Andreas Wesemann asks what government-backed deposit insurance is for
DURING AMERICAS mini banking crisis in March, when Silicon Valley Bank (SVB) and Signature Bank collapsed, the Federal Deposit Insurance Corporation (FDIC) used an emergency systemic-risk authority to provide 100% protection for all depositors of both banks. Since then, regulators in America and elsewhere have pondered the role of deposit insurance in dealing with the consequences of banking meltdowns. As the world contemplates the latest scramble to save a tottering lenderSan Francisco-based First Republic Bank, most of which is to be taken over by JPMorgan Chasecentral banks on both sides of the Atlantic are considering calls to make deposit insurance much more generous to avoid a more general crisis. It is noteworthy, however, that insofar as there have been runs on banks in America, these have been restricted to very specific institutionsand for good reason. SVB, for instance, was technically insolvent because of the (unrealised) losses on its government-bond portfolio. There have been large deposit movements between financial institutions, as you would expect when interest rates rise: higher rates fuel competition for deposits. But depositors are generally good at discriminating between sick and healthy institutions. This tells a more general story. Industry-wide bank runs are extremely rare. The events in America in 1933 that led to the imposition of a national bank holiday by President Franklin Roosevelt are the only episode that comes close to a general bank run in the countrys historyand one that was caused not by confused depositors blindly withdrawing funds but by depositors fears that their state would be the next to declare a bank holiday. Nor was there a general run on banks during the global financial crisis of 2007-09. To the extent there was a run then, it was in shadow banking, for instance certain parts of the asset-backed commercial-paper market. This suggests that the problem of contagionwhere healthy banks are brought to their knees alongside bad ones as jittery depositors pull their money from all institutions indiscriminatelyis much less of a risk than commonly believed. Which invites a question: if bank runs are idiosyncratic, institution-specific events, what is government-backed deposit insurance really for? The argument in favour is that such insurance is required to protect retail depositors when there is a banking crisis, and that without it they would lack confidence in the financial system. This was the rationale for its introduction in America in 1934 and in Europe during the 1970s. Since the early 1970s the number of countries with deposit insurance has increased more than tenfold, from 12 to 146. Meanwhile, the number of banking crises worldwide has risen dramatically: the proportion of countries experiencing a crisis rose from less than 1% in 1950-73 to more than 17% in 1974-2008, according to This Time is Different, a book by Kenneth Rogoff and Carmen Reinhart. While other factors have, to be sure, contributed to this increase in financial instabilitythe collapse of the Bretton Woods fixed-exchange-rate system, and rapid credit growth in the wake of financial deregulationthere is also extensive evidence that deposit insurance can be destabilising. As an IMF working paper published in 2006 put it, deposit insurance tends to be detrimental to bank stability, especially when coverage offered to depositors is extensive, when the scheme is funded and when it is run by the government rather than by the private sector. In periods when insured banks operated alongside uninsured banks (for instance, in Kansas and Oklahoma in the early 20th century) failure rates of uninsured banks were far lower than for their insured counterparts. If the ultimate objective is to reduce the frequency of banking crises rather than just dealing with their consequences, the case against deposit insurance is that dismantling it would force banks to be stronger. Depositors would have to become even more discerning about where to place their money. Riskier institutions would no longer be able to attract depositors from safer banks at no extra cost. As banks could no longer rely on an expectation that in times of crisis all deposits would become insured (as happened in March, and in many countries during the global financial crisis 15 years ago), risk awareness and general behaviour would surely improve. Banking systems can and do operate without deposit insurance. Indeed, two members of the OECD, New Zealand and Israel, offer no explicit guarantee. In New Zealand the authorities concluded that, as Grant Spencer, deputy governor of its central bank from 2007 to 2017, put it, deposit-insurance schemes...blunt the incentives for investors and banks to properly manage risks, and may even increase the chance of bank failure. However, the current government has proposed legislation to introduce insurance up to NZ$100,000 ($62,000) per depositor per bank. That movesaid to be driven more by a desire to protect small depositors than by wider prudential concernsis wrongheaded. There is no sign that New Zealands economic performance has been affected by the absence of deposit insurance, and other than during 2007-09 it has had only two banking crises since the mid-19th century. Introducing deposit insurance is clearly easier than getting rid of it. But the latter can be achieved in three steps. First, the government announces that deposit insurance will be abolished after, say, a two-year transition period. This gives everyone enough time to prepare. Second, all deposits, not just those that are currently insured, are made preferential debts in an insolvency. This ensures that depositors will always rank ahead of all other bank creditors in a crisis. Third, a government savings bank starts offering a full range of current and savings accounts. Anyone who wants their deposits to be insured can then bank with this institutionin return for lower rates of interest. A model for this is Britains National Savings & Investments, the highly popular savings bank that once belonged to the post office. Faced with this arrangement, banks would have to persuade their customers that they, too, are super-safe: higher levels of capital and liquidity would result. Otherwise their customers would vote with their feet (and money). Depositors would be paid higher rates of interest as the implicit insurance subsidy to banks was eliminated. Banks would find that insuring each other, for instance via mutual guarantee schemes (MGSs), becomes necessary to hold on to their customers. In that eventand the precedent for well-run schemes of this kind, for instance in various American states in the 19th century, is encouragingprivately funded insurance would remain available for those who wanted it. MGSs would have to regulate themselves to weed out bad apples, circumscribing the role of the state in the banking system. This, too, could be a welcome consequence of abolishing government-backed deposit insurance. Andreas Wesemann is a partner at Ashcombe Advisers, a corporate-finance firm.