These are the cities that will dominate the global economy in the next 15 years

New York will continue to be the world’s largest urban economy in 2035, but Jakarta will have passed Tokyo to become the most populous city. However, cities in the Indian subcontinent, led by Bengaluru (Bangalore) and Dhaka, will be the fastest-growing. Those are three of the conclusions from a study by Oxford Economics of the way the economies of global cities will develop over the next couple of decades.

The development of megacities, cities with a population of more than 10 million, has become a source of endless analysis and concern. From a UK perspective, it is intriguing to see London is projected to remain the fourth largest urban agglomeration ranked by GDP – after New York, Tokyo and Los Angeles, though by then it will be joint fourth alongside Shanghai. In terms of financial and business services output, London is projected to rise to third place, after New York and Los Angeles, and pull ahead of Tokyo and Paris.

But changes in the pecking order of developed world cities are slight when set against the rise of the emerging world. Thus, of the top 10 cities – or rather urban areas, as defined by the US metropolitan statistical area yardstick – four are expected to be in China: Shanghai, Beijing, Guangzhou and Shenzhen.

As a business centre, Guangzhou trails Chicago and Washington but leads San Francisco and Boston. In terms of total GDP, Jakarta will be number eight, rising from 21 this year. And in terms of growth, of the top 20 fastest-growing urban areas, 17 are in India.

What should we make of all this? These are projections for nearly 20 years ahead, and will inevitably be off the mark in terms of detail, but the broad thrust of the study is likely to be correct. So we have to accept, massive urbanisation will continue for a generation at least, probably longer. That leads to several conclusions.

One is that this will change the planet in social as well as economic terms. We know from our experience in the west, moving from the countryside to a large city is socially liberating. No one knows who you are. People are free from their social backgrounds, or at least more free than they would have been, had they remained on the land.

Another observation is that people are not just richer – though measured living standards are generally higher in great cities than the surrounding countryside – but they benefit in other ways. For example they have, in general, better access to healthcare and education. Probably – though this is a huge generalisation – they are better fed – I recall being told, when visiting Dharavi, the huge informal city near Mumbai airport, that for all its mess and apparent disorder, no one went to bed hungry.

A third conclusion is that urbanisation is a huge investment opportunity. We know a lot about what to do and what not to do when developing attractive cities. There are the basics of water, sewage and transport. We know land should be set aside for parks and recreation. We know informal settlements spring up, and converting these into more organised housing has to be handled with sensitivity.

All this costs money, so part of the wealth generated by a fast-growing city has to be channelled back into its infrastructure.

But alongside the developed world’s knowledge of how to manage cities lies one big lesson: the model a rich city follows is not necessarily best for a city in an emerging market.

For example, the typical US suburban sprawl, while enabling people to have a lot of living space, is inefficient in land use terms.

Water use has to be managed more carefully, and use of the private car needs to be priced wisely – the average fuel consumption of the Chinese car fleet is lower than that of its European equivalent.

Still, living in even an inefficient city is probably more efficient than living in the countryside. It may be that the communications revolution has further tilted the balance towards the city. For example, online delivery services are much cheaper to run in densely populated areas. Road mapping and traffic systems enable a city’s infrastructure to be used more efficiently, whereas they don’t bring a similar advantage to rural areas.

Further, if people choose to move to a city – and that is what is driving urbanisation – they will have reasons for doing so.

I think the challenge of megacities is more about the growing gap between the most successful ones and less successful middle-rankers, rather than the movement to cities as such.

There seems to be a winner-takes-all phenomenon. You can see that in the choice by Amazon of two east coast American cities for its second headquarters: it split operations between the New York and Washington metropolitan areas, numbers one and four in the US in terms of financial and business services output – both of them in the global top ten.

Second-rank cities will have to work out where their competitive advantage lies and play to whatever strengths they have. Thus, Stockholm is western Europe’s fastest-growing city in this study, and San Jose – home to Silicon Valley – is the US’s fastest-growing.

Meanwhile the winners, including London, need to justify their success, for it is a hugely competitive world, and will only become more so.