A key regulatory and public policy goal is maintenance of competitiveness in markets for financial services – especially for small businesses that have limited access to capital markets but thought to be critical to the health of the economy. Regulators in the United States frequently use concentration ratios as the initial filter for assessing market competitiveness in merger reviews, with further analysis of product pricing changes for cases where a concentration threshold is violated. Implicit in this approach is an assumption that higher concentration leads to reduced competition and worse outcomes (higher prices and/or reduction in credit availability and service) for market participants, a prediction that follows from the structure–conduct–performance (SCP) literature (Bain, 1951 and Hannan, 1991).1
The problems associated with using the SCP model to analyze competitiveness in local markets are well known, especially for small, information-opaque firms. The impact of competition on small firm credit availability or loan pricing depends on the incentives of lenders to invest in private information acquisition (Petersen and Rajan, 1995), the lender’s portfolio strategy (Boot and Thakor, 2000), and the impact of technology on remote monitoring costs (Dell’Arricia and Marquez, 2004). For example, Petersen and Rajan (1995) show how increased competition could lead to less availability and higher rates, while Dell’Arricia and Marquez (2004) show how it could lead to more lending.
The empirical evidence related to the effect of bank deposit concentration, the most frequently used proxy for degree of competition, on small firm credit availability and loan terms is also mixed. Some cases find deposit concentration associated with improved small firm credit availability (Petersen and Rajan, 1995 and Zarutskie, 2006), reduced credit availability (DeYoung et al., 1999), or to have no effect (Jayaratne and Wolken, 1999, Cavalluzzo et al., 2002 and Berger et al., 2007b). Berger et al. (2007b) note a further complication in empirical tests for the association between deposit concentration and small firm banking outcomes (e.g. credit availability and loan pricing). Size–structure, i.e. the relative share of deposits scaled by total bank assets (within or outside of a Metropolitan Statistical Area or MSA), appears to be more important than either bank size or concentration alone in determining availability and pricing of small firm lines of credit.
The different theoretical predictions of how competition affects banking outcomes, and the conflicting empirical evidence associated with the use of deposit concentration as a proxy for competition, is problematic for regulators and policy makers. They frequently seek to understand how changes in market structure affect the delivery of banking services to small firms, because of the importance of banks as the primary source of external capital for small firms (Berger and Udell, 1998). Two examples of market structure changes include the determination of the relevant market size, which appears to be expanding for some products, and the role of out-of-market competitors that may have no physical presence in that market (Hannan, 2003 and Hannan and Prager, 2004).
This paper employs an alternative measure of competition among banks for small firm financial business using a survey that asks: “Have you noticed any change in competition for your firm’s business among financial institutions now compared to 3 years ago?” The responses fall into one of five categories: “much less,” “less,” “no change,” “more,” and “much more.” These responses should reflect the actions of banks in the firm’s local market. The advantage of this measure compared to deposit concentration as a proxy for competition is its lack of dependence on defining market size or the number (or share) of banks in that market. The responses to this question, described in detail in Section 3.2 below, include any contact (mail, telephone, personal contact, etc) that could be from in- or out-of-market banks.
We first address whether small firm reports of change in bank competition relate to deposit concentration, the traditional proxy for market structure. Our results suggest a complex association between the changes in bank competition, market size, the level and change in deposit concentration, and size–structure within local markets. The primary finding is that changes in bank competition in non-urban markets are negatively related to both the level and change in deposit concentration, e.g. a higher the level of concentration is associated with less frequent bank competition as is an increase in the level of concentration. Offsetting the effect of concentration is a positive association between density and size–structure, where size–structure is the single-market bank deposit share. These effects are strongest in non-Metropolitan Statistical Areas.
Next we address whether changes in bank competition affect small firm banking outcomes regardless of the degree of credit usage. These outcomes are responses to a question about changes (“better,” “no change,” or “worse”) in various attributes of banking products (e.g. service delivery, credit availability, and lending terms). The inclusion of non-credit outcomes is a unique contribution to understanding the effect of bank competition on all small firms, especially for firms that just use banks for transaction purposes. This analysis is motivated by two assumptions: (1) small firms have low switching costs; and (2) there are low barriers to new bank entry. Both assumptions have support in the literature. Under these conditions, banks should increase competition to maintain (or increase) market share through improved pricing and/or service.
Two empirical challenges arise in estimating the association between changes in bank competition and small firm banking outcomes. The first is whether causality runs from changes in bank competition to changes in banking outcomes. For example, small firms could erroneously attribute changes in bank competition as a result of a favorable outcome in their most recent credit experience. We devise a test that provides some assurance that owners “know competition when they see it.” Even with this assurance, a second challenge arises from the possibility that the error term is correlated with the change in bank competition. For example, banks may devise a strategy to contact new customers based on a favorable risk/return profile so that improved outcomes are due to firm characteristics, not increased competition. We address this challenge by identifying instruments for the change in bank competition variable.
Our results show that an increase in bank competition has a positive and significant association with reported changes in banking outcomes. Use (or non-use) of credit, as well as firm characteristics associated with information opacity (e.g. years in business or size), does not affect this result. Small firms located in markets with higher levels of deposit concentration report higher levels of satisfaction, but this result appears to be driven by their choice of a small bank as their primary financial institution. Additional tests show that the effect of lender competition on banking outcomes is unaffected by length of time with the current lender and the timing of the most recent loan.
The results of the paper provide additional evidence that traditional definitions of local markets, such as MSA, and proxies for competition in those markets, such as deposit concentration, are incomplete measures for describing the competitive environment for small firms and their primary bank. These results add to the growing evidence of wider markets for some bank services such as Radecki, 1998 and Hannan and Prager, 2004 for deposit rates and Hannan (2003) with respect to out-of-market loans for small firms.
The rest of the paper is organized as follows. Section 2 presents a brief review of the competing theories and empirical evidence related to banking market competition and its effect on the banking outcomes of small firms. Section 3 discusses the survey data and key variable construction. Section 4 empirically examines the relation between changes in bank competition and traditional market structure measures, while Section 5 empirically examines the relation between reports of changes in banking service attributes and changes in bank competition. Section 6 offers conclusions.
This paper addresses two questions of fundamental importance to small firms and their banks as well as policy makers who rely on small firms to promote employment growth. First, how does increased bank competition – the proxy for the state of competition in local markets – affect bank service, credit availability, and loan terms? Second, how does market structure affect a small firm’s perception of bank competition for their business? To answer these questions, we use a unique survey data set that measures small business owner reports of the change in bank competition for their business over the previous 3 years. The responses to this question provide a different measure of the competitive environment because it can include the actions of banks within and outside of the local market regardless of their size or number.
The paper has two principal findings. First, owner reports of changes in bank competition for their financial business are related to deposit concentration – the typical proxy for market competition – but in a complex way. Reports of increased competition are negatively associated with both the change and level of deposit concentration (i.e. more concentrated markets and markets where concentration has increased are associated with less competition). Banks operating all of their offices within a single market (MSA or rural county) are more likely to contact small firms as their share of deposits increase in the local market. These effects are largely limited to firms in non-MSA markets.
The second principal finding of the paper is that increases in bank competition improve ratings of both credit and non-credit banking. The results also reveal a more complex picture of how competition may work on banking outcomes because deposit concentration relates positively to banking outcomes as well. However, this effect appears to relate to the owner’s bank size. Increased deposit concentration improves banking outcomes only for owners with a small bank.
Overall, our results support the notion that the behavior of banks can be as important their number or share of deposits in local markets when analyzing small firm banking outcomes. The analysis provides additional evidence that the appropriate market for small business banking products is not congruent with MSA or rural county and complements the findings of Radecki, 1998 and Hannan and Prager, 2004. Yet our data leave many unanswered questions remain regarding the types of banks that are competing for small firm business. For example, how important is the presence of out-of-market banks in local markets in its effect on the behavior of within-market banks? And, is there a non-linear association between competition and outcomes for small firms that borrow from relationship lenders as suggested by Boot and Thakor (2000)? While these questions remain for future research, they do not weaken the primary findings of the paper.
