The fundamental question in economics is why some nations became rich. It is self-evident that this is not the natural state of affairs. If you go back just a few decades, then no nation by modern standards could be considered wealthy. Yet we still know very little about how to deliver this growth. Some scholars have emphasised the importance of institutions in their analysis, others ideology or property rights, and some look more to geographical factors.
By themselves, these theories are either ill-defined, not grounded in empirical reality, or more often a combination of the two. The sad reality is we have not answered how to deliver growth yet. Where industrial policy made South Korea the success story of the 20th century, similar approaches in Brazil have maintained its status as an emerging economy for decades. We can see below the divergence started in the 1980s and has only expanded since then.
The impact this can have on outcomes can be astounding. For example, since 1980 the difference in life expectancies has grown from 4.8% to 9.2%. Once nations begin growing it is not hard to see why this accelerates. If a country grows at 7% per year their GDP will double every 10 years; if they grow at just 1% per year it will take over 69 years for their GDP to double. (1)
Clearly there is no standard successful approach, but given the massive improvement in outcomes that growth has, it is hard to not be captivated by this research agenda. As Robert Lucas once explained,
“The consequences for human welfare involved in questions like these are simply staggering: Once one starts to think about them, it is hard to think about anything else.”
Even though economists may debate for years why these divergences occur, I think I have come upon a good explanation. Granted it is not original, but I’ve yet to see it laid out well in blog form so I thought I’d take that task upon myself. Those who know me already know what I am going to say it is, and that’s cities.
A Smithian explanation
I’ve always found that if you want to understand a macroeconomic concept you need to understand the micro-foundation to ensure that whatever you’re arguing is grounded within realistic assumptions of the self-interested rational humans that we tend to be. This can be explained best through Adam Smith’s theory of the division of labour.
I won’t bore you all with the details, but effectively if people specialise in a particular task then they will become more productive at that task. He used the example of the pin factory where by breaking the process down into a production line, the individuals making the pins could collectively make vast sums more pins than if they individually worked on them.
However, an important detail about the division of labour is that it is famously limited by the extent of the market. What this means is you can only specialise insofar as you have demand for it. Take the example of a man farming cattle - in an isolated rural village they may also have to be a butcher and a trader simply because without these functions being done his business is unfeasible and there is no one else to do it. He is limited by the nature of the market and forced to act like the unproductive pin factory where people just individually make their own pins without a production line.
Now imagine that same farmer relocates to a big city. He may find that there is a shortage of butchers in that city and be able to specialise in that art, and give up the work as a trader and as a farmer. This can enable the production of meat to grow massively, as the farmers can focus on farming, the butchers on butchery and the traders on trading.
A brief flirtation with Ronald Coase
Now this is not the end of the story. To paraphrase Churchill, it is not even the beginning of the end, but perhaps the end of the beginning. That is thanks to a brilliant British economist called, you guessed it, Ronald Coase. His contribution for these purposes is in his 1960 paper the Problem of Social Cost. Effectively he argues that assuming zero transaction costs and defined property rights people will bargain to make a pareto efficient distribution of resources.
Of course this assumption is a big one. Indeed, Coase was under no delusion of the fact that transaction costs were often massive - it was kinda his whole thing. However, an extension of this is that by lessening transaction costs you are more likely to see this kind of bargaining exist and therefore efficiency gains. This means, for example, rather than fighting wars or suing other parties during a conflict, where transaction costs are lower, you can agree on a situation that improves the lot of both parties.
This creates an incentive to move to places that have these low transaction costs and to reap the benefits of the efficiency gains that will materialise from them. So, people do that.
Throw in a few spill(over)s
So, we now know that when transaction costs are low people make more efficient agreements and when people move to bigger markets they may divide their labour and become more specialised allowing for more production. This, in itself, constitutes a great case for cities. However, their benefits go a lot further due to the concept of spillovers and agglomeration.
Where firms cluster together they create positive externalities - beneficial effects on each other. This has been mentioned previously by the Coasean and Smithian arguments, but the literature on this adds to it. Broadly, we can separate two other types of agglomeration benefits - knowledge spillovers and labour pooling.
Basically a knowledge spillover is an exchange of ideas. When firms are close to each other knowledge travels between them much easier. This allows productivity advances by a given firm to be also held by competitors, but also anyone else who would benefit from these innovations. For example, when Henry Ford introduced the assembly line those proximate to his factory would be more likely to hear of his innovation, and thus following him in producing cars, and other manufactured goods, in this way. These spillovers only work if you can communicate with people and pick up knowledge whether it be directly or indirectly.
The other form of agglomeration worth noting is labour pooling. Effectively this is the idea that when firms cluster amongst their speciality they are close to workers who have the skills to reduce cost. Consider Silicon Valley - if all the best software engineers already live here, hiring staff is only made easier if you are already in the area. The same applies to less specialised skills like car manufacturing as shown by historic clusters in Detroit and the West Midlands of England.
In turn, these both ensure that the development of cities presents significant positive externalities helping all the inhabitants there to become more productive than they would be without them.
So how does this relate to development?
Simply every country to have ever developed has gone through a process of urbanisation to get there. This is not surprising if you look up and consider the benefits of cities listed there. Indeed, it seems impossible to be a successful country and not benefit from having cities - they are simply such powerful vehicles for productivity growth.
Take the example of Britain - it is common knowledge that during the industrial revolution there was a rapid urbanisation of Britain, helped by policies encouraging this like enclosure of common land. However, what is less commonly known is this urbanisation actually pre-dated the start of the industrial revolution. For example, between 1600 and 1750 the urbanisation rate had increased massively from 6% to 18%. This puts England and Scotland as outliers from the European experience which did not share this urbanisation process until later. This likely explains why the industrial revolution started in England and not elsewhere.
For a more modern example we can look at the growth of China. Realistically it was only from the year 2000 onwards that China began its massive growth. The period preceding this was one of unanticipated urbanisation sprung by fiscal decentralisation that allowed municipalities to encourage migration towards urban areas. This was the first era without an explicit anti-urban bias in Chinese policymaking and it was followed by some of the fastest economic growth the world has ever seen. Coincidence? The economic theory does not support that conclusion.
Concluding thoughts
Obviously a huge amount more analysis is needed to develop this theory, and I would appreciate some critical engagement with this idea. But, we know all the benefits of developing cities and allowing more people to access them. The literature on this is extensive, and it is full of excellent contributions from some of the world’s best economists. I plan on digging into the empirics behind this idea more over the next year or so, and may post more blogs on it. But for now I think the secret to growth is urbanisation. My advice to policy-makers is therefore “build it (cities) and they (development) will come”.
You can work this out yourself by dividing a given growth rate into 69 (the log of two times 100).