norm for developing countries.

REFERENCES

Allen, P.R. 1976. Organization and administration of a monetary union. Princeton University, Princeton, N), USA. Studies in international finance 38.

Bhatia, R.J. 1971. Factors influencing money supply in BCEAO countries. IMF Staff Papers, 18, 389–398.

——— 1985. The West African Monetary Union, an analytical review. International Monetary Fund, Washington, DC, USA. Occasional paper 35.

Callier, P. (ed.). 1991. Financial systems and development in Africa. World Bank/Economic Development Institute, Washington, DC, USA.

Canzoneri, B.M.; Rogers, C.A. 1990. Is the European Community an optimal currency area? American Economic Review, 80 (3), 419–433.

Collier, P. 1992. European monetary union and “1992”: opportunities for Africa. The World Economy, 15 (5),633–643.

Collier, P.; Gunning, J. W. 1992. Aid and exchange rate adjustment in African trade liberalizations. Economic Journal, 102 (413), 925–939.

de Cecco, M.; Giovannini, A. (ed.). 1989. A European central bank? Cambridge University Press, New York, NY, USA.

De Macedo, J.B. 1986. Collective pegging to a single currency: the West African Monetary Union. In Edwards, S.; Ahamed, 1. (ed.). Economic adjustment and exchange rates in developing countries. University of Chicago Press, Chicago, IL, USA.

Devarajan, S.; de Melo, J. 1987. Evaluating participation in African monetary unions: a statistical analysis of the CFA zones. World Development, 15 (4), 483–496.

——— 1990. Membership in the CFA zone: Odyssean journey or Trojan horse? World Bank, Washington, DC, USA. Policy Research Working Paper WPS 482.

de Zamaroczy, M. 1992. The reform of Senegal’s banking system. Finance and Development, 29 (1), 14–15.

Downes, P.; Vaez-Zadeh, R. (ed.). 1991. The evolving role of central banks. International Monetary Fund, Washington DC, USA.

Eichengreen, B. 1990. One money for Europe? Lessons from the U.S. currency union. Economic Policy, 10, 117–188.

Elbadawi, I. 1991. Macroeconomic stability and real exchange rate and price convergence in the CFA zone. World Bank, Washington, DC, USA.

Elbadawi, I.; Majd, N. 1992. Fixed parity of the exchange rate and economic performance in the CFA zone. World Bank, Washington, DC, USA. Policy Research Working Paper WPS 830.

Folkerts-Landau, D.; Garber, P.M. 1992. The European central bank: a bank or a monetary policy rule. National Bureau for Economic Research, Boston, MA, USA. Working paper 4016.

Grilli, V.; Masciandaro, D.; Tabellini, G. 1991. Political and monetary institutions and public financial policies in the industrial countries. Economic Policy, 13, 341–392.

Guillaumont, P.; Guillaumont, S. 1984. Zone franc et développement africain. Economica, Paris, France.

——— 1989. The implications of European monetary union for African countries. Journal of Common Market Studies, 28 (2), 139–154.

Guillaumont, P.; Guillaumont, S.; Plane, P. 1988. Participating in African monetary unions: an alternative evaluation. World Development, 16 (5), 569–576.

Honohan, P. 1983. Measures of exchange rate variability for one hundred countries. Applied Economics, 15, 583–602.

——— 1990a. Price and monetary convergence in currency unions. World Bank, Washington, DC, USA. PRE working papers WPS 390.

——— 1990b. Monetary cooperation in the CFA zone. World Bank, Washington, DC, USA. PRE working papers WPS 389.

IMF (International Monetary Fund). 1963. The CFA franc system. IMF Staff Papers, 10, 345–396.

——— 1969. Financial arrangements of countries using the CFA franc. IMF Staff Papers, 16,289–389.

Ishiyama, Y. 1975. The theory of optimum currency areas: a survey. IMF Staff Papers, 22, 344–383.

Julienne, R. 1988. Vingt ans d’institutions monétaires ouest-africaines. Harmattan, Paris, France.

Kaplan, J.J.; Schleiminger, G. 1989. The European payments union. Oxford University Press, New York, NY, USA.

Kydland, F.E.; Prescott, E.C. 1977. Rules rather than discretion: the inconsistency of optimal plans. Journal of Political Economy, 85,473–491.

Laker, J .F. 1981. Fiscal proxies for devaluation: a general review. IMF Staff Papers, 28, 118–143.

M’bet, A.; Niamkey, A. 1990. European economic integration and the franc zone. Centre Ivoirien de Recherches Économiques et Sociales, Abidjan, Côte d’Ivoire.

Medhora, R. 1992a. The West African Monetary Union: institutional arrangements and the link with France. Canadian Journal of Development Studies, 12 (2), 151–180.

——— 1992b. Seigniorage flows in the West African Monetary Union, 1976–89. Weltwirtschafdiches Archiv, 128, 513–526.

——— 1992c. Reserve pooling in the West African Monetary Union. Economia Internazionale, 45 (2), 209–222.

——— 1993. The allocation of seigniorage in the franc zone: the BEAC and BCEAO regions compared. International Development Research Centre, Ottawa, Canada.

Mundell, R.A. 1961. A theory of optimum currency areas. American Economic Review, 51, 657–665.

Neurrisse, A. 1987. Le franc CFA. Librairie Générale de Droit et de Jurisprudence, Paris, France.

Paraire, J. 1988. L’instabilité des taux de change. In Guillaumont, P.; Guillaumont, S. (ed.). Stratégies de développement comparées zone franc et hors zone franc. Economica, Paris, France.

Robson, P. 1971. Current problems of economic integration. United Nations Conference on Trade and Development, Geneva, Switzerland.

Sacerdoti, E. 1991. Central bank operations and independence in a monetary union: BCEAO and BEAC. In Downes, P.; Vaez-Zadeh, R. (ed.). The evolving role of central banks. International Monetary Fund, Washington DC, USA.

Seck, D.; El Nil, Y. 1992. The experience of African countries with financial liberalization (vol. 1 and 2). African Centre For Monetary Studies, Dakar, Senegal.

UMOA (Union Monétaire Ouest-Africaine). 1962. Traite du 12 mai 1962: accords de coopération; statuts de la Banque Centrale; convention de compte d’ opérations. BCEAO, Dakar, Senegal.

——— 1973. Traite constituant l’Union monétaire ouest africaine. BCEAO, Dakar, Senegal.

——— 1989. La nouvelle politique de la monnaie et du crédit de la Banque Centrale des États de l’ Afrique de évolution. BCEAO, Dakar, Senegal.

Vallee, O. 1989. Le prix de l’argent CFA: heurs et malheurs de la zone franc. Karthala, Paris, France.

Vinay, B. 1980. Zone franc et coopération monétaire. Ministère de la Coopération, Paris, France.

Documento(s) 14 de 20