Matt Yglesias writes
David Henderson has a column in Forbes getting some pick up in what passes for conservative movement policy circles alleging that Christina Romer’s co-authored paper on “The Macroeconomic Effects of Tax Changes: Estimates Based on a New Measure of Fiscal Shocks” debunks Barack Obama’s approach to fiscal stimulus.
I’m going to go with the other interpretation: Henderson is wrong.
He writes “The Romers carefully sift through all federal tax cuts and tax increases from 1947 to 2005 to figure out, based on the discussion at the time, whether the changes in tax policy were motivated by a desire to offset the business cycle or by other goals.” Their finding was that tax cuts implemented between 1947 and 2005 that were intended to serve as economic stimulus didn’t work.
I'm not sure I remember correctly and I might be making a fool of myself, but I think that Henderson's error is elementary. The Romers' result shows the changes after a tax cut. The reasonable interpretation of their results for anticyclical tax cuts is that they are measuring the sum of the effect of the tax cut and the recession.
To measure directly if a tax cut works you have to compare events with and without the tax cut while everything else is equal. This is impossible. They distinguish endogenous (countercyclical) and exogenous tax cuts so that they can measure the effect of tax cuts by the changes following exogenous tax cuts. If taxes are cut in response to a political event (Republicans win election) not an economic event, they guess that the economic effects which follow are caused by the tax cut. If there is a recession and a tax cut the following events are the sum of a recession and the effects of a tax cut.