Cracking the Information Problem

Editor’s Note: This is the second part in a three-part dialogue about what role the government should have, if any, in spurring transparency in higher education. Robert Martin, Andrew Gillen, Neal McCluskey, Kevin Carey, and Brian Mannix debate the issue. The first installment can be found here.

Robert E. Martin
Boles Professor of Economics, Emeritus
Centre College:

There are two “information” issues in higher ed. The first involves measuring value-added or teaching productivity, and the second relates to the transparency required to control the principal/agent problem. Let me try the easier issue first, the principal/agent problem (the fact that managers or agents have different goals than the people who hire them, the owners or principals).

In higher education, the principal/agent issue is pervasive. It affects how much of a school’s expenditures go directly to core academic activities (as opposed to more peripheral activities, which the agents might prefer to spend money on), whether the real costs of those activities are rising, and, if so, why they are rising. The National Center for Educational Statistics (NCES) has loads of financial info about higher education revenues and expenditures. But you cannot trace the source and use of funds unless you have sufficient information about real physical operations. In higher education, that means such things as staffing patterns by campus activity, teaching loads for faculty, class sizes, and who actually teaches the undergraduates.

There is no information about faculty teaching loads and virtually no administrative staffing information in the NCES data. This means that no one has the critical real-operations data that would allow anyone to trace where the money comes from and how it is actually spent. If you cannot do that, you have little hope of curbing the principal/agent problem.

The public institutions have done a better job of controlling costs than have the privates. They cannibalize tenure-track positions as they become available and hire adjuncts and non-tenure-track professors to teach undergraduates. The more expensive tenured/tenure-track faculty teach few undergraduates. Although this cuts costs, it tends to shift resources that should go to undergraduate teaching to support graduate instruction, and it tends to weaken the undergraduate program’s influence on campus.

The second information issue is responsible for many of the adverse incentives in higher ed. Consider an informed, talented, and ambitious young person considering a career in higher education. She says to herself, I can gain national prominence, job mobility, and rising real income if I am a successful scholar; on the other hand, if I am really interested in teaching, I will be lucky to get tenure; if I do get tenure, I won’t have any job mobility; and my income will be subject to salary compression. If she is not that interested in scholarship, she is going to choose some other career. Two separate quotes from Derek Bok and David Kirp make the point:

If universities truly compete with one another, why do they neglect their teaching so? At least part of the explanation is that rewards for excellent research far exceed those available for excellent teaching. Successful scientists gain worldwide reputations. They receive abundant recognition, awards and prizes, opportunities to consult, offers from other institutions, and salary increases to counter these offers. In contrast, the successful teacher is often unknown beyond her own campus. Her rewards are limited to the satisfaction of a job well done and the gratitude and approbation of her students – all pleasures well worth having but seldom comparable to the fame and other, more tangible benefits given to the accomplished researcher. Derek Bok, Our Underachieving Colleges, p. 160.

Although superb teaching can bring popularity on the campus, no one becomes a superstar because of classroom prowess; no one makes a reputation, either in a discipline or as a media sage, by introducing twenty-year-olds to Wittgenstein or Einstein. It is not surprising, then, that many luminaries are not interested in offering introductory courses. David Kirp, Shakespeare, Einstein, and the Bottom Line, p. 122.

This incentive imbalance is also responsible for “mission creep” in higher ed. The point here is that the market for world class-gifted college teachers does not exist, even though there should be a vigorous derived demand for those services from the public. Why does the market not exist? Because we have no metrics for teaching value-added.

Higher education spends little effort trying to measure teaching value-added and asserts there are no metrics to do so. When one is proposed, the response is that it has some flaw, so it cannot be used. Note that the ways we measure scholarly productivity are also not perfect, yet we have a vigorous, well functioning market for scholars. This is George Stigler’s critical point about the relationship between efficiency and market perfection: Markets do not have to be perfect in order to be efficient.

My bottom line here is this: I know from both study and experience that we have information problems in higher ed and that they tend to persist. I am not sure how society should go about fixing these problems. In other words, I do not know if it will require government intervention. I do not believe after all this time the solution will arise spontaneously. Something has to push it in this direction.

Andrew Gillen
Research Director
Center for College Affordability and Productivity:

At the risk of ignoring the sage advice that it is better to remain silent and be thought a fool than to speak out and remove all doubt, I thought I’d join the conversation.

Let me begin by saying that when it comes to purely private transactions, I generally view them as none of the government’s business.

However, the fact of the matter is that we currently have, and for the foreseeable future will continue to have, massive public funding of higher education. Combined with the information issues we observe, this makes me quite sympathetic to the Martin and Carey camp. [Editor’s note: Kevin Carey of the Education Sector previously proposed that the federal government mandate that schools make public their student learning outcomes.]

My thinking on this is still evolving (spurred very much by this discussion), but here goes.

As I see it, there are two reasons to fall into the pro-regulation crowd.

First, if for no other reason than that schools must be accountable for how the public’s money is being spent, some sort of evaluation of the outcome is necessary. Right now, it seems as though we have public funding with reporting requirements of inputs (and a few outputs), but not of outcomes. This would be analogous to the Department of Transportation only requiring contractors to report what they received and what they spent, without bothering to ask how many miles of road were paved, or if the bridge collapsed from shoddy construction. Adding measures of outcomes in this context seems quite logical.

A second reason would be imperfect information (“information asymmetries” in econospeak), which leads to market failure. What I’m thinking here is that schools are stuck in a situation (a suboptimal Nash equilibrium) where it is only beneficial to reveal information that is selected to make them look good. So if you’re a school with a high faculty-to-student ratio, you advertise that, while if you’re one with small class size, you advertise that. (Building on the frog discussion, frogs can croak, meow, or moo, and there is no correlation between size and which sound they are best at, so they only do the one that makes them look good). The result is lots of little bits of information that aren’t all that useful if you’re a student trying to determine which school would do a better job educating you, because they aren’t easily found or comparable.

As I see it, there are a few reasons to fall into the anti-regulation crowd.

First, as a few people indicated, all the information you need may already be available.

But I don’t buy that. Saying that you can find out all the information you would want seems a bit too much like saying “all the laws of the land are publicly available, so you would never need to hire a lawyer.” These are unrepeated transactions where the outcome is very difficult to determine, so it is not surprising that many decisions are heavily influenced by reputation. At the release of the Hess and Carey paper on graduation awhile back, someone asked former president of George Washington University Stephen Joel Trachtenberg for his advice to potential college students. His answer was something along the lines of, “Go to the richest school you can get into and assume they’re using the money wisely.” That doesn’t inspire confidence that students have the information they need to make informed choices.

Second, as I just noted above, outcomes are very difficult to determine. It is quite possible that the government simply wouldn’t be able to figure out what should be reported. While some things would be pretty straightforward (job placement rates and distribution of starting salary by major, GPA, etc.) this is a very real concern when it comes to evaluating many other outcomes. I’m pretty sure a method could be devised to indicate whether you knew more about calculus upon graduation, but some subjects aren’t so clear-cut. Not to mention that subject-specific knowledge is only part of what universities impart to their students.

Third, even if the government does require all this information, it may not solve the information problem. Perhaps is simply not possible to even get in the ballpark of determining outcomes with the diversity of schools and their missions, as well as all the non-classroom based experiences/impacts their students have.

Fourth, the reliance on reputation is not bad if a school has earned the reputation. While there is some truth to the link between reputation and performance, I don’t think I need to go here with this group—the connection is often stagnant. If you want some convincing, compare the turnover of Fortune 500 companies to the turnover in the U.S. News rankings over the last 20 years. I find it highly implausible that Harvard has been doing everything right for the past 200 years.

Fifth, it would be too costly and bureaucratic.

Sixth, regulations requiring more information would inevitably be followed by regulations mandating how colleges should conduct their business. I am too inexperienced and naïve to comment much on this one. I’d like to think that politicians would have the good sense to leave schools alone, but counting on politicians to have good sense seems like the dumbest thing I’ve written in a while. The experience
of K-12 and of constant “mission creep” of government further scares me.

To wrap up (sorry for the long length), I’m not really sure which camp I fall into. I find 1 & 2 in the pro side quite convincing, but I’m worried about 2 on the anti side, and scared to death of 6.

While I would love it if a bunch of Consumer Reports-type organizations existed so that it was in the schools’ interests to make this information available, they don’t. There are some glimmers of hope that non-governmental solutions will arise. The public is
getting quite riled up about the cost of higher education, leading to closer looks at the industry, and the exposing of some of the more undesirable tendencies. In addition, websites such as payscale.com are sprouting up that provide a start for evaluating outcomes (at least financial ones). Our rankings (CCAP and
Forbes) take advantage of some of these, but our rankings still quite rudimentary due to a lack of good publicly available and standardized information.

We all seem to be in general agreement that one of the worst outcomes would be government mandates about how schools operate. If additional regulation leads to that, I would be strongly opposed. However, I’m not convinced that the one necessarily leads to the other, and am quite partial to the idea that additional reporting requirements would be beneficial on the grounds that a) there appears to be an imperfect information-based market failure, and/or b) that it would be nice to know what we are getting in return for the massive amount of taxpayer money poured into higher ed.

Neal McCluskey
Associate Director
Center for Educational Freedom, Cato Institute

I would like to throw something else out there, perhaps to people with more concrete knowledge on these things than I have. Inspired by Kevin, I did a little research on the Student Right to Know Act last night, and it appears to me that while the law certainly did not lead to a federal takeover of school—the worst, but hardly the only possible deleterious outcome of regulation—it probably didn’t do a lot of good, either. It likely drove up paperwork compliance costs without making colleges do things they weren’t already doing. For instance, the law required making graduation rates available, but this came after U.S. News started reporting graduation rates in its rankings (that started in 1988, from what I have found).

In other words, the feds started requiring data after schools were already providing it. I would guess this was the case for many of the law’s other requirements as well. I have also been struck by the effect of such laws as the Clery Act, which requires universities to disclose crime statistics. It appears to often confuse schools or inspire outright evasion rather than being a tool for getting quick, accurate safety information to students. (This is much like the NCLB’s “persistently dangerous school” provisions.) And then there’s Family Educational Rights and Privacy Act. It was probably well intended, but it now seems to be used as a wall behind which schools can hide unflattering information. All these things, it seems, appear to be instances of either regulatory waste, outright failure, or unintended consequences. But perhaps I’m missing something, because I have not spent much time specifically researching either the Student-Right-to-Know Actor the Clery Act.

Kevin Carey
Policy Director
Education Sector:

A few more minor points: Endowments aren’t part of the U.S. News & World Report rankings calculation. And while there’s surely been some gaming of those elements that are part of the U.S. News rankings, I don’t think the dishonesty and bad faith of certain universities in that regard is per se an argument against reporting data.

Some people will always respond to incentives badly. I don’t recall anyone reacting to the Enron and Worldcom scandals by saying that the government-mandated requirement to report accurate quarterly financial results forced Jeff Skilling and Bernie Ebbers to lie, and should thus be done away with.

As for the claim that the current system is doing a good job in ensuring college student learning results, I would encourage everyone to read The Literacy of America’s College Students whose conclusions haven’t been challenged as far as I know. This study, sponsored by the Pew Charitable Trusts, found that more than 75 percent of recent graduates at 2-year colleges and more than 50 percent of students from 4-year colleges “do not score at the proficient level of literacy. This means that they lack the skills to perform complex literacy tasks, such as comparing credit card offers with different interest rates or summarizing the arguments of newspaper editorials.”

Robert E. Martin:

A substantial part of what is wrong with higher education stems from the fact that there are few real rewards for exceptional teaching. The reason those rewards do not exist is because we do not measure teaching productivity. The market for exceptional teaching fails to exist because any institution trying to hire exceptional teachers from other institutions faces an adverse selection problem that kills the market. The institution seeking to hire exceptional teachers does not know who is a good teacher and who is a bad teacher (no metric for measuring teaching), but the home institution from which the raiding institution tries to hire does know who is good and who is bad. Therefore, the raider knows the home institution will counter outside offers made to good teachers and will not counter offers made to bad teachers; thus, trying to hire good teachers adversely selects for bad teachers when we do not have a metric for teaching productivity. The result: no market for exceptional teachers and no rewards for exceptional teachers.

This creates the following problems:
1. Ambitious people with teaching talent do not choose academic careers—a prior selection bias against good teaching in the academy;
2. Rewards go to researchers, so faculty neglect teaching in order to research;
3. “Mission creep”: Two-year institutions want to be four-year institutions, four-year institutions want to have graduate programs, and institutions with graduate programs want to become research institutions.

Mission creep is hugely wasteful and populates higher ed with low-quality graduate programs.

Brian Mannix
Former Associate Administrator
Environmental Protection Agency:

That’s an excellent description of a classic “lemon effect” in the market for good teachers. Let’s all acknowledge the existence of such a market imperfection (not failure). As such, it is an opportunity for some entrepreneur to come up with a solution—e.g., a way for teachers themselves to credibly demonstrate their abilities to potential employers.
It is also, of course, an opportunity to ask whether some change in the rules might improve the market outcome. For the reasons others have stated, I am skeptical that any proactive intervention by governments in this market will be helpful.

Certainly it makes sense for government (preferably courts, but perhaps also regulatory agencies) to prohibit fraudulent misrepresentation in the marketplace. Hence, we have strict rules about disclosing the accurate mileage of a used automobile. Beyond that, it is difficult to define a role for government that (1) will be useful, and (2) is capable of competently fulfilling without dangerous side effects. Should there be legal recourse
against those who falsify degrees and awards on their resumes? Sure. But let’s be careful. Mission creep at higher education institutions is a vastly smaller problem than mission creep in governmental institutions.

Editor’s note: The third and last installment will address whether a non-governmental solution is possible, offer some negatives that might be caused by federal intervention, and include a comparison with the federal No Child Left Behind Act. The paper that sparked this discussion, Robert Martin’s The Revenue-to-Cost Spiral, is available on this website.

Dialogue Participants:

Jane S. Shaw
President
John W. Pope Center for Higher Education Policy