David R. Henderson  

Technology and Health Care

The Nuances of EU Unemployment... Afghanistan's Natural State...

Last week, I was at an event in Washington, D.C. with about 20 other people--politicians, economists, journalists, and think tankers. One of the economists, Greg Mankiw, made a presentation in which he discussed the reasons for increasing health expenditures as a percent of GDP. One of the two reasons he gave is that improvements in technology have allowed us to do more, to do things we weren't able to do in the past.

One of the other economists, Don Boudreaux, challenged Greg, saying:

Our usual view is that improvements in technology bring costs down. Computers today are much cheaper than they were a few years ago. Why should health care be different?

Greg had a beautiful answer:

Technology does bring costs down. What percent of his income did your grandfather spend on computers? I bet it was zero. Technology brings down the cost and price of health care too. How many hip replacements were done in 1950? None. So the price of a hip replacement was effectively infinity. It's much lower now.

Comments and Sharing

COMMENTS (22 to date)
HJ writes:

What is also different is that the end consumers do not bear all costs and thus are insulated from price changes, leading both consumers and suppliers to care less about cost-efficiency.

Keith Hennessey has a similar conclusion:

"America spends more on health care each year primarily because we demand more and better health care each year. We just don’t know that we’re demanding it, because government policies push us toward high-premium low-deductible health insurance that increases our demand for high-quality but low-value technology improvements."


Dr. T writes:

This was my answer to Don Boudreaux:

The two most prominent reasons for why health care technology continually gets more expensive are:

1. An historic "cost-plus" mentality that afflicts the inventors, producers, vendors, and buyers. No one balks at five million dollar MRI installations because the owners can quickly recover costs by billing $500-$1500 per MRI exam.

2. An inability of most clinicians to use technologies optimally. If there's the tiniest doubt about a diagnosis and the patient is insured, then clinicians readily order expensive testing: imaging studies, endoscopy studies, echocardiograms, esoteric lab tests, etc. Most radiologists and pathologists do not say "no" to these requests, because they also profit from them, and they don't want to annoy their clinician customers. (I'm a clinical pathologist. When I rejected an expensive lab test as unneeded, I usually got a poorly reasoned argument.)

If medical technology had the same level of competitiveness as personal computer technology, and if clinicians knew the strengths and weaknesses of the technologies, then the overall cost of medicine would decline while quality would improve. Howver, I know of no way to achieve these outcomes. Most medical technology companies are near-monopolies with little need to compete on cost. And, it is notoriously difficult to educate physicians and alter their behavior. Only large financial incentives have done this. But, how do we reimburse a physician for not doing something of little or no utility?

adam writes:

Drug patents.

MikeP writes:

Contrasting with Dr. T, this was my answer to Don Boudreaux:

Yet the price of moving a ton of freight from New York to London, or a half-ton of family members from home to the regional shopping mall, has fallen continually.

Why should medical care be different?

The correct analogy is not to the price of traveling from New York to London then versus now or the price of traveling from home to the mall then versus now.

The correct analogy is between traveling from home to the mall then and from New York to London now.

Technology may make what we did yesterday cheaper today, but it also makes what yesterday would have been too expensive to do within reach today and what yesterday would have been impossible to do possible today.

It is not hard to believe that the latter two effects overwhelm the former. The result is that average cost of technology per treatment is higher.

Adding the facts that we're richer today than yesterday -- and that we have more third parties paying the bills -- exacerbates the situation, but does not provide the whole story.

Zac Gochenour writes:

"Our usual view is that improvements in technology bring costs down. Computers today are much cheaper than they were a few years ago. Why should health care be different?"

Health care is different because its not actually about results, its about showing that you care.

Mankiw's answer isn't entirely wrong but it misses the mark. You can't explain the increase in health care expenditure this way unless health care actually makes people healthier on the margin.

David R. Henderson writes:

You way overgeneralized. Hip replacements are not primarily about showing that you care.

kebko writes:

Zac did say Mankiw wasn't entirely wrong.
To defend Zac's point, I will suggest that the next time your wife's doctor suggests getting a $600 MRI when a $100 X-Ray would show him enough information, tell her you think she should do the X-Ray instead & see how she reacts. I recommend having your running shoes on when you make the recommendation, even if you're paying cash.

happyjuggler0 writes:

I love that hip replacements were brought up.

We've all heard that we should be spending less on healthcare in the US because it hasn't brought us longer lives than countries with lower per capita health care spending.

There are at least two problems with this argument. The first is that adjusting for accidents and deliberate deaths (homicides, self dense killing, military deaths etc.), the US actually has higher life expectancy than everyone else.

But as my first paragraph alludes to, there is another reason why you can't use life expectancy tables to claim the US is spending too much. That reason is quality of life healthcare spending, such as hip replacements. Who wants to bet that the US has fewer such procedures than the UK or Canada does, or anyone else for that matter? I sure don't want to take that bet.

Biomed Tim writes:

"You can't explain the increase in health care expenditure this way unless health care actually makes people healthier on the margin."

It may not make people healthier but it may be increasing their consumer surplus nevertheless.

John Booke writes:

I sold medical diagnostic devices for 30 years. The last 10 years as a salesman I had to do one of two things to be successful: 1) convince the doctor that the device would save him time and 2) convince the doctor he could bill for more procedures. For the final 5 years I had to spend 50% of my time convincing hospital and private practice IT people that the data could move through their networks successfully.

Jose writes:

I'm waiting for the day when Doctors offices and Hospitals publish a menu of services with a price list that is available in advance for patients to examine.

Tim Worstall writes:

Weirdly, here's an extract of n email I sent to Don B just last week.

"There's also a further point. The cost of health care is not rising at all. The cost of treating diabetes has gone from infinite in the 1920s to a few dollars a day for insulin now (gross simplification yes). It's that there are an increasing number of health cares available rather than that health care is more expensive per unit."

Maybe Don's not reading emails from me then?

Greg writes:

Medical providers are also in a bit of a technology arms race. They believe dedicated cardiac units, 64-slice CT machines, flat screen TVs in reception areas, and so on will draw in consumers and make them more money.

Unit writes:

I'm confused. In short:

Greg: Healthcare costs are rising because of technology.

Don: but shouldn't costs be coming down?

Greg: yes they are coming down from infinity, because of new technology.

Huh? Does this make any sense?

roversaurus writes:

That was an interesting answer to Don B.

It make me wonder. Is there any data on total
technology spending? Perhaps tech spending has
gone up as a whole while individual tech gets cheaper? Just like total health care spending has gone up while individual procedures have gotten cheaper.

ed writes:

Unit: *Costs* (per unit) are falling but total *expenditures* are rising. (According to the post, Mankiw was talking about expenditures rather than costs.)

Dr. T writes:

roversaurus wrote: "I'm confused. In short:

Greg: Healthcare costs are rising because of technology.

Don: but shouldn't costs be coming down?

Greg: yes they are coming down from infinity, because of new technology.

Huh? Does this make any sense?"

No, it's a ridiculous argument. We're talking about costs of medical technology, not the entire fields of medicine and pharmacology. As we know, in most areas, technology costs decline over time. But, looking at examples in medicine such as irradiation devices to treat cancer, we see that inflation-adjusted costs continue to rise. The excuse is always that today's medical technologies are better, faster, stronger, smarter, more communicative, prettier, sexier, etc. However, the incremental gains rarely are worth the massively increased prices. I spent nearly twenty years directing medical labs that purchased (over those years) millions of dollars of instruments. The prices continually increased while the value went up slightly, and in some cases went down. This happened because there was so little pressure by administrators to control technology costs (to meet budget, we just had to lay off medical technologists instead).

In the posts above and at Cafe Hayek I read lots of economic theory arguments about this topic by people who have not worked in healthcare. Most of the theories I read do not match the reality (see post 2).

Zac Gochenour writes:

There is more to "Computers today are much cheaper than they were a few years ago" than meets the eye.

It is more accurate to say that as a result of technological improvement, computing performance has gotten cheaper. $210 bought me my first ever cpu, a 300 mhz Pentium 2 in 1998. A 400mhz Pentium 2 now costs $9. So say P1 is the equilibrium price of one "unit" of computer performance in 1999, and P2 for 2009, so P2

It is also true that the average selling price of PCs has fallen although their performance has increased. So, Q2>Q1 and P1*Q1 > P2*Q2. This does not necessarily follow from the first point nor is it contradicted by the first point.

Now let's relate to health care. First, let's assume that health care is to PC as health is to computer performance, so when you buy health care you are really buying health. We know that people are spending more money on health care, so (A) P1*Q1

One explanation is that B is not true, but it doesn't seem likely that improvements in health care technology raised the cost per unit of health. Another explanation is that C is not true (most people would say this), but its an empirical result -- people are not healthier as a result of greater health care expenditure. A is also an empirical result.

Now let's change the assumption. Say health care is actually a mix of health and signaling. So, A is P1*Q1 + S1 P1*Q1-P2*Q2+S1. Since Q1=Q2, S2 > Q(P1-P2) + S1.

In plain English, if it has actually gotten cheaper to be healthy yet we spend more on health care to be just as healthy as before, that means we now spend on signaling at least what we were spending before, plus what we saved by the technological improvement.

Biomed Tim says "It may not make people healthier but it may be increasing their consumer surplus nevertheless." Hmm but are they? All the benefit from signaling comes from spending more than others. If everyone spent less, they would all benefit.

John Fast writes:

As far as I can tell this is simply an example of the substitution effect which is a standard part of basic microeconomics.

Obviously Professor Boettke is familiar with this phenomenon. So what am I overlooking here?

Chris writes:

[Comment removed for supplying false email address. Email the webmaster@econlib.org to request restoring this comment. A valid email address is required to post comments on EconLog.--Econlib Ed.]

Kartik writes:

Greg Mankiw's answer is a good one. In the long run, it is accurate.

Traditionally, the cost of extending human lifespan to 120 has been infinity. Today, it is still infinity, but is rapidly approaching a finite number.

By 2030, such a lifespan may be within reach for wealthy (98th percentile) people.

By 2050, it will affordable to most people.



People are not optimizing their healthcare resources. A lifetime spent in a healthy diet (high on fruits and vegs, low on almost everything else), exercise, yoga, etc. will avoid costly diseases like cancer, heart disease, Alzheimer's etc. Healthy foods are actually cheaper than unhealthy foods (those who disagree don't actually know how to shop for vegetables).

People incur costs of $300,000 to treat their cancer, and still die. Instead, a healthy life might have avoided that cost altogether, AND added years to life.

So people are simply not viewing health from an ROI perspective, and a proactive vs. reactive perspective.

Kartik writes:

Again, all the talk about life expectancies merely shows that Americans don't really WANT life expectancy to rise badly enough to make some small sacrifices.

US life expectancy will immediately rise by 5 years, while also saving tons of money, if all Americans simply :

1) Ate a diet high in fruits and vegetables, and low on sugary foods, fried foods, processed foods, and anything that was not considered 'food' 60 years ago.
2) Keep red meat to a minimum (1-2 times a month)
3) Exercise 3 times a week, 30 minutes each.
4) Meditation, 10-15 minutes a day.
5) No smoking, and minimal drinking (beer and red wine only, no more than 1 per day).

If every American did these 5 things, life expectancy would rise 5 years, and $1 Trillion a year would be saved both in consumer spending as well as healthcare costs.

Longevity + a Booming economy, all so easily.

This is not secret info. Everyone knows it. Yet, Americans can't even do this much.

Comments for this entry have been closed
Return to top