Why?

One popular explanation of the rise and survival of Donald Trump is summarized by Senator Elissa Slotkin, who said that some of her constituents tell her "it’s like I’m a Stage 4 cancer patient. My life has been getting worse, from my grandfather to my father, from my father to me, and my kids are going to do worse than me, so I need experimental chemo. Trump is my experimental chemo."  That is, things were getting worse for a long time, under both Democrats and mainstream Republicans, so they got desperate and turned to someone that promised to be completely different.  My last two posts have been directed against this analysis:  most people don't think that their lives have been getting worse, or that they're worse off than their parents.  Then how would I explain the rise and survival of Trump?  

His initial success is mostly because of American political institutions.  First, it's possible for an outsider to become the leader of an existing party.  Members of Congress never had any special role in nominations, and state party leaders are weaker than they used to be.   You don't even have to appeal to the voters in general:  the Iowa caucuses and first few primaries have an outsize influence, so there's a good deal of unpredictability in who gets the nomination.  Second, the two party system is extremely strong, probably stronger than anywhere else in the world, and our system of electing the president helps to keep it that way.  Most nations which elect a president have a runoff if no one gets a majority in the first round.  The United States doesn't:  if no candidate gets a majority in the Electoral College, it goes to the House of Representatives.  That means that if a political party nominates a candidate who many members find unacceptable, it doesn't make sense for them to break away and nominate another candidate.  In 1912, the Republicans nominated William Howard Taft, but former President Theodore Roosevelt ran as an independent.  He ran ahead of Taft, so if we had a runoff it would have been Roosevelt against Woodrow Wilson, and Roosevelt might well have won (he got 27.4% and Taft got 23.2%, making a total of 50.6%).  But as it was, Wilson won with a large electoral college majority despite getting only 41.8% of the popular vote.  

In March 2016, after Trump had become the front-runner for the nomination, an Opinion Research/CNN poll asked Republicans how they would feel if Trump got the nomination.  21% said they would be "angry" and another 12% said "dissatisfied."  The survey also asked if "you would like to see another Republican run for President as a third-party candidate or not," about 35% of those who said they would be dissatisfied or angry said no.  They were asked a follow-up question about whether that was because they would be comfortable with Trump as the nominee or because it would lead to a Democratic win, and about 90% said that it was because it would lead to a Democratic win.  

So the choice comes down to supporting your party's nominee or helping the other party's nominee to get elected.  There were a few previous cases when many members of the party were unhappy with the nominee, most recently McGovern in 1972 and Goldwater in 1964.  In those cases, many party leaders declined to endorse or even openly opposed them.  But partisanship has become stronger, so fewer people are willing to do that today.  According to Wikipedia, 31 Republican Senators and 19 Republican governors endorsed Trump in 2016, against only 14 and 4 for Goldwater in 1964.  

So Trump won in 2016 because he got the Republican nomination, and then party loyalty and negative partisanship took over.  But why did he get the Republican nomination?  It's natural to think that an important event must have deep roots in social and economic conditions--that voters turned to an outsider because they were profoundly dissatisfied with something.  But that's not necessarily the case--there are always people people who are discontented and looking for something different, and the nature of our system of nominating a president means that they have a chance.  For example, it's not clear that Republicans were more dissatisfied in 1964 than they had been before.  There are also changes in general mood that don't seem to reflect social conditions: e. g. people seem to have been discontented in the early 1990s, without any obvious reason.  So I think that there was a significant risk of a Trump-like figure before Trump came along.  

Then there's the question of why Trump survived and came back from his defeat in 2020.  I'll address that in a future post (probably not the next one, but pretty soon).  

[Some data from the Roper Center for Public Opinion Research]

Now and then

 My last post looked at subjective income:  whether you think your income is above average, average, or below average.  The number of people who say that their incomes are average has declined, while the number who say their incomes are below average income has grown--but so has the number who say their incomes are above average.  So there's more dispersion, but no trend in the average rating (there are short-term changes reflecting economic conditions).  This doesn't fit one popular story about the rise and return of Trump:  that people felt like they were falling behind for a long period of time, so they lost faith in "elites" of both parties and turned to an outsider out of frustration.  But that's only one possible comparison:  you sometimes hear that people think that they are no better off, or even worse off, than their parents were.  

Last year (almost exactly one year ago), I wrote about a question that was asked a few times between 1951 and 2016:  "Comparing your present family circumstances with those when you were a child, would you say you are better off, or worse off, than your parents were then?"  In order to get a more complete picture of change, I identified questions that seemed similar--that asked people to compare themselves to their parents at the same age or their family when growing up.  The one that's been asked most frequently is in the GSS starting in 1996:  "Compared to your parents when they were the age you are now, Do you think your own standard of living now is much better, somewhat better, about the same, somewhat worse, or much worse than theirs was?"  Some of the questions had five response categories, some had three, and some just asked better or worse.  I dealt with this by considering the ratio of better to worse answers.  The results:

The ratio was always greater than one, meaning more "better off" than "worse off" answers.  The closest was in September 1980, when 44% said better off, 33% worse off, and 22% about the same.  It seems like opinions have been getting less favorable in the 21st century, but they are still more favorable than in the early 1980s or even the early 1990s (as I've noted before, people seemed to feel negative in the first half of the 1990s).  Specifically, the ratio in 2016 was almost exactly equal to the median of all surveys (there were two surveys:  Gallup found 72% better off and 20% worse off, GSS 59% and 17%, both giving a ratio of 3.6).  

So these data also don't suggest that the rise of Trump was a result of economic frustration.  There are some other interesting things in the figure, like the strongly favorable opinions in the 1980s--the economy was doing well, but no better than in the late 1990s, but the main thing is that people have not been particularly dissatisfied with their living standards in recent years.

[Data from the Roper Center for Public Opinion Research]

Above and below

 Many people have observed that in the last 5 or 10 years, a gap has emerged between public ratings of economic conditions and standard measures like unemployment and inflation:  people say that the economy is bad even when the statistics look pretty good.  One view is that the gap is due to politics or the nature of media coverage; the other is that it's because the standard statistics miss something about people's experience of the economy.  In the New York Times, David French writes that he used to favor the first view, but has now shifted to the second.   He says that what the standard statistics miss is the development of complex pricing systems; rather than offering a standard product at a standard price, companies now offer different levels at different prices.  For example, airlines used to distinguish between first class, business class, and economy--now they make a lot of additional distinctions, charging for a little bit of extra legroom, boarding in the first group, and so on.  His idea is that now people are constantly being reminded that other people are getting premium service--they are discontented because they're more aware of what they are missing.  

This change seems likely to have had the most impact on the middle and upper-middle classes:  people who could afford something beyond the necessities, but not at the premium level.  You have to be able to afford to fly before you're aware of all the extras you can't afford.   Since 1972, the General Social Survey has asked "Compared with American families in general, would you say your family income is far below average, below average, average, above average, or far above average?"  If people have reacted in the way that French suggests, the percent who see themselves as "average" or "above average" will have declined in the last decade or so.

The percent who say their income is average:

That's been declining pretty steadily over the whole period.  Now, (somewhat) above and below average.  

They both go up and down with short-term economic conditions--for example, below average rose and above average fell from 2008 to 2010.  But they both have upward trends, and the percent rating themselves as "above average" reached its highest level ever in 2024.  

Finally, far above and far below average:

They both have clear upward trends.  

The basic pattern is that there's more dispersion, but no change in the mean.  In a general way, this matches real changes in the distribution of income:  inequality has increased since the 1970s.  Although the increase has slowed down or stopped in the last 15 years or so, it may take time for people to become aware of that.  In any case, average ratings of your relative economic position haven't declined in recent years.

Although French identifies a real change, it doesn't seem to have affected people's perceptions of their economic position.  I think that's because it can work in both directions--people may look up at the people who get something better, but they may also look at those who get something worse and congratulate themselves at being able to afford an upgrade.  There are also people who could afford an upgrade but pass it up and congratulate themselves at getting a bargain.  

So we're still left with the paradox:  people think their own economic situation is pretty good, but that "the economy" is in bad shape.  Of course, this is just one possible comparison:  looking at other people today.   Another comparison that people talk about is with previous generations:  for example, where your parents were at the same age.  I'll look at that in my next post.

Whose idea was that?

A lot of the commentary after Paul Ehrlich's death said that his ideas were popular among liberals and/or "elites."  For example, Nicholas Eberstadt writes:  "In retrospect, what may look most amazing about Ehrlich’s career is the company he managed to keep. Despite his harsh and jarring rhetoric, his strident ideology, and his proclivity for veering off toward pseudo-science, Ehrlich was embraced into the bosom of the American academy. .... But perhaps this shouldn’t surprise at all. Though polemical and extreme in so many of his formulations, Ehrlich’s pronouncements on the human condition were largely in consonance with the moral panic about the 'population explosion' that swept through the American Establishment during the Cold War era."* But the New York Times obituary mentioned a detail that suggests a different possibility--Ehrlich was a frequent guest on Johnny Carson's Tonight Show.  So maybe his ideas were popular among the sort of people who watched the Tonight Show:  that is, a broad range of people.  

In 1974, a Gallup Poll asked "SOME PEOPLE FEEL THAT THE WORLD WILL REACH THE POINT SOMEDAY WHERE, BECAUSE OF POPULATION AND ECONOMIC GROWTH, THERE WON'T BE ENOUGH WATER, LAND, FOOD, AND OTHER NATURAL RESOURCES FOR EVERYBODY. OTHER PEOPLE BELIEVE THAT THE WORLD CAN CONTINUE TO GROW WITHOUT RUNNING INTO SERIOUS SHORTAGES BECAUSE SOMEBODY WILL ALWAYS BE ABLE TO SOLVE THESE PROBLEMS. DO YOU, YOURSELF, FEEL THAT SOONER OR LATER WORLD POPULATION AND ECONOMIC GROWTH WILL HAVE TO BE REGULATED TO AVOID SERIOUS SHORTAGES, OR NOT?"  62% said yes and 30% said no.  It was asked again in 1976:   65% said yes and 27% said no.  

Using the 1976 survey, here are "yes" answers by self-rated ideology 

Very liberal                                54%                
Moderately liberal                     76%
Middle of the road                     74%

Moderately conservative            66%
Very conservative                       58%
Don't know                                 70%

Support seems to have been somewhat higher in the middle (and those who didn't choose a label) and lower in the extremes.  Why wasn't there a straightforward relationship?  I think it's because there were two offsetting factors:  on the one hand, environmentalism was associated with concern about overpopulation; on the other hand, the Malthusian position suggested that trying to help poor people would be futile or harmful, giving it an affinity with conservatism, 

Not college graduate                      67%

College graduate                            79%

College graduates were more likely to think that regulation would be necessary.  But if we restrict it to whites:

Not college graduate                      73%

College graduate                            80%

The difference by education is smaller (and not statistically significant).  The reason that restricting it to whites makes a difference is that blacks were much less likely to think that regulation would be necessary (divided about 50/50) and less likely to be college graduates.  I considered a few other group differences:  men and younger people were a bit more likely to agree, and there were no clear differences by religion.  I'm not sure why race was so important--I just tried it because it's a standard control variable.  But the general point is that it wasn't just the "Establishment": most people were concerned about the "population explosion," to the point of supporting a policy that would now be regarded as pretty extreme. That's not hard to understand:  world population was growing rapidly, and people often think in terms of a fixed stock of resources.  

*As a sociologist, I have to note that this is an example of the dumbing down of the term "moral panic" to  be just a way of dismissing something as not a real problem.  In the original sense, the "moral" part was important.  

[Data from the Roper Center for Public Opinion Research]

Bad to worse?

In late 2022, I had a couple of posts about perceptions of moral conditions.  I concluded that assessments had become more negative in the 1960s, and after that there might have been some further decline.  It turns out that Adam Mastroianni and Daniel Gilbert were doing a similar study, which was published in Nature in 2023.  Their conclusions are completely different:

"A linear model indicated that the proportion of participants who reported moral decline was not significantly influenced by the year in which the survey was administered, b = 0.07, 95% confidence interval (CI) = [−0.11, 0.24], t(175) = 0.77, P = 0.45, adjusted R2 = −0.002, and the same model fit in a Bayesian framework indicated strong evidence of no effect (Bayes Factor of 0.04), which is to say that US Americans have been reporting moral decline at the same rate for as long as researchers have been asking them about it. (These and all tests we report are two-tailed)."

A problem with their model is that it omits a potentially important variable, or a lot of potentially important variables, depending on how you look at it--the specific question asked.*  Their sample of 177 cases includes over 70 distinct questions.  Some of these differences are small, but some are substantial.   For example, "Would you say that people are more willing, less willing, or about as willing to help their neighbors as they were twenty-five years ago?" and "In the last eight years, do you think crime has increased, decreased, or stayed about the same?" are clearly different questions, although they both involve the general topic of change in moral conditions.  So you should consider a model that includes dummy variables for the different questions in addition to the time trend.  The data set is not available to the public (the numbers are the property of the data archives), so I can't fit that model.  However, I can consider two questions that were asked frequently (data are available from the Roper Center).  One is "how satisfied are you with the direction that the country is going in at this time in terms of morals and ethics?...Very satisfied, somewhat satisfied, not very satisfied, not at all satisfied"; the other is "right now, do you think the state of moral values in this country as a whole is getting better or getting worse?"

  The figure shows positive minus negative responses to both questions over time.  For the first question, there is a clear downward trend (a t-ratio of about 10 if you regress the summary measure on time).  For the second, there's no clear trend, but the numbers are consistent with the hypothesis of a drop after 2004.  In addition to the trend, the responses for the first question show some short-term variation--e.g., opinions were more positive in October-December 2001 than in January-March 2001.  It's easy to think of an explanation for that.  

While I'm at it, here's the estimated assessment of moral conditions, adjusting for question, in the data set that I compiled.  It's not the same as the Mastroianni/Gilbert data, but there's some overlap--mine includes the "getting better or getting worse" question, but not "how satisfied are you with the direction."

*Another problem is the appeal to a "Bayesian framework" to claim that there is positive evidence of no change, but I've written about that at length here and here, so I'll leave that aside for now.

[Data from the Roper Center for Public Opinion Research]

No problem

 Adrian Vermeule, a professor at Harvard Law School, writes that "the Supreme Court faces a serious problem in the court of public opinion . . . . If the Court, having invalidated the President’s tariffs, also invalidates the birthright citizenship order . .   the Court will have invalidated the President’s two main or signature issues, on which he has campaigned since 2016 and twice won the Presidency."  

In the past year, a number of surveys have asked about birthright citizenship.  The wording varies, so I'll summarize by giving percent in favor of keeping birthright citizenship and percent in favor of eliminating it:

                     Keep     Eliminate
Jan 2025       61%          30%
Jan 2025       56%          43%
Feb 2025       55%          31%
Feb 2025       56%          39%
April 2025     67%          31%
May 2025      54%          28%
June 2025      74%          23%
June 2025      64%          31%
Nov 2025      72%          28%
Dec 2025       70%         24%

Average          63%        31%

All ten surveys showed a majority in favor of keeping birthright citizenship.  The narrowest margin (56%-43%) was for a question that mentioned Trump's executive order: "As you may know, Donald Trump signed an executive order arguing that children born in the United States are only US citizens by birth if they have at least one parent who is a US citizen or a legal permanent resident. Several states and outside groups have sued the Trump administration, arguing that there is a longstanding constitutional guarantee that children born in the US are automatically US citizens by birth. All in all, do you approve or disapprove of Trump's executive order limiting citizenship?"   The widest margin (72%-28%) was for "The Supreme Court is expected to hear arguments asking whether the 14th Amendment’s provision that those 'born in the United States, and subject to the jurisdiction thereof,' are U.S. citizens was intended to only apply to newly freed slaves after the Civil War and should not apply to a non-citizen‘s child who is born in the United States today. How do you think the Supreme Court should rule?"  There were some questions that just asked for opinions without giving arguments on either side:  for example, in May 2025 "Do you support or oppose . . . ending birthright citizenship, which makes anyone born in the United States a citizen" got 28% support and 54% opposed.  Although the small number of surveys and variation in wording means that there's a lot of uncertainty, support for birthright citizenship may have increased over the year.  But clearly the Supreme Court will have more trouble in the "court of public opinion" if it supports Trump than if it opposes him on this issue.  

I found only three questions on the subject before 2025--one from late 2024, one from 2023, and one from 2015.  This is relevent to Vermeule's claim that it is "one of his signature issues, on which he has campaigned since 2016."  The general idea of "getting tough" on illegal immigration was certainly a central part of Trump's appeal, but ending birthright citizenship was not a major issue.  The Trump Social Media Archive shows only one mention of birthright citizenship during the 2016 campaign (charging that Ted Cruz had changed positions, but not giving Trump's own position) and none in 2020.  There were a few in 2024 reposting articles supporting Trump's position, but nothing in his own words.  That is, he didn't campaign on the issue--he (or Stephen Miller) just decided to elevate it after his election.   This is part of a general pattern in which the second Trump administration has been more extreme than the first.  After January 6, 2021, mainstream Republicans temporarily distanced themselves from him, and the people who stuck with Trump during his exile have had a lot of influence in his second term.

[Data from the Roper Center for Public Opinion Research]  

Old and new

 I'll start with the new.  The SAVE Act requires proof of citizenship in order to register to vote.  If it were implemented, which party would it help?  The most convenient form of proof is a passport--of course, people who don't have a passport could assemble the necessary documentation, but it's safe to say that many of them would not get around to doing it.   There aren't many surveys that include questions about both passports and politics, but Pew had one in February 2016.  52.5% of Democrats and 46% of Republicans said they had a passport; independents were in between at 48%.  The survey didn't have any questions about anticipated vote in November, but it did ask who they wanted to get their party's nomination.  Among Republicans who had a passport, John Kasich led with 33%, followed by Ted Cruz with 23% and Donald Trump with 20% (the rest were undecided or scattered among other candidates).  Among Republicans without a passport, Trump led with 31%, then Cruz with 25%, and Kasich third with 20%.  The difference by passport status was smaller but still apparent after controlling for education.   That is, if this pattern held today, it's not just Republicans that would be more affected, but especially Trump's base.  I think the pattern would hold, because distrust of foreign countries is a central part of Trump's worldview, and people who distrust foreign countries are less likely to be interested in getting a passport.  

And now, following up on two older posts:

1.  In Munich, Alexandria Ocasio-Cortez said that high levels of inequality were to blame for the growing strength of authoritarian movements.  In December, I had a post about a paper by Eli Rau and Susan Stokes which found that income inequality was associated with a higher risk of "democratic erosion."  I found that this was still the case after controlling for another potential influence, presidential vs. parliamentary system.  Rau and Stokes noted that "democratic erosion" was becoming more common, and suggested that inequality could explain this trend:  they spoke of "the suspicion that rising inequality is playing a role in the wave of cases of democratic erosion."  However, they didn't directly look at changes in inequality.  This is the average of their measure of inequality by year (adjusting for changes in the nations in the sample):

It peaked around 2005, and in 2020 was lower than it had been in 1995.  Although nations with higher inequality seem to have higher risk of democratic erosion, inequality is not behind the rise in democratic erosion.

2.  My most recent post was about the difference between two presidential ratings:  one by political scientists, the other by a collection of conservative journalists, activists, and academics.  I compared ratings by the conservative (PragerU) sample to Republicans in the political science (APSA) sample and concluded that the differences weren't entirely due to ideology.  It occurred to me that I could improve the analysis by considering party differences in the APSA sample.  That is, maybe rankings depend on ideology and the PragerU sample was farther to the right than the APSA Republicans were.  You can estimate a model where the rating of each president by each group is the product of scores for the president and the sample.  If you do this, the difference between PragerU and the APSA Republicans is about 1.5 times as large as the difference between APSA Republicans and Democrats.  If you fit this model, the largest residuals (by absolute value) are for McKinley, Kennedy, GW Bush, Polk, Taft, and Madison.  McKinley, Bush, Taft and Polk get relatively better ratings from APSA Republicans, and Kennedy and Madison get relatively better ratings from PragerU.  For most presidents, the APSA Republican ratings are in between the APSA Democrats and PragerU; for these six, the PragerU ratings are in between the two APSA groups.  For example, LBJ got 74 from APSA Democrats, 55 from APSA Republicans, and 35 from PragerU; JFK got 66, 50, and 57.

In a general way, my conclusion that there was something beyond politics holds up, but my focus on Wilson and Coolidge was misplaced.  Basically, there's a big ideological/partisan difference in the rating of these presidents, which shows up within the APSA sample and between the APSA Republicans and PragerU.  As far as what that "something" is, I would say it reflects differences in knowledge.  For example, I gather that most experts regard Polk as an effective president, although there's disagreement about the merits of his policies (Republicans tend to be more favorable).  APSA Republicans would know that, while for many in the PragerU sample, he would just be another one of the nondescript run of presidents who came in between Jackson and Lincoln.   In my last post, I also said that the PragerU sample gave higher ratings to pre-1900 presidents; there's still some evidence for that, although it's weaker than in my previous analysis.

[Some data from the Roper Center for Public Opinion Research]