A few simple calculations will help us understand how bad things are. During normal times, there are enough jobs for at least 61 per cent of the American population. Based on 220m people of working age today, there should be at least 134m jobs. Surveys report there are fewer than 130m. Just to return to some sort of normal state, employment would have to increase by 4m. That is the "job gap". It is the sum of jobs that have been actually lost plus those that growth should have added since the beginning of 2001.
But if the US wants to get back to normal, it cannot just close the current job gap. Unlike the populations of Europe and Japan, the American population is growing. Over the past few decades, the US has been adding people at a rate of roughly 1.25 per cent a year. That means that over the next year the number of Americans of working age will increase by 2.75m. Since 61 per cent of these people need jobs, keeping the gap from growing means creating nearly 150,000 new jobs per month. Shrinking it means creating jobs at a rate faster than 150,000 per month.
When the American economy is really booming, as it was in the late 1990s, it is capable of creating something like 250,000 to 300,000 jobs a month. With steady job growth at the high end of that range it would take at least three years to get rid of the current gap. That is the disturbing arithmetic of the American labour market.
I can find two explanations for where the job gap has come from. One has to do with continued adjustment to the excesses of the late 1990s, and the other focuses on the nature of the Bush tax cuts.
In the past, workers who were laid off could expect to be recalled when business picked up. Not this time. Temporary lay-offs have been much less frequent, as industries that were hardest hit by the recession have been forced to shrink permanently. The bubble led to over-investment and over-employment, and now it is time to settle up. Sectors such as high technology, where employment grew rapidly during the boom, are the ones that shed jobs in the downturn.
Turning to the tax cuts, Bush administration officials have said that the fiscal policy shift has spurred on the recovery. The nature of the cuts leads me to be very suspicious of their claims. It is true that the government has cut taxes, thus increasing many people's income. But a number of the changes, such as the reduction in capital gains tax, were designed to raise saving and reduce the cost of capital, not increase consumer spending.
As capital becomes less expensive, we expect to see more of it. This has both short-run and long-run implications. In the long run, more capital is good for everyone. It means higher productivity, higher growth and higher incomes.
But in the short run, there is adjustment. When the cost of capital falls, companies substitute capital for workers. As they do, output rises and employment falls. This drives output per worker sharply upwards. Over the past two years labour productivity has gone up by more than 4 per cent a year, roughly twice the average of the previous five years.
Is there anything policymakers can do to close the job gap in the short run? On the monetary policy side, the Federal Reserve will almost surely keep interest rates low until the US economy has recovered the 4m jobs lost since January 2001, and added the ones needed to keep up with natural population growth. By itself that is likely to take quite a while.
As for fiscal policy, with government borrowing high, and rising, tax cuts to lower labour costs are not in the offing. Therefore, the choice is either to wait and hope, or to rescind some of the tax cuts that lowered the cost of capital. But given the lags in the system, if Mr Bush and his advisers want to change the tax code again, they had better get started. Beyond that, there is little anyone can do to eliminate the job gap by next November.
Stephen Cecchetti is Professor of Economics at Brandeis University, Massachusetts, and a Research Associate at the National Bureau of Economic Research.