A new report, “The Dreaded ‘P’ Word,” published by the Lumina Foundation, amasses a huge amount of statistical data in an effort to help state officials figure out if their public higher education systems are high or low in productivity, that is, in producing college graduates. Author Patrick Kelly writes that his findings “clearly point to the reality that some [states] perform better than others with the resources they have.”
Kelly also writes that his analysis “refutes the argument that more funding always leads to better performance.”
With the second statement, I’m in complete agreement. It can’t be said often enough that there’s no automatic benefit to pouring more tax dollars into public higher education. The opposite is actually more likely; as Richard Vedder showed in his book Going Broke by Degree, states that spend more on higher education actually reduce their economic growth.
But what about the first statement? If policy makers analyze data the right way, will they know how to change their higher education systems? I’m very skeptical.
Before getting into the specifics of this study, there is something we should keep in mind about statistics: they are often used as an excuse for government action when no action is called for. That point was driven home to me many years ago when I read about a man who served as financial secretary of Hong Kong in the days when it was a British colony, Sir John Cowperthwaite.
He was once asked why the government in Hong Kong didn’t collect economic statistics. He replied that they would serve no useful purpose and would simply encourage politicians back in Britain to think they could do something to improve on the laissez-faire policies he followed. In one of his last writings, Milton Friedman paid tribute to Cowperthwaite and his “positive non-intervention” approach.
Kelly has gone to great lengths to come up with statistical measures of state higher education performance. It wasn’t easy. “The most difficult barrier to conducting productivity analyses in postsecondary education,” he writes, ”is the lack of available data on the institutional costs of producing college degrees.”
Let’s stop right there.
Colleges and universities, much less systems of them, do not produce degrees, certificates or any other sort of educational credential. Individual students earn them. This isn’t just semantic quibbling. It makes a great deal of difference whether you think of higher education as a producer of credentials, or as a market in which people can shop for educational services they want. The former encourages the kind of thinking that plagued the Soviet economy (“Comrades, we must meet or exceed our production goals!”) whereas the latter encourages businesslike thinking (“How do we make our services as appealing as possible?”).
Think I’m exaggerating about the central planning mindset that educational statistics encourage? Kelly writes that states should think about “creat(ing) and sustain(ing) an economy that can retain the graduates they produce.” Why? People will go where they find the best employment, and state borders have nothing to do with that. Indiana “produces” more engineers than employers in the state need, but that doesn’t mean that politicians need to do anything to attract more companies that employ engineers.
Nor do states have to do something if they have a large proportion of what Kelly calls “undereducated adults.” If an individual thinks he would benefit from more education, he has no shortage of opportunities to obtain it.
For the sake of argument, let’s take the report’s statistics seriously. How productive is North Carolina’s public higher education system?
Measuring the graduation rate per 100 full-time equivalent students enrolled, North Carolina is at 21, which is somewhat below the national average of 23. (Colorado is tops at 30.) That statistic will probably lead to demands that UNC schools need to “improve” and get more students through to their degrees.
I’d look at it differently. If a somewhat higher percentage of students who enroll in college in North Carolina than in Colorado later decide that completing the degree isn’t the best thing to do, what does it matter?
On another measure, the percentage of degrees awarded in scientific, math, and engineering fields, North Carolina looks quite “productive,” with 13.1 percent versus the national average of 11.8 percent. Florida is last with just 8.2 percent. Since those fields tend to be in high demand and generally pay well, we might cheer.
Don’t bother. Again, high numbers aren’t necessarily better than low ones. (Tops, by the way, is South Dakota.)
State higher education leaders should be thinking of improving efficiency, but these statistics don’t help. They should be asking questions like, “Are we spending more than necessary on administrative staff?” “Are professors doing too much research and not enough teaching?” That business-like approach will serve students and taxpayers best.