Chicago Booth Review Podcast This 100-Year-Old Pattern Explains Trump’s Victory

Lubos Pastor: People see through campaigns. I think people are smart. People see the Democratic campaign, they filter through it. People see the Republican, they filter through it, and then they vote the way that is best for them.

Hal Weitzman: Donald Trump's victory in the presidential election took many pundits by surprise. Voters consistently said the economy was their top issue, but even though the US economy is currently the world's top performer, they nevertheless voted for a change in government. That led many observers to claim that Trump's supporters had a warped view of the economy, concentrating more on inflation than on growth. Chicago Booth's Lubos Pastor, disagrees. His research suggests that voters knew the economy was growing, and that was in fact the very reason why they wanted a change of government.

Welcome to the Chicago Booth Review Podcast, where we bring you groundbreaking academic research in a clear and straightforward way. I'm Hal Weitzman, and today I'm talking with Lubos Pastor about how a model with a 100-year record of predicting presidential outcomes was proved correct again. Pastor's model is relatively simple. When the economy is strong, more voters opt for the Republican candidate in elections, while a weak economy favors Democrats. This pattern has held since 1927, regardless of who the candidates or the incumbents are. The reason has to do with how the economy shapes people's attitude to risk, which also explains in other phenomenon, the stock market performs much better under Democratic presidents than under Republicans. Lubos Pastor, welcome to the Chicago Booth Review Podcast.

Lubos Pastor: Thanks for having me.

Hal Weitzman: Now, listen, we had to have you on after in the wake of this victory for Donald Trump and the Republicans because you have this fascinating thesis that a strong economy favors Republicans in elections, a weak economy favors Democrats, regardless of who the candidates or the incumbents are. We could've used your research to predict better than the polls to predict the election outcome. Were you thinking of that on election night?

Lubos Pastor: I was. We published this paper, Pietro Veronesi, my colleague and I, published this paper in 2020, and we started working on it in 2015. Ever since, I pay attention to the election, the presidential election every four years and see whether it works out, and all three times out of sample, our model predicted who's going to win. I'll be curious four years from now as well.

Hal Weitzman: Because you've taken this back historically and looked, and that pattern has held since when?

Lubos Pastor: Well, we go back to 1927, which is the beginning of our in-house crisp database.

Hal Weitzman: It's a hundred years now, and it's held every time, and it held this time as well.

Lubos Pastor: That's about right, yeah.

Hal Weitzman: What's so fascinating about it is I feel like if there's any common wisdom about the economy's effect on politics is that a bad economy will almost necessarily lead to a change in government, and you disagree with that. Voters in this election repeatedly told pollsters that the economy was their top issue, and so the pundits who think economic performance is critical had two reactions. One is they were surprised because the economy was strong, so why shouldn't the incumbents win? Kamala Harris or the Democrats in general, they were supposed to have won. The other reaction seems to have been that the Trump supporters were mistaken. They were acting on false information. They thought the economy was bad, the Republicans were constantly telling them that. They were reinforcing the sense of inflation, but your research takes a very different approach. Explain to us what your research tells us about why Trump won.

Lubos Pastor: There's a lot here. First of all, you've mentioned the number one electoral predictor, which is that a strong economy tends to favor the incumbent. This has a name in political economy, it's called retrospective voting. People assess the state of the economy. If the economy is good, they're going to re-elect the incumbent. If the economy is bad, they're going to kick the incumbent out and bringing something new. It's not that I disagree with this, I actually agree with this. This thing is real. There's a fair amount of truth to it. It's just not the whole story. It's just not the whole story, and it wasn't the whole story this time either, as you could tell, because the economy under Biden, certainly in the past couple of years, has been strong.

The US has been doing really well in 2023, 2024. It emerged from COVID stronger than any other country in the West. The gap between the US and the rest of the West has widened in recent years. The unemployment rate is low, the US economy is doing very well. Retrospective voting would say in a strong economy, Kamala Harris should be elected, Democrats should be re-elected, and they weren't. Well, this is where our paper comes in because our main thesis is not that a strong economy favors the incumbent. Our main thesis is that a strong economy favors Republicans regardless of who the incumbent is, and a weak economy favors Democrats regardless of who the incumbent is.

Hal Weitzman: Okay, and so tell us why do you come to that conclusion? What's the mechanics?

Lubos Pastor: First of all, we have a rigorous model that explains all this, but actually, there's a simple way of summarizing what's going on. When the economy is weak, people are more risk averse, people are more reluctant to take risks, they tend to be scared, and they're more likely to demand social insurance. They demand redistribution. Their demand for social safety net goes up. When the economy is weak, they end up voting Democrat because Democrats are the party that's more likely to provide this social redistribution. Democrats are well known to provide a stronger social safety net than Republicans. When the economy is strong, people are much more tolerant of risks, they're much more willing to take on business risks, so they vote for the party that allows them to keep more of their gains, which would be the Republican party.

Hal Weitzman: Well, what about this sense that you did get in this election, that there was a disconnect between the experience that people had of the economy or what they call the economy, which they mean prices they pay in the store, prices they pay at the pump, versus the headline figures that you're talking about, GDP, jobless rates?

Lubos Pastor: Well, so the story you just mentioned is indeed, that's the leading story in the media. Journalists trying to explain how come Democrats were kicked out despite the strong economy. They always talk about people being confused. Some Americans were confused about the economy. They looked at the price of eggs, and they didn't understand that the economy was strong. To me, this mainstream view in the media is a tad arrogant because it views American voters as irrational, as confused. I think American voters are smart. In our view, and according to our model, nobody was confused. Everybody was acting perfectly rationally. Americans understood the economy was strong, and they acted in their own self-interest. See, look, if you're a journalist and you want to tell the story why Democrats lost, this is the story you'll tell, Americans were confused. You'll always find two or three voters who will say, "Oh, the price of eggs is high, and that's why I voted for Trump." That's what they could say. You don't need to tell the story.

Hal Weitzman: It's not just the voters, it's also the way that the campaign was constructed.

Lubos Pastor: One way is what people say in the campaign, another way is how people think. People see through campaigns. I think people are smart. People see the Democratic campaign, they filter through it. People see the Republican, they filter through it, and then they vote the way that is best for them.

Hal Weitzman: Are you saying that most people who voted for Trump knew that the economy was doing well, and that's actually why they voted for Trump, or you're saying it doesn't really matter that their revealed preference comes out? Are they consciously voting for Trump and discounting everything he says about the Democrats screwing up the economy?

Lubos Pastor: In our simple model, there's no role for campaigns. People talk in the campaigns, see through the campaigns.

Hal Weitzman: Just to clarify, you're not looking primarily or principally at the data, you're looking first at your model.

Lubos Pastor: We look at both the model and the data. We have a new model with a new mechanism, and we look at the data as well. Just to respond to your question, in our setting, people are smart. They know the economy is strong right now, and in a strong economy, they're much more willing to take risks, so they're much more willing to vote Republican. Let me actually, so you mentioned data. If you think about this thesis that a weak economy favors Democrats and vice versa, you find a lot of support for it in historical data. Look at the past hundred years, take the single biggest crisis we've had in this country, the Great Depression. At the peak of the Great Depression, November, 1932, people elected a Democrat, Franklin Delano Roosevelt. Take the second biggest crisis we've had, the Financial Crisis of 2008. In November, 2008, Barack Obama, another Democrat, was elected. I can give you examples of JFK, Bill Clinton, Jimmy Carter, all of these democratic presidents were elected either during or after recessions.

These are all cases that we knew about while writing our paper. After we wrote our paper, the first draft of our paper, three more elections took place. The first one, November, 2016, as you know, Donald Trump got elected, a Republican. The economy was strong under Donald Trump. Why were Democrats kicked out in 2016 when the economy was strong? Retrospective voting has to look for, again, for confused voters. We don't. In our setting, that's exactly what you would expect. In a strong economy, people want to take risks. People are more likely to vote Republican. Four years later, November, 2020, we were in the midst of the COVID crisis. It's like a textbook example of our model. People were scared, literally. Not just of the economy, but of all kinds of risks, health risks. In November, 2020, they elected a Democrat. Again, their demand for redistribution, social insurance led them to vote Democratic. A few days ago, when the economy is strong, again, people elected a Republican. I'd say that's three for three out of sample.

Hal Weitzman: Absolutely. It's just that the campaigns, it's almost it would be rational then, according to your model for the Republicans to emphasize how well the economy is doing, and the Democrats, the incumbents to say, "Yes, everything is awful. You need us back."

Lubos Pastor: Yeah, good point. The simple story I'm telling is one in which there are no campaigns. Voters look at the state of the economy, they understand what's going on. Surely, to understand why Democrats were messaging something, Republicans were messaging something else, we would need to go outside of our very simple framework.

Hal Weitzman: To what extent do you think, I mean, we look at the election, there were so many different issues that people were voting on. The economy was just one of them. How do we think about the economy versus cultural issues?

Lubos Pastor: Well, you did say at the outset that people told pollsters that the economy was their number one issue.

Hal Weitzman: That's true.

Lubos Pastor: In our model, there's only one thing going on, the economy. By the way, in our model, there's one difference between Democrats and Republicans, and that's the tax rate. Democrats like high tax rate and more redistribution. Republicans like a low tax rate and less redistribution. There's nothing else going on. No abortion, no guns, no immigration, nothing. Now, did those other issues play a role? They always play a role. People care about much more than just the level of the tax rate. However, I would argue that those other issues tend to cancel out in a typical election.

Take abortion, on the issue of abortion, I think it's pretty clear the median voter was on the side of the Democratic party. Take another issue, take issues like immigration or identity politics. On those issues, the median voter was very clearly on the side of the Republican party. You have all these non-economic issues, and some people are swayed one way, some people are swayed the other way. On average, they cancel out because both parties go after the median voter. Both parties change their messaging to go after the median voter. In the end, you end up with the economics playing the lead role.

Hal Weitzman: It sounds like, as you say, you've gone back through history, the past hundred years. The famous line, it's the is economy stupid. It's always been the economy is stupid.

Lubos Pastor: Look, I'm sure you'll find elections when one of these cultural issues ended up dominating. By the way, even in our model, you get some uncertain, some randomness, because I simplified things a bit earlier on. What happens in our model is when people are sufficiently risk-averse, which tends to happen during crises, that's when they always vote Democrat. When people's risk aversion is sufficiently low, like during booms. that-

Hal Weitzman: Meaning, they're willing to take more of a risk.

Lubos Pastor: Exactly. That's when they always vote Republican. If risk aversion is in between, then the outcome is indeterminate. Then cultural issues decide the outcome of the election, even in our simple model.

Hal Weitzman: If you're enjoying this podcast, there's another University of Chicago podcast network show that you should check out, it's called The Pie. Economists are always talking about the pie, how it grows and shrinks, how it's sliced, and who gets the biggest share. Join host Tess Vigeland as she talks with leading economists about their cutting-edge research and key events of the day. Hear how the economic pie is at the heart of issues like the aftermath of a global pandemic, jobs, energy policy, and much more.

Lubos, in the first half we talked about your research about the economy. When the economy is doing well, voters tend to vote Republican, and when the economy is doing badly, they not tend to, they do, they vote Democrat, and this has happened for the past hundred years. There's something that motivated that, which is a pattern that is well known, called the presidential puzzle, which is about how the stock market behaves under Democrat or Republican presidents. Tell us about that.

Lubos Pastor: This is a pattern that was identified by other researchers, by for example, Santa-Clara and Valkanov in a 2003 paper. They looked back at stock market returns from 1927 through 1998, and they discovered that under Democratic presidents, the stock market did better on average than under Republican presidents. We added 17 years to their data, so we went from 1927 to 2015 and found an even stronger pattern. Over this period of almost 90 years, the average stock market access return under Democrats was 11% per year higher than under Republicans. It's hard to talk about it. It seems like a puzzle because we didn't really know where it was coming from. A risk appears the same, about the same under Democrats and Republicans, for example. It just wasn't clear where it's coming from.

Santa-Clara and Valkanov dubbed it a puzzle. What could be causing this pattern? That's what we ask in our paper. We argue that it's not that Democrats are somehow causing high stock market returns or that Republicans are causing low market returns. We argue that it's not what presidents do, but rather when they get elected, that explains the pattern. As we discussed in the first half, in times of trouble, when people's risk aversion is high, that's when voters tend to elect Democrats. Well, think about what happens in times of trouble when risk aversion is high, stock valuations tend to be low and expected stock market returns tend to be high. When a Democrat comes in-

Hal Weitzman: Just explain that, Lubos.

Lubos Pastor: Let me explain on an example. Take November, 2008, and we're voting in the election, and people voted for Barack Obama to become president. This was in the middle of the Financial Crisis. Stocks had just lost almost half of their value in the 2007, 2008 election, so stock valuations were really low. People were highly risk averse, risk premium were very large in the market. What that means in an environment like that, the expected market return going forward is high because people require compensation for holding risky assets during such periods. This is called a risk premium. You demand a higher expected return on stocks than on, let's say, a Treasury bill, because you are concerned about risk in the middle of the financial crisis. After Obama came in, of course the expected market return was high, and it ended up being realized. Whereas take the opposite case when the economy is booming like right now, stocks are hitting historical records.

The stock market has hit the historical high, I don't know, 50 times or something this year. Stock prices are really high. Stock valuations are really stretched. It's in periods like this that voters elect a Republican, as we discussed previously. This is when risk conversion is low. Going forward, risk premia are low and expected market returns are low. On average, you should expect to see lower market returns under Republicans and higher market returns under Democrats according to this story, according to our model. It's not because Democrats are somehow great at manufacturing high-market returns, or that Republicans are somehow terrible at destroying market value. It's just it's a byproduct of when they get elected.

Hal Weitzman: It just suggests that presidents can't claim any credit for, or take any blame for anything that happens in the markets.

Lubos Pastor: That's what we're saying, basically.

Hal Weitzman: Maybe in the economy more generally as well. Just to be clear, when do you take these states from? When Donald Trump was elected, the stock market went up like 4% or something the following day. Is it from the day that they're elected or is it there's this gap in America, isn't it, between the election and the swearing in?

Lubos Pastor: We date everything from the swearing in. You start at the end of January, so Donald Trump's clock will begin ticking at the end of January. All this run up we just experienced was under Joe Biden.

Hal Weitzman: How do you count that? Because even before the election, markets were predicting that he would win.

Lubos Pastor: Yes. There are two separate things going on. One is the presidential puzzle that we've talked about, 11% per year difference between Democrats and Republicans. The other thing is, and this is well known, this is not our paper, other people have found that the stock market tends to go up on average when a Republican gets elected, especially if it's a surprise. This time a Republican got elected, and like you said, the stock market responded positively. Even before the election, it looks like the market was pricing in an elevated likelihood of Trump victory. This effect, however, the second effect is smaller in magnitude compared to the one we just talked about.

The presidential puzzle is 11% per year, so it's like almost 50% compounded in a four-year presidential term, whereas the run up we're talking about in response to Trump's surprise election or somewhat surprise election, that's like single-digit percentage points. We've done the analysis both ways. If you let Trump's clock tick in November rather than in January, and you do this for every president, the presidential puzzle is going to be slightly smaller. It won't be 11% per year, it'll be a slightly smaller number, but it'll still be highly statistically significant.

Hal Weitzman: In 2016, of course, we ran an article in Chicago Booth Review about your research that said something like, you can look it up, please, "Don't expect too much from a Trump win in terms of the stock market." Sure enough, the returns that stock market investors enjoyed during the Trump years were less than they enjoyed during the Biden years, so it's still playing out.

Lubos Pastor: It turns out since 2016, returns were really high under both Trump and Biden. Under Biden, they were really high, as we know. The stock market is hitting new highs right now as we speak, but even under Trump, stocks did well. Stocks did fall under Trump during COVID. When we were hit by COVID, stocks lost about a third of their value, but then there was massive stimulus in the second half of 2020 way outside of our model, and that ensured high stock market returns even under Trump. With high returns under both Trump and Biden out of sample, I don't think that's going to change the balance over a hundred years.

Hal Weitzman: If you haven't bought stocks yet and you're about, this sounds like, well, they might still go up, but not as much as they would've gone up had Kamala Harris been our president.

Lubos Pastor: If I were to apply our model to the current election, I would've to argue, well, it's not surprising stocks went up recently in response to, and also in anticipation of the Trump win. That is what our model would predict, because why? Because Republicans mean lower taxes and lower taxes are good. That includes the corporate tax rate, and that is good for the stock market, so it's as simple as that. However, all of this is already priced in. Everybody knows Trump is going in, in January. Going forward, I think our main mechanism, which is the risk premium-based mechanism based on risk conversion should kick in, and that's why I believe that over the next four years, the expected market return should be somewhat lower than the historical average. It's just hard with such high stock valuations to predict otherwise.

Hal Weitzman: Now, half of Americans don't own any stocks, so how do we think about them? Because they obviously, half the electorate isn't affected by what you're talking about. Do they still have this risk aversion that you're talking about, same sentiment?

Lubos Pastor: I think so. Look, when the economy is doing well, everybody, I mean, most people get to appreciate it whether they own stock or not. When the economy is doing poorly, most people get to appreciate whether they own stock or not. Everybody votes, everybody has an opportunity to vote. It's true, not everybody gets to price stocks in the market, but as long as people who price stocks are at least somewhat similar to people who don't, this doesn't really make much difference, so our effect should go through.

Hal Weitzman: Lubos, talk briefly about other countries. I know you've looked at the UK. What have you found then, both in terms of the thing we talked about the first half, the correlation or the connection between the economy, economic growth, and which party gets elected and the stock market too, if you have data on that?

Lubos Pastor: We found that the outcome of the US election matters also for other countries, not just for the United States, and we find a similar pattern in other countries as a result. It's perhaps not too surprising because stock markets are correlated across countries, so I don't view that result of ours as somehow particularly important.

Hal Weitzman: Do you find that left-wing parties get elected when, or parties that are generally on?

Lubos Pastor: No, it's the US election that matters for everyone.

Hal Weitzman: I see.

Lubos Pastor: In the Netherlands, it doesn't matter whether it's the left-wing party or the right-wing party that wins because Dutch stocks are not priced by Dutch investors. In the global market, people hold stocks globally and it turns out there's one election that rules them all, and that's the US election.

Hal Weitzman: What about the fact that, so in the UK or in a European country, when the economy is doing well, do they tend to gravitate to the right-of-center party?

Lubos Pastor: There's research in political science showing that during crises, in particular, banking crises, left-wing parties tend to get elected, not just in the US, but elsewhere. There's also research in political science showing, like looking across US states that states that have higher unemployment rates, so again, a weaker economy, are more likely to vote for the left-wing party. Yes, there are elements of what you're describing in political science, but what matters for the stock market turns out to be just the US election.

Hal Weitzman: Lubos, have you solved the presidential puzzle?

Lubos Pastor: Well, we do propose a simple, rational explanation in which the outcome people have described as the presidential puzzle is not puzzling. Again, the puzzle was that stocks do better under Democrats than under Republicans, and that's exactly what happens in our simple models in which everybody is rational, nobody is confused. At least in that sense, I would argue, yes, I'm happy not to call it a puzzle anymore.

Hal Weitzman: Looking forward four years, we should expect stock market may perform well, but not as well as it might've done under a Democratic president.

Lubos Pastor: That's what the model predicts.

Hal Weitzman: That if the economy is doing badly, we will have a change in leadership in the White House.

Lubos Pastor: That is what the model predicts, indeed, but it's hard to tell what's going to come four years from now.

Hal Weitzman: All right. We'll have you back in four years and we'll both see.

Lubos Pastor: Sounds good.

Hal Weitzman: Lubos Pastor, thank you very much for coming on the Chicago Booth Review Podcast.

Lubos Pastor: My pleasure. Thanks for having me.

Hal Weitzman: That's it for this episode of the Chicago Booth Review Podcast, part of the University of Chicago Podcast Network. For more research, analysis, and insights, visit our website, chicagobooth.edu/review. When you're there, sign up for our weekly newsletter so you never miss the latest in business-focused academic research. This episode was produced by Josh Stunkel. If you enjoyed it, please subscribe, and please do leave us a five-star review. Until next time, I'm Hal Weitzman. Thanks for listening.