Tuesday, February 13, 2007

The Myth of Superstar Cities

Joel Kotkin had a major op-ed in the Wall Street Journal today on "The Myth of Superstar Cities" (permalink). They promoted it on the front page, and it even got Richard Florida of creative class fame riled up. Here are some key excerpts, especially those relating to Houston.

Over the past 15 years, it has been opportunistic newcomers -- Houston, Charlotte, Las Vegas, Phoenix, Dallas, Riverside -- that have created the most new jobs and gained the most net domestic migration. In contrast there has been virtually negligible long-term net growth in jobs or positive domestic migration to places like New York, Los Angeles, Boston or the San Francisco Bay Area.

What as much as anything distinguishes elite places -- what Wharton real-estate professor Joe Gyourko calls "the superstar cities" -- are their absurdly high real-estate prices. New York, Boston, San Francisco and Los Angeles have long been more expensive than, say, Dallas, Houston or Phoenix -- but in recent years the difference in price, he calculates, has increased beyond all reason. San Francisco prices since 1950, for example, have grown at twice the national rate for the 50 largest metropolitan areas.

This is good news for those who hold property, but has been less than a blessing for those middle-class families who might want to enter these markets. In some superstar cities less than 10% of households can afford a median-priced home. Nationally the average is about 50%.

Mr. Gyourko traces these surging prices to two basic causes. First, there remains in superstar cities a remarkable concentration of very high-earning families who can bid up real estate. The second factor lies with the regulatory and tax regimes, which greatly limit the production of housing and job opportunities, particularly for middle-income families, not only in the city cores but in surrounding areas.


So what about the rest of the hoi polloi -- what is their urban future? For the most part, sadly, not in the "superstar" cities. Middle-class people have been fleeing the expensive cities for more affordable ones for a generation, and the migration has continued as the price differentials have grown.

Fortunately the jobs are headed in the same direction. After all, companies depend not only on elite MBAs but upon on the collective skills of middle managers, technicians and skilled laborers. Most companies also tend to be more mindful of basic costs, taxes and regulations than the average hedge-fund manager or trustafarian.

This perhaps explains why the largest companies -- with the notable exception of Silicon Valley -- have continued to move toward the more opportunistic cities. New York and its environs, for example, had 140 such firms in 1960; in 2006 the number had dropped to less than half that, some of those running with only skeleton top management. Houston, in contrast, had only one Fortune 500 company in 1960; today it is home to over 20. Houston companies tend to staff heavily locally; this is one reason the city was able to replace New York and other high-cost locales as the nation's unchallenged energy capital. Another example of this trend is Charlotte's rise as the nation's second-ranked banking center in terms of assets, surpassing San Francisco, Chicago and Los Angeles, indeed all superstar cities except New York.

The non-superstar cities have become the nation's most prodigious centers for job creation. Between 1990 and 2006, job growth in Las Vegas averaged over 6% annually; Phoenix and Riverside well over 3%; Houston, Atlanta, Dallas and Charlotte right around 2%. New York City, L.A., Boston, Chicago and San Francisco all remained well less than 1%.

Since 2000, these divergences have, if anything, actually widened. One reason is that superstar cities have continued to hemorrhage prodigious numbers of blue-collar jobs, including in fields such as manufacturing and warehousing that once sustained many urban working-class families.

To be sure, the superstar cities still likely boast far more high-six- and seven-figure incomes in finance and other business services. But in any industry this covers only a relatively small minority of workers. Overall, according to data collected by Pepperdine University's Mike Shires, the average real income -- after factoring in taxes and the cost of living -- of workers in professional business services is actually higher in places like Phoenix, Denver, Houston and Dallas than in the pricey environs of San Francisco, New York, Boston or L.A.

Some urban boosters see these shifts to the high end as evidence of superiority. After all, they argue, only the "best" remain, while immigrants, the poor and ordinary middle-income slobs migrate out to the suburbs and other less elite regions.

Yet, even here, the demographic trends are not nearly so promising. Over the past decade college-educated workers -- who once disproportionately migrated to the superstar cities -- now appear to be tilting instead to more affordable, family-friendly places. Since 2000, Riverside, Phoenix, Charlotte, Las Vegas and Dallas all have been among the big net gainers with such migrants. In contrast New York, Boston, L.A. and even the Bay Area, a big winner in the 1990s, appear to have become among the highest net losers. The big outlier here, as in many things, is Washington, D.C., where an ever expanding federal government and its satellites continue to draw in ever more educated workers.

Another intriguing shift is taking place among immigrants, the group who did much in the 1990s to help reverse demographic and economic decline in places like New York. Recent census data suggests they are increasingly likely to move to more affordable, business-friendly places such as Houston, Dallas, Charlotte and Phoenix.


This is something of an oddity, where the fashionable "left" defines successful urbanism by its ability to lure the superaffluent, the hypereducated and the avant garde -- or what Dr. Florida calls "the greatest number of the most skilled people." One wonders what true progressives like Harry Truman or Fiorella La Guardia would think of such an approach.

La Guardia or Truman understood that great cities become so, in large part, due to the strivings of the upwardly mobile middle class and families, not the elites of any stripe. It may well be true, as Mr. Gyourko argues, that as the nation grows to 400 million or more there could be a niche for 10 to 20 such "productive resorts" serving as "enclaves of the wealthy." But the urban future -- today as in past generations -- will belong mostly to places that continue to draw and nurture the middle class, which has driven the rise of most successful capitalist cities.

Hear, hear. If you're interested in my humble take on Superstar Cities (with only slightly less publicity and readership than the Wall Street Journal ;-), I did a whole week of posts on it last month you can find here.


At 12:52 AM, February 15, 2007, Anonymous Anonymous said...

I think that some of these "Superstar" cities are just maxed out geographically. For places like Boston or San Fransisco to be competitive with Houston, Dallas, Phoenix, or Atlanta, cities with almost infinite cheap land around them, would entail serious sacrifices to the landscape, and even then I don't think they could do it. Notice that LA is not growing, but Riverside is - this is because LA is full, and the growth has moved out to Riverside. The only "superstar" city this doesn't explain is Chicago.

Trivia question: So what was the one Fortune 500 company in Houston in 1960? My guess would either be Humble Oil & Refining or The Tennessee Co.

At 7:31 AM, February 15, 2007, Blogger Tory Gattis said...

Yeah, I think that F500 hq count came from actual paper archives. No easy way to confirm.

As far as maxed out, there's always the option of allowing more density to create more space to meet demand. I agree there are practical limits to that too, but I don't think most of these metros are there.

At 10:51 AM, February 15, 2007, Anonymous Anonymous said...

Trivia answer?
In 1960, Humble was part of Standard Oil, whose corporate headquarters was still in New York.

I think the answer is Texaco.


At 10:53 AM, February 15, 2007, Anonymous Anonymous said...


At 10:54 AM, February 15, 2007, Anonymous Anonymous said...

I don't know, Boston, SF, New York... pretty dense. A lot of people don't want to see their city get torn up just so they can land more Fortune 500's. And even if Boston were to say, "Okay, let's start letting highrises be built everywhere and widening all the existing roads so that we can attract more mid-level jobs," could they still beat Houston and Dallas and the other land-ocean cities? I doubt it.

At some point you draw a line in the sand. There are more important things than economic expansion. I think if you were to live in one of these places for a significant length of time, you'd see that the policies they've adopted make sense.

At 11:08 AM, February 15, 2007, Anonymous Anonymous said...

Texaco was headquartered in New York for most of the twentieth century. Gulf Oil was ranked 7th, and the Tennessee Corp. was 473rd. They owned the two tallest buildings in Houston at the time... so which one was based in Houston?

At 2:19 PM, February 15, 2007, Blogger Tory Gattis said...

It's not just landing F500 hqs - it's having an environment of opportunity and upward social mobility and affordable quality of life. Can they beat Dallas or Houston? Probably not. But if they offered substantial density around their transit stops - for instance (which they generally don't, according to a recent study at Harvard) - they could offer a lot more affordable home ownership opportunity (condos, townhomes). They also need more transportation infrastructure to the far suburbs (transit + freeways) to open up new housing areas within a realistic commute. Then existing businesses would be more likely to consider local expansion rather than stay static or even constrict, move, or outsource/offshore.

At 2:23 PM, February 15, 2007, Blogger Tory Gattis said...

According to this, Gulf Oil was based in Pittsburgh of all places. Can't find anything on Tennessee Corp, but if it's between those two, it's the only one left.


At 3:40 PM, February 15, 2007, Anonymous Anonymous said...

It's hard to put density around suburban transit stops when those are in incorporated towns that don't want to become more dense.

You can't really judge how these places work with such rough statistics... I've lived in Boston, and they don't seem to be worried about upward social mobility and affordable quality of life. The economy and the mindset are different there... they don't mind living in the same neighborhood as their parents and grandparents, often even in the same house. We are a young city - we're infatuated with growth - but these are mature cities, they've had their growth era, and now they're interested in other things.

Have you ever lived in one of the "superstar" cities?

At 5:17 PM, February 15, 2007, Blogger Tory Gattis said...

Nope. Travelled pretty extensively to many of them though. I agree a lot of those towns and residents don't want new development/density - they perceive it as crowding, plus their house goes up more in value the more supply is restricted. But citizens and leaders need a better understanding of the broader society costs of those decisions, which I don't think they have now.

At 5:43 PM, February 15, 2007, Anonymous Anonymous said...

In 1950, CONtinental Oil CO moved to Houston. ;-)

Of the cities cited, two are growing directly as a result of how expensive the two major population centers in CA have become has become: Phoenix and Las Vegas. How they differentiate Riverside from LA is a lot like trying to draw the line on the Houston area at Spring Creek and talk about how The Woodlands is growing but Houston is not.

At 6:43 PM, February 15, 2007, Anonymous Anonymous said...

Here are two websites from the United States Census bureau which may be of interest to Houston Strategies readers on this topic:


The above site has metropolitan area
populations and source links.


The second link has links to the populations, land areas, and average densities of the top "100 Urban places" in the U.S. This is almost certainly misleading because following in the great tradition of analysis about urban areas, the areas in question are almost certainly the populations, land areas, and average densities of the top 100 cities and not "urban places." The meaning of my remark is that "urban areas" can be used to denote cities, borroughs, counties, MSA's, SMSA's, or CMSA's, so take your pick.

Setting that matter aside, there are certainly items of note and interest. For example, if one looks at Census figures for 1950:


and compares them with 1990 figures:


Some observations that really stand out are the fates of various cities over time. The cities of San Francisco and New York have essentially remained stagnant in population. The city of Boston has suffered a near 30 percent drop in population. All three of these cities have remained nearly the same size geographically.

The city of Chicago has suffered roughly a 23 percent drop in population despite having increased in size by 20 square miles. Philadelphia has suffered a similar fate. The biggest city loser was Detroit with a 40+ percent drop in population.

Meanwhile here in our fair city, it is clear that we gained enormously in population, but that was because of an extremely aggressive annexation program which was effectively kneecapped in 1978 when Congress passed revisions to the 1965 Civil Rights Act whose intent was to prevent annexations by cities whose outcome would significantly dillute minoritiy voting strength. (Hint - do read your history and remember which party was in control of both Congress and the Presidency at that time).

Afterwards, the CoH's population has continued to grow, but the population growth has largely been because of a slow densification of our area and much less due to any new annexations. I can testify to this because I can remember stretches of Westheimer outside the 610 Loop in the 1980's which were not developed but were developed during the 1990's. The densification has gone from an average of about 3,000 to the 3,500 - 4,000 range over a period of 20+ years. Still, all of our northern, western, and southern suburbs are growing much faster than the the CoH itself.

The good news is that the broader urban areas of all of these cities have gained population. Some would resort to the usual explaination that people were fleeing cities because of crime, poor schools, etc, but as always when thinking about matters like these, I come back to Barton Smith's dictum. When the prices in an area rise to high enough levels, that becomes the market signal for an area to stop growing. Those prices can mean housing costs, general living costs, or transportation costs - which could be either the price of transportation or slowness thereof which would eat into people's time. In all, I would have to say that I do agree with the substance of Kotkin's analysis.

At 7:18 PM, February 15, 2007, Blogger Tory Gattis said...

Bill: Riverside is definitely part of the southern CA metro blob, but there's a pretty clear mountain range delineating it from LA. I'd say they're at least as well separated as Baltimore and DC.

Good stuff, Neal. Thanks.

At 9:38 PM, February 15, 2007, Anonymous Anonymous said...

Live in either Chicago or Boston, and you will see that, within city limits, things are about as dense as they can be for anyone's comfort. They used to be a little bit denser, but that was before most people started owning cars. If a city is crammed to begin with, and then everybody starts wanting a car, this pushes the "threshold of discomfort" down as far as density is concerned.

Yes, Chicago did in fact decline in population due to white flight (as did Philadelphia and Detroit). I can walk you through once-bustling middle class neighborhoods in Chicago that were entirely emptied out by gang terror, and today remain stigmatized. Some more familiarity with these places is needed before judgments can be made.

At 9:41 PM, February 15, 2007, Anonymous Anonymous said...

Tory - LA is full. It is denser than New York. That is why they are not growing quickly: because it is prohibitively expensive to do so.

At 11:14 PM, February 16, 2007, Anonymous Anonymous said...

A correction on my earlier post. It was the Voting Rights Act being extended to Texas and not necessarily the Civil Rights Act which was the root cause of the slowdown of Houston's annexations.



Post a Comment

<< Home