Binyamin Applebaum has an excellent article in the New York Times today, titled "Who Wants to Buy a Politician?", on the ineffectiveness, on the margin, of spending on political campaigns.
One reason is that buying elections is economically inefficient. Most voters, like most consumers, have defined preferences that are difficult for advertisers to shift. Chevron spent roughly $3 million during a recent campaign backing, certain City Council candidates in Richmond, Calif., where it operates a major refinery. Voters instead chose a slate of candidates who want to raise taxes. "Campaign spending has an extremely small impact on election outcomes, regardless of who does the spending," the University of Chicago economist Steven Levitt concluded in a 1994 paper. He found that spending an extra $100,000 in a House race might be expected to increase a candidate's vote total by about 0.33 percentage points. Investors appear to agree that companies can't make money by investing in political campaigns. A 2004 study found that changes in campaign-finance laws had no discernible impact on the share prices of companies that made donations.
One clarification on Levitt: Companies don't donate directly to the campaigns of candidates for the House of Representatives, the Senate, or the Presidency. That has been illegal since 1907. The previous year, a racist Democratic Senator from South Carolina, "Pitchfork Ben" Tillman, upset that corporations were giving money to Republican candidates who were critical of Jim Crow laws, introduced legislation banning corporate donations. It was signed by Teddy Roosevelt in 1907.
Applebaum covers ground that is covered in the entry, "Campaign Finance," in The Concise Encyclopedia of Economics. Here's one excerpt:
Every two years, public-interest groups and media pundits lament the fact that winning candidates typically far outspend their rivals. They infer from this that campaign spending drives electoral results. Most systematic studies, however, find no effect of marginal campaign spending on the electoral success of candidates.
How can this be so? The best explanation to date is that competent candidates are adept at both convincing contributors to give money and convincing voters to give their vote. Consequently, the finding that campaign spending and electoral success are highly correlated exaggerates the importance of money to a candidate's chances of winning. To gauge the causal relationship between campaign spending and electoral success, it is necessary to isolate the effects of increases in campaign spending that are unrelated to a candidate's direct appeal to voters. For example, wealthy candidates are able to spend more money on their campaigns for reasons that have little to do with their popularity among voters. Consider the experience of Senator Jon Corzine (D-N.J.), who defeated a weak Republican opponent to gain election to the Senate in 2000. Corzine spent sixty million dollars, mostly from his personal fortune, on his Senate campaign. Many observers pointed to this episode as an example of how a wealthy individual can buy elective office. Despite his record spending, however, Corzine's vote total ran behind that of the average House Democrat in New Jersey and behind the Democratic nominee for president, Al Gore, even though Gore did very little campaigning in strongly Democratic New Jersey. There is even some evidence that Corzine's wealth was a liability, given that many yard signs urged his Republican opponent to "make him spend it all!"
HT2 Tyler Cowen, who emphasizes a different, but also interesting, part of the article: Political spending just isn't that high.