We Need More Universities

We Also Need Better Data About the Ones We Have!

Noah Smith has a good new article at Bloomberg View. I think he’s spot on on almost everything says here about the need for policies that work to undo much of the (policy-driven!) concentration of opportunity in America by lifting up depressed places. In this, I think there is left-right agreement in principle, as Ross Douthat shows, if not always in policy details.

But because I’m an assinine person and I pick fights over details, let me tell you where I think Noah is wrong. Here:

Okay, he’s wrong here. We don’t have a “steady clip” of new universities. But it’s not necessarily Noah’s fault that he’s wrong. If you google “how many universities are there” you’ll eventually land here, with this reputable-looking table from a very reputable source, the National Center for Education Statistics (NCES):

But that chart looks wrong to me. Why? Because I’ve been working on this issue for a while, and have data on when universities were founded or accredited. Let’s start with foundings. Using publicly available records (i.e. Wikipedia), I have a database of over 3,300 universities and their date of founding, running up through 1980. My data cuts out at 1980. Here’s the dates of founding:

So low-ish foundings until the 1810s, then a steady rise until 1890, then stability until 1950, then a huuuuuge spike, then a collapse. What happened?

Well, first of all, let me tell you what that line looks like if I classify schools based on words that appear in their names:

Now look, just using names is a seriously haphazard methodology. But it’s better than no distinguishing characteristics, as the NCES data gave. Here, we can see that “university” foundings remained low-but-steady from the 1910s to the 1970s, well below founding-rates of the mid- and late-1800s. Meanwhile, the boom in the 1960s was about half driven by vocational, community, and technical colleges, as well as by other non-university colleges, which may or may not be community colleges that just didn’t have my keywords in their name.

The point is, university founding rates seem to be fairly low, and most new higher ed entrants are probably not full-fledged universities.

Let’s look at another data source: accreditations!

Here I have founding and accreditation data for all institutions with “college” or “university” in their name.

As you can see, there are some significant differences, though through 1980 accreditations come to about 1,500, while foundings are just 1,300: this makes sense as many accreditations may be for foundings that occurred before 1900.

Most notably, both data sources show spikes in the 1940s and 1950s, and again in the 1960s. Both show declines to 1980. But where my founding data cuts out, my accreditation data continues, and shows an ongoing decline in new entrants.

If accreditation and founding data is to be believed, Americans are launching the fewest new higher ed institutions since WWII or before.

But wait. Why does NCES data show growth?

Well, first of all, there are a few more recent years than the data Noah showed, though finding the data is a pain. Here’s total institutions, including community colleges, according to NCES:

As you can see, NCES actually tracks a decline in total institutions in the last 2 years. So even the data that Noah cites would seem to suggest that all is not well in American higher ed.

But what are these schools? Why did they keep rising?

Well, I don’t have an annual list of NCES schools per se… but I do have IPEDS data which allows me to track schools that report any students in major surveys of students and institutions, which should be the vast majority of the schools we care about. Sure, we may miss some small ones, idiosyncratic schools, unaccredited schools, etc, but really, IPEDS should catch the vast majority of schools. Maybe we can use IPEDS data to track new entrants indirectly.

Yes, we can! Here are IPEDS entrances and exits from the data, 1980–2015.

Entrances are rising! So we’re adding plenty of new universities!

But what kind of university?

Well, I can’t go back and classify every university in my sample. I did restrict myself to primarily BA-granting institutions. But I didn’t restrict to nonprofits. Let’s zoom in on that spike in 2013; what drove entrances?

Well, of the 46 entrants, 23 have “DeVry University” in their name. Because every campus of a for-profit institution is counted separately. Here’s the summary stats:

I don’t have figures for other years on hand as I have to look up each school individually and read its history to figure out what happened, so it’s labor intensive. But from eye-balling, this mix looks pretty standard. Lots of for-profits, some small religious colleges, a speckling of status-upgrades, branch campuses, or consolidations, occasional exchange-oriented schools for international students, and then a hefty amount of schools that have existed for a long time but just didn’t bother reporting their data.

I should say, many of the religious schools are either Jewish, or oriented towards Asian-American Christians. And in 2015, we got a Muslim school, to my knowledge the first Muslim liberal arts school in the US. So what educational entreneurship is occurring is mostly being driven by religious and immigrant communities, or else by for-profits who spin off dozens of campuses all at once.

Lest you think this is all just an artifact of classification, I decided to zoom in on a state I know well and go through all the historical records I could find to identify how many schools there had ever been. So here’s the history of higher education in Kentucky, based on my trying to identify even long-defunct schools:

Here’s the number of people per 4-year-school in Kentucky:

As you can see, for Kentucky, foundings basically stopped dead after the 1960s, even among for-profits. The result is that the population per institution has risen steadily. For 4-year schools, we have as few universities per capita as we had in the 1880s. Including all schools, we’ve only slipped back to the 1940s… but still!

Now to be clear, it’s reasonable to think a permanent downward trend shouldn’t exist. Plus, maybe we over-built in the 60s! So if 50,000 Kentuckians per institution was too few, fine; maybe 60,000 is better, or maybe 70,000, which we’ve now reached. If 70,000 is ideal, then we should found another university in Kentucky now, because population is still growing!

But if what we care about is 4-year schools, the situation is even more severe. Again, maybe 100,000 people per school is too low. I think that’s possible. But surely we weren’t over-institutioned in 1920, when we had about 120,000 people per institution? If 120,000 or even 140,000 is the right level, we need more schools!

In sum, I have shown you from 4 sources (foundings, accreditations, IPEDS entrants, Kentucky historical information) that new entrance of 4-year nonprofit institutions, the kind of institution we mean when we say colleges or universities benefit their communities, has slowed markedly since the 1970s or 1980s. We cannot say with certainty what the optimal rate of entrance is. But let’s consider that question.

Why No Supply?

If I told you there was a commodity where the price reliably rose several points above the rate of inflation, and the quantity of demand was rising several points faster than population growth, but that there was a diminishing rate of new entrants in the marketplace… what explanations would you offer?

You might expect to find supply restraints or other barriers to entry. But for higher ed, it’s not easy to see formal barriers. The reality is starting a university is not terribly complex. Many communities launch small institutions that grow over time into successful ones, and the regulatory requirements are not excessively high at those early stages. Sure, there are probably opportunities for improvement, but it’s probably easier in principle to start a new college than a new day care.

The problem is, nobody will attend your college. If you launch a shoestring university without accreditation, who would ever attend? Nobody would believe in the product you’re selling, and for good reason!

No, to be successful, schools need prestige. That’s what you’re really paying for. Whether this is “signalling” or whether prestige itself signals quality of human capital improvement doesn’t matter for the mental model I’m building here; all that matters is that a school only succeeds if it can signal prestige to customers.

What signals prestige? Well, all the usual things: facilities, advertising, a sleek website, signs of wealth and resources, etc. Award-winning faculty of course helps, too, but faculty themselves respond to prestige.

Longevity helps signal prestige, but a startup can’t have longevity, so this key prestige-signal can’t be duplicated. It’s a costly signal. To the extent longevity indicates prestige and to the extent prestige determines demand for a university, the higher ed market will have systematically inefficiently low entrance.

But maybe longevity is just one factor. Maybe we also think that prestige can come from participation in a values-based community, like a religion, or even a cultural or ethnic group. Religious or ethnic groups are in fact a major piece of the university-entrepreneurship millieu, so this seems to fit. To the extent that religious or ethnic group identity facilitates startup prestige accumulation, and to the extent religious or ethnic identity declines in relevance, we would expect the extent of the higher-ed-entrance-deficit to grow.

But a lot of schools are nominally religious at best. What else could signal commitment to quality or provision of prestige? Well, how about nonprofit status. If you’re not in it for the money you must be in it for the prestige. However, while small-scale startups are possible, another key determinant of prestige is probably size and faculty quality, which will relate to size. Size means there are big startup costs. Big startup costs means you need entrepreneurs to make big investments, taking big risks. Big risks means entrepreneurs expect big profits… but to signal quality you have to be nonprofit! Because nonprofit status is integral to private university prestige- and quality-signalling, the risk associated with launching a substantial new university cannot be compensated for by entrepreneurial returns (i.e. profits).

Sidenote: one genuine new startup university called Ave Maria exists in Florida. The billionaire founder spend hundreds of millions to launch a successful university, and built an entire town around it, with the university at the center, in the 2000s. It continues today, and is doing well. However, the founder owned not only the university land, but the town land. In theory a savvy big-money developer could get entrepreneurial returns by accruing 100% of the returns to rising land value around a university. This is basically the underlying economic principle of land-grant universities, but privatized. However, this model has huge startup costs, and is very rare. Also, note Ave Maria is a devoutly Catholic university!

Finally, an institution could signal prestige by being a government-run institution. But that subjects them to political decisionmaking, which doesn’t really solve our problem.

So then I think there is a reasonably plausible case to be made that there may be systematic under-entrance in the higher ed market. With entrance suppressed by key features of the market relating to what it means to be a university, price can be bid way up even as enrollees expand too, because expansion at a given university does not create price competition in the way expansion at two universities would.

Let me unpack that.

Why Not Expand Universities?

Douthat and Noah both suggest satellite campuses or expansion. I disagree. Universities set prices at least partly on a centralized basis, which means satellite campuses or system-partners are rarely price-competers. But we want some price competition. In fact, satellites can be the opposite of price-competers, as they can ride the prestige of the home university and draw in students paying a high price, even if the satellite quality of education is far below the flagship, as is often the case. And if satellites are much cheaper, they usually function as transfer-mills, but transfer mills don’t produce any benefits for local communities, because research doesn’t happen, para-university civil society doesn’t become robust, and students don’t put down long-term educational roots. What you need are schools with robust undergrad-to-PhD programs to keep people around for 4, 6, 9, 12 years, and to draw in older people with lower migration propensities.

Creating more satellite campuses will crowd out new institutions who may have more long-run quality and growth potential, but whose start-up-prestige can’t match the longevity-prestige of a satellite, as the satellite gets flagship prestige regardless of quality. And if the satellite produces a discount version, it may totally fail in regional economic development goals.

Rather, we need new institutions. We need institutions that have immediate prestige; that are credible from day one. But we also need institutions that are in struggling areas, so places where the returns to educational entrepreneurship might not immediately be high. And we also need institutions that don’t strain states’ already struggling higher ed budgets.

Boy, have I got a proposal for you.


Public-Private Universities Are the Answer

The problems we face are: (1) the regional returns to higher education are too localized, (2) the price of higher education is bid up very high, (3) the traditional entrepreneurial player, state governments, is financially strained or unwilling, (4) private entrance is systematically suppressed by unavoidable market features.

So here’s what we do. We divert a few hundred million dollars each year from elsewhere in the Federal budget. Some comes from the Department of Education, but we get different Federal agencies to partner with universities to develop specialized programs in partnership with them. We also ask states and localities to make nominal contributions. We could even ask private businesses to take an ownership stake. Anybody who chips in gets a board member.

We make a $100 million investment per school in Year 0, then pledge to subsidize the per-student cost at a rate of maybe 75% in Year 1, dropping to 0% in, say, Year 20. In other words, we guarantee several years of fierce price competitiveness and large-scale local buy-in.

At the end, all government ongoing support is cut off, and it’s a private university with government board members. No funding support, but still the quality-control/prestige of having explicit governmental backing and management. If going to the Huntington National University also means you get special internships at the FBI’s counter-narcotics division, then all the better!

This policy would create price discipline in an inflated market, allow for large-scale innovation in how schools are run provided that institutions are fairly autonomous (you could even allow some board members to be elected democratically in the local community!), and create a standing policy of regional diversification. If we found 5 universities a year and the targeted final enrollment was about 5,000 students, and if the cost to education a student was $20,000/yr, and if we ignore inflation (for simplicity here), here’s what this program would cost:

That’s assuming we only do new foundings for 20 years. At the end of it, we have 100 new, good-quality mid-sized universities in formerly depressed local areas with valuable programs that directly connect people to jobs, other regions, their community, and their government.

For reference, $1.4 billion a year is only 2% of the Department of Education’s 2017 budget request.

For just TWO PERCENT of what we already spend on Federal-government-education-spending, we could turn the population depressive trajectory of university-less towns around and launch them into the 21st century with universities that pioneer a new model of higher ed policies.

I could rhapsodize on this for a while, but I’ll spare you. Mull it over. Let me know what you think.

Check out my Podcast about the history of American migration.

If you like this post and want to see more research like it, I’d love for you to share it on Twitter or Facebook. Or, just as valuable for me, you can click the recommend button at the bottom of the page. Thanks!

Follow me on Twitter to keep up with what I’m writing and reading. Follow my Medium Collection at In a State of Migration if you want updates when I write new posts. And if you’re writing about migration too, feel free to submit a post to the collection!

I’m a native of Wilmore, Kentucky, a graduate of Transylvania University, and also the George Washington University’s Elliott School. My real job is as an economist at USDA’s Foreign Agricultural Service, where I analyze and forecast cotton market conditions. I’m married to a kickass Kentucky woman named Ruth.

My posts are not endorsed by and do not in any way represent the opinions of the United States government or any branch, department, agency, or division of it. My writing represents exclusively my own opinions. I did not receive any financial support or remuneration from any party for this research.