Skip to content

The Allure of Donald Trump

  • by

The primary season is now well underway, and the Trump bandwagon continues to gather pace. Like most observers, I thought it would run out of steam well before this stage. Trump delights in the kinds of vicious attacks and stupidities that would derail any other candidate. His lack of shame and indifference to truth give him a kind of imperviousness to criticism. His candidacy no longer seems funny: it now arouses more horror than humor for many observers. Given that Trump is so awful – so bereft of genuine ideas, of intelligence, and obviously of decency – what explains his poll numbers? 

While many Republicans disapprove of him strongly (a sufficient number, I suspect, to sink him when the field thins and the non-Trump vote is less divided among candidates), around one third support him. No doubt there are many reasons. Disillusionment with mainstream politicians (on both sides) is understandable, perhaps even warranted. So is despair at the loss of jobs in the US and the state of the economy. Trump’s very failure to offer specific proposals may allow people to project any fantasies they like onto him. And his scapegoating of minorities offers the white working class, which has seen real wages fall and unemployment rise, a target for their anger.

I want to suggest that in addition to all these factors, a psychological mechanism may be playing a role in explaining Trump’s attractions to a substantial minority of voters. Trump is a very wealthy man. By one primary indicator, that means he is a successful person. We are often unable to judge accurately what makes a person successful. For this reason, we are disposed to imitate them. We don’t (consciously) hope thereby to copy some of the behaviour which makes them successful: rather, we have evolved to have mechanisms that dispose us to imitate them because doing so raises the probability that we will imitate the behaviour that makes them successful. All kinds of things might play a role in causing success; sorting out what is cause and what is mere correlation is notoriously difficult. Better to imitate more widely: hence the evolution of our prestige bias.

It may be that the prestige bias helps to explain why celebrities are often paid to endorse consumer goods. No one consciously believes that wearing the same scent as Gwyneth Paltrow or the same watch as Brad Pitt will lead to movie contracts, or money, or being sexually irresistible, but we might nevertheless look more favourably on the products they endorse due to the fact that they are prestigious individuals. Similarly, I suggest, the fact that a successful businessman, one who repeatedly emphasises his wealth, puts forward what he claims are solutions for the many real problems confronting the white working class disposes them to give his words much more weight than they deserve.

It is worth noting that low information individuals – those with less capacity, due to limited resources or education – to identify what factors actually explain the success of individuals seem to be especially susceptible to the prestige bias. That makes evolutionary sense: if imitation is relatively low cost to the individual, then it is adaptive to copy broadly, even if for every behaviour copied there is only a small chance of thereby achieving success. If the prestige bias explains some of Trump’s attractiveness, then we ought to see stronger support for him among the less well educated and the less wealthy. And that’s precisely what we observe. No doubt many factors, rational and irrational, explain why people consider voting for Trump. But the prestige bias may explain a great deal of his attractiveness.

Share on

24 Comment on this post

  1. Sadly, The Rise of Trump provides an excellent opportunity for us to hone our skills regarding heuristics and biases. And it seems to me that your example of prestige bias aligns perfectly with Kahneman & Frederick’s definition of a heuristic: a process by which the brain assesses a target attribute by another property that comes more readily to mind. [Other definitions exist, but as long as we are going to be discussing prestige bias, I will quote a Nobel laureate in the hope that….]

    One can easily imagine other attributes that come to mind more readily than Trump’s policies. For example, he is arguably quick witted. And projects strength, even if comes in the form of bombast. And fearlessness, even if it arrives accompanied by foolishness. Each of these would fit as nicely as prestige in the narrative that you have put forth. Of course then, there is the outlier: why don’t people think FAKE FAKE FAKE when see that crazy hair?

    An alternative (but certainly not mutually exclusive) hypothesis has been put forward by Dilbert’s creator Doug Adams, who suggests that Trump is a ‘Master Persuader’, again using clever psychological principles but these derive from the illusionist’s handbook. Not only has he put forward this thesis, but over the months that he has been writing about it, he has been making a series of predictions about how Trump’s apparent gaffes would be interpreted as strengths, and as a scientist I am pretty impressed with the accuracy of his insights. For a full list of his posts on the topic, see his reading list http://blog.dilbert.com/post/139541975641/the-trump-master-persuader-index-and-reading-list.

    1. I must say, I avoid reading Scott Adams. His silliness on evolution and on gender infuriate me. They even spoil my reading of Dilbert!

  2. You’ve nailed it, Neil! I’ve been trying to figure out why he’s been so popular and I think you’ve provided a nice line of reasoning as to one of the major reasons this is so. Though, I wonder if you’re being overly optimistic of the masses in the states. Prior to considering prestige bias I was thinking more of implicit racial and/or racial bias? Given that he is clearly a racist and sexist, at least to my ears, do you think many gravitate toward him because when is saying it “like it is” he’s really saying sexist and racist things, and, given that many people do not see themselves as biased against women or minorities he provides them with the avenue to say the sorts of things that they often think but eschew in favor of political correctness?

    1. Justin, I think that his attraction is due to a number of different factors. I think that *explicit* racism and sexism play a role, given how outspoken he is (repeated refrain: “he’s just saying what everyone is thinking”). Here’s an excerpt from a recent New York Times article:

      In one segment, Mr. Smigel/Triumph hires “multiple robotic white women” to pose as Fox News reporters, asking voters to react to fictional Trump scandals, including a proposal to “sterilize Puerto Ricans until we ‘figure this thing out.’” “He’s just saying what everybody else believes,” says one supporter. Another, asked what he would say to people criticizing Mr. Trump for delivering “a three-minute monologue in an insulting Chinese accent,” blows a raspberry into the microphone.

  3. A more obvious reason may be that Trump amply satisfies the ‘anti-Washington’/’anti-Government’/’anti-Establishment’ animus that is very strong in the American community at present. Trump is clearly aware of, and knows how to capitalise on, this sentiment. This is another effect (as well as symptom) of the Tea Party’s pernicious influence on American politics. This ‘anti-Establishment’ attitude would also account for why Bernie Sanders is doing much better than expected against Hillary Clinton for the Democrat nomination (I don’t think the ‘prestige bias’ thesis works very well in the case of Sanders–or Jeremy Corbin in the UK, for that matter).

  4. I agree that’s a big part of it, Robert. In fact, I said so. And I agree that the hypothesis doesn’t explain Sanders or Corbyn. In all 3 cases, being anti-establishment might be a necessary condition of their appeal. But whereas the latter two offer something else to back up the appeal – actual policies – Trump doesn’t and seems not to need to. Further, the Sanders challenge is receding now, but Trump just won Nevada. Something else us needed to explain his appeal. I doubt it can be explained by the prestige bias plus being anti-establishment. The suggestion is that the prestige bias explains part of the attraction (probably by interacting with other factors).

  5. Populism is a major research theme in political science, but because of his background Neil Levy is probably unfamiliar with that work. I don’t think that ‘prestige bias’ would ever be accepted there, as the main explanatory factor for populist success. That said, outside perspectives are often useful, but this explanation seems to fail, on common-sense grounds alone.

    If the Trump voters are so easily influenced, the why doesn’t the Republican establishment simply put up another rich businessman to oppose him? The party has enough rich businessmen – probably more than any other political party in the world. Why don’t the Democrats also put up a rich businessman, to catch all the easily influenced ill-educated voters? And why doesn’t the prestige bias make those voters support Hillary Clinton, one of America’s most successful women?

    The prestige bias explanation belongs to a class of theories and pseudo-theories, which attribute quasi-hypnotic powers to individual politicians, especially those in other countries. These theories assume that voters are very easily misled, deceived, and influenced, and presume that there is no effective counter-strategy. Adolf Hitler allegedly mesmerised the Germans with his speeches, Slobodan Milosevic mesmerised the Serbs, and now ISIS ‘brainwashes’ teenagers into running away to become a jihadist or jihadi bride.

    I suspect this kind of explanation is popular, because it lets everyone else off the hook. Trump is seen a a jester leading the gullible peasants. No-one has to rethink their own positions and policies, which might be an uncomfortable exercise.

    In reality, the primary explanation for the success of Trump, is that he is a competent representative of a section of the population, that is currently unrepresented in the party spectrum. By ‘competent’ I mean that he has an organisation, is seen als potentially electable, and accurately reflects the concerns of his own voters. This is also the primary explanation for the rise of UKIP, the Front National, the Dansk Folkeparti, the AfD, Geert Wilders’ PVV, and so on. Their succes is essentially a restructuring of the party landscape, to conform more closely with the attitudes and aspirations of the electorate. For the same reason, we can expect that sooner or later there will be successful women’s parties in western democracies. Two-party systems tend to exclude new parties, but if enough voters are dissatisfied, then sooner or later there will be a shift in the political spectrum.

  6. If you’d like a response, Paul, please address your points to the actual claims made. Do not accuse me of claiming that prestige bias is the main explanatory factor, for instance.

    1. That is what Neil Levy wrote:

      But the prestige bias may explain a great deal of his attractiveness.

      He makes other claims about Trump, which would not be accepted as explanatory by political scientists. I don’t take that as an infallible norm, because most political scientists are also dismissive of populist movements, which they see as short-term disruptions of ‘normal politics’. Nevertheless, simply disparaging Trump and his voters can not offer an alternative explanation for his success, or for comparable populist success in other countries.

  7. Fair enough, Paul. Now show me where I disparaged Trump voters. We are all subject to heuristics and biases. They are an indispensable part of reasoning. They can also lead us astray

    1. Neil Levy wrote that the ‘white working class’ give Trump’s words more weight than they deserve because he is ” a successful businessman, one who repeatedly emphasises his wealth”. He describes these voters further as low information individuals especially susceptible to the prestige bias. That is effectively saying that they are stupid, and certainly no evidence is offered for their status as ‘low information individuals’. As it stands, this is not a hypothesis that political science could test. Even the term ‘white working class’ is misplaced here: it is British usage, and has specific cultural connotations in Britain (white-van man, UKIP, football, and so on)., which are not relevant in the US.

      1. What’s with the weird third personal, Paul? I won’t reply again; its pointless. Why would my usage of “white working class” carry Britain-specific connotations (I am not British, by the way). I used white working class in its correct sociological meaning. There is plenty of evidence that Trump voters are indeed low information voters. I don’t need to test it: I can rely on the existing data. That doesn’t entail stupidity. Nor does vulnerability to the prestige bias entail stupidity. As a matter of fact, general intelligence is not protective against biases generally (under some conditions it actually makes people more biased, by giving them greater ability to rationalise their intuitions).

  8. The author of the article may be correct in his over all theory. but he weakness his case
    by his own bias.

    Mr. Trump most certainly does have policy positions. They are brutal and simple. Build a wall,
    is in fact a policy position. The author may not like the policy, and find it abhorrent. That is quite different
    than saying none exist.

    I suggest part of Trump’s appeal is his policies are simple, and direct, you might start your analysis there.

    1. I freely admit to being biased against Trump. That bias strikes me as a mark of sanity. Trump has some ideas that might just barely count as policies (like the wall), but on many issues he simply says he will fix things. Here is his stated policy on health care, for instance: ““I would end Obamacare and replace it with something terrific, for far less money for the country and for the people”. He is the King Lear of the primaries:

      I will do such things,–
      What they are, yet I know not: but they shall be
      The terrors of the earth.

      1. Why ending Obamacare and replacing it with some other less expensive solution would be so bad? If he said he would implement free healthcare for all at the point of need, then it seems evidence suggests (see endnote) we should be worried. I’m genuinely interested in your response here given that a lot of what you write is so empirically driven.
        Also, we have no evidence being rich does not track implementing good policies better than having good policies as a candidate. Wealth correlates strongly with IQ, which has a strong correlation with overall success across all domains. Of course, he might not want to implement good policies.
        I think Paul Treanor was on track with some of his remarks about Trump rightfully representing a significant part of the eleitorade (although, yes, the way he states things is borderline trollish). The big problem here is why the other part of the electorate is not reacting, and I do fear for the future political stability of the US if they decide to respond after Trump is elected. I wonder what sort of biases has made the American so passive to the possibility of having Trump as a president. Is there a “this must be a joke” bias? People need to start realising it is not a joke.

        http://www.rand.org/pubs/reports/R3055.html
        http://annals.org/article.aspx?articleid=716066
        http://economics.mit.edu/files/10580
        http://www.cato-unbound.org/2007/09/10/robin-hanson/cut-medicine-half

        1. I am absolutely sure that Obamacare is not optimal. But I am equally sure that Trump has nothing better proposed. Indeed he has effectively nothing proposed. Wealth does correlate positively with IQ, but not strongly enough to license inderences about individuals. In this case, we have much better, countervailing evidence. I don’t what you mean when you say he represents a significant part of the electorate. His polling numbers are not in dispute.

          1. I believe the correlation with income is strong enough to license inferences about individuals, most studies report around 0.23 and others up to 0.5 correlation. (The problem is that in the case of Trump he hasn’t produced any wealth more than if he had put his inheritance in an index fund.) Wealth correlates with a better educational background as well and many other things which presumably would make a better leader. It seems one can be licensed to interpret the uneducated voting for Trump because he is wealthy either as a bias or as a rational heuristic that makes a lot of sense in the absence of higher quality information. (I personally have doubts how much higher quality information there really is available to anyone.) That leads to my general point regarding his polling numbers, which is that I don’t think it is helpful to frame Trump’s polling numbers as people being mislead or biased.

            I have few disagreements with anything you claimed in the post. However, I’ve seen an abundance of explanations of Trump’s numbers based on his voters being mislead while his numbers keep going up. Perhaps there’s some level of causation between those two. An analysis focused on how his voters might be right could be more fruitful at this point.

            1. I think you will find that more recent studies tend to show a correlation smaller than 0.23. But in any case, there is absolutely no evidence that discriminates between the wealth of Cruz, Bush and Trump. Even Rubio, with a much lower net worth, is probably in the income bracket that is the highest for which we have data. And in fact there is every theoretical reason to expect that variance above $100,000 will be explained by lots of other factors.

              1. A large 2006 meta-analysis concluded the correlation was .23 and IQ’s predictive power the highest among other predictors:
                “The overall correlations were .56 (between intelligence and education), .43 (between intelligence and occupation), and .20 (between intelligence and income). Exclusion of the samples that were too old (over 18) at the time of testing or too young (below 30) at the measurement of success resulted in somewhat larger correlations: .56, .45, and .23, respectively. (…) Meta-analysis demonstrated that parental SES and academic performance are indeed positively related to career success but the predictive power of these variables is not stronger than that of intelligence (see Table 1). In fact, intelligence exhibited several correlations with the measures of success that were larger than the respective correlations for other predictors suggesting that intelligence is, after all, a better predictor of success.”
                http://www.emilkirkegaard.dk/en/wp-content/uploads/Intelligence-and-socioeconomic-success-A-meta-analytic-review-of-longitudinal-research.pdf

                Regarding the correlation between IQ and income at the top of the distribution, there are a lot of studies showing the correlation holds strong (even stronger sometimes) there. Here’s a few I had in hand:

                “It is remarkable that even in this very high-IQ sample, where the range of observed IQs is clearly restricted, IQ still has a positive and statistically highly significant association with lifetime earnings.”
                http://www.iza.org/conference_files/CoNoCoSk2011/gensowski_m6556.pdf

                “Using 174 studies of the relationship between 9 scales of the General Aptitude Test Battery and job performance, with a mean sample size of 210, and using a power polynomial approach, which has been shown to have higher power than the r vs. eta test, the issue of linearity was reexamined. Similar to previous findings, nonlinear relationships were not found at levels substantially greater than would be expected by chance.”
                http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=9406160022&site=ehost-live

                “… the data reported here on secured doctorates, math–science PhDs, income, patents, and tenure track positions at top U.S. universities collectively falsify the idea that after a certain point more ability does not matter. (…) When sample sizes are sufficient to establish statistical confidence and criteria with high ceilings are employed, measures that validly assess individual differences within the top 1% of ability reveal important outcome differences between the able and the exceptionally able (even on outcomes that are exceedingly rare). A recent 20-year longitudinal study of 380 profoundly gifted participants (Lubinski et al., in press), the top 1 in 10,000 on quantitative or verbal reasoning (viz, SAT–M 700 or SAT–V 630, before age 13), reinforces this idea. These participants, by their mid-30s, secured tenure track positions at top U.S. universities at the same rate as a comparison group of 586 1st and 2nd-year graduate students attending top-15 math–science training programs and tracked for 10 years.”
                https://www.psychologytoday.com/files/attachments/56143/creativity-and-occupational-accomplishments-among-intellectually-precocious-youths.pdf

                Here’s an image using data from the above study and for other studies for lower IQ values: https://pumpkinperson.files.wordpress.com/2014/11/dollarbar.png

                “Empirical research fails to substantiate the threshold hypothesis (Park, Lubinski, & Benbow, 2007, 2008; Sackett Borneman, & Connelly, 2008). Studies cited in support of the threshold hypothesis typically suffer from at least one of several methodological weaknesses that reduce statistical power and artificially attenuate the relationship between ability and accomplishment at exceptional levels. When a study lacks sufficient statistical power to detect a real relationship between cognitive ability and outcome variables above this threshold, some researchers mistakenly interpret their inability to find a significant relationship between cognitive ability and outcome variables as evidence supporting the threshold hypothesis. Figure 1 is based on data from Park et al. (2007, 2008) that falsifies the threshold hypothesis. In this study, nearly 2,500 participants took the SAT-Math by age 13 as part of a talent search. Figure 1 presents outcomes they achieved over the following 25 years. It shows that ability differences matter among participants in the top 1% in mathematical ability. For example, in the bottom quartile (Q1) of the top 1% in Figure 1, 15.4% have earned a doctorate (a PhD, JD, or MD). But the corresponding proportion for the top quartile (Q4) is 33.2%: more than twice as many people in the top quartile have earned a doctorate as in the bottom quartile!”
                https://my.vanderbilt.edu/smpy/files/2013/02/Kell-Lubinski-Benbow-20131.pdf

                “Even within a group in the top 0.0000001% of wealth and a group of CEOs who were compensated quite highly (well within the top 1% of wealth), there were differences in the education and ability level between those who earned more money compared to those who earned less. The analyses in Table 3a and b demonstrate that even within billionaires and CEOs, higher education and ability level is related to higher net worth and compensation. Prior research demonstrated that even within a group in the top 1% in ability, higher ability is associated with higher income (Wai et al., 2005). The analyses in Table 3c demonstrated that even within the top 1% of ability, higher ability is associated with higher net worth and compensation. Therefore, this study adds to, expands, and strengthens the literature linking education, ability, and wealth (Murray, 1998; Nyborg & Jensen, 2001; Zax & Rees, 2002), and provides further evidence that does not support an ability threshold hypothesis (Kuncel & Hezlett, 2010; Park et al., 2007; Wai et al., 2005) — or the idea that more ability does not matter beyond a certain point in predicting real world outcomes.”
                https://www.psychologytoday.com/files/attachments/56143/wai-americas-elite-2013.pdf

              2. As I said, more recent studies find a lower correlation than .23. Except for your last reference your citations are silent on income differences that discriminate between Cruz, Bush and Trump. The last use college as a proxy for g, and finds that billionaires are more likely to go to the most elite schools. I don’t think that’s a good proxy at all, especially in the absence of an attempt to control for legacies and parents’ income.

                1. But all of this is completely beside the point. Why argue about whether great wealth like that possessed by Trump is evidence for high LA, when we have the pronouncements of the man himself to go on? It’s like someone doubting that Einstein was intelligent because they haven’t heard of the Zürich Polytechnic. Proxies should drop out when direct evidence is available.

    2. “Mr. Trump most certainly does have policy positions. They are brutal and simple.”

      And implausible, ineffective, and politically untenable. A former CIA Director thinks the Military might disobey orders from Trump, if (as Trump has indicated) those orders would amount to war crimes (torture; murdering family members of enemy combatants). As President Trump may be more or less idiotic than his current rhetoric suggests. Grievance politicians frequently exercise power more responsibly than they threaten to, and there may be particular reasons why a billionaire may be content with tinkering rather than with radicalism. (On the basis that radical experiments usually have unpredictable outcomes.) But Neil is right that most of Trump’s stuff is not policy – he shows very little grasp of policy – but rather emotional outbursts on policy issues, which is something else entirely.

      The Primaries have shown (again – we’ve seen this movie before in Europe and Latin America) that you don’t beat populists through superior policy ideas. One point of Neil’s post is the reasonable point that some voters don’t know about policy, and hence aren’t well-positioned to vote on the best ideas. In a world of variable talents, this is unsurprising. Another reasonable point he made is that, to those non-policy voters, Trump is pitching his “prestige” as a proxy for his suitability. Hence Trump’s hostility when Rubio claimed that Trump was less successful than he should have been, given his priviliged start in life. This is a promising line for anti-Trumpers because, in one sweep, it both attacks a source of his appeal, and points out the social distance between Trump and his voters.

  9. Trump has clear policies, which certainly on immigration correspond to the positions of populist parties in Europe.

    We need to control the admission of new low-earning workers in order to: help wages grow, get teenagers back to work, aid minorities’ rise into the middle class, help schools and communities falling behind, and to ensure our immigrant members of the national family become part of the American dream.

    He does not have a full policy platform, but that is not necessary for a populist party to be successful. It only needs to address the issues which anger its electorate. And that is the reality which needs to be considered: a section of the population is angry at mainstream politics, in Europe we would say ‘the political class’, for ignoring their concerns and fears. The denialist position which Neil Levy advances, in effect tells the political class that they can continue to ignore these voters. I think that is a dangerous strategy.

Comments are closed.