Exploring the Eclipse of Equality

Rigor Without Mortis: Do-gooders and do-badders

In Intellectual History on March 25, 2010 at 10:12 am

This will be a quick shot that reveals more of my prejudice than my wit, but I had to say a few words about a column by Nicholas Kristof today. I should say first that I really enjoy his columns in general. He shocks me out of my dogmatic slumber on occasion and keeps me focused on the world when I might drift back into the details of American agonistic minutiae. But he got me going today in a bad way and it seemed to be worth explaining why that is. Here is the offending paragraph:

We’re getting a much better handle on what policies can overcome poverty. We’re now seeing more experiments, modeled after randomized drug trials, that measure carefully whether an approach works and how cost-effective it is. Partly this reflects the rise of economists (at the expense of political scientists and do-gooders) and the rigor they pack in their briefcases.

Now, Kristof goes on to explore a set of ideas from MDRC (formerly known as manpower demonstration research corporation) that may well make sense. I know quite a bit about MDRC and I grant that their experimental approach to generating knowledge about social policy is grounded in the most solid epistemological soil that we can find. I always open my research methods classes with a few statements to the effect that if you want to know the causes of things, do experiments. Observational methods will simply not do. There are third variables out there to confound you, endogeneity problems in which eggs become chickens unknown to the researcher, and mechanisms that could plausibly link the cause to the effect are left implicit. But we have known this since Gallileo and so it should not be big news. Even the experimental approach discussed here is not news in that MDRC dates from 1974, a time when we used words like manpower to describe workers.

You may say, what a bunch of dolts social scientists are; they have not even learned how experimental science works. We finally have a new group of people out there doing experiments to discover if social policies are working. Now we will not waste so much money on welfare queens and other freeloaders, but instead can ground our knowledge in facts. Fine, but there are a few problems. First, most social experiments are impractical. Imagine an experiment in which we assign genders to subjects by random draw and then put them through exercises intended to discover the influence that masculinity has on their performance. Second, experiments, where practical, are often unethical. We have become far more concerned about the ethics of human subjects research and it is the experiments that serve as the money shots in the training sequences that portray abuse. Think about Milgram’s studies of authority and obedience. Now admit it: you use his findings all the time, but condemn the mode of discovery even more often.

This is only the tip of the iceberg on ethics of experimentation, but writing as a do-gooder political science type, let me point out something that you may not have thought about. Who is that we do these experiments on? The subjects that MDRC takes as its subjects are often poor, marginalized people. They are also more likely (I suspect with cause) to be women of color. We want to make sure that the money we spend in supporting them (pronouns might deserve scare quotes) is not wasted. Therefore we do experiments. We do not run experiments in determining which scheme of executive compensation is most useful. One could imagine that more money is at stake in the American, versus European, theory that one should make the CEO a member of a qualitatively distinct economic class from his or her employees than in social support policies. Why we (perhaps the shareholders) don’t do this is complicated, but has as much to do with the power of the research subject as it does with any practical consideration. I could go on with examples, but I won’t bore you.

My final statement is not directed at MDRC, which is a research outfit that I like and support, whatever I have just said about social experiments (although I do hope you now react to that term with a just a hint of revulsion). Instead I get to my pet peeve. So Kristof has been hoodwinked into believing that it is the economists who are rigorous (by which he should mean appropriately systematic) and who are finally getting us to the root causes of poverty after all these years of do-gooders wasting our time.  With apologies to my economist friends (yes I have them just as I have Libertarian friends, although I can’t tell what formally distinguishes them whatever they may think of politics), I think that economics has less to contribute here than does Psychology. Since when was economics an experimental subject, barring rare exceptions?

Since the days when the Institutionalists and their ilk were purged from the field, Economics has been a theoretical discipline. If the data don’t fit the theory, we might first question the data. Here we reach the point of sour grapes and disciplinary cat fighting, but I believe that the rational actor, equilibrium tested theorizing places them closer to Galileo ‘s Simplicio and Pope Urban VIII than to his Salviati and experimental science. As all practical economists discover when they use their widely legitimated rigor to take opportunities to make big league policy at Treasury, the World Bank or the Federal Reserve, the politics quickly overwhelm the economics. When you haven’t the first clue what a science of politics would entail because you have committed your career to closed form mathematical catechism, it becomes quite possible that if you are not a do-gooder in social science, you are a do-badder.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: