An alert reader sent me a link to this post (and presentation) by Peter Orszag, Director of the Office of Management and Budget. It gives a synopsis of a talk he gave to the Association of American Universities suggesting important links between the difficult economics of higher education and the difficult economics of health care.

As longtime readers know, I've been linking the two for years. (For example, see here.) The fact that folks in a position to matter are starting to make the connection gives me hope. They haven't really cracked it yet, but the fact that they're at least connecting the two suggests the possibility of eventual progress.

Orszag's argument, which is true as far as it goes, is severalfold. First, education is a key driver of economic productivity, and of economic equality. When access to quality education becomes the province of the elite, we should expect to see economic polarization accelerate. Second, education competes for tax dollars with other expenditures, notably health care. As the costs of health care have increased relentlessly, government support for higher education has been squeezed to make up some of the difference. Finally, the Obama administration is committed to 'bending the curve' (love that!) of cost increase in health care, specifically to free up resources for higher ed. Since a healthy and educated population is more productive than a sick and ignorant one, these policies are both 'hard headed and soft hearted.'

Okay, cool. But even though I'm sympathetic, Orszag is only scratching the surface. To make real and lasting progress, you have to go deeper.

Both health care and higher ed have been subject to incessant cost spirals, and for many of the same reasons. And our current political debate gets those reasons almost completely wrong. Leaning on one to feed the other is a defensible short-term political choice, but it doesn't get at the underlying drivers of cost increase. It doesn't solve the long term problem.

First, and most basically, health care and education are exceptions to the rule that industries only spend money on innovations that are likely to pay off. (Law enforcement is another exception; the cost spiral there is also incessant.) In most industries, a given firm will spend money only on those experiments or innovations that seem likely to generate more money (in either increased revenue or decreased costs) than they cost. That's why local auto repair shops often have ten-year-old computers with dot matrix printers. They don't see the payoff to getting something cutting-edge, so they don't.

In health care, innovations typically add cost; they're justified instead by their results. We've made a judgment as a society – and I'm on board with this, by the way – that extending good life is worth serious money. That's true even when extending that life involves both expensive treatments now and additional treatments later for the afflictions that the patient now lives long enough to get. In olden times, fatal heart attacks at 50 were common. Now, many people who would have died from those live long enough to get cancer, too. That's not to deny for a moment that this is a net gain for humanity. It's just to say that the cost spiral is a direct and predictable consequence of medical success.

In higher ed, similarly, we adopt innovations not because they pay off for us, but because they pay off for other industries. We prepare students to work in the current economy, at jobs as they're currently configured. That means not just professors and libraries, but human patient simulators, multimedia labs, imposing IT infrastructures, and the lab technicians to tend to it all. We don't capture any of the economic gains from these investments. Our students do, and the economy as a whole does, but we don't (at least not directly). We upgrade even when it isn't 'efficient,' because we couldn't do our job otherwise. So we have the normal inflationary cost increases, plus the cost of backfilling decreased public support, plus the cost of health insurance in a labor-intensive sector, plus the cost of keeping up with industries that keep changing. That adds up. We take the edge off by exploiting the hell out of adjunct faculty, but there are limits to that.

(We also have an ever-increasing number of unfunded mandates for which we have to hire, and pay, staff. It's our version of 'defensive medicine' – unproductive spending made necessary by arbitrary external interventions. I shudder involuntarily whenever I hear the phrase “reporting requirements.”)

Second, both industries are paid by their input. Hospitals and doctors are paid by the procedure. Colleges are paid by the course. A cynical view would suggest that procedures and course requirements would multiply. A more nuanced view would suggest that from the institution's point of view, innovations that might save money by getting away from input-based payment will get strangled in the crib. (In my darker moments, I imagine that this is part of the subtext of the opposition to outcomes assessment.) When you're paid by time – whether seat time or bed time – the only way to increase productivity (I said productivity, not quality) in economic terms is to raise prices. There is no other way. So we do.

Third, in both industries it's extraordinarily difficult for the novice consumer to judge the quality of one institution against another. To compete for students or patients, both have to spend money in ways that novices will notice. That may or may not correlate with the core function. To the extent that it doesn't, it's yet another cost item. In a competitive environment, expect those to increase continually.

Fourth, and predictably, both sectors are beset with parasites. In health care, they're called “HMO's.” In education, they're called “private student loan providers.” These parasites crawl into the interstices of the system and suck the lifeblood, growing fat on diverted nutrients. They must be expunged. By replacing profit-seeking middlemen with, say, single-payer systems, tremendous recurring savings can be realized. For proof, compare the administrative costs of Social Security to the average administrative costs of 401(k)'s. The math is irrefutable.

'Bending the curve' of cost increase in these two sectors without just watering down the product won't be easy. It will involve upending some longstanding arrangements, and tackling head-on the 'productivity' problem without which no progress can stand. (Interestingly, public support through progressive taxation solves the 'capturing the gains' problem elegantly. If a given institution's public support were correlated, say, to the taxes paid by its graduates...) I'm happy to see a commitment to increased Pell grants, but over the long term, that doesn't bend the curve. The problem is far deeper than a mere moral commitment will solve, as welcome as that commitment is. The political debate, up to this point, has elided the productivity point completely, but it's key to the entire enterprise.

Still, I'm glad to see that folks in the halls of power are finally starting to connect the dots, even if only in a basic way. After so many years of darkness, a glimmer of light is remarkably refreshing.