On the Latin American Credit Drivers

ABSTRACT

We add to the literature about credit in Latin America by assessing what has been driving the recent and heterogeneous expansion of credit to gross domestic product based on supply and demand variables. We chose working with these emerging economies due to the low levels of human capital, the divergent patterns of evolution of economic variables and the vulnerability of credit expansion. According to balanced panel estimations, our main findings in terms of public policy suggest that credit reflects a financial deepening characterized by a higher bank concentration and by a policy able to stimulate saving even practicing lower deposit interest rates.

KEY WORDS:

• Credit to GDP in Latin America
• bank deposits to GDP
• bank concentration
• real deposit interest rate

JEL Codes:

• O54
• O16
• G18

Acknowledgments

First draft July 2015. I am thankful to anonymous referees and the editor for their insightful comments. I thank Andrei Simonassi, Giovanni Beviláqua, Jaime De Jesus Filho, and seminar participants at Federal University of Ceará for their comments and suggestions. The usual disclaimer applies.

Funding

The financial support from CNPQ-Brazil is gratefully acknowledged.

Notes

^1. See Sheng, Bortoluzzo, and Dos Santos (2013) for a recent discussion about trade credit in this region during financial crisis, for instance.

^2. In the next section, we describe the specific source for each variable analyzed here in Table 1.

^3. Domestic deposit money banks mean commercial banks and other financial institutions that accept transferable deposits, such as demand deposits.

^4. Total assets include total earning assets, cash and due from banks, foreclosed real estate, fixed assets, goodwill, other intangibles, current tax assets, deferred tax, discontinued operations, and other assets.

^5. In our main exercise, we follow the empirical-related literature, but we also perform our procedure taking into account lagged variables in order to evidence or not some changes in our conclusions due to the timing of some effects.

^6. See Wooldridge (2002) for a further discussion about our estimation setup.

^7. These results are not reported here, but they are available upon request.