The debate over state polarization and campaign finance laws continues

One of the fundamental arguments in the “Political Realism” debate is whether or not strong political parties could make government work better. One way to assess party strength is to look at how much money parties can raise and spend.

In this vein, political scientists Ray LaRaja and Brian Schaffner have claimed that removing limits on party funding activity would make politics less polarized. I’ve been skeptical of this claim. In fact, in a short analysis, I found that the opposite is more likely the case—that states with limits on party fundraising appear to be less polarized, though I cautioned against inferring too much from this pattern.

LaRaja and Schaffner have now responded and previewed their forthcoming book, Campaign Finance and Political Polarization: When Purists Prevail, which will be out this fall from the University of Michigan Press. So, a response to their response is now in order.

I’ll start by granting a point of agreement: LaRaja and Schaffner note that I didn’t re-produce their analysis. I didn’t do this because, based on what they’ve written, it’s not clear exactly which states they consider to be “Parties Unlimited” and “Parties Limited” states. So, until they make their list public, it will be impossible to conduct a precise replication of their analysis.

The good news is they’ve promised to make their data public in the future. As they write in their recent post, “we will be posting all the data necessary to replicate (and challenge) our results upon publication of our book this fall, and we look forward to seeing what others find when they dig into the data.” They also note in their analysis that “11 states changed their laws on party limits during the period of our study (1993-2012).” Assembling this list, they note, was “possibly the most painstaking work we did on this book.” For now, their list of changes remains a well-kept secret, though the changes appear to be driving their analysis. So it will be good when all the relevant data and categorization choices are clear and on the table.

A lot depends on which states fall into which categories. But, there is a more fundamental question: does it make sense to dichotomize states into “Parties Unlimited” and “Parties Limited” states? States with limits vary considerably. Some states limit the money into parties, but allow unlimited flows to candidates; some states allow unlimited money into parties, but limit money from parties to candidates. Some limits are high, some are low. Some have exceptions for party-building activities. Rules vary between primary and general elections, as well.

Consider California. There are limits on how much parties can raise from individuals, but those limits are quite high (they are now at $35,200), and also only cover the party accounts that go to state candidates (so, for example, ballot measures are exempt or general party activities are exempted). California also has no limits on how much parties can transfer to candidates. So should California be a “limits” or “no limits” state? California also has the most polarized legislature, as measured by the distance between party medians. Depending on how you choose to classify states, you can get very different results, especially when you are only working with 20 states (LaRaja and Schaffner limit their claims to the 20 states with the most professionalized legislatures, as per the Squire Index).

LaRaja and Schaffner’s response presents a time series regression model to “calculate the predicted level of polarization over time in a state that limited party fundraising … and spending to on where those limits were removed.” But if states that removed limits became less polarized following the removal of those limits, why not tell us what those states were, and report the actual polarization trends in the states? Put another way: Why rely on model predictions when there are real world data? Grounding this debate in the trajectories of actual states would lend some realism to the claims. Then we could debate examples.

For example, as Thomas Mann and E.J. Dionne note in a recent Brookings paper, two of the states with no limits are Texas and North Carolina. As Mann and Dionne write, “The behavior of their legislatures in recent years cannot, on any plausible definition, be described as ‘moderate.’” However, neither Texas nor North Carolina shows up as excessively polarized when polarization is merely a measure of voting patterns. Moreover, if parties are so pragmatic, why did the North Carolina Republican Party (which could raise and spend unlimited sums of money) fail to stop a takeover by multi-millionaire right-wing extremist Art Pope?

This takes us to questions of how party leaders actually behave. LaRaja and Shaffner show evidence in their response that parties give more money to moderate incumbents than to extreme incumbents. This should not be surprising. Presumably, moderate incumbents are more likely to be in competitive races, since moderates are more likely to represent competitive districts.

The more relevant question is what types of candidates parties recruit. Thankfully, we have answers to this courtesy of excellent work by David Broockman, Nicholas Carnes, Melody Crowder-Meyer, and Christopher Skovron, who surveyed 6,000 county-level political party leaders. They found that, “party leaders…use their influence to discourage moderates from seeking office: they strongly prefer candidates at least as ideologically polarized as their median party member. Republican party leaders show this preference especially.”

Their findings also reinforce something that should be apparent to students of polarization—that polarization is asymmetric. Republicans have moved far to the right. Democrats have mostly stayed in place. Let me quote Broockman et al.’s paper at further length, because the findings are extremely relevant to this debate:

“Republicans are much more likely to, unprompted, mention ideology as an important factor for candidates. Our evidence suggests that not only do Republicans care more about ideology, it is also readily accessible when they think of candidate recruitment. It seems likely, then, that Republicans are much more active in recruiting ideologically polarized candidates than Democrats are.”

“Democratic chairs are most inclined to support candidates who are middle-of-the-road or slightly left with respect to the party, while Republicans prefer candidates who espouse an ideology matching or more conservative than their party. In fact, while Democratic chairs are less likely to support very liberal candidates than those nearer to their party average, Republican chairs seem to give very conservative primary candidates the same boost that Democrats give to moderates.”

This does suggest that perhaps giving parties more money and therefore more control over candidates would produce moderation in blue states, but exacerbate polarization in red states. Unfortunately, there is nothing in LaRaja and Schaffner’s analysis that addresses this possibility.

The importance of recruitment also suggests that what we really want to know is who controls the actual recruitment mechanisms in the first place. It’s possible that states with limits might have strong party recruitment mechanisms. If what we really care about is the strength of party machines, why not try to measure that more directly?

LaRaja and Schaffner seem to envision parties being run by hard-headed pragmatists who can determine outcomes with money alone. They seem to assume that if parties can get billionaires to fund them, this will enable party leaders to support more moderate candidates. They seem to ignore that the billionaires may have a few ideas of their own about how they think government should be run (see, e.g. North Carolina).

This gets to a final point, about whether we ought to care if parties rely on small or large donors. LaRaja and Schaffner dismiss the case for small donors, noting that: “the endless romanticizing of small donors as being emblematic of American voters has no empirical grounding.” They go on to note that the ideological distribution of small donors and the ideological distribution of large donors “are nearly identical,” and therefore, “[p]utting more emphasis on ideological small donors may even make our politics worse as politicians streamline their messages to cater to this minority of individuals with more extreme views.”

Let’s grant that large and small donors have the same ideological distribution. If there is no difference, then there’s no reason to think that relying more on small donors would make politics any more extreme. However, since there are many more small donors than there are large donors, a small-donor matching system would allow less extreme candidates the ability to seek out less extreme donors from a larger population of potential donors. We know large donors are polarized, so relying more heavily on them doesn’t give parties much room to moderate. Of course, this presumes that large donors want to shape party positions. But that seems a safe bet.

There are also good (small-d) democratic reasons to support small-donor programs: they bring more participants into the political process; they orient politicians to think differently about whom they represent, and they probably make politics an attractive profession for a broader set of potential candidates. I’d even trade off some polarization for a small donor system. Fortunately, based on their data, it doesn’t appear that I’d even have to.

Finally, and perhaps most importantly, polarization is a function of many, many things, and it’s hard to imagine how changing limits on what parties can raise and spend would have much of an influence given the many other factors. Consider this thoughtful systems map developed by the Hewlett Foundation to analyze American politics as a system: it describes multiple factors that might influence levels of polarization. Systems thinking warns us to be careful of putting too much focus on a single point of leverage without thinking about the larger systems dynamics. This is why many reform skeptics are cautious about unintended consequences—thinking about a single variable in the absence of a larger context usually has unexpected results.

Moreover, as Mann and Dionne explain, we need to be cautious of applying lessons from the states to Washington:

“The gridlock in Washington is a consequence of the ideological polarization of the parties buttressed by vast party networks, their strategic opposition to one another throughout the legislative process fueled by the intense competition for control of the White House and Congress, the prevalence of divided party government, and the asymmetry between the parties that leads Republicans to eschew negotiation and compromise.”

“The situation in the states is dramatically different. Most now have unified party governments, and gridlock is the exception, not the rule. There is little evidence of moderation in the Republican- controlled states, whatever their campaign finance laws.”

I’m sure we will continue this debate for many months to come, especially after the publication of Campaign Finance and Political Polarization: When Purists Prevail this fall. I’m glad that LaRaja and Schaffner are bringing valuable data to this important question. It’s certainly far from settled.