Romesh Vaitilingam: Welcome to Vox Talks, a series of audio interviews with leading economists from around the world. My name is Romesh Vaitilingam, and today's interview is with the economic historian, Professor Sascha Becker, deputy director of the Centre for Competitive Advantage in the Global Economy at the University of Warwick. We talked about Sascha's research on the importance of education in driving economic development in the Industrial Revolution.

Professor Sascha Becker: So when studying the existing literature on the Industrial Revolution, the thing we noticed is that there is quite a lot of work looking at England and how England industrialized, and most of that work concentrates on the textile sector. Much less work has been done with data that really covers large parts of the country, in continental Europe, and in particular in Prussia, which was one of the first countries on the continent to industrialize. We were kind of wondering how the industrialization there got going and whether education, which nowadays people see as very important to growth and development, had any impact on the adoption of these new technologies that were developed in England. What we found also with the literature on England is that there, people typically don't find education to be a key driver of industrialization, which is kind of puzzling.

But then, when you look more in detail of how these results come about, then the findings are often at a relatively local level. People look at how industrialization got going in Lancashire and Shropshire and have less of a comparison regionally whether areas that had lower literacy levels in, say, the northeast of England would industrialize later also because of these lower literacy levels.

In the Prussian context, we are in the fortunate situation that since the industrialization there started later due to, amongst other, wars the Napoleonic Wars, et cetera, in Prussia industrialization starts after a statistical office has been founded, so that there we are in the fortunate situation that we can look before the industrialization, something we cannot really do with good data in the English context.

So what we did is we got hold of the first census that was ever done in Prussia, in 1816, where we have, for more than 300 counties, information on school enrollment rates. So we know how many kids went to school in the age bracket 6 to 14, which, on paper, was compulsory schooling age. Or let's say there was a recommendation that kids should go to school, but it was not really strictly enforced, so that in some areas there was more of an effort to send kids to school than in others.

So that then, when we look in the middle of the century, around 1850, where we have the first factory census, so we can count how many people work in different kinds of factories, whether those counties where literacy rates were higher or where school enrollment rates were higher were those that had higher rates of industrialization in the middle of the century. We do find these effects, so that higher literacy rates in 1816 matter for industrialization in the middle of the century.

What we also can do, because the data is pretty detailed, is that we can split up the sectors into textiles, metal, fabrication, and all other sectors, which includes stuff like paper, wood, rubber, kind of sectors. And, similar to what people find for England, we do not find education to matter in the textile sector, but it does matter in the other sectors. A little less so in the metal sector, which is also pretty much resource dependent, you can only produce relatively low cost metal kind of fabrics if you have the related primary resources there. Whereas all other sectors that don't depend either on sheep or on metals in the ground, education is a key driver of developing those sectors.

Romesh: How do you think the link actually happens between education and industrialization? Is there some issue around having people who are well educated who are able to work in these industries? Or is there some issue of, in a sense, that the Industrial Revolution started in Britain and there might have been some ability to access some of that know how - that R&D, if you like – and then implement it in Germany?

Sascha: So on that, we think about several possible channels, but data wise we can't pin them down, so we can't say the share going to this particular channel is so and so. But the possible channels we think about is that, on the one hand, the probability of having an entrepreneur, of someone who is driven to develop something and of someone inventing something, is higher the more educated the population is. So, just probabilistically, the likelihood of having a good entrepreneur, an inventor around is higher if you have more educated people. Then, also, reading and writing was probably key to being able to read manuals for these machines. So, very often, machines were shipped over from England and were brought to Prussia. And then you needed people to run them and to really be able to understand also how to fix them when they are broken, and there literacy came in as being important.

But also, literacy was probably key in supplementary sectors, like accountancy, that were key in getting these factories going.

Romesh: In the case of Britain, do you think it's just been a lack of data that we haven't been able to find this link? Or do you think there might have been a difference there in that Britain was the first industrializing country, whereas Prussia was coming along, following on in its wake?

Sascha: So, theoretically, some people say that there is a difference between being the leader and the follower. So we also, in our work, say that there are models where people describe that adopting existing technologies requires relatively basic education, like reading and writing whereas inventing requires the super brain who comes up with a new spinning jenny, a new machine, that kind of type. And in the Prussian context, you only have to be able to adopt that which was already out there. In terms of data, yes, I think the fact that people concentrated on textiles where textiles were, in a sense, a sector where the transition was somewhat smoother. What we also find in the Prussian context is, in the factory data, they have information about people who work at machines, but you still have hand driven looms in the factory. So people would, at some point, move out of their homes, where they would weave, and go to the factories and be side by side to people who operate the machine.

So this sector, in a sense, had a smoother transition, where people learned more, on a step by step basis, these new technologies, whereas it might have been more disruptive in other sectors. In the English context, I recently talked quite a bit with Bob Allen in Oxford, and he says, also, there education may be more at work than we realize so far. So he gives the example of Lancashire, where he says, "Why is it that Lancashire was one of the first areas to industrialize?"

And it turns out, when you go back some centuries, that Lancashire was the hub of the clock making industry. And this knowledge that people had built up, of being able to work with little gears and mechanics, was helpful then, also, to build larger machines. Of course, then you can go back even further and ask, "So why was it that the clock making industry set up there in 1500?"

But then we end up with Adam and Eve. But also there, a bit, the idea is that education was there. There was some human capital that was useful, then, to get industrialized processes going.

Romesh: You've established a clear link between education and industrialization in Prussia, but you're carrying on with this research program. What are the kinds of issues you're looking at in your further work on this data?

Sascha: So, we have looked at other issues that go away from the industrialization in the stricter sense, where we look at the trade off between education and fertility. So, in modern day research, people tend to find, in developing countries, that at some point families have to choose whether they want to have more kids and have too little money to give all of them good education, or to have fewer kids but send them to Harvard. And this is, to some extent, a modern idea, and many historians reject that this kind of trade off was going on in earlier periods of history. And we were interested in seeing whether, maybe already in the 19th century, parents were trading off the quantity and quality of children. This actually has some theoretical background, because of the work that has been done by Oded Galor and co authors on the so called “unified growth theory” tries to explain the development, from the early days up to today, now in a unified framework. Earlier on, people said we had a period in which there was essentially no growth, then the Industrial Revolution came in and suddenly economies in Western Europe started growing. And what Oded Galor and co authors try to say, that this was an evolutionary process that was smoothly gaining ground.

And one key element there is demographics. So, they brought into the whole growth literature the idea that fertility rates matter and the way people also trade off education and fertility. And that's what we then look at, also, with this Prussian data, again, looking at the county level and knowing how many children females gave birth to, and how high the enrollment rates in school were. And we do see the same trade off there that we find in modern day data, so that is the nice piece of evidence that sustains the arguments made in the theoretical literature.

Romesh: Final question, Sascha. As you mentioned, the pressure on developing countries these days is always "Invest in education. Invest in human capital. That is the way to growth." As you mentioned, the English data maybe questions that. Your data seems to suggest that that is a very good idea. What's your perspective on the light that your work can shed on this modern day debate for developing countries?

Sascha: Well, I'm always very reserved when it comes to drawing bold conclusions out of restricted work that I do, but I personally am surprised in how many different contexts, both in modern data and in history, you find education to matter. And this earlier work, where we traced down the developments also across the German lands to the Reformation, where Martin Luther wanted people to read the Bible, and that was kind of a key driver to push in Protestant areas, the first schooling efforts.

And you find those areas that became Protestant earlier on to have higher development levels even today, so after 500 years. And I find it quite striking how important it is. And these effects may sometimes take a while to really take off, but in the long run, they have such a huge effect that it is unwise to stop education efforts today, even if you don't see the immediate payoff. But it does matter so much in the long run.

Romesh: Sascha Becker, thank you very much.