China bulls bet on an urban fiction

More urbanisation is the latest go-to argument for Chinese growth, but it’s not the prosperity machine many observers think it is.

The latest default bull argument supporting higher levels of growth in China than I believe possible is the urbanisation argument. Beijing is planning another major urbanisation push, and according to this argument China can resolve the problem of wasted investment by investing in the urbanisation process, that is it can engage in a massive investment program related to the need to build infrastructure for all the newly urbanised. 


Like so many of the earlier bull arguments, however, this new belief that urbanisation is the answer to China’s growth slowdown is based on at least one fallacy and probably more. The first and obvious reason is that urbanisation is not an act of God, and therefore indifferent to earthly conditions. Urbanisation itself responds to growth. Countries do not grow because they urbanise, in other words, they urbanise because they are growing and there are more good, productive jobs in the cities than in the countryside. In that sense urbanisation is not a growth machine. It is simply a pro-cyclical process that accommodates growth when growth is rising and reduces it when it falls.

And pro-cyclicality is a bad thing, not a good thing. It means that increases in growth are enhanced but reductions in growth are exacerbated, so it adds costly volatility to the economy. As the economy slows, in other words, urbanisation itself slows, thus subtracting economic activity. In May I re-read Eric Hobsbawm’s excellent bookThe Age of Capital, and remember his pointing out that within three years after the 1873 crisis the number of European immigrants landing on Ellis Island dropped by nearly 70 per cent. Do not expect, in other words, migration to stay high once growth drops.

The second, more important, reason is a little more complex and has to do with the source of urbanisation. Let me explain by using an example. Let us assume the state government of Illinois decides to raze Chicago to the ground and build an identical new city a few miles away. Would Illinois and the US become richer?

It depends. If the new Chicago were exactly the same as the old Chicago, and the workers in the new Chicago produced just as much as the workers in the old Chicago did before it was razed to the ground, then Illinois and the US would become poorer by “re-urbanising”. There might be the temporary illusion of wealth created by all the new building, but remember that this new building would have to be paid for by taxes, and the resulting reduction in household income over the repayment period would also cause a reduction in household consumption during that period that reduced jobs and productive expenditures elsewhere. What is more, if people with productive jobs left their jobs to build the new Chicago, the true economic cost of transferring workers from those jobs to building the new Chicago would be very high.

Urbanisation accommodates, it doesn‘t cause, growth

But let us suppose that working and living conditions in the old Chicago were so bad, and unemployment so high, that the creation of the new Chicago required very few productive new workers (most of the workers were previously unemployed and so producing nothing) and caused a surge in productivity at a very low real cost as the facilities in the new Chicago greatly outclassed the facilities in the old Chicago, allowing workers to be many times more productive. If the higher output generated by Chicagoans in the new Chicago were greater than the cost of building the new city, then Illinois and the US would indeed be richer.

This is the key. It is not the act of building all this stuff that creates wealth or real, long-term growth. It is only if building the stuff caused overall productivity to rise by more than the cost of capital and labour employed in building it that a society gets richer.

The same is true of urbanisation. Of course if 300 million people move from the Chinese countryside to the cities, we would need to build lots of new apartment buildings (assuming the existing apartment buildings are already full, which is an heroic assumption in China), roads, hospitals, schools, and so on to accommodate them, but these do not make China any richer. They have to be paid for, and their cost is equal to the cost of capital (resources, machines, steel, copper, etc.) plus the total amount of labour that workers building all these things were doing (and no longer can do) before these things were built. We can assume, if we like, that all of the labour was previously unemployed, in which case the labour cost of urbanising will be zero, but there will still be a tremendous cost of resources.

Paying for these resources would require transfers from other sectors of the economy, so that the economic activity created by all this building would be exactly equal to the economic activity subtracted from the economy. Why? Let’s say that it costs $1 trillion RMB to house all these urban migrants. This $1 trillion would be paid for by direct or indirect taxes on Chinese households, which would of course reduce their disposable income.

As their disposable income dropped, they would spend less on food, shoes, holidays, education, health and lots of other things, and so the only way China would be “wealthier” is if the value of the new stuff created was greater than the value of the stuff that would have been created but no longer can be. If you believe that migrant workers would prefer to stay home but only move to the cities because conditions at home are terrible – a plausible assumption – then it is hard to believe that Chinese households are really happier by spending more on apartments to house migrants and less on food, education, clothing, and so on.

So does this mean that urbanisation must always result in less growth and lower overall wealth for China? Of course not. It can result in greater overall wealth if there are so many jobs in the cities whose productivity levels are so much greater than the productivity levels in the countryside that the increase in overall Chinese productivity is greater than the cost of building all this stuff for 300 million migrant workers.

Once again this is the key. If China is growing so quickly that it desperately needs to move people out of low-productivity jobs in the country and into high productivity jobs in the city, then urbanisation is wealth enhancing for China. But urbanisation itself does not make China richer. It only allows China to become richer if there is already desperate need for workers in the cities.

So as in the case of razing Chicago to the ground, urbanisation itself will not cause growth. It will allow growth to happen if the conditions are already there for rapid growth, and if urbanisation allows a transfer of labour from lower productivity jobs to higher productivity jobs. If China isn’t already growing quickly, however, forced urbanisation will make it poorer, not richer.

But what if you are Keynesian? Didn’t Keynes say that simply creating work would make an economy richer, even if that work consisted of digging holes and filling them up again? Isn’t it better to build an apartment building than to dig a hole and fill it up again?

No, and Keynes never said that. Keynesian pump priming only works if there is high unemployment caused by the self-reinforcing tendency of lost jobs leading to lower demand leading to more lost jobs as factories fire workers. Spending in itself does not create wealth, and Keynes never said that it would. Spending can create a process that under some conditions can cause the economy to grow faster and under other conditions simply creates an unnecessary expense.

If unemployment in China is extremely high and rising, and if as it rises it causes demand to drop, which causes factories to fire workers, therefore causing unemployment to rise and demand to drop even further in a devastating downward spiral, then building buildings, roads, and services for 300 million new city dwellers is positive for China’s economy. But it is positive only because it prevents the economy from collapsing in on itself.

Notice what this means. 'Keynesian' spending on urbanisation is not a way to keep GDP growth sustainably above 7 per cent. It is a way, if China’s economy slows so dramatically that unemployment is surging, for Beijing to respond with a massive make-work program to prevent the economy from spiralling out of control.

This why the argument that China is forcefully urbanising, and that this urbanisation will guarantee growth rates above 7-8 per cent for many more years, is almost sheer nonsense. Urbanisation is a cost, and if China borrows to fund this cost, its debt will grow enormously. It is only if the resulting increase in productivity on other projects – that is on economic projects that have nothing to do with housing the new migrants – is much greater than the increase in debt that China will get richer, not poorer.

Michael Pettis is a senior associate at the Carnegie Endowment for International Peace and a finance professor at Peking University’s Guanghua School of Management. He blogs at China Financial Markets

No carousel of other videos after the video ends.

Want access to our latest research and new buy ideas?

Start a free 15 day trial and gain access to our research, recommendations and market-beating model portfolios.

Sign up for free

Related Articles