Consumers were piling up debt in the years leading up to the recent US financial crisis. Total debt of all households in 2007 surged to 130 percent of their disposable income, compared with only 92 percent in 2000, according to data from the Federal Reserve Board. When the crisis hit, banks tightened lending and consumers could no longer get credit as easily as they did before. The adjustment was sharp and painful for many people.
The pressure on households to repay debts quickly and save more right after a credit shock can lead to a large fall in consumption. In fact, the resulting contraction in output can be severe in the short run, because households initially find themselves far below their desired level of precautionary savings, according to a recent paper, "Credit Crises, Precautionary Savings and the Liquidity Trap," by Chicago Booth professor Veronica Guerrieri and Massachusetts Institute of Technology professor Guido Lorenzoni.
When credit suddenly tightens, highly indebted households can no longer roll over their loans. Those with very little savings worry it will be difficult to get a loan should they need one in the future. Add the risk of job loss, which increases with the threat of a recession, and saving and cutting down on consumption begin to feel more urgent than they had before.
As households save more and spend less, the central bank may step in to boost consumption. But the recession that follows can be so deep that the central bank may not be able to bring interest rates down to the level where it can sufficiently stimulate spending, because the nominal interest rate cannot fall below zero. The fact that a credit crisis can lead the economy into a "liquidity trap" seems consistent with historical experience. "Liquidity traps typically follow large credit crunches, just as we have seen in Japan in the 1990s and in the United States during the Great Depression," says Guerrieri.
Overshooting Interest Rates and Output
To find out how households respond to a credit crunch, Guerrieri and Lorenzoni analyze what happens when the borrowing limit of households is unexpectedly reduced to a level that would bring down the ratio of US household debt to GDP by about 10 percentage points from its peak right before the crisis.
When a credit shock hits, households with debts exceeding this new borrowing limit would have to drastically reduce their debts, and those close to the limit would have to save more as a precaution. "Losing a job prior to the crisis may not have been so bad, because people could go to a bank for a loan or use a credit card, and then pay back the loan when they found a job again," says Guerrieri. But with poor job prospects and without the help of credit markets to smooth consumption, households know they need savings to get them through the rough times.
Real interest rates drop dramatically in the short run because households affected by the tighter borrowing limit are eager to put their money into safe assets, such as savings accounts or Treasury bonds. They are willing to accept low interest rates. Indeed, Guerrieri and Lorenzoni find that the drop in interest rates "overshoots" right after a credit crisis. "Households were not ready when the shock hit and so they had to deleverage faster initially," says Guerrieri.
As households accumulate more safe assets, they start to move closer to their desired level of precautionary savings. The borrowing constraint becomes less worrisome and interest rates start to move up as households gradually save less than they did right after the shock. Interest rates eventually settle at a lower rate to reflect tighter credit conditions.
That real interest rates would fall to such a low level gives a sense of just how much households would be willing to reduce consumption in order to pay back their debts and increase savings.
In theory, households would like to work more in response to a credit crunch in order to boost their earnings and thus their ability to get out of debt. If goods and labor markets worked perfectly, this increase in labor supply would dampen the effects of a credit crunch on aggregate activity. However, in a recession the number of jobs created is driven more by the firms’ ability to sell goods and services than by workers’ willingness to work. Therefore, the labor supply effect would likely be weak.
Following a similar pattern as interest rates, output drops on impact as a result of households cutting back significantly on consumption. In other words, output contracts strongly in the short term before gradually settling at a level lower than its pre-crisis level.
Government Policy and the Liquidity Trap
In normal times, the central bank intervenes in the economy by using its nominal policy rate to influence the real interest rates that matter for households’ consumption and saving decisions. When the economy is weak, the central bank typically cuts nominal interest rates to stimulate consumption. The authors find, however, that the response of households to a credit crisis can be so strong that the nominal interest rate required to pull the economy out of a recession falls below zero. Because the central bank cannot reduce rates below zero, interest rates effectively remain higher than they should be, exacerbating the recession. The inability to stimulate the economy by lowering the interest rate is known as the liquidity trap.
To ease the effects of a severe downturn, the government may consider increasing the supply of bonds. This is beneficial for two reasons. First, the supply of liquid assets for people to put their savings into goes up, thus reducing the downward pressure on real interest rates. Second, the revenue raised from issuing more bonds can be used to increase government transfers or reduce taxes in the short run to help people cope with the recession.
Indeed, the authors find that increasing the supply of government bonds helps dampen the overshooting effect on real interest rates and output. They also find that increasing transfers by temporarily raising unemployment benefits is more effective in alleviating the effects of the recession than reducing lump sum taxes. This type of policy allows the government to target the most highly indebted and creditconstrained households, who are more likely to cut their consumption in a credit crisis.
Flight to Liquidity
Durable goods, items such as cars and household appliances that can be used for many years, are an important part of consumption. Because purchases of durable goods are often financed with credit, they are particularly sensitive to changes in credit conditions. Not surprisingly, durable goods consumption experienced the most noticeable decline of all components of consumption in the recent US recession. A large fraction of household debt was used to purchase durables, in particular housing. Right before the crisis, consumer credit without mortgages accounted for only 18 percent of GDP in 2006, compared with 96 percent with mortgages.
Households that borrowed heavily to buy houses included not just cash-strapped first-time home buyers, but also families who desired bigger homes and were able to finance these purchases quite easily through bank loans. They, too, were affected by tighter borrowing constraints. In analyzing both durable and nondurable goods in their research, the authors obtained a much broader picture of how households respond to a credit crisis.
Consumers also face a more interesting portfolio problem with the introduction of durable goods. Instead of safe assets such as bonds or savings accounts, households can choose to allocate their precautionary savings towards investment in durables. In this case, a shock that increases the precautionary demand for safe assets could lead to an increase in demand for these long-lived but less liquid assets.
The study shows, however, that in the event of a credit crisis, households can actually favor bonds over durable goods, which the authors say is a form of "flight to liquidity." This exacerbates the economic downturn, as households not only spend less on nondurable goods but also shift their portfolio of assets away from durable goods.
Veronica Guerrieri is associate professor of economics and Charles M. Harper Faculty Fellow at the University of Chicago Booth School of Business.