Austin Contrarian

Randal O'Toole reviews The Gated City

Here:

Didn’t we read this book already? Only last time it was called The Triumph of the City and it was written by Edward Glaeser, a fine economist whose view of cities is slightly shaded by the fact that he grew up in Manhattan.  Avent’s bio doesn’t say whether he grew up in New York [AC: Raleigh, NC], but he did go to school in London and currently lives in Washington.  I can’t help but feel he is another city boy who loves cities so much he wants to impose them on everyone else.

The Antiplanner would never step into another city again if I didn’t have to as a part of my work.  But I don’t think everyone would be better off living my lifestyle; I just think people ought to have a choice.  Avent’s book simply stimulates the smart-growth advocates who want to impose all sorts of policies on urbanites and suburbanites to force them to live in denser communities.

I infer from this that (i) O'Toole didn't read the book; (ii) he didn't read the book carefully, or (iii) he did read the book carefully but is misrepresenting the conclusion.  Ryan does not advocate any coercive policies in The Gated City.  He asks that the housing market be allowed to function like it's supposed to:   the market should be allowed to supply more housing when and where there is demand for more housing which means -- gasp -- places should be allowed to grow denser when there is demand for it.  It's sad that it takes an economics-laden, research-intensive book to justify this principle, and even sadder that so many self-described libertarians denounce this principle, but such is the state of land-use policy in the United States today.

Like O'Toole, "I just think people ought to have a choice."  People do have different preferences.  Some people like dense, bustling, noisy places.  And some people -- including, by his own admission, O'Toole -- hate them.  The problem is that the people who hate cities try to impose their preference on the places where cities want to be. 

After all, only a small portion of even the densest metropolitan area is genuinely urban.  Most "city" dwellers actually live in places that are relatively sparesely populated.  This would still be true even if the law allowed the demand for urban places to be fully satiated, even in places like San Francisco and San Jose.  People who hate dense environments have and will always have thousands of suburbs and small towns to choose from.  If you don't like dense places, don't live in dense or densifying places. 

The libertarian position accordingly ought to be that we should allow the market to provide density where and when people want it; the people who don't like that particular product can buy a different product.  That occasionally might require a density-hating libertarian to accept change to his own neighborhood.  It occasionally might require him to move to another neighborhood if he finds the change too uncongenial.  But libertarians are supposed to embrace dynamic markets.  They deride people who fight market-driven change as "statists."   Don't preach creative destruction for me but not for thee.

The view that we should accommodate density ought to be the progressive position, too.  Allowing housing to keep up with demand keeps home prices low, which certainly is more equitable and egalitarian than propping them up.  Allowing housing to keep up with demand lets people move freely from low-opportunity cities to high-opportunity cities.  Allowing housing to keep up with demand facilitates the density necessary to support transit, preserves open space on the city's periphery, and probably reduces VMT and the other things that environmentalists should care about.  Fighting to preserve a neighborhood just because you happen to live there and like it the way it is is not only statist, it's selfish, even if you can't identify the specific people who will be hurt.

I've always thought that the libertarian position and the progressive position on land-use policy ought to be approximately the same.  Thus it is not surprising per se that so many "libertarians" and "progressives" agree on land-use policy.  What is strange is that so many "libertarians" and "progressives" agree because the libertarians turn into command-and-control statists  and the progressives into regressive elitists.  

Land prices vs. entitlement prices

Matt Yglesias quibbles with The Gated City because "it’s a perpetuation of the American habit of mixing up discussions of the price of houses with discussion of the price of land":

Land is a speculative commodity, comparable to a bond or a stock or oil futures. You might lose money buying Texas Hill Country land and you might make money. The house is more like a bulldozer or a camel. It’s expensive, and its potential resale value is a relevant issue when you’re considering a purchase, but it’s not an investment. Five years from now, your house, like your bulldozer, is going to be older, more broken, and cheaper than it is today. It’s the land the house/bulldozer is on that might be more expensive. That might be because it becomes more desirable as a place to live, or because climactic shifts make the lend better-suited for high-value agriculture, or because oil is found on the property, or because it acquires a better road connection, whatever. There are lots of perfectly ordinary reasons for land to go up or down in price. A house, by contrast, is a large decaying physical object.

This is a reasonable point but I don't know what it has to do with Ryan's thesis.   The price of a piece of land actually has three components:  (1) the value of the physical fixtures, (2) the price of the land, and (3)  the value of the things you can legally do with the land.

You might think that (2) and (3) are the same but they are not.  Suppose we calculated the value of all land in Los Angeles by subtracting the value of all structures and improvements from the market value of the land.  The number we got, of course, partly would reflect what the city lets people do with the land.  Now suppose we calculated what the value would be if the City let people build as densely as they wanted.  Those numbers would not be the same.  The difference would be due to restrictions on development.  (The first number would not necessarily be higher than the second for the same reason that the total value of the output of a monopolized market might be smaller than the total value of the output of a perfectly competitive market.) 

This raw calculation is intractable so let's take a more tractable example.   A 47-story condo tower uses just as much land as a 46-story condo tower.  If you could build as tall as you wanted, and held the price of land constant, then the price of the condos on the 47th floor should equal their cost to build.  But Glaeser and Gyourko showed that in Manhattan in 2004, the price of the marginal condo fetched as much as a $300 psf premium over the cost of construction.  That premium reflected the value of the entitlement, not the value of the land.   

Obviously, the value of the right to build can rise or fall with demand, just as the value of land can rise or fall with demand.  So why draw a distinction between the value of land and the value of entitlements?  Because the supply of land is fixed; the supply of entitlements is not.  In an efficient, equitable world, a city would set the price of the entitlement to build one housing unit equal to the social cost of that housing unit.  In high-value, high demand places, where additional people produce net positive returns, the value of the entitlement should be zero or close to it.  Instead, it is worth hundreds of thousands of dollars.  That inflated price produces dead-weight losses for the residents of the city and the country as a whole, and that's what Ryan, Glaeser and a host of others are complaining about. 

I'm pretty sure Matt understands this, but his post unfortunately will lead to a perpetuation of the American habit of mixing up discussions of the price of land with the price of entitlements.

Trillion-dollar bills

Michael Clemens, in Economics and Emigration:  Trillion-Dollar Bills on the Sidewalk?:

What is the greatest single class of distortions in the global economy? One contender for this title is the tightly binding constraints on emigration from poor countries.  Vast numbers of people in low-income countries want to emigrate from those countries but cannot.  How large are the economic losses caused by barriers to emigration?  Research on this question has been distinguished by its rarity and obscurity, but the few estimates we have should make economists' jaws hit their desks.  The gains to eliminating migration barriers amount to large fractions of world GDP—one or two orders of magnitude larger than the gains from dropping all remaining restrictions on international flows of goods and capital.  When it comes to policies that restrict emigration, there appear to be trillion-dollar bills on the sidewalk.

The same logic holds for interstate migration, by the way.  The United States would be a much richer country if more people lived in San Jose and Manhattan and fewer people lived in Mississippi and West Virginia.   While there are no laws barring someone from Ohio from moving to D.C.  to look for work after college, cities can and do restrict interstate immigration.  They can't issue visas but they control the supply of new housing and there's really not much incentive to move to a city without a place to live.  Since little new housing is built in Manhattan, San Francisco and San Jose, would-be immigrants must outbid  one another for phantom visas.  That's a good result, I suppose, if you own property in Manhattan or San Jose, but it's not good for the immigrants and it's not good for the country as a whole.  I don't think it's good for Manhattan or San Jose, either.  Texas, despite its parochial reputation, is far more willing to accept newcomers than, say, San Jose, and I agree that Texas's strong economic growth has been the product of its population boom, rather than vice versa.

NIMBYism and affordable housing

The federal government underwrites affordable housing in part by providing tax credits to developers who build affordable housing.  The Texas Department of Housing and Community Affairs parcels out these tax credits in Texas.  A lot rides on TDHCA's decisions and competition for the tax credits is fierce.

This Statesman article on TDHCA's scoring system is interesting because it explains how NIMBYism is baked into TDHCA's evaluations.  Although projects are scored on several criteria, local opposition (or support) carries a lot of weight.  The scoring system effectively counts local reaction twice:  it weights letters from neighborhood associations and it weights letters from the legislators who represent these neighborhoods, who, naturally, can be counted upon to do their constituents' bidding.  There is a 52-point swing between unqualified opposition and unqualified support, a big deal given the maximum score is 226.   (Although I am calling this "TDHCA's system," in fairness I should note that it is merely implementing statutory standards.)

It's worth stressing that these are not zoning cases.  Tax-credit subsidized housing can only be built where zoning allows it.  Developers, in other words, merely propose multi-family projects for parcels of land that the city has already determined are suitable for multi-family housing.  Concerns about traffic, noise, overcrowding, etc. should have been resolved when the parcel was zoned.

Weighting the neighborhood's approval and the approval of its proxy is of course bad policy.  Homeowners are notoriously risk averse.  Their risk aversion causes them sometimes to oppose a project that has a remote chance of depressing property values even though it has a better chance of raising property values.  The premise of the tax-credit program is that cities need more affordable housing.   Politicians like to talk about affordable housing in the abstract but ultimately it has to be built somewhere, which means that neighborhood opposition has to be discounted rather than emphasized.  Unless, that is, we are happy to relegate it to neighborhoods that are already full of affordable housing, which is the effect of current policy. 

I also want to point out that the TDHCA's system institutionalizes a sort of "ward courtesy" at the state level.  The Texas Legislature attempted to reform that system this session, but Perry vetoed it for murky reasons.  (Texas Housers has a detailed account.)  The proposed reform would have entrusted the decision to support or oppose the project to the local elected body rather than a single legislator, probably in the sound belief that councilmembers are more likely to make reasonable decision in the aggregate than individually.

Micro homes aren't crazy

Randal O'Toole often says the "right" thing about inner-city development.  He says he would abolish parking minimums of any sort and relax inner-city anti-density zoning; I would do the same.  And asserting that most American families with children prefer single-family homes to small apartments does not make one an anti-city bigot.

But there is a difference between being a market advocate and being a dogmatic shill for suburbanism.  O'Toole constantly arms those who claim he's the latter.  Exhibit number 731:  this piece ridiculing the plan by a homebuilder -- D.R. Horton, a national, production homebuilder -- to build "micro-homes" in Portland.   These "micro" homes are single-family homes that range from 364 to 687 squae feet.  According to O'Toole, the demand for these tiny houses is a pernicious consequence of Portland's growth boundary.   And while he doesn't "wish ill to D.R. Horton,"

any Portlandia residents crazy enough to buy Horton’s micro-homes are fooling themselves if they think they are somehow morally superior to people elsewhere who can buy much nicer homes for far less money. Oregon’s elite has created an artificial land shortage and is using that shortage to stick it to young people and other newcomers to the region. That’s plain disgusting.

I'm not sure why he thinks micro houses are absurd.  I imagine these houses are about the same size as the typical apartment but have more privacy and less noise.  They might attract people who'd otherwise be put off by dense city living.

Perhaps a bigger point is that this type of development is exactly what the plain vanilla model of urban land prices predicts, even in an unfettered land market.  Land prices near the city core are high, at least if it's a desirable place, because the cost of getting there is low.  Land prices in the suburbs are low, because the cost of getting to the core is high.  High land prices cause central property owners to make do with less land and, inevitably, smaller homes.  Low land prices cause suburban property owners to take more land and build bigger dwellings on them.

Sometimes demand for living near the city center rises.  Transportation costs might rise, making proximity to the center a relatively better deal.   Or the city might improve the amenities or reduce the crime in the city center, making it relatively more attractive.  Or property owners' preferences might shift to a more urban habitat.  Or -- and this is O'Toole's implicit claim -- the city might choose to raise the cost of suburban development, either by limiting the supply of land or adopting other anti-growth policies.  Any of these will cause denser infill development with smaller homes and less suburban development.  

Which of these explains Portland's rising central-city prices is an empirical question.   I don't presume to know the answer.  Pointing to suburban growth boundaries is not an absurd hypothesis, but it's not obviously true, either -- central city prices have been rising in lots of places without suburban growth boundaries (including Austin).  O'Toole's bald assertion that growth boundaries are to blame isn't compelling evidence.

But O'Toole isn't just making bald empirical claims.  He's dismissing a hard-nosed, for-profit homebuilder's proposal to supply exactly the kind of housing that the economic model predicts.  And he is ridiculing as "crazy" buyers who also would behave as the models predict.  This not the market position; it's just the opposite.  And it is the just the kind of shilling which draws so much criticism, including from market sympathizers like me.  

Shared-lane rail won't manage congestion

David Parvo at the Placemaking Institute defends the City's plan to run streetcars in shared lanes through downtown Austin.  He argues streetcars are an economic development tool and a tool to "help expedite a much needed generational shift that will better manage congestion in the long-term."  Read the original, but his response to Mike Dahmus's criticism that streetcars will only worsen congestion itself deserves a response:

[O]ne hallmark of a successful city is congestion; the more successful a city becomes, the more congested it will be.  And, thus, congestion can never be solved but only managed.

Those who try to solve congestion are being "utopian."

This is wrong.  The hallmark of a successful city is crowds of people going places; congestion is the hallmark of a city that can't manage those crowds of people.  Congestion is equivalent to a tax on going where you want to go.  It's a tax paid in lost time rather than money but it has the same effect:  it discourages trips at the margin.  It manages crowds only by shrinking them. 

Running a shared-lane streetcar through downtown Austin will not manage congestion, unless by "manage congestion" one means "make it worse."  Shared-lane rail will be slower than buses.  A lot of people will ride at first, but in the long run they won't settle for a slower, less convenient form of travel.  Shared-lane streetcars won't expand the flow of people through and within Downtown but will merely reduce the throughput of our roads and buses.  They will make it more expensive -- measured in lost time -- to get downtown, or to get from UT to South Congress, or to get from Caesar Chavez to 12th Street.  They will hurt Downtown in the long run.  

Reserved guideway rail on the other hand will increase the carrying capacity of our transportation network.  People will be able to get around Downtown or from UT to South Congress more conveniently and (I hope) more quickly than by using cars or buses.  Reserved guideway rail will not eliminate congestion for cars and buses.  I'm no utopian.  But it will give people a congestion-free alternative for getting around.  And that alternative will become more valuable as central Austin grows denser.  Shared-lane streetcars will be progressively less valuable as the streets grow more congested.  

Let me add that I know Dave and some of the guys at the Alliance for Public Transportation and in DAA and DANA and CNU who support the in-street rail plan.  These guys share my preferences and goal for Downtown.  We disagree on the method.  As Dave notes, Glenn Gadbois & Co. do deserve the credit for getting rail back on the agenda.  But they should not settle for half measures that will make things worse.  Go all in.   

NB.  Dahmus links this in the comments.  Skip mine and read that instead.  (Should I have put this suggestion at the beginning rather than the end?)

Urban planners should learn urban economics

Stephen Smith writes about a Manhattan residential project in which 40% of the units will be set aside for affordable housing, likely due to political pressure.  He asks "How much affordable housing is enough?" and then explains the unintended consequences of large affordable housing mandates:

Obviously, I oppose setting aside this much of new developments for affordable housing. People tend to think of different segments of the real estate market as distinct – how on earth could limiting the number of rich people on Far West Side make prices rise in Bed-Stuy? – but they are inextricably linked. You might not shed a tear for the 20-something banker who won’t get his Hudson River view, but when he then decides to buy in, say, Brooklyn Heights, the higher housing prices are going to have a ripple effect throughout the borough, pushing the envelope of gentrification further and further, until finally the knock-on effects reach places that the Far West Side 2o-something banker who got priced out of the Far West Side would never dream of living.

He is describing filtering.  If you don't understand filtering, you don't understand the dynamics of housing markets.  But my impression is that few urban planners understand this principle.  This is a symptom of a broader problem.  You will never catch me making over-generalizations on this blog, so I can safely say that most urban planners know nothing about urban economics.

O.K.  Donald Shoup is an urban planner and has done first-rate work on the economics of parking.  (But then his Ph.D. is in economics.  Does he count as a counter-example?)  Shoup's ideas have percolated down to rank-and-file planners.  Planners who work at the transportation end of the planning spectrum seem to know a lot of economics.  You probably can't get out of a top-notch school like M.I.T. without learning a bunch of economics.   

But the average city planner can get by without learning much, if any, urban economics.  For example, here is the curriculum for the University of Texas's masters degree in community and regional planning.   The core curriculum does not appear to include any urban economics.  (UT grads, correct me if I'm wrong.)  UT planners can satisfy their elective requirements with courses like "growth management" and "deep democracy" and "historic preservation practice."  The program does not even offer a course in urban economics, although a couple of the electives might touch on the basics.

UT graduates are not required to learn about the bid-rent gradient, filtering, Tiebout sorting or read Krugman's Development, Geography and Economic Theory.  I'm sure they commit The Life and Death of Great American Cities to memory (and it is a great book), but are they required to read Jacobs' The Economy of Cities?

Some practicing city planners do learn these things, I know.  But all planners should be made to.  That planners aren't required to learn it signals students that urban economics really isn't very important.  People shouldn't be allowed to meddle with housing markets without knowing this stuff.

Chart o' the day

Prepared by Ryan Robinson and presented here without comment:

Ryan has taken his act on the road, presenting his demographic analysis at a number of (reportedly packed) public forums.  Here is a nice summary of his presentation by Jackie Stone.

Rings of population change

From Stephen Von Worley at Data Pointed (via Flowing Data), maps of block-by-block population changes from 2000 and 2010 for a bunch of American cities.  They show that growth continues to occur mostly at the fringe of cities, which is hardly surprising.  The changes in central city populations are more interesting because you can tell at a glance which central cities are healthy and which are not.

"[D]eep blue indicates that the population doubled (or more), pure red means that everyone left, grey denotes no change, and the intermediate tones represent the spectrum of increases and decreases in-between."

Here is Austin:

You can compare Ryan Robinson's map of the population changes by census tract. 

It would be wise to exercise a little caution with these maps.  Census blocks frequently change from one census to the next, which makes comparisons tricky.  I'm not sure how the author addressed this problem, if he addressed it at all.  [Update:  Steve emailed me to say that he "used the block data for each Census to compute the gridded population for that year, then took the difference.  Nice work!]

Here is what a 25% loss in population (Detroit) looks like:

A cool graph of city densities

I want to point to this wonderful graph of the density of eight American cities by Fedor Manin at We Alone on Earth.  I'm referring specifically to the top graph.  (I have a problem with the bottom one.)

Manin computed the density of each census block group in each city's urbanized area.   He then calculated the percentage of the population living at at least the density given on the x-axis.  The graph gives a snapshot of each city's weighted density:  if city A's weighted density is greater than city B's, then city A's curve mostly lies to the right of city B's. 

 

Note the graph is logarithmic.  

Manin observes that it is strange that the density graphs for Boston and New York are straight lines.  I think it's strange, too.  Because the graph is logarithmic, the straight lines mean the densities of Boston and New York obey (approximately) a power law, while the density functions of the typical city are (approximately) linear. 

The bottom graph also gives a visual snapshot of weighted density.  It depicts the number of people living within 5% of the given density. 

It is perfectly reasonable to smooth the curve using a range like this.  But the graph is distorted because the range increases with density.    E.g., the range for 10³/km² is 950-1,050 persons/km², while the range for 10⁵/km² is 95,00-105,000 persons/km².   The range for 10⁵/km² is thus 100 larger than the range for 10³/km².   This distorts the graph, biasing it in favor of high densities.

Manin should have taken the average density over each range; i.e., he should have divided the number of people living within 5% of the given density by the length of the interval.   

H/t Market Urbanism and Human Transit.