The Trouble With Smart – Part 1

The way Andrew Leigh MP spanked former ALP Senator John Black for his views  on the Federal implications of the Queensland election reveals a core problem for policy and economic research.

This is Leigh’s slapdown tweet (courtesy of Catallaxy):

It’s important to understand that Leigh is a former ANU economist with a Harvard PhD and Masters in Public Administration. He has written extensively on the Australian economy and on inequality.

It’s good that someone with this background has entered politics. It’s not so good that he has bought the myth that what people do in these fields is science, or that a smug dismissal based on authority is justified in any field where, frankly, almost no evidence is in any way convincing.

It’s an occupational hazard for those who make a career out of accumulating academic credentials, understanding abstract models and taking their own research papers seriously.

But engage in most social or economic questions that matter – as opposed to hard physics or engineering – and you find there is no science involved at all. There is an array of academic papers with competing measures of statistical validity – measures that no-one other than the authors truly believe indicate anything much – proposing conflicting views that reason and evidence alone cannot separate.

Or there are various theoretical models that only convince those predisposed to accept them.

This is very clear in current debates in macroeconomics.

We can see prejudice-masquerading-as-economic-debate over stimulus and monetary policy among US economists. Highly informed, abstractly impressive debate, but one where it is clear that people’s primary theoretical commitment is psychological and emotional, and that a lifetime’s research is a search for rationalisation of that commitment.

It is even clearer in ethical theory, where no school offers anything convincing, just a series of useful insights. Yet there are academics, writers and activists committed to one school or another on some basis. What basis could that be?

If evidence is inconclusive for separating theoretical perspectives, how else is it possible to explain the very strong commitments researchers, writers and academics have other than pre-existing emotional predisp0sitions?

Harsh call. I have been told I am far too cynical to ever be an academic. But in areas where experimental evidence is either impossible or irredeemably inconclusive, reality cannot apply the necessary discipline to our capacity to rationalise anything.

Worse, far too often the rigour that is applied to the data and research method seems to disappear in the Discussion section of policy and economic papers. It is stunning how many unstated presumptions and ethical assumptions slide into Discussion to colour the inferences and implied actions from a paper’s data.

Like many of my recent posts, this is opening up into a book.  Topic: the limitations of the Smart Class – those people who equate credentials with insight or morality, and abstract intelligence with a pathway to truth. And who put far too much faith in rigorous research methods in social and economic areas.

It’s a class that isn’t confined to academia. It is a core identity element for many in the media, bureaucracy and for high-earning corporates.

The real trouble with smart is that it isn’t smart enough to recognise its own limitations, and the limitations of their areas of study. And that it is far too willing to see its own activity as impartial and pure, rather than driven by exactly the same biases, prejudices and needs to maintain identity as all the poor slobs without higher degrees from Ivy League universities.


After direct feedback I need to clarify a few things: Firstly, the core problem I identify is bipartisan. It is a challenge to me that the Smart Class is almost as established in the Liberal Party as anywhere else.

Secondly, the great problem it poses beyond the academic is that it justifies the growth in government intervention we have seen over the last 40 years.  Now that there is the added rationalisation of behavioural economics and nudge thinking, it poses the threat of intervention without a policy debate about that intervention.

Thirdly, this post is not meant to be a complete argument on the quality of social research, to reopen old debates from Hume, the Vienna School, the philosophy of science or anything similar. It is a blog post. I may write an article, I may write a book. Gimme a chance.

Fourthly, one of the key supports for my intuitions is the fact that an informed reader of social, policy or economic research can almost always predict a paper’s conclusion based on who the author is and/or who the sponsoring institution is. That should immediately raise any thoughtful person’s eyebrows.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.