論西非貨幣聯(lián)盟的經(jīng)濟(jì)可取性：一種貨幣適合所有人嗎

1、On the economic desirability of the West African monetary union:would one currency fit all? HYPERLINK l _bookmark0 1Ccile Couharde HYPERLINK l _bookmark0 , Carl Grekou HYPERLINK l _bookmark0 , and Valrie Mignon HYPERLINK l _bookmark0 IntroductionDuring the year 2019, the longstanding monetary union

2、project of the Economic Com- munity of West African States (ECOWAS) has undergone a considerable acceleration. This surge, mainly influenced by the popular rejection of the CFA franc reached its peak on June 29th in Abuja during the 55th ordinary session of the ECOWAS Conference of Heads of State an

3、d Government where the latter agreed to launch a new currency, the “ ECO” by 2020. HYPERLINK l _bookmark0 2 However, given the difficulties for the ECO to be implemented, the roadmap foresees a gradual approach based upon the achievement of the required ten convergence criteria. On December 21st of

4、the same year, as expected, the president of Cte dIvoire, on behalf of his counterparts from the WAEMU, announced that (i ) the ECO would be adopted first by the WAEMU countries, and (i i ) the latter will maintain the fixed parity with the euro inherited from the CFA franc. HYPERLINK l _bookmark0 3

5、 The members of the West African Monetary Zone (WAMZ) criticized the decision to peg the ECO to the Euro, ar- guing that this choice will maintain the scope of regional monetary policy decisions. In this sense, the switch from the CFA Franc to the ECO appears purely cosmetic. Moreover, this decision

6、 contrasts with the currency basket peg retained for the ECO in its ECOWAS design. HYPERLINK l _bookmark0 4 Inexorably, if the ECOWAS monetary union comes to existence one day, the1We are grateful to Thomas Grjebine and Sbastien Jean for helpful remarks and suggestions.EconomiX-CNRS, University of P

7、aris Nanterre, France. Email: HYPERLINK mailto:cecile.couhardeparisnanterre.fr cecile.couhardeparisnanterre.fr.EconomiX-CNRS, University of Paris Nanterre and CEPII, France. Email: HYPERLINK mailto:carl.grekoucepii.fr carl.grekoucepii.fr.EconomiX-CNRS, University of Paris Nanterre and CEPII, France.

8、 Corresponding author : Valrie Mignon, EconomiX-CNRS, University of Paris Nanterre, 200 avenue de la Rpublique, 92001 Nanterre Cedex, France. Phone: 33 1 40 97 58 60. E-mail: HYPERLINK mailto:valerie.mignonparisnanterre.fr valerie.mignonparisnanterre.fr2The CFA franc is the currency shared by eight

9、countries of the ECOWAS. These countries form the WestAfrican Economic and Monetary Union (WAEMU thereafter) since 1945. This latter is composed of Benin, Burkina Faso, Cte dIvoire, Guinea-Bissau, Mali, Niger, Senegal, and Togo. The ECOWAS is formed by the WAEMU countries, Cabo Verde, Gambia, Ghana,

10、 Guinea, Liberia, Nigeria, and Sierra Leone.3However, the countries will not have to keep 50% of their reserves in the French treasury, and there willno longer be a French representative on the currency unions board.4In January 2020, the WAMZ Convergence Council excluding Cape Verde condemned the un

11、ilateral WAEMU decision to launch the ECO as a rebrand of the CFA franc.different countries would have to agree on a basket of currencies. This paper tackles this issue and aims at studying the economic desirability of the ECO project.More specifically, building on the approach developed by Coudert

12、et al. (2019), we investigate, from a policy coordination viewpoint, the sustainability of the ECO in its var- ious forms that can also be seen as the different steps towards the ECOWAS monetary union. The key take-away from Coudert et al. (2019) is the following: since the main cost associated with

13、 participation in a monetary union is the loss of the monetary policy autonomy, a country would be more likely or less armed to join a monetary union if its “optimal” or sustainable exchange rate path coincides with that of the other members. Adopting such an approach presents several benefits.First

14、, in currency unions, the least advanced economies generally endure the common monetary policy which, either imposed by the leading countries or resulting from an unfor- tunate combination of circumstances, rarely fits with the economic structures and, in turn, the national “needs”. Are the member s

15、tates willing to subordinate national interests to regional interests? Is Nigeria accounting for two-thirds of the zones GDP and broadly half of the population willing to renounce its monetary policy autonomy? Is the naira affordable as an anchor by the other countries? Given the logical negative an

16、swers ex- pected to these crucial questions, the compatibility of the “optimal” or sustainable national monetary policies is essential for such a monetary unions feasibility and sustainability.The European Monetary Union (EMU) crisis has stressed the importance of the ade- quacy between the member s

17、tates sustainable exchange rate paths. Indeed, the mismatch between the core and periphery countries regarding the exchange rate level has led to the build-up of large macroeconomic imbalances, which may have been erroneously interpreted as catching-up phenomena. Given the loss of the exchange rate

18、as an adjustment tool, countries were forced to adjust via internal devaluations, which, in addition to being socially costly and lengthy processes, undermined the growth momentum. This was particularly the case in Greece. However, the difficulties were not restricted to the countries undergo- ing e

19、conomic hardships but spread to the others, even the healthier ones, challenging the unions perennation.Second, as detailed below, we define the “optimal” or “sustainable” exchange rate path as the equilibrium exchange rate path allowing simultaneously both internal and external balances which are c

20、entral for a stable and sustained economic development (Alberola et al., 1999; Berg and Miao, 2010; Schrder, 2013), and maximizing welfare (Engel, 2011). Our approach therefore fully accounts for the countries economic and structural charac- teristics as well as their interdependencies, and addresse

21、s the issue of sustainability at boththe country and the monetary union levels simultaneously.Third, by focusing on a unique but all-encompassing policy indicator, our approach allows us to avoid the problem of inconclusiveness associated with the multiple criteria considered in the traditional Opti

22、mum Currency Area (OCA) theory framework. Overall, we can gather in a unified framework different aspects of monetary union viability, and identify the “natural” member countries.To this end, we rely on a two-step methodology. First, we use the Hierarchical Ascen- dant Classification (HAC) method an

23、d factor analysis, and identify two groups of countries with relatively similar sustainable exchange rate paths. The first group is formed of the WAEMU countries Benin, Burkina Faso, Cte dIvoire, Guinea-Bissau, Mali, Niger, Senegal, and Togo and Cabo Verde. The second one is made of Gambia, Ghana, G

24、uinea, Nigeria, and Sierra Leone. Liberia, although belonging to this second set of coun- tries, is an outlier. In a second step, we address the issue of the sustainable exchange rate regime (ERR)i.e., the regime underlying the sustainable exchange rate path by inferring the weights associated with

25、hypothetical basket pegs. Our findings show that no single currency peg nor a freely floating ERR would be desirable for any of the considered countries. Instead, we suggest that a basket peg with a certain degree of flexibility would be preferable. Overall, our results tilt in favor of the ECO as a

26、 common but not unique currency, and the necessity of two distinct ECO zones in a first phase. Within each zone, countries would peg with some flexibility their national currency to the ECO, which itself would be defined by the consistent currencies basket. Due to differences regarding the adjustmen

27、t capacities, this first stage of the implementation of the ECO should be long enough to make both nominal and real convergence as feasible as possible before pro- ceeding further. In this sense, we go further than previous studies (e.g. Bnassy-Qur and Coupet, 2005; Tsangarides and Qureshi, 2008; Du

28、frnot and Sugimoto, 2013) by showing that countries heterogeneity is not an insurmountable obstacle for their integration desire.The rest of the paper is organized as follows. Section 2 provides some background elements regarding the ECOWAS countries and a review of the related literature. Section 3

29、 is devoted to the methodology and data. In Section 4, we focus on heterogeneity between ECOWAS countries, and present the results of HAC and factor analysis. Section 5 reports our findings related to the ERR choice. Finally, Section 6 concludes.Economic convergence within the ECOWAS: where do we st

30、and?The developments regarding the ECO during 2019 took place in a favorable context, marked by considerable GDP growth rates. Indeed, as can be seen in the top left panel of Figure 1, the average GDP growth rate of the ECOWAS over the 2014-2018 period is equal to 4.4%. The WAEMU countries display h

31、igher growth rates around 6% on average with Cte dIvoire showing the highest one (8.15%). In the non-WAEMU countries, the picture is less homogeneous with, on the one hand, Ghana and Guinea exhibiting high GDP growth rates comparable to WAEMU countries, and, on the other hand, Sierra Leone displayin

32、g a negative growth rate. On its part, Nigeria registered an average GDP growth rate of 2% over the same period. This value is well below the 6% rate recorded during the first half of the decade.Figure 1 Favorable context despite structural headwindsNote: Data on GDP growth rates and terms of trade

33、are from the World Development Indicators database (World Bank). The bottom left map is based on the results from a clustering method (Hierarchical Ascendant Classification) implemented by the authors. The input data are the output gaps derived by applying the Hodrick-Prescott filter to GDP (World E

34、conomic Outlook, IMF) over the 1980-2018 period. Statistics on intra-community trade flows are from the EQCHANGE database (CEPII).The case of Nigeria is very interesting because it illustrates some structural weak-nesses characterizing these countries, namely their strong dependence on commodities.

35、HYPERLINK l _bookmark0 5 This strong dependency on commodities coupled with a very low diversification of their exports, exposes these countries to a sequence of different and considerable terms of trade shocks see the latterss dispersion in the top-right chart of Figure 1. HYPERLINK l _bookmark0 6

36、Besides the fact that this constitutes a perennial source of shock asymmetry hindering the convergence process, one consequence is the asynchronicity of business cycles in the zone. As the bottom-left chart of Figure 1 shows, we can broadly distinguish three groups of countries with more or less syn

37、chronicity of the cycles. HYPERLINK l _bookmark0 7 However, this asynchronicity also reflects the weakness of endogenous growth factors as the intra-community trade (bottom-right chart of Figure 1). In almost half-century of its existence, intra-ECOWAS trade had in- creased from 9% in 1975 to 16% in

38、 2018. Intra-WAEMU trade, around 6% in 1975, is now about 12%. For the non-WAEMU countries, the trade flows have barely gained two percentage points. This weak level of trade integration is again explained by the special- ization of these economies in commodities, the destination of their exports to

39、ward rich countries, and, most importantly, by the near absence of industries and, in turn, of real wealth creation. As displayed in Table 1, the manufacturing value added (as a share of GDP) is, on average, around 8.4% in the ECOWAS over the 2010-2018 period, average pulled up by the WAEMU zone.Tab

40、le 1 Some indicators (average 2010-2018)ECOWASWAEMUNon WAEMUManufacturing, value added (% GDP)8.410.56.3Foreign direct investment, net inflows (% GDP) a4.43.35.9Domestic credit provided by financial sector (% GDP) b26.227.224.4Lending interest rate (%) c9.95.217.6Inflation, consumer prices (annual %

41、)4.91.48.9Revenue, excluding grants (% GDP) d18.419.617.3Tax revenue (% GDP) d15.115.814.3Population growth (annual %)2.62.92.4Age dependency ratio (% of working-age population)86.091.180.2Source: World Development Indicators (World Bank)“ a”: excluding Liberia (43%); “b”: excluding Cabo Verde (83%)

42、 and Liberia (850%); “ c ”: Missing data: Ghana, Guinea; “d ”: Missing data: Cte dIvoire, Ghana, Mali, Togo.Overall, while many economies are in rapid expansion, the nature of this expansion5The fall in the price of oil in 2014 has significantly impacted Nigeria, and resulted in a considerable decre

43、ase in the growth prospects.6As shown, terms of trade shocks obviously vary across time, but also between countries with differencesthat can evolve between around 5 and 15 pp. depending on the considered year.7Countries (displayed in red on the graph) such as Nigeria or Sierra Leonne are characteriz

44、ed by large asynchronicity compared to economies (shown in green) such as Cote dIvoire.is not satisfactory to meet the structural challenges these economies are facing. Among these challenges, funding is key since these countries attract very few capitals: the foreign direct investment net inflows a

45、re evaluated at 4.4% of the GDP in the ECOWAS, and only 3.3% on average for the WAEMU; the banking conditions are not favorable with high interest rates. On their part, the governments have very little room given that their revenues are, on average, around 20% of the GDP, with an average rate of fis

46、cal pressure amounting to 15%. This picture has to be completed by the demographic context again marked by significant growth rates and relatively high age dependency ratios and the Sahel band countries security situation.In light of these characteristics, one may question the desirability of the mo

47、netary union as well as the reasons behind the acceleration of this project given the large structural differences and weaknesses. These concerns are even more legitimate given that the countries do not satisfy the ex ante conditions for the viability of a monetary union, as specified by the Optimum

48、 Currency Area (OCA) theory.The OCA theory, initially developed by Mundell (1961) and extended by McKinnon (1963) and Kenen (1969) generally serves as a frame of reference for analyzing the im- plications related to participation in a monetary union. This literature suggests that two economies have

49、an interest in sharing the same monetary policy (i.e., a peg or a common currency) if and only if the shocks they face are symmetrical. However, if the shocks are asymmetrical, Mundell suggests that the loss of the exchange rate as an adjustment tool is less serious if alternative mechanisms are ava

50、ilable e.g., price and wage flexibility, labor mobility, fiscal transfers. Following Mundell (1961), McKinnon (1963) puts forward the importance of regional trade integration (the trade openness of each country vis-vis the other member countries), and argued that the more these economies trade with

51、each other, the higher the interest in stabilizing the exchange rates. Later on, Kenen (1969)emphasized the importance of diversification to mitigate the effects of specific shocks. HYPERLINK l _bookmark0 8Hence, based on the OCA theory, the West African countries do not satisfy the ex antecondition

52、s for the “optimality” of a monetary union and, therefore, should not form oneespecially since they do not meet the convergence criteria they agreed on (see FigureA.1 in Appendix A).8Fleming (1962) and Mundell (1963) (thereafter, Mundell-Fleming) also enriched the debate by showingthat a fixed ERR s

53、hould be preferably chosen by small open economies whose trade is more concentrated with member countries in this case, trade and welfare gains are maximized thanks to lower exchange rate variability while for countries with a higher incidence of real shocks, adopting flexible ERR is a better choice

54、.This conclusion is also shared by several empirical studies focusing on the West African monetary union project that put forward the heterogeneity of countries as well as that of the shocks they face as the major impediments. Bnassy-Qur and Coupet (2005), using a set of variables stemming from the

55、OCA theory and from the fear of floating literature (see Calvo and Reinhart, 2002) rely on cluster analysis to investigate the mon- etary arrangements in 17 Central and West African countries. Their results indicate that, although the creation of a monetary union including Nigeria is not economicall

56、y viable, a union with the core of the WAEMU and Ghana, Gambia, and Sierra Leone can be relevant. Tsangarides and Qureshi (2008) also rely on cluster analysis to investigate the homogeneity of the potential members in terms of economic characteristics inspired by the OCA literature and the convergen

57、ce criteria set by these countries. They show that coun- tries belonging to WAEMU and WAMZ (West African Monetary Zone; i.e., non-WAEMU ECOWAS countries) do not form a homogenous group. Mobilizing the same method- ology, Coulibaly and Gnimassoun (2013) focus on the convergence and co-movements betwe

58、en West African countries exchange rate misalignments. Their results show that the WAEMU area has a core composed of Burkina Faso, Mali, Niger, and Senegal, which can be joined by Ghana, Sierra Leone, and the Gambia. Bangak (2008) focuses on the relationship between bilateral exchange rates and some

59、 OCA criteria variables. In line with those of Bnassy-Qur and Coupet (2005) and Coulibaly and Gnimassoun (2013), his results indicate the existence of a core of the WAEMU to which Ghana could be linked. He also underlines that Nigeria should not be part of the WAMZ as well as the WAEMU. Houssa (2008

60、), using a dynamic factor model, shows that supply shocks in the ECOWAS are more important than demand shocks. He finds a positive correlation of demand shockswith a temporary effect on output and significant asymmetry regarding supply shocks.As a result, he concludes that a monetary union would be