Tuesday, January 23, 2018

Meet Jordan Peterson. Or On the Channel 4 Interview About Pay Gap.

Jordan Peterson is a clinical psychologist, a professor of psychology in Canada, and the iconic guru of a new self-improvement movement that has a very large mostly male following (1).

He has come out with a book: 12 Rules for Life.  An Antidote to Chaos.  It is going to sell well. Google tells me that many areas already have men's groups based on Peterson's ideas, and the viewership of his YouTube videos is usually in several hundreds of thousands and more.  The comments attached to the videos are full of gratitude and adoration.  Clearly the advice Peterson gives has helped many, though it seems that it is men his message is aimed at and it is men that it largely seems to truly affect (2).

I had never heard about him until I came across a piece on the online harassment a female television interviewer stated that she received after a bout with Peterson concerning the messages of his new book.  That made me watch the interview and then spend several hours online trying to understand the psyche of professor Peterson.

And fascinating he is!  An excellent speaker, telling his followers to fix their lives in all sorts of perfectly innocent-seeming and wholesome ways.  A self-defined traditionalist, a firm believer in biological sex differences in lots of stuff (he is a supporter of Damore),  a firm believer in hierarchies as the weapon against chaos and also a firm believer that hierarchies are not based on power but largely on competence, and that status is the due reward for competence (3).

It's possible for someone like me (a feminine girly goddess with no brain and no logical ability, but lots and lots of agreeableness, bares fangs) to enjoy his philosophical arguments, to agree with some of them, to feel persuaded about others, to sharpen my counterarguments with others,  and then — suddenly, like a lightning out of a blue sky — to feel truly frightened by that shadow side of his which just might be what motivates his public speaking on these issues.

And that is his argument that the far-left in American and Canadian universities is going to take over the world, using postmodernism as its weapon.  Postmodernists, Peterson believes, are Marxists, and the radical far left in universities has already destroyed the humanities and social science departments and will go on to destroy the Western world.  Millions and millions will die, just as they did when other authoritarians gained power: Mao, Stalin and Lenin.

Peterson argues that the radical far left is authoritarian and focused on group identity politics, whereas he is an individualist, except that he believes that women are largely held back by their own biological (?) preferences for staying at home and for lower-paid jobs, and in case if they are not already quite convinced by those instinctual nudges, he reminds us of the great importance of breast-feeding, the mother-and-infant connection and so on.

A fascinating man, and quite fun to study.  But one blog post is not long enough to address all the sides of professor Peterson, his Jungian side, his Christian side and so on.

Rather, I want to use the rest of this post to analyze the Channel 4 interview with him in Britain.  The interviewer is called Cathy Newman.  She was unprepared for debating someone who had spent over a decade honing his arguments and someone who was willing to use words such as "multivariate analysis" while talking to a layperson (4).  That's why Peterson "won" the interview, though whether his arguments have more inherent validity is up for further evaluation.

Let's do the latter.  Here is the interview:

I have picked three issues to analyze.  I could have done more issues, but space is limited, and these three show what he leaves out when he argues that the gender pay gap is not that much to do with discrimination, that the leaders of the largest firms are there largely because they work so very hard (and women could choose to do that, too) and that the markets are not dominated by men because women carry out 80% of consumer decisions.

First, from 5:23 on, the two discuss the wage gap between men and women in the UK.  He makes a correct point, the same I have often made, and that is the importance of not comparing gross earnings differences between men and women but net earnings differences.  By the former I mean the actual average earnings that are reported per day, week or month, and by the latter I mean the earnings differences which still remain after we control for any differences in education, experience, time worked per wider time unit and so on.

The controls are necessary, because many different variables affect earnings.  More educated and more experienced people, for instance, earn more, ceteris paribus,  and so on.  The multivariate analyses of various types allow us to hold those other variables constant while seeing what earnings gap still remains.  That gap may be due to labor market discrimination or it may be due to omitted non-discriminatory factors.

But it's also possible, when adding explanatory variables to such analyses, that we over-control for other variables, thereby biasing the calculations.  Take adding "occupation" as one of the controlling variables when measuring net earnings differences.  Adding it makes sense, right?  After all, different occupations pay different amounts.

But what happens if women are not hired for certain occupations at all?  Or if they are driven out of them with sexual or other type of workplace harassment?  Then controlling for occupation will understate the true impact of discrimination. 

Then there's the problem of explanatory variables which are correlated within the sample.  Suppose that you decided to control for the length of a worker's hair in the analyses and suppose, then, that you found this variable explained some amount of the gross earnings gap.

How would you interpret it?  After all, in this culture women, on average, have longer hair than men, on average, and the reasons for paying less to long-haired men might be completely different from the reasons for paying less to women, most of whom have longer hair than men.

Peterson uses "agreeableness" in this manner, arguing that women are more likely to be agreeable than men and that agreeable people are paid less. He  seems to believe that greater agreeableness is an inherited sex-linked biological trait, though he also points out that the actual differences in  average self-reported (!) agreeableness between men and women are very small.  But if agreeableness is not as much biological but what we inculcate in our daughters, then in some wider sense this variable is also correlated with societal sex roles.

I was unable to find the multivariate analysis that Peterson used for his agreeableness comment, so I can't tell you what the remaining unexplained difference between male and female workers in that study might have been.  Peterson didn't tell us, but there might be one.

Second,  at 12:23 Newman asks Peterson what his explanation for the great scarcity of women in the very top business posts might be.

Peterson's answer to that is a textbook conservative answer to similar questions:  A few (very few) very hard-working, ambitious, talented and single-minded men are willing to sacrifice everything in striving for those posts.  Women could do the same if they chose to do so.

That is the "it's-your-choice-but-you-don't-actually-have-a-choice-because-your-biology-makes-you-want-children-and-men-won't-mind-them" generic conservative response.  I have written about it many times before.

What it leaves out are all the other factors which determine who will end up leading the largest corporations.  Though Newman uses British data, the Fortune 500 data from the US looks very similar.  In mid-2017,  6.4% of the Fortune 500 CEOs were women.

But there's more:

The Most Powerful Women team reported on Wednesday that 32 of the CEOs on this year’s list are women. It’s an historic high-water mark that pushes their representation to 6.4%.
Seventy-three percent of the senior executives, men and women, are white. The rest are 21% Asian, 3% Latino/a, 2% black, 0.6% two or more races, 0.2% Native American and 0.1% Native Hawaiian or Pacific Islander.
Compared to the demographics of the overall employed workforce, Asian and white workers at these 16 companies are overrepresented in senior leadership by 15 and 10 percentage points, respectively. Latino/a and black executives are underrepresented by 9 and 13 percentage points.
People who represent two or more races are three times as common in the overall workforce as they are in senior executive roles. Meanwhile, Native Americans and Hawaiians or Pacific Islanders are underrepresented by 0.3 and 0.4 percentage points.

Peterson's thesis would argue that for some reason white and Asian men are much more likely to be willing to pay the heavy price to climb to the top of these hierarchies than Latinos or African-American men.

There's even more:  It's quite possible that some fraction of the top leaders of the business world were born with all that money or even inherited their CEOship.

And, finally, Peterson says nothing about the very real possibility that women face additional hurdles on their way up the hierarchy, and not just hurdles having to do with the sexual division of labor about children and household responsibilities.

By omitting those arguments and by sparring against someone unprepared  in a particular research field he comes across as having won the argument.  But the true argument is much, much longer and more complicated.  Even multivariate!

Third, at roughly 29:50, Newman and Peter debate the markets.  Peterson argues that markets are what rewards masculine characteristics in the managers of firms, Newman states that markets are controlled by men.  Peterson hits back by stating that 80% of consumer decisions are made by women.

Another victory for Peterson, right?  Except for two things:

First, markets have two sides, and the supply sides of most markets are, indeed, still largely controlled by men, though what such control might mean probably looks very different in oligopolies and more competitive industries.  Still, the relevant market here is the market for hiring and promotion people, all the way to CEOship, not some general market for, say, breakfast cereal.

Second, that 80% consumer decision-making that women do is one I have met earlier on various manosphere sites, and it's always used to suggest that women control the whole economy.

The last time I came across it I spent a lot of time trying to find the study it came from.  It seems that the 80% figure is from some marketing study. Women do much more daily grocery shopping, tend to buy the children's clothes and perform most routine shopping tasks for families which also may contain men.  That's because the job of shopping is still to some extent assigned to women inside families.

In other words, that women carry out the actual shopping does not mean that they alone (or at all) determine what kind of car the family drives, what kind of dwelling the family has, what kind of electronics the family uses and so on.

That concludes my off-the-cuff comments to the three points I picked from the interview, though I could have picked more examples (5).

Jordan Peterson is going to do very well indeed.  He is a great speaker, he has spent years honing his shtick, and he offers both individual online "therapy" with many good points which might be applicable to everyone, and also a defense of traditional masculinity and social conservatism based on biological (innate) differences between men and women (6). 

(1) I watched one very long lecture in which he gives his views on white privilege (he believes that the argument is poorly sourced and is, in any case, majority privilege, though he doesn't address male privilege the same way).   The question and answer session at the end had zero women asking questions but several serious and conservatively dressed white young men.  The bad lighting in that room makes it hard to say if there's any racial or much gender diversity among the audience, but my impression is that there is not.

It's worth noting that white supremacists like him quite a bit, but that doesn't mean that he likes them.  Still, one famous neo-Nazi liked an interview of Peterson by Stefan Molyneux in which, according to the linked article,  Peterson suggested that political correctness is caused by the maternal instinct running amok:

In some sense it doesn't much matter if Peterson hates the Alt Right.  They seem to like him, and so does the more traditional right.

(2)  I haven't read his book yet, so the following is a very preliminary impression:  My guess is that Peterson fights for the values of traditional masculinity.

(3)  Another very preliminary observation:  Peterson argues that lobsters have hierarchy and that therefore there is no such thing as patriarchy, because other animals, even those quite different from us, have hierarchies.  That seems to confuse the question whether humans are likely to have hierarchies with the question of how the different rungs on those ladders are or should be staffed.  But elsewhere he seems to regard hierarchies as the "order" which fights against "chaos."

(4)  Conor Friedman in the Atlantic has a somewhat different take from that.  He urges for a calm and nuanced debate and careful listening to the other side's messages.  That would be a very good idea in all political debating, of course.

Not to defend Newman's unpreparedness, but there's a slight difference in debating something like the way national income statistics should be kept and the question whether people with Conor Friedman's genes can ever hope to climb any hierarchies.   Or earn equal salaries, on average, with people who possess Echidne's superior genes.

The last sentence was a joke, just in case I need to elaborate.

(5)  For instance, though Peterson is incorrect when he argues that there are more female physicians than male physicians (even in Canada), his use of the large number of female physicians (perhaps in comparison to the small number of female engineers) as proof of preferences affecting outcomes doesn't mention the fact that almost all physicians were men not that long ago.

It's unlikely that the increased numbers of women entering medical school is caused by some kind of change in underlying  sex differences in personality traits.  He might of course argue that all sex discrimination is now over so what we see are women freely choosing certain occupations and avoiding others.

(6)  A certain irony is associated with Peterson reviving the focus on inherited sex differences and his simultaneous fear and loathing of the far left radicals on college campuses.

Peterson became famous in 2016 for refusing to comply with a rule that would have required him to use gender-neutral pronouns when speaking about trans people among the university's students and staff if those individuals requested such use.

The irony is in the new argument that gender identity itself could be largely biologically determined.  More about all that, including the very difficult question what "gender" might mean in this context, can be found in this article.

Note that the view that gender identity could be biologically determined is NOT the same as the view that gender itself is biologically determined.  Gender is constructed differently in different cultures, and the way it is constructed is linked to the ways women are oppressed in a specific culture.

Still, the greater use of essentialist arguments is not limited just to the right side of the political spectrum.