You might stop banking crises and bailouts, but not bank failures

During hearings after the recent failures of Silicon Valley Bank, Signature Bank and First Republic Bank — including the last one by the Committee on Oversight Accountability’s Subcommittee on Health Care and Financial Services — several members on both sides of the aisle have asked how can we “end,” “prevent” or “stop” bank failures. Not banking crises or bailouts, but failures. 

Perhaps that just reflects people misspeaking in the heat of the moment. If so, that’s understandable. But if those words were spoken as intended, whatever initiatives follow won’t likely end bank failures.

In the face of uncertainty, firms use adaptation, imitation and trial-and-error to generate profits to survive. But most don’t survive. Bureau of Labor Statistics data shows that firm survival rates fall as firms age. After five years, only about half created in a given year survive. After 10 years, only about 35 percent survive, and so on. Those principles apply to banks, too, though survival rates for banks are likely higher because of the steps required to obtain a bank charter and bank supervision.

Since the FDIC first began reporting bank failures in 1934, there have been an average of 46 per year and a median of seven. While the size of each recently failed bank alarmed many, the assets of those banks combined made up only 2.4 percent of the $23 trillion in total commercial bank assets. In 2008 and 2009, the assets of failed banks peaked at 13.6 percent and 17.7 percent of the roughly $12 trillion in total commercial bank assets at the time, respectively. Keep in mind that we still have about 4,000 commercial banks, which is probably too many.

Understanding U.S. bank failures and banking crises requires an understanding of what happened before the FDIC. Census data on U.S. commercial banks show their number growing throughout the 19th and early 20th centuries, first exceeding 1,000 in 1854, 4,000 in 1884, 10,000 in 1900 and 20,000 in 1908. We reached a peak of almost 31,000 in 1921 — and that does not include savings banks, credit unions and other institutions that also provide loans and facilitate payments.

That dramatic rise in the number of banks was an unintended consequence of the constitutional prohibition on states issuing their own currencies. Lacking that option, states turned to chartering and taxing banks. And since states could not tax out-of-state banks, they simply restricted them from operating within their jurisdictions. Many states even restricted branching within their borders.

Having so many banks led to another unintended consequence: banking crises. These tens of thousands of largely local banks were too geographically undiversified to weather regional economic shocks. Between 1825 and 1929, there were seven major crises and 20 minor ones. By the end of the Great Depression in 1933, fewer than 15,000 commercial banks remained.

Financial historian Eugene White found that this dramatic decline was partly due to the new Federal Reserve’s creation of the discount window, through which troubled banks could borrow even as they tended to remain lost causes. Before then, weak banks tended to choose to close before failing, similar to Silvergate’s recent, voluntary liquidation. That’s because, before the FDIC, a bank’s shareholders could not only lose their investments in the event of a failure — they could also be made to pay additional penalties as they were subjected to contingent (double, triple or even unlimited) liability.

With the discount window in place, weak banks could now borrow and attempt to prolong their existence, and many kept operating until they failed. Since then, while the frequency of U.S. banking crises has declined, the presence of weak banks increases the duration, severity and costs of these crises.

While bank failures are inevitable, crises and bailouts need not be.

Economic historians point out that Canada has had bank failures and recessions and bank merger waves but has never experienced what we would consider a banking crisis since its confederation in 1867. Even though there are fewer than 100 banks in Canada, the largest have always operated coast-to-coast — meaning they had geographically diversified asset holdings and could attract funding from all regions — and they also tended to be well funded with equity capital.

As regulators contemplate bank merger guideline revisions, they should keep the past, as well as the experience of our neighbor, in mind. Allowing banks to operate coast-to-coast while having banks fund with more equity (and less debt) works to reduce the likelihood of bank failures and eliminate banking crises and bailouts.

Stephen Matteo Miller is a senior research fellow and member of the Program on Financial Regulation at the Mercatus Center at George Mason University.