I The West African Monetary Union and the Theory of Currency Unions
- Rattan Bhatia
- Published Date:
- May 1985
Until 1984, the West African Monetary Union (WAMU) consisted of six West African countries—Benin, Burkina Faso, Ivory Coast, Niger, Senegal, and Togo. (Mali withdrew from the Union in 1961 and rejoined in 1984; it is therefore excluded from this analysis, which deals with a period when it was not a member.) For nearly two decades these countries have had a freely circulating common currency issued by the Central Bank of West African States, the Banque: Centrale des Etats de l’Afrique de l’Ouest (BCEAO), which was formally established in 1962. The BCEAO implements the same monetary policy for the entire WAMU area, and its statutes cannot be unilaterally altered by a member government, although they can be amended by unanimous agreement. The currency, the CFA franc, is pegged to the French franc at an exchange rate of CFAF 50 = F 1 that has remained unchanged since 1948. France, which is represented on the Bank’s Board of Directors, ensures unlimited convertibility of the CFA franc into French francs through an operations account at the French Treasury, which holds the foreign exchange reserves of all the member countries and which handles the BCEAO’s foreign exchange transactions.
The year 1974 marked a watershed in the history of the BCEAO; it was then that its statutes were substantially revised to enable it to play a more active monetary role in the development of member economies. Under the old statutes, the BCEAO was a passive “cofinancier” of members’ demand for credit at relatively low interest rates, primarily through its rediscount window. Its monetary policy was highly centralized at its then headquarters in Paris, and was coordinated hardly at all with the fiscal policies of its individual members. The main reason for revising the statutes in 1974 was to improve on these shortcomings, so that monetary policy could be used more actively to promote the economic development of member economies, and to encourage their economic integration.
The aim of this study is to analyze the implementation of monetary policy in the WAMU area since 1963, as an interesting experiment in what Machlupaptly describes as complete monetary integration (Machlup, 1976). It also reflects the role a central bank may play in developing countries. Unfortunately, very little of what has been published on this interesting experiment can throw light on such questions as whether the Union has functioned smoothly with the necessary degree of fiscal harmonization and economic integration, or whether the functional and organizational changes in the BCEAO have increased the effectiveness of its monetary policy. This study will analyze some of these issues in their historical context.
Optimum Currency Areas and Fiscal Integration
The rationale for any monetary or currency union is provided by the general framework of the theory of optimum currency areas. It was Mundell among economic theorists who first perceived the key element of an optimum currency area to be a high degree of factor mobility (Mundell, 1961). His reasoning runs as follows. If two regions, A and B, are initially in internal and external equilibrium, and an exogenous shock shifts total demand from goods produced in B to those produced in A, unemployment will lend to rise in B and prices in A. Since A and B have separate national currencies, the adjustment to the shock will depend on the exchange rate regime that these two countries adopt. With flexible rates, a depreciation of B’s currency relative to A’s will ease unemployment in B by making its goods more competitive, and inflation will fall in A as demand for its goods falls off. With fixed rates, on the other hand, the entire burden of adjustment—a reduction in real income, output, and employment—will fall on B if A succeeds in preventing a rise in prices and its goods remain more competitive.
Now consider the case where A and B are regions that do not correspond to national boundaries so the tool of the exchange rate cannot be used. A shift of demand toward A’s goods will pose a dilemma; if the monetary authorities move to reduce unemployment in the depressed region. B, they will aggravate inflation in the prosperous region. A, by increasing B’s demand for A’s goods. If they move to mitigate inflation in A by restrictive policy, they will aggravate unemployment in B. This result is caused by Mundell’s definition of regions as “areas within each of which there is factor mobility, but between which there is factor immobility” (Mundell, 1961, p. 658). Hence, price stability and full employment are compatible goals if and only if each region has its own currency and can use its exchange rate to offset demand shifts, or alternatively, factors are mobile between regions in an optimum currency area, but are immobile among currency areas.
When the currencies of small countries are pegged, Mundell puts the argument thus: “If you throw the devaluation instrument away (say by pegging), you want to have some protection in case your country gets into unemployment or excessive internal debt problems. If there is a shift of demand from your own country onto the goods of the metropolitan center, that will cause unemployment at home. If it can be mitigated by a labor flow to the metropolitan center, the problem will be less acute” (Mundell, 1961, pp. 366–67).
Mundell’s thesis has been criticized on two grounds. First, it is suggested that capital mobility need not always be equilibrating. In the above example, if equilibrium is attained through higher unemployment in B and higher inflation in A, the opportunities for investment would then rise in A and fall in B. With mobile capital, if the fall in investment exceeds the fall in savings in B resulting from the initial demand shift, and if the rise in investment exceeds the rise in savings in A, then unemployment increases in B, and inflation increase in A (unless it lends to B). Hence, capital mobility can aggravate regional disparities. Second, it is argued that Mundell overemphasized the potential mobility of labor, since social and cultural differences, as well as economic and psychic moving costs, can make labor a quasi-fixed factor even within a currency area. Hence, “labor mobility is an inadequate substitute for more conventional payments adjustment instruments—demand management and exchange rate variation” (Ishiyama, 1975, p. 349).
Despite these criticisms, the adjustment dilemma posed by Mundell for a country comprising many regions, or a union comprising many countries, is real. Many economists see it as providing a rationale for greater integration of fiscal and financial policies in monetary unions, Ingram (1973) and Scitovsky (1967) use a variant of Mundell’s original argument to suggest that a high degree of financial integration should be a key characteristic of an optimum currency area. As Ishiyama puts it: “under a high degree of financial integration, the need for exchange rate changes would be eliminated because only fractional changes in interest rates would evoke sufficient equilibrating capital movements across national frontiers” (Ishiyama, 1975, p. 355). However, as observed earlier, equilibrating capital movements need not occur, as surplus countries are not always willing to lend to deficit countries. Moreover, while capital may finance a payments deficit, it does not correct the basic imbalance. Capital movements are not, therefore, a long-term solution to external disequilibrium and cannot be a substitute for adjustment in the exchange rate, or in the level of prices or employment.
The question of how far policy cohesion contributes to the success of an optimum currency area is quite important, as emphasized by Haberler (1970) and Tower and Willett (1976), among others. Policy cohesion is facilitated if there is a substantial agreement among member nations on how far they are willing to trade inflation off against unemployment, as well as the extent to which they will willingly surrender sovereignty to a common central bank. Haberler (1970) and Fleming (1971) specifically single out a similar rate of inflation among members as a qualifying condition for an optimum currency area, because differences in rates of inflation and productivity growth are primary sources of external payments disturbances. Kindleberger (1973), while agreeing with this precondition, remains pessimistic about unions among countries, For him, the nation-state is the optimum currency area, as conflicting sovereignty problems are bound to occur in a union of countries. Indeed, the question of how well policy objectives are integrated is central to understanding how a monetary union functions. But divergent policies are possible even in a union, if the burden of adjustment is carried by one nation or a group of nations. The WAMU is a notable example of a union where the costs of divergent policies are borne by the constituent members but mitigated by external support. Ishiyama (1975) prefers the cost-benefit approach to optimum currency areas. He sees many benefits of a common currency. It reduces currency-conversion costs, eliminates speculative capital flows, saves on exchange reserves, and may accelerate fiscal integration. But it also results in a loss of autonomy in national monetary policy, a possible loss of autonomy in fiscal policy, a possible worsening of the inflation-unemployment trade-off, and an increase in regional disparities.
The main point that emerges from the theory underlying the link between monetary and “real” factors is of vital importance to an optimum currency area, particularly when member states have broadly similar economic characteristics and experiences. This is well illustrated by the problems faced by the WAMU since 1974. Until that year, a relatively weak BCEAO was perceived to be adequate, as some members of the WAMU shared similar experiences and characteristics. However, as they began to experience diverging trends in their external and budgetary deficits, a stronger monetary authority and a greater degree of fiscal coordination were felt to be necessary.
Effectively coordinated fiscal policy is viewed as essential to ease regional adjustment (this is pointed out by Tower and Willett (1976)); such integration is necessary to make up for factor immobility, or to structure programs to encourage factor mobility, and thus to strengthen the monetary union and enhance the impact of its operations. This is not to deny the validity of the Mundellian criterion of the existence of substantial factor mobility as an indispensable condition for an optimum currency area. Indeed, Corden’s (1972) contention that monetary integration does not require parallel fiscal integration has been supported by the experience of the WAMU so far, mainly because governments have financed budget deficits through borrowing in the community or international capital markets. Had the deficits been financed by taxation, as he argues, “efficient short-term management” would have required some harmonization or control of overall budgetary policies, to minimize the distorting interregional flows caused by taxes on capital movements, like corporate income, interest and dividend taxes, and commodity taxes based on destination. The WAMU certainly satisfies Corden’s notion of limited fiscal integration (“talk” plus no flow-distorting taxes).
Other writers have given stronger reasons for fiscal integration. Allen (1976) recommends centralized fiscal policy for five reasons. First, given economies of scale in collecting and processing information, a centralized government would operate with more complete information than individual national governments. (Presumably this would lead to better, more efficient, public policy.) Second, national governments always formulate policy with a concern for spillover effects from other member governments’ policies. A centralized government would be free from this problem, but even so, it must be concerned with the regional impact of its policies, if not for reasons of equity and efficiency, then at least to minimize political discontent among member nations. Third, a centralized government would enjoy more confidence and a higher credit rating on capital markets than its least creditworthy members, giving it a larger borrowing capability than the sum of the capabilities of member governments (’"the debt of … a central fiscal authority would be considered no more risky than the debt of any one national government and considerably less risky than the debt of some of the national governments, notably those that most wanted to borrow” (Allen, 1976, p. 63)).
The fourth argument made for fiscal integration is that it would make redistribution policies more effective because they would be more comprehensive. While differing national policies, such as divergent income tax rates, will cause the rich to move from areas where the concern for equity is more into areas where the concern for equity is less, aggravating the income differentials between nations, this would not occur in a union. Centralization maximizes a union welfare function, however defined, not necessarily conflicting with the national welfare functions. Finally, in investment policy, national governments may provide “beggar-thy-neighbor” incentives to industry or foreign interest, while a uniform common policy would not, in theory, lead to ruinous competition among areas to attract investment or other factors of production.
These arguments in favor of centralized fiscal policy are highly relevant to the future of the WAMU. If the eventual goal of a monetary union is economic integration, then centralization is obviously unavoidable. It is all a question of the degree of centralization that is consistent with a monetary union, and this will be determined by the willingness of the member states to continue their association with the union without a sense of grievance about unfair treatment. Since the option of dropping out is more real for most nations within a union than it is for most units within a nation, more attention should be given to the regional impact of union policies than to national policies. If, as Ishiyama (1975) has argued, there is evidence that a common currency may worsen the inflation-unemployment trade-off, then it is imperative to remedy the situation before its impact is transmitted to the union.
A point made by Williamson (1976) should be mentioned—that monetary integration leads to fiscal integration by increasing unemployment in depressed regions. According to him, in monetary unions the “demonstration effect” is strong; wage settlements in depressed regions are influenced by the higher wage settlements in the richer regions, thereby accentuating unemployment in the depressed regions. Since “unemployment is more visible than poverty,” some fiscal policy will be initiated sooner or later to reduce it. Williamson’s point, however, is more relevant to integration among developed, industrial countries than to a union of developing countries. First, relative wages are important in developed countries, where awareness of wage differences among groups of workers is pronounced. Second, since a labor-absorbing traditional agricultural sector is dominant in developing countries, higher wages do not necessarily result in unemployment, but underemployment or disguised unemployment. So the fiscal integration that may be forthcoming automatically in a union of developed countries may not emerge from a monetary union among developing countries.
There is, therefore, a case for enforced fiscal integration in a union such as the WAMU. In particular, there is a need for a coordinated strategy to diversify and develop the economy. A centralized strategy would be more manageable and efficient than a national approach.
The academic discussion points, therefore, toward a conclusion different from Kindleberger’s, which underscores coherence of an area’s major economic policies as a precondition for an optimum currency area. For a union to be successful, initially cohesive policies among members may not be essential since a currency area may lead to an integration of major economic policies. What is fundamental to such a union is the political will to integrate. This has been demonstrated by the experiment of the WAMU since 1963, and of the BCEAO since 1974.