Cities and the returns to education

The more education you have, the bigger the payoff to living in a city

A recent Wall Street Journal article painted the nation’s rural areas as its new inner cities, with high rates of poverty, limited economic opportunity and a range of social problems.  While the aggregate data mask enormous variation in the nation’s non-metro areas, it got us thinking about the relationship between the payoffs to education and living in the cities.

The US Department of Agriculture’s Economic Research Service has analyzed data from the Census Bureau on the comparative earnings of urban and rural residents by educational level. The definition of “rural” in this case is the approximately one in seven Americans who live outside the nation’s metropolitan areas.  “Urban” is everything else.

The data show the regular stair-step relationship between educational attainment an average earnings: the more education you have, the higher your earnings, on average. This holds for both rural and urban areas.  But especially for those who live in urban areas, the returns to education are significantly higher. In 2015, those with a graduate or professional degree living in an urban area earned about $18,000 more than their rural counterparts, urban workers with a bachelor’s degree earned about $10,000 more than those with a BA in rural communities.  But at the lower levels of educational attainment, the differences almost disappear. High school dropouts earn almost exactly the same whether they live in an urban area or not.

The importance of urban location to capitalizing on educational attainment is even more apparent when we compute proportionately how much more urban workers earn than their similarly educated rural counterparts. Those with a professional or graduate degree earn one-third more  than their rural counterparts. Those with just a BA earn 25 percent more.  those with more modest education’s also earn a premium if they live in urban areas, but it declines to zero for those who haven’t finished high school.

Percent by which average urban earnings exceed rural earnings

It’s also the case that the return to education is higher within urban areas than it is in rural areas.  For example, in rural areas, a four-year college graduate makes about 50 percent more than a high school graduate.  But in urban areas, a four-year college graduate makes 75 percent more than a high school graduate. So its also the case that the economic incentives (and rewards) for an education are relatively higher in urban areas than in urban ones. It’s also likely to mean that these economic returns to education are more obvious to urban residents than it is to rural ones.

This consistent set of relationships between urban-ness and the economic return to education has at least two important implications.

First, it shows that cities increase the productivity of workers. Cities tend to better match workers to jobs that capitalize on their skills and interests. Cities also help workers acquire new skills. And cities facilitate agglomeration economies, particularly by enabling and incentivizing the innovation that makes workers, firms and the entire economy more productive.

Second, it helps explain why well-educated workers tend to live in cities. If you can earn more living in a city you are likely to do so. Since this relationship is stronger for higher educated workers, they have a larger incentive for living in cities. Those with a high school diploma may only see a modest increase, and therefore have fewer incentives to move to (or stay in cities). This may be affected by cost of living differences, too. If rural areas are cheaper-as they are for many things, particularly housing-this may make them more attractive to less educated workers (who generally spend a higher share of their income on housing, and are therefore relatively more sensitive to housing cost differentials). It also suggests that it will be very difficult to encourage talented workers to move to (or stay in) rural areas:

The hidden bias of big data

So-called smart cities have an achilles heel: data is biased by the status quo

Streetsblog recently highlighted a new report from Houston’s Kinder Institute, evaluating bike and pedestrian road safety based on user-reported near misses. Kinder got 187 cyclists and pedestrians to record their travel for a week in March, and identify and describe situations in which they narrowly avoided an collison. The idea behind the report is that actual crashes, injuries and deaths capture only a small portion of the actual dangerous situations that active transportation users encounter, and the if we cast our net more widely to look at near-misses, we’ll have a better idea of where danger lies. It’s not a bad idea: waiting for crashes or deaths to find out that a road is really dangerous is unfortunate.

But the Streetsblog article highlights an important caveat to using relying on this kind of data. Because pedestrians and cyclists self-censor their route choices to generally avoid the scariest and most dangerous road segments, even this methodology can produce a kind of “false-positive” giving the impression that a roadway is safe, because there are few or no reported crashes or near misses.  The report’s author, Dian Nostrikasari writes:

The near-misses that bicyclists and pedestrians sometimes experience may affect their future travel decisions and prompt them to avoid roads they know are dangerous. That, in turn, could reduce the number of collisions at particular intersections. On paper, that could make an area seem safe, even if they aren’t.

While it may seem like this is a minor technical issue it isn’t. It reveals a fundamental problem with our over-reliance on data-driven planning methodologies. As we’ve noted at City Observatory, while we have copious metrics for assessing car travel (traffic volumes, speeds, level of service, number of crashes, delay), we have precious few measures of bike or pedestrian use (or safety). This imparts a subtle but pervasive bias to planning processes, and creates the illusion that we’re managing scientifically, when in fact we’re ignoring many aspects of the system for which we essentially have no data. But more importantly, even the bike and ped measures we do have reflect the very constrained levels of activity that occur on a system that for decades has been optimized for vehicles and is hostile to people walking and cycling. As we wrote last year:

An exacting count of existing patterns of activity will only further enshrine a status quo where cars are dominant. For example, perfectly instrumented count of pedestrians, bicycles, cars in Houston would show—correctly—little to no bike or pedestrian activity. And no amount of calculation of vehicle flows will reveal whether a city is providing a high quality of life for its residents, much less meeting their desires for the kinds of places they really want to live in.

This is not a problem that can be solved by more or better data. In our view, that’s a fundamental flaw in the “Smart City” visions we hear so much about. When we gather data about vehicle movement, and parking, and even current bike and pedestrian activity, we largely miss the opportunity to talk about radically different alternatives. As we discussed earlier this year, rather than relying exclusively, or even primarily on this kind of data, we ought to be talking more about the kind of places we want to live in and the way we want to enjoy them. That’ll be a much better guide to the future that an excessive reliance on data.

A hat-tip to Streetsblog’s Stephen Miller for excellent reporting here.


Integration and the Kumbaya gap

Gentrifying neighborhoods produce more mixing, but don’t automatically generate universal social interaction. What should we make of that?

In one idealized view of the world, economically integrated neighborhoods would have widespread and deep social interactions among people from different backgrounds. We’d tend to be color-blind and class-blind, and no more (or less) likely to interact with people from different groups than with people similar to ourselves. In practice, even in neighborhoods with a high degree of racial or income diversity, it still tends to be the case that people primarily associate with people like themselves. Even in the most integrated neighborhoods, there’s a “kumbaya” gap. Should we we regard that as a sign of failure?

That’s the argument that Derek Hyra makes about gentrifying neighborhoods, like U Street, in Washington DC. Blacks and whites, rich and poor live in close proximity to one another, but primarily associate only with people like themselves in daily live. Last week’s CityLab article interviewing Hyra is entitled: “Gentrification doesn’t mean diversity.” The article’s URL is “gentrifying neighborhoods aren’t really diverse.”

The point Hyra actually makes isn’t that the neighborhoods aren’t diverse, per se, but that within the neighborhoods, people still associate primarily with people with similar demographic characteristics. We may have alleviated segregation at one level, but in personal interactions, there’s still “micro-segregation.”

CityLab’s Tanvi Misri interviews Hyra about his new book, —Race, Class and Politics in the Cappuccino City.  Hyra observes that Washington, DC’s U Street neighborhood is now more racially and economically diverse, but notes that its still the case that people mostly associate with others of similar backgrounds in places like churches, stores and coffee shops. His argument seems to be, sure, its great that so-called gentrifying neighborhoods are more integrated, but since people of different races/classes, aren’t socializing directly, its basically a failure.  From the interview:

Elaborate on what’s positive and what’s problematic about this change, and with this perception of the neighborhood.

We have been so segregated in the United States and that now that whites are attracted and willing to move into what was formerly a low-income African-American neighborhood does symbolize some progress, in terms of race relations in the United States. That we have mixed-income, mixed-race neighborhoods, I think, is a very positive thing.

But that diversity not necessarily benefiting the former residents. Most of the mechanisms by which low-income people would benefit from this change are related to social interaction—that low-, middle-, and upper-income people would start to talk to one another. They would problem solve with one another. They would all get involved civically together to bolster their political power. But what we’re really seeing is a micro-level segregation. You see diversity along race, class, sexual orientation overall, but when you get into the civic institutions—the churches, the recreation centers, the restaurants, the clubs, the coffee shops—most of them are segregated. So you’re not getting a meaningful interaction across race, class, and difference. If we think that mixed-income, mixed-race communities are the panacea for poverty, they’re not.

Is the failure to reach maximum kumbaya really an indication that more socioeconomic mixing isn’t a good thing?  We don’t think so, for several reasons  First, unless you first get mixed income, mixed race neighborhoods, you have almost no chance having the opportunity for regular  social interactions. When we live in neighborhoods widely segregated by race and/or income its even more difficult to establish these boundary-crossing personal relationships. Socioeconomic mixing is necessary, even if it alone isn’t sufficient–especially immediately–to produce deeper interactions.

Second, “kumbaya” integration is probably an unrealistic goal: even within our neighborhoods (and socioeconomic groups) we do spend our personal time disproportionately with people who share our own peculiar interests. That’s true even within economically homogenous neighborhoods: people tend to spend much more time and develop stronger relationships with people most like them.

Third:  The evidence of overwhelming that mixed income neighborhoods (kumbaya or not) have big benefits, especially for lower income kids.  They get more resources, can access stronger networks, find better partners and career paths, etc.  The evidence from the Equality of Opportunity project, led by Raj Chetty, the research of Patrick Sharkey, and Eric Chyn’s study of Chicago Housing Authority residents all confirm that simply moving to a more mixed income neighborhood materially improves the life outcomes of poor kids. In addition, an important aspect of the socioeconomic mixing in the civic commons is promoting the kind of interactions that help us develop an awareness–imperfect and incomplete as it may be–that there are real people who have very different lives and expectations than we do.

Fourth, we know what happens when people don’t have this kind of first hand familiarity with a more diverse population. It shows up plainly in the results of the last election.  People who lived in communities with limited exposure to immigrants, or in neighborhoods that were predominantly white, segregated enclaves were much more likely to vote for Donald Trump than Hillary Clinton, even after controlling for other characteristics (party affiliation, age, and income) than others.  After sifting through national polling and demographic data Gallup’s  Jonathan Rothwell concludes:

“The analysis provides clear evidence that those who view Trump favorably are disproportionately living in racially and culturally isolated zip codes and commuting zones. Holding other factors, constant support for Trump is highly elevated in areas with few college graduates and in neighborhoods that standout within the larger commuting zone for being white, segregated enclaves, with little exposure to blacks, Asians, and Hispanics.”

The more separated we are from one another, the more likely we are to not support broad-based policies that promote equality and opportunity.  In the absence of more U Streets, we get policies that produce more and more segregated suburbs and neighborhoods of concentrated poverty.  We shouldn’t fixate on the failure of U Street to achieve some imaginary ideal; instead we should recognize that its essential to do many more “U Streets”  just to offset the scale of the segregation everywhere else. Fighting segregation comes first; Kumbaya will come, if it comes at all, later.

If we set impossibly high expectations about the nature of integration, and when we’re provided with anecdotes that recent and long-time residents in a community don’t associate much with one another, it’s tempting–but wrong–to conclude the whole thing was an epic fail.  As with so much in this field, that makes the perfect the enemy of the good, or at least the somewhat better.


Just ahead: Road pricing?

Trump’s infrastructure package would let states pursue road pricing

A trillion dollars for infrastructure. That’s been the headline talking point for months about the Trump Administration’s policy agenda, but the details have been murky at best. A short white paper prepared for the campaign by now Commerce Secretary Wilbur Ross sketched out a plan for tax credits leveraging private investments, but that has approach appears to be morphing into something different.

If the road ahead were priced correctly, it wouldn’t be congested.

A new budget fact sheet released by the Office of Management and Budget earlier this month, is starting to sharpen the focus on some of the potential details of the administration’s upcoming infrastructure package. It now appears that Transportation secretary Elaine Chao is taking the lead here. She said that the administration plans to implement long run changes in how “projects are regulated, funded, delivered and maintained.”  That appears to be code for a sharply reduced federal role in funding things like highways and transit, and shifting more of the responsibility to state and local governments and the private sector. Apparently, general fund support for the highway trust funded will be pared back, reducing money for new roads, and essentially eliminating future “new starts” transit projects. The popular Tiger-grant program is also on the chopping block. The plan calls for incentivizing non-federal funding, which would include additional state and local funding as well as privately financed projects.

In the realm of transportation, one of the key “incentives” is easing the current restrictions on tolling interstate freeways. Currently, its difficult to impose tolls on federally financed roads. The proposal calls both for expanding an existing policy that allows states to explore road pricing on a pilot basis, and also more generally eliminating the ban on tolling existing roads. From the fact sheet:

  • Incentivize Innovative Approaches to Congestion Mitigation. The Urban Partnership Agreement Program – and its successor, the Congestion Reduction Demonstration Program – provided competitive grants to urbanized areas that were willing to institute a suite of solutions to congestion, including congestion pricing, enhanced transit services, increased telecommuting and flex scheduling, and deployment of advanced technology. Similar programs could provide valuable incentives for localities to think outside of the box in solving long-standing congestion challenges.
  • Liberalize Tolling Policy and Allow Private Investment in Rest Areas. Tolling is generally restricted on interstate highways. This restriction prevents public and private investment in such facilities. We should reduce this restriction and allow the States to assess their transportation needs and weigh the relative merits of tolling assets. The Administration also supports allowing the private sector to construct, operate, and maintain interstate rest areas, which are often overburden and inadequately maintained.

According to Politico, these proposals have already prompted an outcry from truckers and truck stop operators. Truckers do not like paying tolls, and existing truck stop operators recognize that privatized highway rest areas would likely be formidable competitors. But beyond the objections of these two groups, there are actually good reasons to welcome a federal policy that would allow states to implement tolls more broadly.

Despite the widespread mythology that the cars and trucks pay for the roads they drive on, the truth is that gas taxes and vehicle registration fees don’t pay for anything close to the construction and maintenance cost of roads, much less the social and environmental costs they create.  The Congressional Budget Office–the same group that scores health care initiatives–estimates that over the road trucks alone are subsidized to the tune of between $57 and $128 billion per year.

Implementing road pricing more widely is one way to assure that those who benefit from the road system pay the costs of constructing and maintaining it.  And just as importantly, pricing is a way to manage demand and reduce congestion. If we priced peak hour travel more accurately, we would likely diminish traffic, and avoid the need to spend additional billions on road capacity–an expensive lesson that Louisville is learning.

While there’s a lot to be concerned about in the Trump infrastructure proposal–particularly in vague and open-ended suggestions about eliminating the kinds of environmental review standards that have help block some of the worst boondoggles–when it comes to road pricing, urbanists might well be advised to heed the advice of Kim-Mai Cutler:  When life hands me lemons, I make tarte au citron.



Dirt cheap.

Why we’re very skeptical about urban farming.

At City Observatory, we don’t tend to have a lot of content about agriculture. Farming is not an urban activity. But every so often, we read techno-optimistic stories about how a new era of hyper-local food grown in your neighborhood or very nearby,  is just around the corner or coming soon to a city near you. The latest of these appeared at Fast Company last week, in an article asking: “Has this Silicon Valley Startup finally nailed the indoor farming model?”

Adele Peters writes about a company called Plenty, which is working on a vertical farming system. They’ve raised $26 million in venture capital to develop technologies they’re using to growing kale and a range of herbs in a mixture of ground-up plastic bottles in clear vertical tubes, aided by hyper-efficient LED lights that speed-up the growth process. Plenty’s business model calls for growing products closer to customers, so that produce is fresher, better tasting and healthier. And, at least in theory, shipping product shorter distances allows them to grow more fragile or more perishable varieties than crops that have to travel for days or thousands of miles.

There are a lot of cool aspects to this product: Peters extols the tastiness of the product–regaling readers us with stories about how celebrity chef Rick Bayless pronounced the product surprisingly good–and shares Plenty’s claim that vertical farming is 350 times more space efficient than conventional dirt-based agriculture, While the story is replete with VC-pitch based talking points about the efficiency of some aspects of the indoor farming model–purportedly 1 percent as much water use as field crops, 30 miles to the consumer, not 3,000, plus the company uses LED lights are 64 times more cost-effective than those available a few years ago–one fact is conspicuously missing from the narrative:  How much will consumers be asked to pay for indoor-grown kale and basil?  As a comparative price point, its worth noting that its already possible to buy live butter lettuce (root ball attached) for $3.89 and live basil plants (Trader Joes sell’s them nationally for $3-$4.)

There are good reasons to be dubious of the economics of urban agriculture. Plenty’s technology requires that they pay urban or retail prices for things like water, energy, growing substrate and space. With conventional farming, water is cheap or even free (rainfall, or water that landowners get by right), the energy content of food comes directly from the sun (again at zero price, bundled with the land). And rural land is orders of magnitude cheaper than urban land. The average value of an acre of cropland in the US is about $4,000 an acre, or roughly 10 cents a square foot.  Meanwhile, typical prices for bare land in urban areas are frequently on the order of $10 to $20 per square foot (and higher in big cities).  So just for starters, vertical farming has to be 100 times more efficient than conventional farming to offset land costs (and that’s without considering capital construction costs to go indoors). To economists, the high price of urban land isn’t so much an obstacle to urban or indoor farming as much as it is an indicator telling us that we have much better uses for our limited supply of urban land than growing food. Every acre of city land not available for housing, for example, pushes residents further out (often on to farm land), and entails more driving for all trips.

Its also worth noting that the technology for quickly moving fragile and perishable plant-based products across the globe is fully mature, and surprisingly inexpensive. For example, we get tulips from the Netherlands and raspberries from Chile for prices that are competitive with US grown product.

Little surprise then, as Peter’s mentions in the article’s tenth paragraph, that several startups–including Farmed Here, Local Food, and Podponics–that have tried indoor farming have gone belly up, taking millions of dollars of investors funds with them. Whether Plenty will succeed where others have failed most likely hinges on the un-asked question of what price it will charge for its produce.

The headline of this Fast Company article is a classic instance of what we call “Hertz’s law*”, named for our City Observatory colleague Daniel Kay Hertz who observed that when story or blog post titles are framed as a yes/no question, the best answer to that question is almost always “no.”  “Has this Silicon Valley Startup finally nailed the indoor farming model?” In this case, we believe Hertz’s law applies. But we’ll stay tuned to see whether Plenty can make the math work. We just hope the next business reporter that writes a story about this company asks a hard question or two about the retail price of the end product.

By the way, our pessimism isn’t because we’re somehow not highly enamored of fresh greens. Here at City Observatory’s headquarters in Portland, a small patch of arugula grows year-round. It is picked minutes before its served, and travels about 75 feet to where its consumed.

Urban arugula.

While we agree there’s undoubtedly a market for premium grade produce, its probably not a very large one, and competition in the high end of the food business is increasingly fierce (as Whole Foods has learned). Indoor farming is likely to be niche technology for some specialized products, but as long as dirt is cheap, and sunlight is free, it’s unlikely to be much of a challenge to open-air, soil-based farming. While we believe cities do most things better than anywhere else, efficiently growing enough food to feed several billion people is the thing they’ll probably continue to do less well.

* Several commenters have alerted us to the much longer lineage of this observation, which goes under the name “Betteridge’s Law.”

Volunteering as a measure of social capital

Volunteering is one of the hallmarks of community; here are the cities with the highest rates of volunteerism

The decline of the civic commons, the extent to which American’s engage with one another in the public realm, especially across class lines, has been much remarked upon. Our report, Less in Common, explores the many dimensions along with the fabric of our connections to one another has become increasingly strained over several decades: we are less likely to socialize with neighbors, we travel in isolation, increasingly we recreate in private, rather than public space, and as a result, the strength of a shared public realm has deteriorated.

In his book Bowling Alone, Robert Putnam popularized the term “social capital.” Putnam also developed a clever series of statistics for measuring social capital. He looked at survey data about interpersonal trust (can most people be trusted?) as well as behavioral data (do people regularly visit neighbors, attend public meetings, belong to civic organizations?). Putnam’s measures try to capture the extent to which social interaction is underpinned by widely shared norms of openness and reciprocity.

As economist Brad DeLong explains,

. . . at some deep level human sociability is built on gift-exchange—I give you this, you give me that, and rough balance is achieved, but in some sense we both still owe each other and still are under some kind of mutual obligation to do things to further repay each other.

This sense of mutual obligation is important both to society, and the the effective function of markets. When we live in communities, places where most people have a strong sense of mutual obligation to look out for and take care of one another, social problems are lessened and economies run more smoothly.

It’s difficult to come up with a single, clear-cut indicator of social capital, so we and other researchers have ended up relying on a patchwork of different measures to judge the degree to which different cities exhibit high or low levels of civic interaction.

One of the most fundamental of these measures is volunteering. It’s long been a staple of American lore–since DeToqueville–that we regularly engage non-remunerated community activities.

Our data come from the Corporation for National and Community Service. It works with the Census Bureau to conduct a nationally representative survey exploring the degree to which Americans engage in a range of volunteer activities.

Across the nation’s largest metropolitan areas, about 27 percent of adults reported having volunteered in their local community the past year. The volunteering ethic is strongest in Salt Lake City and the Twin Cities of Minneapolis and St. Paul, where more than a third of adults volunteer.  Conversely, volunteering is much lower than the national average in cities such as Miami, New Orleans, New York and Las Vegas.

This measure stands in stark contrast to our measure of “anti-social capital” the number of security guards per capita in each metropolitan area, which we wrote about earlier this year. Not surprisingly, cities that rank high in our measure of anti-social capital (Miami, New Orleans, and Las Vegas) are all in the “top five” for security guards per capita and in the bottom five for volunteering.  Conversely, the cities with the fewest security guards per capita (Minneapolis, Portland, Grand Rapids and Rochester) are all in the top ten for volunteering.

While any ranking always implies that there are winners and losers, we interpret the variation we see here a bit more optimistically. These data imply that what happens in a metropolitan are can affect its degree of social capital. Fixing this problem from the top down may seem daunting, but improving social capital from the bottom up is something than can be done at the community level. No matter where you live, we’re sure there are opportunities for you to volunteer to help make your city a better place.

Why America can’t make up its mind about housing

Here are two ideas that, if you’re like most Americans, you probably mostly agree with:

  1. Government policy should help keep housing broadly affordable, so as not to price out people of low or moderate incomes from entire neighborhoods, cities, or even metropolitan areas.
  2. Government policy should protect residential neighborhoods from things that might negatively impact housing values, because homes are an important investment and wealth-building tool.

Having read them together like that, you’ve probably already jumped ahead to the big reveal, which is that these two ideas are almost entirely mutually exclusive. The first essentially says, “Use housing policy to keep home prices down”; the second says, “Use housing policy to keep home prices up.”

It’s no wonder, then, that housing policy is a bit confused. The same municipal governments that require that housing on scarce urban land be taken up only with resource-intensive, high-building-cost single family homes; that use zoning to separate out unwanted apartments, shops, transit lines, and other uses on the grounds that they might hurt home values; and promote neighborhood beautification and other projects on the grounds that they will raise housing values, also issue affordable housing reports trying to understand why home prices aren’t lower, and levy “impact fees” on new development for the alleged crime of, you know, raising home values.

The problem is that, at least in certain contexts, both of these goals are legitimate and important. Of course, especially in the wake of the Great Recession’s housing market collapse, a number of people have expressed skepticism about homeownership as a wealth-building tool; surely there are less risky—that is to say, less potentially ruinous—ways to a retirement income or college fund than investing hundreds of thousands of dollars in a single asset whose appreciation, as we all now know, is far from guaranteed.

And perhaps, if we could go back to the New Deal and talk this over with President Roosevelt before he inaugurated the era of mass homeownership with a federally regulated and subsidized mortgage market, we would want to make that point. But today, some eighty years later, it’s a bit late. That’s because homeownership has already provided a route to middle-class stability for tens of millions of households—and those households are mostly white. As the Washington Post wrote in a fantastic series earlier this year, public and private housing discrimination has led to a situation where home prices in black neighborhoods—even ones where the vast majority of households have solidly middle-class incomes, or higher—are much, much more unstable than in white neighborhoods.

This inequality is buttressed by the fact that our homeownership subsidies have worked selectively to the benefit of higher-income homeowners: the mortgage interest tax deduction, for example, gives larger tax expenditures to people who own more expensive houses. The more your home appreciates in value, the more benefit you get from the exclusion of home sales from capital gains tax. Ditto the value of excluding imputed rent, and deducting property taxes. The net result, according to the University of Chicago’s Atif Mian and Amir Sufi, is that homeownership has significantly magnified wealth differentials in the US.

One result of all this is that many white families have built generations of wealth through homeownership, while black families have made barely any progress: In fact, The Atlantic reported on a study suggesting that homeownership has been a net financial loss to African Americans since 2000. In 2013, the median white household held $126,000 in wealth from their home, while the figure for the median black household was just $31,000. That gap, in turn, represents a massive difference in the ability of a family to withstand a big financial shock—unexpected unemployment, for example, or a serious medical crisis—that may go some way to explaining why black middle-class workers are much, much more likely than their white counterparts to fall back into poverty.

(And the resistance of many neighborhoods to growing property values is quite strong indeed: look at how much of central and southern Brooklyn has actually lost property value since 2004.)

Giving up entirely on the idea of homeownership as a path to wealth-building would essentially be saying that black Americans have just missed the boat on this one, and will have to remain behind forever. On the contrary, in the relatively few places where housing values finally do go up in mostly black neighborhoods, it represents—at least in part—a kind of justice: giving black homeowners the same access to financial gain that their white counterparts have enjoyed for the better part of a century.  

Unfortunately, that gain also represents a loss for people, especially renters, who can’t afford to pay much more than they already do, and for whom artificially low prices in largely minority neighborhoods meant access to locations that they would not be able to afford in a normal market where race did not play such a major role in housing prices.

There is no good reason for East Garfield Park, 20 minutes west of downtown Chicago, to be so cheap—except for issues related to being a segregated black neighborhood. Credit: Eric Allix Rogers, Flickr
There is no good reason for East Garfield Park, 20 minutes west of downtown Chicago, to be so cheap—except for issues related to being a segregated black neighborhood. Credit: Eric Allix Rogers, Flickr


So how to square that circle? Well, that’s basically the challenge of housing policy in a nutshell. Perhaps a start would be to acknowledge that there is, in fact, a tension here—that “protecting” or “promoting” property values is the same thing as “making housing more expensive.” It’s somewhat discouraging, for example, when community organizations claim that “affordability doesn’t mean housing values have to remain stagnant,” without acknowledging that if housing values aren’t stagnant—ie, they’re growing—that means they’re also becoming less affordable.

But there is some hope. For one, robust production of housing that isn’t priced by the market, and therefore isn’t affected by rising market prices. That can be accomplished through public housing, privately-developed affordable housing with programs like LIHTC, and housing vouchers. At the moment, few places produce non-market housing at anything close to a scale that would provide broad affordability, but there are encouraging examples: Portland, for instance, has created 2,300 units of affordable housing in its redeveloping Pearl District, adjacent to downtown, supported largely with funds from tax increment financing.

In many places, having a wide variety of housing types and sizes can also make room for people of a wide variety of incomes. My street in the Edgewater neighborhood of Chicago, for example, contains a handful of single family homes, whose value at this point probably reaches into the seven figures; expensive newer condo buildings; older multifamily buildings, some of which have large, luxuriously updated units, and others whose apartments are somewhat smaller, or have less up-to-date finishes; and a few single room occupancy buildings, with minimal accommodations. As a result, there is market-rate housing for everyone from upper-middle-class professionals to working-class immigrant families to low-income elderly adults. Of course, that sort of diversity is typical of a pre-zoning “illegal neighborhood”: a vanishingly small proportion of American neighborhoods allow that sort of mix to be created today, which is a large part of the problem.

We are, in conclusion, profoundly conflicted as a nation when it comes to housing: we want it to be affordable, but we also want its prices to rise fast enough to be valuable as a financial investment. That’s a contradiction we need to acknowledge if our housing policy debate—and, ultimately, our housing policy—is going to be coherent and constructive.

My illegal neighborhood

Editor’s note:  City Observatory is pleased to provide this guest commentary by our friend Robert Liberty a keen observer of and advocate for cities.  We first published this post in 2015, but its as timely today as it was then.


by Robert Liberty

For many years I lived in Northwest Portland, Oregon.

It was a part of the city first settled by white pioneers in the 1860s, but development really took off when the streetcar arrived in the first half of the 1900s. (A century later, the old streetcar tracks had to be dug up so they could put down the new streetcar tracks.)

I first moved there in the 1980s by renting a part of a house. Then I moved a few blocks away into a courtyard apartment building of a type built all over the city in the 1940s. There were a dozen one and two-bedroom apartments on two floors around a small courtyard, built on a 15,000 square foot lot (about one-third of an acre, roughly the size of many suburban house lots). There were storage areas and a laundry room in the basement.

Next door to the west was a large single family house, built around World War I. To the south was a one-story three-plex: three tiny apartments slotted into a narrow strip between our building and a large old home.

Kitty-corner across the street was a small restaurant that served breakfast at a few booths and a counter. For a few years, every Saturday, a long black limousine with tinted windows would park near the restaurant and the chauffeur would deliver a hot breakfast to the occupant and then take away the dirty dishes. I never found out who was in the limousine.

Diagonally across the street to the northeast was a warehouse that processed large volumes of “direct mail”—i.e., spam.

Across the street to the north sat another Edwardian house used for offices, a bland three-plex built in the 1970s, and a four-plex that looked like a large single-family home in Dutch Colonial style. I lived in that four-plex happily for many years.

The rest of the street was a mix of large older homes on small lots and small apartment buildings. Both young families and older couples lived in the houses and apartments.

The street was shaded by big trees and it was usually very quiet. The street was so narrow that bigger cars had to queue to pass each other, partly because so many people parked their cars on the street since the apartment buildings provided few or no parking garage spaces.

At the other end of the block was a park that was also served as part of an elementary school’s grounds. The school was built of blond-colored brick and rose three stories. It’s locally famous as being on the migration path for Vaux’s Swifts. Early each fall thousands of the birds would swarm and then spiral down into the decommissioned smokestack of the school incinerator and boiler. Beside the school were some community tennis courts.

Not far from the school was a senior center and some subsidized housing for families of modest means. Scattered here and there in the nearby blocks were grand old houses—some beautifully maintained and very expensive, other cut up into legal and illegal apartments.

Three blocks away was an arterial street, but it wasn’t too much wider than the street in front of my apartment building. I often walked there to buy groceries from a small grocery store and drop off my dry cleaning. Another block or two farther along the arterial was a branch library. Across the street from the grocery store was a small sheet metal fabrication business.

Once, when I was explaining to a reporter how our neighborhood had every possible kind of use and service, I gestured to the sheet metal company to illustrate the presence of light industrial uses. It was then that I realized is was called Schmeer Sheet Metal Works and Fabrication. “See,” I said, “we have the whole schmeer.”

That neighborhood is typical of many older neighborhoods in American cities. And in almost all of American cities and suburbs, that neighborhood would be illegal.

It is illegal to build an apartment building in a district of single family homes. Residential zoning was adopted in order to prevent single family neighborhood property values and families from being degraded by the presence of apartments where immigrants and low-class people lived. (If you think this is an exaggeration read the early history of zoning including the various state and federal supreme court decisions upholding challenges to the constitutionality of residential zoning.)

Residential zoning today has carried class separation to great extremes, which you can see if you travel by air: Over here, big single-family homes on big lots. Over there a mobile home park. In another direction, a pod of apartment buildings. A place of every income, and every income in its (separate) place.

Some affluent cities use their power to regulate development to exclude entire categories of housing from within their border, like apartments and mobile homes.

Typical city zoning makes it illegal to build or operate a warehouse or a light industrial use next to homes and a grocery store. The separation of industrial and commercial uses from residential uses was the very foundation of zoning a century ago.

It is illegal in most cities to build apartment buildings without providing one or more parking spaces for every apartment. The same would be true of grocery stores or office buildings. The neighborhood’s grocery store has fewer than 20 parking spaces.

The street in my old neighborhood does not meet more current design requirements, because it is considered inappropriate to design a street so that car cannot pass each other at any time or location. The street is 27 feet wide, curb to curb. That includes parallel parking on both sides, leaving a travel lane about 12 feet wide. That violates the standards for a local road recommended by the American Association of State Highway and Transportation Officials.

In most cities, you cannot operate a business out of your home if you have employees or customers arriving from other locations.

In too many places, it is effectively illegal to build subsidized housing for families of modest means. Even when it might be legal, local officials can interpret nebulous phrases like “preserve neighborhood character” or complex regulations in way that such housing is never approved.

A senior center, even though it is not a business, would be treated like a commercial use that cannot be allowed next to single family homes.

The elementary school would probably be illegal too because the school property would be too small to meet many state’s standards. The school is located on about 9.8 acres but many of those acres are occupied by a park open to the public at all times. The school, which has 685 students, would require a site of 11.85 acres in California, Texas and Connecticut, 15 acres in New Mexico and 18-20 acres in suburban Pennsylvania.

And then there is the absence of parking places; according to Virginia’s 2010 school design standards, the school should provide parking for all the staff, visitors and about a third of the students. (Apparently the legal driving age in Virginia is much younger than in Oregon.)

Of course, a jumbled neighborhood like mine would probably be regarded by many residential realtors, local officials, and even prospective home purchasers as a bad investment. After all, it’s about as far from the suburban residential model as possible. But in fact, this neighborhood, while providing many apartments (formerly) affordable by lower-income renters, was and is highly sought after.  According to Zillow, homebuyers in this neighborhood pay more than twice as much per square foot to live here than they would in the region’s suburbs.

One reason the prices are so high is because the supply of this kind of neighborhood has been limited by zoning, parking regulations, street design standards, school design standards, and building codes. We need many more neighborhoods like this all across America, so that all of the increasing numbers of people who want to live in places like this can afford to live in them.

Does that mean do away with all regulations? No. But it does mean that we need to stop assuming that everyone wants to, or can afford to, live in a big-house on a big lot in a residential-only neighborhood.  We shouldn’t be making it illegal to build the kind of neighborhoods, like mine that are increasingly popular and in short supply.

An illegal neighborhood in NW Portland.

Robert Liberty has worked over the last 34 years as an attorney, elected official and university program administrator to help implement plans to create livable, sustainable and equitable cities and to conserve the rural lands and resources we need for food, fiber and wildlife.  He has called Portland home for almost half of a century.

Key to prosperity: Talent in the “traded sector” of the economy

“Traded sector” businesses that employ well-educated workers mark a prosperous region

At City Observatory, we regularly stress the importance of education and skills to regional economic success. Statistically, we can explain almost two-thirds of the variation in per capita income among large metropolitan areas just by looking at the educational attainment of the population.

The strong relationship between education and income is a basic fact, but there’s a lot of nuance. A more sophisticated way of understanding the role of talent is to distinguish between what economists call the “traded” and “local” sectors of the economy. Traded sector businesses are those that sell their output in competition with businesses in other states or nations. Local businesses, as the name implies, sell their goods and services primarily or exclusively in a local market. By definition, local businesses tend to be sheltered from competition from other places.

Many well-educated workers are employed in the local sector of the economy. Teachers, health care workers and government employees are all more likely to have a college degree than the typical worker. There tends to be little variation in place to place among the education and skill levels of workers in these industries. Instead, the big variations among metropolitan areas in the skill level of workers is in those trade sector businesses.

A good measure of the knowledge intensity of a region’s traded sector is the share of workers in the traded sector that have a four-year college degree or higher education. Below, we’ve used data from the American Community Survey to compute the fraction of traded sector workers in each metro area with a four-year degree. Slightly fewer than 30 percent of all traded sector workers in large metro areas had a four-year degree. At the top of our rankings are some of the nation’s most recognizable knowledge centers–San Jose, San Francisco, Washington and Boston, all of which have college attainment rates of more than 40 percent in their traded sector industries.


A low level of educational attainment in the traded sector is a good indicator that a region’s exporting industries aren’t based on strong knowledge assets. Two of the lowest ranking cities (Riverside and Las Vegas) disproportionately specialize in industries that are not knowledge intensive: distribution and tourism, respectively.

This measure excludes college-educated workers employed in health care, education and government, which are not considered traded sectors. Data on traded sector employment by educational attainment are taken from the American Community Survey and tabulated from the Integrated Public Use Micro-Sample (IPUMS).

More debate on city revival

Is the urban renaissance over? 

Earlier this week, The New York Times published an op-ed from Jed Kolko–”Seattle Climbs but Austin Sprawls, The Myth of the Return to Cities”–offering up another iteration of his long running argument that the urban rebound is overstated.  His key point: in the aggregate suburbs are growing faster than cities, and that the density of most metropolitan areas continues to decrease.

The article is surely refuting something of a straw man: not everyone is moving back to cities everywhere. And some big metro areas in the sunbelt continue to sprawl rapidly. But even if they aren’t absolutely outperforming suburbs everywhere, cities are relatively much stronger than they were a decade (and especially several decades ago), and are disproportionately attracting the best and the brightest. The chief reason they’re not doing even better–as Kolko concedes–is that we’re not building housing in city centers fast enough.

Kolko’s indicated that  the move back to city centers is real:

That revival is real, but it has mostly been for rich, educated people in particular hyperurban neighborhoods rather than a broad-based return to city living. To be sure, college-educated millennials — at least those without school-age kids — took to the city, and better-paying jobs have shifted there, too. But other groups — older adults, families with kids in school, and people of all ages with lower incomes — either can’t afford or don’t want an urban address.

And coincidentally, at the same time The Wall Street Journal summarized a new report from BuildZoom’s Issi Ronnem which emphasized the growing demand for housing in cities.  Ronnem’s work shows that home sales have accelerated closer to the urban core in most cities, and remain depressed in the suburbs.  The takeaway:

While new home sales within 5 miles of the centers of 10 of the country’s priciest and most densely populated metropolitan areas have surpassed levels from 2000, they remain more than 50% below where they were in 2000 when you go more than 10 miles out.

As Justin Fox pointed out at Bloomberg, the we’re bumping up against the limits of supply in cities. More and more Americans, especially young adults want to live in cities, but the supply of housing there is constrained, and growing only slowly–which is why rents have spiked in cities. And while more households may be locating in less dense neighborhoods, the price of housing in suburbs has fallen relative to that in city centers.


We’ll weigh in soon with our own City Observatory analysis of these data, but in the mean time, we think that several of the points that we raised a few weeks back about Kolko’s last round of research are worth remembering. There are several key reasons why no one should be glum about the continuing prospects for urban growth and revitalization.

First, employment and population data aggregated nationally by county don’t reveal much about the most urban neighborhoods.  We’ve made this point before with Kolko’s analysis of job growth.  County boundaries correspond poorly, if at all, with cities or urban density.

Consider Chicago, which gets tarred with the “Biggest Loser” title in earlier Kolko’s tabulations.  While Cook County saw an overall decline in population, the most urban parts of the city, including its Downtown Loop and Northside neighborhoods, continued to rack up big gains.


A more finely grained geographic analysis shows that the closer you get to the city center in most metros, the stronger has been the performance. While its true that the more outlying parts of some cities are losing population, their cores are becoming increasingly vibrant. As we’ve noted, that notion of critical mass at the neighborhood level is one of the defining characteristics of urban growth.

Second, national aggregations conceal local patterns. To be sure, these data show a renewed growth in the sunbelt. And in the biggest sunbelt metros, a big chunk of the growth is in their sprawling suburbs (Houston, Dallas). But the growth in these metros is not representative of what’s happening in many other places.

Third, there’s a baseline issue here.  City growth has decelerated from the past year or two.  But city growth this decade looks far different than it did a decade ago.  While Kolko’s post just shows the change in city and suburb growth rates over the past few years (and emphasizes the one-year change between 2015 and 2016), his longer blog post on his own website shows the change in population by type of county since 2001. Taking this longer view, its apparent that growth rates in suburbs have declined sharply since the last decade, while growth rates in urban counties were up.

Between 2001 and 2005, low density suburban counties (gray line) were growing more than 2 percent per year–their growth rates since 2011 are about half that. During that same time, the most urban counties (blue line) were growing sluggishly or losing population–and they’re now growing. While urban counties dramatically underperformed suburban ones last decade, the growth patterns have converged dramatically since then. As a result, when judged against the baseline of the previous decade, more urban counties have accelerated, while more suburban ones have faded.

Fourth, in many places we’re bumping up against the (policy-induced) limits to meeting the market demand for urban living. In the early stages of growth, cities can add population by filling vacant housing. But as vacancy rates fall, the only way to accommodate more people is to add more housing, which is a process that (a) takes time, and (b) is too often unfortunately encumbered by NIMBY building restrictions. That suburbs might grow faster than cities for a time is not necessarily evidence of a declining demand for city living, so much as it is evidence that we haven’t expanded the supply of urban living opportunities to meet that demand.


The Geography of Independent Bookstores

Which cities have the strongest concentrations of independent bookstores?

Last week, we explored what we called the “mystery in the bookstore.” There’s a kind of good news/bad news set of narratives about bookselling in the US. After decades of decline in independent bookselling, many cities have seen a rebound by locally run stores. And while that appears to be true in many different locations, the overall trend, in terms of the number of bookstores counted in government statistics, still seems to be downward. We’re not quite sure how to reconcile these to divergent trends, so it remains something of a mystery–although we have some suspicions about what’s happening.

Independent bookstore owners Toni & Candace.

Part of the answer to this mystery lies in geography. The bookstore industry seems to be doing better in some places than in others. So, in typical City Observatory fashion, we set out to quantify the concentration of independent bookstores in different metropolitan areas. For data, we turned to the Indie Bookstore Finder, a web-based directory that let us search by radius in different cities. Like other web-based resources, its likely that not every independent bookstore in the country is included, but the data are extensive and nationwide in scope. We selected a point in the center of the central business district in each of the nation’s 53 largest metropolitan areas (all those with a population of a million or more) and searched for independent bookstores within a 25-mile radius.  The Indie Bookstore finder reported that there were 453 independent bookstores in these metropolitan areas.

There’s a pretty wide distribution in the presence of independent bookstores among metropolitan areas, at least according to this data source.  Two Pacific Northwest cities, Seattle and Portland top the list, with more than 7 independent bookstores per million population.  The typical large metro area has between 3 and 4 indie bookstores per million; three metro areas, Hartford, Jacksonville and Virginia Beach have none (at least none who are captured in this database).

As we noted yesterday government statistics count a lot more establishments as bookstores. For example, according to xxx, there are an estimated 3,500 bookstore establishments in these metro areas, so our “indie bookstore” listing about 10 percent of establishments counted by the Census. Many of the businesses classified as bookstores under the Census data are often auxiliaries to other businesses, such as a bookstore that’s part of a museum, or sell a range of printed material, like greeting cars or magazines (think of an airport bookstore). Given the demographics of the firms in this database, however, they seem to represent the more substantial and “book” focused retailers.

What patents tell us about America’s most innovative cities

Patents rates are a useful indicator of innovative activity

The US is increasingly becoming a knowledge-based economy, and as a result, the markers of wealth are shifting from the kinds of tangible assets that characterized the old industrial economy (like huge factory complexes) to much more intangible assets (the creativity and innovativeness of workers and organizations). Most of our statistical measures of economic activity were crafted for our machine-age economy, so its often a challenge to come up with measures of the new, intangible wealth that characterizes today’s economy.

One useful indicator of innovative capacity is patenting. The US government awards patents for novel ideas, and patent records record not only the name of the inventor, but her location as well. As a result, its possible to map and tabulate the density of patenting in different states and metropolitan areas.  The US Patent and Trademark Office provides a periodic tabulation of patent data, and you can also drill down to particular patent categories to identify the kinds of technologies that are present in a particular location, and even track down the number of patents awarded to particular firms or inventors.

To rank metro areas by innovativeness, we’ve computed the number of patents issued per 100,000 population for each of the nation’s metropolitan areas with more than a million population. San Jose–home of Silicon Valley–is far and away the most prolific patenter among US metro areas (it has about 770 patents per 100,000, a number that flows off the chart).  Other tech centers are also leaders in patents per capita including San Francisco, San Diego, Austin, Seattle, Raleigh and Boston.


As the data show, the distribution of patenting activity is highly skewed to tech centers.  The typical large metropolitan area has just 40 patents per 100,000 workers.  (The gray bar on the chart shows the inter-quartile range; about half of all metro areas get between 20 and 60 patents per 100,000 workers. The least patent-intensive metros include a group of mostly Southern cities (Virginia Beach, Birmingham, and New Orleans).

To be sure, there are important limits to patents as an indicator. Patenting is extremely common in some industries (like biotechnology and semiconductors), where the ownership of intellectual property is both an important source of competitive advantage (and a bevy of patents is an important source of trading stock for working out cross licensing agreements for firms with complementary technologies).  In some industries, patenting is rare, or unknown altogether. New styles and designs for apparel, for example, are seldom patented.


The Week Observed, May 26, 2017

What City Observatory did this week

1. Dirt Cheap. A number of tech startups are exploring techniques for high density urban farming. In theory, new methods, like vertical farming in plastic tubes, can greatly reduce the amount of land and water needed to grow certain crops, and offers the advantage of much shorter distances to consumers. Still, the big question is whether urban farming can produce food at prices that are competitive with conventional agriculture. Dirt is cheap, and the sunlight falling on it is free. Farmland is generally orders of magnitude less expensive than urban sites, so indoor growing has to be even more efficient to offset its cost disadvantages. A recent  story in Fast Company about Bay Area startup Plenty dutifully echoes the talking points of proponents about technical advances, but is silent on how much this urban produce will cost.

2. The return to the city. In an article in the New York Times, Jed Kolko again challenges the idea that everyone is moving to cities. He finds that many US metros have become, in the aggregate, less dense in the past several years. We look at the evidence on the patterns of population and job growth, and note some of the limitations of county level data in judging these trends. It also seems to be the case the movement to city centers is increasingly bumping up against available housing supply, a trend confirmed by the widespread increase in city center housing prices compared to prices on the suburban periphery: It appears that even more Americans would like to live in the center than we have built housing to accommodate.

3. Just ahead: road pricing? The details of the Trump Administration’s infrastructure proposal seem to be coming into clearer focus. The headline $1 trillion number apparently will count on lots of private, state and local resources, and will actually reduce federal funding for many programs. The highway trust fund may lose its general fund transfers, which will mean less money for roads and transit. One bright spot, in our view: the administration is proposing to greatly liberalize state authority to impose tolls on federally funded roadways. This could encourage states to more widely apply road pricing strategies, which would be much more effective at fighting congestion.

Must read

Dan Savage, who mostly writes a sex-advice column for Seattle’s alt-weekly, weighs “Doing something real about gentrification.” As always, he’s readable and provocative–skewering anti-development gentrification critics–“If I meet one more anti-gentrification activist who moved to Seattle ten minutes ago, I shall scream.” But most of the column focuses on wonky bits that will be familiar to City Observatory readers: “Housing scarcity—exacerbated by the ridiculous amount of this city zoned for single-family housing—deserves as much blame for the displacement crisis as gentrification. More. And unlike gentrification (“a once in a lifetime tectonic shift in consumer preferences”), scarcity and single-family zoning are two things we can actually do something about. Rezone huge swaths of the city. Build more units of affordable housing.” Savage also urges a big increase in transit service and subsidies to low income transit users as a way of buffering the economic impact of urban growth. This is smart, opinionated and readable: share it with your less wonky friends.

Todd Litman, Gentrification for Better or Worse. For a counterpoint–if only stylistically–to Dan Savage’s exuberant essay on gentrification, we turn to the soft-spoken and even-handed Canadian, Todd Litman. Litman, who runs the Victoria Transportation Policy Institute, mostly does research on transportation policy, but turns his thorough and balanced analytical approach to the issue of gentrification. You’ll find a two-by-two table that tabulates the benefits and costs of gentrification for existing residents and newcomers, and weighs the relative importance of each. While far less vitriolic than Savage (indeed, not vitriolic at all), Litman reaches many of the same broad conclusions, stressing that its possible to have gentrification without displacement if the housing supply increases, and noting that opposing higher density development in an attempt to prevent gentrification is likely to backfire, driving up rents still further.

Aaron Renn on infrastructure policy. As the discussion of the Trump Administration’s potential one trillion dollar infrastructure policy heats up, you’ll find some very useful background material in Aaron Renn’s essay at American Affairs. He makes a number of very important points that are largely missing from the discussions in Washington. Chief among them: building new infrastructure is unlikely to be a major economic stimulus. “In short, our infrastructure systems are, by and large, built out. There is little need for a major deployment of new infrastructure, or new types of infrastructure. There do not appear to be any huge economic gains to be had from new infrastructure on par with those coming from electrification or building the interstate highway system.” Nor do much vaunted public-private partnerships offer any kind of panacea: private investors want to be paid back, and are adept at shifting risk to the public sector, and maximizing their own profit, often at the expense of taxpayers or system users. The most important tasks in the realm of infrastructure may be prosaic: regular maintenance, competent management, and solvent finance.

New ideas

Are ride-hailing services prompting people to sell their cars? A new survey from Reuters asks recent car sellers if they had replaced the vehicle they just sold.  About a quarter of American adults reported selling (or trading in) a vehicle in the past twelve months. While the vast majority replaced their vehicles, some 9 percent of car sellers said that they had shifted to services like Lyft or Uber, rather than buying a new vehicle. About the same share of prospective car sellers report planning to do the same in year ahead. If this trend holds, it has important implications for urban transportation (as well as the finances of car makers).



The Week Observed, May 5, 2017

What City Observatory did this week

1. Mystery in the Bookstore. In cities around the country, there’s been a noticeable rebound in the local bookstore business. After decades of steady decline, this is a pleasant surprise. One metric, the number of members of the national association of independent bookstores confirms this trend. But broader data compiled by the federal government suggest that overall the number of bookstores has continued to fall. Its a bit of mystery as to why these indicators are headed in different directions. We consider the reasons why this might be the case, and what it says about cities.

2. Immaculate conception of housing. It’s easy to take the existing housing stock for granted–it’s there, in place, and seems like its always been that way. But virtually all of the housing in our cities is the product of some profit-motivated developer building what at the time was considered large, unaffordable and even garish housing for the middle and upper classes. Case-in-point, the humble bungalows that until recently were part of most city’s stock of affordable homes, were, when new, housing for the aspiring households with above average incomes (the lower classes had to content themselves with crowded tenements and rooms in humble boarding houses). This historical context provides a useful perspective for many current housing policy debates.
3. What patents tell us about the geography of the knowledge economy. Most of our economic metrics are the legacy of an older, industrial economy. We know a great deal about the location of manufacturing employment, the occupational composition of large organizations, and the monthly volume of orders for new durable goods. But the knowledge economy–the creation of new ideas and products–which is playing a larger role in driving economic growth, has few well accepted metrics. One indicator we can examine is the location of patents. We’ve gathered data from the US Patent and Trademark Office to compute the number of patents issued in each of the nation’s largest metro areas, and to facilitate comparisons, standardized these by metro population.

Must read

1. Will wealthy parents ever choose to integrate schools?  Writing at The Atlantic, Patrick Wall, asks an important question: will wealthy families, those with the means to choose where they will live, and where they will send their children to school, ever voluntarily choose integrated schools? Wall relates the story of two adjacent public schools in Manhattan, PS 199 and PS 191, who though cheek-by-jowl, have attendance lines that are plainly divided by economic class.With wealthier parents and better supported and prepared students PS 199 is one of the cities best performing public schools, while PS 191 has more children who face learning difficulties and its parents lack the means to add much to the public budget for the school. When the city moved to rectify this imbalance by re-drawing the attendance lines between the two districts to more nearly balance the incomes of the two schools, many wealth parents balked at keeping their kids in the public school system. For them, part of the reason they paid high prices for their homes in this neighborhood was because it entitled them to send their children to a better school. Though this story is just about two schools in one city, it captures in a nutshell many of the problems confronting achieving economic integration, both in schools and in cities generally.

2. Ed Glaeser on reforming land use regulations. Brookings Institution has published a new essay from Harvard’s Ed Glaeser, author of Triumph of the City, and a prodigious scholar on regional economics which offers a non-technical summary of his latest research on the connection between local land use regulations and housing affordability. He and co-author Joseph Gyourko used construction cost data for different metropolitan markets to compute the difference between current market prices and the cost of building new homes. They find that fully 40 percent of American homes are priced below the minimum cost of profitable production, their benchmark for cost. But in a few markets, representing about a tenth of American homes, the average price of home is more than double the estimated cost of production. In such a case, its surprising that the market doesn’t produce more housing. The reason, of course, is that local land use restrictions make it too costly to build new housing. And unfortunately, these markets tend to be among the most productive urban economies in the US, so constraining population growth by limiting housing supply in these markets implies lower national productivity. Glaeser argues for state enacted housing requirements that could override local zoning to allow more housing in high cost areas, and suggests states enact fiscal incentives that would encourage local governments to zone more land for housing.

New ideas

Walking, biking and health: New evidence from the UK.  The British Medical Journal has published a new study–”Association between active commuting and incident cardiovascular disease, cancer, and mortality: prospective cohort study”– of the connection between active commuting and key measures of health. The study followed more than a quarter million 40 to 69 year-old subjects in Britain over 3 years, and looked at the correlation between commuting mode and mortality from all causes, and indicators of cardiac health and cancer. They found that cyclists enjoyed a statistically lower risk of mortality, cancer and heart disease, and that those who walked to work enjoyed a lower risk of cardio-vascular disease. Cycling in particular exhibited a “dose-response relationship” meaning that the more one cycled (as measured by the length of one’s cycle commute) the greater the health benefits. (This was true of the effect of walking on heart disease risk). They conclude that measures to encourage additional active transportation, especially cycling, would improve public health.



The immaculate conception theory of your neighborhood’s origins

A while back, a columnist in Seattle Magazine, Knute Berger, expressed his discontent with modern housing development. As Berger sees it, today’s homebuilding pales in comparison to the virtues of early 20th century bungalow development:

In a rapidly growing city where the haves have more and the have-nots are being squeezed out, the bungalows offer a lesson we ought to relearn. They recall a city figuring out a way to house its people affordably, without excess. To me, they reflect a lack of materialism, housing built not for profit, but for living in. They reflect a modest approach to life, one steeped in a conservation ethic—don’t use more than you need. Seattle culture needs to find a way to get back to those values, and create a built environment that reflects it.

This is one of the more eloquent expressions of something you might call the “immaculate conception theory” of neighborhood development. This narrative is common all around the country, in communities of many ages—from colonial Boston to postwar Minneapolis—and sets powerful background assumptions about what affordable, friendly neighborhoods can and should look like that inform many of our debates about housing. These assumptions mostly revolve around the idea that older housing was built the right way: ethically, modestly, with an eye to community rather than profit. These older values, in turn, highlight the faults of modern buildings: gaudy and wasteful, disruptive to existing communities, and motivated only by money.

Old building: virtuous. New building: villainous.

The problem with the immaculate conception theory is that, like parents swearing that they would never have behaved the way their kids do, it is conveniently forgetful about what actually happened in the past. Taking, just as an example, the kind of housing that Berger romanticizes—the early 20th century bungalow boom—a closer look reveals that it was defined not by mass affordability, efficiency, and respect for traditional communities, but something very nearly the opposite.

To begin with, many of the arbiters of taste at the Seattle Magazines of the bungalow era believed those new bungalow neighborhoods “ruined” the character of the places they were built, just as new apartment buildings are maligned today. They even had a snappy put-down for it: “bungalow disease.” “Tradition has broken down,” wrote the British planner Thomas Sharp, describing a proliferation of bungalows on both sides of the Atlantic, and “taste is utterly debased…. The old trees and hedgerows…have given place to concrete posts and avenues of telegraph poles, to hoardings and enamel advertising signs.” Closer to Berger’s home, Architectural Record reviewed Seattle’s building boom in 1912 and, in an otherwise positive article, pronounced the quality of its new homes “disappointing.”

Critics accused new bungalow neighborhoods not just of being ugly, but of ripping apart the social fabric of the city. One writer argued that in new neighborhoods full of many separate houses, “each building is treated in isolation, nothing binds it to the next one,” and as a result they lacked an “essential” “togetherness.” Another pointed out that the rise of bungalow neighborhoods coincided with the rise of decentralized business districts, as these sprawling areas—bungalows took up much more space per person than either the more modest single family homes or apartment buildings that had come before—encouraged outlying commercial development and car ownership.

Another lovely open field built over and ruined. Credit: Skokie Heritage Museum
Another lovely open field built over and ruined. Credit: Skokie Heritage Museum


Which brings us to our next point: Far from being based on an ethic of efficiency and conservation, early 1900s bungalows represented a dramatic leap to neighborhoods that required higher energy consumption than ever before. This was true, first of all, because bungalows tended to be much larger than existing homes. While Berger marvels that one 1910 bungalow was just 1,600 square feet, the average home size at the time was closer to 1,000 square feet—making the 60 percent larger home look like a veritable McMansion. In addition, many of these new single-family-home neighborhoods, which were built much further from job centers and at much lower densities than older communities, were enabled by the boom in energy-consuming automobiles, and encouraged their use. By the 1920s, one in every two American families had a car—a figure that was much higher in bungalow neighborhoods—and public transit began losing many of its customers to driving. In the same decade, suburban population growth outpaced that of cities for the first time ever.

Finally, the idea that bungalows represented a housing type that was affordable and open to all—and an ethic that valued community instead of money—simply doesn’t describe actual American cities in the 1910s or 20s. Home prices in the 1920s were rising rapidly, leading many people to talk about a housing crisis in terms not so terribly different from today’s. But as Gail Radford describes in her book Modern Housing for America, bungalows weren’t holding the line on cheap homes: in many cases, they represented the luxury housing of their day. Bungalows were so much more expensive than the more modest homes that had preceded them that while the overall cost of living increased by about a factor of two between the 1890s and 1920s, the cost of an entry-level home had increased by a factor of five and a half. Even before the economic crash of 1929, there was a growing foreclosure crisis, strongly suggesting that “housing costs were simply too high in relation to incomes for many families.”

Moreover, the bungalow era coincided with the development of zoning codes—codes that were essential, in fact, to preserving many bungalow neighborhoods’ all-single-family character. The people who advocated for these zoning codes did so by explicitly arguing that they were needed to protect the property values of homeowners and other landowners. In other words, the denizens of the early 20th century cared so much about their houses as a financial investment that they invented an entire new regulatory infrastructure to ensure that they wouldn’t lose their value.

And of course, “not losing their value” was very closely tied to excluding any kinds of people who might threaten the neighborhood’s desirability. It’s impossible to talk about the development of urban American neighborhoods in the early 20th century without acknowledging that this was the period in which modern residential racial segregation emerged—a system of exclusion enforced by covenants, zoning, and violence carried out by the residents of all kinds of neighborhoods. This isn’t some separate issue from how those who were excluding, rather than excluded, built their homes and communities: it’s an integral part of the story, without which those bungalow neighborhoods may have looked quite different.

Community values were also quite contentious in the first half of the 20th century.
Community values were also quite contentious in the first half of the 20th century.


Why have we forgotten all of this? Partly because all the people in these stories are gone. We can’t see the developers laying roads and streetcar tracks to open up huge new areas for subdivisions; we can’t see the disproportionately wealthy people who were able to buy homes in a pre-FHA era when required down payments routinely hit 50 percent. We can’t talk to the people who remember, and miss, what existed in these places before bungalows. All that’s left are the buildings, which over the years have lost their sheen of newness, often becoming more affordable in the process, and allowing us to imagine our own stories about where they came from.

The point here is not that bungalows are “bad.” Given what has happened in the intervening century, a return to bungalow-scale living would be a huge win for sustainability and efficient living in the many postwar suburbs and neighborhoods where homes have ballooned to much larger sizes, and development has become much more sprawling. In many urban communities, bungalows today represent a prized architectural tradition, and a form of single-family home that fits neatly into the kind of mixed urban neighborhood—along with small apartment buildings and local shopping districts—that we have long since made illegal.

But there are important lessons to be learned by looking at what the bungalow era actually looked like, rather than our romantic imagination of it.

One is that everything old was once new, and new things often provoke a backlash. We ought to be humble in believing that our opinions represent some timeless, objective truth, looking backwards or forwards. The same bungalows that seem to us quaint and charming were tacky and soulless to many of the people watching them be built; it seems more than possible that the new apartment buildings we vilify today will be thought of sentimentally by future generations who know them only as an important part of their city since they were born.

A second lesson is that American cities have an impressive history of growing to accommodate new arrivals. Berger leaves out of his column, as is frequently left out in “immaculate conception” stories, that the bungalow era was also the fastest period of urbanization in American history: Between 1900 and 1930, Seattle’s population grew more than fourfold, from 80,000 to over 360,000—a rate of growth approached or exceeded by many other American cities at the time. In the process, millions of rural Americans and immigrants were given the opportunity to live in newly industrializing cities where wages and quality of life were dramatically higher. Today, most of our cities have shut the door on that kind of growth. (Seattle’s growth rate today, while much higher than many other central cities, pales in comparison to the bungalow era Berger wishes we would return to.) As a result, our doors are no longer open to as many people, from this country and others, who would like to make better lives by moving to places where job openings and quality of life are high.

Finally, the bungalow era suggests that building new market-rate housing that’s affordable to working-class and low-income people in urban areas is hard, especially if that housing takes the form of single-family homes. And it’s worse today: while the bungalow builders had the advantage of lots of open land relatively close to center cities, today, that “frontier” has closed. And we’re well aware of the costs—environmental, social, and financial—of continuing to push all of our growth out further and further onto the fringe.

An old Seattle neighborhood with both bungalows and small apartment buildings—which were and are much more affordable and energy efficient.

Rather, the deeply affordable and decent homes of the bungalow era were largely in multifamily buildings. It’s curious that, though more than four in ten of the homes built in the 1920s were in apartment buildings, that kind of construction—and those kinds of people—are entirely absent from Berger’s romantic musings about the time. But they were a crucial source of urban accommodations for people of modest incomes. As the Sightline Institute has pointed out, rooming houses and other small, multifamily homes made up a huge proportion of the affordable housing stock in cities around the country in the early 20th century. Unfortunately, a combination of regulations and market conditions has virtually eliminated that stock in most places. If we want to go return to something the 1910s and 20s got right, bringing back modestly-sized homes in multifamily buildings is a good place to start.

The past does have lessons for us—we’ve made a point of championing the now-“illegal neighborhoods” that American cities built up through the bungalow era. But we have to look at it as it really was, and not through rose-colored glasses, if we want to get them right.

The Week Observed, May 19, 2017

What City Observatory did this week

1. Volunteering as a measure of social capital. Thanks to the work of Robert Putnam, author of Bowling Alone and more recently, Our Kids, there’s a growing understanding of the important role of social capital–the relationships and norms of trust and reciprocity–in making our communities and our economy work better. But measuring social capital is tougher than measuring physical capital (like buildings and machinery). One of the indicators that illustrates the presence of social capital is the extent to which citizens volunteer their time for civic and social purposes. We present data for the largest US metro areas which show the places with the highest and lowest levels of volunteering. Here are some of the high performing cities:

2. Why America is of two minds about housing policy. There’s a fundamental contradiction between two of the most important cornerstones of public policy on housing:  We want housing to be affordable, and we also want housing to be a great investment for homeowners. The first implies that we ought to have few limits on supply and that we ought to encourage home prices (and rents) to stay low, and not increase. The second, in practice, implies the opposite: we enact zoning to “protect” property values, which has the effect of constraining supply, and we have a raft of policies designed to push up home prices. Resolving this contradiction is at the heart of solving our housing problems.

Must read

1. The culprits behind white flight.  Leah Boustan writes at the New York Times about the historical factors that fueled white flight from cities between 1940 and 1970. While there are many competing explanations, Boustan finds that all of them have a bearing on the outcome, a result she describes as comparable to Agatha Christie’s “Murder on the Orient Express” in which all the suspects struck a blow in the crime. Not only did racism fuel white flight to suburbs, but it was aided and abetted by political and fiscal factors, notably the ability of suburbs to enjoy lower taxes for any given level of public services by including higher income households and excluding lower income (and often minority) ones.

2. The disappearance of racial disparities in public housing. A not-unrelated article in the Washington Post’s Wonkblog looks at the role of changes in public housing policies in attenuating racial segregation in US metropolitan areas. They report on the results of a new study from Johns Hopkins University looking at the racial patterns of occupancy of classic public housing and voucher-supported private sector housing. As recently as the 1970s, there was a clear racial division between the two types of housing, with black households disproportionately found in public housing, and white households in voucher supported housing. The latest data suggest that in the aggregate, these racial disparities have all but disappeared. Key factors behind the trend: the demolition of some of the most segregated public housing projects, and the expansion (and apparently more race neutral) availability of housing vouchers relative to public housing.



Mystery in the Bookstore

Signs of a rebound in independent bookstores, but not in the statistics

Lately, there’ve been a spate of stories pointing to a minor renaissance of the independent American bookstore. After decades of glum news and closings, there are more and more instances of independent bookstores opening or expanding. The American Bookseller’s Association points with pride to a seven-year string of increases in its dues paying members. Articles in the New York Times “Indie Bookstores are back with a passion,” and US News “Indie Bookstores remain on a roll,” recount first hand accounts of successful firms.

The independent bookstore is an American icon. It’s hard to picture a city–the classic Main Street–without a local bookstore. Bookstores are one of the categories of customer-facing retail and service businesses we’ve used at City Observatory to create our “storefront index” which measures urban walkability. Founding father Ben Franklin was famously a self-taught intellectual who ran a book shop in Philadelphia. The indie bookseller figures prominently in pop culture, from Meg Ryan’s Shop Around the Corner bookstore owner in You’ve Got Mail, to a host of other films and television. In The Big Sleep, Humphrey Bogart’s Phillip Marlowe takes refuge in Dorothy Malone’s Acme Bookshop while staking out a suspect.

More recently, Portlandia has featured ardently feminist booksellers Candace and Toni, the proprietors of Women and Women First Bookstore.


For a long time, what with the growth of on-line retailer Amazon (which built its business model selling books at a discount) and with the advent of big-box retailing, it seemed like the small independent bookseller was a doomed anachronism. But in the past few years, there’s been a surprising rebound in local bookselling. It turns out that many readers still prefer the printed page, and gladly patronize a knowledgable and attentive local business. And the surviving and thriving local booksellers have changed their business models to emphasize personal service, community, and on-site experiences that larger and virtual competitors have a hard time matching. But while some stores are flourishing, others are floundering: in Memphis, the Booksellers at Laurelwood, one of three remaining city bookstores is closing this month. In Detroit, the city’s oldest–Big Bookstore in Midtown–is closing after eight decades. In St. Louis, it’s the half century old Webster Groves bookshop that’s closing.

One final sign that a shift back to bricks and mortar bookselling is in the cards: even Amazon is opening its own physical stores.

Government data tell a different story

With such upbeat stories in the popular press, we decided to take a quick look at Census data on the number and geography of bookstores, to see if we could corroborate and quantify these trends. We looked to two key data series, the annual County Business Patterns series, tabulated by the Census Bureau using payroll tax records, and the once-every-five years Economic Census, which survey’s the nation’s businesses about sales, wages, and business operations. We focus on the government definition of bookstores, NAICS 451211.  This statistical category includes all kinds of bookstores, from the large national chains to small, independent businesses, as well as college bookstores, and those that are adjuncts to museums.

According to the Economic Census, the number of bookstores in the US has fallen from 12,363 in 1997 to 7,176 in 2012–a loss of more than 5,000 establishments.  That pattern is also reflected more recently in the data reported as part of the County Business Patterns series. These data show the number of bookstores declining by about 30 percent since 2008, from 9,700 to about 6,900.


So here’s our mystery: While there’s been a visible resurgence in bookstores in some locations, the bigger pattern of change remains downward.

We’re not sure what the answer is to this mystery.  There are some of the usual suspects to consider.  First, its likely that many of the bookstores that are closing are the big national chains, like Borders, Barnes and Noble and Waldenbooks.  In market’s where these larger national stores are closing, it may be creating more market space for independent operators to thrive and even occasionally expand. A second factor is that much of the decline in the number of establishments may be among very small bookstores in small towns and rural areas. These are the kinds of places where the threat from Amazon (lower prices, wider selection and convenience) would be a threat.