Campaign Spending and Spurious Correlations: Why Self-Financed Gubernatorial Candidates Lose

Abstract

Critics are quick to accuse wealthy gubernatorial candidates of attempting to "buy" elections. But although self-financed gubernatorial candidates sometimes win, their electoral success does not necessarily imply that voters can be bought. To the contrary, I present evidence that self-financed campaign spending has a far weaker marginal effect on electoral results than externally-financed campaign spending. For every Corzine, there's a DeVos - who spent record amounts in his 2006 attempt to unseat Michigan's incumbent governor, only to lose by an embarrassingly wide margin. Money can't buy the governor's mansion.

This empirical finding presents a theoretical puzzle - why would externally financed spending trump self-finance? The solution lies in strategic incentives facing would-be campaign donors. A candidate's ability to raise funds serves as a crucial indicator of her electoral viability. When it comes to influencing voters, a candidate's ability to raise money matters far more than her ability to spend it. As such, the apparent correlation between campaign spending and votes is largely spurious.