Financial development and income in North Africa.
Keywords North Africa * Financial development * Economic growth * VECM * Arab countries
JEL E20 * G40 * G2
Numerous studies have stressed the important role of the financial sector in economic growth. Emphasis on the effect of financial development can be traced back to the writings of Bagehot (1873) and Schumpeter (1912). The influential work of Gurley and Shaw (1955), Goldsmith (1969), Mckinnon (1973), and Shaw (1973) contributed significantly to the early literature. The topic was revived in more recent work, including Levine (1997), Beck et al. (2000), Benhabib and Spiegel (2000), and Levine et al. (2000). Yet, some scholars have maintained that finance may not be a significant determinant of economic growth and development (Robinson 1952; Lucas 1988; Stem 1989).
Recent empirical literature presents mixed findings. For example, King and Levine (1993a) show that the level of financial development is a predictor of productivity improvement and economic development. On the other hand, Demetriades and Hussein (1996) find little empirical evidence that finance causes growth. Moreover, some studies claim that financial development may have a negative influence on growth. Improved financial development, which leads to better resource allocation, increases returns and may lower saving (income effect); thus possibly causing growth to fall (Bencivenga and Smith 1991; King and Levine 1993b).
The goal of this paper is to explore the short-run dynamics and long-run relationship between income and financial development (banking sector development) in North Africa. We use four indicators of financial development and annual data from Algeria, Egypt, and Morocco, Morocco and Egypt have implemented important financial reforms in the past two decades, but the financial sector in Algeria remains controlled by the state and is relatively underdeveloped (see Creane et al. 2006).
We use a bivariate vector-autoregressive (VAR) model in order to be able to compare our results to those obtained by Demetriades and Hussein (1996) who employs similar methodology and has found mixed evidence on the direction of causality between growth and financial development. Interestingly, the empirical results we obtain indicate that the evidence on the direction of causality is mixed. Moreover, the relationship between financial development and income in both the long and the short run is, in many cases, negative. Only in the case of the ratio of liquid liabilities to GDP (the variable LIQ) do we find a positive long-run relationship between financial development and income in all three countries.
Brief Review of the Empirical Literature
It is widely argued in literature that the main effects of financial development on growth (or economic development) operates through enhancing the functions of the financial system. It is argued that financial development reduces risk. improves the allocation of resources, allows a better access to information about investments, improves monitoring, and increases saving mobilization (Levine 1997).
Recent empirical literature on the relationship between financial development and growth includes Bencivenga and Smith (1991), King and Levine (1993a, b), Demetriades and Hussein (1996), Beck et al. (2000), Baliamoune-Lutz (2003), and Calderon and Liu (2003). Interestingly, the empirical literature contains conflicting findings. Some studies report that finance causes growth (King and Levine 1993a), while others maintain that growth causes financial development (Baliamoune-Lutz 2003). Yet, other authors have found mixed results (Demetriades and Hussein 1996), or bi-directional causality (Luintel and Khan 1999; Calderon and Liu 2003). For example, Calderon and Liu (2003) use data from 109 developing and industrial countries from 1960 to 1994 and report that there is bi-direction causality between financial development and growth. The authors also show that financial deepening contributes more to the causality link between growth and financial development in developing countries than in industrial countries.
Overview of the Financial Sector in Algeria, Egypt and Morocco (1)
Similar to other developing countries, the governments of Morocco, Algeria and Egypt practiced high degrees of financial repression during most of the 1970s and the first half of the 1980s. Giovannini and de Melo (1993) use data on 24 developing countries (including Morocco and Algeria) covering the period 1972-1987 and estimate the revenues from financial repression. The authors show that in the period 1974-1987, Algeria derived revenues from financial repression equal to 4.3% of its GDP and 11.4% of its tax revenue, while Morocco's revenues from financial repression (in 1977-1985) amounted to about 2.3% of its GDP and 8.9% of its tax revenue. This was much higher than the revenues obtained from financial repression in Malaysia, Korea, Thailand, Brazil, and Colombia. The authors also observe that
"financial repression appears to be more relevant to African countries (including North Africa) and least important in Asian countries. It is also much more evident in the recent years (1979-1987) than in the early part of the sample. This last phenomenon is associated with growth of fiscal imbalances among LDC's in the 1980s" (Giovannini and de Melo 1993, p. 959).
Of the three countries, Algeria has the least developed financial system. Three state owned commercial banks--Banque Nationale d' Algerie, Banque Exterieure d'Algerie, and Credit Populaire d'Algerie--have dominated the market since the late 1960s. In the 1980s, Algeria attempted to introduce specialized banks that focus primarily on credit to specific sectors.
The state-owned banks account for over 90% of bank assets and they tend to operate inefficiently with a high ratio of non-performing loans (The Economist Intelligence Unit 2006). Algeria liberalized deposit interest rates in 1990. In 1994, the country introduced limits on banking spreads to replace ceilings on lending rates. The limits on banking spreads were removed in 1995 (see Jbili et al. 1997). Currently, there is no significant participation of foreign banks in the Algerian banking sector.
In comparison with Algeria. Morocco has a more developed banking system. Morocco began timid reforms in the early 1980s when it steadily raised administered interest rates, and accelerated the reforms in the late 1980s and early 1990s. As reported in Baliamoune and Chowdhury (2003), most of the early actions targeted changes in interest rates. The monetary authorities liberalized interest rates on above-1-year deposits in 1985, and 6-month and 3-month deposit rates in 1989 and 1990, respectively. In addition, in 1986-1991 a minimum interest rate replaced the fixed rate, which was in place for all other deposits that were still subject to regulation. In 1992, interest rates on all time deposits were liberalized. As soon as the initial reforms were implemented, real interest rates became positive and have continued this trend since 1987. Finally, in February 1996 lending rates were effectively liberalized (Baliamoune and Chowdhury 2003).
Foreign participation in Morocco's banking sector has been present for many decades, going back to the period of its colonization by the French. However, between 1973 and 1993 Morocco enforced the 'Moroccanization' law that restricted foreign ownership of businesses, including banks, to 49 percent of total capital in order to allow Moroccans to have majority-based control Since the abolition of the 'Moroccanization' decree in September 1993 (agricultural land and some mining industries remain subject to restricted foreign ownership) some foreign financial institutions have managed to become majority shareholders in Moroccan banks. For example, the French financial groups Credit Lyonnais, Societe Generale, and BNP Paribas are majority control holders in, respectively, Credit du Maroc (CDM), Societe Generale Marocaine de Banques (SGMB), and Banque Marocaine pour le Commerce et l'Industrie (BMCI).
In the early 1990s, Egypt undertook major economic reforms, including financial reforms. Egypt introduced treasury bill auctions and removed the official limits on interest rates in January 1991. In October 1992, ceilings on bank loans to the private sector were removed. This reform contributed to a substantial increase in the level of credit to the private sector, which averaged 23% of GDP in 1970-1990 and rose to an annual average of about 40% of GDP in 1991-2001. Lending limits on loans to the public sector were removed in 1993. Real interest rates rose significantly in 1997-1998 and became relatively high (around 5%) and positive, after being negative (about -6%) in the early 1990s (Arestis 2003).
It is important to note that one of the objectives of the early reforms was to make the Egyptian currency more attractive to Egyptians and foreigners by liberalizing credit and interest rates. Subsequent reforms focused on enhancing the financial sector competitiveness by facilitating private participation (through privatization) in financial institutions and by increasing the share of foreign participation (Abu-Bader and Abu-Qarn 2005). In 2000, the share of foreign banks (those with foreign ownership of at least 50%) was 20%, up from 3% in 1995 (Lee 2002).
We use annual data for the period 1960-2001 and four indicators of financial development. First, we use the variable DM, defined as the ratio of deposit money bank claims on domestic nonfinancial real sector to the sum of deposit money bank and Central Bank claims on domestic nonfinancial real sector. Second, we use the variable DMGDP, defined as the ratio of deposit money bank assets to GDP. Our third measure of financial development is LIQ. This variable represents the ratio of liquid liabilities to GDP. The fourth indicator of financial development is PRIVCR, defined as the ratio of private credit by deposit money banks and other financial institutions to GDP. All four indicators are assumed to have a positive correlation with financial development, with higher ratios indicating a more sophisticated financial system.
Data on income (purchasing-power-parity value of real GDP per capita) are from the World Bank World Development Indicators database (2004). Data on financial development indicators are from the World Bank Financial Structure database (2003). The estimations are performed using EViews 5.1.
Unit Root and Co-integration
We begin the empirical analysis by performing unit root tests to determine whether the variables are stationary in levels (integrated of order zero). Table 1 displays the Phillips-Perron unit-root test results. (2) All variables have unit root and are stationary in first difference. Next, we test for co-integration using the Johansen's co-integration rank test. Co-integration test results along with relevant critical values are reported in Table 2.
Table 1 Phillips-Perron unit-root test results Level First Difference Adj. t-Stat Prob. (a) Adj. t-Stat Prob. (a) Algeria Income -1.241483 0.6472 -8.052485 0.0000 DM -2.276819 0.1857 -3.733279 0.0090 LIQ -1.887194 0.6269 -2.839811 0.0069 LIQ -1.635339 0.7450 -2.698763 0.0097 PRIVCR -1.710625 0.7119 -2.009838 0.0450 Egypt Income -0.804944 0.8071 -3.578201 0.0107 DM -1.688954 0.4292 -5.469073 0.0000 DMGDP -2.360686 0.3937 -8.048961 0.0000 LIQ -0.729566 0.8278 -4.881439 0.0000 PRIVCR -0.507135 0.9792 -2.500712 0.0137 Morocco Income -2.418446 0.3651 -9.690226 0.0000 DM -1.578126 0.7844 -5.891320 0.0000 DMGDP 1.969451 0.9998 -3.697722 0.0005 LIQ -1.593331 0.7784 -7.458152 0.0000 PRIVCR -1.760777 0.7052 -5.405416 0.0000 (a) MacKinnon (1996) one-sided p-value. The critical values differ according to the number of observations and lags inclded. More details may be obtained from the author Table 2 Co-integration test results--trace statistic Hypothesized Number of Co-Integration Equations None At Most 1 Trace 0.05 Trace 0.05 Statistic Critical Statistic Critical Value Value Algeria Income and DM 29.54217 25.87211 5.633349 12.51798 Income and DMGDP 28.07436 25.87211 7.015156 12.51798 Income and LIQ 23.85200 20.26184 6.336014 9.164546 Income and PRIVCR (a) 14.01231 20.26184 4.202608 9.164546 Egypt Income and DM 43.11480 25.87211 9.237707 12.51798 Income and DMGDP 27.12544 25.87211 6.218187 12.51798 Income and LIQ 36.65358 25.87211 6.471882 12.51798 Income and PRIVCR 20.26195 18.39771 0.365125 3.841466 Morocco Income and DM 27.10467 25.87211 7.756477 12.51798 Income and DMGDP 21.16960 20.26184 7.229231 9.164546 Income and LIQ 26.16436 20.26184 7.130572 9.164546 Income and PRIVCR 30.61442 25.87211 7.506477 12.51798 The critical values differ according to the number of observations and lags included. Eigenvalue test results are consistent with those associated with the trace test. Thus, we report only the results for the trace statistic (a) Income and PIRVCR in Algeria are not co-integrated
Based on the results from the trace test (and the co-integrating equation), we conclude that all financial indicators, except PRIVCR in Algeria, are co-integrated with income. Relevant coefficients from the co-integrating vectors are displayed in Table 3. We note that the long-run relationship between income and financial development is not consistent across indicators and countries. First, only in Algeria do we find a positive long-run relationship between each financial indicator (three indicators, since PRIVCR is not co-integrated with income) and income. This suggests that the effects of financial development could be more important at low levels of financial development. Second, only in the case of LIQ do we find a positive long-run relationship in all three countries. Third, credit to the private sector (PRIVCR) seems to have a negative long-run relationship with income in Morocco and Egypt.
Table 3 Coefficient on the co-integrating vector (LHS variable: income per capita in log) RHS Algeria Egypt Morocco Variable DM 1.1776 (a) (0.145) -0.7937 (a) (0.104) -0.6311 (a) (0.102) DMGDP 0.2330 (a) (0.071) -3.4056 (a) (1.020) 4.4286 (a) (1.4147) LIQ 2.8296 (a) (0.251) 0.6507 (a) (0.063) 2.7888 (a) (1.187) PRIVCR Not co-integrated -0.8702 (b) (0.461) -0.4618 (a) (0.020) Standard errors in parentheses (a) Significant at 1% (b) Significant at 10% (c) Significant at 5%
Adjustment to the Long-run Equilibrium
We incorporate information about the co-integration relationship in our estimation of the vector autoregressive equation. We estimate a bivariate VECM where the two endogenous variables are income per capita and financial development, and include the co-integrating equation (the error-correction term). The coefficients on the error-correction terms in the VECM estimates are shown in Table 4. The coefficients marked as 'ns' (not significant) are either statistically insignificant or have a positive sign. A meaningful coefficient on the error-correction term (also known as the loading factor or speed of adjustment) must be negative and statistically significant.
Table 4 Speed of adjustment (loading factor) (a) Dependent Income DM Income DMGDP Income LIQ Income PRIVCR Variable Algeria ns 0.8877 1.1097 ns 0.8405 ns nc nc Egypt 1.0031 ns 0.1371 ns 0.3648 ns 0.3132 ns Morocco ns 0.9627 0.1079 ns 0.0550 ns ns 0.8257 All reported coefficients are significant at the 5% level or better and have the correct (negative) sign. Detailed results may be obtained from the author upon request ns Not significant and/or does not have a negative coefficient, nc no evidence of co-integration (a) The numbers shown are the absolute values of the coefficients on the error-correction terms.
The results displayed in Table 4 indicate that the speed of adjustment to the long-run equilibrium between financial development and income, as well as the variable that makes the adjustment differ across countries and across financial development indicators. In the case of DMGDP and LIQ, income adjusts in order to restore the long-run equilibrium in all three countries. In the case of credit to the private sector, income adjusts to the long-run equilibrium in Egypt, whereas in Morocco PRIVCR does the adjustment, restoring about 83% of the disequilibrium in one period (1 year). In the case of the variable DM, financial development does the adjustment in Algeria and Morocco. In Egypt, the adjustment is undertaken by income. In fact, in Egypt income adjusts to the long-run equilibrium in all cases.
Finally, the speed of adjustment varies significantly across countries, ranging from a low of 0.05 in the case of LIQ in Morocco (implying it would take about 20 years to restore the long-run equilibrium) to a high of 1.11 for the adjustment to the long-run relationship between DMGDP and income in Algeria (implying it would take less than 1 year to correct the disequilibrium). It is important to keep in mind that the coefficients from the VECM estimation, including the loading factor (coefficient on the error-correction term), provide us with information only about short-run dynamics and do not provide information on long-run causality. The VECM equation includes, in addition to the error correction term, lags of the endogenous variables as well as a constant and a time trend where appropriate. Tests for the lag length were performed using the Wald test in EViews 5.1.
Granger Causality: Weak Exogeneity Tests
To further explore the short-run dynamics we test for weak exogeneity (Granger causality) and report the results in Table 5. The sign on the coefficients (positive or negative) represents the sign of the statistically significant coefficients on the lagged endogenous variable that is not on the left-hand-side (LHS) of that equation. The first three rows under each variable indicate the chi-square value, the p-value, and whether the short-run effect is positive or negative, respectively. In the fourth row we report the lag length as determined by the results from the lag exclusion test (Wald test). Interestingly, the appropriate lag length is generally short in Algeria and long in Morocco and Egypt. This may be due to country differences in the level of financial development.
Table 5 Granger causality/block exogeneity Wald tests null hypothesis: dependent variable is weakly exogenous Wald statistic, (pvalue), (nature of the effect) Algeria Egypt Dependent Variable Y Fd Y Fd fd = DM 2.6579 6.5618 36.5704 4.3881 [0.264] [0.037] [0.000] [0.820] (positive) (positive) Number of lags 2 8 fd = DMGDP 3.2105 5.6264 22.9800 4.6850 [0.360] [0.131] [0.000] [0.455] (positive) Number of lags 3 5 Fd = LIQ 0.6536 9.2067 6.7373 14.2222 [0.721] [0.010] [0.034] [0.000] (negative) (positive) (negative) Number of lags 2 5 fd = PRIVCR 6.2288 0.4472 8.9382 15.3502 [0.012] [0.503] [0.257] [0.031] (positive) (positive) Number of lags 1 7 Morocco Dependent Variable Y fd fd = DM 7.3087 15.7057 [0.397] [0.027] (positive) Number of lags 7 fd = DMGDP 18.8752 8.1842 [0.015] [0.415] (negative) Number of lags 8 Fd = LIQ 19.4904 33.8783 [0.006] [0.000] (positive) (positive) Number of lags 7 fd = PRIVCR 8.9024 17.4960 [0.350] [0.025] (positive) Number of lags 8 The results are from VECM estimations for the variables that are co- integrated (Table 2) and from VAR estimations for the variables that were not co-integrated. The critical values differ according to the number of observations and lags included. The number of lags included in each equation is based on the results from the Wald tests for lag exclusion. We use the p-value of the Wald test to determine the maximum number of lags fd Financial development
The tests for weak exogeneity also yield results that are not consistent across the three countries. In the case of LIQ, we find evidence of bi-directional causality in Morocco and Egypt. However, while the relationship is positive in both directions in Morocco, in Egypt income seems to have a negative short-run effect on financial development (LIQ). In Egypt and Morocco, income has a positive short-run effect on PRIVCR, while in Algeria PRIVCR has a positive impact on income. Finally, only in Egypt does the supply-leading hypothesis--that financial development causes economic development--seem to hold for most indicators (3 out of 4), DM, LIQ, and DMGDP. (3) Thus, the empirical estimates provide mixed evidence, which is in line with the findings in Demetriades and Hussein (1996), and is in contrast to the findings reported in Luintel and Khan (1999).
Since the three countries are at different stages of financial development it might be interesting to try to relate the difference in the empirical results back to the cross country differences in financial development. A straightforward way of doing this is to focus on the long-run behavior of the variables and examine the relationship between each of the financial indicators and its average value in recent years. Table 6 shows the mean value for each indicator for the period 1990-2001.
Table 6 Average values of financial indicators: 1990-2001 Algeria Egypt Morocco DM 0.75 0.70 0.86 DMGDP 0.37 0.67 0.49 LIQ 0.46 0.81 0.70 PRIVCR 0.12 0.38 0.41
It is worth noting that the differences in the link between income and credit to the private sector (which is often viewed as the most appropriate measure of financial development) could be due to the fact that the mean value of credit to the private sector as a ratio of GDP in Algeria is much lower (0.12) than it is in the other two countries. The average values of PRIVCR in Morocco and Egypt are 0.38 and 0.41, respectively. We did not find evidence of a long-run relationship between income and PRIVCR in Algeria, while there is a negative stable relationship between the two variables in Morocco and Egypt.
Given that a positive link between PRIVCR and income has been documented in many studies, the present results seem to suggest that the relationship between credit to the private sector and income may be nonlinear. In countries with high levels of financial development PRIVCR is positively correlated with income, while in countries with medium (or low) levels of financial development PRIVCR has a negative correlation with income. In addition, the size of urban population may contribute to the effectiveness of financial reforms (financial development). In 2001, urban population as a percent of total population in Algeria, Morocco, and Egypt was 58%, 56%, and 43%, respectively.
This paper explores the relationship between income and four indicators of financial development in Algeria, Egypt and Morocco. In most cases, we find a stable long-run relationship between financial development and income. On the other hand, the evidence on the direction of causality and the nature of the relationship is mixed. The results show that the only financial indicator that has a positive long-run relationship with income in all three countries is the ratio of liquid liabilities to GDP (LIQ). We also find that in the short run income adjusts to correct deviations from the long-run equilibrium between LIQ and income in all three countries. This seems to suggest that finance leads economic development when financial development is represented by the ratio of liquid liabilities.
This study is in the spirit of the work by Demetriades and Hussein (1996), in that we use a bivariate VAR and include, among our four indicators, the two measures of financial development that the authors use, namely the variables LIQ and PRIVCR. In general, the empirical results yield strong support to the findings in Demetriades and Hussein (1996). We show that the relationship between financial development and economic development may differ even across countries with comparable levels of economic development and located within the same region. In the case of credit to the private sector, the differences between the level of financial development in Algeria and the levels in the other two countries may explain the mixed results. Possible reasons for the mixed results on the other financial indicators include differences in banking regulation and supervision, and differences in the size of urban population, especially between Morocco and Egypt.
The results also suggest that we need to distinguish between the short-run dynamics and the long-run effects when studying the link between growth and financial development. Given the mixed results and the disparities among countries, an important implication of these findings is that cross-sectional models do not seem to be suitable for the study of the relationship between financial and economic development. This point is also emphasized in Demetriades and Hussein (1996).
Acknowledgements The author wishes to thank Robert Lensink and Stefan Lutz for enlightening discussions and suggestions, and attendees of the Current Issues in Trade and Finance Session at the 63rd International Atlantic Economic Conference in Madrid, March 14-18, 2007 for helpful comments.
Abu-Bader, S. & Abu-Qarn, A. (2005). Financial development and economic growth: Time series evidence from Egypt. Monaster Center for Economic Research, Ben-Gurion University of the Negev, Discussion Paper no. 05-14.
Arestis, P. (2003). Financial sector reforms in developing countries with special reference to Egypt. Levy Economics Institute Working Paper No. 383.
Bagehot, W. (1873). Lombard Street: A description of the money market. Homewood. IL: Richard D. Irwin 1962 edition.
Baliamoune-Lutz, M. (2003). Financial liberalization and economic growth in Morocco: A test of the supply-leading hypothesis. Journal of Business in Developing Nations, 7, 31-50.
Baliamoune-Lutz, M. (2008). Financial reform and the mobilization of domestic savings: The experience of Morocco. In G. Mavrotas (Ed.), Domestic Resource Mobilization and Financial Development. Palgrave Macmillan. Baliamoune, M. N., & Chowdhury, A. R. (2003). The long-run behavior and short-run dynamics of private savings in Morocco. Savings and Development, 27(2), 135-160.
Beck, T., Levine, R., & Loayza, N. (2000). Finance and the sources of growth. Journal of Financial Economics, 58(1/2), 261-300.
Bencivenga, V. R., & Smith, B. D. (1991). Financial intermediation and endogenous growth. Review of Economic Studies, 58, 195-209.
Benhabib, J., & Spiegel, M. (2000). The role of financial development in growth and investment. Journal of Economic Growth, 5, 341-360.
Bolbol, A., Fatheldin, A., & Omran, M. (2005). Financial development, structure, and economic growth: the case of Egypt, 1974-2002. Research in International Business and Finance, 19, 171-194.
Calderon, C., & Liu, L. (2003). The direction of causality between financial development and economic growth. Journal of Development Economics, 72, 321-334.
Creane, S., Goyal, R., Mobarak, A. M., & Sab, R. (2006). Measuring financial development in the Middle East and North Africa: A new database. IMF Staff Papers, 53(3), 479-511.
Demetriades, P. O., & Hussein, K. A. (1996). Does financial development cause economic growth? Time-series evidence from 16 countries. Journal of Development Economics, 51, 387-411.
Giovannini, A., & de Melo, M. (1993). Government revenue from financial repression. The American Economic Review, 83(4), 953-963.
Goldsmith, R. W. (1969). Financial structure and development. New Haven, Connecticut: Yale University Press.
Gurley, J. G., & Shaw, E. S. (1955). Financial aspects of economic development. American Economic Review, 45, 515-538.
Jbili, A., Enders, K., & Treichel, V. (1997). Financial sector reforms in Algeria, Morocco, and Tunisia: A preliminary assessment. IMF Working Paper 97/81, International Monetary Fund, Washington, DC.
King, R. G., & Levine, R. (1993a). Finance and growth: Schumpeter might be right. Quarterly Journal of Economics, 108, 717-737.
King, R. G., & Levine, R. (1993b). Finance, entrepreneurship and growth: Theory and evidence. Journal of Monetary Economics, 32, 1-30.
Lee, Jung-Kun (2002). Financial liberalization and foreign bank entry in MENA. Washington, DC.: The World Bank.
Levine, R. (1997). Financial development and economic growth: Views and agenda. Journal of Economic Literature, 35, 688-726.
Levine, R., Loayza, N., & Beck, T. (2000). Financial intermediation and growth: Causality and causes. Journal of Monetary Economics, 46, 31-77.
Lucas, R. E. (1988). On the mechanics of economic development. Journal of Monetary Economics, 22, 3-42.
Luintel, K. B., & Khan, M. (1999). A quantitative reassessment of the finance-growth nexus: Evidence from a multivariate VAR. Journal of Development Economics, 60, 381-405.
McKinnon, R. (1973). Money and capital in economic development. Washington, DC: Brookings Institution.
MacKinnon, J. G. (1996). Numerical distribution functions for unit root and cointegration tests. Journal of Applied Econometrics, II, 601-618.
Omran, M. (2002). Testing for a significant change in the Egyptian economy under the economic reform programme era. UNU-WIDER Discussion Paper No. 2002/59
Robinson, J. (1952). The generalization of the general theory. In the rate of interest, and other essays pp. 67-142. London: Macmillan.
Schumpeter, J. A. (1912). Theorie der wirtschaftlichen Entwicklung: Eine Untersuchung uber Unternehmergewinn, Kapital, Kredit, Zins, und den Konjunturzyklus. (Leipzig: Dunker und Humblot). Translated as The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest and the Business Cycle, tr. Redvers Opie. Cambridge: Harvard University Press, 1934.
Shaw, E. (1973). Financial deepening in economic development. New York: Oxford University Press.
Stern, N. (1989). The economics of development: A survey. Economic Journal, 99, 597-785.
The Economist Intelligence Unit (2006). Country report: Algeria. London, UK.
World Bank (2003). Financial structure database. Washington D.C.
World Bank (2004). World development indicators database. Washington D.C.
Published online; 2 September 2008
University of North Florida, I UNF Drive, Jacksonville, FL, USA
(1) For more detailed discussions see Jbili et al. (1997) on the reforms in Morocco and Algeria; Baliamoune and Chowdhury (2003) and Baliamoune-Lutz (2003, 2008) on financial reforms in Morocco; and Omran (2002), and Bolbol et al. (2005) on the reforms in Egypt.
(2) We also use the Augmented Dickey-Fuller unit-root test and the results are consistent with those derived from the Phillips-Perron (PP) test. We report only the latter since the PP test is known to be more appropriate if we suspect structural breaks in the data.
(3) The coefficient on the error-correction term in the equation linking PRIVCR to income in Egypt is statistically significant and indicates that income adjusts to restore the equilibrium relationship, which implies short-run (Granger) causality. However, because the long-run relationship is negative, we may not say that the supply-leading hypothesis holds.