Global Journal of Management and Business Research, B: Economics and Commerce, Volume 20 Issue 1

b.The OECD zone The table below shows the results: Table 3: GMM model regression (the 25 OECD countries) FINANCE Coef. Std.Err. t P > |t| FINANCE ( t-1) -0.003 0.002 -1.31 0.203 RGDPC -0.896 0.718 -1.25 0.224 INFLATION 0.052 0.140 0.37 0.716 INTECO -.0.502 0.225 -2.23 0.036** INSTPOL -0.199 0.116 -1.72 0.098* INSTFIN 2.063 0.007 314.03 0.000*** CONSTANT 3.625 2.626 1.38 0.180 Hansen test for overid. restrictions chi2 (98) = 22.20 prob>chi2 = 1.000 Arellano-Bond test for AR (1) z = -0.46 pr> z = 0.648 Arellano-Bond test for AR (2) z = -2.13 pr> z = 0.033** Prob> F = 0.000*** F(5, 24) = 662886.55 Source: Author Notes: INTECO= Economic Institutions; INSTPOL =Political Institutions; INSTFIN= Financial Institutions; RGDPC = Gross Domestic Product per capita. The Arenallo and Bond dynamic panel system GMM estimations (Stata xtabond2 command) is used to estimate this model. P-value*, **, *** indicate respectively 10%,5%and 1%, of significance levels. The Hansen test is accepted the over-identification restrictions. The null hypothesis of the absence of first-order serial correlation (AR1) is accepted, but the absence of second-order serial correlation (AR2) is rejected. These results show us that in the OECD, as in the WAEMU countries, the signs and the significance of the different variables are similar. The results are similar in detail to those obtained above. Indeed, as in the WAEMU zone, the new indicator has its relevance as to the impact it has on the functioning of the financial sector. - The coefficient of the new indicator is positive (+ 2.06) and significant. - As for the gross domestic product and inflation, their coefficients are not significant, as in the estimate on the countries of the WAEMU zone. Therefore, they cannot be interpreted reliably. - And finally, as with the WAEMU area, with OECD countries, we get a coefficient of the delayed variable of non-significant financial development. At this level, too, the GMM system model could have been replaced by the techniques for estimating static panel models (what we do after that). After using the GMM System model estimation method and obtaining results showing the non- significance of the delayed variable coefficient, weconcluded that a static panel estimation technique could have estimated our model. The next part will be devoted to this task. ii. Static panel estimation (fixed and random effects model) a. The WAEMU Zone We have obtained results that support those obtained during our regressions by the GMM System method. First of all: - Global significance tests of both models (Fixed Effects and Random Effects) show that both models are significant. - The signs of the coefficients for the two (2) models are almost identical. - Apart from the Economic Growth variable, whose significance is only certain at a threshold of 10%, all other variables are significant. © 20 20 Global Journals 32 Global Journal of Management and Business Research Volume XX Issue I Version I Year 2020 ( ) B Institutional Quality and Financial Development in West Africa Economic and Monetary Union