He writes:
We do not need a second stimulus plan: As the US unemployment rate has risen to 9.5 per cent from 8.1 per cent since the $787bn fiscal stimulus package was enacted in February, many Democrats have become very nervous. They say that another large stimulus may be needed to keep unemployment from rising.... Another stimulus would be a grave mistake. The first one was justified by extraordinary circumstances. But it must be given time to work. People should not allow their impatience to lead to the adoption of policies that will not only fail to reduce unemployment this year, but could stoke inflation in the not-too-distant future....
The forecast also showed the unemployment rate peaking at 8 per cent with the stimulus and 9 per cent without. Obviously this was wrong. Yet it would be incorrect to conclude that the stimulus was doomed to failure, as many Republicans and conservative economists argued.... [T]he Romer-Bernstein document presents reasonable estimates of how quickly different forms of spending would raise gross domestic product. Tax cuts and government transfers are slow to have an effect and have a low multiplier, raising GDP less than $1 for every $1 increase in the deficit even when fully effective after two years. By contrast, government purchases stimulate growth much more quickly and have a higher multiplier, raising GDP by $1.57 for every $1 spent. Unfortunately, the low-impact spending has been the fastest to come online while the high-impact spending is dribbling out very slowly.
In a recent report to the International Monetary Fund, Doug Elmendorf, Congressional Budget Office director, looked at the rates of spending for different components of the stimulus package. He estimates that by the end of fiscal year 2009, which falls on September 30, 32 per cent of the income transfers for things such as food stamps and extended unemployment benefits will have been spent and 31 per cent of the tax cuts will have been disbursed. By the end of fiscal year 2010 virtually all of the money allocated to these programmes will have been spent.
However, just 11 per cent of the discretionary spending on highways, mass transit, energy efficiency and other programmes involving direct government purchases will have been spent by the end of this fiscal year. Even by the end of 2010 less than half the funds will have been disbursed and by the end of 2011 more than a quarter of the money will be unspent. Consequently, it is hardly surprising that five months after the stimulus bill passed it has not yet affected the unemployment rate....
What all this means is that it is foolish to think that any sort of stimulus that is enacted now will have an impact on the economy any time soon. We just have to wait for the medicine we have already taken to work. Pushing ahead with another stimulus will only make it harder to tighten fiscal policy down the road to keep inflation in check.
It's a balance of risks. Any second stimulus package passed this fall would have little impact on the economy until late 2010, that is true. But come late 2010 we might really need more demand to curb unemployment. On balance the inflationary risks of having an extra stimulus hit the economy in late 2010 if it is not needed are outweighed by the deflationary risks of not having an extra stimulus hit the economy in late 2010 if it is needed.
It is like driving a car with its windshield painted black by looking in the rear view mirror.
