You might be interested in Fourastié's model (which directly inspired Baumol, as he himself acknowledged), as it also includes a demand component. This captures the idea that as people become wealthier, they want to consume more services (and relatively fewer goods), including those for which the State is a key provider such as education and healthcare. https://read.dukeupress.edu/hope/article-abstract/52/1/101/140690/Technical-Progress-and-Structural-Change-in-Jean
Maybe I am missing something, but what is the justification for assuming that public sector services are less 'progressive' than the economy in general? Postal services and rubbish collection, for example, can be extensively automated. Amazon deliver stuff by drone in some countries. Even healthcare doesn't seem like it should be immune to automation, but, leaving that aside, there have been lots of advances in medical technology that must make it more 'productive' (i.e. more effective at saving/extending lives per unit of expenditure).
Maybe you will say that, while these industries are not inherently 'non-progressive', the fact that they are publicly run makes them non-progressive and hence subject to cost disease, but that just seems like a roundabout way of saying that public services are wastefully run.
As I explain in the post, the distinction between a "progressive" sector with productivity growth and a "non-progressive" sector with no productivity growth is a simplification of the model to make it easier to explain the phenomenon, but in truth even the most paradigmatic examples of "non-progressive" sectors have experienced some productivity growth. All that is required for Baumol's cost disease to exist is that productivity growth be unequal across sectors.
In the case of government-provided sectors, I don't think anyone denies that productivity couldn't be raised or even that it hasn't in fact been raised to some extent, but in most cases productivity growth has clearly been more limited than in manufacturing and I don't think it's just because the government is not as efficient as the private sector. It could be that technological innovation, such as AI, will change the situation by allowing very rapid productivity growth in sectors where it has been slow up until now, but this hasn't happened yet.
Why is productivity growth obviously higher in manufacturing than other sectors? Are we living through a second industrial revolution? Amazon doesn't manufacture very much, but they demonstrate enormous productivity growth. On top of that, manufacturing is a small proportion of most westenr economies (16% in the US), so even if it was true, it doesn't seem obvious that this would be a catalyst of cost disease in the public sector. On top of that, productivity has been flat in the UK and Spain 2005-2020 and grew significantly in Germany, so, according to your model, public services should have got worse in Germany, but not in the UK, which is ... not evidently the case.
In the rest of your post, you do a bit of Maths, but at the key point you just assume it is obvious when, not only is it not obvious, it doesn't even appear to be true. Most public services are not irreducibly labour intensive in the way music concerts are. If productivity in them is not rising in line with priductivity in the general economy (which, to reiterate, in many western countries isn't even rising) it seems that they are very poorly run.
Again, the distinction between a "progressive" sector and a "non-progressive" sector is a simplification, just as the claim that productivity is growing quickly in manufacturing but slowly in services, obviously manufacturing and services are highly heterogeneous with respect to productivity growth, but the claim is true on the whole and it's a good first approximation that is useful to explain the general phenomenon.
The reason why I didn't dwell on the claim that the kind of goods and services produced or purchased by the government have low productivity relative to other sectors is because it's not really controversial. For instance, in the US, education, health care and social assistance have had negative growth since the 1970s: https://www.brookings.edu/articles/explaining-a-productive-decade/. If you look at table 5 in this paper, you'll see that my admittedly simplified dichotomy between manufacturing and services is basically right. You may not realize that productivity has growth a lot in manufacturing, but it has, at least relative to most services.
It's also not the case that, for Baumol's cost disease to exist, manufacturing or indeed the "progressive" sector as a whole be a large proportion of the economy. In fact, Baumol's model predicts that the share of the "progressive" sector in both employment and GDP will go down over time, which is exactly what happened! All that's required is for the labor market not to be completely segmented, which is clearly not the case.
"For instance, in the US, education, health care and social assistance have had negative growth since the 1970s"
Let's walk through this. All of these are examples of public sector services. We all agree that public sector services have low productivity growth. Your claim is that this is because productivity growth is inherently low in the service sector (and not because the public sector is badly run, even though many government services aren't notably labour intensive), but, in order to demonstrate this, you need to point to low productivity growth in non-manufacturing industries that are not part of the public sector.
In the table you point to, average productivity growth in durable goods was 3.57, 7.69 and 6.04 between the years 1998-95, 95-2000, and 2000-5 respectively. In nondurable goods it was 2.26, 1.78, and 4.26. Conversely, wholesale trade was 2.24, 5.41, and 3.64 and Information was 3.7, 2.48 and 8.85, so roughly the same as manufacturing. Transportation and warehousing was 3.00, 2.48 and 2.12, so a bit less than manufacturing. What exactly is garbage collection except moving things around and putting them in holes in the ground? The fact that government services have declining productivity growth under conditions of widespread productivity growth is not obviously explicable by any other factor than that they are publicly owned, which is what you would predict anyway by reading economics books from a 100 years ago.
This is actually pretty obvious. If anything should be subject to cost disease it is the restaurant industry. Do people across the western world habitually complain that restaurants have got worse, that they can't find a restaurant, or that restaurants are too expensive for them to afford?
I work in a school. We have a decked out computer room that is hardly ever used, a bunch of 3D printers which I think are literally never used, and an army of therapists and support teachers coming in and out to manage children who literally just need a smack. But spiralling costs in education can't be because we are spending lunatic sums of money on useless rubbish. No! It must be something really complex and hard to understand!
Nevertheless, I congratulate you on actually making an effort to defend the assertions in an article rather than just repeating that it's obvious and/or declaring you will write an article about it and never doing it. This must represent a great feat of self denial on your part.
Again, for the umpteenth time, the manufacturing vs. services dichotomy is a simplification. There is obviously a lot of heterogeneity within each category with respect to productivity growth, but on the whole the claim is clearly true, it has obvious implications for government-provided services and again that's not really controversial.
I don't know about garbage collection specifically, because I haven't studied the issue, but in general we should expect to see less productivity growth in the kind of activities performed by the government for that reason and indeed that's the case when you look at how costs have evolved in countries like the US where more of the relevant services are provided by the private sector instead of the government.
While some of the activities in the "education, health care and social assistance" category are performed by the government, a lot of them are done by the private sector in the US and costs in the private sector have also exploded for that kind of activities just as they have in the public sector and often even more.
In particular, in the case of health care (by far the largest activity in that category), the vast majority of those services are provided by the private sector in the US because there is no NHS over there and costs have still exploded. In countries where a larger share of medical services are provided by the government, they will increase for the same reason, even if there is more waste and mismanagement in the public sector.
(By the way, I don't deny that there is more waste and mismanagement in the public sector on average, but it doesn't follow that productivity will be lower in the public sector because it also has advantages such as the fact that it can more easily achieve economies of scale. It really depends on the details and the inference you're making is very simplistic.)
The same thing is true for education, a very large government expenditure pretty much everywhere in the world, where costs have exploded in the private and public sector alike. You're harping on stuff like the cost of 3D printers and fancy computer rooms, but that's just dumb because the overwhelming majority of costs in education comes from teacher salaries and benefits.
There are decades worth of research showing that the Baumol effect can explain the data very well, it doesn't explain everything obviously because no single theory ever does but still nothing else comes even close to explaining as much as it does, but here you are explaining that it must be because of 3D printers and therapists or whatnot...
There is a reason why even libertarians think that Baumol's cost disease explains so much. For instance, if you read the paper Tabarrok co-wrote with Helland on the topic (which is cited in the paper I linked to above), you'll see that they go over most of the kind of things you are talking about and conclude that they can't explain that much whereas the Baumol effect can. That's not because Tabarrok is a fan of the government!
What's funny is that the main point of my post consisted in showing that it's not as obvious as many people think to what extent Baumol's cost disease explains the perceived decline of public services, so it's not as if I had a simplistic view of the issue, but here I am having to defend the obvious fact that it's still a significant part of the story because you do.
"As Random Critical Analysis convincingly argued, this is certainly true for health care, which even in the US is largely provisioned by the government."
"In particular, in the case of health care (by far the largest activity in that category), the vast majority of those services are provided by the private sector in the US because there is no NHS over there and costs have still exploded."
Which one of these is going to be your argument going forward, or are you just going to switch arbitrarily?
An example would be infrastructure: the UK may be unusually bad at this, but maintaining public infrastructure, let alone building new, seems to be in crisis across the West. But productivity would seem to have enormously increased: we have modelling technology and machinery that the people who built the 1970s motorway system couldn't have dreamed of.
But productivity in construction has not "enormously increased"! In fact the sector is notorious for low productivity growth, if you look at the paper I shared above it has even been negative in the US during the past few decades (at least as of 2005), which is why people are constantly talking about how to increase productivity in construction. It's not clear exactly why that is and I think people debate that a lot, but that productivity growth in the sector has been low relative to other sectors of the economy is not really controversial.
To add an additional point, the way that cost disease is supposed to primarily manifest itself is high labour costs, but UK public infrastructure doesn't even have this because of cheap foreign labour.
When I visited my parents in the UK this summer, I went to the dump and over 9/10ths of the staff were Africans, presumably working for close to minimum wage, but, when I was growing up, we had weekly bin collections, and now they have fortnightly and you are capped at 6 dump visits a month.
First, as I said in response to another comment, if Baumol's cost disease puts upward pressure on costs and the government faces a budget constraint (resulting in a squeeze), it's going to hire less qualified people to contain the costs.
So even putting aside that in many countries, I'm not sure about the UK but that's definitely the case in France, minimum wage has increased faster than wages in general (I also wouldn't assume that those workers are working for minimum wage by the way), labor costs should relate to productivity. If you pay the same price for less productive workers, your labor costs have actually increased, even if the wages you pay workers have not or not that much.
But perhaps even more importantly, the labor costs of government services can't be reduced to the wages of the workers you actually see in the streets, there are also all kinds of support workers such as people in management that are needed and their salaries may have increased a lot more.
To be clear, I'm not saying Baumol's cost disease is the only explanation and in fact I don't believe that for a second, only that it's probably a big part of the story. As I say in the post, there is definitely plenty of waste and mismanagement in government, but also other things like growing regulations that can push productivity down and, since you mention the UK, definitely have over there in a pretty massive way.
=> I will post my twitter comment here too because the comment section feels so lonely otherwise ;-)
Interesting.
However, Baumol disease as an explanation does not really work for sector such as education and healthcare where the number of employees has increased or stayed stable. It could explain the inflation (those people are now payed better), but not the decrease in quality (the absolute amount of resource is the same or higher).
I think a good candidate is increased cognitive sorting, particularly among gender line. For instance, a lot of smartt women who used to be a teacher because it was the only path for them now go to work in well paid private sector position).
In the same vein, low skill immigration plays a role: we can see in France that postal service, delivery service, etc, are more and more done by immigrants with low conscientiousness, low literacy, and low language skills. Unsurprisingly, it leads to a decrease in service quality, since in the past, such service would have been delivered by more skilled natives.
As I said on Twitter, if Baumol is putting upward pressure on costs and people have a hard budget constraint, this will push people to seek ways to contain the rise in labor costs, one of which is to hire less qualified people. So the other phenomena you're talking about are not necessarily unrelated to Baumol's cost disease. To be clear, I'm not saying it's the only explanation, but I think it's almost certainly a big part of the story.
You might be interested in Fourastié's model (which directly inspired Baumol, as he himself acknowledged), as it also includes a demand component. This captures the idea that as people become wealthier, they want to consume more services (and relatively fewer goods), including those for which the State is a key provider such as education and healthcare. https://read.dukeupress.edu/hope/article-abstract/52/1/101/140690/Technical-Progress-and-Structural-Change-in-Jean
Thanks, I had never heard of this work, but it does seem interesting.
Maybe I am missing something, but what is the justification for assuming that public sector services are less 'progressive' than the economy in general? Postal services and rubbish collection, for example, can be extensively automated. Amazon deliver stuff by drone in some countries. Even healthcare doesn't seem like it should be immune to automation, but, leaving that aside, there have been lots of advances in medical technology that must make it more 'productive' (i.e. more effective at saving/extending lives per unit of expenditure).
Maybe you will say that, while these industries are not inherently 'non-progressive', the fact that they are publicly run makes them non-progressive and hence subject to cost disease, but that just seems like a roundabout way of saying that public services are wastefully run.
As I explain in the post, the distinction between a "progressive" sector with productivity growth and a "non-progressive" sector with no productivity growth is a simplification of the model to make it easier to explain the phenomenon, but in truth even the most paradigmatic examples of "non-progressive" sectors have experienced some productivity growth. All that is required for Baumol's cost disease to exist is that productivity growth be unequal across sectors.
In the case of government-provided sectors, I don't think anyone denies that productivity couldn't be raised or even that it hasn't in fact been raised to some extent, but in most cases productivity growth has clearly been more limited than in manufacturing and I don't think it's just because the government is not as efficient as the private sector. It could be that technological innovation, such as AI, will change the situation by allowing very rapid productivity growth in sectors where it has been slow up until now, but this hasn't happened yet.
Why is productivity growth obviously higher in manufacturing than other sectors? Are we living through a second industrial revolution? Amazon doesn't manufacture very much, but they demonstrate enormous productivity growth. On top of that, manufacturing is a small proportion of most westenr economies (16% in the US), so even if it was true, it doesn't seem obvious that this would be a catalyst of cost disease in the public sector. On top of that, productivity has been flat in the UK and Spain 2005-2020 and grew significantly in Germany, so, according to your model, public services should have got worse in Germany, but not in the UK, which is ... not evidently the case.
In the rest of your post, you do a bit of Maths, but at the key point you just assume it is obvious when, not only is it not obvious, it doesn't even appear to be true. Most public services are not irreducibly labour intensive in the way music concerts are. If productivity in them is not rising in line with priductivity in the general economy (which, to reiterate, in many western countries isn't even rising) it seems that they are very poorly run.
Again, the distinction between a "progressive" sector and a "non-progressive" sector is a simplification, just as the claim that productivity is growing quickly in manufacturing but slowly in services, obviously manufacturing and services are highly heterogeneous with respect to productivity growth, but the claim is true on the whole and it's a good first approximation that is useful to explain the general phenomenon.
The reason why I didn't dwell on the claim that the kind of goods and services produced or purchased by the government have low productivity relative to other sectors is because it's not really controversial. For instance, in the US, education, health care and social assistance have had negative growth since the 1970s: https://www.brookings.edu/articles/explaining-a-productive-decade/. If you look at table 5 in this paper, you'll see that my admittedly simplified dichotomy between manufacturing and services is basically right. You may not realize that productivity has growth a lot in manufacturing, but it has, at least relative to most services.
It's also not the case that, for Baumol's cost disease to exist, manufacturing or indeed the "progressive" sector as a whole be a large proportion of the economy. In fact, Baumol's model predicts that the share of the "progressive" sector in both employment and GDP will go down over time, which is exactly what happened! All that's required is for the labor market not to be completely segmented, which is clearly not the case.
"For instance, in the US, education, health care and social assistance have had negative growth since the 1970s"
Let's walk through this. All of these are examples of public sector services. We all agree that public sector services have low productivity growth. Your claim is that this is because productivity growth is inherently low in the service sector (and not because the public sector is badly run, even though many government services aren't notably labour intensive), but, in order to demonstrate this, you need to point to low productivity growth in non-manufacturing industries that are not part of the public sector.
In the table you point to, average productivity growth in durable goods was 3.57, 7.69 and 6.04 between the years 1998-95, 95-2000, and 2000-5 respectively. In nondurable goods it was 2.26, 1.78, and 4.26. Conversely, wholesale trade was 2.24, 5.41, and 3.64 and Information was 3.7, 2.48 and 8.85, so roughly the same as manufacturing. Transportation and warehousing was 3.00, 2.48 and 2.12, so a bit less than manufacturing. What exactly is garbage collection except moving things around and putting them in holes in the ground? The fact that government services have declining productivity growth under conditions of widespread productivity growth is not obviously explicable by any other factor than that they are publicly owned, which is what you would predict anyway by reading economics books from a 100 years ago.
This is actually pretty obvious. If anything should be subject to cost disease it is the restaurant industry. Do people across the western world habitually complain that restaurants have got worse, that they can't find a restaurant, or that restaurants are too expensive for them to afford?
I work in a school. We have a decked out computer room that is hardly ever used, a bunch of 3D printers which I think are literally never used, and an army of therapists and support teachers coming in and out to manage children who literally just need a smack. But spiralling costs in education can't be because we are spending lunatic sums of money on useless rubbish. No! It must be something really complex and hard to understand!
Nevertheless, I congratulate you on actually making an effort to defend the assertions in an article rather than just repeating that it's obvious and/or declaring you will write an article about it and never doing it. This must represent a great feat of self denial on your part.
Again, for the umpteenth time, the manufacturing vs. services dichotomy is a simplification. There is obviously a lot of heterogeneity within each category with respect to productivity growth, but on the whole the claim is clearly true, it has obvious implications for government-provided services and again that's not really controversial.
I don't know about garbage collection specifically, because I haven't studied the issue, but in general we should expect to see less productivity growth in the kind of activities performed by the government for that reason and indeed that's the case when you look at how costs have evolved in countries like the US where more of the relevant services are provided by the private sector instead of the government.
While some of the activities in the "education, health care and social assistance" category are performed by the government, a lot of them are done by the private sector in the US and costs in the private sector have also exploded for that kind of activities just as they have in the public sector and often even more.
In particular, in the case of health care (by far the largest activity in that category), the vast majority of those services are provided by the private sector in the US because there is no NHS over there and costs have still exploded. In countries where a larger share of medical services are provided by the government, they will increase for the same reason, even if there is more waste and mismanagement in the public sector.
(By the way, I don't deny that there is more waste and mismanagement in the public sector on average, but it doesn't follow that productivity will be lower in the public sector because it also has advantages such as the fact that it can more easily achieve economies of scale. It really depends on the details and the inference you're making is very simplistic.)
The same thing is true for education, a very large government expenditure pretty much everywhere in the world, where costs have exploded in the private and public sector alike. You're harping on stuff like the cost of 3D printers and fancy computer rooms, but that's just dumb because the overwhelming majority of costs in education comes from teacher salaries and benefits.
There are decades worth of research showing that the Baumol effect can explain the data very well, it doesn't explain everything obviously because no single theory ever does but still nothing else comes even close to explaining as much as it does, but here you are explaining that it must be because of 3D printers and therapists or whatnot...
There is a reason why even libertarians think that Baumol's cost disease explains so much. For instance, if you read the paper Tabarrok co-wrote with Helland on the topic (which is cited in the paper I linked to above), you'll see that they go over most of the kind of things you are talking about and conclude that they can't explain that much whereas the Baumol effect can. That's not because Tabarrok is a fan of the government!
What's funny is that the main point of my post consisted in showing that it's not as obvious as many people think to what extent Baumol's cost disease explains the perceived decline of public services, so it's not as if I had a simplistic view of the issue, but here I am having to defend the obvious fact that it's still a significant part of the story because you do.
"As Random Critical Analysis convincingly argued, this is certainly true for health care, which even in the US is largely provisioned by the government."
"In particular, in the case of health care (by far the largest activity in that category), the vast majority of those services are provided by the private sector in the US because there is no NHS over there and costs have still exploded."
Which one of these is going to be your argument going forward, or are you just going to switch arbitrarily?
An example would be infrastructure: the UK may be unusually bad at this, but maintaining public infrastructure, let alone building new, seems to be in crisis across the West. But productivity would seem to have enormously increased: we have modelling technology and machinery that the people who built the 1970s motorway system couldn't have dreamed of.
But productivity in construction has not "enormously increased"! In fact the sector is notorious for low productivity growth, if you look at the paper I shared above it has even been negative in the US during the past few decades (at least as of 2005), which is why people are constantly talking about how to increase productivity in construction. It's not clear exactly why that is and I think people debate that a lot, but that productivity growth in the sector has been low relative to other sectors of the economy is not really controversial.
To add an additional point, the way that cost disease is supposed to primarily manifest itself is high labour costs, but UK public infrastructure doesn't even have this because of cheap foreign labour.
When I visited my parents in the UK this summer, I went to the dump and over 9/10ths of the staff were Africans, presumably working for close to minimum wage, but, when I was growing up, we had weekly bin collections, and now they have fortnightly and you are capped at 6 dump visits a month.
First, as I said in response to another comment, if Baumol's cost disease puts upward pressure on costs and the government faces a budget constraint (resulting in a squeeze), it's going to hire less qualified people to contain the costs.
So even putting aside that in many countries, I'm not sure about the UK but that's definitely the case in France, minimum wage has increased faster than wages in general (I also wouldn't assume that those workers are working for minimum wage by the way), labor costs should relate to productivity. If you pay the same price for less productive workers, your labor costs have actually increased, even if the wages you pay workers have not or not that much.
But perhaps even more importantly, the labor costs of government services can't be reduced to the wages of the workers you actually see in the streets, there are also all kinds of support workers such as people in management that are needed and their salaries may have increased a lot more.
To be clear, I'm not saying Baumol's cost disease is the only explanation and in fact I don't believe that for a second, only that it's probably a big part of the story. As I say in the post, there is definitely plenty of waste and mismanagement in government, but also other things like growing regulations that can push productivity down and, since you mention the UK, definitely have over there in a pretty massive way.
"there are also all kinds of support workers such as people in management that are needed and their salaries may have increased a lot more."
You're getting close! Just remove the word 'needed'!
=> I will post my twitter comment here too because the comment section feels so lonely otherwise ;-)
Interesting.
However, Baumol disease as an explanation does not really work for sector such as education and healthcare where the number of employees has increased or stayed stable. It could explain the inflation (those people are now payed better), but not the decrease in quality (the absolute amount of resource is the same or higher).
I think a good candidate is increased cognitive sorting, particularly among gender line. For instance, a lot of smartt women who used to be a teacher because it was the only path for them now go to work in well paid private sector position).
In the same vein, low skill immigration plays a role: we can see in France that postal service, delivery service, etc, are more and more done by immigrants with low conscientiousness, low literacy, and low language skills. Unsurprisingly, it leads to a decrease in service quality, since in the past, such service would have been delivered by more skilled natives.
As I said on Twitter, if Baumol is putting upward pressure on costs and people have a hard budget constraint, this will push people to seek ways to contain the rise in labor costs, one of which is to hire less qualified people. So the other phenomena you're talking about are not necessarily unrelated to Baumol's cost disease. To be clear, I'm not saying it's the only explanation, but I think it's almost certainly a big part of the story.