The past two decades have seen a large increase in foreign bank entry across the globe. The increase in foreign bank participation has been especially strong in the transition countries of Central and Eastern Europe and Latin America, reaching well above 80% of the number of banks in several countries (Claessens et al. 2008). The effects of foreign bank participation on lending to small and medium enterprises (SMEs) have been a controversial issue among academics and policymakers alike. The recent crisis has rekindled this debate, with multinational banks based in OECD countries forced to retreat from developing and emerging markets. Similarly, SME lending has been an important topic during the ongoing crisis in the Eurozone and globally.
What does theory say?
Theory and previous empirical work is ambiguous on the effect of foreign bank entry on SME lending. On the one hand, theories on organisational structure of banks suggest that, given their hierarchical organisational structure, foreign banks tend to lend to large and transparent firms relying on ‘hard’, easy to process information. Domestic banks, given their decentralised structure, are, on the other hand, better equipped to extend loans to small and opaque firms based on ‘soft’ information (Stein 2002). Detragiache et al. (2008), for example, show that foreign banks tend to ‘cherry pick’ clients and extend loans only to large and transparent firms because they are better at monitoring hard information. On the other hand, Berger and Udell (2006) argue that only differentiating between transactional (hard information) and relationship (soft information) lending is oversimplified. Large foreign banks may be able to overcome their informational disadvantage with the help of alternative transactional lending technologies, which are better suited for small and opaque firms. Hence, foreign banks may be able to target the same clientele as domestic banks by employing different lending technologies. Whether foreign banks cater to different clienteles and/or use different lending technologies for the same clientele is ultimately an empirical question.
Previous micro-level evidence has compared the clientele of domestic with the clientele of foreign banks. Most prominently, Mian (2006) shows that clients of foreign banks in Pakistan are of larger size, more transparent, in larger cities and more likely to be foreign-owned, inferring from that the lending techniques foreign banks apply. This analysis, however, confounds two effects – differences in clientele and differences in lending techniques. Do foreign banks use different lending techniques because they have different clienteles or do they use different lending techniques even for the same customers of domestic banks? Our findings suggest that both factors play a role, as we will discuss below. While banks employ different lending technologies to firms of different size, domestic and foreign banks employ different lending technologies even when lending to the same borrower.
We have access to the entire Bolivian credit registry for the period between January 1998 and December 2003. For each loan, we have information on the origination and maturity dates, contract terms, and ex-post performance. For each borrower, we have information about their industry, physical location, legal structure, bank lending relationships, and whether they have been delinquent or defaulted on any loan. We focus on commercial loans denominated in US dollars from one of the 13 commercial banks that were active in Bolivia during this period. All in all, this yields 32,279 loans to 2,672 firms. Consistent with previous research, the clients of foreign banks are, on average, larger and riskier firms with ‘weaker’ bank-lending relationships. In particular, loans originated by foreign banks are less likely to be given to sole proprietorships and more likely to be given to joint stock companies, which are typically larger firms. In order to understand whether differences in contract terms between domestic and foreign bank loans are solely due to their different clienteles or also due to the use of different lending technologies, we hold the clientele constant and compare the contract terms of domestic and foreign bank loans to the same firm in the same month. To this end, we restrict our analysis to a sub-sample of loans to firms that receive a new loan from at least one foreign and one domestic bank in the same month. The restriction results in a sub-sample of 5,137 loans to 287 firms. This sub-sample constitutes 25% of the total lending amount of the entire sample. In a robustness exercise we confirm our key findings for the larger sample of firms albeit the less strict controls for different clienteles.
Our main findings
Exploring variation in loan contract terms for a sample of firms that borrow from both domestic and foreign banks in the same month, we find that:
- Foreign banks charge loan interest rates that are on average between 89 and 107 basis points lower than the interest rates of domestic banks, which constitutes a 9% discount relative to the interest rate of domestic bank loans in the sample.
- Foreign bank loans are, on average, 27 percentage points more likely to have collateral – a large effect given that only 33% of all loan contracts in our sample include collateral – and have maturities that are up to 33% shorter than domestic bank loans, which, at the average maturity of nine months, implies a difference of two to three months.
- Domestic banks base their loan pricing on the length of their relationship with the borrower, especially in the case of smaller firms, while foreign banks have a more transaction-based pricing approach, relying on borrower ratings and collateral, especially for larger firms.
Figure 1. Differences in loan conditionality between domestic and foreign banks
Our findings are consistent with significant differences between foreign and domestic banks in how they cater to enterprise borrowers. While foreign banks rely more on collateral and shorter maturity as disciplining tools and hard information as input for the loan pricing, domestic banks rely more on relationship and soft information as lending technologies, but compensate with higher interest rates.
We are also able to distinguish between two different forms of foreign bank entry, as Bolivia had four foreign branches and three foreign subsidiaries during the sample period. Foreign branches are typically smaller operations focusing on small segments of the overall market, such as wholesale operations and investment banking, with less focus on retail operations. We find that both foreign branches and subsidiaries charge lower interest rates, with the discount being stronger for branches. On the other hand, there are interesting differences in the risk mitigation tools branches and subsidiaries use. While foreign branches focus on having collateral for their loans, foreign subsidiaries focus more on having shorter maturity loans.
What about performance?
If foreign and domestic banks set their loan conditions and price their loans in an optimal way, then we should not observe any differences in arrears, i.e., borrowers should be as likely to repay loans given to them by domestic or foreign banks in the same month. We find, however, that foreign bank loans are more likely to fall into arrears than domestic bank loans. This effect is, however, stronger among uncollateralised and higher-maturity loans, which gives support to the risk mitigation techniques of foreign banks to grant shorter maturities and ask for collateral.
Our research provides micro-insights into the variation in lending techniques across banks with domestic and foreign ownership, by holding constant the sample of borrowers. Our results also allow an assessment of the effects of foreign bank entry as a function of the informational and contractual frameworks of countries. As foreign banks depend more on collateral and credit ratings, our results suggest that foreign banks will not be able to lend to SMEs in countries where collateral rights cannot be effectively created and enforced and in markets with little information available about enterprises.
Beck, Thorsten, Vasso Ioannidou, and Larissa Schäfer (2012), “Foreigners vs. Natives: Bank Lending Technologies and Loan Pricing”, CentER and ECB Discussion Paper, forthcoming.
Berger, Allen N and Gregory F Udell (2006), “A More Complete Conceptual Framework for SME Finance”, Journal of Banking & Finance, 30:2945-2966.
Claessens, Stijn, Neeltje Van Horen, Tugba Gurcanlar, and Joaquin Mercado Sapiani (2008), “Foreign Bank Presence in Developing Countries 1995-2006: Data and Trends”, (Mimeo).
Detragiache, Enrica, Poonam Gupta, and Thierry Tressel (2008), “Foreign Banks in Poor Countries: Theory and Evidence”, Journal of Finance, 63:2123-2160.
Mian, Atif (2006), “Distance Constraints: The Limits of Foreign Lending in Poor Economies”, Journal of Finance, 61:1465-1505
Stein, Jeremy C (2002), “Information Production and Capital Allocation: Decentralized versus Hierarchical Firms”, Journal of Finance, 57:1891-1921.