E59 -

Why is my smartphone cheap while my healthcare is crazy expensive? Alex Tabarrok explains the Baumol Effect.

One of the most remarkable aspects of the last few generations is that for the first time in human history, at least to this degree, stuff has been getting cheaper while human labor gets more valuable. It’s a technology‐​enabled humanist revolution! At the same time, labor‐​intensive sectors like healthcare and education have become more expensive relative to the declining price of goods. Economists call this the “Baumol effect,” though it’s sometimes referred to as the “cost disease.” But economist Alex Tabarrok joins the show to discuss how that curse might actually be a blessing in disguise and how the Baumol effect radically disrupts our preconceived notions about effective government policies.

Why are some prices getting higher while innovation causes the lowering of other prices? Why has the price of education gone up? What is the Baumol Effect? How can we substitute for skilled labor?

Further Reading:

Why Are the Prices So Damn High?, written by Eric Helland and Alexander Tabarrok

Marginal Revolution

Stubborn Attachments, written by Tyler Cowen

The Automation Revolution is Upon Us, Building Tomorrow Podcast

Will Artificial Intelligence Take Your Job?, Building Tomorrow Podcast

On Innovation: Don’t Ask for Permission, Building Tomorrow Podcast



00:05 Paul Matzko: Welcome to Building Tomorrow, a show about the ways tech and innovation are making the world a better place. I’m your host, Paul Matzko, and today, I have a very basic question. Why are the prices so damn high? Why are some things getting more expensive, even as technological innovation are making the goods we use much less expensive? Our smartphones every year get better and cheaper. Our cars, better and cheaper. But why are things like education and healthcare getting so much more expensive, year after year? Today, I have a special guest with me, George Mason University economist, Alex Tabarrok, the Bartley J. Madden Chair of Economics, who is also one‐​half of the most interesting economics blog in the world, Marginal REVOLUTION. Welcome to the show Alex.

00:52 Alex Tabarrok: It’s great to be here.

00:53 Paul Matzko: Now, you and Eric Helland just released a new short book, you can actually download it free online, we’ll put a link to it in the show notes, titled Why Are the Prices So Damn High? Okay, so let’s start real basic. What prices are high and which ones aren’t?

01:09 Alex Tabarrok: Right. So this was sort of motivated by kind of a famous chart, Mark Perry and other people, you can find it on Twitter, which shows that, since the 1950s, there’s been tremendous decreases in the prices of things like home appliances. The very first microwave, around 1950, it cost over $10,000 and you literally needed plumbing in order to make it work. Now you can buy a great microwave for 75 bucks. And in general, home appliances, they’ve fallen by a factor of more than four, clothing and shoes are way down, telecommunications services are way down, I think we’re all familiar with that. On the other end of the spectrum, however, you see tremendous increases in prices for things like higher education, lower education, and, of course, healthcare.

02:06 Alex Tabarrok: Higher education has gone up by more than a factor of four since the 1950s. Healthcare, of course, is up. Professional services are up as well. So, the question motivating our short book or booklet was, why the difference? Why are some of these prices going up? Because that, at least, when you first look at it, that seems anomalous. That seems weird. Right? I mean, capitalism should be driving prices down. Falling prices, increasing productivity, technological advances, that’s the norm. That is what we sort of expect. So it seems, at least on first glance, that there must be something wrong with these sectors. And so we try and look at that question.

02:53 Paul Matzko: Yeah, there’s a kind of intuitive logic, too. Well, if we do things better, we make better widgets, we make… Everything gets better. Smartphones get better over time, our cars get better over time. Why don’t we see that across all sectors? That kind of level of improvement. And that makes intuitive sense but what I found really compelling about the explanation in your book for this discrepancy, you point to something called the Baumol effect ? I think I’m pronouncing it right, the Baumol effect. What is that and where does it come from?

03:27 Alex Tabarrok: Right. So, William Baumol was a economist. He should have won the Nobel Prize. He got a little bit too old and died before he could win.

03:36 Paul Matzko: Oh, poor guy.

03:38 Alex Tabarrok: But he was definitely a… He was a contender.

03:41 Paul Matzko: Okay.

03:42 Alex Tabarrok: And that’s where the name comes from. So, the Baumol effect is… There’s a kind of a famous example. It may not actually be the best example but it’s a kind of famous example and I’ll just step through it briefly. And that is, the string quartet example. So you take a string quartet in 1826, it takes four people 40 minutes to run through a Beethoven’s string quartet. 2019, it still takes four people 40 minutes to run through the same string quartet. So in over 200 years or almost 200 years, there’s been zero increase in productivity in string quartet performances. And yet, at the same time, of course, we know that in other areas of the economy, there’s been huge increases in productivity.

04:35 Alex Tabarrok: Now the key is to think of what that means for wages. So you go back to 1826 and the average wage is like a dollar an hour because people just aren’t that productive at producing things. You come to 2019 and the average wage is about $25 an hour. So what this means is that, in 1826, to pull those four workers away from other areas of the economy, you had to pay them $4 in total, a dollar each. In 2019, to pull those same four workers from the other areas of the economy, you’ve gotta pay them $100. So the wages of these four workers and the cost of producing the string quartet has gone up by a factor of 25. Productivity hasn’t gone up at all, therefore, prices must rise.


05:47 Paul Matzko: So we see this kind of shift from fields, from domains that are good producing intensive, they make stuff, and they’ve become more efficient in their production processes. They can make the stuff cheaper and we are seeing the shift. Those gains are then moving to labor‐​intensive sectors or services‐​intensive sectors, which I guess is like, stuff is becoming cheaper while people and labor is becoming more expensive, which seems great from a humanitarian perspective. I mean, this seems people first, in a sense.

06:28 Alex Tabarrok: Yeah, so we have had a situation where manufacturing, in particular, has gone much much, more productive, and because… In part, because it’s become more productive, there are fewer people. We need fewer people in manufacturing than we used to. And people have moved to services, and services, for a variety of reasons, not all services, by the way, but services for a variety of reasons, productivity growth has been slower. And that’s actually one reason why the economy has become more slowly growing over time, because we’ve shifted away from the fast productivity increasing industries, towards slower productivity increasing industries.

07:13 Paul Matzko: Yeah. One point in the book, you note that one of the other terms, so we can call it the Baumol effect, but it’s often also called the Baumol cost disease, which carries a negative connotation. No one ever says, “Oh a disease is good,” right?

07:27 Alex Tabarrok: Right, yes.

07:27 Paul Matzko: Disease, Eesh! But you make your argument that cost disease is mis‐​named, that is, “Not a disease but a blessing.” What do you mean for that? Flesh that out for us.

07:37 Alex Tabarrok: Right. So what this means is that, any time we have differential productivity increases, that is one industry is increasing at a faster productivity rate than another industry, we’re gonna have changes in relative prices, and this is kind of a deep… A simple but a deep point is that, excuse me. This is a simple but kind of a deep point, it’s that in a capitalist economy, all prices are relative prices. So in a capitalist economy, what happens when the reason why, for example, education has become more expensive, what that means is that we must now give up more other goods in order to get an increase in education. So why do we have to give up more of other goods? Well, simply because we’ve become more productive at producing other goods, right?

08:37 Paul Matzko: Yeah.

08:38 Alex Tabarrok: So that’s why the opportunity cost, the opportunity cost of more education is, we have to give up other goods. And because we’ve become more productive at producing other goods, we have to give up more of them in order to get a little bit more education. But that’s not a bad thing. The fact that we’re becoming more productive at producing cars is a good thing. Exactly right, all prices are relative prices. And so, another way of thinking about this is, when we’re thinking that everything should become cheaper over time, we’re thinking that things should become more affordable over time. And that can be true even as prices go up, because since all prices are relative prices, we cannot have, it is impossible to have all prices falling.

09:34 Alex Tabarrok: All real prices cannot fall, because prices represent opportunity costs and all opportunity costs cannot fall, for the reasons you just stated, if you’re becoming better at producing one thing, then the price of doing the other thing has got to go up. And since prices represent opportunity costs, and all opportunity costs cannot fall, we cannot have all prices falling. What we can have, and in fact, what we primarily see, is that things are becoming more affordable. And this actually is a reason why, even as the price of education has gone up, and as the price of healthcare has gone up, we’ve actually bought more of these things.

10:19 Paul Matzko: It’s contrary to what… So, much of the conversation about education, healthcare, is there is an affordability crisis to use the word you just used. There’s an affordability crisis that’s getting too expensive, that itself is a problem, it’s a cost disease. And so what you’re saying is, we’re thinking of that the wrong way, that what’s going on here with the Baumol effect is that, other more productive fields are freeing up our… In a sense, freeing up resources that can be spent on these lower productivity fields. So maybe, how about walk us through a specific domain. How about education? So flip our understanding on the head from this, every newspaper you open trumpets headlines about the affordability crisis, the student loan debt crisis, etcetera. Help us flip that conversation on its head with the Baumol effect.

11:18 Alex Tabarrok: Right, sure. So I think the case for education is a very good one for the Baumol effect, because think about a teacher in 1950. A teacher in 1950, standing before a classroom, basically talking, lecturing, using chalk. Instead, today, the same teacher would be standing in front of the classroom using a PowerPoint. I’m not that big in improvement and productivity, so I’m gonna say a decline in productivity, right?


11:49 Paul Matzko: Yeah, yeah, right.

11:51 Alex Tabarrok: But we’re not seeing teachers teaching more people in a faster amount of time. So the productivity of education has not gone up really at all. Of course, the productivity of many other things have gone up, so in order to get a teacher into the classroom, and teachers are pretty highly educated themselves, they’re relatively skilled workers. The price of skilled workers in particular, has gone way up, so if you wanna pull a teacher into a classroom, you’ve gotta pull them away from a industry which has increased in productivity, and you’re bringing them into education, where there’s been no increase in productivity or not much increase in productivity.

12:34 Alex Tabarrok: And so, of course, the price of education has gone up. Now, at the same time, we’re richer precisely because we can afford all of these other goods, manufactured goods have become so cheap, we can now afford actually to have more teachers, and in fact that’s what we have. We have since the ‘1950s, more than doubled the number of teachers per student, and that’s a good thing, or it may be a good thing, but it’s a reflection of the fact that we are actually richer so we were buying more of these goods, even as the price goes up. And there’s an interesting aspect of this is that, in the book, we look at a lot of the other kind of popular explanations for why prices have gone up. Things like administrative bloat, we have too many principals, too many administrators.

13:30 Paul Matzko: Yep, yep.

13:31 Alex Tabarrok: Things like, “Oh, it’s the teacher’s unions.” And just very briefly, on teacher’s unions, they probably raise wages in education relative to other fields a little bit but the best evidence is like 5%. Okay, not a huge effect by any means. So we go through a number of these other explanations and find them wanting, but there’s another puzzle and that is this. If any of these other explanations are true, that is, if education is becoming more expensive because of inefficiency or bloat or regulation. If any of those explanations are true, the only rational thing to do is to consume less education, right? If the cost… If the real cost is actually going up, that is the… If you actually have to invest more resources because of inefficiency, you should consume less. The Baumol effect, in contrast, is completely consistent with consuming more even as the price goes up because the Baumol effect happens when we get richer.


14:55 Paul Matzko: So students aren’t responding, it… They wouldn’t be responding rationally to rising real cost of college education because we know two things. First of all, college education is getting more expensive. We also know that it’s as desirable… It’s gotten more desirable over the last 40 years. The percentage of people who go to college and graduate from college keeps going up. So you actually have more demand, not less. Which I… But that wouldn’t be explained by the standard I suppose right and left‐​wing explanations for that.

15:31 Paul Matzko: Now, I’ve used them before like the administrative bloat, the way in which government guaranteed loans and grants lead to tuition inflation. I’ve used all those standard explanations, but I get that point you’re making, which is, look, that wouldn’t explain… If that was all true, if the universities were building useless lazy rivers, entertainment facilities, which they are, but if that was the cause of this increasingly expensive college education, you would expect people to respond by, “I don’t want or need as much of that.”

16:14 Alex Tabarrok: Right, exactly.

16:15 Paul Matzko: Right. Okay, okay.

16:16 Alex Tabarrok: And I think this is one of the… And I agree with you completely. Because all of these other alternative explanations, they’re very intuitive and they’re very plausible, because all of these things are true.


16:28 Paul Matzko: Right, right, right.

16:29 Alex Tabarrok: Right. I teach at a university, and we have two Olympic size swimming pools.

16:37 Paul Matzko: Nice, but do you have a lazy river yet though Alex?

16:39 Alex Tabarrok: We do not have a lazy… We do not have a lazy river. I don’t think.

16:43 Paul Matzko: That’s right.

16:45 Alex Tabarrok: Or climbing walls, but it’s a pretty nice place to work.

16:48 Paul Matzko: Yeah, yeah.

16:48 Alex Tabarrok: That’s for sure. And it is true probably, there are too many administrators alright. But first of all, all of these things were true 50 years ago as well, and people were complaining about them as well. Complaining about people in the ivory tower, eggheads in the ivory tower and coddled students who are having a free ride. These complaints go back as long as education does. So as I say in the book, if you think it’s lazy rivers, the rivers have to be getting longer and lazier every single year in order to explain this very long‐​term steady increase in prices. This appears to be something which goes back, I mean in our data, we take it back to 1950, but actually for at least 100 years, prices have been rising steadily year after year and it doesn’t seem to change very much what particular policies we have, which is not to say we couldn’t do better.

18:02 Paul Matzko: Sure, yeah.

18:03 Alex Tabarrok: But it’s not the explanation for this very long‐​term rise in prices over time.

18:11 Paul Matzko: So there are a couple of responses that I think are baked into society. The economy to Baumol Cost effect is something gets more expensive relative to things being produced by a high productivity sector. People naturally are going to look for substitutes, often technology‐​based substitutes. And those substitute technologies move from the high productivity to low productivity sectors. I suppose an example would be with our string quartet… Listening to live string quartet music is getting more expensive relative to a century ago. While what’s a substitute that’s maybe… It’s not the same. It’s not exactly same but I guess like it’s recorded music, CDs, Spotify. Flesh out a little bit more for us. What role the substitutes play and how far does that go and what don’t subs… So walk us through the substitute question.

19:11 Alex Tabarrok: Yeah. Yeah, if we can find a technological substitute for one of these goods which has become more expensive, that’s great. And recorded music, absolutely, that’s superb. As you pointed, it’s not quite a sub… It’s not a perfect substitute because people are still willing to pay to hear live music. So it can’t be a perfect substitute, but it is good. And we see this, we mentioned home appliances. It used to be the case that people, fairly wealthy people, would have domestic servants, right?

19:49 Paul Matzko: Mm‐​hmm, mm‐​hmm.

19:49 Alex Tabarrok: And that’s just not true any longer. They can’t afford them, right?

19:52 Paul Matzko: Yeah. Right, right.

19:55 Alex Tabarrok: And one reason is they can’t afford them, but the other reason is because the substitutes, the dishwasher, the washing machine, have become very good substitutes. Right?

20:05 Paul Matzko: Mm‐​hmm.

20:08 Alex Tabarrok: And so we hope for that, in some sectors of the economy, we just haven’t found good enough substitutes. I’m a big fan of online education. I do quite a bit of it myself at Marginal Revolution University. Get a plug in there.

20:29 Paul Matzko: Yeah, that’s right.

20:32 Alex Tabarrok: And I’d like to see more of it. But for whatever reason universities have been around for over a 1000 years, Oxford is like a 1000 years ago, and they’re basically follow the same model, which is the professor standing in front of the classroom, and lecturing, and that hasn’t changed very much in a 1000 years. So there’s something very powerful there. I don’t… It’s hard to put my finger on exactly what that is. You might have said, “Well books, books are a substitute.” Now you can carry the book, you don’t have to go listen to the professor, you can carry the book with you, and you could read it wherever you want, and it’s much cheaper than having a professor. So once we have the invention of books we won’t need professors lecturing anymore. That, of course, turned out to be false.

21:20 Alex Tabarrok: People thought the same thing about a television, and that turned out to be false. Radio, there was a big effort when radio was invented to broadcast lessons, broadcast these lessons throughout the world, and again that failed as well. So there are some goods where for a variety of difficult to articulate reasons, it’s just hard to substitute away from that labor component. And I think education is one and healthcare is another.

21:54 Paul Matzko: One of the things that’s really interesting about this too like speaking of education, health care, so a very common conversation in tech innovation circles is to where this understanding of the Baumol effect is not robust is to think that, okay, for tech folks will look at these sectors, look at education, look at healthcare, and they see the rising cost, and they see that, “Well that must mean a market opportunity for us, like this must be a field that is ripe for innovation,” and then they propose a variety of changes that typically fail or underperform. These sectors have been the graveyard of many an entrepreneur who made their first fortune in Silicon Valley doing tech and manufacturing productive efficiency type gains. And then they tried to… But is that a mistake, because they’re thinking about those fields and the Baumol effect in the wrong way?

22:52 Alex Tabarrok: Yeah, I think possibly. So they may be thinking about it in terms of there’s some inefficiency there and here’s a way to eliminate that inefficiency, and I think that’s probably incorrect. However, I’m very supportive of the idea that really the only thing which in the long run is going to reduce prices in these industries is if we can find a way of making skilled labor more productive or substituting for skilled labor, and I think artificial intelligence and robots are really the best hope we have, because they’re the closest substitute that we have for skilled labor, and so if there’s a way… If there’s a way forward, that is the way to do it to have some productivity improvements in these sectors and artificial intelligence and robots are probably the closest thing that we have so far.

23:56 Paul Matzko: You give examples in the book using machine learning to replace radiologists or to well at least supplement. You wouldn’t need as many radiologists if you have the AI screening scans of people’s lungs for lung cancer that kind of thing.

24:11 Alex Tabarrok: Right, right, and we have done this in several fields, field which comes up often is laser eye surgery where prices have come down and people often point to this. Indeed, I think I started this meme many years ago in [24:30] ____ revolution, and I said, “Look laser eye surgery. It’s typically not insured, and that’s why it’s unsubsidized, and that’s why prices have come down. It’s more market, it’s more sold in a market.”

24:45 Paul Matzko: Yep, I’ve used it too. Yep.

24:46 Alex Tabarrok: Yeah, yeah, which I think is true. But thinking about it in the context of the Baumol effect, I think what is also the case is that a huge portion of laser eye surgery is the machine, is the laser. Right? And actually the doctor sees you for five minutes, right? And the laser goes at it, measures your eye, and zap, zap, zap, zap, and the slice and dice, and whatever, and it happens very, very quickly. And most of the work which has gone into reducing the price has been making the lasers better, and more accurate, and so forth. And so, that’s a case where there’s been a big machine component to the price, and that’s the part I think, which has fallen price over time.

25:39 Paul Matzko: Alex, how dare you undermine my preconceived libertarian explanation senses? It’s horrible though.

25:45 Alex Tabarrok: Well mine too, mine too.

25:46 Paul Matzko: Yeah, yeah. Well and it makes sense then that so, fields, medical fields that are more diagnostic, diagnosticians that hasn’t been mechanized in nearly to the same degree now. It might happen in the future here with machine learning, and AI, but that’s more labor intensive rather than machine based, so that makes sense. I find that quite convincing. Now there’s something else I wanna ask you on this line. So if this all happens, if there are ways using artificial intelligence and the like to replace skilled professors, education professionals, doctors my concern or my interest here… We interviewed an economic historian named Carl Frey, about his book, The technology Trap recently, and he basically was applying the observations from the first industrial revolution to the coming wave of automation replacing white collar jobs in the future.

26:53 Paul Matzko: But in there, he talks about the risk of essentially a second [27:01] ____ that as white collar skilled workers are replaced it leads to much more unrest backlash pushback, ludditism, than when working class workers, factory workers are replaced that the replacement of blue‐​collar labor versus white collar labor. So my question here is this do you see a similar issue that as these sectors that had relatively low productivity growth over the last generation or so as they now are going to be exposed potentially to labor replacing innovation, do you expect that to be a cause of pushback, a cause of backlash against productivity reforms in those sectors?

27:49 Alex Tabarrok: So yes and no. Let me start with the no, at the present point in time, I don’t see evidence for this. And I see most of the social backlash as coming for low growth, not from high growth, right? So if the robots really were taking over, we would see massive improvements in productivity and great increases in output. And that is not what we’re seeing, to use my colleague’s phrase, we’re in the great stagnation. Productivity growth is actually down. And I think more social unrest comes from low growth than comes from high growth because when you have high growth, there’s always enough to go around and people are less concerned with distributing the pie when the pie is growing already. It’s when the pie is fixed in size or even shrinking that people become much more concerned about dividing the pie. And it’s when you divide the pie that you have social conflict because if somebody’s getting more somebody else must be getting less if the pie isn’t growing. So at the present point in time, I don’t see the danger of the robots here in the United States.

29:14 Alex Tabarrok: Now, elsewhere, in the world, I do think that there are significant problems. If you take a country like India for example, India just because of population growth, it needs a million new workers… A million new jobs every month and because manufacturing has become more automated and has become cheaper and is not growing, we need less of it, the route which countries in the past, like Japan and then Korea and then China, the route which those countries took to become developed… More developed countries working through manufacturing and exports that is becoming less available to the developing countries today and so that I think could be a problem. So the robots, I don’t see as a problem for us, but they are a problem for India, let’s put it that way.

30:19 Paul Matzko: Innovation always comes with, even if it has medium or long‐​term benefits which I generally believe I’m pro‐​growth or pro‐​innovation here, it does come with on the short term an social and economic disruption, there are people who lose in the short term even if on the net as a society, we gain in the long term. And people who don’t like disruption in general, there’s an… We like stability and the status quo. It’s comfortable often, but do you believe that we have… Does this generation have an ethical obligation to future generations to pursue growth in labor productivity even if it comes at the net cost to ourselves, in terms of that disruption? And how does the Baumol effect play into that kind of future ethical obligation?

31:10 Alex Tabarrok: Yeah, so my colleague Tyler Cowen, has a book which I highly recommend, Stubborn Attachments. And yeah, he makes the argument that future people ought to count in our Calculus, in our benefit cost calculations much more than the political process allows for. There’s just no reason to discount future lives compared to our lives today. And so this, yes, does put a lot of weight then on saying innovation, it’s even better than we thought because it’ll benefit most of us but it’ll really benefit future people, right?

31:53 Paul Matzko: Yeah, yeah.

31:54 Alex Tabarrok: Our children and our grandchildren. So, yeah, I take that argument very seriously.


32:04 Paul Matzko: At building tomorrow we’re a tech‐​focused podcast, and we are pro‐​innovation and pro‐​growth. As I was thinking through the implications of the conversation with Alex, that we just had, I started to wonder if understanding the Baumol effect was key really, to properly appreciating how much technological innovation has done to improve all our lives. We mentioned this during the interview that it’s easy for us to appreciate just how much better our lives are when our stuff gets cheaper, better phones, cheaper phones, better cars, cheaper cars. But the problem is that’s being waged against the downsides of innovation, all that social and economic disruption. And a nicer phone compared to, “I can’t afford healthcare,” doesn’t feel like a fair trade‐​off for a lot of Americans.

32:51 Paul Matzko: It’s easy to see where the anti‐​tech backlash is coming from. But what’s less intuitive, but just as important is that we need to realize that the wealth created by gains and productive efficiency are spilling over to benefit people who don’t work in tech. And in ways that don’t just involve new gadgets. The upside is not just a nicer or better or cheaper smartphone. The upside is also that doctors are making more money, teachers are making more money, social workers, violinists, you name it, through no effort on their own people who aren’t in tech are being better compensated because we as a society have embraced technological innovation despite the disruptions.

33:30 Paul Matzko: And I think if we properly appreciate that fact, there would be less kind of anti‐​tech backlash in American society today. So we can thank the Baumol effect for making all of our lives better, not just giving us nicer stuff and neater gadgets and toys.


33:48 Paul Matzko: Thanks for listening, Building Tomorrow is produced by Tess Terrible. If you enjoy Building Tomorrow, please subscribe to us on iTunes or wherever you get your podcast. If you’d like to learn more about libertarianism find us on the web at www​.lib​er​tar​i​an​ism​.org.