Printer Friendly

Monetary Union, Competitiveness and Raw Commodity Dependence: Insights from Africa.

Introduction

Two African economic and monetary communities, the "West African Economic and Monetary Union" (WAEMU) and the "Central African Economic and Monetary Community" (CAEMC) have been using the franc CFA (standing for African Financial Community, in French) as a currency since 1945. Eight countries belong to the WEAMU, and six countries belong to CAEMC. The franc CFA zone is an excellent economics case study because it combines two key features: an unaccomplished economic integration and a double monetary constraint. The economic integration in both these two communities still lags today due to the establishment process of the franc CFA zone. Monetary integration was first imposed by the French and is part of the WAEMU and CAEMC's post-colonial legacy. This legacy is an obstacle that prevents CFA members from meeting the criteria which would make it beneficial for members to join a monetary union. Following Mundell (1961) and McKinnon's (1963) definition of optimal currency area (OCA) criteria, the recent literature has stressed that the CFA zone is a poor match due to the lack of trade integration and the low synchronization of business cycle. Furthermore, the recent literature states that the CFA zone is not an optimal currency area (Benassy-Quere and Coupet 2005; Karras 2006; Tapsoba 2008; Debrun et al. 2005; Loureiro et al. 2011; Couharde et al. 2013a, b; Gnimassoun and Coulibaly 2014).

Another unique feature of the CFA zone is that it is the only monetary union that uses a common currency which is itself pegged to another one. Therefore, one could infer that CFA countries are subject to a double constraint regarding their monetary policy. Not only are CFA countries constrained by the common policy determined by their common central bank (respectively, the BCEAO--Banque Centrale des Etats de lAfrique de l'Ouest--for the West and the BEAC--Banque des Etats de l'Afrique Centrale--for the Central region), but they are also required to align their common monetary policy to that of the eurozone. On top of these monetary constraints, the export structure of the CFA economies is heavily dependent on commodities. All of these constraints are challenges for the macroeconomic policy of the CFA member states.

The academic literature assessing the adverse consequences from belonging to a less-than-optimal monetary zone has yet to look at the transmission channel challenges introduced by the peg to the euro of the franc CFA. To the best of our knowledge, no study has explored the relationships between exchange rates and key macroeconomic variables in sub-Saharan African (SSA) countries. There exists however the literature arguing that the peg contributes to macroeconomic stability in the WAEMU and CAEMC. A number of researchers highlight the benefits of the stability of the monetary zone, in particular for its positive effect on the attractivity for foreign direct investment (FDI) compared to other SSA countries (Suliman et al. 2015 for US pegs; Reinhart and Rogoff 2003; Ramirez and Tsangarides 2007; Bangake and Eggoh 2012, for the euro pegs). Yet, no paper has focused on the transmission channels of FDI in CFA economies, leaving open the question of the macroeconomic benefits of the monetary union.

This paper aims to fill the gap left by the existing literature by addressing the following question: to what extent are the constraints on monetary policy able to affect trade, competitiveness and growth of countries that are highly dependent on commodities? To answer this question, we perform a comparative analysis of two sets of SSA countries: those belonging to the franc CFA monetary zone and those integrated in an economic union but not in a monetary zone--also referred to as extraCFA countries. A panel VAR analysis is performed on these two subsets of countries to assess the impact that the peg has on the relationships between growth rate, current account, real effective exchange rate, foreign direct investment and terms of trade.

The paper contributes in several ways to the existing literature on the franc CFA zone and the impact of the peg to the euro area. The analysis compares SSA countries that belong only to an economic union to those which have integrated further through a monetary union to estimate the cost-benefit trade-off of belonging to the CFA zone and its double monetary constraint. In addition, the paper contributes to the literature related to the competitiveness and trade balance challenges faced by developing economies which are dependent on commodities.

The first major result shows the inefficiency of investment transmission channels in CFA economies. Even if the peg sends a positive signal to foreign investors, FDI is mainly driven by the prices of exported commodities. Moreover, the results show the absence of a positive impact that foreign investment has on the trade balance in CFA countries, in contrast to extra-CFA countries where there is a significant impact. The second major result is the rejection of the Marshall-Lerner condition in the CFA zone. As opposed to other monetary zones where members suffer from an inability to stimulate exports through the nominal exchange rate tool, the competitiveness of CFA members is not impacted by this monetary constraint. The rest of the paper is organized as follows.

"Combining Commodity Dependence and a Peg" section develops the theoretical challenge associated with the combination of the peg and commodity dependence, "Empirical Strategy" section presents the empirical strategy, the dataset, and the PVAR model. "Results and Interpretation" section presents the results, "Transmission Mechanisms and Public Policies" section gives policy recommendations, and "Conclusion" section presents the conclusion of the paper.

Combining Commodity Dependence and a Peg

Commodity dependence makes the SSA economies vulnerable to external shocks such as changes in the commodity prices. However, CFA and extra-CFA countries are not subject to the same monetary constraint to address these shocks. Three main challenges stand out for the members of the CFA zone.

The first challenge is the central bank's inability to use the nominal exchange rate channel to respond to shocks, which makes it difficult in the short term to react to terms of trade shocks (Cashin et al. 2004). This potentially creates an inherent instability whenever there exists a poor synchronization of business cycles between the pegged economy and the reference economy (notably with CFA data, see Loureiro et al. 2011; Nubukpo 2015). The second challenge is related to the direct pass-through of a monetary shock in the eurozone to CFA countries (Couharde et al. 2013a), and the third challenge relates to currency misalignments. The loss of autonomy in monetary policy makes it difficult to address long-term imbalances, in particular overvaluations. These are usually associated with deviation from equilibrium, less sustainable current accounts or slower economic growth (Razin and Collins 1997; Sachs and Warner 1997; Rodrik 2008). Nonetheless, exchange rate stability and the structural adjustment within a pegged area could be beneficial for the long-term growth if business cycles are not too unsynchronized (Kocenda et al. 2013). Couharde et al. (2013a) worked with CFA zone data and came to the conclusions that in the context of strong wage and price rigidity and in the absence of nominal adjustment, there exists important inertia in the dynamics of misalignments before returning to equilibrium. Combined with the fact that competitiveness in this zone is mainly driven by the exchange rate of the eurozone, they warn against a potential persistence of misalignments.

Trade in SSA countries is heavily driven by commodities as shown in Table 1. Regarding the CFA zone, a decomposition of export data shows that the entire zone relies almost exclusively on raw commodities from the primary sector. Matching these raw exports with the COMTRADE database shows that most of these exports are priced in international markets, and almost always in dollars. In seven out of 12 countries of the CFA zone (1), commodities priced in international markets represent at least 90% of total exports. Concerning the comparable SSA countries not belonging to the CFA zone, there is a strong dependence on commodities exports as well. Specifically, in six out of nine countries, 90% or more of exports are commodities priced on international markets. From this subsample, only Gambia has more than 10% of its exports that are non-primary sector commodities.

Economies of the WAEMU are dominated by the export of agricultural and mining commodities, whereas the economies of central CFA are dominated by oil exports. This dependence on raw commodities is of prime importance. Such a specific pattern has the potential to alter the expected relationship of macroeconomic variables within the pegged zone. Notably, there are two important implications that must be kept in mind.

First, the peg and potential currency overvaluation are not expected to negatively impact exports of raw commodities. Because these small open economies are price takers, price competitiveness seems unable to explain trade balance fluctuations in the area. And as mentioned, the large majority of commodities are priced in foreign currency, in this case dollars. Moreover, it must be noted that the production of raw commodities in SSA countries is dependent on imported inputs priced in dollars (e.g., agricultural machinery). Therefore, if countries are heavily dependent on commodities for their exports, a rise of the real effective exchange rate (later referred as REER) may not affect international competitiveness as one would expect. In this specific case, an appreciation of the CFA, through an appreciation of the EUR/USD exchange rate, may lead to more competitive commodity production. However, the Marshall-Lerner condition asserts a different reasoning. According to this condition, the volume effect overtakes the price effect in the middle run, and thus a real depreciation boosts exports. Second, an extensive literature showed that countries depending on commodities were more vulnerable to Dutch disease mechanisms (Auty and Warhurst 1993). If so, then the peg could act as a protection against currency appreciation and thus prevent the loss of competitiveness in non-extractive sectors.

In order to test these hypotheses, we analyze the (1) relationships that exist between trade and exchange rate, growth and its main driver such as investment and (2) how these relationships differ between the CFA zone and the extra-CFA countries.

Empirical Strategy

The Panel Vector Auto-Regressive model is an efficient tool to capture linear interdependencies among several time series for a comparative analysis of different geographical areas. Examples are the study of macroeconomic imbalances in the eurozone (Gnimassoun and Mignon 2016), the domination of shocks generated outside a country on domestic variables (Canova and Pappa 2006), the relevance of fiscal and monetary interactions (Canova and Pappa 2007) and the channels of transmission to international shocks (Canova and Ciccarelli 2013). One PVAR model by Fauzel et al. (2014) has been run using CFA data on the impact of FDI; however, no publication has explored macroeconomic interactions with a specific focus on terms of trade. This paper aims to fill this gap in the literature by providing a PVAR estimation sub-divided in two samples: a CFA and an extra-CFA sample. The comparative analysis between these two subsamples will provide evidence for the specific interactions associated with CFA zone members compared to their extra-CFA counterparts.

The Data and Construction of the Terms of Trade Proxy

Current account, output growth and FDI variables are taken from the World Economic Outlook for the period 1980-2015. Real effective exchange rate (REER) data are from the Bruegel database. World prices of main commodities traded from and to the CFA zone are issued from UN COMTRADE database, with the notable exception of the uranium price which is derived from the Euratom Supply Agency database. Hence, we make the assumption that all countries of the sample are price takers on all commodities that are listed on the UN COMTRADE database, in addition to the uranium price. The sample includes 26 African countries that are all members of an economic union (cf "Appendix 1" for member lists):

* 14 countries from the two CFA monetary unions: WAEMU and CAEMC whose currency is pegged to the euro.

* 5 countries belonging only to the 'Economic Community of Central African States' (2) (ECCAS economic community), which are not CFA members.

* 6 countries from the "West African Monetary Zone" (3) (note that the name is misleading since these countries do not share a common currency yet: WAMZ is currently only an economic community and not a monetary union).

ECCAS and WAMZ members are a good comparison to the CFA sample. ECCAS is a larger economic union that encompasses CAEMC. WAMZ and WAEMU have launched discussions in the 1980s regarding an eventual merging of their communities to create a larger economic and monetary union. The latter discussions are still ongoing.

We construct a country-specific price index of the main commodities traded. This way, we can study the macroeconomic interactions following an exogenous shock on the price of the main traded commodities (which we associate with the terms of trade) both in the CFA zone and in comparable extra-CFA countries. Deaton (1999) and later Sissoko and Dibooglu (2006) have constructed a somewhat similar index. The commodity terms of trade annual index is defined as follows:

[tot.sub.i,t] = [N.summation over (j=1)] [X.sub.ij][P.sub.jt] - [M.sub.ij] [P.sub.ij]

With N being the total of main traded commodities j, [X.sub.ij] being the share of total export of commodity j in country i (average over the period of interest), [P.sub.jt] being the price of the main commodity j in the year t, [M.sub.ij] being the share of total export of commodity j in country i (average over the period of interest) and [tot.sub.i,t] being the proxy for terms of trade based on commodity price for country i in year t.

Based on a COMTRADE dataset on the period 1995-2005, we determine the mean share of each exported or imported commodity. For each country, we then multiply these shares by the annual price of each commodity extracted from the UNCTAD database. We assume that this average composition is constant over the time period 1980-2015.

The Panel VAR Model

A panel unit root test is performed (Pesaran 2007, cf "Appendix 2"), to ensure that the four macroeconomic variables are stationary at one percent confidence interval and thus are suitable for the PVAR model. Following Arellano and Bover (1995), the model is instrumented with available future observations. Before running the PVAR, we systematically test for the optimal number of lags that should be used relying on MAIC, MBIC and MQIB criteria. For all of the samples and subsamples, the optimal number of lag is one. The Vector Auto-Regressive model we run is (1) which follows the Abrigo and Love (2016) methodology:

[mathematical expression not reproducible] (1)

With [DELTA]GDP the growth rate, BCA the current account balance as share of GDP; [DELTA]REER is the first difference of real effective exchange rate; FDI is the level foreign direct investment as share of GDP. [DELTA][tot.sub.it] the commodity terms of trade index and Y1994 a dummy for the year 1994 when the franc CFA were devalued by 50%. The PVAR approach does not impose any a priori constraint on the relationship between the four macroeconomic variables. Variables are ordered from the presumably most exogenous to the least exogenous.

It is important to note that no empirical or theoretical studies using the same variables can provide guidance on the ordering of the four endogenous variables in the PVAR. In the case of the present paper, there is no consensus in the literature regarding the endogeneity of the real exchange rate vis-a-vis other variables. On the one hand, the REER can be viewed as the closest from a purely endogenous variable, which is the result of internal and external balances (Edwards 1989; Hinkle and Montiel 1999; Ricci et al. 2013). On the other hand, another important part of the literature considers the REER as an explanatory variable (among others: Rodrik 2008). The same input/output debate exists with growth and current account variables. The order specified in Eq. (1) is chosen. Foreign investment is considered as the most exogenous variable. We choose also to follow the literature considering real exchange rate as an explanatory and thus being the second most exogenous of the variables. Finally, GDP is considered to be more endogenous than BCA because the sample is composed of economies driven by exports of commodities. Hence, the volume and value of these commodities affect growth in turn. Nonetheless, we are aware of the lack of empirical and theoretical support for this specific ordering. It is for this reason why we test the robustness of the results by changing the order of the variables. Following those tests, no changes are found from the results presented below.

As an additional robustness check, cross-sectional dependence in the errors is tested. Since the growth equation has been simplified, common shocks and unobserved components could ultimately become part of the error term and bias the results. Following De Hoyos and Sarafidis (2006), Frees and Pesaran tests are performed to test cross-sectional dependence of the two subsamples and the tests do not reject the null hypothesis of cross-sectional independence at the 5% confidence interval. Since we do not expect a strong interdependency between cross-sectional units, robustness of the results is confirmed.

Results and Interpretation

We find in both samples strong causality over a 1-year lag between the four key macroeconomic variables, namely growth rate, current account, real effective exchange rate, foreign direct investment. CFA and extra-CFA samples display similar patterns but also different transmission channels. These differences suggest that economic interactions within the CFA zone are significantly different compared to other SSA countries. As a consequence, public policy implications derived from the results differ for the CFA countries. Table 2 presents the estimates of the PVAR model, and Fig. 1 displays the impulse response functions (IRF) with 5 percent standard error bands generated through Monte Carlo simulations. Comparative results of the two samples are presented by macroeconomic variables: after exploring the relationships between growth and trade balance, we will explore real exchange rate and the drivers of foreign direct investment.

Growth and Trade Balance

Regarding the impact on growth, estimates display similar signs for both CFA and extra-CFA samples, though significance level differs between the two samples. Investment positively impacts growth with a slightly stronger significance level in extra-CFA countries (1 vs. 5%). Impulse response functions, displayed in Fig. 1, show that this effect is more persistent in extra-CFA countries since the impact is still significant at 5% level after 3 years, which is not the case in CFA countries. In other words, spillover effects introduced by a surge in investment are more long-lasting in extra-CFA countries than in their CFA counterparts. Real exchange rate appreciation negatively impacts growth even if the impact is more statistically significant in CFA countries (1% significance) versus extra-CFA countries (10% significance). Estimates for the impact of the current account on growth are positive but significant only in the case of the extra-CFA sample. Interestingly, the lagged value of growth positively affects growth only in extra-CFA countries, suggesting for positive spillovers that are not observed in CFA countries.

As far as trade is concerned, three main results stand out when looking at the BCA variable in Fig. 1 or alternatively in Table 2.

First, the terms of trade proxy has a very strong impact on BCA in both subsets which are heavily dependent on commodities. Even though the estimate is more important for CFA countries than for extra-CFA countries (0.16 vs. 0.10), they are both significant at a one percent confidence interval. In other words, the trade balance depends highly on the price that is fixed on international markets for the main commodities traded.

Secondly, the real exchange rate has a highly significant impact on BCA but of opposite signs in both samples, meaning that it is positive in CFA countries and negative in other SSA countries. An appreciation of the REER has a significant and negative impact on the BCA for extra-CFA members (1% confidence interval). This result is in opposition to what is observed in the CFA sample where the impact is positive and significant and supports the arguments for a rejection of the Marshall-Lerner condition in the CFA zone. Even if advancing a formal proof of this hypothesis requires an estimation of price elasticities and is therefore beyond the scope of this paper, several arguments can be brought up to support this idea. As detailed in "Combining Commodity Dependence and a Peg" section, if the price effect overtakes the volume effect, an appreciation of the REER leads to a positive effect on the trade balance. In contrast, in the extra-CFA zone, a real appreciation reduces the trade balance.

For a better understanding of the question, we will be looking at the price and then the volume effect. First, with regard to the price effect, the members of the monetary zone rely heavily on imports of capital goods and equipment. These include agricultural machinery, capital-intensive goods and luxury items (Chassem 2011) with a significant and increasing part (4) of these goods which are priced neither in CFA franc nor in euro (Bagnai et al. 2015). Moreover, these goods are not substitutable with domestic goods. As a consequence, these imports positively benefit from an appreciation of the currency. This provides an explanation for the positive impact of a real appreciation on the current account.

Secondly, regarding the volume effect, one can expect it to be extremely limited because CFA members export almost exclusively raw materials. As detailed above in "Combining Commodity Dependence and a Peg" section, small open economies are price takers for these goods, and thus price competitiveness is negligible in the determination of export volumes. Considering that the volume effect is ruled out, we then suspect that the value effect on imports cancels out the negative price effect of commodity exports priced in dollars and then converted to franc CFA.

To sum up this last point, the challenge of competitiveness is very different in the CFA zone compared to other monetary zones. Since the Marshall-Lerner condition does not hold, even if the constraint of the peg was removed, stimulating exports temporarily through nominal devaluation would not an efficient tool for the members of the monetary zone. As a consequence, the root causes of competitiveness issues in the CFA zone do not lie in the inability of the member states to respond to currency overvaluations using monetary policy tools. The main competitiveness challenge for the CFA members is their high dependence on raw commodities for which they act as price takers.

Still related to the impacts on BCA, there is strong evidence for a positive impact of FDI on BCA (at 5% confidence interval) for the extra-CFA sample, which remains significant for 3 years as shown in Fig. 1. However, this is not observed in the CFA sample. For this subsample, a positive shock on foreign investment is negatively impacting BCA, yet remaining outside significance level boundaries. In other words, these observations suggest that transmission mechanisms are not driven the same way in our two samples. On the extra-CFA side, investment drives positive spillover effects that improve trade balance with a persistent effect over a few years. On the other side, CFA countries do not experience the same transmission mechanism between investment and the trade balance.

Real Exchange Rate

The dummy for the year 1994 has a strong and negative impact only in the CFA sample. The 1994 devaluation occurred on January 12, and the REER variable is an annual average. Interestingly enough, the estimates give us the average annual pass-through following the 50% devaluation against its peg, namely the French Franc. The estimates show a 39% impact of the 1994 dummy on the REER. This result is aligned with the previous literature that showed the low pass-through effect in the pegged CFA countries (Borensztein and De Gregorio 1999; Razafimahefa 2012). In these economies, the pass-through is essentially explained through higher import prices. Nonetheless, the pass-through remains limited due to the inflation target in line with the pegged economy. Furthermore, looking at the macroeconomic variables impacting the real exchange rate, estimates are comparable between the two samples with the notable exception of impact of investment, which is negative but only significant for CFA countries.

Investment

The first subsection of Table 2 highlights the contributions of the macroeconomic variables to investment. The two samples display different results. In extra-CFA countries, the impact of BCA on investment is negative, though at a low confidence interval (10%) and remains significant for less than five years. Conversely, in CFA countries, the impact of an improved current account has a positive yet non-significant impact. Furthermore, FDI positively reacts to an increase in terms of trade in CFA at one percent confidence interval, whereas the estimate is negative and non-significant in the extra-CFA sample. These results suggest a strong pro-cyclical investment behavior with main exported commodity prices in the CFA zone, aligned with Ramirez and Tsangarides (2007).

Moreover, estimates of the impact of real exchange rate on investment differ for the two sets of countries. On the one hand, a real appreciation of exchange rate has a positive impact on investment in extra-CFA countries, whereas the impulse response function in CFA countries does not show any significant impact of REER, as shown on Fig. 1. In other words, this result suggests that the perception from foreign investors is positively associated with a real appreciation. From their point of view, a real depreciation in extra-CFA countries makes investment less appealing.

Transmission Mechanisms and Public Policies

The above-mentioned results help us understand the main difference between transmission mechanisms in CFA and extra-CFA countries.

First, we learn that investment transmission channels seem to be ineffective in CFA countries compared to non-CFA countries. Let's recall that according to recent studies (Alguacil et al. 2008; Fauzel et al. 2014) two distinct mechanisms are at play in the FDI transmission mechanism toward BCA: first the introduction of more efficient technologies in the local production and secondly the improvement in the local human capital. A few patterns can contribute to the explanation of this specific result in the CFA zone.

De Mello (1999) has demonstrated that the enhancing power of FDI is conditional on the degree of complementarity and substitution between FDI and domestic investment. However, CFA countries are not very technologically advanced and very little domestic investment displays complementary or substitutable patterns. Countries belonging to the monetary zone feature much lower level of domestic investment compared to other SSA countries (first highlighted by Devarajan and De Melo 1987). Finally, even if foreign investment in physical assets is fixed, profits from these investments are moving at will between the franc CFA zone and the eurozone (France prior to 1999).

To this extent, the total freedom of capital movements with the Euro zone contributes to a form of capital export through profit repatriation from the CFA economies. Fofack and Ndikumana (2014) show that capital flight discourages domestic investment and retards the adjustment of output to its long-term growth rate. This mechanism could keep these countries with a trade deficit, even when raw commodity production increases. This restricted impact of foreign investment on the real economies of the CFA zone observed in this paper contributes to the broader literature on FDI in sub-Saharan Africa. More specifically, this issue puts an additional constraint on the CFA countries which suffer already from a low attraction for investment (Asiedu 2002).

We also learn that relationships between FDI and real exchange rate display a different pattern in the two subsamples. In fact, an appreciation positively and significantly impacts direct investment only in the case of extra-CFA countries. Additionally, we draw from Udomkerdmongkol et al. (2009), who contributed to the literature on the drivers of FDI. Using a sample of 16 emerging economies (including three African countries, namely Morocco, South Africa and Tunisia) and relying on a within-group fixed effect approach, the authors found a positive causal relationship between current devaluation of local currency and the attractivity for FDI. However, the results below do not reach the same conclusions when applied to SSA data.

For the extra-CFA sample, a currency appreciation in real terms positively benefits FDI. This result must be considered in the context of hyper-inflation and devaluation cycles occurring in flexible exchange rate regime countries (Razafimahefa 2012; Nguyen et al. 2017). Thus, foreign investors could consider a real appreciation as a signal of macroeconomic stability. Within the CFA zone, the picture is somewhat different. The REER has no significant impact on FDI dynamics. This is an expected result in the sense that the hard peg to the euro and its guarantee have been a way to import trust and have strengthened the confidence in the business environment. As a consequence, the REER in the CFA zone, mainly driven by the euro exchange rate, has no significant impact on FDI over the period of interest.

Rather than contradicting the findings of Udomkerdmongkol et al. (2009), the results of this paper tend to highlight how different is the pegged CFA zone when compared to other emerging economies. In the CFA zone, the guarantee of the French Treasury to maintain the peg sends a strong and positive signal to investors, to the point where they display no preference between holding franc CFA or euro currency. As a consequence, the mechanisms described in Udomkerdmongkol et al. (2009) do not hold up when applied in a CFA context.

In terms of public policy, complementary domestic investment would be an efficient way to address the issue by allowing for a better absorption of more advanced technologies and for benefiting from the positive spillovers of FDI on the economy. As far as the behavior of foreign investment is concerned, the CFA zone benefits from the good signal sent by the peg, the stability of the exchange rate and the free movement of capital with the euro zone. In contrast, foreign investment in the extra-CFA countries suffers from the bad signal of devaluation, usually associated with inflationary pressure. Nonetheless, investments in the CFA zone seem to be mainly driven by the prices of the main exported commodities. This result comes as a call for CFA countries to better monitor inbound foreign investment in order to insure that they are targeted at the right sectors, those contributing to long-term growth. Additionally, as far as public policies are concerned, the conclusions of the paper recommend that the CFA economies diversify their export structures away from a high dependence on one, or a few raw commodities. This diversification will then foster more sustainable growth, allowing CFA economies to reap the benefits of the sound macroeconomic stability introduced by the peg.

In terms of limitation, even if the present analysis strongly suggests that the Marshall-Lerner condition does not hold in the CFA zone, an in-depth analysis of long-term export and import demand elasticities would confirm these results. Moreover, further analysis could also integrate a dynamic term of trade index in order to account for any major shift in the export structure of the SSA economies over the period of interest.

Conclusion

This paper contributed to the literature on the challenges introduced by commodity dependence in a monetary union by investigating the interactions between four key macroeconomic variables: growth, balance of current account, real effective exchange rate and foreign investment as well as commodity terms of trade. A PVAR model run on two different subsamples of SSA countries helped to distinguish the different macroeconomic transmission channels in the pegged CFA zone and in a comparable extra-CFA sample. Our results highlight two key challenges faced by the franc CFA monetary zone members. First, even if the peg contributed to creating a good macroeconomic environment that attracts investors to the CFA zone, these investments fail to generate strong spillover effects such as those observed in extra-CFA countries. In terms of public policy, the results of this paper indicate that CFA economies would benefit from both the monitoring of inbound foreign investment and the promotion of domestic investment into sectors contributing to long-term growth which would essentially foster a diversification of their export structure. Furthermore, the results show that the Marshall-Lerner condition is not verified in the CFA sample. As opposed to other monetary zone members who suffer from an inability to stimulate exports through the nominal exchange rate tool, the competitiveness of CFA members is not impacted by this monetary constraint. The main obstacle to competitiveness lies in the dependence on raw commodities for which CFA countries act as price takers.

https://doi.org/10.1057/s41294-019-00080-6

Acknowledgements The author is grateful to Olivier Damette, Blaise Gnimassoun and to the editor Paul Wachtel for their useful and constructive comments. I would also like to thank participants to the 2017 INFER and the 2017 AFSE conferences, Quentin Batreau and anonymous referees for their helpful comments. All remaining errors are the author's responsibility.

Appendix 1: List of Countries

Extra-CFA countries belonging to WAMZ: Gambia, Ghana, Guinea, Liberia, Nigeria, Sierra Leone

Extra-CFA countries belonging to ECCAS: Angola, Burundi, Dem. Rep. of Congo, Rwanda, Sao Tome e Principe

CFA countries (WEAMU): Benin, Burkina Faso, Ivory Coast, Guinea Bissau, Mali, Niger, Senegal, Togo

CFA countries (CEMAC): Cameroon, Central African Republic, Chad, Rep. of Congo, Equatorial Guinea, Gabon

Appendix 2: Panel Unit Root Test in Presence of Cross-Sectional Dependence (Pesaran 2007)

See Table 3.

References

Abrigo, M.R., and I. Love. 2016. Estimation of panel vector autoregression in Stata. Stata Journal 16 (3): 778-804.

Alguacil, M., A. Cuadros, and V. Orts. 2008. EU enlargement and inward FDI. Review of Development Economics 12 (3): 594-604.

Arellano, M., and O. Bover. 1995. Another look at the instrumental variable estimation of error-components models. Journal of Econometrics 68 (1): 29-51.

Asiedu, E. 2002. On the determinants of foreign direct investment to developing countries: Is Africa different? World Development 30 (1): 107-119.

Auty, R., and A. Warhurst. 1993. Sustainable development in mineral exporting economies. Resources Policy 19(1): 14-29.

Bangake, C., and J.C. Eggoh. 2012. Pooled mean group estimation on international capital mobility in African countries. Research in Economics 66 (1): 7-17.

Benassy-Quere, A., and M. Coupet. 2005. On the adequacy of monetary arrangements in Sub-Saharan Africa. The World Economy 28 (3): 349-373.

Borensztein, E., and J. De Gregorio. 1999. Devaluation and inflation after currency crises. Washington: International Monetary Fund.

Canova, F., and M. Ciccarelli. 2013. Panel vector autoregressive models: A survey. In VAR models in macroeconomics--New developments and applications: Essays in honor of Christopher A. Sims (pp. 205-246). Emerald Group Publishing Limited.

Canova, F., and E. Pappa. 2006. Does it cost to be virtuous? The macroeconomic effects of fiscal constraints. In NBER International Seminar on Macroeconomics 2004 (pp. 327-370). The MIT Press.

Canova, F., and E. Pappa. 2007. Price differentials in monetary unions: The role of fiscal shocks. The Economic Journal 117 (520): 713-737.

Cashin, P., C.J. McDermott, and C. Pattillo. 2004. Terms of trade shocks in Africa: Are they short-lived or long-lived? Journal of Development Economics 73 (2): 121-144.

Chassem, N. P. 2011. Long-run effects of real exchange rate on the nominal and real trade balance in African Franc zone. MPRA Paper, (30252).

Couharde, C., I. Coulibaly, and O. Damette. 2013. Anchor currency and real exchange rate dynamics in the CFA Franc zone. Economic Modelling 33: 722-732.

Couharde, C., I. Coulibaly, D. Guerreiro, and V. Mignon. 2013. Revisiting the theory of optimum currency areas: Is the CFA franc zone sustainable? Journal of Macroeconomics 38: 428-441.

De Hoyos, R.E., and V. Sarafidis. (2006). Testing for cross-sectional dependence in panel-data models. Stata Journal 6 (4): 482.

Hinkle, L.E., and P.J. Monteil. (1999). Exchange rate misalignment: Concepts and measurement for developing countries. Oxford University Press.

De Mello, L.R. 1999. Foreign direct investment-led growth: Evidence from time series and panel data. Oxford Economic Papers 51 (1): 133-151.

Deaton, A. 1999, Commodity prices and growth in Africa. The Journal of Economic Perspectives 13 (3): 23-40.

Debrun, X., P. Masson, and C. Pattillo. 2005. Monetary union in West Africa: Who might gain, who might lose, and why? Canadian Journal of Economics/Revue canadienne d'economique 38 (2): 454-481.

Devarajan, S., and J. De Melo. 1987. Evaluating participation in African monetary unions: A statistical analysis of the CFA zones. World Development 15 (4): 483-496.

Edwards, S. 1989. Real exchange rates, devaluation, and adjustment: Exchange rate policy in developing countries. Cambridge, MA: MIT press.

Fauzel, S., B. Seetanah, and R.V. Sannassee. 2014. A PVAR approach to the modeling of FDI and spill overs affects in Africa. International Journal of Business and Economics 13 (2): 181.

Fofack, H., and L. Ndikumana. 2014. Capital flight and monetary policy in Africa. In Capital flight from Africa: Causes, effects, and policy issues (pp. 130-153). Oxford: Oxford University Press.

Gnimassoun, B., and I. Coulibaly. 2014. Current account sustainability in Sub-Saharan Africa: Does the exchange rate regime matter? Economic Modelling 40: 208-226.

Gnimassoun, B., and V. Mignon. 2016. How do macroeconomic imbalances interact? Evidence from a panel VAR analysis. Macroeconomic Dynamics 20 (7): 1717-1741.

Karras, G. 2006. Is Africa an optimum currency area? A comparison of macroeconomic costs and benefits. Journal of African Economies 16 (2): 234-258.

Kocenda, E., M. Maurel, and G. Schnabl. 2013. Short-and long-term growth effects of exchange rate adjustment. Review of International Economics 21 (1): 137-150.

Loureiro, J., M.M. Martins, and A.P. Ribeiro. 2011. Anchoring to the euro (and grouped together)? The case of African countries. Journal of African Economies 21 (1): 28-64.

McKinnon, R. I. 1963. Optimum currency areas. The American Economic Review, 717-725.

Mundell, R.A. 1961. A theory of optimum currency areas. The American Economic Review 51 (4): 657-665. https://doi.org/10.3386/w3949.

Nguyen, A. D., J. Dridi, F. D. Unsal, and O. H. Williams. 2017. On the drivers of inflation in Sub-Saharan Africa. International Economics.

Nubukpo, K. 2015. Le franc CFA, un frein a l'emergence des economies africaines? L'Economie politique 4: 71-79.

Pesaran, M.H. 2007. A simple panel unit root test in the presence of cross-section dependence. Journal of Applied Econometrics 22 (2): 265-312.

Ramirez, G., and C. G. Tsangarides. 2007. Competitiveness in the CFA franc zone (No. 7-212). International Monetary Fund.

Razafimahefa, I. 2012. Exchange rate pass-through in Sub-Saharan African economies and its determinants. IMF Working Paper, WP/12/141.

Razin, O., and S. M. Collins. 1997. Real exchange rate misalignments and growth (No. w6174). National Bureau of Economic Research.

Reinhart, M. C., and M. K. Rogoff. 2003. FDI to Africa: the role of price stability and currency instability (No. 3-10). International Monetary Fund.

Ricci, L.A., G.I.A.N. Milessi-Ferretti, and J. Lee. 2013. Real exchange rates and fundamentals: A cross-country perspective. Journal of Money, Credit and Banking 45 (5): 845-865.

Rodrik, D. 2008. The real exchange rate and economic growth. Brookings Papers on Economic Activity 2008 (2): 365-412.

Sachs, J.D., and A.M. Warner. 1997. Sources of slow growth in African economies. Journal of African Economies 6 (3): 335-376.

Sissoko, Y., and S. Dibooglu. 2006. The exchange rate system and macroeconomic fluctuations in subSaharan Africa. Economic Systems 30 (2): 141-156.

Suliman, A., K. Elmawazini, and M.Z. Shariff. 2015. Exchange rates and foreign direct investment: Evidence for sub-Saharan Africa. The Journal of Developing Areas 49 (2): 203-226.

Tapsoba, S.J.A. 2008. Trade intensity and business cycle synchronicity in Africa. Journal of African Economies 18 (2): 287-318.

Udomkerdmongkol, M., O. Morrissey, and H. Gorg. 2009. Exchange rates and outward foreign direct investment: US FDI in emerging economies. Review of Development Economics 13 (4): 754-764.

Alexandre Henry [1]

[mail] Alexandre Henry

henry.alexandre7@gmail.com

[1] Laboratoire BETA, Universite de Lorraine, 6 rue des Michottes, 54000 Nancy, France

(1) UNCTAD COMTRADE data was not available for Chad and Equatorial Guinea

(2) This Economic Community is born in 1983 and regroups 11 states: CAEMC members as well as Angola, Burundi, Dem. Rep. of the Congo, Rwanda and Sao Tome and Principe.

(3) Monetary Union project launched in 2000 regrouping six west African countries namely: Gambia, Ghana, Guinea, Nigeria, Sierra Leone and Liberia.

(4) According to UN COMTRADE data, imports from the EU represented 43% of total imports of the CFA zone over the period 1980-2010. It is important to highlight the fact that this share is on a downward trend. The share has been reduced to 31% on average over the period 2010-2015. An increasing share of these imported goods is priced in dollars or other Asian currencies.

Caption: Fig. 1 Impulse response functions, a CFA sample, b non-CFA sample
Table 1 Share of the commodity in total exports
Source: Authors' calculations from COMTRADE
data-set, average 1995-2015

(a) CFA countries

Country             Share of         Share of other   Other
                    commodity        commodities      exports
                    (price fixed     (natural         (%)
                    on int'l         resources
                    markets) (%)     only) (%)

Benin               91               2                6
Burkina Faso        90               10               0
Cameroon            97               3                0
Congo               85               0                15
Central Afr. Rep.   97               0                3
Ivory Coast         93               4                4
Gabon               100              0                0
Guinea Bissau       0                100              0
Mali                97               3                0
Niger               71               21               7
Togo                59               0                41
Senegal             45               25               30

(b) Extra-CFA countries

Country             Share of         Share of other   Other
                    commodity        commodities      exports
                    (price takers)   (agricultural    (%)
                    (%)              only) (%)

Angola              98               2                0
Burundi             100              0                0
Sao Tome et P.      100              0                0
Rwanda              81               17               3
Gambia              0                37               63
Ghana               100              0                0
Guinea              85               10               6
Nigeria             100              0                0
Sierra Leone        97               3                0

Share is expressed as share in terms of value, not volume.
In addition, only exports that represent at least
2% of the total exports are taken into account

Table 2 PVAR estimates

Variable            CFA (SE)              Extra CFA (SE)

Equation 1: FDI
L.FDI               0.797 *** (0.0615)    0.419 *** (0.100)
L.D.REER            -0.0466 (1.43)        7.37 *** (2.29)
L.BCA               0.0158 (0.0328)       -0.128 * (0.0679)
L.D.GDP             0.0263 (0.0355)       -0.0501 (0.0339)
D.ToT               0.0387 *** (0.0143)   -0.0211 (0.0201)

Equation 2: REER
L.FDI               -0.004 *** (0.001)    -0.000860 (0.00180)
L.D.REER            -0.075 (0.049)        0.321 *** (0.065)
L.BCA               -0.001 (0.005)        -0.001 (0.001)
L.D.GDP             0.002 (0.001)         -0.001 (0.001)
D.ToT               0.000 (0.000)         -0.000 (0.005)
Y1994               -0.341 *** (0.094)    --

Equation 3: BCA
L.FDI               -0.072 (0.102)        0.198 ** (0.0796)
L.D.REER            14.61 *** (3.18)      -16.7 *** (2.94)
L.BCA               0.460 *** (0.065)     0.775 *** (0.0876)
L.D.GDP             -0.062 (0.061)        0.00241 (0.0548)
D.ToT               0.164 *** (0.024)     0.101 *** (0.0262)

Equation 4: GDP
L.FDI               0.188 ** (0.0955)     0.360 *** (0.0908)
L.D.REER            -8.93 *** (3.43)      -3.11 * (1.73)
L.BCA               0.088 (0.057)         0.190 ** (0.0937)
L.D.GDP             -0.116 (0.071)        0.369*** (0.135)
D.ToT               -0.020 (0.016)        -0.0360 (0.0252)

Significance Levels * 10%; ** 5%; *** 1%. For the sake of space,
the exogenous variable Y1994 is only represented for the REER
equation where it is significant

Tabic 3 PURT test

Variable   Extra-CFA               CFA

           Stat-test     P value   Stat-test     P value

D.GDP      81.035 ***    0.000     171.503 ***   0.000
BCA        33.701 ***    0.008     68.695 ***    0.000
FDI        22.054 ***    0.008     67.939 ***    0.000
D.REER     109.716 ***   0.000     224.542 ***   0.000

Tests are performed with a constant and trend.
Lag is set to 1 Significance Levels * 10%; ** 5%; *** 1%
COPYRIGHT 2019 Association for Comparative Economic Studies
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2019 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Henry, Alexandre
Publication:Comparative Economic Studies
Article Type:Report
Geographic Code:60SUB
Date:Jun 1, 2019
Words:7230
Previous Article:Labor Market Conditions and Geographic Mobility in the Eurozone.
Next Article:Trade in Fuels and Environmental Regulation: A Two-Sided Story.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters