If banking crises affected only banks, they wouldn’t be as serious a concern. But banking crises are almost always associated with recessions, and severe banking crises are associated with the worst economic slumps. Furthermore, history shows that recessions caused in part by banking crises inflict sustained economic damage, with economies taking years to recover.

A severe banking crisis is one in which a large fraction of the banking system either fails outright (that is, goes bankrupt) or suffers a major loss of confidence and must be bailed out by the government. Such crises almost invariably lead to deep recessions, which are usually followed by slow recoveries.

Figure 17-2 illustrates this phenomenon by tracking unemployment in the aftermath of three banking crises widely separated in space and time: the Panic of 1893, the Swedish banking crisis of 1991, and the American financial crisis of 2008 that spawned the Great Recession. In the figure, t represents the year of the crisis: 1893 for the United States, 1991 for Sweden. As the figure shows, these three crises, spanning over 100 years and continents apart, produced similarly devastating results: unemployment shot up and came down only slowly and erratically so that the number of jobless remained high by pre-crisis standards for many years. A third line shows U.S. unemployment in the aftermath of the 2008 crisis.

These historical examples are typical. Figure 17-3, taken from a widely cited study by the economists Carmen Reinhart and Kenneth Rogoff, compares employment performance in the wake of a number of severe banking crises. The bars on the left show the rise in the unemployment rate during and following the crisis; the bars on the right show the time it took before unemployment began to fall. The numbers are shocking: on average, severe banking crises have been followed by a 7-percentage-point rise in the unemployment rate, and in many cases it has taken four years or more before the unemployment rate even begins to fall, let alone returns to pre-crisis levels.

It’s not difficult to see why banking crises normally lead to recessions. There are three main reasons: a credit crunch arising from reduced availability of credit, financial distress caused by a debt overhang, and the loss of monetary policy effectiveness.

Recall from Chapter 14 how the Fed normally responds to a recession: it engages in open-market operations, purchasing short-term government debt from banks. This leaves banks with excess reserves, which they lend out, leading to a fall in interest rates and causing an economic expansion through increased consumer and investment spending.

Under normal conditions, this policy response is highly effective. In the aftermath of a banking crisis, though, the whole process tends to break down. Banks, fearing runs by depositors or a loss of confidence by their creditors, tend to hold on to excess reserves rather than lend them out. Meanwhile, businesses and consumers, finding themselves in financial difficulty due to the plunge in asset prices, may be unwilling to borrow even if interest rates fall. As a result, even very low interest rates may not be enough to push the economy back to full employment.

In the previous chapter we described the problem of the economy’s falling into a liquidity trap, when even pushing short-term interest rates to zero isn’t enough. In fact, all the historical episodes in which the zero bound on interest rates became an important constraint on policy—the 1930s, Japan in the 1990s, and a number of countries after 2008—have occurred after a major banking crisis.

The inability of the usual tools of monetary policy to offset the macroeconomic devastation caused by banking crises is the major reason such crises produce deep, prolonged slumps. The obvious solution is to look for other policy tools. In fact, governments do typically take a variety of special steps when banks are in crisis.

Before the Great Depression, policy makers often adopted a laissez-faire attitude toward banking crises, allowing banks to fail in the belief that market forces should be allowed to work. Since the catastrophe of the 1930s, though, almost all policy makers have believed that it’s necessary to take steps to contain the damage from bank failures. In general, central banks and governments take three main kinds of action in an effort to limit the fallout from banking crises:

1. Lender of Last Resort An institution, usually a country’s central bank, that provides funds to financial institutions when they are unable to borrow from the private credit markets is a lender of last resort. In particular, the central bank can provide cash to a bank that is facing a run by depositors but is fundamentally solvent, making it unnecessary for the bank to engage in fire sales of its assets to raise cash. This acts as a lifeline, working to prevent a loss of confidence in the bank’s solvency from turning into a self-fulfilling prophecy.

Did the Federal Reserve act as a lender of last resort in the 2008 financial crisis? Very much so. Figure 17-4 shows borrowing by banks from the Fed between 2005 and 2010: commercial banks borrowed negligible amounts from the central bank before the crisis, but their borrowing rose to $700 billion in the months following Lehman’s failure. To get a sense of how large this borrowing was, note that total bank reserves before the crisis were less than $50 billion—so these loans were 14 times the banks’ initial reserves.

2. Government Guarantees There are limits, though, to how much a lender of last resort can accomplish: it can’t restore confidence in a bank if there is good reason to believe the bank is fundamentally insolvent. If the public believes that the bank’s assets aren’t worth enough to cover its debts even if it doesn’t have to sell these assets on short notice, a lender of last resort isn’t going to help much. And in major banking crises there are often good reasons to believe that many banks are truly bankrupt.

As we have already learned, in such cases governments often step in to guarantee banks’ liabilities. In 2007, a bank run hit the British bank, Northern Rock, ceasing only when the British government stepped in and guaranteed all deposits at the bank, regardless of size. Ireland’s government eventually stepped in to guarantee repayment of not just deposits at all of the nation’s banks, but all bank debts. Sweden did the same thing after its 1991 banking crisis.

When governments take on banks’ risk, they often demand a quid pro quo; namely, they often take ownership of the banks they are rescuing—an action called nationalization. Northern Rock was nationalized in 2008. Sweden nationalized a significant part of its banking system in 1992. In the United States, the Federal Deposit Insurance Corporation routinely seizes banks that are no longer solvent; it seized 140 banks in 2009. Ireland, however, chose not to seize any of the banks whose debts were guaranteed by taxpayers.

These government takeovers are almost always temporary. In general, modern governments want to save banks, not run them. So governments eventually reprivatize nationalized banks, selling them to private buyers, as soon as they believe they can.

3. Provider of Direct Financing As we learned in Chapter 14, during the depths of the 2008 financial crisis the Federal Reserve expanded its operations beyond the usual measures of open-market operations and lending to depository banks. It also began lending to shadow banks and buying commercial paper—short-term bonds issued by private companies—as well as buying the debt of Fannie Mae and Freddie Mac, the government-sponsored home mortgage agencies. In this way, the Fed provided credit to keep the economy afloat when private credit markets had dried up.