A Short-run Schumpeterian Trip to Embroynic African Monetary Zones

Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 1. Introduction It is now an economic fact that, the spectre of the European Monetary Union (EMU) crisis is looming substantially and scaring potential monetary zones. With renewed interest in the economics of monetary union following this EMU crisis, very few papers have recently examined the feasibility of the proposed African monetary zones (Tsangarides & Qureshi, 2008; Asongu, 2012ab; Alagidede et al., 2011). Moreover, studies on the proposed West African Monetary Zone (WAMZ) (Debrun et al., 2005; Celasun & Justiniano, 2005) and the embryonic East African Monetary Zone (EAMZ) (Mkenda, 2001; Buigut & Valev, 2005) over the past decade are scarce. Hitherto, the focus of these studies has been on the optimality of the proposed currency areas (Mkenda, 2001; Asongu, 2012a; Buigut & Valev, 2005), costs and benefits of candidate countries (Debrun et al., 2005) and adjustments to shocks (Celasun & Justiniano, 2005; Alagidede et al., 2011; Asongu, 2012b). Results of the works are broadly consistent with one fact: the need for greater improvements in structural and institutional characteristics (that will facilitate convergence) in light of a paramount lesson of the EMU crisis1 (Willet, 2011; Willet & Srisorn, 2011). In spite of the substantially documented role finance plays in the economic growth of a monetary union (De Avila, 2003), little (if nothing) is known about evidence of the financegrowth nexus in the proposed WAMZ and EAMZ. According to De Avila, the analysis of the main channels through which policy changes may affect growth indicate that, the harmonization process has impacted growth (via increase in the level of efficiency of financial intermediation) and the liberalization of capital controls has principally affected growth through improvements in the degree of efficiency in financial intermediation (p.4). In the experience of the EMU (Vickers, 2000), embryonic African monetary zones constitute ideal scenarios to analyze the finance-growth nexus. They also present the opportunity of shedding light on some of the unresolved issues on causality between finance and growth in sub-Saharan Africa (SSA)2. In light of the above, this study is a short-run trip to the proposed monetary unions in Africa. We assess the Schumpeterian thesis for the positive spillovers of financial services on growth. Causality analysis is performed on seven financial development and three growth indicators. Schumpeter postulated that an efficient financial system greatly helps in economic prosperity. As emphasized by King & Levine (1993), Schumpeter disputed that, well-functioning banks spur technological innovation by offering funding to entrepreneurs that have the best chances of successfully implementing innovative products and production process. Opposed to this mainstream consensus are sympathizers of Andersen & Tarp (2003) who have concluded that, contrary to what Schumpeterian authors claim, the positive link between financial development and growth has not been sufficiently documented in recent empirical works. Andersen & Tarp have vehemently argued that, turning to the empirical evidence, the alleged first-order effect whereby financial development causes growth is not adequately supported by econometric work. Hence, they conclude that the empirical evidence on the finance-growth nexus does not yield any clear-cut picture (p. 1). This second school of thought has recently been supported by Asongu (2011a) in a meta-study of 186 papers on the finance-growth nexus. It will therefore be interesting to examine the positions of the embryonic African monetary zones in light of the above debate. The rest of the paper is organized as follows. Section 2 presents the data and discusses the methodology. The empirical analysis is covered in Section 3. Section 4 concludes. 1 Serious disequilibria in a monetary union result from arrangements not designed to be robust to a variety of shocks. 2 See “Finance and Growth: A Schumpeterian Trip to Africa” by Baonza (2011) for more details. 860 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 2. Data and Methodology 2.1 Data We examine a sample of 4 West and 5 East African countries with data from African Development Indicators (ADI) and the Financial Development and Structure Database (FDSD) of the World Bank for the period 1980-2010. Guinea is left-out of the WAMZ due to data constraints. The summary statistics of the variables and details on the countries investigated are presented in Panel A and Panel B respectively of Appendix 1. Variable definitions and corresponding sources are presented in Appendix 2. A number of theoretical papers on finance and growth that emerged following the insights of the early endogenous growth models (Romer, 1990; Grossman & Helpman, 1991; Lucas, 1988) have documented three main channels to growth: 1) the rise in the rate of private savings; 2) increase in the efficiency of the financial intermediation process and; 3) the rise in the social productivity of capital (Pagano, 1993). Within the framework of our study, only the first two points are taken into consideration. For organizational purposes, the financial variables are presented in terms of financial intermediary dynamics of depth (money), activity (credit), efficiency and size. Firstly, from a financial depth standpoint, we are consistent with the FDSD and recent African finance literature (Asongu, 2012c) in measuring financial depth both from overall-economic and financial system perspectives with indicators of broad money supply (M2/GDP) and financial system deposits (Fdgdp) respectively. Whereas the former represents the monetary base plus demand, saving and time deposits, the latter denotes liquid liabilities of the financial system. It is interesting to distinguish between these two aggregates of money supply because, since we are dealing exclusively with African countries, a great chunk of the monetary base does not transit through the banking sector. Secondly, financial activity is appreciated in terms of credit allocation. Thus, the paper seeks to appreciate the ability of banks to grant credit to economic operators. We use measurements of both banking-system-activity and financial-system-activity in terms of “private domestic credit by deposit banks: Pcrb” and “private credit by deposit banks and other financial institutions: Pcrbof” respectively. Thirdly, financial intermediary size is measured in terms of deposit bank assets as a proportion of total assets (deposit bank assets plus central bank assets). Fourthly, financial efficiency3 appreciates the ability of deposits (money) to be converted into credit (financial activity). This fourth measure appreciates the fundamental role of banks in transforming mobilized deposits (savings) into credit for businesses or the private sector (Asongu, 2011b). Accordingly, we adopt indicators of banking-system-efficiency and financial-system-efficiency (respectively ‘bank credit on bank deposits: Bcbd’ and ‘financial system credit on financial system deposits: Fcfd’). The correlation analysis presented in Appendix 3 shows that, employment of two variables in almost every financial dynamic category is a form of robustness check. Hence, we are able to cross-check financial system results with those of the banking system for the most part. Three measures of economic growth are employed: GDP growth, GDP per capita growth and real GDP output. While the first two are in growth rate, the last is in natural logarithm. 2.2 Methodology The estimation technique typically follows mainstream literature on testing the short-run effect of financial variables on economic activity (Starr, 2005). The approach entails unit tests to examine the stationarity properties of the variables before a Granger causality approach is used to examine the short-term effects (Engle & Granger, 1987). Impulse response functions are used to further assess the tendencies of significant Granger causality results. 3 By financial efficiency here, we neither refer to the profitability-related concept (notion) nor to the production efficiency of decision making units in the financial sector (via Data Envelopment Analysis). 861 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 3. Empirical analysis 3.1 Unit root tests The assessment of stationarity is based on two types of first generational panel unit root tests. When the variables exhibit unit roots in levels, we accordingly test for stationarity in their first differences. Employment of the Granger causality approach requires that the variables do not have a unit root (or are stationary). Two main types of panel unit root tests have been documented: first generational (that is based cross-sectional independence) and the second generational (which supposes cross-sectional dependence). A necessary condition for the employment of the latter generational test is a cross-sectional dependence test which is only applicable if the number of cross-sections (N) in the panel is above the number of periods in the cross-sections (T). Given that we have 31 periods (T) and 5(or 4) cross-sections (N), we are limited to the first generational type. Therefore, both the Levin, Lin & Chu (LLC, 2002) and Im, Pesaran & Shin (IPS, 2003) tests are employed. While the former is a homogenous based panel unit root test (with a common unit as null hypothesis), the latter is a heterogeneous oriented test (with individual unit roots as null hypotheses). In case of conflicting results, IPS (2003) takes precedence over LLC (2002) in decision making because, consistent with Maddala & Wu (1999), the alternative hypothesis of LLC (2002) is too powerful. In line with Liew (2004), goodness of fit (or optimal lag selection) for model specification is ensured by the Hannan-Quinn Information Criterion (HQC) and the Akaike Information Criterion (AIC) for the LLC (2002) and IPS (2003) tests respectively. Table 1: Panel unit root tests Panel A: Unit root tests for the WAMZ F. Depth (Money) M2 Fdgdp Level First difference Finance Fin. Efficiency F. Activity (Credit) F. Size BcBd FcFd Pcrb Pcrbof Dbacba LLC tests for homogenous panel c ct c ct 0.879 -0.828 -5.01*** -3.58*** 1.252 0.200 -2.81*** -4.14*** -0.738 0.691 -6.65*** -6.20*** -2.89*** -0.125 -3.80*** -3.46*** c ct c ct 0.103 -0.828 -6.47*** -5.54*** 0.647 -0.121 -4.71*** -5.52*** 0.101 1.616 -6.79*** -6.42*** -1.52* -1.34* -4.10*** -3.86*** 2.150 2.390 -2.10** -2.82*** 2.142 2.612 -1.130 -2.30** 3.028 0.047 -8.82*** -4.57*** Economic Growth GDP growth rates Real GDPg GDPpcg Output -6.24*** -6.23*** n.a n.a -6.16*** -6.71*** n.a n.a 3.229 -1.024 -6.61*** -6.49*** -5.77*** -5.89*** n.a n.a -5.62*** -6.10*** n.a n.a 3.865 -0.159 -7.36*** -7.93*** IPS tests for heterogeneous panel Level First difference 2.513 3.685 -3.33*** -3.15*** 2.398 3.840 -2.39*** -2.98*** 1.844 -0.799 -9.36*** -9.05*** Panel B: Unit root tests for the EAMZ F. Depth (Money) M2 Fdgdp Level First difference Finance Fin. Efficiency F. Activity (Credit) F. Size BcBd FcFd Pcrb Pcrbof Dbacba LLC tests for homogenous panel c ct c ct 4.969 3.126 -3.36*** -3.74*** 5.386 2.463 -2.86*** -3.08*** -0.461 0.304 -9.25*** -9.10*** c ct c ct 4.028 2.126 -3.71*** -3.29*** 5.061 2.289 -3.66*** -3.20*** -1.324* 0.002 -8.73*** -8.94*** -0.774 1.517 -1.86** 1.054 2.478 2.778 -0.135 -0.888 2.009 2.631 -2.80*** -6.60*** Economic Growth GDP growth rates Real GDPg GDPpcg Output 0.912 0.566 -9.67*** -4.63*** -5.25*** -5.17*** n.a n.a -6.26*** -0.861 n.a n.a 1.459 1.730 -7.03*** -5.40*** 1.192 0.260 -10.7*** -6.15*** -4.94*** -4.54*** n.a n.a -6.09*** -3.15*** n.a n.a 2.358 -0.026 -6.88*** -4.80*** IPS tests for heterogeneous panel Level First difference -1.70** -2.49*** n.a n.a 2.234 -0.227 -3.16*** -3.26*** 1.817 -0.430 -3.62*** -4.95*** Notes: ***, **, *denote significance at 1%, 5% and 10% respectively. ‘c’ and ‘ct’: ‘constant’ and ‘constant and trend’ respectively. Maximum lag is 8 and optimal lags are chosen via HQC for LLC test and AIC for IPS test. LLC: Levin, Lin & Chu (2002). IPS: Im, Pesaran & Shin (2003). M2: Money Supply. Fdgdp: Liquid Liabilities. BcBd: Banking System Efficiency. FcFd: Financial System Efficiency. Pcrb: Banking System Activity. Pcrbof: Financial System Activity. Dbacba: Deposit Bank Assets on Total Assets. GDP: Gross Domestic Product. GDPg: GDP growth. GDPpcg: GDP per capita growth. WAMZ: West African Monetary Zone. EAMZ: East African Monetary Zone. 862 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 Table 1 above shows results for the panel unit root tests. While Panel A presents the findings for the WAMZ, those of Panel B are of the EAMZ. For both monetary zones, while the financial variables are overwhelmingly integrated in the first order (i.e: they can be differenced once to be stationary), the economic variables are stationary in levels (with the exception of real output). 3.2 Granger causality for finance and growth Let us consider the following basic bivariate finite-order VAR models: Growthi ,t   ijGrowthi ,t  j   ijFinancei ,t  j  i   i ,t (1) Financei ,t   ijFinancei ,t  j  ijGrowthi ,t  j  i   i ,t (2) p q j 1 j 0 p q j 1 j 0 where, Growth denotes economic prosperity (GDP growth, GDP per capita growth or real GDP output) while, Finance represents financial development dynamics (of depth, efficiency, activity and size). Simple Granger causality is based on the assessment of how past values of a financial indicator could help past values of a growth indicator in explaining the present value of the growth indicator (Eq. 1). In the same vein, it also implies investigating how past values of growth variables are significant in helping the past values of financial variables to explain the present value of financial variables (Eq. 2). In mainstream literature, this model is applied on variables that do not exhibit unit root (in levels for the most part). Within our framework, we are applying this test to all ‘finance and growth’ pairs in both ‘first difference’ and levels for three reasons: (1) ensure comparability; (2) consistency with application of the model to stationary variables and; (3) robustness checks in case we might have missed-out something in the unit root test specifications. In light of the above, the resulting VAR models in first difference are the following: Growthi ,t   ijGrowthi ,t  j   ijFinancei ,t  j  i   i ,t (3) Financei ,t   ijFinancei ,t  j   ijGrowthi ,t  j  i   i ,t (4) p q j 1 j 0 p q j 1 j 0 The null hypothesis of Eq. (4) is the position that, ‘Growth does not Granger cause Finance’. Accordingly, a rejection of the null hypothesis is captured by the significant Fstatistics, which is the Wald statistics for the joint hypothesis that estimated parameters of lagged values equal zero. Optimal lag selection for goodness of fit is in accordance with Liew (2004). 863 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 Table 2: Short-run Granger causality analysis for the WAMZ Panel A: Finance and GDP growth Null Hypothesis: Finance does not cause GDP growth Financial Depth (Money) M2 Fdgdp Levels 1st Difference Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba 0.331 0.378 0.152 0.185 0.628 0.623 1.044 D[M2] D[Fdgdp] D[BcBd] D[FcFd] D[Pcrb] D[Pcrbof] D[Dbacba] 0.108 0.030 1.050 0.893 0.988 0.963 0.016 Null Hypothesis: GDP growth does not cause Finance Financial Depth (Money) M2 Fdgdp Levels st 1 Difference Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba 0.392 0.365 0.808 1.177 0.912 0.793 3.324** D[M2] D[Fdgdp] D[BcBd] D[FcFd] D[Pcrb] D[Pcrbof] D[Dbacba] 0.405 0.302 1.418 1.738 0.017 0.027 2.160 Panel B: Finance and GDP per capita growth Null Hypothesis: Finance does not cause GDP per capita growth Financial Depth (Money) M2 Fdgdp Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba Levels 0.171 0.222 0.054 0.031 0.331 0.341 0.880 1st Difference D[M2] 0.134 D[Fdgdp] 0.029 D[BcBd] 0.839 D[FcFd] 0.631 D[Pcrb] 0.934 D[Pcrbof] 0.904 D[Dbacba] 0.015 Null Hypothesis: GDP per capita growth does not cause Finance Financial Depth (Money) M2 Fdgdp Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba Levels 0.291 0.249 1.024 1.341 1.024 0.909 3.405** 1st Difference D[M2] 0.412 D[Fdgdp] 0.305 D[BcBd] 1.431 D[FcFd] 1.825 D[Pcrb] 0.019 D[Pcrbof] 0.029 D[Dbacba] 2.233 Panel C: Finance and Real GDP Output Null Hypothesis: Finance does not cause Real GDP Output Financial Depth (Money) M2 Fdgdp Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba Levels 0.242 0.115 0.068 0.032 0.210 0.197 0.952 1st Difference D[M2] 0.118 D[Fdgdp] 0.054 D[BcBd] 0.120 D[FcFd] 0.033 D[Pcrb] 0.112 D[Pcrbof] 0.156 D[Dbacba] 2.151 Null Hypothesis: Real GDP Output does not cause Finance Financial Depth (Money) M2 Fdgdp Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba Levels 1.531 1.512 8.126*** 9.216*** 9.742*** 10.35*** 0.779 1st Difference M2 1.215 Fdgdp 1.297 BcBd 2.370* FcFd 2.675* Pcrb 7.351*** Pcrbof 8.01*** Dbacba 2.070 M2: Money Supply. Fdgdp: Liquid liabilities. BcBd: Bank credit on Bank deposit (Banking System Efficiency). FcFd: Financial credit on Financial deposits (Financial System Efficiency). Pcrb: Private domestic credit from deposit banks (Banking System Activity). Pcrbof: Private domestic credit from deposit banks and other financial institutions (Financial System Activity). Dbacba: Deposit bank assets on Total assets (Banking System Size). Fin: Financial. WAMZ: West African Monetary Zone. 864 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 Table 3: Short-run Granger causality analysis for the EAMZ Panel A: Finance and GDP growth Null Hypothesis: Finance does not cause GDP growth Financial Depth (Money) M2 Fdgdp Levels st 1 Difference Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba 0.021 0.074 3.732** 7.306*** 1.174 1.912 1.404 D[M2] D[Fdgdp] D[BcBd] D[FcFd] D[Pcrb] D[Pcrbof] D[Dbacba] 0.032 0.052 0.571 2.864* 2.801* 2.088 0.015 Null Hypothesis: GDP growth does not cause Finance Financial Depth (Money) M2 Fdgdp Levels 1st Difference Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba 1.249 1.333 0.048 3.050* 2.399* 2.506* 0.695 D[M2] D[Fdgdp] D[BcBd] D[FcFd] D[Pcrb] D[Pcrbof] D[Dbacba] 0.172 0.042 0.522 2.319 2.175 1.311 0.617 Panel B: Finance and GDP per capita growth Null Hypothesis: Finance does not cause GDP per capita growth Financial Depth (Money) M2 Fdgdp Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba Levels 0.258 0.087 6.269*** 8.292*** 2.227 3.551** 1.245 1st Difference D[M2] 0.248 D[Fdgdp] 0.297 D[BcBd] 0.891 D[FcFd] 2.810* D[Pcrb] 3.715** D[Pcrbof] 3.042* D[Dbacba] 0.082 Null Hypothesis: GDP per capita growth does not cause Finance Financial Depth (Money) M2 Fdgdp Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba Levels 1.589 1.675 0.016 2.342 3.232** 2.935* 0.797 1st Difference D[M2] 0.211 D[Fdgdp] 0.146 D[BcBd] 0.416 D[FcFd] 2.040 D[Pcrb] 1.671 D[Pcrbof] 0.937 D[Dbacba] 0.926 Panel C: Finance and Real GDP Output Null Hypothesis: Finance does not cause Real GDP Output Financial Depth (Money) M2 Fdgdp Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba Levels 0.175 0.163 3.387** 4.183** 0.368 1.338 0.581 1st Difference D[M2] 1.486 D[Fdgdp] 1.357 D[BcBd] 0.764 D[FcFd] 3.256** D[Pcrb] 0.949 D[Pcrbof] 1.516 D[Dbacba] 0.390 Null Hypothesis: Real GDP Output does not cause Finance Financial Depth (Money) M2 Fdgdp Financial Efficiency BcBd FcFd Fin. Activity (Credit) Pcrb Pcrbof Fin. Size Dbacba Levels 0.608 0.675 0.707 1.368 0.359 0.143 3.055* 1st Difference M2 0.279 Fdgdp 0.464 BcBd 1.687 FcFd 1.809 Pcrb 0.472 Pcrbof 0.415 Dbacba 3.764** M2: Money Supply. Fdgdp: Liquid liabilities. BcBd: Bank credit on Bank deposit (Banking System Efficiency). FcFd: Financial credit on Financial deposits (Financial System Efficiency). Pcrb: Private domestic credit from deposit banks (Banking System Activity). Pcrbof: Private domestic credit from deposit banks and other financial institutions (Financial System Activity). Dbacba: Deposit bank assets on Total assets (Banking System Size). Fin: Financial. Fin: Financial. EAMZ: East African Monetary Zone. Table 2 and Table 3 above present Granger causality results for the WAMZ and the EAMZ respectively. Regardless of tables, Panel A, Panel B and Panel C show ‘Finance and GDP growth’, ‘Finance and GDP per capita growth’ and ‘Finance and real GDP output’ 865 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 causality estimations respectively. The Schumpeterian thesis is based on the top-half of each panel which has a null hypothesis of: ‘Finance does not Granger cause Growth’. The bottom halves (with null hypotheses: ‘Growth does not Granger cause Finance’) are relevant complementary assessments of tendencies in the finance-growth nexus. From the results in Table 2, the following could be established: (1) there is overwhelmingly no evidence of finance causing growth; (2) real GDP output causes financial allocation efficiency and financial activity and; (3) the scanty evidence of GDP growth and GDP per capita growth causing financial size is not very robust because of ‘level significance’4. The following conclusions could be derived from Table 3: (1) financial allocation efficiency is instrumental in GDP growth, GDP per capita growth and real GDP output, while financial activity causes only GDP growth and GDP per capita growth and; (2) the evidence of growth causing financial development can only be validated for financial size (Panel C) with respect to real GDP output because it is both significant in levels and first difference5. The simple fact that we have seen evidence of Granger causality flowing from some financial variables to growth dynamics is not enough to draw any economic inferences. Hence, the impulse-response functions (IRFs) of such relationships should provide additional material on the scale and timing of a one standard deviation shock in the financial variables and the responses of the growth dynamics. 3.3 Impulse response for the EAMZ Using a Choleski decomposition on a VAR with ordering: 1) financial variable, 2) growth dynamic; we compute IRFs for the finance-growth nexus. We know from intuition that the Schumpeterian thesis advocates for positive spillovers of financial services on growth. Hence, we expect positive shocks in financial services (financial system efficiency, banking system activity and financial system activity) to improve growth dynamics at least in the shortrun because of the long-run neutrality of money. Appendix 4-9 show graphs corresponding to the IRFs. The dotted lines are the two standard deviation bands, which are used to measure the significance (Agénor et al., 1997, p. 19). It could be observed that, but for the responses of GDP growth (GDP per capita growth) to financial system efficiency in Appendix 4 (6)6, there is an overwhelming significant positive short-run impact on the temporary components of the growth dynamics. Convergence of the effect to zero towards the 10th year confirms the longrun neutrality of monetary policy variables on real output (growth). 3.4 Robustness checks In order to ensure that our results and estimations are robust, we have checked and performed the following. (1) For almost every financial variable (depth, efficiency or activity), two indicators have been used. Hence, the findings have broadly encapsulated measures of financial development dynamics both from banking and financial system perspectives. (2) Three measures of economic growth have been employed as well to capture growth both from overall economic, per capita and real output standpoints. (3) Both homogenous and heterogeneous assumptions have been considered in the unit root tests. (4) Optimal lag selection for model specifications has been consistent with the goodness of fit 4 It should be recalled that financial size for the WAMZ is stationary only in first difference (see Panel A in Table 1). 5 Financial size for the EAMZ is also stationary only in first difference (see Panel B of Table 1). 6 A possible explanation for these initial negative responses is the substantially documented evidence of surplus liquidity issues in African financial institutions (Saxegaard, 2006; Fouda, 2009). 866 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 recommendations of Liew (2004)7. (5) Granger causality has been performed both in level and first difference equations. (6) Impulse response functions have been used to further assess the tendencies of significant Granger causality results and correspondingly, the Schumpeterian thesis. 3.5 Monetary policy implications The traditional discretionary monetary policy arrangement favors a short-run effect of changes in monetary policy variables on economic activity (especially real output). This favors arrangements such as international economic integration (monetary unions and inflation targeting for example). Results of the EAMZ are broadly consistent with this traditional strand. The significant absence of any short-run effect of monetary policy on output in the WAMZ is consistent with the non-traditional strand of policy regimes that limit the ability of monetary authorities to use policy to offset output fluctuations. Thus, the inability of monetary policy to affect short-run real GDP is in line with the stance of Week (2010) who views this International Monetary Fund (IMF) oriented approach as absurdly inappropriate because a vast majority of SSA countries lack the instruments to make monetary policy effective. Hence, the monetary authority in the potential WAMZ may not use policy instruments in the short-run to offset adverse shocks to output by pursuing either an expansionary or a contractionary policy. 4. Conclusion With the spectre of the Euro crisis looming substantially large and scaring potential monetary unions, this study has been a short-run trip to embryonic African monetary zones to assess the Schumpeterian thesis for positive spillovers of financial services on growth. Causality analysis has been performed with seven financial development and three growth indicators in the proposed West African Monetary Zone (WAMZ) and East African Monetary Zone (EAMZ). The journey has been promising for the EAMZ and lamentable for the WAMZ. Results of the EAMZ are broadly consistent with the traditional discretionary monetary policy arrangements while those of the WAMZ are in line with the non-traditional strand of regimes in which policy instruments in the short-run cannot be used to offset adverse shocks to output. Acknowledgement The author is highly indebted to the editor and referees for their useful comments. “The major findings in the current simulation study are previewed as follows. First, these criteria managed to pick up the correct lag length at least half of the time in small sample. Second, this performance increases substantially as sample size grows. Third, with relatively large sample (120 or more observations), HQC is found to outdo the rest in correctly identifying the true lag length. In contrast, AIC and FPE should be a better choice for smaller sample. Fourth, AIC and FPE are found to produce the least probability of under estimation among all criteria under study. Finally, the problem of over estimation, however, is negligible in all cases. The findings in this simulation study, besides providing formal groundwork supportive of the popular choice of AIC in previous empirical researches, may as well serve as useful guiding principles for future economic researches in the determination of autoregressive lag length” (Liew, 2004, p. 2). 7 867 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 Appendices Appendix 1: Summary Statistics and Presentation of Countries Panel A: Summary Statistics Economic Growth Finance Growth Rates Real Output Fin. Depth Fin. Efficiency Fin. Activity Fin. Size GDPg GDPpcg M2 Fdgdp BcBd FcFd Pcrb Pcrbof Dbacba West Mean 3.459 0.740 9.521 0.226 0.154 0.625 0.629 0.096 0.099 0.502 African Monetary Zone (WAMZ) S.D Min. Max. Obser. 5.499 -19.01 27.462 124 5.108 -18.63 22.61 124 0.855 8.248 11.31 124 0.116 0.091 0.796 114 0.093 0.045 0.600 114 0.347 0.173 2.103 117 0.326 0.209 1.812 114 0.066 0.014 0.350 114 0.068 0.014 0.368 114 0.273 0.054 1.350 117 East African Monetary Zone (EAMZ) Mean S.D Min. Max. Obser. 4.077 6.606 -50.24 35.22 143 1.208 6.246 -46.89 37.83 143 9.581 0.456 8.774 10.49 147 0.224 0.118 0.046 0.498 134 0.171 0.110 0.026 0.414 134 0.676 0.282 0.070 1.609 146 0.819 0.357 0.139 1.968 134 0.112 0.074 0.011 0.255 134 0.137 0.097 0.011 0.349 134 0.628 0.198 0.110 0.999 141 Panel B: Presentation of countries West African Monetary Zone (WAMZ) East African Monetary Zone (EAMZ) The Gambia, Ghana, Nigeria, Sierra Leone Burundi, Kenya, Rwanda, Uganda, Tanzania S.D: Standard Deviation. Min: Minimum. Max: Maximum. Obser : Observations. Fin: Financial. Appendix 2: Variable Definitions Variables Signs Variable Definitions Sources Economic Prosperity GDPg GDP Growth (Annual %) World Bank (WDI) Per Capita Economic Prosperity GDPpcg GDP Per Capita Growth (Annual %) World Bank (WDI) Real Output Output Logarithm of Real GDP World Bank (WDI) Economic financial depth (Money Supply) M2 Monetary Base plus demand, saving and time deposits (% of GDP) World Bank (FDSD) Financial system depth (Liquid liabilities) Fdgdp Financial system deposits (% of GDP) World Bank (FDSD) Banking system allocation efficiency BcBd Bank credit on Bank deposits World Bank (FDSD) Financial system allocation efficiency FcFd Financial system credit on Financial system deposits World Bank (FDSD) Banking system activity Pcrb Private credit by deposit banks (% of GDP) World Bank (FDSD) Financial system activity Pcrbof Private credit by deposit banks and other financial institutions (% of GDP) World Bank (FDSD) Banking System Size Dbacba Deposit bank assets/ Total assets (Deposit bank assets plus Central bank assets) World Bank (FDSD) Infl: Inflation. M2: Money Supply. Fdgdp: Liquid liabilities. BcBd: Bank credit on Bank deposits. FcFd: Financial system credit on Financial system deposits. Pcrb: Private domestic credit by deposit banks. Pcrbof: Private domestic credit by deposit banks and other financial institutions. WDI: World Development Indicators. FDSD: Financial Development and Structure Database. GDP: Gross Domestic Product. 868 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 Appendix 3: Correlation Matrices Panel A: West African Monetary Zone (WAMZ) Financial Depth Fin. Efficiency Financial Activity Economic Growth GDPg 1.000 GDPpcg 0.985 1.000 Output 0.080 0.124 1.000 GDPpcg 0.951 1.000 Fdgdp 0.109 0.065 -0.105 0.990 1.000 BcBd 0.069 0.055 0.294 0.020 0.062 1.000 FcFd 0.062 0.043 0.238 0.022 0.056 0.966 1.000 Pcrb 0.101 0.057 0.108 0.646 0.682 0.746 0.731 1.000 Pcrbof 0.100 0.057 0.150 0.634 0.675 0.745 0.735 0.994 1.000 Panel B: East African Monetary Zone (EAMZ) Financial Depth Fin. Efficiency Financial Activity Economic Growth GDPg 1.000 M2 0.097 0.050 -0.175 1.000 Output 0.205 0.173 1.000 M2 -0.115 -0.150 0.427 1.000 Fdgdp -0.072 -0.110 0.497 0.989 1.000 BcBd -0.162 -0.162 -0.447 0.148 0.106 1.000 FcFd -0.357 -0.344 -0.665 0.010 -0.057 0.870 1.000 Pcrb -0.199 -0.224 0.215 0.893 0.884 0.450 0.278 1.000 Pcrbof -0.243 -0.276 0.152 0.912 0.900 0.461 0.344 0.953 1.000 F. Size Dbacba 0.183 0.127 0.079 0.478 0.537 0.528 0.547 0.780 0.766 1.000 GDPg GDPpcg Output M2 Fdgdp BcBd FcFd Pcrb Pcrbof Dbacba F. Size Dbacba 0.008 -0.012 0.374 0.583 0.576 0.234 0.079 0.600 0.533 1.000 GDPg GDPpcg Output M2 Fdgdp BcBd FcFd Pcrb Pcrbof Dbacba M2: Money Supply. Fdgdp: Liquid liabilities. BcBd: Bank credit on Bank deposit (Banking System Efficiency). FcFd: Financial credit on Financial deposits (Financial System Efficiency). Pcrb: Private domestic credit by deposit banks (Banking System Activity). Pcrbof: Private credit from deposit banks and other financial institutions (Financial System Activity). Dbacba: Deposit bank asset on Total assets (Banking System Size). Fin: Financial. Fin: Financial. Appendix 4: Financial System Efficiency and GDP growth (EAMZ) Response to Cholesky One S.D. Innovations ± 2 S.E. Res pons e of D(FCFD) to D(FCFD) Res pons e of D(FCFD) to D(GDPG) .08 .08 .06 .06 .04 .04 .02 .02 .00 .00 -.02 -.02 -.04 -.04 1 2 3 4 5 6 7 8 9 10 1 2 Res pons e of D(GDPG) to D(FCFD) 4 3 3 2 2 1 1 0 0 -1 -1 -2 -2 -3 -3 2 3 4 5 6 7 8 4 5 6 7 8 9 10 Res pons e of D(GDPG) to D(GDPG) 4 1 3 9 10 1 869 2 3 4 5 6 7 8 9 10 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 Appendix 5: Banking System Activity and GDP growth (EAMZ) Response to Cholesky One S.D. Innovations ± 2 S.E. Res pons e of D(PCRDBGDP) to D(PCRDBGDP) Res pons e of D(PCRDBGDP) to D(GDPG) .016 .016 .012 .012 .008 .008 .004 .004 .000 .000 -.004 -.004 1 2 3 4 5 6 7 8 9 10 1 2 Res pons e of D(GDPG) to D(PCRDBGDP) 3 4 5 6 7 8 9 10 9 10 Res pons e of D(GDPG) to D(GDPG) 4 4 3 3 2 2 1 1 0 0 -1 -1 -2 -2 -3 -3 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 Appendix 6: Financial System Efficiency and GDP per capita growth (EAMZ) Response to Cholesky One S.D. Innovations ± 2 S.E. Res pons e of D(FCFD) to D(FCFD) Res pons e of D(FCFD) to D(GDPPCG) .08 .08 .06 .06 .04 .04 .02 .02 .00 .00 -.02 -.02 -.04 -.04 1 2 3 4 5 6 7 8 9 10 1 Res pons e of D(GDPPCG) to D(FCFD) 2 3 4 5 6 7 8 9 10 Res pons e of D(GDPPCG) to D(GDPPCG) 4 4 3 3 2 2 1 1 0 0 -1 -1 -2 -2 -3 -3 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Appendix 7: Banking System Activity and GDP per capita growth (EAMZ) Response to Cholesky One S.D. Innovations ± 2 S.E. Res pons e of D(PCRDBGDP) to D(PCRDBGDP) Res pons e of D(PCRDBGDP) to D(GDPPCG) .016 .016 .012 .012 .008 .008 .004 .004 .000 .000 -.004 -.004 1 2 3 4 5 6 7 8 9 10 1 Res pons e of D(GDPPCG) to D(PCRDBGDP) 4 3 3 2 2 1 1 0 0 -1 -1 -2 -2 -3 -3 2 3 4 5 6 7 8 9 3 4 5 6 7 8 9 10 Res pons e of D(GDPPCG) to D(GDPPCG) 4 1 2 10 1 870 2 3 4 5 6 7 8 9 10 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 Appendix 8: Financial System Activity and GDP per capita growth (EAMZ) Response to Cholesky One S.D. Innovations ± 2 S.E. Res pons e of D(PCRDBOFGDP) to D(PCRDBOFGDP) Res pons e of D(PCRDBOFGDP) to D(GDPPCG) .020 .020 .015 .015 .010 .010 .005 .005 .000 .000 -.005 -.005 1 2 3 4 5 6 7 8 9 10 1 Res pons e of D(GDPPCG) to D(PCRDBOFGDP) 2 3 4 5 6 7 8 9 10 Res pons e of D(GDPPCG) to D(GDPPCG) 4 4 3 3 2 2 1 1 0 0 -1 -1 -2 -2 -3 -3 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Appendix 9: Financial System Efficiency and real GDP output (EAMZ) Response to Cholesky One S.D. Innovations ± 2 S.E. Res pons e of D(FCFD) to D(FCFD) Res pons e of D(FCFD) to D(LOGREALGDP) .08 .08 .04 .04 .00 .00 -.04 -.04 1 2 3 4 5 6 7 8 9 10 1 Res pons e of D(LOGREALGDP) to D(FCFD) 2 3 4 5 6 7 8 9 10 Res pons e of D(LOGREALGDP) to D(LOGREALGDP) .06 .06 .04 .04 .02 .02 .00 .00 -.02 -.02 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 References Agénor, P, R., McDermott, C. J., and Ucer, E. M. (1997) “Fiscal Imbalances, Capital Inflows, and the Real Exchange Rate: The Case of Turkey” IMF Working Paper 97/1. Alagidede, P., Coleman, S., and Cuestas, J. C. (2011) “Inflationary shocks and common economic trends: Implications for West African Monetary Union membership”, Journal of Policy Modeling: Forthcoming. Andersen, T., and Tarp, F. (2003) “Financial Liberalization, Financial Development and Economic growth in LDCs” Journal of International Development 15, 189-209. Asongu, S. A. (2011a) “Finance and growth: Schumpeter might be wrong in our era. New evidence from Meta-analysis” MPRA Paper No. 32559. 871 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 Asongu, S. A. (2011b) “Why do French civil-law countries have higher levels of financial efficiency?” Journal of Advanced Research in Law and Economics 2(2), 94-108. Asongu, S. A. (2012a) “Are Proposed African Monetary Unions Optimal Currency Areas? Real, Monetary and Fiscal Policy Convergence Analysis” African Journal of Economic and Management Studies: Forthcoming. Asongu, S. A. (2012b) “REER Imbalances and Macroeconomic Adjustments in the Proposed West African Monetary Union” African Governance and Development Institute Working Paper. Asongu, S. A. (2012c) “African Financial Development Dynamics: Big Time Convergence” African Journal of Economic and Management Studies: Forthcoming. Baonza, J. D. (2011) “Finance and Growth: A Schumpeterian Trip to Africa”, Barcelona GSE. http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1898932 (Accessed on: 02/11/2012). Buigut, S. K., and Valev, N. T. (2005) “Is the Proposed East African Monetary Union an Optimal Currency Area? A Structural Vector Autoregression Analysis” World Development 33(12), 260-267. Celasun, O., and Justiniano, A. (2005) “Synchronization of output fluctuations in West Africa: Implications for monetary unification” IMF Working Paper. De Avila, D. R. (2003) “Finance and Growth in the EU: New Evidence from the Liberalization and Harmonization of the Banking Industry” European Central Bank Working Paper No. 266. Debrun, X., Masson, P., and Pattillo, C. (2005) “Monetary union in West Africa: Who might gain, who might lose and why?” Canadian Journal of Economics 38(2), 454-481. Engle, R. F., and Granger, W. J. (1987) “Cointegration and error correction: Representation, estimation and testing” Econometrica 55, 251-276. Fouda, O. J. P. (2009) “The excess liquidity of banks in Franc zone: how to explain the paradox in the CEMAC” Revue Africaine de l’Integration 3(2), 1-56. Grossman, G. N., and Helpman, E. (1991) Innovation and Growth in the Global Economy The MIT Press, Cambridge, Massachusetts. Im, K. S., Pesaran, M. H., and Shin, Y. (2003) “Testing for unit roots in heterogeneous panels” Journal of Econometrics 115, 53-74. King, R. G., and Levine, R. (1993) “Finance and Growth: Schumpeter Might Be Right” The Quarterly Journal of Economics MIT Press 108(3), 717-737. Levin, A., Lin, C. F., and Chu, C. S. (2002) “Unit root tests in panel data: asymptotic and finite-sample properties” Journal of Econometrics, 108, 1-24. Liew, V. K. (2004) “Which lag selection criteria should we employ?” Economics Bulletin 872 Economics Bulletin, 2013, Vol. 33 No. 1 pp. 859-873 3(33), 1-9. Lucas, R. E. (1988) “On the Mechanics of Economic Growth” Journal of Monetary Economics 22, 3-42. Maddala, G. S., and Wu, S. (1999) “A comparative study of unit root tests with panel data and a new simple test” Oxford Bulletin of Economics and Statistics 61, 631-52. Mkenda, B. K. (2001) “Is EA an optimum currency area?” Working Papers in Economics, No. 41 School of Economics and Commercial Law, Goteborg University. Pagano, M. (1993) “Financial Markets and Growth: An Overview” European Economic Review 37, 613-622. Romer, P. M. (1990) “Endogenous Technological Change” Journal of Political Economy 98(5), S71-S102. Saxegaard, M. (2006) “Excess liquidity and effectiveness of monetary policy: evidence from sub-Saharan Africa” IMF Working Paper 06/115. Starr, M. (2005) “Does money matter in the CIS? Effects of monetary policy on output and prices” Journal of Comparative Economics 33, 441-461. Tsangarides, C. G., and Qureshi, M. S. (2008) “Monetary Union Membership in West Africa: A Cluster Analysis” World Development 36(7), 1261-1279. Vickers, J. (2000) “Monetary union and economic growth”, National Bank of Belgium Working Paper Research No. 10. Weeks, J. (2010) “Why Monetary Policy is Irrelevant in Africa South of the Sahara”, School of Oriental and African Studies, Center for Development and Policy Research, Development Viewpoint No. 53. http://www.soas.ac.uk/cdpr/publications/dv/file59766.pdf (accessed: 29/10/2012). Willett, T. D., (2011). “Some Lessons for Economists from the Financial Crisis”. In Indian Growth and Development Review 3(2), 186-208. Willett, T. D., and Srisorn, N., (2011, June). “Some Lessons for Asia from the Euro Crisis”, TheClaremont Colleges, Workshop on Global Growth and Economic Governance Implications for Asia: George Mason University. 873