The Facts About Incumbency: A Response to Professor Hayward

Professor Hayward replied to my earlier blog post, and was somewhat dismissive of the Brennan Center research ...

May 21, 2009

In a post on The Hill Blog, The Incumbency Problem Has Everything to do with Money, I wrote that the availability of low contribution limits and public financing help challengers in elections against incumbents.  Professor Hayward replied here, and was somewhat dismissive of the Brennan Center research inspired by Randall v. Sorrell that proves these points.

After paraphrasing Prof. Hayward's statements at our recent conference for the blog, I checked the transcript.  Here's what Prof. Hayward said, "So ask yourselves, and this is my closing thought: as passionate reformers, how much of what you dislike about political funding is a problem of incumbency rather than a problem of money?"  Given this, I do not think it was misleading to write: "panelist Professor Allison Hayward, a skeptic of campaign finance reform, asked whether reformers should really focus more on incumbency than they do on limits on money in politics."

We certainly welcome her thoughts about the problem of incumbency.  Here at the Brennan Center, we spend a lot of time contemplating how this particular feature of the 21st Century American democracy can be addressed.  For example, we researched the best ways to conduct redistricting. But an undeniable factor in how incumbents from the New York City Mayor to the President get to keep their positions is the way in which we finance elections.  In short, money matters.

Of course, there was a lot going on in the Randall case besides concern over electoral competition.  As noted in our report, "In Randall the Supreme Court considered a law with many facets.  The Vermont law included both expenditure and contribution limits, and included limits on parties and restrictions on volunteers.  In its final opinion, the Court rejected all elements of the law." 

However, we do take issue with her blithe dismissal of our findings, if for no other reason than that they actually have constitutional significance.  One of the reasons that the Brennan Center partnered with Dr. Thomas Stratmann, an economist from Prof. Hayward's own George Mason University, to study the effect of contribution limits on incumbency is because understanding the relationships between contributions and incumbency is fundamentally a mathematical problem.  While people may debate many aspects of campaign finance, this is actually one of the areas in which the data produce real answers.

In the study, we looked at single-member state-assembly elections from 1980 to 2006 and compared them to the contribution limits in effect at the time.  Vermont's limits were the lowest in the country.  To consider only Vermont's limits would have created far too small a sample size for robust results. 

Instead, we split states into four cohorts: those with limits of (1) $500 and less (2) $501-$1,000, (3) $1,001-$2,000 and (4) over $2,000.  This gave us the ability to compare, for the first time, whether there was a difference between, say, a limit of $1,000 and a limit of $500.  (Incidentally, as to the critique of our analysis vis-à-vis gerrymandering, to the extent that some states have stricter gerrymandering rules than others, Dr. Stratmann controlled for this with state dummy variables.  He also controlled for other factors as well to rule out the problem of endogeneity.)   

All of the data pointed in the same direction.  Contribution limits lead to more competitive elections by every single measure, including unseating incumbents, and contribution limits lowered the fundraising spreads between incumbents and challengers.  Furthermore, the lower the limits, the bigger these pro-competitive effects were.  Public financing in Maine and Minnesota had the same pro-competitive effects.

We did not know that this would be the case when we started the study.  Previous studies had been split in their predictions.  It could have been, for example, that $1,000 limits were pro-competitive, but that at the $500 limit level, systems became wildly uncompetitive.  But there was no spike in uncompetitiveness in states with low limits.  Indeed, we found just the opposite.  

Randall stated that one of the reasons Vermont's $200 limit was struck down just six years after upholding Missouri's $275 limit was a concern that low limits bolstered incumbency.  Thus, given our data, I do not think it an overstatement to say, in this particular factual matter, it turns out the Court guessed wrongly. 

Moreover, the Brennan Center report's conclusions directly answer Prof. Hayward's rather plaintive question: incumbency is inextricably tied to campaign finance rules, and the notion that we can fix the problem by tackling incumbency alone is both wrong-headed and empirically uniformed.