Wharton's Eric Gilje discusses his research on the impact of the fracking boom on voter preferences.

What happens to voter preferences in an area that sees a sudden influx of wealth? According to the research paper, “Voter Preferences and Political Change: Evidence from the Political Economy of Shale Booms,” voters whose fortunes swelled as a result of the shale boom switched to the Republican party from the Democratic party en masse — and they elected politicians who reflected their newfound beliefs, according to Wharton finance professor Erik Gilje, who co-authored the paper along with Viktar Fedaseyeu, a professor at Bocconi University in Italy and Philip E. Strahan, a professor at Boston College.

The boom in shale oil, which uses a controversial new technology called ‘fracking,’ has minted sudden oil and gas millionaires. Gilje said this wealth effect led voters to support the party that backs energy development, which has implications for congressional races in these states. The paper studied voter data in seven states from 1996 to 2012.

The research showed that before the shale boom, less than half of the House seats were held by Republicans in the areas studied. After the boom, the GOP held more than 80% of the seats. Voter shifts affected political races from the county and district level all the way to presidential elections. Moreover, their changing mindsets also spilled over into other arenas, such as civil rights, labor and tax policies, among others.

Knowledge at Wharton recently spoke to Gilje about his research. An edited transcript of the conversation appears below.

Knowledge at Wharton: We’re here with Wharton finance professor Erik Gilje to talk about one of his research papers, which actually found a link between voter preferences and the shale boom. So tell us about your paper.

Gilje: Basically, what we are trying to do is trace out how political change occurs in the U.S. electoral system. There are two sets of hypotheses. One is that when voter preferences shift, your elected officials change their positions to adapt to the new preferences. The other is when voter preferences shift, they throw out their old elected representatives and bring in somebody new. So we kind of test each of these two hypotheses within the context of shale discoveries.

Shale discoveries are an interesting place to look at this because when these discoveries occur, there’s a large change in voter preferences toward becoming more conservative and supportive of issues that will help underpin the development of shale. What we see overall is that there’s a shift in voting for more conservative political candidates — Republican candidates — and that linked with this, rather than existing Democratic candidates changing their views, … they lose their jobs and Republican candidates replace them.

“What we see overall is that there’s a shift in voting for more conservative political candidates — Republican candidates.”

So, in terms of which of these two hypotheses seem to dominate, it’s the case that you basically find a new representative that fits your preferences as opposed to your elected representative adapting their preference towards your views — which is kind of interesting because we typically think of politicians as saying whatever they need to say to get elected. This evidence would suggest otherwise.

Knowledge at Wharton: So you actually gathered data from seven states — I understand those are red states.

Gilje: Some of them are red, some of them are, I guess, purple. You have Pennsylvania, West Virginia, North Dakota, Arkansas, Oklahoma, Louisiana, Texas and many of these states actually were maybe not always considered red states, certainly prior to shale discoveries.

Knowledge at Wharton: What are some key takeaways from your paper?

Gilje: The key takeaways are these, that in essence you see that the mechanism through which the political change occurs is through bringing in new representatives and that we find several other effects in that when you bring in a new representative, you get a lot of other things with that, not just people that help protect shale but also people that may vote differently on social issues or other issues that are unrelated to shale.

And then lastly, we kind of have a final result in the paper where we show that even the Democrats that adjusted their voting records slightly to become more conservative still lost their jobs, suggesting that maybe they couldn’t credibly convey that they had adapted to the new preferences of their congressional districts.

Knowledge at Wharton: I think what really grabbed me about your paper was your major finding that when there’s a big, positive economic shock to an area that more Democrats become Republicans. Can you tell me why?

Gilje: This is one interesting aspect of the analysis where we’re actually able to obtain access to exit poll data and so some people might say “Well, this result was driven by the changes in the electoral makeup, people moving into the area”, but in fact, through this exit poll data, people are asked about their prior political preferences and we actually show that people that self-identify as historically being more liberal, shift to be more conservative when these shale discoveries occur.

It’s maybe not surprising when you think about the positions that each of these political parties have tended to have on energy development and shale. Also, you have in these areas large increases in income, large increases in job growth and that basically, people in these areas adapt their preferences to try to protect and insure that this development continues. And historically, the Republican Party has been more consistent with doing that with energy development.

“Since the beginning of the shale boom, 17 Democratic seats have shifted to Republicans … that’s half of the current Republican majority in Congress.”

Knowledge at Wharton: Do you think you can handicap the race for us this election year?

Gilje: What the results suggest is not necessarily about one presidential candidate versus another because the results are really at the congressional district level, but they suggest that the magnitude of the shift that has occurred that’s linked to shale discoveries are quite large. In our paper, we say that in the aggregate, since the beginning of the shale boom, 17 Democratic seats have shifted to Republicans and in terms of economic magnitudes, that’s half of the current Republican majority in congress.

So to me, what it suggests is that if the Democratic party does not adapt its views or seek out these energy voters in a way that they haven’t been [sought out] before — gaining control of the U.S. House of Representatives will be challenging for them because a large [number] of the Republican [seats] exists due to their preference and support of shale development and so until that part of the Democratic platform maybe changes or adapts to that aspect of the electoral calculus, I think it will be challenging on at least the House side for Democrats to make significant headway.

Knowledge at Wharton: Did that come as a surprise to you?

Gilje: The magnitude certainly came as a surprise. I think we expected to see some shift towards Republicans but when you actually trace out the effect on house seats, compare that to the current majority in Congress, what you end up seeing is that this was quite important for the shift in the U.S. House that we’ve seen over the last 15 years — and this is linked with both shale oil and shale gas development, essentially. When we plot this out, we see one shift that occurs when shale gas development happens in the mid-2000s and then another one that occurs later on in 2010, 2011, 2012, when shale oil development really gets going.

Knowledge at Wharton: Were there other findings that surprised you?

Gilje: We looked at different interest groups’ scoring of congressmen and what we find is even those that adapt a little bit are not able to maintain their seats and so that suggests that more significant steps would need to be taken to demonstrate they’ve adapted to the new voter preferences.

Knowledge at Wharton: What are some practical implications of your findings?

Gilje: I would say in practice, it documents a few things. First, it shows that there are a lot of spillover effects into other policy areas when one policy area changes. So the shift in energy preferences by voters leads to shifts in other areas — social issues, tax issues that maybe you may not have expected. And I would say that’s probably one of the most interesting practical aspects of it.

Knowledge at Wharton: What sets your research apart from prior work in this area?

“If the Democratic Party does not adapt its views or seek out these energy voters … gaining control of the U.S. House of Representatives will be challenging for them.”

Gilje: Prior work by and large has not focused on the spillover effects. You can observe something in the data to see whether people actually change their votes one way or another. What we’re showing is that because of the electoral system that we have, that having an elected representative, they have their views and you basically kind of rank how important your views are and you may end up getting some things in there with that electoral representative you may not have anticipated.

The other aspect, which is new, is to show that even those who do shift more to the right, it doesn’t seem to matter for voters that much. So it’s either a question of credibility — they can’t credibly shift — or a question of the nature of the party structure that even if they do shift, the rest of the party … may not be conducive to advancing the voter preferences that have now changed in these districts.

Knowledge at Wharton: How are you going to follow up your research?

Gilje: There are a couple different avenues we’re exploring. One avenue would be to look at the sorts of people who seem more apt to shift their preferences or their particular characteristics or demographics. Another is to look at how this political change spills over and affects firms. So you might imagine that firms are working to build political relationships, to get favors of some sort, or preferences of some sort in different aspects of government — whether it’s regulation or laws. Now, you have a shift that’s occurred because of the energy issue and so you might expect that some firms are now going to be more out of favor because they were connected to politicians that lost their seats. Some other firms may be more in favor because now they’re connected to politicians that are in power.