World | Commentary

Why is Tokyo defying population outflows among world's 'superstar' cities?

by Justin Fox

Bloomberg

Like a lot of big cities in the developed world, Amsterdam lost population in the 1960s, 1970s and early 1980s as its inhabitants opted for newer dwellings and more space outside the city. During those same decades, newcomers arrived in large numbers from former and current Dutch territories as well as Turkey and Morocco. But there weren’t enough of them to make up for the domestic exodus, and their presence led some longtime Amsterdammers to decide the city was no longer for them.

After 1985 — again like a lot of big cities in the developed world — Amsterdam stopped shrinking. New arrivals from overseas were still coming, and more and more of them were having kids (one of whom, just to keep American readers engaged here, is the new shortstop for the Philadelphia Phillies). Educated young people from elsewhere in the Netherlands began to stream in, too, attracted rather than repelled by the city’s increasingly diverse, cosmopolitan feel, and drawn in by a growing assortment of good jobs. Starting in 2005, the city began to experience sustained net inflows of domestic migrants for the first time since at least 1950.

In 2015, that domestic inflow reversed. By 2018, the net outflows were of a scale not seen since the early 1980s.

The inflow from abroad has actually spiked back up recently — albeit from very different places, with U.K. citizens now the biggest group of foreigners in the city and Italians leading the list of new arrivals in 2018. Together with a big birth-death differential (because of earlier migration patterns, and probably also the insanely steep stairs in older apartment buildings, there aren’t a lot of old people in Amsterdam), this has kept the city growing. Last November its population hit an estimated 873,200, finally surpassing the previous record of 872,428 set in 1959. But the accelerating exodus to elsewhere in the country seems significant, in part because similar things have been happening in Berlin, London, New York, Paris, Sydney and Toronto. The one exception I found in my admittedly less-than-exhaustive search of what are frequently called “superstar cities” was Tokyo, which is still experiencing big domestic inflows.

Berlin, London, Sydney and Toronto have, like Amsterdam, continued to grow despite domestic emigration. New York and Paris have resumed shrinking, though, and my sense is that some of the other superstars may follow in their wake before too long. The great rich-world revival of the big city seems to be fading, or at least entering a new phase.

Population outflows from New York, Chicago and Los Angeles have already been getting some attention in the U.S. (some of it from me). Partisan politics often intrude into this discussion, which is one reason it can be helpful to internationalize things. Big cities in other rich countries tend to be run by left-leaning politicians, just as in the U.S., but more-centralized national governments generally mean that taxes and other policies don’t vary nearly as much between regions. Other forces must be at work as well.

The focus here is on Amsterdam because: 1) I was there last week; 2) I spent the 1982-1983 school year in nearby Haarlem, among classmates who later became part of that late-1980s moving-to-Amsterdam wave; and 3) Statistics Netherlands and the Amsterdam city office of research, information and statistics have more relevant data on offer than their counterparts just about anywhere else. Statistics Netherlands reports, for example, that 27- to 40-year-olds with kids have been driving the recent Amsterdam departure wave. Younger Dutch people are still moving to the city, but the supply of those in their early 20s is shrinking, with even bigger declines to come.

Similar demographic trends are at work in most other rich countries, meaning that the prime domestic candidates for moving to superstar cities are getting to be in shorter supply.

They’re also finding it harder to afford housing there. Purchase prices for private dwellings in Amsterdam have doubled just since late 2014 (rents are up a less spectacular 16 percent, still the highest rate of increase in the country).

Such price increases will be familiar to residents (and former residents) of other superstar cities. They are an indication that this isn’t the 1960s and 1970s all over again — these cities continue to generate good jobs and possess great appeal. But they are also an indication that the superstar city phenomenon has its limits.

It is notable that the one major exception I found to the trend of domestic emigration from superstar cities is Tokyo, which thanks to the Japanese custom of tearing down housing that’s more than 30 years old has seen a lot more new construction than other big cities, and despite average rents that are a little higher than Amsterdam’s offers great price variation among neighborhoods.

Thanks to the Japanese custom of tearing down housing that’s more than 30 years old, Tokyo has seen a lot more new construction than other big cities and offers great price variation among neighborhoods. | BLOOMBERG
Thanks to the Japanese custom of tearing down housing that’s more than 30 years old, Tokyo has seen a lot more new construction than other big cities and offers great price variation among neighborhoods. | BLOOMBERG

Tearing things down every 30 years isn’t really going to fly in Amsterdam, where much of the city center dates to the Dutch “golden age” of the 1600s. It remains in that form because over the 150 difficult years after 1700 the city’s population barely budged, while the price of real estate along the posh Herengracht canal fell 41 percent in real terms. From today’s perspective that long depression seems like kind of a lucky break, and while I’m all for cities growing and evolving even I don’t think Amsterdam should fill in its canals to make room for high rises.

The resulting World Heritage site, though, has proved so charming and lovable that it is to some extent being loved to death. One high school friend complained that the central city is starting to feel like Venice — a place where tourism is the only industry. A couple of years ago, a 33-year-old letter writer to the local newspaper Het Parool offered this sad account:

I arrive in my neighborhood, stick the key in the lock of my little apartment, where my landlord has seen the value go up by €100,000 over the past 2½ years. My home, in the neighborhood with the most Airbnb stays in Amsterdam. My home, where I’ve just welcomed the third expat couple in a year and a half as neighbors. My neighborhood, where I am now addressed in English at my favorite sidewalk cafe. Where my boyfriend and I feel somewhat out of place. We, highly educated, cosmopolitan world-improvers with good jobs, who have traveled over half the world and speak multiple languages fluently.

Sound familiar? (Well, except for the part about speaking multiple languages fluently — that’s kind of a Dutch thing.) One can hear similar complaints from residents of the West Village in New York, Shoreditch in London, Prenzlauer Berg in Berlin and so on. They are easy enough to mock — “I gentrified this neighborhood, and now that it’s being gentrified even more I am outraged.” Also, most cities aren’t superstar cities, and even most neighborhoods in superstar cities aren’t overflowing with affluent expats and tourists.

Still, this disaffection has become a widespread enough phenomenon that it bears paying attention to. One cause of the superstars’ rise has been the increasing economic importance of knowledge work, and the tendency of that work to grow in value the more knowledge workers are crammed in the same place. By this logic, the best policy is just to encourage the superstar cities to keep growing. A paper by economists Chang-Tai Hsieh and Enrico Moretti that I’ve cited approvingly in the past estimated that local restrictions on new housing construction just in San Francisco, San Jose, California, and New York had cost the average U.S. worker $3,685 in annual wages.

Again, Tokyo’s example seems to indicate that there’s something to this analysis. But in other countries the housing to meet the needs of residents of superstar cities, if it’s being built at all, is generally being built somewhere else. Those departing Amsterdam usually end up in cities a short train ride away. Most emigrants from London are similarly headed elsewhere in the south of England, those from Berlin to the surrounding state of Brandenburg.

The internet does allow more white-collar workers to do their jobs remotely, though, and in some places they are moving even farther away from superstar cities. The entire Ile de France region around Paris is now shedding residents to the rest of the country, with the Gironde and Loire-Atlantique departements (home of Bordeaux and Nantes, respectively) along the Atlantic coast experiencing the biggest net inflows. And of course lots of New Yorkers are moving to Florida, as they’ve been doing for more than a century.

Even at the height of its renaissance a few years back, New York City was experiencing net outflows of residents to elsewhere in the U.S. To some extent that’s the natural state of big cities, which are prone to migration deficits because they attract young single people and then often lose them after they’ve had kids. But net migration out of the city more than doubled from 2012 to 2018, according to Internal Revenue Service data. Something is going on, and it’s not just going on in New York.

Coronavirus banner