Reducing official corruption is the main justification for limits on campaign contributions. A new analysis by the Center for Competitive Politics (CCP) casts doubt on that rationale, finding that strict contribution limits are not associated with lower levels of corruption. Looking at convictions of public officials on corruption charges during the decade from 1997 through 2006 (the most recent year for which data are available from the U.S. Justice Department's Public Integrity Section), CCP Research Director Laura Renz calculated a conviction rate per 100,000 residents for each state. She divided the states into three categories based on their campaign finance rules: those with high limits (above $5,000) or no limits on contributions to candidates, those with limits between $1,001 and $5,000, and those with limits of $1,000 or less. Renz found no obvious relationship between conviction rates and contribution limits:
Overall, the distribution of corruption among all 50 states is random, at least when compared to contribution limits. All three states with the lowest corruption rates have no or very high limits on contributions to candidates for state legislature, while 2 of the 3 states with the highest corruption rates also have no or very high limits.
Of states with low limits on contributions, 5 make the "Top 10" for high corruption rates while 4 make the "Bottom 10" with very low corruption rates. Half of the states with moderate contribution limits are in the "Medium Corruption" group.
According to Renz, "This analysis demonstrates that there is no credible reason to believe that contribution limits on what citizens can give to candidates for public office [have] any effect on restraining corruption by elected officials or otherwise cleaning up any 'culture of corruption' that may exist." That conclusion seems too strong to me. It's fair to say that the analysis provides little support for the theory that contribution limits reduce corruption, but it's a pretty simple analysis that does not take into account the timing of the contribution limits or the possibility that some states imposed them in response to cases of corruption.
If that's true, the coexistence of strict limits and a high corruption rate would not necessarily show the limits are ineffective; you'd want to see what happened to corruption after the limits were imposed. If the tightening of restrictions tends to be followed by a decline in corruption across different states, that would be prima facie evidence that the limits work (though it would not be conclusive, since other factors associated with the adoption of new limits, such as publicity and increased scrutiny, might be driving the declines in corruption). And if no such pattern appears, that would count as evidence against the effectiveness of contribution limits (though again, not conclusive, since new restrictions might be accompanied by intensified law enforcement efforts, leading to more convictions).
More fundamentally, I'm not sure convictions on public corruption charges are the right measure by which to judge the effectiveness of contribution limits. Corruption might involve dispensing favors to campaign contributors without an explicit quid pro quo, in which case it probably would not lead to criminal charges, let alone a conviction. Then again, as Renz notes, attempts to curtail this sort of corruption by restricting contributions might increase corruption convictions, making bribes more tempting by making it harder to raise money legally.
This sort of analysis is a bit risky, since CCP does not want to concede that limiting political speech by limiting campaign contributions would be constitutional if only it worked as advertised. At the same time, it's perfectly legitimate to demand evidence that the official rationale for carving out this exception to the First Amendment holds water in the real world.