The impact of financial development, economic growth, income inequality on poverty: evidence from India

Abstract

This paper examines the impact of financial development, economic growth and income inequality on poverty in India from 1970 to 2015 by employing the autoregressive distributed lag (ARDL) bounds testing procedure. The findings reveal a robust long-run relationship between financial development, economic growth, inequality and poverty. Results show that financial development and economic growth help in poverty reduction in India, whereas income inequality and inflation aggravate poverty. Empirical evidence of the Granger-causality test supports the presence of unidirectional causality from financial development and economic growth to poverty. Moreover, bidirectional causality exists between inequality and poverty. The present study provides evidence on which the policymakers may proceed with detailed investigation of how specific financial sector policies and interventions can be deployed as effective instruments for achieving favorable economic growth and income distribution. The study recommends that policies geared toward increasing financial development and economic growth should be adopted to reduce the high level of poverty and inequality currently prevailing in India.

Appendix A1: Principle component analysis (PCA) and construction of FDI

Table 8 Principal component analysis for financial development index.

The Principal Component Analysis (PCA) is a special case of more general method of factor analysis. It transforms original set of variables into smaller set of linear combinations that account for most of the variance of the original set. The aim of PCA is to construct out of a set of variables, X$_{j}$’s (j = 1, 2, ..., k) new variables (Pi) called ‘ Principal Components’, which are linear combinations of X’s. The first principal component (P1) is determined as the linear combination of X$_{1}$, X$_{2}$,...,X$_\mathrm{m }$provided that the variance contribution is maximum. The second principal component (P2), independent from the first principal component, is determined as to provide a maximum contribution to total variance left after the variance explained by the first principal component, then the third and the other principal components are determined as to provide the maximum contribution to the remaining variance and independent from each other. The aim here is to determine a$_{ij}$ coefficients providing the linear combinations of variables based on the specified conditions. The following formula is used to construct financial sector development index.

$$\begin{aligned} { FDI}=\mathop \sum \limits _{i=1}^j a_i \frac{X_{ij} }{Sd(X_i )} \end{aligned}$$

where FDI is the financial development index; Sd $=$ Standard Deviation; $X_{ij}$ = $i\mathrm{th}$ items in $j\mathrm{th}$ year; $a_{i}$, = Factor loadings as derived by PCA. Measuring financial development is a complicated procedure because there is no clear-cut definition of financial development and no thumb rule about the inclusion of variables. Bandiera et al. (2000) stated that an ideal index of financial sector development should include various aspects of regulatory and institutional reforms. Inclusion of all the policy variables separately in the same model might cause multi-co linearity. To avoid it, this study uses three different types of financial development indicators to construct the financial development index. All the variables are taken in their natural logarithm. The variables are taken from 1970 to 2015.

The results obtained from principal component analysis are presented in Table 8. Eigenvalues suggest that the first principal component explains about 93.8% of the standardized variance explains 91.68 percent, the second principal component explains another 4.59% and the last principal component accounts for only 1.53% of the variation. It can be easily concluded that the first principal component is better than other components/combination of variables because it explains higher level of variations. Thus, the first eigenvector values are used as a weight to construct a Financial Development Index (FDI) and denoted as FDI. The variables LCREDIT, LGDFC and LM3 are individually contribution the standardized variance of the first principal component, i.e., 57.64, 58.60 and 56.94%, respectively. We have used these contribution as the basis of weighting to construct a financial development index (FDI).