The Urbanist Musings of Pete Saunders

The Orthodoxy of Supply Side Urbanism: Wrong

Many economists and urbanists are coming together to arrive at a consensus that our nation’s affordable housing crisis in cities is due in large part to an artificial reduction of supply from land use restrictions (aka exclusionary zoning).  These supply side urbanists (as I call them) are — partly — wrong.

Just this week, Paul Krugman, Princeton economist, New York Times opinion columnist, Nobel Prize winner and a famously liberal opinion maker, chose team supply side in the affordable housing debate.  Here, he talks about the inflection point that New York and other cities passed some 15 years ago, when affluent whites formed a critical mass in the Big Apple and impacted housing affordability.  He worries about the impact on the rest of the city’s residents:

“(W)hat about all the people, surely a large majority, who are being priced out of America’s urban revival? Does it have to be that way?

The answer, surely, is no, at least not to the extent we’re seeing now. Rising demand for urban living by the elite could be met largely by increasing supply. There’s still room to build, even in New York, especially upward. Yet while there is something of a building boom in the city, it’s far smaller than the soaring prices warrant, mainly because land use restrictions are in the way.”

Conservative urbanist and demographer Wendell Cox has touted this point of view for some time, and in this piece seems to relish the developing consensus on the matter, as the Chair of the White House Council of Economic Advisors arrives at the same conclusion:

(White House economist Jason) Furman starts with the fundamentals: “Basic economic theory predicts—and many empirical studies confirm—that housing markets in which supply cannot keep up with demand will see housing prices rise.”

Furman cites research by Christopher Mayer of the University of Pennsylvania and C. Tsuriel Somerville of the University of British Columbia who “conclude that land use regulation and levels of new housing construction are inversely correlated, with the ability of housing supply to expand to meet greater demand being much lower in the most heavily regulated metro areas.”

Others arriving at this position include Harvard economist Edward Glaeser, one of the earliest developers of the idea, and economist and writer Matt Yglesias. 

When great minds, across the political spectrum, start to agree on something, they must be right, right?

I’ve always questioned orthodoxy, and even as the best minds in urbanism and economics converge on a supply solution to addressing housing affordability, I remain skeptical.  Why?  Mostly because there are multiple housing markets within cities, even more within metro areas, and a potential solution for one housing market segment may have disastrous impacts on others.  

This is a matter of demand.

Relaxing zoning will help the cities where the demand for housing is the greatest.  Relaxing zoning in suburbs will add units to places that have artificially suppressed supply, particularly within metro areas that have a strong housing demand.  However, I fail to see how it helps cities and metros that don’t have the red-hot housing demand of New York, or the Bay Area, or other areas.  

Relaxing zoning in cities or metro areas with a lower housing demand, which is the case for much of the Rust Belt and many low density Sun Belt cities, in my opinion would lead to a concentration of new housing development in the most in-demand neighborhoods within them, at the exclusion of other neighborhoods and suburbs.  It would serve the affluent, it would be clustered, and would potentially decrease affordability and further increase inequality.

brought this point up in a post last August, when I looked at rates of residential density and housing vacancy for the 25 largest cities in the U.S.  Using 2010 Census data, I saw that I was able to form quadrants of density and vacancy for the 25 cities.  The quadrants look like this:

In 2010, the density/vacancy averages for the 25 largest cities was 6,269 persons per square mile, and 11.2% vacant units within them.  If you plot the cities in a graph, it looks like this:

As one would expect, New York, San Francisco, Boston, Los Angeles and Seattle, which could be called America’s gentrification pioneers, are firmly in the high density/low vacancy quadrant.  Even after the fallout resulting from the Great Recession, each had housing vacancy rates under 10 percent.  I’d imagine in the five years since the numbers have dropped even more.  Relatedly, Austin and San Jose are in the low density/low vacancy quadrant.  They’re joined by cities like San Diego, Denver, Charlotte and Nashville.

If we use housing vacancy rates as an indicator, these are the in-demand cities, and increasing supply is an appropriate response.  In cities that are already high density, doing so may be problematic as residents reach their limits on what is possible and acceptable in their cities.  In the low density cities, one could reason that there is space to add the needed units.  

But let’s look at the group of cities that are in the high housing vacancy category, making them presumably low demand cities.  Chicago, Philadelphia and Washington, DC fall into the high density/high vacancy camp, while Houston, Dallas, Memphis, Indianapolis, Jacksonville and Detroit are among those in the low density/high vacancy camp.  One could argue that the Chicago-Philly-DC arrangement could be a best case scenario for the low density cities like Houston and Indianapolis, but anyone who knows anything about how gentrification has proceeded in our nation’s large cities would know that issues of bifurcation, where not all residents of the city are recipients of the benefits of emerging affluence, are probably strongest in those three.

These cities are the low demand cities and increasing supply might create some complications.  Most, if not all, of the cities have some type of gentrification activity taking place already.  Any relaxation of zoning regulation in these cities would be a signal to developers to build new housing for those at the high end of the income spectrum, and at the exclusion of the middle or working classes.  In fact, I envision the growth of affluent enclaves within cities, should zoning be relaxed.  Little construction would occur outside of the enclave, and there would be little incentive for public investment there as well — leading to an accelerated downward cycle for already troubled neighborhoods.

Cities must address their demand issues before acting to address supply.  In my opinion, there are many young and affluent potential city residents who are attracted to a small sliver of a given city, and neglect other parts of the city that have many of the characteristics they seek.  However, their laser focus on the most in-demand neighborhoods artificially raises prices within those hotspots, and leads to the erroneous conclusion that housing is more scarce, and therefore more costly, than it really is.  High vacancy rates in low demand cities suggest this is the case.

In all, supply side urbanists and economists must not neglect the demand side of the equation.  A more nuanced approach should be applied to land use regulation, both within and beyond city boundaries.

11 Responses to “The Orthodoxy of Supply Side Urbanism: Wrong”

  1. Anonymous

    Having lived in Chicago, NYC, and now Albany (which firmly occupies the low demand/high vacancy quadrant), I very much see your point about the variability inherent in housing markets. I guess my challenge is this: a lot of my Chicagoland peers (I went to a relatively wealthy suburban private high school, but my family lives in the city) will move to the city, but *only to one or two neighborhoods*. I'm sure you and I agree that that's far from ideal, and firmly rooted in racism, classism, etc. BUT can restricting supply in Lincoln Park, Wicker Park, and Lakeview force people to explore other neighborhoods that could use their investment? I'm not so sure it works that way. I'd like it to! But I fear people will take their toys and their fancy coffees and go right back to the suburbs. And if I have a choice between keeping them in the city, where maybe they will eventually tread carefully into exploring other neighborhoods, and letting them migrate back to the 'burbs, I will always choose the former. And that means expanding supply in high-demand neighborhoods so they maintain some measure of integration and affordability.


  2. SFB

    Paul Krugman is no longer at Princeton University. He is at CUNY now. As for the rest of your piece, people either complain about disinvestment in low-income areas, or they complain about gentrification. Probably best if we just lay off the arbitrary housing restrictions and let people move to where they want to live, to the greatest extent possible.


  3. Alon

    The problem with dividing the US into quadrants is that it treats Seattle and New York as equally dense. It also treats New York's single-family neighborhoods (Gravesend, etc.) as equivalent to the Upper East Side. The reality is that no city neighborhood in the developed world is at the limit of density today. Some areas built around towers in parks are hard to densify, but these aren't the neighborhoods in question. The hot US neighborhoods have single-family houses and low-rise apartment buildings, and a handful of high-rises; even the eastern part of the Upper East Side, which is already the cheapest neighborhood in Manhattan south of Harlem, is full of low-rises. Low-rises are replaceable. Knock down a 6-story building, put up a 50-story building, and there, you have more density.Now, is this acceptable to existing residents? No, but not because it's higher density; it's because it's more people who are not those residents' children. The residents, as an empowered political group, are selfish nativists who care more about preservation of their political power than about other people's affordable housing. Few pay market rates – they either own or are rent-controlled for life. Why do they care about reducing market-rate rents? If anything, the owners want higher market-rate rents, and the rent-controlled do not want to be politically diluted by new residents.


  4. arcady

    Another problem is when places move from one quadrant to another. Somerville, MA was a relatively poor area before it started getting gentrified, and along with other urban neighborhoods around Boston had a period of decline from the 50s through the 80s, about the same time that the current zoning codes were put in place. And even assuming they were doing an okay job of managing that decline, they became grossly inadequate when decline turned into growth.


  5. Phineas

    Glaeser and Krugman are making their arguments about the places with high demand, that may have relatively higher density, but not nearly as much as if density restrictions were relaxed. This blog seems to miss the issue that's being contested. It states, \”In cities that are already high density, doing so may be problematic as residents reach their limits on what is possible and acceptable in their cities.\” The whole question is whether it would really be problematic and whether people should shift their notions of acceptably dense development.


  6. Pete Saunders

    I think you miss the point of my post here. I know Glaeser, Krug man and others are addressing the needs of cities where demand is high and the affordability crisis is acute. Beyond that I think they are absolutely sparking the right debate for such cities and metros. Reducing restrictions can add units and address affordability.But is the response for New York or the Bay Area or DC or Seattle necessarily the one for Cleveland or Buffalo? Where demand is stagnant? It's often presented as a one-size-fits-all solution, and I see it actually hurting more cities – maybe not more people, but more places – than it helps.Here's another reason why. Gentrification or revitalization starts as a rehab activity. Pioneers find an \”undiscovered\” neighborhood and renovate a home. More do the same. After some time demand grows for more units and pressure grows for new construction. That's where many coastal cities are today.But many Rust Belt cities are in the early stages of the process. There is actual rehab and rebirth happening, even in my hometown of Detroit. But relaxing zoning could have the impact of stalling the outward spread of that process to other neighborhoods that could benefit from reinvestment. The incentive to expand the revitalization footprint is lost.I'm all for the right solutions for the right cities. But we can't impose one solution on them all.


  7. simval84

    I don't get the point. So if we relaxed zoning, most new housing would concentrate in neighborhoods that are in high-demand and leave low-demand neighborhoods… and the bad thing is? If there is a place close to rapid transit, with plenty of stores within walking distance and good urban design, how is it not a good thing for new housing to be concentrated there? Why should we wish to force people to live in crappy neighborhoods instead?The idea that if developers are allowed more freedom, they would build only housing in rich areas strikes me as wrong. For one thing, it ignores filtering, providing new higher density housing to the rich gets the rich to stop going after older housing, which filters down to the middle-class. An example of this is Tokyo, I once collected data about condo prices in Shibuya (a local, very high demand central neighborhood). New condos' prices were no smaller than 1 000$ per square foot, however, condos built in the 80s were going for 600$ per square foot.Oh, and the fact that Japanese zoning is strongly pro-development doesn't prevent developers from still building affordable housing in suburbs.Allowing higher densities in less desirable neighborhoods may also help revigorate limited areas with higher desirability. For example, a commercial street can have some desirability even in an undesirable neighborhood, attracting some new developments which will revigorate the street and start a renewal process from these seeds of desirability.Of course, some areas may be completely abandoned, so what? The low-density character of most American cities and suburbs means that not all cities can be upgraded to walkable urbanism. In fact, if we were to achieve the objective of moving everyone into a walkable neighborhood, it's likely that 75% or more of current residential areas would effectively become abandoned, because the population density requirement of walkable urbanism is much higher than the average population density of North American settlements.


  8. Kevin Klinkenberg

    I think there's some interesting nuance here, Pete, and some areas of potential agreement. I'm very much in favor of not only loosening up but blowing up a lot of our permitting processes for urbanism. Most are just terrible, and about 100 light years beyond what was intended when the city planning movement was birthed. In fact, most of our processes were conceived as part and parcel of the suburbanization of the country. They completely ignore how and why the older cities we love were built in the first place. So, yeah, I'm certainly in the camp that much of that needs to go away or be completely rethought, which sounds like I'm a supply-side guy.But… I'm also one that believes that cities dominated by tall towers are not in fact good for humans (or good for cities generally) and I prefer caps on height of 6-8 stories. That's really the traditional urban model, and includes pretty much every city in the world that is beloved. I know others disagree and love their towers, but in my opinion towers should be an exception, not a blanket allowance. So that all said, it seems there a couple of routes to take the supply-side thinking: 1) just let it all go, and let the market do what it wants, where it wants. I suspect much of your analysis is correct, that desirable places will continue to get more desirable and more dense, and that outward pressure to other neighborhoods will only depend on the strength of the overall region's economy. 2) Relax the rules dramatically, but have some sort of effective \”cap\” by neighborhood, so that it does in fact encourage more outward push for redevelopment. I should note that effectively any historic district with teeth to it already does this – they fundamentally limit the density and market by not allowing tear/downs and keeping development to a certain scale. Rightly or wrongly, that's a reality.


  9. simval84

    I don't think there is much reason to fear skyscrapers, they just don't make a lot of economic sense in general. Skyscrapers are very expensive to build per square foot, and they are not to everyone's taste to live in. However, I think they have their place in business districts, to keep offices as close to rapid transit stations or transit hubs (like bus terminals) as possible. Even Paris has been forced to build tons of them in La Défense to take some pressure off of the historical center. I do not agree at all with blanket height limits, these strike me as completely arbitrary and basically imposing aesthetics on everyone else. That being said, there is a compromise, which is a slant plane limit that limits the height of buildings based on the width of the road and the setback. This allows buildings to be tall while protecting access to the sun and the pedestrian perspective.That being said, there is a problem with a system without much regulation: the risk of land speculation. The value of land depends on the profit that is expected from developing it. If you don't limit density at all, that creates uncertainty wiuth regards to the value of the land, the lot owner will thus likely require land value depending on optimistic expectations of development. For example, if a skyscraper is built in a neighborhood, all land owners may start pricing their land as if a skyscraper was going to be built there, making land too expensive for less dense options.There are two ways of dealing with this issue:1- Limit density (best to go with Floor Area Ratio if that is the desired goal)2- Make holding land or small, decrepit building more expensive through a land value tax.If you go the first route, then it would be best to have a dynamic density limit, based not on an arbitrary number set in by a planner, but by a multiple of the average existing density, and maybe a FAR transfer system, to get incremental development going.


  10. Alan

    I'm from Milwaukee and I live in Brooklyn now after going to school in Pittsburgh— three fallen manufacturing centers at differing points along the housing-demand spectrum.Prices in rust belt centers like Milwaukee generally seem low enough that it's unlikely there would be much new construction even if zoning were wholly repealed— witness the fact that much of land at the edge of Milwaukee's downtown still sits vacant nearly 15 years after Norquist demolished the Park East freeway.Pittsburgh is more of a middle ground, and is seeing a healthy level of new investment after a great many years of decline. But many there don't necessarily seem to have a desire for gentrification to be spread widely throughout the city— you can see a great deal of lamentation of the gentrification in the East Liberty neighborhood, even as there are full city blocks of vacant land within a short walk.


  11. D Holmes

    I think we have a middle class/working class wage stagnation \”crisis,\” and an affordable housing \”challenge\” for a handful of US global elite cities. Anyone that can't afford to live in one of the most desirable neighborhoods, should perhaps move to a suburb, downsize, or move to a metro area that offers the best standard of living for the wage they can earn. Crisis solved. Even better if the businesses that can't afford to pay their workers enough for them to live within reasonable commuting distance of their place of business, move to a lower cost metro area.I agree with your key point which is that relaxed zoning is a possible partial solution to a challenge occurring in a handful of US cities, but is presented as a prescription for a wide range of housing issues in other cities. My sense is that too many of the \”leading\” urbanists live in that handful of cities with an affordable housing \”challenge\” and are somewhat myopic in their focus relative to the challenges facing most cities. Same with the gentrification debates.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Basic HTML is allowed. Your email address will not be published.

Subscribe to this comment feed via RSS

%d bloggers like this: