Krugman has conveniently stopped worrying about productivity and switched to worrying about jobs. The author of The Age of Diminished Expectations has to have something to worry about. Krugman recently wrote The Return of Depression Economics, and has now turned to advocating Depression policies. In a recent New York Times column, he writes that "Franklin Roosevelt's Work Progress Administration put the unemployed to work doing all kinds of useful things; why not do something similar now?"
Well, unemployment rate was 20.3 percent in 1935 when the WPA began, and it was still 19.1 percent three years later. Today, the unemployment rate is only 5.8 percent, much lower than the 6.9 percent unemployment rate in 1993, when Krugman was more concerned about productivity.
To propose reviving Depression-era programs when unemployment is unusually low and productivity unusually high seems eccentric. So does the arithmetic by which Krugman defended his nostalgia for the WPA. He took a 10-year estimate of revenues supposedly lost by the Bush tax plan and divided it by the 1.4 million extra jobs the administration predicts would be added over just two years. Dividing the two-year job gain by the 10-year tax cut allowed Krugman to pretend the new jobs would cost $500,000 each.
Replying on his web page to those who criticized this new math, such as Don Lufkin, Krugman said, "No serious economist thinks that a tax cut or spending increase will have any effect on employment more than a couple of years from now." That remark defines seriousness as 1962 demand-side economics, thus ignoring work incentives. Lower marginal tax rates encourage more people to join and remain in the labor force (particularly spouses and the otherwise prematurely retired), which can leave employment permanently higher.
For Krugman to treat spending increases as equivalent to lower tax rates also ignores the well-documented fact that government purchases hurt private employment by drawing real resources away from productive private uses. In any case, Krugman's device of dividing a two-year job gain by a 10-year revenue loss was deliberate fraud, pure and simple.
I do agree, however, that the administration's constant repetition of the estimated 1.4 million new jobs is nearly as annoying as the opposition's constant repetition of the estimated $726 billon revenue loss. Both figures are just estimates, and both miss all the important points. There are many reasons for easing the tax rates on dividends, for example, that are equally valid regardless of any impact on jobs or productivity. For one thing, punitive taxes on dividends are widely avoided in ways that do damage to both the economy and corporate governance (e.g., by fostering unbearable corporate debt).
Benefits to the economy from faster growth should definitely not be measured only in terms of a promised 1 percent addition to the 134 million jobs we already have. Faster economic growth raises incomes and wealth for those who have jobs too, not just for the 8 million between jobs.
It makes little sense for the administration to promote faster growth merely as a method of bringing unemployment down a bit faster. But it makes no sense at all for Paul Krugman to judge a whole package of varied tax proposals merely in terms of their short-term impact on employment. If we just wanted to "create jobs" and didn't give a damn about productivity and prosperity, then we could all grab a rake and work for Paul Krugman's new WPA.