The other day I published a graph on student housing in Canada and the United States that seemed to grab a lot of people’s attention.  It was this one:

Figure 1: Student Living Arrangements, Canada vs. US

People seemed kind of shocked by this and wondered what causes the differences.  So I thought I’d take a shot at answering this.

(caveat: remember, this is data from a multi-campus survey and I genuinely have no idea how representative this is of the student body as a whole.  The ACHA-NCHA survey seems to skew more towards 4-year institutions than the Canadian sample, and it’s also skewed geographically towards western states.  Do with that info what you will)

Anyways, my take on this is basically that you need to take into consideration several centuries worth of history to really get the Canada-US difference.  Well into the nineteenth century, the principal model for US higher education was Cambridge and Oxford, which were residential colleges.  Canada, on the other hand, looked at least as much to Scottish universities for inspiration, and the Scots more or less abandoned the college model during the eighteenth century.  This meant that students were free to live at home, or in cheaper accommodations in the city, which some scholars think contributed to Scotland having a much more accessible system of higher education at the time (though let’s not get carried away, this was the eighteenth century and everything is relative).

Then there’s the way major public universities got established in the two countries.  In the US, it happened because of the Morrill Acts, which created the “Land-Grant” Universities which continue to dominate the higher education systems of the Midwest and the South.  The point of land-grant institutions was to bring education to the people, and at the time, the American population was still mostly rural.  Also, these new universities often had missions to spread “practical knowledge” to farmers (a key goal of A&M – that is, agricultural and mechanical – universities), which tended to support the establishment of schools outside the big cities.  Finally, Americans at the time – like Europeans – believed in separating students from the hurly-burly of city life because of its corrupting influence.  The difference was that Europeans achieved usually achieved this by walling off their campuses (e.g. La Sapienza in Rome), while Americans did it by sticking their flagship public campuses out in the boonies (e.g. Illinois Urbana-Champaign).   And as a result of sticking so many universities in small towns, a residential system of higher education emerged more or less naturally.

In Canada, none of this happened because the development of our system lagged the Americans’ by a few decades.  Our big nineteenth-century universities – Queen’s excepted – were located in big cities.  Out west, provincial universities, which were the equivalent of the American land-grants, didn’t get built until the population urbanized, which is why the Universities of Manitoba, Saskatchewan and Alberta are in Winnipeg, Saskatoon and Edmonton instead of Steinbach, Estevan and Okotoks.  The corollary of having universities in big cities was that it was easier to follow the Scottish non-residential model.

The Americans could have ditched the residential model during the transition to a mass higher education in the 1950s, but by that time it had become ingrained as the norm because it was how all the prestigious institutions did things.  And of course, the Americans have some pretty distinctive forms of student housing too.  Fraternities and sororities, often considered a form of off-campus housing in Canada, are very much part of the campus housing scene in at least some parts of the US (witness the University of Alabama’s issuing over $100 million in bonds to spruce up its fraternities).

In short, the answer to the question of why Americans are so much more likely to live on campus than Canadian students is “historical quirks and path dependency”.  Given the impact these tendencies have on affordability, that’s a deeply unsatisfying answer, but it’s a worthwhile reminder that in a battle between sound policy and historical path dependency, the latter often wins.