Glaeser, Again
- Posted by ryan on July 28th, 2008 filed in Cities, In the News
The Overhead Wire directs us to a Wall Street Journal interview with Ed Glaeser. I don’t really disagree with most of what he says, but I found the last bit kind of interesting:
There’s been a segment of urban developers who have been enthusiastic about the model in Europe for quite some time [because] it’s much more environmentally sensitive. But there are bad aspects as well.
While there certainly seem to be some smart things done in Europe, it’s a mistake to think they’ve got it right and we’ve got it wrong. There are many good things that came out of giving Americans the opportunity to live in big houses on the edge of urban areas.
If you think about the lifestyle of ordinary Americans living on the fringe of Houston or Dallas, for example, compared to what their lifestyle would be in an older European city — living in a walk-up apartment there compared to a 2,500-square-foot house here they bought for $130,000 with a 24-minute commute — it’s extraordinary in the low-cost areas of this country what a $60,000 family income gets you.
There’s a reason Atlanta, Dallas, Houston and Phoenix are our four fastest-growing areas. They offer an astonishingly high standard of living for ordinary Americans.
New York City is a great place to be really rich and not a terrible place to be really poor, but it’s a pretty hard place to live on $60,000 a year. You don’t experience anywhere near the basic standard of living you would in Houston on the same income.
Now, I obviously feel very strongly that it’s irresponsible to discuss the outward expansion of suburbs without talking about the unpriced externalities that helped make it possible–greenhouse gas emissions, in particular. And I think it would be nice to acknowledge the opportunity costs of the entire set of government policies that created our current urban landscape. Exurban growth has clearly been very good for some people. It has been very bad for others, however, and a rigorous analysis would attempt to see whether the numbers actually add up. No reasonable economist would say, “There are many good things that came out of giving Americans the opportunity to continue working in high-wage manufacturing plants thanks to the massive tariffs we levied.” That’s because we’ve been trained to focus on net gains from trade, and not just on the positives accruing to the folks intended to benefit from our policy choices.
Next, I have a real problem with him saying, “There’s a reason Atlanta, Dallas, Houston and Phoenix are our four fastest-growing areas. They offer an astonishingly high standard of living for ordinary Americans.” His research explicitly connects growth in these cities with growth in housing supply. Now I guess he means that cheap housing translates into higher real wages, but that’s not how this reads.
And that circles back to my earlier criticism of Glaeser. Rather than assuming a wide range of housing tastes and arguing that the numbers all demonstrate how many more people would love to live in New York and San Francisco but they can’t because of high prices, he frames the issue as one where the sunbelt auto-cities are in every way superior to alternatives. And that is absolutely, factually wrong. As anyone who regularly reads Glaeser’s academic papers can tell you.
July 28th, 2008 at 12:54 pm
Ed Glaeser: a 2,500-square-foot house here they bought for $130,000 with a 24-minute commute
I’m dying to know how many people in Atlanta, Dallas, Houston, and Phoenix pulled that hat trick - buying a house that big, that close to their jobs, for that cheap.
I’m sure Glaeser must have some stats to back that up.
July 28th, 2008 at 9:36 pm
Now, I obviously feel very strongly that it’s irresponsible to discuss the outward expansion of suburbs without talking about the unpriced externalities that helped make it possible–greenhouse gas emissions, in particular
According to the Brookings Institution (http://www.brookings.edu/reports/2008/~/media/Files/rc/papers/2008/05_carbon_footprint_sarzynski/tables.pdf), New York City metro residents produced 1.5 metric tonnes of carbon per capita in 2005. For comparison, Atlanta metro residents were at 2.7. So, for a family of four, the Atlanta family would be about 5 tonnes higher.
Now, social costs of carbon are naturally controversial and difficult to calculate, but crudely using Wikipedia (http://en.wikipedia.org/wiki/Carbon_tax), the highest calculated social cost is $350/tonne. Si that family, under an aggressively anti-carbon scenario, would pay $1400 more.
Cost-of-living calculations are likewise troublesome, but using CNN (http://cgi.money.cnn.com/tools/costofliving/costofliving.html), $60,000 in Atlanta is equivalent to $134,552 in New York City.
Now, I actually agree with a carbon tax, and there’s all sorts of caveats with this calculation (metro area vs. metro area, so doesn’t directly address city vs. suburb, for starters) but unless I’ve messed up the math here (far from impossible, I calculated this quickly), it’s very hard for me to see how pricing carbon would meaningfully impact the cost of living difference between New York and the rapidly-growing Southern and Southwestern cities.
July 28th, 2008 at 11:35 pm
The thing is, however, that it’s not about New York versus Atlanta with the carbon tax - rather, it’s about what form the new growth in Atlanta will take with that added cost.
July 29th, 2008 at 12:34 am
It’s most certainly about New York versus Atlanta. Glaeser explicitly cites those cities as examples of the kind of comparison he is talking about.
I don’t really understand Ryan’s second objection to Glaeser’s piece. And Ryan’s claim in his last paragraph is just silly. Where does Glaeser say that he thinks auto-cities are “in every way” superior to alternatives, or anything even close to that? Ryan’s attacking a strawman here.
As for the difference in greenhouse-gas externality costs, Glaeser addresses this issue in a paper I think Ryan has mentioned before, The Greenness of Cities. In that paper, Glaeser and Kahn calculate that the social cost from excess carbon emissions for an additional house in Houston is only about $500 greater than for an additional house in San Francisco. For Boston suburb vs. Boston city, it’s about $200. These numbers are far lower than the value of the benefits Glaeser estimates for suburban households vs urban households (lower housing costs, etc.), so they don’t really affect Glaeser’s basic argument much.
July 29th, 2008 at 9:01 am
When I say it’s not about New York versus Atlanta, I mean that the goal of a carbon tax wouldn’t be to equal the cost of living between those two places - the effect would be greater within the Atlanta metro area (and within the New York metro area) than it would shift the balance between them.
July 29th, 2008 at 8:11 pm
Having lived in SF, grown up in Houston, and now living north of Dallas, I can see what Glaeser means; “exurbs” in SF means, in effect, Castro Valley and now Tracy, both many miles from anywhere. That’s where the “affordable” homes are, by and for the average middle class income. Modesto and Bakersfield being one rung lower.
I would love to live in SF again, but buying a home in SF seems daunting to me, financially. People do it, though, even folks in my area of financial comfort (by this I mean $70-100,000 of SF wages).
How? You just figure it out. I don’t think it was a coincidence that, of my time in SF, nearly all my friends were from Jersey/NYC. They weren’t afraid of the cost of living; they knew a living could be made. Somebody has to do the work, and it ain’t Miami we’re talking about.
But, I lost my job, and within months was back in Texas. Less safety net, easier to scrape by. “Average” (and mediocre) by my estimation, although Austin, our last bastion of true affordable middle class near-urban comfort, is being turned into Little California as we speak.
July 30th, 2008 at 12:00 am
“low-tech cyclist Says:
July 28th, 2008 at 12:54 pm
Ed Glaeser: a 2,500-square-foot house here they bought for $130,000 with a 24-minute commute
I’m dying to know how many people in Atlanta, Dallas, Houston, and Phoenix pulled that hat trick - buying a house that big, that close to their jobs, for that cheap.
I’m sure Glaeser must have some stats to back that up.”
I know that folks in the Northeast and West have a difficult time believing it, but you really can buy a nice big house with a fairly short commute to work for $100-200K here. My house is 2K SF, 15 minutes from the heart of Dallas, cost $160K — nice neighborhood, new construction.
July 31st, 2008 at 7:00 pm
The livability of Houston (Phoenix etc.) over San Francisco is true at somewhat higher income levels too.
I have lived in San Francisco and currently live in Houston. My spouse and I were offered jobs in San Francisco that would have given us incomes in excess of $900,000. We looked closely and we couldn’t begin to live as well in San Francisco as we could on 25% less in Houston. We have a 4,000+ sq ft house on an acre with a pool, with great schools, within 25 minutes of downtown in a family value oriented town. To get even close to that in San Francisco (if you can find it) was way beyond any reasonable means. We turned the jobs down and continue to enjoy life in Houston.
August 12th, 2008 at 3:23 pm
Ryan says: “Now, I obviously feel very strongly that it’s irresponsible to discuss the outward expansion of suburbs without talking about the unpriced externalities that helped make it possible–greenhouse gas emissions, in particular.”
Sorry, but this argument is pointless, unless you consider ALL externalities, positive and negative. And it’s impossible.
Just one example: lower commute times may result, for most households, in more family time, or more leisure time, or more time for social relationships. How do you value this externality ?
Auto commutes, with no overcrowding, a good FM radio and AC, means more comfort, more relaxed times when you come back home: how do you value it ?
But it also means perhaps some more road mortality… But how much, since more car density on roads means lower average speeds and less fatalities ? How can we value that, positively or negatively… But taxpayer funded, unprofitable common transportation systems add another cost against compact cities. And so forth.
The number of factors you have to take in account to consider all + and - externalities is far beyond any human capability.
So the purely financial approach taken by glaeser, which is what really matters to households, makes pretty much sense here.
Considering only “some cherry picked negative externalities” is just a way to falsely demonstrate everything and its opposite.
August 12th, 2008 at 3:47 pm
Several things, Vincent. One, you appear not to understand externalities. If the costs and benefits of a decision are primarily internal, then that’s not an externality. A driver weighing how driving will affect his mood at home doesn’t fit the bill.
Secondly, if shorter commute times are the priority, that doesn’t imply anything about which modes we should finance and how. Magically assuming that roads generate short commutes is not analytically sound.
And finally, “driving is enjoyable” is a subjective judgment. “Increasing gas taxes by x amount will lead to y decline in greenhouse gas emissions” is not subjective. “X toll will result in freely flowing traffic where X-1 toll will not” is not subjective. These are efficiency improving measures, based on social costs we can define and assess.
August 7th, 2009 at 4:32 pm
vincent Says:
This is an obvious false dichotomy … either take each and every single social cost and benefit into account, or just take into account the costs and benefits to the individual under the current system of taxes and subsidies.
And not considering any external costs and benefits at all is just a way to say “the way things are is the way things ought to be”.
For example, consider the subset of what are clear financial costs, which clearly ought to be included by Glaeser and are not, of the direct public subsidies and cross-subsidies. The fact that large numbers of people take subsidized type of good “a” is not a demonstration of a preference for that good given its cost, its a demonstration of a preference for that good given the portion of the cost that the purchaser pays for.
Indeed, in a question such as the revealed preference for suburban housing, where there are direct development subsidies for greenfield development, a capital gains tax structure biased toward rolling over windfall development gains into greenfield development, cross-subsidies from more efficient per hook-up established and infill development to less efficient per hookup greenfield sprawl development, a wide range of auto commute transport system operating subsidies and cross-subsidies from urban to suburban road building and maintenance, the expressed preference under what is a substantial net subsidy quite clearly reveals a smaller underlying preference.