Michael Steele’s Gaffes: An Empirical Assessment
In the wake of Michael Steele’s most recent gaffe, political pundits and observers opined on the topic of Steele’s effects (broadly defined) on the Republican Party. With little variation, the subject was posed in the following form: How much damage has Steele caused? The answers to this question, however, had some variability. One side argued essentially no effect—where the argument was (1) the American public doesn’t understand what a party chairman does and thusly; (2) the public doesn’t care what some unknown party officialhas to say. The other, and more common, side argued major effects—where the argument was (1) Steele has damaged the Republican image; (2) Steele’s policy statements are outside the party’s mainstream and therefore create disunity; and (3) donors have been less willing to open their checkbooks for Steele. Though these competing perspectives are amenable to empirical scrutiny, the question has been rarely addressed in this manner.
Let’s address the money issue first, since it’s the most specific and easiest to answer. Has Steele hurt the Republican Party’s fund-raising abilities? Fund-raising is, after all, a party chairman’s primary responsibility. I think the answer to this question is “sort of…actually, it depends.” First, through Republicans are generally out raising Democrats, the gap between the two is relatively minor. Chart? Chart. What you can see is that Republicans pulled in $78.7 million in the second quarter of this year compared to $86.4 million for Democrats in the same period. The (relatively minor) size of this gap is surprising given the energy within the Republican base and the near universal perception that Republicans will take back the House. But the central question requires a counterfactual. Would Republicans be raising more money with someone other than Michael Steele as chairman? While we can’t answer this definitively for obvious reasons, I think the answer is yes. For example, we do know for certain that Republican donors have been giving less to RNC in favor of other GOP party organizations. In April, The Daily Caller published a story citing FEC filings which reported that some of the RNC’s top donors have failed to contribute in the past year, donating instead to the NRSC and NRCC. Thus, while I am of the opinion that Steele has hurt Republicans’ fundraising in a general sense, it is important to bear in mind that many donors have simply shifted their contributions to other party organizations. So the answer is yes, with an important caveat.
The other issue—do people care about Steele’s gaffes—is also easily addressed with empirics if one is so inclined. The following figure consists of Google Trends data for the search term “Michael Steele.” The figure charts variation in people searching for information about Steele via Google over time—from the point when Steele announced his intentions to run for RNC chairman (November 2008) to just after his most recent Afghanistan gaffe. Though this is certainly not perfect data, I think there is one key benefit to using Google searchers in this fashion: unlike news coverage or public opinion surveys, internet searches require a deliberate act on the part of the individual. Conceivably, if people didn’t care what Steele had to say, they wouldn’t make an effort to find information about him.
On this figure, the appropriately colored red dots coincide with the points in time where Steele made a public gaffe. One dot is at the exact time period of his gaffe, the second is the subsequent period. Each point in time corresponds to one week. The green dots demarcate Steele’s decision to run for the RNC chairmanship and his subsequent election as RNC chair. Here is that data:
March 2009: Steele said that Rush Limbaugh’s show is “incendiary” and “ugly.” Steele quickly apologized.
March 2009: Steele drew heat from conservative Republicans for saying in an interview that abortion is an “individual choice.”
April 2009: Steele promised to bring a “hip-hop” makeover to the Republican Party. Democrats and pundits ridiculed these statements.
January 2009: On Sean Hannity’s show, Steele said he didn’t think Republicans could take the House in 2010.
March and April 2010: Two events occurred around the same period. The first, though technically not a gaffe, was the revelation that the RNC paid $2000 in expenses to a bondage themed strip club. On the heels of this, Steele claimed that he has a “slimmer margin for error” because of his race.
July 2010: Steele makes the claim that Afghanistan is a “war of Obama’s choosing.” He also suggests that the war in Afghanistan was a poor decision, contrary to the opinion of most of his fellow partisans.
We can see that the time series spikes when Steele made a public gaffe. Again, though the causal inference is not perfect, I would claim that this indicates people do care (to what extent is debatable) what Steele has to say. This point is a critical position advocated by Steele’s (few) defenders, who claim the opposite of what the data appears to show. Still, two of the gaffes appear to have had little to no effect (his claim that abortion is an “individual choice” and his promise to bring a “hip-hop” makeover to the Republican party). To confirm that the spikes in the time series are statistically meaningful events, I estimated an ARIMA (1,0,0) intervention analysis using pulse-functions to operationalize the main effect (i.e. the red dots). The green events were included in the model as a control. The autoregressive term and control were statistically significant. Standard post-estimation diagnostics confirm that the residuals are white noise. The ARIMA results confirm what seems obvious based on a visual inspection of the data—Steele’s gaffes are meaningful events. I think this provides at least some evidence that people do in fact care what party chairmen (at least, the current RNC chairman) have to say.
In summary, I don’t think Steele’s gaffes are benign comments by a marginal Republican. There appears to be some quantitative evidence reported here that he hurts fundraising for Republicans generally (the RNC specifically) and that individuals care to some extent what he has to say.