Do influential people develop more conventional opinions?

Following up on Robin's question I think the answer is yes, mostly. We are talking about the time series here, as people rise in influence.  I see a few mechanisms:

1. People "sell out" to become more influential.

2. As people become more influential, they are less interested in offending their new status quo-oriented friends.

3. As people become more influential, their opinion of the status quo rises, because they see it rewarding them and thus meritorious.

4. The status quo is good at spotting interesting, unusual people who will evolve (sell out?) and elevating them to positions of influence.

5. Oddballs who are influential arrive first at where the status quo is later headed, and eventually they end up looking conventional.

6. Influential people are asked to write increasingly on general interest topics ("How to Be Nice to Dogs") and thus they find it harder to be truly unconventional.  They cultivate skills of conventionality because that is what they are paid for or allowed to express.

Can you think of other mechanisms?  Who are some test cases for these hypotheses?


The whole academic literature on financial analyst behavior confirm most of the mechanism you spotted.

As people become more influential, the status quo moves towards the positions they originally had, making their opinions more conventional.

You can tell this story causally or not, with the influence of the individual being either phenomenal (the cause of the change in the status quo) or epiphenomenal (with their increased influence being a result of the status quo shifting towards their position).

Perhaps the way to become influential has something to do with how to seem deep.

"The part I experienced as effortful was picking a response understandable in one inferential step and then phrasing it for maximum impact."

It argues that in order to strike a chord you have to be a step ahead of those listening to you. Not several steps and not right next to them, but one inferential step ahead. If you want to maximize your influence, you cannot, thus, argue far from the status quo.

Perhaps the following is true (p=0.3)

I think that some artists and musicians are good examples. They are original and sometimes do revolutionary stuff when young and poor but become repetitive and boring once they become rich and bourgeois, dating celebrities and populating cocktail parties. The artists that remain doing original and exciting stuff are often those that in some way or another reject the mainstream social life and conventions (e.g. by doing lots of drugs).

I believe there is a good paper on this topic. It explains why artists sell out as they become famous and the theory applies to thinkers as well!

ABSTRACT: Artists face choices between the pecuniary benefits of selling to the market and the non-pecuniary benefits of creating to please their own tastes. We examine how changes in wages, lump sum income, and capital-labor ratios affect the artist's pursuit of self-satisfaction versus market sales. Using our model of labor supply as a guide, we consider the economic forces behind the high/low culture split, why some artistic media offer greater scope for the avant-garde than others, why so many artists dislike the market, and how economic growth and taxation affect the quantity and form of different kinds of art.

Cowen, Tyler and Alexander Tabarrok. 2000. An Economic Theory of Avant-Garde and Popular Art, or High and Low Culture. Southern Economic Journal 67(2): 232-253.

It takes a while for really good ideas to permeate and get accepted.

So you have to spend some time being a cheerleader for your ideas, which also takes time away from developing another really good idea.

And in the end, nobody wants to challenge you, they just want some more good ideas.

7) People more rigorously self-censor their less conventional opinions as they become more influential, because they perceive the negative consequences (personal and societal) of being incorrect are greater.

You can test your hypothesis # 1 with the recent transformation of Larry Summers. You can start your test with a detailed analysis of his speech last week at the Brooking Institution. In my view it's a total "sellout" of his integrity and expertise. G. Mankiw has already pointed out some of his reversals, but there are several others that are worse.

show me an influential person and i will show you a person who trades in the moral outrage of the masses.

E. Barandian,

"total transformation of Larry Summers," what? Well, in his speech
at Brookings he left behind his more critical views of labor unions from
some of his research of about 20 years ago, but other than that, I am
hard pressed to think of anything else that Summers has publicly said
or supported since becoming a top adviser to Obama that is obviously
out of line with his previous views. Keep in mind that we are not
talking about someone who has ever been particularly unconventional or
wildly innovative, although without doubt brilliant as well as very
self-confident to the point of overweening arrogance. And as for being
influential and powerful, he was Chief Economist of the World Bank in
the early 1990s, served in the Treasury ending up as Secretary under
President Clinton, and then was President of Harvard after that. If
he "sold out" to become influential it was probably more like 20 years

More generally, I would say that Tyler's list is pretty good, and I
can think of many people who would fit each of the categories, with
many fitting several of them at a time. Probably the one that has
the least overlap with the others is #5, which I find kind of cool.

I think pat and Scott above have the right point. When you're young and unknown, you can take more risks--sometimes, you'll look like a fool, but hey, everyone does that sometimes when they're young. When you're established and well-known, the risk of looking like a fool gets higher, both because being a fool is less tolerable (you're supposed to be an expert in this area, how could you say something so dumb?) and because you have more at stake (not a possible career in the field if you do everything right, but an existing well-established career in the field).

I suspect we would benefit from more mechanisms to make it easier for innovative, established people to continue to be iconoclastic. Other than tenure, I'm not sure what that would be.

Your anti-spam verification was impassible for me using either Firefox or Camino on Mac OS/X. In Firefox (with a flash blocker), I couldn't even see the challenge, in Camino, I could see it, but the page refused to accept my comment unless I turned on Javascript. Is there any way to fix this?

I can think of some exceptions, the current Pope being one. If anything his criticism of the status quo (both inside and outside of the Church) has increased with his influence. You'd think 1 would likely for a Pope, but 2 or 3 are not important for people in truly powerful positions. In fact, if your power is secure, destroying the status quo increases your power relative to your rivals.

Isn't it just that influential means you are conventional opinion. The marginal person doesn't actually set opinion on the majority of topics, they just agree or disagree with the people who do. Once you are influential, you are a person they agree with, so you ARE conventional opinion.

E.g. I doubt artists stop being creative or cease to acquire new skills, people just expect it from them - so they are the trend and their idiosyncrasies the modern style of the art form.

It would be nice to think the marginal person has some input in setting convention, but in most areas they simply don't and just choose who to follow.

Which is also, really why people don't like censorship in academia or insiders dictating what is high art isn't it? Because the only time influential people have input in the process is when new influential people are chosen.

E.g. How many people were willing to endorse pedophilia after Allen Ginsberg endorsed it.

Let's face it, had the system continued to make money Cramer would still be considered the norm, albeit not unconventional.

I agree that they do, in fact, offer more conventional opinions. But I also think that even if they didn't, people would perceive them to.

This is because when people are obscure, they are likely only to be noticed for their more outrageous or at least novel claims. When they become famous, people will look to them even on mundane matters, where their view might be consistent with the mainstream consensus. This might be the same view they had before, but before they got well-known, no one was asking.

To influence people, meaning, to lead people, you have to tell them what they're expecting to hear about 90% of the time. That makes them trust you. After that, they'll allow themselves to be led in an unexpected direction about 10% of the time. So if your goal is to lead people in a new direction, step one is to spend a fair amount of time urging them to accept things that are universally accepted truths.

E. Barandian,

First of all, arguing that markets are not always stable is in fact one
of the few somewhat unconventional things that Summers wrote about back
in the late 1980s and early 1990s, most famously in some very influential
papers on speculative bubbles that were coauthored with DeLong, Shleifer,
and Waldman. I did not read his speech in full detail, so I do not know
if he used the cobweb theorem or not. I was under the impression that
he was falling back more on his own old work on speculative bubbles,
which does not at all depend on the cobweb theorem. You are not going
to try and tell us that there are no speculative bubbles and crashes
because that would mean that market participants are stupid and we
know that they are not, are you?

BTW, it is an empirical fact that there is a about a six year corn-hog
cycle and a fourteen year cattle cycle, the latter more clearly and
definitively in the data than the former, with most of those studying
this coming out of ag econ departments. Most would attribute these to
the cobweb theorem operating, even though John Muth cooked up rational
expectations to supposedly "disprove" the cobweb theorem. But are you
one of those foolish enough to be asserting that rational expectations
holds given what has been going on in the markets more generally recently?
Last time I saw anybody arguing for ratex it was based on how the
consistent underforecasts by agents throughout the 1970s were offset by
the consistent overforecasts of inflation throughout the 1980s. What a
pathetic joke.

A correction to my last comment. I have just checked up on the corn-hog and
cattle cycles. The hog cycle is more like four years long rather than six, and
the cattle cycle is about ten-twelve years. However, both are quite persistent
in the data, and the cobweb theorem continues to be invoked, particularly in
regard to the corn-hog cycle.

Check your data, E. Barandian, before unloading so resolutely.

Conventional opinions are more likely to be right because they have stood the test of time; For example, summer is warmer than winter.

Conventional opinions are usually defined as those of an older generation, which has seen more and so naturally takes more facts into account; This accounts for some convergence in thinking, matching their shared histories.

Younger people know only their world but know it better. In times of change, the young know the future better, but that soon becomes conventional.

Here's another theory:

7. Most people have very few unconventional ideas. When they articulate an unconventional idea, this often leads to them becoming influential. It does not necessarily lead to them having subsequent unconventional ideas.

A lot of great theories have been listed here. The next step is to try to see what behavioral details might distinguish these theories.

one becomes influential by having certain opinions--not that you change ('sell out'), but you ascend to a certain social position precisely because you are accepted by the opinions that are accepted at the time.

i am sure one could come up with a list of academic/government types who fit this model.

I think volume and quality of feedback is a factor.

If a 20 year old student comes up with an unconventional opinion: we should introduce capital punishment for companies that go bankrupt
Ta will be largely ignored. Some other students may call him an idiot.

If the minister for trade and industry comes up with the same statement, ta will receive a great deal of feedback about why this is a bad idea. This will include educated, fact laden, well thought out opinion, as well as media crap.

Even if both people have exactly the same knowledge, opinion and reasoning before hand, the minister will be far more likely to moderate the opinion.

Test cases for these hypotheses:

David Gregory

David Broder

Eliezer Yudkowsky,

Very nice comment. Won't your test be hard to really do? Even if we had a good data set etc., low-status women will express fewer unconventional opinions than will low-status men. Thus, you just don't have have that much variation to work with, do you?

I think Jonathan and JP nailed it. Influence brings greater knowledge, and greater knowledge brings a new understanding of the problems and challenges we face. That the traditional solutions are not always wrong any more than they are always right. Learning to distinguish between the two without prejudice is, in my mind, a large part of maturing.

Another possibility is that older people are more conventional, and that it takes time for people to develop their influence, so influential people tend to be older.

Then, you would think that younger influential people would be more likely to be more unconventional than older influential people, which I think is probably true.

I agree that this is related to JP's comment above, in that there is a process of learning that conventions exist for a reason (sometimes a good reason).

It could also be that as our brain ages, it becomes less able to learn truly new things.

I agree somewhat, but there are major exceptions. Some influencers are influencers because they stay ahead of the curve. So ... long before #5 happens and these influencers begin to look conventional, they are off and running to the Next Big Thing.

Heh, I'm number five.

Comments for this post are closed