David Brooks gets it wrong about education and health care

Correcting David Brooks’ errors of fact and interpretation could be a full time job.

Today he points out that spending on elementary and secondary education and on health care have gone up tremendously in this country since 1960 and yet that productivity in these two sectors of the economy has not gone up at all. And he blames the greater spending in these sectors on two factors, lack of international competition which, he implies, leads them to wind up under the control of government, which as we all know, is a recipe for inefficiency.

I’m not going to spend all day finding all the evidence to show what’s wrong with Brooks’ argument. But I’m going to point to some basic things we all know (or should know).

Are outcomes really stagnant in education?

First, let’s look at what outcomes we are getting from the education and health care sectors. Are we getting more for the greater amounts of money we are spending?

The high school drop-out rates in 1960 was 27.2%. By 1970 it had dropped to 15%. By 1990 it dropped to 12.1% By 2009, the last year for which I could easily find data it dropped to 8.1%. That’s a huge difference. It means that our schools are doing better. And, of course it also means that our schools are teaching more kids for longer. Sending would go up for that reason alone.

I don’t have a lot of faith in the college entrance exams. The correlation between scores on them and both first year college grades and later life success are pretty negligible. But if you take them seriously consider this: over the last thirty years when the scores can be compared (and making corrections for some changes in how they are scored) SAT and ACT scores have pretty much held steady with former declining slightly and the later increasing a bit. This has occurred even though the percentage of high school students taking the test has skyrocketed as more and more of our high school graduates go on to college. We would assume that, as more kids take the tests, those with lesser academic talents would bring down the scores. That this hasn’t happened is a sign of good things happening in our schools. Either our teachers are effective at bringing kids with lesser talents up to higher levels. Or our schools have become more effective at encouraging kids with talent from lower socioeconomic stations to apply to college. Schools have probably done both things and that is a sign of progress (not just in our schools but our culture.)

Are outcomes really stagnant in medical care?

In the area of medicine the signs of progress are simply unmistakable.  The age-adjusted risk of dying dropped 60 percent from 1935 to 2010. There have been huge improvement in both treatment and access to health care especially for infants. But the elderly have done very well as well, largely due to Medicare. Treatments for most diseases have improved dramatically. Pick a common cancer and you will see that survival rates have gone up substantially over the last fifty years. Per capita deaths from heart disease over the last sixty years has fallen by half. For stroke they have fallen by two thirds.  

Someone my age may remember a grandparent with heart disease in the early sixties. The treatments available then were basically the same that were available in the 1940s. They took a baby aspirin every day and had nitroglycerine tablets for angina pain. There were no cholesterol lowering drugs, and no one knew the impact of cholesterol on heart disease. There was concern about high blood pressure but there were no medical treatments for it. There was no bypass surgery, cardiac catheterization or heart transplants. Our grandparents died of untreated heart disease. Our parents are living today much longer with it.

So the whole idea that there has been no improvements in medical care is just laughable. We are paying lot more. But we get a lot more, too.

Is productivity increasing?

Of course, improvements in both education and medical care at higher cost does not mean that productivity has grown. Do we get more for each hour of effort put in by our teachers and doctors? Or are we just spending more?

It’s hard to doubt that, to some extent we are seeing productivity increases. An operation for cervical disk fusion—which I had three months ago—may take a little longer than it did fifty years ago. But the improvements in technology and results are dramatic. That’s evidence of greater productivity. A cardiologist who spends an hour giving a full examination and can then prescribe medications that improve one’s quality of life and reduce the risks of death is being far more productive than a doctor who cannot do so.  

Teachers who set higher standards for all of their students—not just those from upper classes—and who have better ideas about how to meet them are clearly being more productive.

Why productivity increases are small in labor intensive activity

But amidst this evidence for increased productivity, we should remember something very important: there are some areas of activity where productivity increases are very difficult to attain. They are the ones where human skill cannot be mechanized or computerized. Let me give two famous examples, due to William Baumol. We are unlikely to see productivity improvements in the performance of string quartets. That’s not to say that the violinists, violists, and cellists today are not, on average, more skilled than they were fifty years ago. They probably are. And that probably means that the average performance of Beethoven string quartets is better and more consistently good than they were fifty years ago. But there is a limit to productivity improvements in this field. A Beethoven string quartet can probably be played by the best performers today at twice the tempo it could be played fifty years ago. But no one would consider that an improvement in quality, would they? Nor is there any hope of developing any time soon robots who can play a Beethoven string quartet with any degree of musicality.

Take another example: haircuts. The average hair cutter is probably more skilled and more consistently good than fifty years ago. And some haircuts may be a bit more complicated (although those layered cuts of the 70s were probably as complicated as any today). But even if they are better than ever, there is upper limit to productivity in hair cutting. There is only so fast a human being can comb and cut hair without running into the risk of causing bodily injury. And, here too robots are not going to replace human beings in the near future.

Much the same is true in teaching and medicine. Technology can enhance productivity in medicine by providing new pharmaceuticals and devices and perhaps by making information more readily available to physicians. But diagnosis, prescription, and procedures will continue to be carried out, as they have for the last century, by highly skilled professionals who develop those skills through a lengthy period of training.

Education is an even more difficult area in which to enhance productivity. Rote practice can be done by computers today and will be done more effectively by them in the future. But skills in writing, reading interpretation, and in the development of hypotheses and evaluation of evidence, can only be taught by talented and skilled human beings. And, indeed, the more we hand over rote teaching to computers, the more skilled and talented teachers will have to be to handle higher level instruction.

And let me give one more example: the writing of op-eds. I’m sure that, in real terms, David Brooks makes a lot more money today than James Reston did in the 1960s for writing his columns for the Times. But is he much more productive than Reston? Hardly. Reston wrote three columns a week for most of his tenure. Brooks writes two and does a little blogging on the side. Can we increase the productivity of writing newspaper columns. In some ways, yes. Brooks has computerized access to information  of a kind that that Reston probably could not imagine.  I wouldn’t be surprised to find out that Reston had some assistance that Brooks does not have in doing research as a result. But here again, there are limits to how much more productive a writer can be. So far, there is no science and technology that can make us think faster or writer better about difficult subjects. And any effort to think faster is likely to be as counter-productive as playing Beethoven string quartets faster.

Our limited ability to improve productivity in education and medical care (and music and hair cutting and column writing) is why we have been paying a lot more for these things over the last fifty years while the prices of electronic equipment and, in real terms, taking into account quality improvements, food, cars, and housing have declined. I once paid two bucks for a haircut. I now pay fifty for a much better haircut. But the same old one would today cost $20. A concert that once cost $5 now costs $50 or more. And the costs of education and medical care have gone up as well. The reason is that in fields dominated by highly skilled labor, we have to pay people more even though their productivity does not go up that fast. If we didn’t pay them more, no one would do this work.

There is nothing wrong with paying relatively more for goods that can’t be produced ever more efficiently. It’s just a fact of life. The only thing that could stop this from happening is if there were no more productivity in the production of electronics or cars or housing. And that would be a real problem. Rather than complain about the increasing costs of labor intensive, low productivity growth goods and services, then, we should be happy that rapid productivity growth in so many areas gives us the wealth to afford to pay more for them.

The problem of taxes

Government is responsible for providing elementary and secondary education and increasingly for health care as well. Government has to be responsible for education and medical research because they are public goods that benefits all of us. And it has to be responsible for medical care because justice requires that we all have access to good care, no matter our income.  And thus, to pay for these goods, government has and will continue to take a larger proportion of our income over time. Again, there is nothing wrong with this. Even as government takes a higher proportion of our income, we get a lot more bang for the bucks that are left because of productivity improvements in the provision of many goods and services.

Government can do better. But not by getting smaller.

All this is not to say that government can’t do a better job. We do, as Brooks points out, waste a huge amount of money in medical care. The reason, however is not that we have too much government but, rather, that we have too little. The fee for service medical system that served us well when medicine couldn’t do much and thus didn’t cost very much has failed us. Government expansion of the same system through Medicare, which provided funding for health care but which did not change how we deliver and pay for it, exacerbated the problem. (Which is not to say that Medicare is inefficient. Its administrative costs are far lower than those of private insurance.) The ACA’s expansion of health care alone could also make things worse but there are little known provisions in the law that will gradually force changes in how we deliver and pay for medical care which, if we continue to improve them, will wring a great the waste out of the system.

We can certainly save a lot of money. But at some point, those savings will come to an end. We are going to pay a higher percentage of our income for medical care as the costs of TVs, cell phones and, again, in real terms, cars, houses, and food continue their decline.

And much the same is true in education. We can, I’m sure, become more productive by, for example, expanding the use of technology to replace teachers who still do too much rote instruction. But while this will improve instruction, it won’t necessarily make education cheaper. Indeed, as we ask teachers to do more higher level instruction–which we need in our higher tech world– we will need to raise the quality of our teachers and this will cost money. And, at some point the limits to productivity increases in education will force us to spend more of our income in this sector.

Can American political culture adjust to the 21st century?

The real question for a country like America, which has a large right wing that is allergic to taxes, is whether we can adjust to a world in which more of our income goes through the government to pay for critical services that are not subject to rapid productivity improvement. If we can’t, then education and health care will decline in the US. And so will our strength as a country.

David Brooks likes to take the long view and point to the impact of culture on politics. If he were really to do that in talking about education and medical care, he would ask whether a political culture is fit for our world if it leads us to think we have a natural right to low taxes. And he certainly wouldn’t give aid and comfort to that culture.

Edit PDF    Send article as PDF