Why don’t cities grow without limit?

In other words, why don’t they put everything into Atlanta or Los Angeles?  Paul Krugman has a good blog post on that topic, here is one of his points:

…once upon a time dispersed agriculture ensured that small cities serving rural hinterlands would survive. But for generations we have lived in an economy in which smaller cities have nothing going for them except historical luck, which eventually tends to run out.

Krugman suggests that eventually many smaller cities will indeed fade away, although the process of equilibration may be a long and slow one.  All of his points are well-founded, nonetheless I can see a few factors favoring the continuing existence of small cities on a greater scale than many might be expecting:

1. As Alex points out on Twitter, rents are permanently lower, and many people don’t value big city amenities very much.

2. Congestion is likely to be lower.  Why should the larger city have worse traffic if it has proportionately more roads?  That may require a blog post of its own, but part of the problem is geographic specialization within the larger city, which is not simply some number of smaller cities placed side by side.  In other words, sometimes you have to drive all the way across town.  Many people don’t like geographic specialization, but wish to find most everything in a small downtown or Walmart (or on Amazon).   From this point you can see that Amazon may favor larger cities more than small towns.  If it bugs you that in a large city all the shopping of a particular kind is on the other side of town, just order those goods on-line and stay within your cozy neighborhood.

3. Governance may become worse in a very large city.  Furthermore, separate and specialized lobbies, as would correspond to geographically specialized parts of a large city, may be a bad influence.  Here is a paper on the public choice of mega-city governance.

4. Very large, rich, and famous cities tend to become financial centers, or perhaps movie-making centers, and that is not in the interest of all city residents.  Some of this is a matter of rents, in other regards a matter of culture and ethos.  Anonymity also increases with size, as does (I think) sexual promiscuity.  Smaller locales will have more faux conformism and more real conformism too, which some people prefer.  People not wanting to live amongst all the specialization of major cities really is a significant and enduring factor in these comparisons.

5. If you are building a firm for eventual export success, you will prefer to put that firm in a larger city to begin with (“built to scale”).  That in turn tends to price out companies and people with less interest in exporting.  The larger city will become all that much more globally oriented, which not everyone will wish to pay for or even wish to have at zero price.

6. If I were offered an extra 50% of total salary (nominal, to make this comparison in real terms across all goods and services eliminates the very difference in locales) to move from Fairfax to Washington, D.C. (15-20 miles away), I would decline the offer.

7. The very fact that smaller cities are used to consume non-pecuniary amenities suggests their inhabitants are more diversified than it may appear at first.  The shift of gdp into services further enhances this diversification, and the new crop of semi-small cities may be more resilient than the older lot dependent on manufacturing.

8. A significant and enduring trend is the move into warmer and sunnier climates.  So while Rochester and Flint decline, Chattanooga and Birmingham are on the rise.  I predict the more time you spend in the South, the more optimistic you will be about small and mid-size cities.

9. Here is a good Duranton and Puga piece on the costs and benefits of city size.  Here is a short McKinsey piece on complexity as a limit on size.  Here is a discussion of city size in Civilization VI.


Wrt "Anonymity also increases with size, as does (I think) sexual promiscuity." You could argue instead that it is harder to make people conform to norms as anonymity increases.

Love your work Tyler. Happy 2018!

"But for generations we have lived in an economy in which smaller cities have nothing going for them...."

Paul Krugman does not live in a small city. That's a lot going for them right there.

Honestly, I stop reading when I see Krugman's name. I've read he used to say intelligent things but now he's just a (especially low quality) partisan commentator.

Costs - benefits - equilibrium. It's pretty simple and not worth a post. If Krugman doesn't understand, he's never left a big city and/or he doesn't understand basic economics.

This particular topic isn't really a partisan one unless you need to make it so.

In theory, no. But in reality? It is hard to escape the notion that Krugman means that the Hillary voting urban sophisticates are the wave of the future and the deplorable Trump voters in the hinterland will die out.

This is partly because of the nature of the answer but also because Krugman has such a long history of hyper-partisanship that it is hard to believe he is otherwise this time.

It will be partisan to the extent things like infrastructure dollars need to be allocated which is why the Professor is getting out ahead already on this...

Urbanization is THE partisan issue.

Liberals rely on cities for all their power. They draw in lots of cash and voters. Then they use that power to the detriment of everyone loving outside of cities.

Population density is the cause of nearly every social ill, and liberals capitalize on this.

Breaking up cities would be the greatest boon to ending the political death spiral.

What a crock of resentment!
In actual fact governing structures in the US are strongly weighted toward rural areas. Many years ago my father ( a Republican) used to complain how gerrymandering in the Michigan legislature tended to give the "farmers" (rural people) too much power and they used it to screw over the cities and suburbs.

Re: Population density is the cause of nearly every social ill, and liberals capitalize on this.

Um no,. Ever hear of "original sin"? Even if you're not theologically inclined there are messed up people everywhere and no place is freer of human dysfunction.

Re: Breaking up cities would be the greatest boon to ending the political death spiral.

Pol Pot lives.

To follow on Jan's point, one of Krugman's areas of academic expertise is economic geography. I'm not sure there is too much distance between Krugman and, say, Enrico Moretti on this issue although the latter may be slightly more optimistic about the future of small cities with niche specialties.

Here are the 15 fastest growing cities in the US according to the Census Bureah (% not absolute, 50,000+ people):

1. Conroe City, Texas (Houston, The Woodlands, Sugar Land)
2. Frisco City, Texas (Dallas, Fort Worth, Arlington)
3. McKinney City, Texas (Dallas, Fort Worth, Arlington)
4. Greenville, S.C. (Greenville, Anderson, Maudlin)
5. Georgetown, Texas (Austin, Round Rock)
6. Bend, Ore. (Bend, Redmond)
7. Buckeye, Ariz.(Phoenix-Mesa-Scottsdale)
8. Bonita Springs, Fla. (Cape Coral-Fort Myers)
9. New Braunfels, Texas (San Antonio, New Braunfels)
10. Murfreesboro, Tenn. (Nashville, Davidson, Murfreesboro, Franklin)
11. Lehi, Utah (Provo- Orem)
12. Cedar Park, Texas (Austin-Round Rock)
13. Meridian, Idaho (Boise City)
14. Ankeny, Iowa (Des Moines-West Des Moines)
15. Fort Myers, Fla. (Cape Coral-Fort Myers)

Not exactly Krugman country! I do think his views on this are heavily politicized (blue state cities: diverse and democtatic, Athenian even! Red state cities: yechhh!, basket of deplorables) and he's also looking at post-industial legacy parts of the country in decline as opposed to new areas springing up. And interestingly he doesn't discuss the highly prevalent trend of workers serving national and large city markets by leveraging technology and living in lower, cost, less-congested areas.

With regard to #13, Meridian, Idaho it, and its northern neighbor, Eagle, are upscale refugee camps for Californians fleeing the charms of the once Golden State.

As a family with millions in California real estate, we could certainly "flee" or a ski town like Bend, or ski adjacent like Provo.

Great fly fishing at both as well.

Sadly (from my perspective) we are divided on the necessity of big city amenities.

Mr. Krugman has never let the facts get in the way when there is a larger partisan point to be made/scored. Megacities support his outlook. I do believe that he is looking at the future through a lens from the past. I would think that today's tech will lead to the opposite developments. The 15 fastest cities are probably just a start. Just as breakthroughs in farming equipment got us off the farms and into cities I think modern tech will get millions back to the countryside trying to get a more balanced life between work and nature. Here's hoping.

I wouldn't exactly describe Austin, TX as the opposite of Krugman country. Sure, it has moderate and conservative suburbs but so do New Jersey and Connecticut. You are reading entirely too much into this.

"Fastest growing" is usually statistical nonsense. A pizza shop that went from one worker to fifteen might be America's 'fastest growing business'....yet that's statistical noise if Google grows 15%.

LOL - Paul Krugman is a fraud, a complete idiot.

> 6. If I were offered an extra 50% of salary to move from Fairfax to Washington, D.C. (15-20 miles away), I would decline the offer.


Not trying to be confrontational, I'm genuinely curious about your reasoning and motives. I feel like this single point could be expanded to an entire classic long-form Tyler Cowen post all of its own.

Because the homicide rate is 4x higher.

You just might want to check a finer breakdown of that data. https://www.washingtonpost.com/graphics/local/homicides/

You will notice that roughly a quarter of DC (think the left bank of Rock Creek Park, roughly) has no murders at all. On the other hand, those DC areas with a large number of clustered homicides are likely drug market areas - markets fueled in large part by residents of places like Fairfax.

And I still find it amazing that DC's homicide rate is a quarter of the late 80s, and yet so many people talk about how things have gotten worse there.

Yeah, you can live a very wide swathe of DC with very little concern about any crime. Not everyone can afford to live in those parts of the city but Tyler certainly could.

I think the concerns about violent crime in DC are driven by the real one-year spike in murders we saw in 2015. Of course that was preceded by a long, steady decrease in murders, and in 2016 and 2017 the decline continued. A small share of those 2015 murders and other violent crimes were in parts of the city that have been recently gentrified and that naturally brought even more attention to it.

This is what I do not get. Look we all know the reality. Jan here and Prior clearly know the reality too. They have a map showing the reality. But for some reason we are not allowed to mention it.

Yes, you can live in all sorts of places in Washington with no crime. But you are not allowed to mention what those places have in common? Let's check the map. No homicides in Annandale and Merrifield. An enormous cluster in Hillcrest Heights. But it is racism to suggest why that might be.

Why is saying openly what both Jan and Prior know to be true racist? You cannot have large cities in America because the sort of people who live in Annandale really do not want to live near the people who live in Hillcrest Heights. And perhaps vice versa. Homicide is one reason.

Did you perform your analysis by income level?

'But you are not allowed to mention what those places have in common?'

You mean that the rich live on the left bank of Rock Creek? It is an open enough secret to anyone familiar with DC, after all.

Or the one where a few places in Falls Church used to have a comparable enough murder rate to DC, for the same reason? That being drugs, by the way.

'But it is racism to suggest why that might be.'

Or stupidity - income is a considerably more useful tool for predicting criminality in the U.S.

'Why is saying openly what both Jan and Prior know to be true racist?'

Well, if one rejects that race is a useful construct for looking at things, it is actually pretty easy. But then, regarding what is below, let's look at using your perspectiive regarding what is going on. The one where America's cities are being overrun by those parasites who threaten America's prosperity in large East Coast cities because they are being filled with Irish, Italian, and Jewish immigrants - oh wait, this is the 21st century, not the 19th. A modern era where the term white continues to be a fantastically malleable term in American discourse, and we conveniently forget what racists generations ago believed when it is inconvenient to whatever argument a modern racist is making. And remember, no less august an American than Benjamin Franklin was decrying America's sad future fate unless real Americans stopped the wrong swarthy Europeans from immigrating. Like Germans - https://en.wikipedia.org/wiki/Observations_Concerning_the_Increase_of_Mankind,_Peopling_of_Countries,_etc.

Hillcrest Heights is not that poor actually:

The median income for a household in the CDP was $46,367, and the median income for a family was $52,573. Males had a median income of $34,198 versus $34,558 for females. The per capita income for the CDP was $22,620. About 6.7% of families and 7.4% of the population were below the poverty line, including 9.0% of those under age 18 and 11.5% of those age 65 or over.

Compare with one of the poorest counties I could find. Take Lake County

The racial makeup of the county was 66.63% White, 31.19% Black or African American, 0.39% Native American, 0.14% Asian, 0.62% from other races, and 1.03% from two or more races. 1.37% of the population were Hispanic or Latino of any race. ... The median income for a household in the county was $21,995, and the median income for a family was $30,339. Males had a median income of $25,082 versus $18,700 for females. The per capita income for the county was $10,794. About 19.90% of families and 23.60% of the population were below the poverty line, including 35.10% of those under age 18 and 25.10% of those age 65 or over.

Admittedly a low income goes much further in a rural county than it does in DC. Income does not seem to be a factor though.

You...didn't do the analysis.

clockwork_prior January 1, 2018 at 6:45 am

You mean that the rich live on the left bank of Rock Creek? It is an open enough secret to anyone familiar with DC, after all.

The rich you say? Would you care to name a crime-ridden White suburb of Washington? Apart, that is, from Falls Church below. You know, somewhere with a lower crime rate than Virginia and getting on for half the national average for most crimes?

Falls Church, VA Count 0 3 8 7 16 196 8 2
Falls Church, VA Per 100,000 0.0 21.7 58.0 50.7 116.0 1,420.9 58.0 14.5 689
Virginia Per 100,000 4.1 17.2 51.5 112.9 277.7 1,560.5 92.1 NA 912
U.S. Per 100,000 4.5 26.4 102.2 232.5 542.5 1,837.3 216.2 NA 1,423

Or the one where a few places in Falls Church used to have a comparable enough murder rate to DC, for the same reason? That being drugs, by the way.

Sorry but what? Falls Church is the richest county in the US. With the lowest levels of poverty. When precisely did it have this massive murder rate? This Falls Church, right?


Or stupidity – income is a considerably more useful tool for predicting criminality in the U.S.

The entirely point of Charles Murray's latest book is that it was not. And it isn't. There are plenty of places poorer than Hillcrest Heights. With much lower crime rates.

'Would you care to name a crime-ridden White suburb of Washington?'

Manassas Park, at least in the past. And then, again from the past, much of the Rt. 1 corridor down past Quantico - which is not a 'suburb,' of course. Reston also had a bad reputation when I was younger, but I'm fairly sure that is no longer true. But to be honest - I did not bother to check whether Manassas Park fits your definition of whiteness - it used to, though.

'Falls Church is the richest county in the US.'

Falls Church is not a county, I talked about a couple of Falls Church neighborhoods in the past when DC's homicide rate was 4x times higher than today, and I am fairly confident that the drug trade was dried up by the mid-90s.

'The entirely point of Charles Murray’s latest book is that it was not.'

This is the same Charles Murray who knew, several decades ago, which group needed mass incarceration by looking at them, right?

Manassass Park? This Manassass Park?


How much safer can a place get? Safer than 72% of other American cities. Come on, even for you this is utterly underwhelming. Especially as even if you were right, you would simply be proving that crime is *not* linked to poverty. The median household income in MP is over $60,000 a year.

Causation is always hard to work out. I do not make a claim about causation. But a whole raft of things are correlated with each other. Race being one of them. Poverty for instance. Crime is another. Strongly correlated in many cases. As with crime. Race remains one of the best predictors of crime - more so than poverty.

Yet you and Jan insist that it is racist to point out what the evidence actually is. You talk about in gentle euphemisms. But you know. I think that sort of dishonesty to oneself is corrupting and corroding. You all know. You should admit it.

So Much For Subtlety January 1, 2018 at 6:44 pm

"Yet you and Jan insist that it is racist to point out what the evidence actually is. You talk about in gentle euphemisms. But you know. I think that sort of dishonesty to oneself is corrupting and corroding. You all know. You should admit it."

I actually think you're right about some of these unmentionable subjects, but I'm not convinced that "honesty" really is the best policy. I just don't know. Some kind of cautious, humble and sensitive honesty would be helpful, but I'm not seeing it anywhere. Even if I were a WASP, I'm pretty sure I'd rather be governed by New York Times commentators (blech!) than iSteve ones. Tyler Cowen's readers are, on balance, vastly superior to both.

Tyler travels so much, I'd bet it's better for him to enjoy lower cost, simpler living, as he can save that money for the better dollar-for-dollar entertainment within his emerging market travel destinations.


Also, Fairfax is an atypical small city: it is really a suburb of Washington, D.C.

Atypical in what sense? Perhaps compared to all small cities in the US. There are a *TON* of rural cities, which are really, really small. But there are also a ton of small and medium sized cities in the suburbs where a large proportion of Americans live. I think the median American lives in a city of 60,000, if you line up cities from smallest to largest.

Krugman's discussion of small cities is very clearly about those that are not suburbs of other larger cities.

I don’t understand. Tyler gave himself a 50% nominal raise to account for price differences between the locales. Couldn’t he just take the extra money and travel even more?

The real Tyler has a different color comment line. And doesn't the real Tyler eat Whole Foods smoked trout every morning?

It's not quite a fair point in favor of "small cities." It is a point in favor of suburbs in the orbit of major cities. Would Tyler move to the suburbs of Cleveland or Charolette, NC?

Not sure I agree. Not sure what Tyler's arguing but I would rather be close to downtown in a small city than out in a suburb of a large one.
The thing is that having lived in the orbit of DC in suburban Northern Virginia, I found that I almost never went into DC for anything. It took too long. It was a hassle due to congestion. There wasn't than much there that I wanted to see or do. Where I'm living right now (smaller city), I actually have a better chance of getting to see a major theater production because it's just way easier to get to the downtown theaters and the tickets are less likely to be sold out. I just have to wait for one to go on tour. Not that I'm that big on theater anyway. There was simply no value to me to living in the orbit of DC, just to have car access to DC, when I could get a more affordable house within walking distance of the downtown of a smaller city.

The question of economic opportunity is more important than physical distance from downtown amenities. That makes even my revised question a little unfair: as an academic, Tyler may have more attractive opportunities in small cities (college towns) compared to his private sector peers. If you spend 10 years building a career in an extremely specialized field, you are relatively limited in the number of metro areas you can consider relocating to without also considering a career change.

That's part is true, though you might be surprised how many specialized industries actually do locate significant offices in smaller cities. Not all the major universities are located in large cities, and many specialized tech industries tend to be located in college towns like Ann Arbor Michigan. If you are in a specialized field, there's a pretty decent chance that you'll wind up living in a small city even if you would prefer to live in New York.

'If you are building a firm for eventual export success, you will prefer to put that firm in a larger city to begin with (“built to scale”).'

Well, pretty much the entire Mittelstand in Baden-Württemberg is a contradiction to this, as even Stuttgart has a smaller population than DC.

North east Italy is another example, no big cities but a lot of export.

Switzerland is also a candidate, though not Austria - Wien is quite a dominating city, with about a third of Austria's population within its metro area.

using "wien" for "vienna" is as clockwork_prior as it gets.

What, no love for the 'ü'?

(And true story - my sister was born in Vienna. And she has spent a lot of time explaining to various people that no, she was not born in Austria.)

But surprisingly , a quarter of Austrian exports is generated in the province of Oberösterreich, outside of the Viennese metropolitan region. Vienna is the origin of just 13.5 pc of Austrian exports despite housing more than a fifth of its population. So probably, for Austria, it does not pay to locate your export industry in the metropolitan region.

We might also speculate that

1) Inequality could be a drag on city growth;

when you only own one house and that house is a large percentage of your net assets, you can't easily liquidate that and move, because a) you're living in it and because b) you don't have enough assets to go towards the expense of moving and the expense of getting a new house in another city.

China had their great migration over the past 20+ years, where there was a huge transfer of population from rural areas to cities on the coast, as the poor and middle class saw skyrocketing standards of living. The US similarly saw its great migration during the mid-20th century, again when standards of living were rising their fastest for the poor and middle class.

2) Economic growth patterns of the country changes the risk equation.

A reverse way of interpreting the above situation would be to note that when there are greatly rising standards of living and economic growth, "moving to the city" is a fairly easy bet that you'll be able to improve your standards of living by finding work there, despite the increased expenses of city living. With slower growth and much less rise in standard of living among the poor and middle class, that becomes a much riskier bet and may make a significant marginal difference.

3) City jobs may be particularly increasingly relying on skilled labor, rather than unskilled labor.

In the past, you could move to a city and work in a factory; hence almost any rural farmhand could move. If you need a degree in computer science to get a city job, well, that's going to limit large potential population shifts.

3-b) City jobs may be increasingly relying on accreditation from fancy schools.

Given that a lot of the best colleges are themselves in cities/on the coasts, and people tend to go to college *relatively* near where they grew up (or at least in-state tuition makes a big difference), this could cause people to get indirectly geographically-eliminated from the city job pool, by having degrees from less fancy schools, and hence never move to the city to begin with. This would generate a feedback loop, where people in cities, tend to stay in cities, and people not from cities, tend to not be in cities, over the generations.

4) Why has no one mentioned family?

The number one consideration for people who end up staying in their home town or state is family. (http://www.pewsocialtrends.org/2008/12/17/who-moves-who-stays-put-wheres-home/)

"Asked why they have not left their hometown, ―stayers cite major reasons such as the tug of family ties (74%), the desire to remain where they grew up (69%) and their belief that their communities are good places to raise children (59%)"

We might group all of those reasons together and say essentially that historical population dispersal has a lasting impact on where future populations end up, for a variety of social reasons where people prefer to stay where they grew up, all else equal.

Combine this with #2 and say that this is another factor to overcome; when economic growth is high, it tips the balance towards moving; when it's lower, reasons to stay prevail.

5) You don't have to be in a big city to get 'entertained' anymore.

A lot of entertainment used to involve needing to be *physically present*; live music, going to cinemas, seeing the latest play, etc. But you can get all of that on an iPhone now-- what do I need to move to a big city and endure all those costs for, when I can stay in suburbia or ruralville and get like 80-90% of the same entertainment value on the cheap?

So much number 4. 66% of households are a family, and most of us have extended family, so this comes up in one way shape or form. Ignoring things like the desire to be around people we want to be around (for whatever reason), is missing a huge chunk of any puzzle

There's nothing La Traviata on an iPhone.

People in smaller cities seem nicer-behaved. Higher rents also means larger cities are denser, another congesting and polluting factor. Larger cities are more likely to have criminal areas. Also, more inequality. And problems related to diversity (not just in the mainstream political sense, but just that more types of people imply more conficts), but this is probably just point #4. Point #3 seems related to decentralization.

Since I started reading Steve Sailor I always ctrl-f for immigration. And of course race.

The history of the US in the Twentieth Century is that White people create great places to live. They are successful. They attract Black people to move there. Very quickly those places become horror zones and White people flee. The North-East was doing fine. Full of cities that were growing. Then the Great Migration started. Philadelphia for instance grew to 2 million by 1050. It is now down to 1.5 million. The only North-East city that consistently gets voted a good place to live is Pittsburgh. Which as of 2010 was 66% White.

California grew and did very well. Then it attracted a large non-White population. LA is now about 30% non-Hispanic White. So it will be interesting to see if Hispanics (just under 50% of the population) will follow the African American or the White example. It is not looking good so far. In the same way Whites are a minority in San Francisco but with East Asians, mainly Chinese, as the largest non-White minority. Their politics have always been dysfunctional in China so they are likely to continue to be dysfunctional in California.

In the end, economic growth and innovation takes places in areas that are mainly White. So Atlanta is looking a good bet. Maybe even New York. The Pacific North-West has been doing well. The problem is that as these places grow they will attract minorities and begin the long slow descent into Detroit. So people will have to move and economic growth will take place somewhere else.

Only in an ethnically homogeneous place like South Korea can half the population live in one city.

Same as the old year - without any subtlety, though one never knows what sort of ethnic cleansing might occur in the comment section.

Was it? I don't think so. Even if it was you are asking the wrong question. The right question is "is it true?"

But OK let me rephrase - White people create great places to live because of their unearned White privilege. They cause economic growth by their unchecked and yet creative pillage of the world's resources. They are Ice People. Black people, because of slavery, systemic racism and institutional discrimination, are denied the benefits of increasing property values. And because White people are evil. They deliberately move away from areas with large numbers of Black people, taking all the jobs with them and leaving nothing but oppression and debt.

Is that racist? If not, I am happy to discuss that. The reality is the same either way.

'Was it?'

No one is clueless enough to ask that question in good faith, especially when you straight out start by saying that you search for 'race' when reading.

'I don’t think so.'

Well, what term would you prefer to use when using race as a filter?

'White people create great places to live'

Actually, when talking about the mid-Atlantic and its colonial heritage, you are talking about white people using slaves to create great places for white people to live. Though that was in the 18th century, not the 20th, of course.

'Is that racist?'

Even you are not clueless enough to ask that question in good faith after what you wrote.

'The reality is the same either way.'

Which reality? The one where America's cities are being overrun by those parasites who threaten America's prosperity in large East Coast cities because they are being filled with Irish, Italian, and Jewish immigrants - oh wait, this is the 21st century, not the 19th. A modern era where the term white continues to be a fantastically malleable term in American discourse, and we conveniently forget what racists generations ago believed when it is inconvenient to whatever argument a modern racist is making. And remember, no less august an American than Benjamin Franklin was decrying America's sad future fate unless real Americans stopped the wrong swarthy Europeans from immigrating. Like Germans - https://en.wikipedia.org/wiki/Observations_Concerning_the_Increase_of_Mankind,_Peopling_of_Countries,_etc.

"Since I started reading Steve Sailor I always ctrl-f for immigration. And of course race."

Steve Sailer comes up with brilliant ideas and is a fabulous writer, but unfortunately cultivates a nasty racist following. A good example of how cures can be worse than diseases.

By some strange coincidence, Sailerites always come to the conclusion that anyone who isn't a White non-Jew is too smart or too stupid, too violent or too peaceful, and so forth.

See, ethnic cleansing - the New Year is clearly bringing us a burnished comment section.

One where commenting 'Of course' to someone's text searching habits is just too harsh to be allowed in the sort of enlightened public discourse that marks this web site.

Some of us like Asians, too!

Does it feel at all weird to view everything through the lens of race first and foremost? Does it produce any insights that make your life better? Does it change your behavior? Have you told any friends that you do this?

Ask him about his attitudes toward marriage - only a select few here where able to read them in the past.

In all sincerity, I would like to see those questions asked of the New York Times. But it's an impossibility, I guess. It would be the barber who shaves only those who don't shave themselves, giving himself a shave.

Such navel gazing has been successful for minorities wouldn't you say?

They're just doing what we're being taught by society at large. Being colorblind was tried by many with earnest effort but it just wasn't appreciated. There was always some systemic racist Boogeyman around the corner. So they adapted.

Being colorblind was tried by many with earnest effort but it just wasn’t appreciated.

Being colorblind works even less well for minorities.

You realize that Hispanics settled California first, BEFORE white people, right? It's why all the major cities have Spanish names.

Not according to the Census Bureau. Back when they first introduced "Hispanic" (I believe in those days they called it "Spanish-heritage", which had the virtue of being more accurate), a colleague called their help line to see if he, as a Spanish-American (i.e. the native-born child of a married couple who had emigrated from Spain) qualified. The Census Bureau guy rattled off the list of the qualifying countries-of-origin, on which two were conspicuous by their absence: "...What about Spain? Or Portugal?" "They don't count."

...But perhaps you meant "Los Indios"? ;-) I long ago concluded that the "Hispanic" category in this country was a kind of negative-image of what it was in central and south America: the more "Hispanic" blood you had, the less Hispanic you were.

It would be far more accurate (if, alas, less politically-correct) to replace the term "Hispanic" with "Indio" or "mestizo."

'to replace the term “Hispanic” with “Indio” or “mestizo”

Except those terms describe two distinct groups of people - there aren't exactly too many mestizo Mayans, for example.

Well, if you want to call them "mestizos", the native American population was around 200,000-250,000 under Spanish colonial rule. Whatever you call them, they were there first.

There were maybe a few thousand of them?

I really don’t not have a dog in this fight. But this is one of my historical fact pet peeves.

To get this out of the the way: Hispanic Americans are Americans. They’ve been fighting and dying for this country for hundreds of years. They’re part of the fabric of this nation, and Spanish is going to officially be a second language of this country in the near future. #SailerWillHaveChicanoGrandkids

That being said:

The Spanish planted some flags and set up some missions/institutionalized slavery and brutally treated the native population.

However, they did not “settle” almost anything north of the current border. There was not a massive refugee/ethnic cleansing aspect to the Mexican American War.

They did not control the border, they could not even police northern Mexico internally. Apache slave raids were common and ...productive. Mexico never “held” much of anything in northern Mexico, and they still do not hold northern Mexico, let alone California.

They’ve always been a dysfunctional country, and seem destined to remain so.

I have to say I am surprised at the odd tolerance of the moderators. That did come out harsher than I intended and other people got deleted for comments that even I thought were reasonable objections.

That said, I notice no one has bothered to engage with anything I said - whether factually or as a matter of opinion. I am deplorable it seems and that is all that is to be said.

Judah Benjamin Hur January 1, 2018 at 6:10 am

By some strange coincidence, Sailerites always come to the conclusion that anyone who isn’t a White non-Jew is too smart or too stupid, too violent or too peaceful, and so forth.

And yet Western populations are hugely, uniquely, unprecedentedly successful. Why this should be I do not know. But if you try to single out any factor, European populations tend to fall between African populations on one side and East Asian populations on the other. I would think that East Asia should have lead the industrial revolution but they still are actually having problems catching up. Europe does seem to be some sort of Golden Mean.

I did like Hazel Meade's comment. Being color-blind has sure worked out for some minorities. The Irish and Black communities have decided that political solutions are best. That has not worked out well for either although the Irish have had fewer problems assimilating. The East Asians and Jews have opted for economic solutions. America's turn to a color-blind policy has worked out very well for both.

As for Potato's objections, sure, the Spanish made it to California first. So what? It is irrelevant to my comment. In California's Golden Age it was solidly White and most Hispanics in California now are not descendents of those that were there in 1700. They are recent arrivals from south of the present border. As we all know. We will have to see how Hispanic politics go. Mexico has had problems with a sensible economic policy for a long time but they seem to be showing they can make progress. Maybe there has been a shift away from Chavez-type politics in Latin America. Perhaps not. California is one of the states pushing to make the Little Sisters of the Poor pay for abortions and is insisting that everyone has to sell abortion services. But is that a good sign of assimilation or not?

If Hispanics were taking over California, I would expect them to be banning abortion, not making Little Sisters of the Poor pay for morning-after pills. But I don't see why conservatives would have a problem with Californians becoming more Catholic either. They were happy enough to vote for a Mormon in 2012.

“The problem is that as these places grow they will attract minorities and begin the long slow descent into Detroit. So people will have to move and economic growth will take place somewhere else.“

Thanks for clarifying that you see “minorities” and “people” as two distinct categories.

yes yes he is very deplorable
now respond

"In the end, economic growth and innovation takes places in areas that are mainly White. So Atlanta is looking a good bet."

A good bet for what? Atlanta isn't exactly mainly white.

Cities once established are amazingly self-perpetuating. There are cities today that were founded literally thousands of years ago based on economic benefits that became irrelevant thousands of years ago but they are still going strong. An example is cities that were on overland caravan routes from the Far East that were superseded by maritime trade already in ancient times.

7. is an excellent point, I wonder if some Baumol cost disease theory comes into play here? Do cities monopolize certain service sector resources? Entertainment is an obvious one, but surely there are more examples. Will we accept a certain commoditization as "good enough"? We're getting less optimistic on tech continually delivering this to us, which may just be our lack of patience, but food for thought...

Another excellent post. Really enjoy reading these lately.

1) This was my first thought as well.

6) I wonder what personal preferences exactly Tyler is getting at here that don't appear under the other considerations listed.
We recently moved out of DC to the suburbs to buy a house. A number of our friends are staying in the city and seem to be trying to live the suburban lifestyle there. While the city is more convenient in a lot of ways (e.g. easy commute to work, walk to lots of things), eventually they seem to run into the more problems than benefits. Driving is a pain in the ass, the yard is tiny and there are few parks for kids, the housing stock is older and buildings often need lots of work, schools are generally worse, much lower value at restaurants, etc.

A comment section polished to a gratifying sheen, one that only reflects what its owners wish.

Krugman has no children and can't seem to imagine why large families wouldn't want to live in major cities.

Why would Paul Krugman know any ultra-Orthodox Jews? They don't belong to the same social or religious or political organizations as secular Jews. He probably know about as many devout, pro-life Catholics as ultra-Orthodox Jews--very few in either case, and not enough to make him consider their politics, their religion, or their lifestyle worthy of consideration.

In any case the Lubavitchers and other, similar groups in New York have mostly succeeded in carving out what amount to proprietary niches for themselves and are thus at least partly immunized from what they see as the corrupting influences of the Big City.

He likely knows some because pretty much everything you have said there is not true. The Chabad are largely a New York-based movement of Jewish people whose ancestors came from Belarus, Poland or thereabouts. Krugman is a New York-based Jewish person whose ancestors came from Belarus, Poland or thereabouts. Krugman is a liberal and New York Jews tend to vote Left - although that may be changing. New York's Jewish community is not that big. I would think a lot of people know each other. What is more the Lubavitch are famous for their outreach to secular Jews. It is highly likely that Krugman or some of his secular friends have children or friends who have become "born again". Even if none of them did, the Lubavitch are the most dynamic, intellectually innovative and, well, prominent Jewish community in America. No one can ignore them.

Krugman may well know some pro-life Catholics. Although he would be a lot less likely to meet them at MIT or CCNY. They may cluster in urban areas but they are unlikely to do so as strongly as the Chabad. Or do so in New York.

You have to work really hard to be offended by what I said. Really you do.

I think a Jewish kid beat up SMFS at some point in his youth.

That’s the only logical explanation for a random anti-Semitic tangent to a conversation about the Krugster.

Krugman is a liberal priest, and he preaches to the choir.

His work on symmetrical trade is still brilliant, and I choose to remember the Pre-priest Krugman, slaying sacred cows in the NYT like minimum wage and trade restrictions.

There’s still an economist in there somewhere, just dying to write about inefficiency.

Sorry but what are you talking about? What anti-Semitic tangent? Is it even worth asking what you think I said? I think not

What is a smaller city and are we thinking also exurbs of bigger metropols? Quite a big difference if we are talking about town with population 10 000 or population of 100 000. And dont know what Krugman is exactly talking about, but I dont see Washington vs Fairfax as relevant point here. Fairfax is a suburb.

Right. This whole discussion would be helped with a definition of small city. Are we talking less than 100k less than 500k?

Krugman mentions Rochester 200k population and it has indeed been declining. But a city like Greensboro NC of similar size has been growing 20% per decade.

The core city population in the Genesee Valley has been declining. This is due to intrametropolitan migration, not to a decline in the population of the dense settlement as a whole. The dense settlement is growing slowly, not shrinking. (And Rochester is not a small city).

Yes, "heterogeneous preferences" can summarize a number of these points, and is the primary reason for the persistence of small cities.

One thought experiment I like to do sometimes is ask, "If everyone had a teleportation device (but couldn't teleport across international boundaries), where would they choose to live?" It's a fun way to think about urban dynamics. I think (residential) population would be distributed similarly to how it is now, even in this new scenario. Why? Heterogeneous preferences.

Since I reside in a small town I too disagreed with Krugman. However, not mentioned by him or Cowen is the likely decline in mobility among seniors, who have led the exodus from large cities to small, mostly in places with warmer climates, bringing with them economic opportunities for younger people who follow. One can't appreciate what I mean unless one has seen the spectacular growth of places like The Villages in what formerly was nothing. My great niece and her husband have carved out a very successful business there, an opportunity that did not exist before the great migration. What happens when the current seniors are gone and future seniors don't have the resources tb be as mobile and to move from large cities to small.

Well, for some,

Its the social policies of cities

With respect to abortion and gay rights

Have caused them to move to the

Vatican City.

You’re terrible at Haikus.

What "big city amenities"? That's an obsolete concept.

Fifty years ago people would travel from smaller cities all over the US to New York in order to shop, take in a show, dine at fine restaurants, visit museums -- all those "big city amenities."

Now? If you walk around New York the most striking thing you'll see is how all the stores are national chains. You can literally shop as well in Des Moines as in New York because the stores are all the same anyway. (And that's not even mentioning Amazon and other e-commerce.)

Cosmopolitan dining has spread across the country, too. I doubt there's a city over 50,000 people without at least one artisanal slow-food, farm-to-table, or molecular gastronomy restaurant. Thing which were exotic thirty years ago like espresso, sushi, or Korean and Indian food, are now supermarket staples across the country.

Theater is _maybe_ one draw which large cities still have -- but many complain it has become a "theme park ride" and I suspect you could transplant Phantom of the Opera or Cats to Disneyworld and shut down Broadway for good. Museums? There are literally hundreds of good museums in middle-sized cities, and meanwhile the big-name museums seem to be racing each other to become irrelevant via political virtue signalling and artistic faddishness.

No, I think Krugman's article was pure wishful thinking on his part. "Someday everyone will be a New Yorker or an Angelino. We'll get rid of those flyover people who don't agree with me. And then we'll all live in perfect uniform diversity forever!"

The biggest cities still have more range among the following categories: 1. Friends. A higher portion of big city dwellers will be incomers and you will not be struggling to catch up with their life experiences as you would in a small city of say one thousand people. Ironically, this is also true of very far-away areas like mining towns. 1a. Potential marriage partners. This is a big one for young people, until they marry. Maybe some married people too, those cads! 2. Potential jobs so your spouse can work too. If everyone works at the mill, and you are married to a non-miller, your spouse is going to take a big lifetime income hit. As you say, I don't think people live in New York to access Broadway. Look at public policy in Kansas or Nebraska and see that these demographic trends are recognised by God-fearing Republicans who aren't just virtue-signalling SJW cucks, rootless cosmopolitans, Jew-lovers or whatever phrase you guys around here prefer.

So people that live in NYC have more friends than people that live in Omaha? It is to laugh.

Small cities , not rural towns. Small cities have a diverse enough economy that your spouse is going to find a job, and your chances of meeting a marriage partner is just as great. You're not exactly going to have time to sift through the entire dating population in New York to find the optimal partner. A pool of 50K potential dates should be plenty.

This. I'm saying just about exactly the same thing further down. Small cities have everything that big cities have (except maybe theater), but without the congestion and with more affordable housing.

I know it sounds a bit trivial, but the biggest annoyance going from Chicago/St. Louis/San Diego to a smaller city like Birmingham regarded shop hours. Pretty much anything food related closed around 9pm weekday, which was problematic as a grad student who was often at work until that time. Also the lack of 24 hour coffee shops. The food options themselves were actually quite good.

I generally agree with you, but I think you overstate the case. Large cities offer major entertainment and cultural benefits, but they come at a very high price, especially for non-affluent parents.

There are quality of life issues not normally addressed in these discussions. Is the city big enough to support a Trader Joe's, a Costco? Those are the kinds of questions that are far more relevant on a daily basis. I want to live near a "major league" city to have a decent level of shopping and cultural options, but definitely not a large city with expensive real estate.

The most important thing is having good public schools (which is basically the same thing as a nice neighborhood). If I had to live on the moon to achieve that, I'd be satisfied.

How do you determine, in advance of your children's going there, what is or is not a "good public school"? Certainly all good public schools have well-maintained terrazzo floors and modern HVAC systems but that can't be the definition of one. The quality of the instruction might be a clue. Do realtors arrange for prospective home buyers to spend a day sampling classes and evaluating the teachers? It seems that most parents would be advised to park their Prius in eye shot of the school when the buses are loading and unloading and check the average melanin content of the student body complexion. That's the real measure of a public school's quality.

There's no question that a demographic profile would give you a pretty good idea about school quality (% of Asians + Whites). So would test scores. You can find both at greatschools.org. And, yes, I agree that the quality of school is largely determined by the students. I'd compare students to athletes on a team. The coach matters, sometimes a lot, but the quality of players is the primary factor. I'd also point out that good neighborhoods/schools attract smart people of all races and ethnicities.

By the way, anyone who is moving should visit greatschools.org and look at elementary school data. That will give you a much better idea of the future of your neighborhood than census data.

"The view of the McKinsey Global Institute (MGI) is that there is, in theory, no limit set by technology or infrastructure to how big or how fast cities can grow — but only if business and government leaders are able to manage the increased complexity that comes with bigger city size."
The short answer is that once we get past a point in city size, then outliers cannot get a fix on the shortest path to the center.

At any given technology level, we are restricted by transaction rate, The farther out one goes, the more transaction one needs to discover the cheapest deliveries from the center. Hence, onceexceeding the transaction rate, we start making micro centers in the outer ring.

For example, 7/11 stores are the maximum spread we can do to cover the outer belt in a concentric ring. The number of 7/11 stores we can use to ring the city is limited by the rate in and out of the store for suppliers and customers. Thus new technology makes denser cities if it can increase transaction rates.

It is just like atomic orbitals, the transaction rate of light makes an upper bound on the atomic size nature accommodates. The way to measure the effect, economically is to watch how the outliers shop. The more frequent trips they make to 7/1, the more likely the city is at its limit, otherwise shoppers would go to the center to shop.

We can get qt the solution with math, irregardless of what is acking the city.
Consider that 7/11 stores get deliveries from somewhere close to center, they they have to spread their goods over a outer belt ring which expands as a circle. If we talk dense, we talk circles.

Delivery trucks and customers are rational, in the sense that we never deliver 7 eggs per half dozen, we do not do irrational remainder, everything is conserved and quantized by case and container. Computation equivalence ells us the 7/11 system must approximate Pi, with a rational fraction. The greater the transaction rate, the more digits of precision we use to spread goods.

So we can expect a sharp cut off point in city size, atomic size, multiverse size for any given transaction rate.

An important issue, we should understand this. I go one more round with my 7/11 stores.

let us say we want to expand the radius beyond the 7/11 stores ringing our circle. My spread of goods needs to be more granular to keep the logistics connections back to the center. (Remember the assumption, circular city). So, lets break out the 22/7 stores, a better estimate of the curvature. We have more precision, our goods portions are smaller and more frequent retail transactions.

7/11 sells a have dozen eggs per carton. Tell me how you are going to subdivide that except with taco trucks? Taco trucks do not maintain the logistics link to the center, you have emitted a particle, the orbit cannot hold, a combinatorial limit.

The textbook economics of city size focused on marginal benefits vs marginal costs of city (however bounded and defined) size. Textbook writers found "optimal" city sizes. But cites are about spatial arrangements. Emergent spatial arrangements bend the (imaginary) MC and MB curves. People in cities want good access and space. Good access to many things, via many modes (electronic and physical). All of this involves complex trade offs. Trouble is that emergent arrangements are always up against the durability of structures as well as the durability of land use regulation and thinking.

Textbook logistics will tell us the optimum truck sizes and shopping cart sizes based on the marginal ability to engage in a transaction.

>A significant and enduring trend is the move into warmer and sunnier climates.

Well, that shoots Global Warming Theory in the head, doesn't it?

"2. Congestion is likely to be lower."

Almost all cities of any consequence are river, lake or ocean ports. Water transport availability was the reason for their establishment. While this factor remains important in many cases, surface transport is now more significant and the once vital waterways have become an impediment to trains, trucks and cars. Even smaller cities, Missoula, MT, for instance, have serious traffic issues related to bridges over three rivers.

Some of our major inland cities (e.g., Atlanta, Dallas) came into existence as major ground transportation nexuses. The railroad broke the monopoly ports had on transportation.

I find smaller cities much more livable than big ones.
Many people think of big cities as the places that have hip, walkable neighborhoods filled with diverse restaurants and shops.
But actually you can get almost all of that in a small city too. In fact, small cities are often more walkable since there are more residential neighborhoods within walking distance of downtown, and those homes are not priced out of the reach of all but the extremely rich.
What you can't get in small cities tends to be elite cultural institutions like theater. If you want to see innovative theater every weekend, you probably want to live in New York. But if all you're looking for is a trendy neighborhood with some nice restaurants and coffee shops there are lots of places you can get that. You don't have to live in Greenwich Village. The smaller cities also have more room for development, so they can afford to put in things like bike paths.
Of course if you get TOO small you also end up with too little of that - the place won't gel into a vibrant cultural center until it reaches a critical mass. Two restaurants one bar and a coffee shop won't cut it. There's maybe a happy medium between 300K and 1Million where the downtown is a decent enough size to generate some cultural activity - maybe have a theater where broadway productions tour, and isn't cut up by freeways or cut off from residential areas. And you still have parks and can still get out to the countryside and get to work without too much congestion.

Yeah I am not sure that 300k-1 mil is a small city though. I’d think more along the lines of 100k-300k.

But even that size can be quite nice. A city that size will usually have an airport with connections to major hubs. Will often have a university, have some nice restaurants, will have decent schools, a decent hospital, and a critical mass educated upper middle income folks.

I'm counting the population of the total metro area, actually. Wikipedia says that Houston only has 2.3 million people but the metro area has 6 million. The Tucson metro area is 1 million, 500K in the city proper. Ann Arbor michigan (definite a "small city") has 100K in the city proper, but 300K in the total metro area.

Ann Arbor et al is now an appendage of Greater Detroit.

That's like calling Istanbul and appendage of greater Troy.

Ann Arbor has been on the edge, and part, of the Detroit Metro area as long as I can remember (I grew up very near to it)

That’s like calling Istanbul and appendage of greater Troy.

Sober up.

You can see on this population density map


That there is no longer a discrete dense settlement around Ann Arbor. Tract development in Wayne and Washtenaw Counties has now made for a continuous urban field.

I still haven't figured out why anyone would like to live in a city. The price for housing is ridiculous. The time it takes to get anywhere is ridiculous.
I commute 15 minutes through scenery people come from around the world to see. I know the people I deal with every day. People who live here are very well travelled; their budgets allow them to explore the world instead of paying for an absurd mortgage. 15 minutes to world class skiing, with a very good likelihood of cutting fresh powder. Bears, wolves, deer, elk wandering through our yards. Little to no crime.

I have had the opportunity to work with crews from the larger centers. Skilled technical trades. We chat and after a few minutes know each other's caliber and experience. The question is how. How do I make a living in a small town. They have families, they don't see their kids. I'm the lucky one. They make likely twice what I do. It buys them a worse lifestyle.

You live in Boulder Colorado? (Taking a guess...)


Lots of places like this around. Every morning where we have our shop all the other tenants talk about their skiing, mountain biking, mountaineering, kayaking plans for the next day off. The shots in this movie were most within a few blocks of where I work.

Well, that's a very small town. I would imagine that the food options would be fairly boring and you'd have to get into the car and drive to get anywhere. And no delivery except maybe pizza and the one Chinese restaurant around.

To each his own Hazel.

Heterogeneous preferences are the reason we can all live somewhere and afford it.

Well, yeah, I'm returning ot the larger point of the thread - there are reasons why people move to big cities, but lots of smaller cites have those things too. Rural towns definitely don't have hip walkable cafe districts.

6 The whole idea in that article on Civ 6 was made impossible just a few weeks after the article was released, in 2016, because a far less extreme version of the same principles was too dominant a strategy: Adjacency bonuses are now limited to 1 per building, so instead of dense megalopolises, instead the game favors clusters of 2 or 3 cities that have shared factories, stadiums and the like.

Perhaps the most interesting part of Civ 6 though is its computer opponents. A game of Civ goes for up to 500 turns by default, which is probably 20 hours or so at a regular pace. Given that people don't like to lose, and that the game is ultimately just a power fantasy, the AI pretty much goes 'dumb' and twiddles its thumbs at about turn 150 in each and all difficulty settings. So if you reach, say, the medieval era, all a player has to do to win is build armies and throw them at the poor computer. It's amazing to me that something like this is all rational behavior for a videogame company.

Geoffrey West’s recent book “Scale” addresses this topic in an interesting way.

"1. As Alex points out on Twitter, rents are permanently lower, and many people don’t value big city amenities very much."

But aren't rents a function of demand, which is the very phenomenon the post seeks to interrogate? In other words, is it not circular to cite higher rents as the reason more people don't want to live in the big city?

Perhaps the response here is that rents are also a function of supply, and that some cities face geographic constraints on their ability to grow horizontally. Vertical growth, in turn, could be constrained by restrictive ordinances, which is related to point 3.

Yes, but you still need to move all those people around.

Rents are also impacted by alternative uses of the space/land (think of it as a rolling auction, where the current use has to remain the highest and best use or else be displaced). In larger cities, residential rents in/near the city center are pushed upward by office rents in the core.

Manufacturing: it doesn't make sense in these mega cities. The rents are too high. The regulation is too suffocating. The politics are unstable.

that's particularly true for export oriented business where global price competitiveness matters. Build to scale? No.

Factories do not get better by being massive. It's just as well to scatter smaller ones around to where land is cheap, transport Is efficient, and a living wage is modest.

I'm surprised that #8 gets so little attention as we talk about climate change. As average surface temperatures are increasing, the effect for Americans is being swamped by "voluntrary" climate change. For example, I recently moved from upstate NY to Bethesda, MD, which constitutes a personal climate change of +8C - a value that climate scientists consider "catastrophic" (don't worry, I'm ok). Maybe as climate change gains steam and the warmer weather comes to people where they now live, we will see a reduction in the north-to-south population flow. I'd like someone to calculate the rate of temperature increase through voluntary climate change (net north-to-south human migration) vs greenhouse effect climate change.

But if the Antarctic ice sheets melt, A/C won't save you, that's why people worry about global warming, peak oil, and the like. Modern economics is all about GDP growth without accounting for externalities (because they cannot be easily quantified). Sort of like not counting housewives and the black market, but different.

I grew up in the Corn Belt. The agricultural work force is smaller than two generations ago. Alternatives like Caterpillar and John Deere have outsourced. Retail jobs started declining when Walmart showed up, and on-line shopping delivered the coup de grace. The small towns around my old home have a high median age with nursing homes and funerals as the main industries. I don’t think that this is good, but that is the way it is. Illinois is shrinking in population; I guess that declines in Peoria, Moline, and Rockford are greater than the Chicago conurbation.

Good analysis, and the answer is: "Because look at Manila". Also I think city size is a power law so in theory you won't have an upper limit (again, Manila), but obviously some cities are more optimal than others.

Cities like Manila have rent-seeking benefits that make DC seem like a libertarian haven. Formal rules affecting non-Manila area business and informal effects of having all the power brokers concentrated in one place make it hard for national concerns not to have some central presence. Even rich political families in the provinces will often send their kids to Manila for some schooling. This is true to a large extent for Thailand, Malaysia, Indonesia, or for that matter Russia or Brazil.

In a first world political environment without the kind of centralizing rules of France or Mexico, the Philippines would have concentrated cities, but it's likely there would be 2 or three reasonable challengers to Metro Manila and Cebu especially would be larger and more developed.

We shouldn't ignore the not insubstantial efforts of the lefty politicians to destroy small town America. Here in NY our governor has bequested a higher minimum wage and paid leave for his subjects. But outside of NYC and its environs NY from an economic perspective doesn't have much going for it. A recent WSJ article focused on some town in western New York State that a few decades ago was a very substantial city because of heavy industry, which is now long gone. Its main street is empty store fronts because of the Amazon effect and low economic activity in the region. The article talked about a young woman who had graduated from the local high school and the best job she could find was working as a part time pizza delivery person. Her aspiration was a full time pizza delivery position. Well, if she ever gets that job she will also have paid leave and a higher minimum wage. And she can thank her Governor Andrew Cuomo for his kindness and generosity.

Could you explain how "lefty politicians" created conditions and results that you bemoan?

It's false to claim that Chattanooga and Birmingham are on the rise. Chattanooga is stagnant while Birmingham has lost 30% of its people in 50 years. The growth in the southeast has been in big cities, Raleigh-Durham and Atlanta.

Greater Birmingham is a dense settlement within Jefferson and Shelby Counties in Alabama. The sum of the population of Jefferson and Shelby Counties has grown by 18% since 1980.

That’s a 0.4% compound annual growth rate. Not exactly blowing the doors off. (Less than half the US population growth rate over the same period.)

The post by Krugman has an interesting theme. #9 is appropriately posited by Tyler as a counterpoint to keep in mind when discussing the growth of cities (i.e. city size and optimal size). But the papers linked are, IMO, very poor in quality or relevance. Especially the McKinsey piece, which seems to me like self-celebrating corporate pap. I believe this paper offers a more comprehensive and useful overview: http://www.mopt.org.pt/uploads/1/8/5/5/1855409/camagni_capello_caragliu_final.pdf

"Congestion is likely to be lower. Why should the larger city have worse traffic if it has proportionately more roads?"

There's any number of reasons why the question, well, begs the question. The only way you could maintain "proportionate" traffic levels is to widen the roads, not merely to have more of them. Also, larger cities tend to be denser, at least on a localized basis: even Los Angeles is a network of dense cities separated by low-density hills and valleys. And even then, there's only so many ways that you can plausibly build transport arteries from point A to point B.

New York is an interesting case in point. New York reached its limits more than a century ago, and even then, it was the product of an administrative fusion of two large, adjacent cities, New York (Manhattan + The Bronx) and Brooklyn (Brooklyn, Queens, and Staten Island). There were certainly peripheral areas that were /de facto/ parts of the urban zone--Nassau and Westchester in New York State, and several adjacent counties in northeastern New Jersey and southwestern Connecticut--but in an era when any even medium-distance transportation was by rail, there are limits t how far you can coax people into traveling between where they live and where they work.

If anything, I'm expecting more decentralization as a result of telecommuting.

I agree. I think Krugman is behind the times on this. Why is Boise Idaho growing so fast? Why did hundreds of thousands of people move to Colorado in the last decade?
Also, newer industries are going to open facilities out in suburbs or small cities not in dense urban cores, because there's more space to build a new building there.

Regarding telecommuting: I think that would be a boon to smaller towns and more remote suburbs. However, corporate America is suddenly disenamored of it, starting with IBM's recent move to eliminate work-at-home privileges.

To say that corporate America is disenamored simply because of a choice by one, or even a few, is overgeneralization, and very poor logic. Bad form. That's how one starts fake news.

I'm curious how much time you've spent in the South (point #8). I grew up in Florida, neighborhood went from 1/2-a-mile-to-reach-pavement to 9-units-per-acre while I was there, and only continued after I left (and I still visit, friends, family and in-laws are all there). What I saw was sprawl-growth that continued till resource limits were hit; roads become usually-clogged, water got scarce (county I grew up in, there is a separate water system for irrigation with treated sewage; any of y'all got that where you live? Tampa was debating whether to desalinate sea water or sewage, because sewage is easier, but drinking sewage is a hard sell.)

Lived some years in Houston, it seemed like a similar process there; roads are expanded to relieve traffic, then traffic expands to fill roads.

Spent some time in Silicon Valley, I think I got to see one of the last cherry orchards on El Camino ripped out for housing, same. That giant suburban blob was once a bunch of small towns and agriculture (see e.g., "Plane Crash at Los Gatos / Deportee").

I am currently reading Scale by Geoffrey West. Not sure I'm convinced, but definitely interesting!

Chapter 7 is about how cities change with size. His thesis is that within countries a wide range of good (e.g., wages, patents, diversity) and bad (e.g., poverty, disease, crime) characteristics of cities scale superlinearly with the population of the city - the power is about 1.15.

So, if one city in the US is 2 times as big as another, these characteristics will be 2^1.15=2.2 times higher. If one city is 10 times as big as another, the characteristics will be 10^1.15=14.1 times higher. And so on. [Same for cities WITHIN Japan, or WITHIN England, etc. The 1.15 rule would apply across cities WITHIN each country, but the curves would be different ACROSS countries reflecting their different national cultures/economies/etc.] These are of course general trends - you can always find counter examples.

What he's basically saying is that good things get better and bad things get worse as cities grow, and they both change more than just linearly. There is no free lunch.

Anyway, is anybody surprised that good and bad characteristics BOTH grow with city size? I'm surprised by the apparently ubiquitous 1.15 scaling rule, but not by the basic idea.

Given that, it's not surprising that different people react differently to cities of different sizes where both the good and bad have been accentuated.

…once upon a time dispersed agriculture ensured that small cities serving rural hinterlands would survive. But for generations we have lived in an economy in which smaller cities have nothing going for them except historical luck, which eventually tends to run out.
Krugman suggests that eventually many smaller cities will indeed fade away, although the process of equilibration may be a long and slow one. All of his points are well-founded, nonetheless I can see a few factors favoring the continuing existence of small cities on a greater scale than many might be expecting:

"Well founded"?

You might suggest that your interns take time out from posting vulgar comments here to actually investigate whether, in fact, small cities are 'fading away'. The Census Bureau puts the data online. You'll find some examples of demographic stagnation and small incremental declines among dense settlements with populations between 50,000 and 200,000; you'd be hard put to find one example of a city fading away. (And his attribution of the position of extant cities to 'historical luck' is inane).

Here's a test. Have a look at census data for Nebraska and Kansas. Some counties in those states reached their peak population around about 1930. You do see therein sharp demographic declines in local areas, but you'll discover that population centers which had reached a critical mass ca. 1950 have been able to hold their own in the intervening decades. The critical mass is around about 15,000 persons.

Now look at some population figures re counties enveloping small cities (reflecting estimated changes, 2010 to 2015/16):

Cumberland County, Me (Portland), +3.7%
Chittenden County, Vt (Burlington), +3.2%
Hillsborough County, NH (Manchester, Nashua), 1.8%
Broome County, NY (Binghamton), -2.6%
Oneida County, NY (Utica), -1.6%
Erie County, Pa (Erie) -1.6%
Kanawha County, WVa (Charleston) -3.5%
Stark County, Ohio (Canton), -0.6%
Muskegon County, Mich (Muskegon), +0.7%
Saginaw County, Mich (Saginaw), -3.9%
Calhoun County, Mich (Battle Creek), -1.3%
Brown County, Wisc (Green Bay), +5.0%
Madison County, Ind (Anderson), -1.8%
Vigo County, Indiana (Terre Haute), + 0.1%
Woodbury County, Iowa (Sioux City), +0.6%
Black Hawk County, Iowa (Waterloo), +1.4%
Buchanan County, Mo. (St. Joseph)., -0.3%
Frederick County, Va / Winchester, +7.0%
Daviess County, Ky (Owensboro), +3.1%
Sullivan County, Tn (Kingsport), -0.1%
Carrabus County, NC (Concord), + 13.0%
Spartanburg County, Sc (Spartanburg), +6.0%
Dougherty County, Ga (Albany), -4.8%
Clark County, Ga (Athens), +6.9%
St. Lucie County, Fla (Port St. Lucie), +10.3%
Escambia County, Fl. (Pensacola), +6.9%
Lauderdale County, Ala (Florence), -0.4%
Harrison County, Miss. (Biloxi/Gulfport), +8.6%
St. Tammany Parish, La (Slidell), +8.5%
Sebastian County, Ark (Fort Smith), +1.6%
Comanche County, Okla (Lawton), -1.6%
Wichita County, Tx (Wichita Falls), +0.3%
Jefferson County, Tx (Beaumont), +1.0%
Santa Fe County, NM (Santa Fe), +3.1%
Pueblo County, Colo (Pueblo), +3.8%
Cache County, Utah (Logan), +9.0%
Natrona County, Wyo (Caspar), +7.4%
Cascade County, Mont. (Great Falls), +0.5%
Bannock County, Idaho (Pocatello), +1.9%
Deschutes County, Ore (Bend), +14.9%
Yakima County, Wash. (Yakima), +2.6%
Shasta County, Calif (Redding), +1.4%

I'm seeing regional problems and underperformance, not fading away.

I'm not disputing your assertion, but it IS true that 80% or more of US population lives within 50 miles of a major metro area. I live on the Front Range in Colorado, for example, and though I've probably been to San Francisco more times than Denver, it still influences all the suburbs in many non-obvious ways. I suspect the same is true for a number of towns in the 10,000 to 100,000 population range.

Fairfax basically IS DC.

The biggest problem with small cities is limited mate selection.

A city of 90,000 will have about 3,000 women between the ages of 20 and 25. That's adequate (for those who didn't have the sense to marry their college girlfriend or their 2d cousin).

This has to be a parody.

Art Deco recommending cosanguination?

What the f.

You forgot the simple individual question of how many "possible" jobs are available within commuting distance. The more specialized you are, the larger of the city required.

Megan McArdle pointed out the flaw in everyone’s argument when discussing where London fianance Jobs May move post brexit. The primary issue is how good is it to be rich in a given city. Once you’re making 500k or 750k in Milwaukee or Cincinnati you’ve hit the limit on how well you’re can live. In NYC or LA or SF, with each $500k increase in income. you’re able
to access ever more exciting lifestyle choices.

Reading the comments here and the concern about affordable housing, it seems safe to say many of those commenting are of fairly modest means. As such they are unlikely to be the dynamic job creators that drive a city’s economic dynamism.

That must be what accounts for the low value real estate in places like Aspen, CO and Bigfork, MT. Poverty stricken Ted Turner was, and perhaps still is, one of the biggest landowners in Montana. What exactly are the more exciting life style choices available to only the super wealthy in NYC, LA or SF? Maybe having dinner with Harvey Weinstein once was. Running into Mrs. Bill Clinton on one of her strolls down Madison Ave. would be cool. But those aren't "life style choices". The lifestyle choices available in a major city are pretty much the same for everyone. You live in a building surrounded by other buildings, ride around in cars or public transportation and have a large variety of dining choices but the most likely excitement will be as a witness to a crime or the victim of same.

Can I ask you a serious question? Do you honestly think that incoherent rant made even one valid point?

I mean honestly, Aspen population 6,871, Bigfoot population 4,281 vs. Manhattan population 1,641,000? That’s your argument? Seriously? How stupid are you?

the most likely excitement will be as a witness to a crime or the victim of same

Sigh....the stupid, it burns.


with each $500k increase in income. you’re able to access ever more exciting lifestyle choices.

That's probably important to know for the overwhelming majority of the MR commentariat. But, what would be an example of an "exciting life style choice" that a millionaire resident of NYC or LA would have that a similarly wealthy resident of Milwaukee would not?

many of those commenting are of fairly modest means. As such they are unlikely to be the dynamic job creators that drive a city’s economic dynamism.

Dynamism in business is driven by entrepreneurs, most of whom have started small, not giant bureaucratic enterprises, who also inevitably started small. In fact, with the notable exceptions of the legal and finance industries, many large businesses have their headquarters in other than megalopolis cities. Seventeen Fortune 500 companies are based in netherworld Minnesota, ten in Colorado, seven in Oregon, and ten in Wisconsin.

Not to be argumentative, OK maybe to be argumentative, . .. 44 out of 500 isn't much of a percentage.

On the basis of the NYT link, each NYC resident in 2017 had the opportunity of witnessing or being the victim of about 260 reported felonies on any given day.

Chuck, this is far from the dumbest thing you’ve said on this website, or even on this page, but I do feel the need to point out that those 260 reported felonies are in a city of 8.5 million people, not counting commuters andd visitors. Assuming you have 100 notable human interactions a day and felonies are evenly distributed (very aggressive assumptions on both ends) you might see a felony a year if you never leave town. Another angle: this is the equivalent of 3 felonies a day in a small city of 100,000, which would be wholly unremarkable just about anywhere in the world. If there aren’t a few felonies a day in your town (on average) and your police force is bigger than +/- 20 officers you’re setting money on fire. Just to pick a city of about 100k at random, Renton, WA has a force size of 148.

Oh, sure. The comment isn't an observation of life in the Big Apple anyway. First of all, it's a refutation of the link supplied by John, a link that describes a reduction in violent crime in NYC. It doesn't state that there is NO felony activity there, when, in fact, there is. The reality is that for most people witnessing a crime or maybe a serious accident is the most exciting thing they'll see that day or month or year. That's why stuff like that is televised on the evening news. Because it's exciting (perhaps a little depressing as well).

Second,this is actually true in places like Little Rock and Memphis, too. So, what's so "exciting" about life in NYC in comparison to lesser cities? Well, OK, the Allegheny Courthouse in Pittsburgh would be an unlikely target for terrorist plane hijackers but I'm not sure that's the kind of excitement that anyone craves.

I'm still curious as to what an "exciting lifestyle choice" might be and why I couldn't have it in Tampa as well as NYC, especially if I've got all that money.

I totally agree. When I was only making $500K a year and living in Portland, my lifestyle options were limited to having orgies at the local BDSM clubs, but when I moved to LA and started making upwards of $2 million a year I was able to import all the top quality hookers and coke to party in international waters on my private yaugt. Doesn't everyone know these things? (By everyone, I mean everyone of any importance).

First of all, very few people are going to be faced with the question of what to do with each $500k increase in income.

Second, just what are these exciting high income lifestyle choices? I'm personally living well below my means in part because I'm not seeing lots of exciting things I want to spend money on.

Third, for some high income people the attraction of high income is multiple residences. I know people who spend their time moving between 2-3 residences, none of them in NYC, LA, SF. They like different climates, social environments, scenery. Big cities not doing it for them.

People in places like London or New York who have time and money tend to leave the city at every available opportunity. The best thing about these cities once the novelty of the dining and entertainment options wears off is the availability of non-stop flights (or places to park one's private jet for the super-rich) to lots of more interesting places. The one thing these places can offer is networking opportunities with their high concentrations of wealthy, well-connected people. That said, many if not most of the Americans in the Forbes 100 list don't live within the city limits of New York or L.A. Silicon Valley billionaires prefer the staid suburbs of San Jose and the Waltons and Warren Buffett live in "flyover country."

I figure that the amenity effect for megacities is largely indirect. For ordinary folks, the commute is typically monstrous, and art, culture, and amenities are astronomically expensive, and prohibitively time-consuming to reach via glacially slow buses and slow, crammed subways. People in affordable sections of NYC often go their whole lives without ever setting foot in Lincoln Center or MoMA, except maybe once on a high school class trip. The amenities don't matter because they are almost unreachable. And never mind that anyone of ordinary means must arrive by hot stuffy transit, so they will even *smell* like they don't fit in!

But for the managerial/billionaire/rentier class who command ordinary folks and dictate where they must live, the amenities are useful for signaling - i.e. for looking down one's nose at the philistine deplorables. After all, who among the bossy, overpaid rentiers would deign to live in Podunk (or, as one teacher called it when I was growing up in NYC, Squedunk)? And this extends to many of the other factors beyond just amenities. Much preferable to live in a conspicuously *classy* place, especially if one can readily afford taxis or limos.

Thus the herding into megacities is mainly about forced pseudo-choice commanded as a matter of fashion by an aristocracy, not about anything voluntary. That places it outside the scope of economic analysis (of anyone but maybe the aristocracy.)

Have you considered the possibility that big-city residents don't mind these things as much as you do, and thus make different choices? Human variation in tastes seems a little more plausible than a monstrous elitist conspiracy.

There is much more Palinesque pseudo-populism than analysis packed into this comment. I grew up in a solidly middle class family and we would take day trips about two hours in each direction (including parking and taking public transit into Manhattan) to visit museums in New York. If you have the time, energy and money to visit, say, Six Flags, you can easily afford a day trip into Manhattan to visit a museum or two (some of which are quite affordable and/or give discounts to children and students) if you live anywhere in the general vicinity.

There are many global surveys comparing livability (defined different ways) across cities. Why not use those data to infer something on this question? It strikes me that winners are usually 1-3mln population cities in rich countries with access to sea/mountains, temperate climate and good public transport infrastructure (typical winners Copenhagen, Zurich, Geneva, Auckland etc). The mega cities never win - and they dont grow much btw. What does the growth/size curve looks like for cities across the world right now???

9. Driverless cars

I wouldn't mind living farther away if I could easily sleep or work when I travel to the big city.

Any time I spend commuting is wasted time for me (and most days I use a subway, not a car). A small separation from work is OK-- the commute creates wind-up and wind-down time. But for me that time span is ideally no greater than 20 minutes.

The "Henry George theorem" says that, at some size, the marginal quality of additional locations to live and work will decline faster than the agglomeration benefits can increase. Nothing has changed except that, for some group of potential residents, that equilibrium size might be larger than before.

"So while Rochester and Flint decline, Chattanooga and Birmingham are on the rise. "

That's kind of a lazy analysis. Yes, Flint is in bad shape. But Traverse City (much farther north and *much* snowier than Flint) is thriving. As is Ann Arbor -- 45 miles south of Flint with the same weather and similar inland location as Flint.

And, why, BTW, is the Traverse City metro area (hundreds of miles from the nearest major city) booming (40% growth since 2000)? Ready access to natural beauty -- something where large metro areas rate poorly. The larger the metro area grows, the harder it is to escape to rural and wilderness areas AND the more people you have to share with once you get there. California, for example, has wonderful national parks (Yosemite, Sequoia/Kings Canyon) but they're too damn close to tens of millions of people and are generally overcrowded. Utah's parks offer a much better experience. If you care about that, Salt Lake City is a much better option than LA or SF.

While we're at it, (1) Monroe County, NY does not have a declining population, and (2) Greater Rochester isn't a small city.

Traverse City attracts a lot of retirees with good incomes. However if you have to work for your income northern Michigan is not so favorable a place. Tourism, and some agriculture, are about the only industries they have, and that's very seasonal.

Krugman would be more convincing if the major metropolis of the word were actually diversifying. In 1950 about 30% of Detroit's payroll was tied up in automobiles. Last I checked over 40% of NYC's payrolls were in finance, and that was rising over the last 20 years. So exactly what industries is NYC capturing? How about Boston? Well it is has been a major educational/medical center and continues to dominate payrolls. Chicago? Are they not in the middle of retrenchment?

Further, are all these old industries actually dying? Automobiles are bigger business than ever, machinery is only increasing in value, even agribusiness has had pretty stable payrolls. The fact that these industries have decamped to Right to Work states, foreign countries, and the like is not about cities failing to pick up new industries, it is about the industries we have not sticking around.

I suspect that insofar as there is any trouble for small cities, the biggest problem is trying too hard to be the next New York (or Raleigh or Nashville). It would be pretty idiotic to run every city as though it is a financial sector with a highly skewed income distribution and virtually no physical footprint. I suspect that many small cities will do quite well precisely because some industry will set up shop where it can have a disproportionate weight. I cannot imagine that everyone will always want to play second fiddle to finance, though perhaps block-chain or some other innovation will diminish the role, power, and profitability of finance.

I am also extremely skeptical of all the claims about metropoli having some special magic that attracts residents. When I look at all the docs I know, who can literally get jobs just about anywhere, I have yet to hear a single doc say they moved to NYC for the theater or Seattle for the restaurants. What draws them to these areas is the ability to find a properly statused job for their spouse as a lawyer/accountant/politician/MBA executive/engineer. Family, cost of living, space, and climate have all drawn many of them to smaller cities (e.g. Cleveland, Toledo, Decataur, Terre Haute, Prescott, Provo, Billings, Pierre, Biloxi, Mobile, Grand Rapids, Colorado Springs, Manchester, Bangor).

Yep -- there are network effects to a given city. High Finance in NY, tech in SJ, TV/movies in LA, insurance in Des Moines, Oil in Houston, Government in DC. If you don't belong to the network, the city does not make sense.

For this reason, cities are limited in size to their network effect industry. The priciness of the city though seems tied to land-use rules and density. Density is expensive. SF, NY are surrounded by water. Los Angeles is surrounded by mountains and has strong laws against building height.

Sprawl holds down costs.

"The fact that these industries have decamped to Right to Work states..."

With respect to autos, that's not really what has happened. It's true that transplant have put their assembly plants in RTW states where they can avoid the UAW, but that consideration doesn't apply to white collar employees -- many have set up R&D/engineering centers to take advantage of the automotive talent in SE Michigan. For example:


No, Toyota doesn't build any cars in SE Michigan and probably never will, but Apple doesn't assemble iPhones in Silicon Valley either.

That’s the whole point.

Companies are moving jobs to more acceptable locales. An automotive engineer can keep his job, who cares. The actual jobs disappear and people get angry and vote for Trump.

How do these theories interact with Economists' love of bashing zoning regulations?


Recently had to drive down to the NYC and NJ area for the first time (relatively new to US) from my MA small town. The traffic and congestion was of the sort that made me question how any sane person would want to live there. And it still surprise me that properties in many parts of Manhattan (at least many parts of it) is really quite worn down; not really particularly attractive.

So I see plenty of reasons to prefer (some) small towns over e.g. NYC or DC, and as Tyler, I would prefer status quo (relatively charming small town) over a move to e.g. NYC with a 50% increase in nominal compensation. Such preferences are probably quite common, and will make many small towns survive for a very long time despite (on average) less dynamic and rewarding labor markets.

Most people in NYC don't drive on a daily basis, which is how they cope with this.

When I visit (from near Boston) I take mass transit to Amtrak, train to NYC, then either walk or bike to where I need to be. I don't know if that helps with "and NJ area".

What's the effect of metropolitan size on fertility of a 100 IQ white American woman?

I don't think the difference is huge, but I do believe it's real.

An interrelated point to keep in mind is that inland big cities typically can expand almost 360 degrees, while coastal big cities can only expand inland. So, even if all else were equal, housing costs would be higher in geographically constrained San Francisco than in Dallas.

(One reason Los Angeles is so huge is because it started out somewhat inland (City Hall is 16 or 17 miles east of the Santa Monica Pier) and initially expanded toward the ocean, not away from it like most coastal cities do.)

Yep. And remember that the only remaining places that Dems thrive is coastal cities -- enormously expensive places to live that cater almost entirely to the very rich and the very poor (who are willing to put up with the insane crime rates and truly dismal living conditions). Naturally they obsess every day over income inequality.

Dem voters either live in these places or have graduated from them to live out their late years quietly. Vermont is nothing more than a retirement community for New Yorkers.

A number of 'inland' cities are similarly constrained. Chicago, particularly, but also Cleveland, Milwaukee, and Detroit (unless you consider southern Ontario part of the Detroit metro area). The same is true of Toronto.

Richard Henry Dana Jr.'s Two Years Before The Mast describes the ranchos that comprised the populated areas of the California coast in 1834. They were self-sufficient Mexican haciendas whose source of income was the sale of cow hides for the New England shoe makers. From these ranches sprang in amazingly short order the big cities of San Diego, Los Angeles and San Francisco.

10. Fear of extreme dangerous events is associated with cities. Terror attack or global economic meltdown or other extreme events are greater events in big cities. Ditto crime spikes.

11. Online buying is hugely reducing the advantage of big cities for greater product selection. The demand for really big bookstores has plummeted everywhere.

12. An aversion to crowds or preference for nature. I know lots of people who live as close to wilderness as possible. "Green acres is the place to be."

13. Avoiding cities means avoiding regulations: Less need for regulations if the number of interactions with other people is lower.

14. Avoiding cities means avoiding a lot of complexity. Less cognitive load needed for dealing with humanity. Don't have to think about transit strikes, protest marches, which pay car park has an opening, etc.

Re: Fear of extreme dangerous events is associated with cities. Terror attack or global economic meltdown or other extreme events are greater events in big cities. Ditto crime spikes.

Yet you are more likely to die in a car crash than in any of those things, and suburban (and rural) car culture is thus a bigger killer than terrorism (duh) or urban crime. The latter also tends to be heavily concentrated in low income neighborhoods and involve people mired in the drug trade.

Crime rates (notably including murder rates) in the US geographically vary by over two orders of magnitude, probably 3 or more orders of magnitude if you take into account neighborhood level crime rates. https://en.wikipedia.org/wiki/List_of_United_States_cities_by_crime_rate

Death rates from cars also vary greatly by type of car driven, where you drive, whether you text or drink or do drugs and drive, and other factors.

If you let them, they do! Tokyo is the largest urban area in the world, and it's still adding population at a surprisingly fast rate, despite the fact that Japan as a whole is shrinking. Chinese cities are showing no sign of slowing down, even the ones that surpass 20 million in the metro area. American and European cities don't do this because we have much stricter zoning. But there's certainly enough demand – prices well in excess of construction cost – for New York, Los Angeles, London and Paris to grow much, much faster than they do. If you look at Tokyo, it's not hard to imagine a completely unzoned New York metropolitan area growing to 40 million or more.

My theory: introverts like the countryside or big cities; extroverts like small towns (in the latter, you interact with lots of acquaintances while in the former, your interaction is much more limited to close friends or relatives), . Krugman is clearly an (extreme?) introvert, than he favors big cities (and thinks that small towns only exist to serve the rural areas).

I'm struck by the question of just how one defines the large city? If we're calling DC a large city that seems odd to me. Then I started to think about New York -- which I generally think of being much larger than it is because I don't really differentiate all the sub-cities that make up the big city I think of when I hear New York. This is more obvious to me when I think of Manila. Manila itself is rather small but the metro area that most people would think of as Manila in the large is a mash of 16 different cities.

So that leads to the question of a big city "production" process. I suspect most Big Cities are actually the result of the growth of geographically proximate small cities. Was there specialization and trade among these smaller cities earlier that then helps lead to the apparent "drive across town for shopping for X but stay where you are to shop for Y" type results suggested? If so is that truly a Big City outcome or simply a result from the integration (and often a remaining political separation) of the small cities growth butting up against others?

What would the city that was designed as a Big City look like if built to become just that would exhibit that type of geographical specialization beyond the case were certain activities require the physical geography of a given area (which then begs the question why the big city cannot build the appropriate geography).

But one might also ask: What happened to LDMR? Does that basic economic concept not apply here? Or are the margins continually pushed outwards for some reason so we never really reach them in a dynamic, real world setting?

I'm going to suggest that "big city" means any city with a population (in the core city, not the whole metro area) of 500K people. That gives us the following 34 US cities:
New York, Los Angeles, Chicago, Houston, Phoenix, Philadelphia, San Antonio, San Diego, Dallas, San Jose, Austin, Jacksonville, San Francisco, Columbus, Indianapolis, Fort Worth, Charlotte, Seattle, Denver, El Paso, Washington, Boston, Detroit, Nashville, Memphis, Portland, Oklahoma City, Las Vegas, Louisville, Baltimore, Milwaukee, Albuquerque, Tucson, Fresno

Comments for this post are closed