Bank and Nonbank Competition for Small Business Credit: Evidence from the 1987 and 1993 National Surveys of Small Business Finances
Cole, Rebel A., Wolken, John D., Woodburn, R. Louise, Federal Reserve Bulletin
Using newly available data from the Board's 1993 National Survey of Small Business Finances together with data from the 1987 survey, we analyze competition between banks and nonbanks in the U.S. market for small business credit. According to many academics and banking practitioners, the U.S. commercial banking industry has declined. In particular, during the late 1980s and the early 1990s, the record number of bank failures and mergers reduced the number of commercial banks in the United States. Also, there has been an apparent decline in commercial banks' share of lending. These occurrences have raised questions about the changing role of commercial banks in providing credit to key sectors, including business lending.
Several explanations have been advanced for the decline in banks' share of business lending. In particular, technological changes in communications, information storage, and other sectors of the economy - as well as globalization - have enabled an increasing number of large firms to gain direct access to money and capital markets. The same technological changes have facilitated competition from nonbank sources. Nonbanks consist of thrift institutions (savings and loan associations, savings banks, and credit unions), finance companies, insurance companies, mortgage companies, leasing companies, brokerage firms, other business firms, families and individuals, and government sources of credit.
We explore nonbank competition as an explanation for the decline in banks' share of business lending by examining sources of credit used by small firms.(2) Credit here is defined as loans and capital leases, excluding credit card debt and trade credit. Because small firms are unlikely to have direct access to money and capital markets, any decline in banks' share of the aggregate dollar amount of credit provided to these firms would be consistent with the view that nonbanks are eroding this share. If banks have provided a constant or increasing share of the credit used by small firms, such evidence would run counter to the view that nonbanks are eroding this share.
We analyze the bank and nonbank shares of the dollar amount of outstanding credit to small businesses, including how these shares have changed from 1987 to 1993. We also examine the incidence of small business borrowing from banks and nonbanks, which is defined as the percentage of firms using credit of a certain type or from a particular source. The incidence data provide a more representative view of the credit services used by a "typical" small firm than do the share data because larger firms have a greater influence on market shares than on incidence. This distinction is important because the larger firms in the survey account for the majority of the dollar amount outstanding of small business credit but for only a small proportion of the number of firms. For example, among small businesses, firms with more than $1 million in sales account for more than two-thirds of credit but less than one-fifth of the number of firms.
The sources for these data - the 1987 and 1993 National Surveys of Small Business Finances (NSSBF) - are unique.(3) The NSSBF is a nationally representative survey of small businesses sponsored by the Federal Reserve Board and the U.S. Small Business Administration to collect information about the sources and types of financial services obtained by small businesses. The surveys are designed to be representative of small businesses generally and provide data on bank and nonbank shares of the small business credit market.(4) The NSSBF was conducted first in 1987 and again in 1993, making it possible to examine changes in market share over that period. Although the two surveys had somewhat different focuses, the data collected are sufficiently similar to allow comparisons of bank and nonbank market shares across time.(5) However, differences in the coverage of the two surveys preclude comparisons of actual dollar amounts. …