February 8, 2008 12:38

New column: Do presidents matter to the economy?

I have a column in the new Time with Hillary C. and Barry O. on the cover and online here. It begins:

For decades, scholars have been churning out studies on the impact the economy has on presidential elections. The not-very-surprising message of most of them: economic trouble is bad news for the party that occupies the White House.

The seemingly far more important question of what impact Presidents have on the economy has been studied less, with far less conclusive results. See, it's not just journalists who obsess over campaign horse races and neglect issues. It's economists and political scientists too.

The reason is that separating cause and effect on Presidents and the economy is hard. Yes, over the past half-century, Democratic administrations have seen faster economic growth--and better stock-market performance--than Republican ones. But the sample size is so small that you really can't rule out luck.

A perhaps more solid result, because it jibes with the parties' priorities, is Princeton political scientist Larry Bartels' finding that income inequality increases more under Republicans than under Democrats. But a case can also be made that it doesn't matter who's in charge. A study of political leadership and economic growth around the world by economists Ben Jones of Northwestern University and Ben Olken of Harvard found that changes at the top made a big difference--but only in dictatorships. Read more.

The Larry Bartels study can be downloaded here and the Jones-Olken study here. The best-known studies of the impact of the economy on presidential elections are those of Yale's Ray Fair, which you can read all about here. Fair even has a calculator that lets you type in your guesses about inflation and GDP growth leading up to the November election and generate your very own election forecast.

As for my contention that "over the past half-century, Democratic administrations have seen faster economic growth--and better stock-market performance--than Republican ones," it was actually supposed to be changed to "since World War II," but that got lost in the shuffle. I'm pretty sure it's true over both time periods, though. The evidence for the former comes from the book Political Cycles and the Macroeconomy by Alberto Alesina, Nouriel Roubini and Gerald Cohen. There's also a review of the matter in this paper by Allan Drazen. On stock market performance, there's "The Presidential Puzzle: Political Cycles and the Stock Market," by Pedro Santa-Clara and Rossen Valkanov. Another recent paper by Erik Snowberg, Justin Wolfers, and Eric Zitzewitz finds that the stock market's immediate reaction to Republican election victories is strongly positive, but they don't know if that's because investors are deluded or because they know something about future Republican vs. Democratic policies that no economist can quantify.

On economic growth, I'm willing to buy that, up through the 1970s, there might have been a clear cause-and-effect relationship, given that Republican administrations were more likely to push for balanced budgets and other austerity measures that would slow growth in the short term. But the GOP discovered the joys of deficit spending in the early 1980s, and I really think the dramatic economic outperformance by the one Democratic administration since then was more a matter of lucky timing than anything else. I do think the Clintonistas did a good job of adapting their policies to the economic environment, which hasn't been so much the case for Bush 2. But that environment wasn't Bill Clinton's creation.