The Presence of Multinational Banks and the Supply and Quality of Credit in Emerging Economies

Abstract

Barriers to entry for multinational banks (MNBs) have been reduced in many countries. This paper studies the effect of the presence of MNBs on the supply and quality of credit in emerging economies. This study uses data from the Bank for International Settlements and the International Monetary Fund. The results indicate that the credit supply declines in response to increased competition from MNBs. However, the adverse effect of MNBs on the credit supply is less pronounced when the presence of MNBs is larger. The paper also provides some tentative results on the effect of MNBs on the quality of loans. The results suggest that banks shift their portfolios away from loans as a result of increased international financial competition, thereby reducing default risk, which is also reflected in a negative relationship between MNB presence and the chance of a banking crisis occurring.

Key Words:

• Emerging economies
• multinational banks
• credit supply
• quality of loan portfolios

Jel Classifications:

• F14
• F23
• F36

Notes

1. Following Jones (1992), MNBs are banks that establish operations abroad, compared to international banks that operate across international borders, but do not establish a physical presence in other countries.

2. It is theoretically possible that a large enough share of banks has a low enough capital base to be enticed to expand loans for higher risk projects on a large enough scale to lead to more credit at lower quality in the aggregate. While this can be observed for substantial sectors of the financial market, e.g. in the case of the savings and loan crisis in the US, it is an unlikely phenomenon in the aggregate. For instance, even with comparatively high bad loan portfolios in Central and Eastern Europe countries (CEECs) after the end of communism, credit did not rise and quality did not deteriorate. On the contrary, most CEECs remained remarkably stable in the face of the Asian and Russian financial crisis because of improved bank portfolios (Weller & Morzuch, 2001).

3. Offshore banking centers and developed countries are not included. The end is set for 1998 because banking crises are only documented until then in a comprehensive manner.

4. Only countries with a reported MNB presence are included in the study. Thus, the subsequent discussion focuses on the impact of differential sizes of MNB presence on the supply of credit and its quality.

5. The divergence in total credit growth and private sector growth in Eastern Europe is likely an artifact of the privatization of state‐owned enterprises. This should automatically increase private sector credit because previous public sector credit was reclassified. Because of this shift in classification, total credit growth may be the more consistent measure for Eastern Europe.

6. Detailed calculations available from the author.

7. While Nickell (1981) found that panel data estimates are biased when lagged dependent variables are included on the RHS, this bias is negligible with more than 10 periods in the time series.

8. Ideally, per capita personal income should be used instead of per capita GDP. However, personal income data are not available on a consistent basis.

9. The calculated F‐statistic with 1 and 445 degrees of freedom is 170.17. Therefore, the null hypothesis that the additional variable does not improve the efficiency of the estimate is rejected at the 1% level.

10. A comparison of the respective adjusted R ² in Tables 6 and 7 shows that the inclusion of the linear and quadratic terms of the credit market shares provides a better fit than the logarithmic specification.

11. The deposit to loan ratio also increases if the financial sector uses alternative funding methods. For instance, a growing issuance of asset‐backed securities, such as construction bonds, should offset some of the use of deposits for loans, all else equal.

12. It should be noted that fewer loans may only temporarily increase stability. After a while, greater financial constraints may hamper real sector growth and contribute to unrealized speculative expectations (Odedokun, 1996; Levine, 1997; Kaminsky & Reinhart, 1999; Weller, 2001).

13. The capital to asset ratio may not capture all increases in financial stability in banks’ balance sheets. Depending on accounting practices, loan loss provisions, built up in response to MNB presence, may not be counted as capital. There is some evidence that banks increase loan loss provisions in response to international financial competition. For instance, Crystal et al. (2001) found that loan loss reserves relative to non‐performing loans rose in Colombia and Chile from 1997 to 2000 when the presence of MNBs grew, although it fell in Argentina.

14. The literature suggests that the areas most likely to be underserved as a result of the growing MNB presence are small and medium‐sized enterprises, start‐ups, rural producers, among others (Weller & Scher, 2001).