Publications

Resist the Data Siren Song

It seems easy: collect data, process data, publish data, and everyone becomes better informed and wiser. It’s seductive, and it was clear listening to President Barack Obama and Senator Marco Rubio (R-FL) that both are under data’s spell when it comes to budget-busting higher education. But the main college problem isn’t a shortage of useful information — it’s massive federal student aid discouraging its use.

In his State of the Union address, Obama celebrated federal student aid but then lamented that “taxpayers cannot continue to subsidize the soaring cost of higher education.” His solution? Change college accreditation to include measures of “affordability and value,” and publish information such as loan default rates to help consumers become better informed.

Rubio’s take on the affordability problem was almost identical: laud student aid, lament price inflation, and declare that “we must give students more information on the costs and benefits of the student loans they are taking out.”

Our root college problem isn’t too little information. It is too much federal government.”

But Rubio is doing more than talking. He has co-sponsored a bill with Senator Ron Wyden (D-OR) that would create a massive database of individual-level education and earnings information. The data would eventually be used to tie student outcomes to schools. But useful information is already in abundance, yet every year millions of students major in things with little prospect for good pay, or attend schools with poor outcomes.

The infamous U.S. News and World Report rankings have many flaws, but a degree from a top-25, national university almost certainly carries more weight than from a 30th-ranked regional. U.S. News also furnishes four- and six-year graduation rates for most schools, as well as lots of financial aid information. And U.S. News is hardly alone in the college-evaluation game, with numerous outlets ranking schools based on varying criteria.

What about the employment and earnings prospects for different fields of study? In addition to the Bureau of Labor Statistics furnishing data for myriad occupations, PayScale.com provides breakdowns of starting and mid-career earnings for numerous majors. Bachelor’s in psychology? The average starting salary is $35,200. Music? That kicks off at $34,600. Petroleum engineering? $98,000.

Despite such information being readily available, every year throngs of new grads walk off with diplomas in poorly paying areas. According to federal figures, in the 2009-10 academic year there were 97,216 bachelor’s degrees awarded in psychology, 91,842 in performing and visual arts, but only 72,654 in all varieties of engineering.

The root problem isn’t that useful information isn’t out there. It’s that too many students don’t heed it. They have too little future orientation, and more importantly, pay big parts of their bills with other people’s money. Some is from the Bank of Mom and Dad, but a huge amount is from taxpayers who haven’t any ability to say “no.” Inflation-adjusted aid ballooned from $4,071 per-student in the 1983-84 academic year to $14,745 in 2011-12, and is given away with almost no regard for a student’s academic record, major, or institution.

Unfortunately, the problem with new federal data collection isn’t just that it would address a nonexistent failure, or fail to attack hugely distorting subsidies. It would also introduce new dangers. For one thing, while the Wyden-Rubio bill says data collected will never be “personally identifiable,” such promises have been defeated before through “re-identification” techniques. And who knows what might be created in the future to bypass privacy protections?

Then there are all the shenanigans that could be justified using cherry-picked data. Already we’ve seen some in Washington demonize for-profit institutions while lionizing community colleges, despite the fact that the former often perform better than the latter. But community colleges seem cute, and their loan-default rates are lower because they get most of their subsidies directly rather than through students.

Finally, while we want people only pursuing studies they can afford, why should politicians decide that majoring in engineering is inherently more valuable than poetry? There may be huge, non-financial benefits to many majors, and politicians shouldn’t put their thumbs on the scale for any of them.

“More data” is a Siren song many federal politicians can’t seem to resist. But resist they must, because our root college problem isn’t too little information. It is too much federal government.

Neal McCluskey is associate director of the Cato Institute’s Center for Educational Freedom and author of the book <a title=”http://store.cato.org/books/feds-classroom-how-big-government-corrupts-cripples-compromises-american-education-paperback>Feds in the Classroom: How Big Government Corrupts, Cripples, and Compromises American Education.