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

exceeds a certain threshold, the relationship between finance and growth is positive, while it becomes negative below the threshold. The intuitive explanation for this result is that financial development lowers transaction costs on private investment, but also reduces the revenue of seignior age usable for public investment. It is supportive of growth only if the government can obtain other revenue to finance infrastructure, that is, if the institutional quality is sufficient to allow the collection of taxes other than by tax Inflationary. If the institutional quality is too low, Seignior age's revenue loss cannot be offset by the collection of new taxes, and the infrastructure necessary for development cannot be programmed. Our literature review concludes with the result that financial development is not conceivable without a sound institutional framework conducive to the development of economic and financial activities. This brings an additional guarantee to our idea of building from the outset of our research, an indicator of financial development that incorporates the quality of the institutions in determining the level of efficiency of the financial sector. III. M ethodology a) Creating a new financial development indicator We calculated our development index through two steps. First, we calculated a composite index of the quality of institutions. For this, we referred to the databases of World Governance Indicators, December 2018, built thanks to the work of Kaufman and al. This is a database with indicators relating to 6 variables of institutional development, mainly the voice and accountability, political stability and no Violence, government effectiveness, regulatory quality, the rule of law, and control of corruption. We extracted data about each of these variables from this basis to build an index successively for the quality of political institutions and then an index for the quality of economic institutions. Each variable is rated between -2.5 and +2.5. We combined these institutional variables with six financial variables whose data were derived from the Global Financial Development Database (GFDD) 2017. These variables are bank credit to bank deposit, deposit money bank asset to GDP, domestic credit to the private sector, Private credit by deposit money banks and other financial institutions to GDP, Liquid liabilities to GDP, and Financial system deposits to GDP. After ensuring the availability of data on all dimensions of our final indicator of financial development, we selected a sample of 97 countries, including countries from all continents around the world. And it’s from 1996 to 2016, which is the time interval within which we obtain data. Finally, we used the Principal Component Analysis method on the XLSTAT in Excel software to get our financial indicator. b) Estimation method in static and dynamic panel data: the fixed effects model with random effects, the GMM model in System - The Fixed effects and random effects models  Fixed effects model This model, also known as the covariance model , assumes that Ui and Vi are constant, non- random effects, which therefore change the value of the econometric equation constant according to the values i and t. This is an estimate that is carried out by the Ordinary Least Squares (OLS), after an addition to the explanatory variables of the indicator variables, or dummy variables, associated with individuals i and periods t (less an individual and a period to not create co linearity with the Constant. Assuming that the random cross-disturbance Wit satisfies the conventional assumptions of the OLS (i.e., they are centered, homoscedastic, independent, and normal), the estimates are optimal and allow for particular Fisher Tests to test the need for the terms U i or V t . The fixed- effects model is: = α + 1 + 2 + 3 + 4 + 5 + Where FINANCE is financial development, INTECO is economic institutions, INSTPO is political institutions, INSTFIN is financial institutions, RGDPC is real GDP per capita, the subscripts i and t index countries and time respectively. Also, the specification contains an unobservable country-specific effect α and error-term .  The random-effects model This model, also called the compound error model , assumes the random U i , V t . The basic specification assumes: o The centered U i , V t, and W it (zero expectation) o The respective U i , V t, and W it homoscedastic and standard deviation σ u, σ v, Σ w. o U i , V t, and W it are not correlated and independent The idea of this modeling is that the three no longer practice on the constant of the model, but really on the random disturbance Є . The method then aims to clarify these effects to take them into account to refine the estimate. Under the assumptions indicated, the variance of the Є hazard is: ( ) = ( ∗ ) + ( ∗ ) + ( ∗ ) Although fixed-effects and random-effects models appear to be different, the second is generally recommended. Tests (notably Hausman) allow testing both hypotheses. And from the moment when the main © 20 20 Global Journals 26 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