Baumol’s Cost Disease Comes to Britain

Britain’s incoming Labour government got a nasty welcoming present in July when independent pay review bodies told ministers that millions of public sector workers should receive a 5.5% pay increase. The cost of this accounts for about half of the £20 billion “black hole” in the budget which Chancellor Rachel Reeves claims to have inherited. 

Some noted that these increases were not conditional on increases in productivity. This is true, but that illustrates a phenomenon which will put increasing pressure on budgets across the developed world. It is known as “Baumol’s cost disease.” 

This states, as economist William J. Baumol explained, that:

…the costs of health care, education, the live performing arts, and a number of other economic activities known as the “personal services” are condemned to rise at a rate significantly greater than the economy’s rate of inflation…This is so because the quantity of labor required to produce these services is difficult to reduce.

To see why, consider two workers; one works in a car factory, the other is a teacher. The car manufacturer’s output of cars per year can be increased with more or better machines or with a more efficient combination of the capital (the machines) and labor (their work) inputs into production (this “entrepreneurship” falls under “Total Factor Productivity”). This increase in productivity will, ceteris paribus, drive an increase in the car manufacturer’s wage. Baumol called this “the progressive sector.” 

For the teacher, it is much harder to increase productivity. Their labor is a much greater share of the factors of production – indeed, the labor largely is the product, or service – and it isn’t as easy to substitute it for capital. A teacher with 20 kids in his or her classroom could send recordings of their lessons to 100 kids to watch at home, but this would be an inferior service. Likewise, we could cram 100 kids into the teacher’s class and increase “kids taught per year,” but, again, this would be an inferior service. Baumol called this “the stagnant sector.”    

This is where the “cost disease” comes in. If we want people to teach, we must pay them at least what they could get in other sectors, and those other sectors include those, like car manufacture, where productivity increases are driving wage increases. Productivity driven wage increases in these other sectors will, then, drag up wages in sectors where there have not been the productivity increases to justify the wage increases by themselves. 

This disease is especially acute in sectors which are labor intensive and cannot be made much less so, like education or personal care. Britain’s pay hikes cover more than 500,000 teachers and about 1.3 million National Health Service staff. We can, and should, embrace measures like telehealth which could increase productivity in the healthcare sector, but these are unlikely to be more than marginal when the service is so inherently “hands on,” or labor intensive.     

‘Stagnant sector’ industries are often government monopolies or include a significant element of government provision. But not always. The cost of the average cremation in Britain increased by 43% between 2011 and 2020, for example, while the overall Consumer Price Index increased by just 16%. When you see that famous chart showing price changes in different sectors of the economy grouped as private or public sector, you might be seeing the effects of the Cost Disease at least as much as the effects of relative government inefficiency. 

Baumol sounded an optimistic note, writing:

The flip side of the cost disease-the near universality of rising productivity in the economy as a whole-means that we can afford healthcare, education, and the other personal services despite their disturbingly persistent rates of cost increase.

“The only thing that will change, in terms of the cost to us,” Baumol writes:

…is how we will have to divide our money among these items.

Table 1 illustrates this. With growth of 1.5% annually, GDP per capita rises from $66,814 to $296,121 by 2023. If health spending rises from 15% of income to 40% over that period, health spending rises from $10,022 to $118,449. But, because of the overall growth in per capita GDP, non-health spending can also increase – from $56,792 to $177,673 – while decreasing as share of all spending – from 85% to 60%. If that growth doesn’t materialize, neither will the wage increases in the progressive sector required to drag up wages in the stagnant sector. “The only thing that will change, in terms of the cost to us,” Baumol writes, “is how we will have to divide our money among these items.” 

One implication of Baumol’s analysis is that the public sector share of GDP will increase dramatically. “The experiences of planned economies indicate that this is not a promising arrangement,” he notes. A gifted economist, Baumol was also a master of understatement. 

 


READER COMMENTS

Warren Platts
Oct 10 2024 at 1:15pm

I can see why economists like to call it a disease, because a disease implies a possible cure. And the obvious cure is to manage for increasing income inequality through rampant immigration and unfettered, unilateral free trade (cf. Bryan Caplan). This of course will raise the real wages of professional economists and others within the same or higher class.

Some, however, would argue the “cure” is worse than the disease. They’d like to call it Baumol’s Blessing since rising real wages equals a rising standard of living. Moreover, as Marriner Eccles pointed out long ago, mass production requires mass consumption. So unless you’re prepared to go full-on mercantilist and hope the ROW can afford to buy your excess production, your own citizens need money in their pockets to justify investing in increased production to satisfy the increased demand. Thus Baumol’s Blessing will actually grow the economy faster than Baumol’s Cure…

Matthias
Oct 10 2024 at 11:25pm

Foreigners are humans, too

Why do you care for the unwashed masses in country A, but not for the poor wretches who happened to be born in country B?

And if restrictions on trade and migration are as useful as you say, why not restrict trade and migration between federal states, too?

This is all very mysterious to me.

Warren Platts
Oct 11 2024 at 1:21pm

Why do you not care for the U.S. working class? And it is obvious you do not because otherwise you would not support policies that put downward pressure on U.S. working class real wages in an effort to cure their cost disease. That sentiment is what is mysterious to me.

Kat Messer
Oct 11 2024 at 1:48pm

I don’t get the sense anyone hates the working class. But there’s a lot of undercurrents going on and questions also persist about why do these specific jobs matter when we can create other ones that we can shift others into. It’s a natural economic cycle, but it seems like some folks just aren’t comfortable with that.

That said, in such an interconnected world, we are able to do more for less, which I honestly see as prepping us for working on the problems more pressing to society. Climate change for example is a real pressing issue that will without constraint likely kill us. When you put things in perspective, that warehouse job can just be handled by a robot so the human can go focus on re-educating themself and finding a new job that matters.

Warren Platts
Oct 12 2024 at 2:09pm

Climate change for example is a real pressing issue that will without constraint likely kill us.

I’d be more worried about nuclear war. It’s much more likely now than it has been in decades thanks largely to China’s expansion of its nuclear program that’s currently affordable thanks to wonderful free trade…

Chinese nuclear weapons, 2024 – Bulletin of the Atomic Scientists (thebulletin.org)

Mike Burnson
Oct 16 2024 at 6:33pm

Kat, with all due respect, your comment about mythical “climate change” is untethered to reality. The total human contribution to so-called greenhouse gases – the physical chemistry is nothing like a greenhouse – is a trifling 1/3 of 1%; 99.7% is nature. To put that into perspective, the sum total of human emissions since the dawn of the Industrial Revolution remains far below a single year of natural activity. Further, carbon dioxide absorption wavelengths are effectively opaque by 300 ppm; anything more is nothing other than plant food.

The only constant about climate is that it is always changing. We are currently below the global temperatures of the Medieval Warming Period or the Roman Warming Period. Human impact is too infinitesimal to even measure with modern instrumentation.

Jon Murphy
Oct 10 2024 at 5:39pm

“The only thing that will change, in terms of the cost to us,” Baumol writes, “is how we will have to divide our money among these items.”

I am reminded of a line from Paul Krugman:

productivity isn’t everything, but in the long run it is almost everything

Warren Platts
Oct 11 2024 at 1:26pm

productivity isn’t everything, but in the long run it is almost everything

Exactly why the USA should manage for — or at least not complain about — high wages for workers because high wages provide a powerful incentive to invest in new, labor-saving devices.

Jon Murphy
Oct 11 2024 at 1:30pm

Agreed.  And it’s precisely why I so stauchly oppose protectionism, whose result is to reduce productivity

Jon Murphy
Oct 11 2024 at 1:41pm

Well, I should say “partially agreed.”  One doesn’t want to just invest in capital for the sake of it.  As Solow showed 70 years ago, that is not the path to the wealth of nations.

Warren Platts
Oct 12 2024 at 8:45am

precisely why I so stauchly oppose protectionism, whose result is to reduce productivity

There’s a lot to unpack there. I guess by “productivity” you are speaking of total factor productivity for the U.S. as a whole. However, I reckon the productivity that really counts in terms of people’s standard of living is labor productivity. You could perhaps show that invention X has a lower total factor productivity even though it doubles labor productivity. But the effect of forgoing X will be to put downward pressure on real wages and hence standards of living.

Jon Murphy
Oct 12 2024 at 9:05am

You could perhaps show that invention X has a lower total factor productivity even though it doubles labor productivity. But the effect of forgoing X will be to put downward pressure on real wages and hence standards of living.

…no.  That is completely backward.

Warren Platts
Oct 12 2024 at 1:32pm

How so? If you’re an owner of capital, what you want is the highest return on your capital that you can get for the smallest outlay of capital that you can get away with. Thus, from the owner of capital’s perspective, cheap, low-wage labor is a beautiful thing because then there is no need to buy expensive labor-saving devices. As a result, the labor productivity is much less, but the total factor productivity is much greater because it’s so much cheaper to buy short-handled shovels compared to steam shovels!

Jon Murphy
Oct 12 2024 at 5:03pm

None of what you say makes economic sense.

john hare
Oct 12 2024 at 5:12pm

Just ignore th millions of jobs that are not done because there ane no low wage people available to do them. And ignore the long term skilled workers stagnating at their positions because they have to sweep the floor instead of building product.

Jon Murphy
Oct 12 2024 at 7:35pm

I’m hung up on how reducing productivity increases productivity. Must be that 5D chess thing I keep hearing about.

Warren Platts
Oct 13 2024 at 12:18am

I’m hung up on how reducing productivity increases productivity.

There are different kinds of productivity. The standard (non-Humpty Dumpty) definition of labor productivity is output per hour of labor. Capital productivity (Q/K) is output per unit of capital input. So if invention X doubles your labor productivity, but requires tripling your capital inputs, then technically your overall productivity would go down. Right?

Whether to buy invention X depends on the wage level. For example, a widget prior to invention X requires 5 capital inputs (measured at $1 each), but requires 1 hour of labor to produce. However, incorporating invention X triples the capital input per widget, but cuts required labor in half.

In this example, if wages are less than $20/hour, you’re better off not buying X and vice versa if wages are greater than $20/hour. However, if wages are $25/hour, you’re even more better off if you can offshore production to China where wages are $8/hour using the primitive technology. This will cure Baumol’s cost disease — unless there’s a tariff or something.

Warren Platts
Oct 13 2024 at 12:32am

Just ignore the millions of jobs that are not done because there are no low wage people available to do them. And ignore the long-term skilled workers stagnating at their positions because they have to sweep the floor instead of building product.

John, if you’re unwilling to pay the market clearing price for labor, then yes of course you’re going to have a hard time filling slots. Moreover, if sweeping chores are taking too much time, you might want to consider investing in an industrial floor sweeping machine like the Uline H-2760 industrial push sweeper. Using it, a guy can sweep 25,000 square feet in an hour. That’s about 10 times faster than doing it manually. But it will set you back $700…

Matthias
Oct 10 2024 at 11:34pm

Another way to look at the ‘cost disease’ mechanism is to say that the real wages of people in the progressing sector grow slower than productivity, and in the stagnant sector they grow faster than productivity.

Comments are closed.

RECENT POST

[Editor's Note: Dear readers, you delighted us with all your responses to Professor Cutsinger's inaugural EconLog Price Theory Problem. Thanks! We reprint that question below, followed by his suggested solution. Next week, we'll post problem #2. Long live price theory!] Question: In his book, Basic Economics...

Read More

Adam Smith famously commented on how specialization increased productivity in a pin factory, where different individuals specialized in each subtask involved in manufacturing even a simple object.I thought of that anecdote when reading Razib Khan's account of the difference between Neanderthals and early modern humans ...

Read More

Britain’s incoming Labour government got a nasty welcoming present in July when independent pay review bodies told ministers that millions of public sector workers should receive a 5.5% pay increase. The cost of this accounts for about half of the £20 billion “black hole” in the budget which Chancellor Rachel Re...

Read More