In this episode of Ideas Untrapped, I sit down with economist Oliver Kim to explore the complexities of African economic growth and the challenges surrounding industrialisation. We discuss why Africa has struggled to replicate the manufacturing successes of East Asia, touching on issues such as labour costs, political economy, and the global market environment. Oliver also shares his thoughts on the importance of state capacity and regional integration and how to rethink GDP statistics in development research. Oliver Kim is an economic historian and a research fellow at Open Philanthropy. He also writes excellent blog Global Developments.

Transcript

Tobi:Welcome, Oliver, to the show. I've been a fan for a while, and it's fantastic talking to you. So thank you so much for coming on Ideas Untrapped.

My first question to you involves something you wrote a couple of months ago where you talked about African prices, which is always a puzzle that I've been interested in. So, to restate it as simply as possible, we know that manufacturing in Africa has not grown as much, at least relative to other sub-regions in the world. And there are some theories or findings that suggest that it’s because labour cost is too high. And there's a bit of back and forth in the debates about how unique that is to Africa as a continent. So can you shed more light [on that]?

Because you see a lot of comparisons, maybe Ethiopia and Bangladesh…the unit labour cost and how high it is. So, is that really the constraints? What are the nuances based on what you discussed in that blogpost?

Oliver:Yeah. Just to quickly summarise. Africa has kind of missed out on the manufacturing revolution that, for instance, propelled East Asia…so when you think of the East Asian tigers, China, to rapid rates of growth and poverty alleviation. And, i think in some countries, actually, the share of manufacturing value-added or the share of manufacturing employment is the same or lower than where it was in the 1970s immediately after independence. So, from a developmental standpoint, this is a bit of a puzzle and from a poverty alleviation standpoint, it's a tragedy because this is the only sort of way that we know how to lift large numbers of people out of poverty in a rapid sort of fashion. That’s how China did it; that's how earlier, Korea, Taiwan, and Japan did it.

From a prices standpoint, the problem that economists have identified is that labour costs are too high relative to the level of productivity. That's an important qualified statement to make. So most developing countries are poor [and] as a feature of a developing country, one thing that's true is that incomes are relatively low, wages are relatively low, and so labour is relatively cheap. It's also true that if you're a foreign firm deciding where to site a factory, you don't just care about the labour cost. You also care about the productivity of the workforce. And so it works out that what you care about is like the amount of productivity divided by the cost of hiring additional worker.

And on that metric, which is typically measured in something that's called a unit labour cost (the amount that it costs to produce one unit of output), a lot of sub-Saharan African countries turn out looking relatively poor, especially compared to their peers [at] similar sort of income levels.

So there's sort of two dimensions of this problem. One is the productivity side, and then the other is the cost side. On average, it appears basically that African countries have wages that are actually relatively high for their level of development. And so this becomes a further mystery, like why is this the case? One hypothesis that's been put forward in a couple of papers by the folks at the Center for Global Development is that it's because prices are too high. So this is like one step up the causal chain. If prices are high and the goods and services that are to buy cost too much, then you have to pay people a higher wage basically to afford that.

Of course, the sort of factors behind this, I think, are incredibly complex. I think one major, sort of, historical and fundamental feature that I would point to is that historically labour in Africa, sub-Saharan Africa has been relatively scarce. So this is the contrast I guess, with East Asia and potentially South Asia, where population density is incredibly high and labour is constantly in surplus.

So historically, you know, China, East Asia is like one of the most densely populated regions of the world. The opposite is kind of true in Africa. Now the population has grown a lot, but historically you just had actually a lot more land than people. And if you look at the deep history of African sort of polities, a lot of them were trying to economise more on people than on land. So like in East Asia and Western Europe, you know, you had states with very clearly defined boundaries and political control was defined by control over land.

In Africa, there are states, but there are also instances basically where political control was defined more by control over people. And so there was more fluidity in terms of like territorial boundaries. And so control basically of labour, potentially through slavery also, was a way of a political state to assert power. That's a bit of a digression, but historically speaking, you had relatively low population density. I think that's part of the factor into capitalism. why labour was relatively scarce and maybe why wages are low. So in the present day, maybe that's starting to change a little bit. But looking at the sort of deep fundamental factors, it appears that maybe wages are potentially, quote unquote, “too high to enable a sort of African manufacturing revolution.”

Tobi:Yeah, maybe I read that wrong. But one of the things you discussed in that particular essay was the Assam non-linear model or something like that. It was a U-shaped relationship between GDP and price levels, which, again, maybe I'm wrong about this, the conclusion that sort of came out of that, that this might not necessarily be a problem that is unique to Africa. So can you shed more light on that?

Oliver:Yeah. So let me talk about what I talked about in the blog post. So economists have this relationship. It's a purely like empirical one. So if you go out in the world and you observe things, it's called the Balassa-Samuelson relationship, where basically it appears that as countries get richer, prices of things like haircuts and services seem to go up almost more than proportionally, right?

So like, you know, if I go to Switzerland, which is a very rich country, a haircut costs like $30 or something, something ridiculous. Actually, it's probably more than that. It's like $50 or something, 50 Swiss francs versus, you know, if I go to a Kenyosi in Kenya or whatever to go to a barber, that same haircut, which effectively is not that differentiated in terms of quality. Like a haircut is a haircut. Like if I ask for a buzz cut, it's the same thing. It's the same product, but that product in Kenya probably costs a dollar or possibly less. And so this sort of weird differential where richer places seem to have higher prices is known as the Balassa-Samuelson effect. And if you think about like the sort of underlying theoretical mechanism here, basically richer countries have higher productivity. Higher productivity shows up in like higher productivity in let's say like the manufacturing sector or higher tech kind of sectors.

For like service sectors, which everybody needs, right? So everybody needs like barbers, everybody needs janitors, teachers, things that basically don't increase that much in productivity. Even these sectors need to face higher wages in these rich countries in order for them to be able to compete with the manufacturing sector where productivity has gone up a lot, right?

So like people can switch jobs, people can move between different sectors of the economy. And so the price basically of these service sector goods where productivity actually hasn't gone up that much have to sort of keep pace.

And that's how you end up with this phenomenon where - the same haircut, essentially the same quality, costs the same amount across two places with very different incomes. Now, the way that economists typically have thought about this is that there's a linear relationship. So if you drew like a scatterplot of countries by their income levels and their price levels, you'd just get something like a line. That's like the theorised kind of relationship, a linear Balassa-Samuelson relationship. By that logic, if you put African countries on that scatterplot, it looks like basically that their prices are too high.

So they're lying above this line on the scatterplot between GDP and price levels. What I was arguing in this piece is there's some research, in particular a paper from the Journal of International Economics by Hassan, I forget his first name, I think from like 2017 or so, basically arguing that maybe this Balassa-Samuelson relationship is not actually linear. Maybe it's actually nonlinear, right? There's no actually like really strong reason that this thing has to be linear. It's just like it's the easiest model to write down.

And this is like a common flaw amongst economists is that you go with the first thing that's like easiest mathematically to do. And then you forget that it's just a simplification and you start to treat it like a feature of reality. But so he writes out a more complicated model where, you know, as is