Are Mergers Responsible for the Surge in New Bank Charters

Abstract

After stagnating for many years, the rate of new bank formation increased sharply in the second half of the 1990s. The financial press attributes this development to the high volume of bank mergers, which are said to have encouraged new entry by reducing service to some bank customers. It is commonly asserted, for example, that many new banks serve small businesses whose banks were taken over by larger banks uninterested in making small business loans. Most banking experts agree that such an increase in new banks in response to mergers would be healthy, helping maintain competition in local banking markets and offset reductions in service.> The view that mergers encourage new bank formation has recently come into question. Examining data on new bank charters and mergers in the 1990s, a study released early last year concluded that mergers have actually discouraged new bank formation. Shortly thereafter, another study came to the opposite conclusion, finding that mergers encourage new entry. Taken together, these studies raise two important questions. First, is merger activity positively or negatively related to new bank formation? Second, if mergers are positively related to new bank formation, which types of mergers account for the link?> Keeton reexamines the relationship between mergers and new bank charters, distinguishing more carefully than the other two studies between different types of mergers. The results, based on data for the second half of the 1990s, provide strong support for the view that mergers encourage the formation of new banks. Specifically, the author finds that markets with more merger activity experienced higher rates of new bank formation, and that the mergers with the strongest link to new bank formation were those in which small banks were taken over by large banks or local banks by distant banks.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 47 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.