The way banks deal with the possibility of bad loans is to set aside a pool of money, called the "loan loss reserve." As recession looms, how much banks increase reserves depends on how bad they think the downturn will be and how many debtors may have trouble repaying their loans.
In the last recession, banks were pounded because they underestimated the risks they were taking. Now a new study of 38 large banks from UBS Warburg's equity research team suggests that once again, many banks have been too optimistic and set aside too few reserves to deal with the likely prospect of soaring defaults and slow payments. That could mean a big hit to earnings starting with the fourth quarter.
The problem is that banks have an incentive to under-reserve, because additions to the reserve pool come out of net income. So far, in 2001, most banks have kept reserves at 2000 levels, say analysts at UBS Warburg, despite the worsening economy and rising rates of charge-offs, which represent the cost of loans that have actually gone sour. The hope was to avoid an unnecessary hit to net income if things got better in the second half of 2001.
That's a risky policy now that recession has become a reality and reserves are draining faster. Even before September 11, charge-offs of bad loans had risen by 40% from the second quarter of 2000 to the second quarter of 2001 (chart). Given rising unemployment, further increases are inevitable. Indeed, in the 1990-91 recession, charge-offs peaked after the downturn ended.
The UBS analysts estimated that seven large banks--with outstanding loans totaling more than $800 billion--need to boost their reserves by a combined 40% to deal with the likelihood of a sharp deterioration in their loan portfolios. Such a move would shave $6.4 billion from pretax bank earnings.
To be sure, the shortfall in reserves doesn't mean banks are financially unsound. "I don't think it's a crisis--banks are definitively better capitalized than they were at the beginning of the 1990s," says Michael A. Plodwick, an executive director of research at UBS Warburg and author of the report. Nevertheless, banks are again showing their reluctance to face up quickly to bad news. In the knowledge-based New Economy, it's often suggested that state and local government support for higher education can be an important tool for economic development. The assumption is that expanding local colleges and universities to produce more college graduates creates a better-educated local workforce, which will attract more high-skilled jobs. The leading examples of the effect are often said to be Boston and San Francisco, where the presence of strong universities appears to be a cornerstone of the local economy. Skeptics, though, have argued that college-educated workers are quite mobile, not closely tied to their alma mater at all.
A new study by economist John Bound of the University of Michigan and three co-authors finds there is a relationship between degree production and the concentration of college-educated workers in a state's population in the long run--but it's not anywhere near as large as economic-development officials might want. Based on four decades of Census data and figures on degrees awarded by state, the study reported that "states awarding relatively large numbers of BA degreeshave somewhat higher concentrations of college-educated workers." However, the link is relatively weak--increasing the output of college graduates in a state by 10%, say, has a much smaller effect on the education level of the state workforce, as grads move away to find jobs. Such mobility reduces the economic incentive for states to invest in higher education. How low can inflation go? While the latest government reports show negligible price increases, some economists fear an inflationary backlash when the U.S. economy finally turns around. Near-zero real interest rates, they argue, will spur rapid growth and fuel a period of rising prices.
But James E. Glassman, senior U.S. economist at J.P. Morgan Chase & Co. (JPM), argues the combination of weak demand, excess capacity, and falling energy prices should hold consumer inflation below 2%--perhaps even lower--through at least 2002. Core inflation, which excludes food and energy, could fall to 1.8% next year, predicts Glassman. That would be the lowest level since 1965.
The main reason for the fall in inflation, in Glassman's view, is that the U.S. economy is operating well below potential. For example, U.S. plant utilization is at only 75.5% of capacity for September, and Glassman expects it to sink to 74% by next summer. Moreover, the capital investment boom of the late '90s has left many other countries with large amounts of unneeded capacity as well, making it even harder to increase prices. And energy prices are already dropping: Gasoline, for example, has fallen to an average retail price of $1.21 a gallon, from $1.71 in May. Besides helping consumers at the pump, lower energy costs could hold down the price of goods and services. "Reduced cost pressures on nonenergy businesses could subtract as much as 0.5% from next year's [consumer price index]," says Glassman.
Even if the U.S. economy were to suddenly catch fire again, rampant inflation is not likely to reemerge. Glassman argues that the long-term potential growth rate is 3%-4%, so the economy would have to grow significantly faster than that to use up excess capacity. In fact, by his calculation, the U.S. would have to grow at 9% over the next year to create an inflationary rebound.