Financial inclusion and poverty: a tale of forty-five thousand households

ABSTRACT

Using a new nationally representative survey data covering approximately 45,000 Indian households, we examine the effects of financial inclusion on poverty. We construct a multidimensional indicator of financial inclusion and examine the effects of financial inclusion on multiple measures of poverty including the household Poverty Probability Index (PPI), household deprivation scores, and poverty line. We find that financial inclusion has a strong poverty-reducing effect. This finding is consistent across the different measures of poverty used, and alternative ways of measures financial inclusion. These results underpin the importance of financial inclusion and the need for its promotion across countries.

KEYWORDS:

• Financial inclusion
• banking
• credit
• insurance
• poverty

JEL CLASSIFICATION:

• G21
• G22
• I30

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ These individuals include 16,874 respondents who agreed they were household heads. Note that some respondent did not answer the question about whether or not they were household heads. However, every respondent included is surveyed about their household characteristics.

² The common trend in the literature is to control for household head characteristics in household-level regressions, however, given that all respondents in our survey are not households heads (or at least did not agree that they were household heads), we proceed to control for characteristics of respondents with the assumption that these respondents to an extent, represent their various households. To ensure robustness of our results, we also examine our relationship of interest in the household head sample only in Table A4 (i.e. the sample where respondents agree that they are household heads).

³ See https://www.povertyindex.org/about-ppi for details on the construction of the PPI.

⁴ In our dataset, access to bank captures households that have either a checking, savings or fixed deposit accounts.

⁵ See Lewbel (2012) for details.

⁶ These results are consistent with those reported for the sub-sample where respondents are household heads (Appendix Table A4).

⁷ F-statistics from first stage regressions suggest that our instruments used in both specifications are not weak.