You may remember my exchange with FiveThirtyEight.com’s Nate Silver a few months back, in which we debated whether politics had a role to play in stimulus allocation. Based on recipients’ reports from Recovery.gov, my analysis showed that districts’ party affiliations mattered in where the money was spent. Silver suggested that I should control for the money going through state capitals, since that might be the reason why I found such a connection. I did, and I still found the same result. I couldn’t tell how much politics mattered compared to other factors, just that it mattered.
A new paper by Jason Reifler of Loyola University of Chicago and Jeffrey Lazarus of Georgia State University sheds some more light on this issue. They find that the partisan bias does exist. More interestingly, they explain how and why it happens:
Money from the recent economic stimulus package is disproportionately going to Democratic House districts, leading to a debate over the cause of this trend.… Thus while partisanship is influencing the distribution of stimulus funds, the key locus of this partisanship was in the writing of a bill which advances Democratic policy goals such as clean energy, health care, education, and research, not in any post-passage activity.
In other words, the partisanship is more in the design of the stimulus bill (i.e., which agencies the money is going to) than in the distribution of the funds to try to favor the reelection of Democratic members of Congress. Thanks to this paper, I know more today than I did yesterday.
The whole paper is here.