More interesting and wholly random studies. (I'm going to start collecting this stuff.) First, a fun fact: Over the last 40 years, immigrants have been segregating themselves from native populations at an increasing rate, even as racial segregation as a whole has remained constant or declined for many of the fastest-growing immigrant groups. (Mexicans, for instance.) Why is that? Well, three possible theories:

1. Immigrants simply prefer to live among groups with similar cultural characteristics, and away from the Anglo-Saxon-dominated, English-speaking natives. Perhaps. If this is true, then we'd expect more assimilation from immigrants with cultural ties or whatever to the bulk of the native population. Yes? Yes.

2. The natives are trying to get away! Either natives are fleeing from immigrant populations, or they're trying to restrict where immigrants can actually live. Either one could explain segregation.

3. Economic segregation explains it all. Because of the changing nature of the American city, in which cars are important in some areas but not others, you might see immigrants cluster around areas with easy access to public transportation.

So which is the winning theory? Well, David Cutler, Edward Glaeser, and Jacob Vigdor ran the numbers on this (PDF), and the results are pretty interesting. First, Theory #1 is true to some extent—immigrant populations with more experience in the U.S. and those from countries where the language is linguistically similar to English are less segregated. As you'd expect! But the trio found that this only explains a small amount of segregation. Nevertheless, because more and more immigrants are coming from the Caribbean and West Africa, and these groups are more dissimilar to native populations (as opposed to, say, Europeans), this explains some of the rise.

Theory #3 turns out to be a pretty good one: More and more immigrants are coming from poorer countries, they're less able to afford cars, and hence more likely to move to areas with good public transportation. (This doesn't quite explain Los Angeles, but then, nothing really does.) Immigrants from better-developed countries, meanwhile, are less likely to follow these patterns.

As for theory #2, well, as far back as 1970, immigrants were still paying a premium for housing in segregated neighborhoods, meaning that either their housing choices were restricted by natives or, alternatively, they were actively seeking out segregation. But nowadays, say the researchers, it seems that immigrants are mostly occupying neighborhoods that are cheap because they've fallen out of favor with natives for reasons not related to ethnic composition. So it seems that natives are fleeing to the suburbs, as we know, but not necessarily to get away from the ethnic immigrants. It sure looks like that's what's happening, obviously, but no. (The paper does this by looking at housing premiums; I can vaguely follow along but not entirely.) Very interesting! But no word on why the natives are fleeing. Because they think the city is more dangerous? Fools!

UPDATE: Aha! Much, much more on the great urban exodus from the ever-tireless Garance Franke-Ruta.