A university can give a city a museum you can enter for nothing and a housing market you struggle to enter at all. Ann Arbor makes that contrast unusually tangible: the University of Michigan Museum of Art is free to everyone, while the university's expanding enrollment adds to the number of people who need somewhere to live. The public benefit and the housing pressure come from the same institution.
But that does not make an expensive college town a law of nature. Compare Ann Arbor with Boulder and Madison and a more interesting question emerges: how does each city respond when an institution grows? A university supplies people, activities and jobs. It does not, by itself, determine how many apartments are available outside its gates.
Ann Arbor: Count People Before Counting Charm
Michigan's official Ann Arbor campus report records 51,225 students in fall 2022, 52,065 in fall 2023 and 52,855 in fall 2024. That is an increase of 1,630 students in two years, or about 3.2%. These are campus headcounts, not an estimate of new renters: some students occupy university housing, share a home, commute or already lived locally. Still, the series identifies a real source of changing demand instead of attributing everything to a vague surge in desirability.
There are at least two housing questions hidden in that increase. Where will the additional students live? And does the available housing suit the people who are not students? A room in a shared house and a self-contained apartment can serve different households even when both appear in a rental search. Counting available bedrooms as though they were interchangeable homes obscures that distinction.
The attraction is equally concrete. UMMA's permanent policy is free admission, without a ticket requirement. You need not be enrolled to make the museum part of your week. That is a university amenity with a genuine public doorway, unlike a laboratory, a student service or a campus facility whose access rules may be narrower. For a prospective resident, the useful inventory is not everything the university owns. It is what you can actually use.
Boulder: A Landscape Is Not a Complete Explanation
CU Boulder reported 38,428 students in fall 2024, an increase of 1,275, or 3.4%, from the previous fall. The university attributed that year's enrollment increase to record retention. That detail matters: housing demand can change because existing students remain enrolled, not only because a city attracts newcomers from elsewhere. A campus calendar has economic consequences beyond the first-year arrival weekend.
Boulder's housing history also contains decisions, not merely mountains. The city's own policy timeline records the repeal of its Residential Growth Management System in 2024. Repeating an old annual growth restriction as though it still governed new housing would turn a useful explanation into a stale one. The physical setting persists; the rules can change.
Neither fact supplies a dollar value for a supposed university premium. Enrollment, building permissions, construction costs and the appeal of the surrounding landscape are different variables. Establishing how much any one contributed to a rent increase would require a stronger study than comparing three attractive towns. What the evidence does establish is that a simple story about fixed supply facing limitless demand misses both institutional changes and policy responses.
Madison: The Counterexample Is More Homes
Madison is useful because its housing evidence does not tell an uninterrupted scarcity story. The city's 2025 Housing Snapshot reports more than 22,400 homes completed between 2015 and 2024, a roughly 20% expansion of the housing stock. Those are homes of all types across the city, not just student apartments on the isthmus.
The report's CoStar series shows stabilized rental vacancy rising from 1.9% in 2022 to 4.8% in 2025. "Stabilized" matters: this is a particular measure of established rental properties, not a Census count of every empty home and not a promise that every newly arriving renter has abundant choices. Nevertheless, it directly challenges the idea that a university city can only become tighter.
The same report finds the lowest vacancy in lower-quality, lower-cost rentals. More choice at one end of the market can coexist with a difficult search at the other. That is the crucial distinction between a city adding housing and a particular household finding a home it can afford. Madison is a counterexample to inevitability, not an example of a solved problem.
Buy Access, Not the Stereotype
These comparisons change what to look for. In Ann Arbor, inspect how a rental fits the campus-driven market and which university resources would become part of your life. In Boulder, separate the enduring mountain setting from housing rules that have changed. In Madison, look beyond the improving headline vacancy rate to the segment in which you would actually rent.
A hypothetical choice makes the point. Two apartments might offer equal space, but only one puts the museum, work and friends within the journeys you are willing to repeat. Its extra rent would be buying those particular connections, not an abstract quantity called college-town character. Conversely, paying for campus proximity makes little sense if nearly all your destinations lie elsewhere.
Compare Ann Arbor, Boulder and Madison as three housing systems around universities, not three versions of the same bargain. The university explains why you might want to be there. The housing response helps determine whether wanting to be there can become living there.
