Saturday, September 15, 2012

Gene on Schelling

Here, here, and here.

The title to my post here is what they call "self-recommending", and several of his post titles are intriguing as well.

I've only read Micromotives and Macrobehavior of his books (and a couple articles). I've been meaning to read Strategy of Conflict, the other one I own, but I would certainly recommend the former.

I read it my senior year of undergrad or maybe the summer before or after - not for a class, just for fun. The analysis that really struck me was his discussion of segregation in the book. It had a big impact on how I think about racial disparities. That year I had also done an economic sociology readings/independent study type course. One of the sociology professors was trying to design a syllabus, she knew I was like the only sociology/economics double major in the vicinity of Williamsburg, Virginia at that time, and that I had an interest in it, so we did an independent study to work through some material. So I was reading Schelling around the time I was reading Durable Inequality, by Charles Tilly with her - and a lot of other sociological work on institutional racism.

And that's really what Schelling is discussing: institutional racism. "Racism" might be a bad word to use there because it is so laden with meaning in this society, but the concept is pure Schelling (I'm not sure he ever used it, though). The point is that you can have macrobehavior that is racist in that it reinforces disparities withouth having a substantial level of racism at the level of individual behavior. We individualize race relations far too much in this country and ignore the macroprocesses and the institutions that govern most racial disparities.

I also read Schelling shortly before I gave Keynes my first good cover-to-cover read (practically on a whim... that turne out to be a good whim for me). Just as Gene connects Schelling to Keynes in the third link, I remember thinking back to Schelling a lot as I worked through the General Theory. If you follow me on Twitter (@D_Kuehn), you would have already known I thought this connection was important:


  1. I remember Tim Harford citing Schelling on segregation in his argument with Dan Ariely over the worth of pure logic in economics. Easterly has a paper arguing that Schelling's model doesn't fit the data. I wasn't able to access it when I first read about it, but it seems publicly available now.

    1. Funny you mention that TGGP. I read Easterly's paper only a week ago... I've been thinking of my own empirical investigation of a Schelling-type model and relevant tipping points.

      In short, it involves South African schools, which experienced a radical shift in racial composition a few years after Apartheid. Now, that may sound entirely obvious, but some schools were affected to a far larger degree than others. Moreover, the proximate cause (at least according to the anecdotal evidence that I have thus far) was new legislation that obliged schools to accepting a minimum number of non-white pupils within their "catchment" areas. Thus, how these catchment areas were defined would appear to have made all the difference. The end result is that schools that were previously all white have become exclusively black within a few short years... While others remain majority white, or enjoy a more healthy mix.

      Okay, I'm blabbing a bit, but the idea has got me quite excited lately. It would seem that you have a nice quasi-experiment and (possibly) a control group depending on how these catchment areas were defined. Identification El Dorado!

      (PS - Data issues aside, this is obviously a pretty sensitive subject. For one thing, I'm nervous to make this an argument about simple racism even in the wake of South Africa's chequered past. I suspect that is much more a story of income disparities and classism... Though, this clearly still qualifies as a form of prejudice.)

  2. You know it's impossible for me to see a post about Schelling without gushing like a little schoolgirl.

    I'll also shamelessly plug this old post in which I described the dynamics of the Segregation Model in some depth. (For those that just want to see things in action -- don't worry! There's a great, short video at the bottom that gets the intuition across very well.)

  3. Oh, by the way, welcome to Twitter.

    About time!


All anonymous comments will be deleted. Consistent pseudonyms are fine.