Is trickle-down mostly local?

Sam Wetherell writes,

the fifty largest metropolitan areas house just 7 percent of the world’s population but generate 40 percent of its growth. These “superstar” cities are becoming gated communities, their vibrancy replaced with deracinated streets full of Airbnbs and empty summer homes.

The high concentration of wealth in a few metro areas is a surprising phenomenon, given that the Internet was supposed to herald the death of distance. Possible explanations:

1. Talent tends to concentrate. So productivity is higher in the superstar cities than elsewhere.

2. Consumption externalities and network effects are powerful. So people who earn high incomes flock to cities that have what they enjoy.

Here is another possibility. Perhaps it is not the cities per se that are attracting wealth. Perhaps it is just the case that wealth is so concentrated that if you happen to have a city with a handful of the wealthiest people living in it, wealth will trickle down locally. The super-rich will put some of their wealth into the non-profit sector, and they will put significant chunks of their donations into local institutions. This raises incomes in the area.

I would put this possibility of local wealth trickle-down as one possible factor, probably a small one, as we attempt to explain the tendency for high incomes to be concentrated in a few cities.

11 thoughts on “Is trickle-down mostly local?

  1. There has been a death of distance, just not for everything, and not where many people expected it to be. I think part of that was simply motivating reasoning / wishful thinking. People wanted to believe it was possible to earn high developed country wages, while enjoying small-town rents and lifestyles, while also somehow being selected over competitors from abroad. This is like a “good, fast, cheap: pick two” trade-off.

    Freer global trade, cheap and fast transport of atoms, and cheap and fast transmission of bits, have indeed made the world a much smaller place. But economies of agglomeration start to dominate under the above conditions combined with high levels of automation, because many jobs can only be done “in person,” that is, in close proximity to to other people.

    Indeed, without political distortions (e.g., immigration restrictions and strictly local occupational licensing), the inertia from historical contingencies in the distribution of populations and capital, and some demand-destruction originating from high rents, it’s rational to forecast that the whole world gets centralized into a few Ultra-cities, or maybe even just one eventually. Indeed, anything that makes it more costly in money or time to travel long distances – such as the practices of the current transportation security system – will only accelerate this process, since it won’t be viable for most people to routinely “commute by air” as was once hoped.

    Let’s say at any moment there are two kinds of jobs: those susceptible to the death of distance, and those that are not. If distance doesn’t matter, then those jobs are going to leave for the cheapest place on earth. Since that is almost never going to be in a developed country, the vast majority of the laborers of a developed country are never going to get to observe or enjoy this “death of distance”. If I can provide call-center customer service anywhere, it will more often be from Mumbai or Manila, not Montana.

    So that leaves only jobs geographically close to some key human counterparty by default. That’s why more and more people are in the “service sector” these days. And that includes all those “Garett Jones workers” building organizational capital and other intangibles, which is part of what is making our economy increasingly illegible according to the aging mainstream economic framework of interpretation, which is becoming more implausible, obsolete, and distant from reality every day.

    Furthermore, if there are “consumption externalities” then they are most likely of the form of services which involve the participation of other local people. Most people can get any kind of good delivered to their door quickly and cheaply, or maybe drive a short distance to get some kind of perishable specialty item. What people are “consuming” then in those niche instances that depend on markets with Mega-city extent is experiences depending on other people. This too is similar to an economy of agglomeration / network effect.

  2. The wealthy have always used a lot of personal service. Lawyers, accountants, university professors to occupy their children, personalized, if not personal, chefs, etc. These services will be more highly compensated than those similarly available en masse.

    Warren Meyer over at Coyote Blog had a good post on the Lifestyle Charity
    http://www.coyoteblog.com/coyote_blog/2016/08/the-lifestyle-charity-fraud.html

    “These charities seem to be run primarily for the financial maintenance and public image enhancement of their leaders and administrators. Most of their funds flow to the salaries, first-class travel, and lifestyle maintenance of their principals.”

    In exchange, these charities provide a vibrant society social calendar and honorary board member and chairman ships.

  3. There are limits to how much people will give up for the benefits of concentration –for example, concentration itself (in urban centers) has spread nationally since the mid-1900’s: from a few large hubs in the North East and the Great Lakes across almost the entire nation. Here the limits were bad climate, lifestyle, economic policy, high cost of living, e.g. The internet was not a factor supporting the moves. Perhaps this big sort/dilution satisfied the need to change climate, lifestyle; and the new urban centers learned the lessons of pursuing (excessively) bad economic policies.

    As to the consumption externalities, I propose two. Consumption number one: “suitable” mates. Consumption number two: “suitable” neighbors with suitable “children”.

  4. I would like to know what the statement in question means. The 50 largest metro areas house only 7% of the world’s population but generate 40% of its growth. Are we talking about population growth or economic growth (GDP growth?). And how is that economic growth being measured. And how is that stat being influenced by China, which has 17 of 50 largest metro areas (India has 6). Or by fact governments tend to be located in major cities and government is a growth industry (but supported by all). Until one know exactly what is meant by statement, and how it is derived, not sure it makes sense to waste a lot of brain power to explain it. And given US only has 3 of 50 largest cities, and most hip cities like Austin, Sydney, Vancouver, Portland, and San Francisco are not among world’s largest 50, stat has nothing to do with rest of article or any points he is trying to make in article. So there are some red flags whether that statistic is meaningful.

  5. Here’s a 50/50 gdp split map of the US. Several versions around. They all pretty much look like the presidents* favorite hand-out – the Electoral College map.

    [Map] America’s GDP split geographically, 50-50

    https://www.reddit.com/r/NoSillySuffix/comments/6p9ebg/map_americas_gdp_split_geographically_5050/

    I’ve lived a few places, including 3 of the 50 cities. Just my observation, but it seems like a sort of velocity is achieved that just builds on itself.

    If you were an artist, where would you want to be? An architect? An interior designer? A lawyer? If you were a young person where would you want to start your career?

    I left the rust belt in 1982 for the big city. With not much. Not even a college degree. That’s the way you have to do it.

    • Moo cow, you map is interesting but unless one knows % of population in these metro areas that make up 50% of GDP, does not really prove point. Your map looks like top 20 to 25 metro areas. Top 25 metro areas have about 42% of US population. So not sure it is big deal.

      • Well. Sure. 40% packed into that (relatively) small area. Isn’t that what this is about? Expand it to top 100 metro statistical areas.

  6. “1. Talent tends to concentrate.”

    C-suite talent needs to engage in conspicuous consumption as part of zero sum status games played with other C-suiters. Actual productive talent can be anywhere, but the higher-status the productive workers are, the nearer the C-suiters want them to be. Boeing’s HQ is in Chicago, but their productive workers are low status and located several states away.

    The finance industry is moving in this direction. In a generation only the C-Suiters will be left in Manhattan, the lower status back office workers will be in lower cost cities.

    Tech productive workers are now relatively higher-status white collar types, so C-Suiters want them near to for social dominance reasons. As their white collar worker become lower in relative status to their bosses, tech will follow the lead of finance and send them to the boonies.

  7. Trickle down or flow up as value becomes communalized, the value of my property becoming dependent on the value of yours.

  8. I think you underestimate just how uncertain employment has gotten, even at the higher end. I’m a software developer (so Silicon Valley and all that) and the norm is to switch jobs every two to four years. If you’re switching jobs at that rate, you have to live where the jobs are, which means the big cities.

  9. It is a form of trickle down being local, at least from a jobs perspective. Enrico Moretti covered it well in his “New Geography of Jobs”.

Comments are closed.