Financial Update (July-September 1996)
When suspension of payments
proves to be beneficial
An episode in 1861 provides a good example of how suspensions of payments by banks, in times of crisis, can prove beneficial. David Sutton explains.
uspensions of paymentsperiods when banks refuse to convert their liabilities into ready cash or other assetshave sullied many a bank's reputation and are almost universally viewed as catastrophic. The mere hint of a bank's suspension still sends investors into a panic, dumping shares and withdrawing deposits. And arguably, elimination of these suspensions was one important impetus that stimulated the creation of the Federal Reserve System. But are suspensions really always a bad thing, both for banks and investors?
Authors Milton Friedman and Anna Schwartz in A Monetary History of the United States, 1867-1960, say no. While indicating that such suspensions were historically not cures for banking panics, they suggest that in 1930 just such a suspension would have halted the scramble for liquidity by bankswhich resulted in additional failures and increased the severity of the Great Depression. Indeed, some would argue that history supports this argument. During the period 1863-1914, banks often responded to similar (albeit lesser-magnitude) types of liquidity problems by simply suspending payments. And while these suspensions were technically against the law, the Controller of the Currency often allowed them as necessary.
Today, the concept of suspension of payments as a policy tool seems to be making a comeback, both subjectively and practically.
The effects of those suspensions, and the purpose served by them, are important in terms of developing theories of banking, understanding the effects of central banking and deposit insurance, and devising sensible banking policies. Today, the concept of suspension of payments as a policy tool seems to be making a comeback, both subjectively and practically. There is a growing body of literature that argues that suspensions of payments may currently be effective in some financial systems. Many emerging market governments seem to agree. They are increasingly implementing suspensions of payments in their banking systems in times of turbulence.
An episode at the start of the Civil War provides a good "controlled experiment" to examine the effects of suspensions of payments and to test Friedman and Schwartz's conjecture. In 1861, 44 percent of banks in Wisconsin and 81 percent of banks in Illinois closed. Gerald Dwyer, a visiting scholar at the Federal Reserve Bank of Atlanta, and Iftekhar Hasan, of the New Jersey Institute of Technology, assess (in Atlanta Fed Working Paper 96-3) whether a suspension of payments policy enacted in Wisconsin partly explains why fewer banks closed there than in Illinois during this episode.
The 37 percent disparity [44 percent against 81 percent] was rather large, particularly as Dwyer and Hasan assert that with few exceptions the banking systems in the two states were similar and the banks' responses to falling prices of their holdings of southern and border state bonds were also similar. The suspension of payments in Wisonsin was one of the few major differences between the two states that could explain this disparity.
The paper also outlines the banking laws of both Illinois and Wisconsin at the time and the events of 1861, discusses the responses by bankers and banking regulators to these events (including the suspension in Wisconsin), and presents a statistical analysis that estimates the effects of suspension on the number of banks closing and noteholders' losses.
Dwyer and Hasan conclude that the evidence indicates that the suspension of payments in Wisconsin reduced the number of banks that closed or failed. More banks would have closed in Illinois in any case, the authors argue, due to smaller capital bases. Nonetheless, with losses relative to capital allowed for, the suspension of payments increasedby almost 30 percentthe probability that a bank in Wisconsin remained open. Noteholders' losses on bank notes were also similarly affected, with losses of 25 percent reduced to 2.5 percent in cases of suspension of payments.