Q: You know, university faculties are famously not Republican, and political science departments, or government departments, as you call it here, are famously stand back from politics, and observe it, rather than get in and into the game. So, did that provide any — did those sort of cultural barriers within the profession affect you?
SHAW: It was a little bit of an issue. And actually, subsequently, I had some professors here at Texas and other places who held me accountable [00:07:00] for the Florida fiasco. According to a few of them, I’m personally responsible for the Iraq and Afghanistan wars. That sort of conversations I find fairly tedious. But I don’t know that — the reaction amongst people who do public opinion voting research was almost universally positive. It was on the order of, wow, this is an incredible experience, how lucky you are to be involved. And I know a few friends of mine in the profession who were involved in the Gore campaign, and we’ve compared notes on what happened on their side of the aisle, which was very useful.
But I think those who actually write about voting, and do research in voting on campaigns and elections were extremely supportive, and it was nice after the election to have conversations with them, because they were as distant as I was prior to that. They knew things; they talked with people. They may have had, you know, maybe second-order involvement in campaigns, but to actually have a chance to be there and [00:08:00] see it unfold, really informed my research in a way that I couldn’t have without that experience.
My favorite example is that after the election, there was a lot of conversation about why the Bush campaign had wasted resources in California. And I actually went to a couple of panels at the Midwest Political Science Association where people discussed this, and talked about how it made no sense, and a good friend of mine, Larry Bartels at Princeton talked about this as an example of how, you know, even supposedly good campaigns made poor resource decisions.
What actually happened was, is that because George H. W. Bush had raised so much money in California, and not spent any money there in 1992, the California Republican Party and California groups were extremely reluctant to open their wallets for George W. Bush, because it didn’t look like California was going to be competitive in 2000. So, the Bush campaign at a variety of levels made assurances to people in California and the Republican [00:09:00] party that Bush would not leave California, hang them out to dry, that he would spend some resources there in support of statewide candidates, and would make a visit there.
Well, it’s October, and Bush had committed I believe on the order of five million dollars in television advertising money to California. In October, the money was still kind of targeted for California, and the campaign — I think Matt Dowd was a major player in this, and Karl Rove also — decided that they would wrap the California expenditure up, and sort of present it as an example of Bush being on the offense. You know, they’d made this commitment, but they thought they should get some mileage out of it. And if you recall in the 2000 election, Washington and Oregon and New Mexico were all battleground states, and so Bush had a planned trip to the Pacific Northwest. He was going to hit Albuquerque or Santa Fe on the way back to Austin. So he figured, well, they’d stop in Los Angeles and Burbank, and they’d do a shot on The Tonight Show. They’d put [00:10:00] the money out there and buy, you know, five million dollars worth of ads in Los Angeles, and San Francisco, and Fresno in support of some of these candidates, and they would see if they could get an earned media hit, that the newspapers would pick it up as an example that Bush was kind of moving into Gore’s territory and on the offense, and Gore was on the defense. And they’d dress it up.
But the commitment had been made, and the commitment was not one of strategy, except insofar as it was a strategy to raise money in California. So I had to go to these conferences afterwards and have political scientists puzzle over this bizarre decision on the part of the Bush campaign to spend resources in California. Well, that was the only way they were going to raise that money, or at least get the California party people to commit to funding the Bush campaign. But you know, I would have not known that, and you know, none of these political scientists had any opportunity — only, if you’d been in the campaign you could know that. If you hadn’t been, there’s no way you’d know it. And I don’t know that the Bush campaign ever sort of openly talked about that. But it was — you know, it was a decision that was [00:11:00] born of strategic necessity, but not the kind we were focusing on.
Q: Well, this kind of brings us to theme of your book, which is how candidates decide, how their organizations decide where to spend their scarce resources, especially money, and their own time. So, begin with that. I mean, what is it that goes into making those very important strategic decisions about where to spend time and money?
SHAW: I think the premise of the book is in some sense taken from econometrics. And I think a lot of what political science does is to beg, borrow, and steal from both economic and psychology, psychological theories. And I assumed — walked into my research on campaigns assuming that candidates and campaigns are rational actors, that they have limited resources, and they have tangible goals that they’re seeking to achieve. And so, [00:12:00] in the case of a presidential election, you’re trying to get 270 electoral votes. That’s the context, the constitutionally-prescribed context. And the question is, how do you do that?
And one of the things that had not been addressed in the literature before I started my research was the significance of, not just television advertising, but the structure of media markets across the country. This seemed to be something that political scientists hadn’t really understood or come to grips with, that the proper unit of analysis, the level at which advertising is disseminated is the media market. So we had some stuff on states, which is appropriate given the Electoral College, some good stuff in the 1970s by Steven Brams, and [Colin Tony?], and Peter Ordeshook, some interesting things kind of modeling presidential campaign strategies. But it was a small conversation; it hadn’t really gone out to the larger campaigns and voting community.
But this was something when you’re in a campaign that’s unavoidable, that the media market is a critical unit [00:13:00] of analysis, and if you go back and look at strategies all the way back into the Eisenhower days, candidates and their strategists think in terms of media market, with media markets being the building block for winning states. You know, how much advertising do we need in a particular market? What’s the political and behavioral, you know, kind of structure in Fresno, Visalia, Monterey, San Diego, you know, if you’re going to take on California.
And so I thought this was interesting. It’s not a profound insight, and I think after the fact, everybody acknowledged that, well yeah, that makes sense, you know. But, no one had really done it at the time, and we certainly didn’t have any data on spending on that level. So, a couple of conversations with media people or with campaign people will tell you that in fact, polling, any kind of good poll will break the electorate down by media market, and they take a look at the structure of opinion, the nature of partisanship, sense of what’s going on, [00:14:00] mood of the country, within each media market within a particular state, whether it’s Ohio, or Florida, or California, or Pennsylvania, and that what you want to do is identify some set of states that you think are critical to your coalition, and then within the states identify media markets that you think need attention in order to get you to your goal of amassing 270 electoral votes.
And so, what quickly dawned on me was that there were a couple of decisions that were made by campaigns. Within the context of the Electoral College, you had to identify states that you thought were closest to that 270 cutoff point. In other words, you have your base states. You have states that you’re probably never going to win, your opponent’s base. And then there are states that could go either way. And you know, the strategy was fairly obvious, which is, you don’t need to spend a lot of resources in mobilizing your very bottom base states, the states that are going to go with you no matter what. And you certainly don’t want to spend any time in your opponent’s base. But you want to concentrate [00:15:00] your resources on those states that could make the difference between 270 — you know, winning 270, and not winning 270.
And that’s an obvious observation; people had made that before. But it hadn’t really been kind of set up or explored in a systematic way. And I wanted to do that. I wanted to make sure the campaigns were doing that. It turns out, they are. And I wanted to get a sense of, OK, well, that’s sort of a loose strategic context, but is there more to it? And what seemed to be consistent was that states would do something like a fivefold classification system. They would have their base, and the opponent’s base. And then they would have three middle —