Robert Allen: ‘We Have The Service Revolution, And How It Is Going To End Is a Big Question.’

– Thank you very much for agreeing to speak with us. Your research is mostly devoted to the explanation of diverging paths of economic development of different countries. According to the data that you have been collecting for over a decade now there were times when real wages grew in line with productivity and those when that was not the case. So what is the explanation that you propose to these patterns in the data? How are wages, technology, and globalization interlinked?

– I’ve been studying this – collecting data on this – since the 1980s actually, so it’s quite a long time that I’ve worked on this subject. When I think about it I think about the last four hundred years, and when I say this I’m thinking really about the industrially advanced country at the time. So for the first couple of hundred years that was Britain that had the Industrial Revolution, and by the late 19^th century the US became the leading country.

In these last four hundred years, I think, there are four periods. So you go from 1620 up to 1770, that’s the run up to the Industrial Revolution, and in that period you’ve got output per worker rising and you’ve got real wages generally rising for almost everybody, men and women too, converging upwards.

Then in the Industrial Revolution output per worker keeps going up too, but you get flat average real wages and you get huge divergence in the labor market with some workers doing really well and some workers doing really badly. Then from the mid 19^th century onward you get this great boom that’s in Britain and it’s also in the US. And the US boom lasts until the 1960s. There are interruptions with the Depression in the 1930s but basically all the data show output per worker just going up and up and up and real wages going up and up and up. And then since the 1960s it’s all falling apart again.

There’s a famous economist, [Joseph] Schumpeter, who talked about capitalist development being a perennial gale of creative destruction. I think there’s a lot to that idea, but the thing is that the gale is not perennial. And so in the first period, 1620 up to the Industrial Revolution, there was this continual expansion within the cottage manufacturing sector and also agriculture. Britain was establishing a world empire – big markets that they could export handicraft manufactures like cloth and hardware and stuff to. But wages get high, and inventors have an incentive to save on those costs. So the Industrial Revolution consists of all these inventions that eliminate hand production methods and replace them with machine based factory methods. That’s creative destruction. The creative part is that productivity is going up, the destruction part is that the old system is destroyed, and lots of people lose out really badly. The workers that gain are people that make machinery or that build buildings – capital accumulation’s important.

So that lasts till the middle of the 19^th century, when all the cottage jobs have been wiped out and everybody’s in factories. And then you get these new booms that come along with that sector expanding and expanding and expanding. In the case of Britain they have a world empire that they can feed their products into and they destroy handicraft manufacturing in all the poor countries in the world, and they create underdeveloped countries when they do that. And the US, they’ve got an internal empire, it’s called “West”, so they “settle the West'', and it just causes the whole system… it’s an endless source of growth. This all lasts till the 1960s. What happens then is two things happen. Partly because wages are very high in the US you get this growth of manufacturing production in developing countries, and this leads to deindustrialization in the US.

But the biggest part of the story is that people start shifting their demand away from manufactured goods into services. So there’s little growth, very small growth in the demand for manufactures. Technical change is continuing in the industry, and that just means that industrial manufacturing jobs decline. So output per worker goes up, but output does not go up much, so workers go down. So this shift in demand means that we have the Service Revolution. The service sector is growing and it’s destroying the manufacturing sector. So that’s the creative destruction process that’s happening right now, and how that’s going to end is a big question.

– And how are these historical processes influencing education?

– There’s a lot of discussion about education and growth, and how they interact with each other, and whether different educational policies can address this inequality problem that’s arisen today. One thing is that growth in computerization and so forth has led to greatly increased demand for some kinds of educated labor, for some levels of technical education. You see a big increase in the number of scientific, technical, professional, managerial jobs that require university degrees. So that kind of change has been one that’s increased the demand for education, right? The other thing the service revolution has done is create this huge growth in demand for people with little education at all: waiters, car park attendants, care home workers. So they don’t need education, and it’s not at all clear that educating them is going to either increase their productivity or help them do anything else. If we want to address the problem of low wage work, we have to do it by either increasing minimum wages or doing other direct interventions in labor markets. Simply educating people isn’t going to solve the problem.

– A lot of researchers have pinpointed this institutional aspect of development. Would you say that institutions do play this very crucial role in the developmental process, or is it not the cause but the consequence, in your view?

– A very deep and tricky question. So there’s a lot of people that think that institutions are important, but they don’t agree on what the important institutions are. So this is one thing that makes it a little bit difficult to come to grips with this. So a lot of people have argued that basically a kind of minimal government setup is the right thing: countries should have secure property rights, a state that doesn’t intervene very much in the economy and that lets the free market operate on its own. I don’t really agree with that.

If we look again at the Industrial Revolution, we compare England and France. In France along the Mediterranean they need irrigation for agriculture to work well. And in the 18^th century there’s a lot of irrigation projects that are proposed, and none of them happen because to get the irrigation system built you’d have to run canals across lots of people’s property, and they refuse to comply, they won’t sell the land to the irrigation company, they won’t let them build canals across their land, they go to court to stop this. So they are acting, and they have very good property rights, so they can prevent that.

So in England at the same time there’s a lot of development projects like the enclosure of agricultural land, the construction of canals and so forth – it all goes ahead. And why? It’s because after the Glorious Revolution of 1688 Parliament was ascended, and it had the power to pass laws that overrode property rights. So they passed thousands of these enclosure acts, canal acts, turnpike acts, they built all this infrastructure.

What do I learn from that? Do I learn that what you need for development is actually sort of weak property rights? Like the government actually has to be able to override private property? This is not a democracy either, it’s not inclusive. The members of the Parliament are all very rich landowners, and so what they’re interested in is passing laws that raise the value of land – and in particular their own land. But together they raise land values in the country. But if you think about this from an economic point of view, that’s perfect. Right? If you do cost/benefit analysis of things like irrigation, the standard measure for the benefit of the project is the rise in the value of land created by the project. And so Britain had an institutional setup in which the government was controlled by landowners who authorized all projects that raised the value of land, so they did just what the World Bank would want them to do!

So on the ground it all worked, but it worked because they had political institutions that were not inclusive – the French had that too – but they had powerful enough non-inclusive institutions to override private property. So maybe that’s the recipe for growth?

The other thing that you learn from studying the Industrial Revolution is that the other thing the government should do to cause growth is to be a successful imperialist because establishing a good empire was a key part of what Britain did that both fostered the Industrial Revolution and also allowed them to continue to expand - and expand at the expense of what now are poor countries. So those are institutional lessons that you can draw from history, but they are not very optimistic ones.

– What was the environmental impact of technological progress in the past and is it different today? And what can we do to somehow fix this problem?

– Historically there was a lot of pollution. Pre-industrial cities were terrible from the public health point of view, they were just death traps, but the Industrial Revolution made everything worse by increasing their use of coal. That was entirely a question of costs. The British invented the steam engines to substitute coal for expensive labor. It’s all gotten worse as more countries have industrialized.

Right now, I think, the world is kind of trapped between two externality issues. “Externalities” is this term in economic theory for secondary effects produced by a primary activity. So pollution is an example. A firm produces a product, and they sell the product to consumers. But if they can burn fossil fuels and fill the atmosphere with carbon dioxide, they’ll do that if it lowers their costs. So this kind of standard solution to that problem is to put a price on carbon, and then they’ll have an incentive to invent technologies that don’t do that.

The other externality is the creation of knowledge. That’s also an externality but that’s a positive externality, that’s a good thing, especially as knowledge has this public good property that means that we all can consume it. If I buy an apple and eat the apple, then you can’t eat the apple because it’s all gone. But If I was Pythagoras and invented his theorem and gave it to you, you could use this theorem to prove something, but somebody else could use this theorem too. Your using it doesn’t preclude that. That means that it is very hard to turn knowledge effectively into private property.

I think it’s a generalization that’s probably true that all basic research in sciences and also basic industrial research is financed by the state, one way or another, or by charities. Dealing with the environment raises all sorts of questions, like what kind of fuel can we use to power airplanes. Private companies... they’re never going to reap the rewards of these inventions, so their incentive to do it isn’t as great as what we need. So I think I’m very much in favor of publicly funded research to address these big questions. We see this in health, right? The covid epidemic: the vaccines development, it’s all paid for by public money. The role of private pharmaceutical companies is simply to commercialize vaccines that have been paid for by the public and, you know, manufacture them and distribute them.

– You have co-authored a few studies on the whaling industry, which seems like a pretty peculiar subject to look into. So what were your main conclusions and, I guess, what can they teach us about today’s economic reality, can we draw some useful results from them and use them today?

– So this was a project about the extinction of the bowhead whale in the Eastern Arctic. Bowhead whales…there were a lot of them in the past, and they have them in the Western Arctic still, but they all get wiped out in the Eastern Arctic by the Dutch and the English, and then there’s some whaling coming out of places like Bremen in Germany too. What they’re doing is they’re going out and hunting these whales for two products. One is they get the bone, the kind of a bone from the gills of the whale, and that’s used to make corsets for fashionable women. And the other product is very low-grade oil, and it’s low-grade because it gets contaminated with the blood of the whale. It’s used as fuel for lamps in people’s houses, and it stinks ‘cause it’s got all this blood in it.

These whales get hunted to death. And so what it shows is that even with pre-industrial technology human beings have a capacity to destroy the environment. Now of course we have industrial technology, so we can do it on a vaster scale.

The other thing is that you could have saved the whale. It all happens because the environment is treated as this sort of open access fishery – anybody can go out and kill them all. If you model this, you can show: yes, you could have had regulations or you could have had user costs that would have saved the whale. It’s also international, right? You know, there’s a number of countries involved. So I think what it means is that what we need is international cooperation on environmental issues. We really need international agreements at a minimum on carbon taxation, so that firms all around the world have to pay for the carbon they want to dump in the atmosphere (that’s causing global warming). That’s the lesson that I draw from history of the first whale extinction.