Does increased housing supply improve affordability?

The study seems like ready-made evidence for the housing supply skeptics who maintain that building more market rate housing will do little or nothing to solve our affordability problems. 

If you want less displacement, build more housing

The more you limit housing, the more you increase displacement

In city after city, we see the same refrain: a neighborhood is starting to attract new residents and new investment, current residents are starting to worry about gentrification. They show up at city council meetings or planning meetings to voice objection to new development. Just make it stop, so our neighborhood will stay the same, and residents won’t be displaced.

This is the technology for avoiding displacement.

They’re often urged to emulate the tactics of NIMBY’s elsewhere, and stem the tide of change by blocking new development.

Slowing or stopping new development, particularly new housing development has exactly the opposite of the desired effect. It constricts the housing supply, drives up rents and fuels displacement.

We’ve seen this time and again. A couple of months back, we profiled two Oakland neighborhoods, Uptown and Fruitvale. Both experienced almost identical increases in rents and home values as the Bay Area city boomed. But Fruitvale, which has built more housing has seen dramatically less demographic change, while Uptown which has built almost no new housing, has seen its population shift.

The same holds for two neighborhoods in Washington DC’s 20003 zip code, Capitol Hill and the Navy Yard. Historic Capitol Hill has organized to largely block most new development; the Navy Yard area (near the new Washington Nationals ballpark) has seen thousands of new apartments built in the past decade. As Greater Greater Washington describes, the addition of new apartments has helped push down rents in the Navy Yard (the orange line) while rents in Capitol Hill (purple) continue to climb.

If you don’t build new housing, you intensify the shortage, raise the rents, and amplify the displacement. This isn’t intuitively obvious. People, unsurprisingly, associate new buildings with new residents, and simplistically assume that if new housing isn’t built, that new people won’t show up, or that they’ll simply go somewhere else.  But that’s not the case.  Recently, (again, as related by Greater Greater Washington) one Washington DC Councilor, Kenyan McDuffie, patiently explained how this works to one of his constituents, who was testifying against a new housing development, worrying that it would lead to greater displacement.  McDuffie said:

So you can’t have a conversation about displacement and say “Don’t build these high end condos.” Perhaps we say no to all this stuff [the proposal before them includes 375 new homes, most market rate and 113 subsidized affordable homes]. It doesn’t mean those folks aren’t going to continue to come and need a place to live, and the price of housing is going to skyrocket because we don’t meet the demand that exists.

There’s no question that a big reason that some low income neighborhoods are seeing development pressure is because wealthier urban neighborhoods and suburbs generally have been so effective in deploying NIMBYist regulations that block development.  As Brookings Institution’s Jenny Schuetz has shown, higher income neighborhoods in Washington have blocked most new housing development.  It seems to have become an article of faith among some community activists that they can only succeed in empowering low income or distressed communities if they equip them to be just as NIMBYist as any snooty suburb or high end enclave. (Indeed, in some jurisdictions, local governments and activists bluntly employ zoning approval processes or regulations like California’s CEQA to extract concessions from developers.)

Sometimes planning discussions get the relationship between more construction and less displacement exactly backwards. Consider Seattle, which for the past several years has been working on a “Grand Bargain” to help promote greater affordability and equity by upzoning land throughout the city. So far, so good.  But, it’s worried about the prospects for displacement in lower income neighborhoods. Erica C. Barnett wrote about this recently at the blog “The C is for Crank” in an article entitled “Anti-Density Activists Race and Social Justice Gotcha Backfires.

Seattle’s answer for displacement concerns: lower the amount of upzoning in these neighborhoods. The city’s strategy, according to Barnett, was to reduce the amount of new housing that could be built in areas where community members raised concerns about displacement, based on the apparent assumption that lower amounts of upzoning will reduce displacement.

But that pretty much contradicts what we know about how displacement happens. Back to Erica C. Barnett:

Whether restricting the creation of housing—any type of housing—will work as a long-time anti-displacement strategy is, of course, another question—one that city council member Teresa Mosqueda posed at last week’s meeting. “I still struggle with the terminology that if we were to do more development—again, through the community lens, led by community organizations and neighborhood leaders who who can talk about the type of housing that they’d like to see—we can actually benefit by seeing increased housing and density requirements in some of these areas that are being called at risk of displacement.

“If they are at risk of displacement, then [it seems like] we would like to see more opportunities for folks to live in those areas and not get pushed out,” Mosqueda concluded.

In the game of musical chairs that is the urban housing market, the only way to make sure that all people find a place to sit–i.e. not be displaced–is to add more chairs. The research that’s been done on the subject, notably by California’s Legislative Analyst Office–and confirmed by the skeptical academics at U.S. Berkeley’s Anti-Displacement project–is that building more market rate housing reduces displacement.

On a visceral, political level, it does seem to be fair to suggest that if rich neighborhoods are able to block new housing, then poor neighborhoods ought to have the same prerogative. But that’s a road that leads nowhere: Empowering everyone to be NIMBYs in the hope that blocking new development will cause the demand for housing to go away or be absorbed somewhere else is a recipe for worsening housing affordability and greater displacement.

This post has been revised to correctly spell Erica C. Barnett’s name.



Is St. Louis Gentrifying?

Gentrification Debates Without Gentrification?

By Todd Swanstrom

Editor’s note: We’re pleased to offer a guest commentary from Todd Swanstrom. Todd is the Des Lee Professor of Community Collaboration and Public Policy Administration at the
University of Missouri – St. Louis.  He is also co-author of Place Matters:  Metropolitics for the Twenty-First Century (

Recent research has provided evidence for the legend that Eskimos have more than fifty words for snow. Unfortunately, we seem to have only one word for economically improving neighborhoods: gentrification. Our profound lack of linguistic nuance is crippling our ability to talk about what kind of future we want to see for urban neighborhoods.

In the early 2000s, I taught courses in urban planning at Saint Louis University. The City of St. Louis is the #1 shrinking city in the world – having lost more than half a million people in the last half of the Twentieth Century (from 856,796 in 1950 to 348,189 in 2000).   For the first time, however, I began to notice that a few neighborhoods were coming back and actually gaining population and investment. I would bring these up in class. Invariably, a student would chime in, “But professor, that’s gentrification!” – bursting my bubble.

I now have a position at the University of Missouri, St. Louis where I spend a great deal of my time working with a coalition of community development nonprofits. I continue to hear anxious cries of “gentrification!” This time, they are directed not at rebounding neighborhoods but at poor African American neighborhoods north of the infamous “Delmar Divide”.

Historically, the biggest divide in the St. Louis region has been east-west, between the central city and its suburbs – with communities becoming progressively wealthier and whiter as you move west out of downtown. Increasingly, however, the biggest divide in the region runs north and south. In the era of Jim Crow, a tangle of public laws and private practices confined African Americans to neighborhoods north of Delmar Boulevard (see map). To this day, the areas north of Delmar tend to be African American and lower income. This now includes North County suburbs like Ferguson. All of the neighborhoods identified as racially and ethnically diverse (yellow) or mixed income (blue) in City Observatory’s recent report on the subject are South of Delmar.

Despite a weak housing market, many activists have warned about gentrification around the new $1.75 billion National Geospatial Agency (NGA) headquarters that is under construction in North St. Louis. When completed it will employ over 3,000 highly skilled professionals with incomes averaging over $90,000 a year. Even though there is little evidence yet that NGA scientists will choose to live in the area, many people are warning about gentrification.  In most of the neighborhoods around the new NGA headquarters, the housing market has collapsed. If you go to, you will find that there are almost no houses for sale in the immediate area around NGa and the few that are often sell for less than $50,000. If young professionals began moving in, there would be plenty of room for them in the vast tracts of vacant land around NGA (see Google Earth image of the area around NGA).

Google Earth: New NGA Headquarters (in yellow)

By contrast, the Central Corridor is booming with growth in medical, biotech, and various tech start-ups. My research on neighborhood change in St. Louis documents that there are, indeed, what I call “gentrification-like” processes going on. Young professionals who work in the Central Corridor are moving in to the Central Corridor and nearby neighborhoods to the south.

This trend, however, does not fit the classic definition of gentrification — understood as high-income households moving into poor, minority neighborhoods and rapidly displacing long-time residents. First, the neighborhood undergoing “gentrification-like” processes are not poor minority neighborhoods. Gentrification is almost completely absent north of Delmar. “Gentrification-like processes” are concentrated on neighborhoods in and just south of the Central Corridor that have always done relatively well.

Second, there is little evidence that “gentrification” is creating economic pressures that are rapidly pushing out large numbers of low-income residents. Indeed, our data shows that if you take out the citywide increases in rents, the rent burden actually fell in many of the “gentrifying” neighborhoods between 2000 and 2016. In 2016, the median rent in these communities was generally affordable to households making considerably less than the median income for the metropolitan area.

These neighborhoods have not become enclaves of rich white people. As the map shows, many of the most racially and economically diverse neighborhoods in the city are “gentrifying” neighborhoods in or just south of the Central Corridor. Indeed, according to a recent City Observatory report, the only census tract in the entire region that is both economically and racially diverse is Tower Grove East (seen in green on the map), a neighborhood just east of Grand Avenue that has witnessed an influx of young professionals.

Clearly, if I presented these findings to the activists in St. Louis who warn about gentrification, it would have absolutely no effect – and for good reason. Those who are worked up about gentrification in St. Louis don’t have some kind of conceptual or empirical confusion that social science research needs to straighten out. People are worried about gentrification for a reason. It is based on their lived experience.

In effect, “gentrification” has become the “g-word”–a kind of default term to express people’s anxieties about powerlessness and widening economic and spatial inequalities. Affordable housing is an issue in St. Louis, but the cause is not so much gentrification as stagnating wages and a private rental market that is not capable of providing decent housing for those at the bottom of the income scale. This is true everywhere – not just in St. Louis.

The forces behind these growing rifts are mostly impersonal and invisible, but neighborhood change, when it happens, is not. Not surprisingly, people focus on the things that they can see with their own eyes. Like many places, St. Louis is becoming a more divided city. Wages are stagnating and rents are rising across the region. Between 2000 and 2016, the median income of renters in St. Louis fell by more than 8 percent after controlling for inflation. The median income of homeowners rose slightly. (Calculated by Alan Mallach using data from U.S. Census, 1999 income, and American Community Survey data, 2012-2016). The Central Corridor is booming while the North Side continues to decline. People project these broader processes of division onto specific neighborhoods.

For the black community, concerns about displacement have a real basis in history. In the 1950s and 1960s, urban renewal and highway building forcibly displaced tens of thousands of African Americans. “Gentrification” is a shout out by people who feel they have little control over their lives and their neighborhoods.

Research shows that in cities across the nation, many more people live in neighborhoods that are going from low poverty to high poverty than from high poverty to low poverty (what we normally think of as gentrification). Especially in older industrial cities like St. Louis, the most pressing problem is not that higher income households are moving toward the poor and pushing them out – but that they continue to move away from the poor, leaving behind communities bereft of opportunities and resources.

We need a more nuanced vocabulary to discuss economically ascending neighborhoods. Right now, we have many neighborhoods experiencing “gentrification-like processes” but not exhibiting the massive displacement pressures roiling cities like Seattle, San Francisco, and New York. In St. Louis, over 90 percent of those who work in the Central Corridor live outside the Central Corridor. I wish many more of them lived in the Central Corridor or in other city neighborhoods. Of course, at some point in the future, this could happen and rents would then rise exponentially causing massive displacement. For this reason, we should act now, while land prices are still low, to preserve an adequate supply of affordable housing through land trusts, nonprofit-owned housing, and other methods.

Today, however, the big disruptive challenge facing older industrial cities like St. Louis is not gentrification but depopulation and disinvestment – not re-urbanization but de-urbanization. Contagious abandonment and the decline of solid working and middle-class neighborhoods are the most pressing issues facing St. Louis – not gentrification. At the same time, we cannot dismiss fears of gentrification. Ultimately, the only way to dispel those fears is to help city residents acquire political power, access to jobs in the new economy, and control over their own communities.

Editor’s note: This post has been revised to correct a reference to the home locations of those who work in the Central Corridor. (August 16).



Going the wrong way: NYC’s new ride-hailing cap

Capping or taxing ride-hailing services isn’t going to solve NYC’s congestion problem

New York’s City Council is moving ahead with a package of measures designed to cap the number of ride-hailed vehicles, like Uber and Lyft, as a way of addressing the city’s growing congestion problem.

This is both a mistake and a blown opportunity.  Just a few months ago, New York seem poised to do the one thing that we know would address congestion: price the use of its streets. That measure foundered, due in part to the ongoing feud between New York Governor Andrew Cuomo and NYC Mayor Bill de Blasio. Instead, its taking the politically expedient (but wrong-headed) position that ride-hailed vehicles alone are responsible for congestion, and that taxing them or capping their numbers will somehow resolve congestion problems.

In some respects, this re-capitulates the historical evolution of taxis in New York. Initially, taxis were un-regulated, and they flooded city streets.  That led the city to implement the medallion system which strictly limited taxi numbers for decades, and which led to medallion owners reaping huge profits from the limited supply. The entry of Uber, Lyft and other ride-hailing services effectively evaded the medallion cap, providing vastly more service (including to underserved areas of the city) and caused medallion prices to collapse.  Not surprisingly, yellow taxi owners and operators have been urging the city limit the number of ride-hailed vehicles.

While the medallion system was explicit about who had the right to operate in the city, we’ve have seen the details of how the city plans to allocate permission to operate ride hailed vehicles.  Will existing license holders be grandfathered? Like medallions, will their licensees capture the value associated with operating in the city?

It’s clear that ride hailed companies are being made the whipping boy of those alarmed about growing traffic congestion.  But that’s wrong, as Robin Chase, Paul Salama, and Charles Komanoff have all argued.  It’s not just ride-hailed traffic that causes congestion, its all traffic.

The cap on additional ride-hailed vehicles is unlikely to do anything to resolve the congestion problem. New York University’s Eric Goldwyn has a clear explanation as to why:  First, the cap doesn’t reduce the number of ride-hailed vehicles. Second, the problem isn’t just, or primarily due to ride-hailed vehicles.

The fundamental reason New York City’s street’s are increasingly clogged is because they aren’t priced appropriately. By giving this valuable resource away, essentially for free, we guarantee that it will be jammed to capacity. The only plausible solution is some form of congestion pricing, which would give users incentives to use other modes (like transit) and would also provide the city with the resources needed to make the transit system more frequent and reliable.

A recent report from Bruce Schaller makes much of the supposed burden ride-hailed cars impose by traveling empty in the city. The NYC proposal includes caps and utilization requirements designed to reduce this deadhead travel, but the most efficient way to encourage greater utilization is to impose a congestion fee. A car takes up the same amount of space on a city street whether it is carrying a paying fare or not, and whether it is a ride-hailed vehicle, a taxi or a privately owned vehicle. If you want to reduce congestion, the way to do it is to give all of them an incentive to use the street efficiently.

Caps and Regulation are a dead end policy approach

The big problem with the approach of capping and taxing ride-hailing services is it takes transportation policy down a dead end route. As we’ve noted with medallions, those who acquire property rights under a capped system have strong incentives to oppose any changes that would lower the value of their investment. A key part of the new regulatory system for ride-hailed vehicles will be an inherently complex system of “utilization regulations” designed to make sure that ride-hailed vehicles don’t travel too many miles without paying passengers. (Never mind that they have a strong financial incentive to get as many paid miles as possible, and that the city probably doesn’t care about deadhead miles at 5am as much as it does at 5pm). Creating one tier of regulations like the caps or utilization requirements, invites subsequent rounds of regulation to deal with the problems they create: already, according to Streetsblog, NYC is considering further regulations to deal with the likely problem that ride-hailed vehicles, if capped, will tend to concentrate their services in the most lucrative markets, effectively reducing access to under-served outlying boroughs and low income neighborhoods.

There’s another, more subtle effect of taking this regulatory route: politically, and in the public discourse, it reinforces the notion that ride-hailing is the cause of congestion. This framing invites continued constraints on ride-hailing when these measures don’t reduce congestion, and makes it that much harder to have a broader discussion about how to solve this problem.

We don’t get many opportunities to rethink the way we price and regulate the street system. Once we head down this road, it’s likely it will be another decade or more before we can have a serious discussion of congestion pricing, which is the only strategy that’s likely to make a meaningful difference.



Whither small towns? Wither small towns?

Rural and small town America faces some tough odds

In an article entitled: “How to save the Troubled American Heartland,” Bloomberg’s very smart Noah Smith shares his thoughts on how to revive the smaller towns of rural. For the most part, he’s in agreement with the ideas expressed by James and Deborah Fallows in their neo-Toquevilliean travelogue: Our Towns: A 100,000-Mile Journey into the Heart of America. The Fallows’ flew their small plane hither and yon across the nation, landing in the places that the pundits fly-over, and trying to glean signs of hope and lessons for success.

They visit towns such as Greenville, SC, and Allentown, PA and report that there are signs of life. Local leaders are valiantly working to revive their economies. And at least a few places are generating jobs, income and economic revival.

The key lessons, as suggested by Smith:  local universities can train kids to succeed in the new economy and help create new jobs. In addition, communities ought to encourage immigration and look to form public-private partnerships.  These are all plausible suggestions as to what small towns and rural areas ought to try. The problem with extracting “best practice” lessons from these relatively successful places is that it doesn’t answer the question of whether the strategy is replicable in different places or scales to all of rural America.

While essays like Smith’s and books like the Fallows’ treat the issue with a broad brush, the question is not what happens to the entire “heartland.” Some places, even rural, isolated ones will flourish, even if rural America as a whole is challenged or in outright decline. The bigger question is whether this is an exercise of sweeping rejuvenation, or something more like battlefield triage. The unstated objective of most rural economic development efforts to to restore some real (or imagined status quo ante, recapitulating a pattern of population and economic activity that existed in some prior year (usually an historic peak).

You can point to some rural and smaller metro places that are doing well. Smith presents this optimistic looking chart showing five smaller communities that have shown real growth since 2000.

But when we look at the entire nation, grouped by population size, it’s apparent that big cities are powering national growth, and that as a group, small town and rural America is getting left further behind. As we’ve argued at City Observatory, much of this has to do with the strong connection between cities, talent, productivity and knowledge based industries. Smart people and the industries that employ them are increasingly gravitating toward urban locations, and are more successful there than in rural ones.

As a group, the nation’s non-metropolitan areas had not fully recovered from the Great Recession as of September 2016, according to data compiled by the Oregon Office of Economic Analysis. Although larger metros experienced proportionately larger job losses in the recession, they’ve collectively grown faster in the recovery, with metros of a million or more population growing almost twice as fast as smaller ones.

There are some stellar examples of successful rural places. Usually they’ve parlayed a signature asset, like an outstanding university or great quality of life into the nucleus of a new economic engine. But these tend to be exceptions.

While this has worked and will continue to work in some rural communities, it’s far from a general prescription for success. The fact that a few rural communities succeed doesn’t mean than these tactics will work everywhere.  Noah Smith is reminded of the key role that Pike Powers played in organizing such a partnership in Austin.  But Austin was also capital of a large state, an established tech center and home to a huge university. (By the way: every economic development effort I’ve ever seen, good, bad or wildly unsuccessful bills itself as a public-private partnership).

Much of the challenge in rural America is about a re-shaping of economic activity.  Key service functions are becoming more centralized as the somewhat larger towns emerge as the medical center, the retail center or the university town for a rural region. The only thing worse than having the WalMart in your town is having it in a neighboring town. There are winners and losers in the rural landscape as this process unfolds.

Having a university is a valuable asset, but essentially we’re not building any new ones, especially in small rural communities.  The same holds for having the region’s principal medical center. So if you’ve got one of these, great:  If not, expect to see economic activity, kids, opportunity and entrepreneurship gravitate to the places that do.

What’s odd is that Smith’s telling overlooks some of the principal forces underpinning the economies of rural and small town America:  cheap housing, social security and farm subsidies. Even as urban areas become more congested and expensive, housing in rural places is affordable. The foundation of a growing number of rural economies is the steady flow of funds from Social Security and Medicare, both to a rural region’s retirees, and to other retirees they might attract from expensive cities. Medicare (and Medicaid) revenues come disproportionately from cities and are spent disproportionately in relatively rural areas (which are older and poorer). And while the federal government has nominally eschewed most industrial policies, it continues to prop up agriculture through a combination of farm subsidies and support for infrastructure projects, like irrrigation and low cost power.  In effect, we do have a federal policy for rural America-even if we don’t acknowledge it as such. The likely cuts to these and other federal programs necessitated by the recent tax cuts and ballooning federal deficit are likely to have a disproportionate effect on rural America.

There will continue to be occasional bright spots in the rural economy, but in most places, its beyond their means to think about job creation on a scale that’s going to reverse the slow process of decline.

Don’t decry new urban housing as “gentrification”

Whenever a distressed neighborhood gets new market rate housing, someone’s bound to cry “Gentrification”.  Here’s why that’s wrong.

This is a guest post from Jason Segedy, Director of Akron’s Planning Department.  This is an excerpt of a two-part essay written by Segedy and published in The American Conservative. Jason writes about a range of urban issues at his excellent blog “Notes from the Underground.”

Residents and community activists who are opposed to new housing often demonize the real estate development profession as being “greedy”, overlooking the fact that their own home was developed by a developer, built by a builder, and sold by a realtor – most likely for a profit. This isn’t to argue that every development professional is a white knight, but it is important to remember that the vast majority of people who work in the real estate and construction sectors are not the enemy of neighborhoods.  Without them, there would be no neighborhoods.

When new housing is proposed, many people who don’t even live in the neighborhood will come out of the woodwork to oppose it.  Sometimes they provoke class conflict by seizing upon the dubious marketing term “luxury housing”.  This is especially true of far-left activists and academics, many of whom reflexively label any new development in a lower or middle-income neighborhood as “gentrification”.

The only reason that I continue to use that word is because other people use it. But it has become a useless word, and means so many different things to so many different people that it is no longer of any descriptive value.

People who describe any new market-rate housing as gentrification need to get their story straight, because they often have contradictory, and self-refuting notions about urban redevelopment.  They may claim to want to see economic and racial integration, but oppose the new housing in lower-income or minority neighborhoods which could actually bring it about.

I have heard self-described gentrification opponents claim that a new housing development will be home to people who “do not look like” those who live in the neighborhood, while simultaneously calling for diversity and inclusion; and remaining blissfully unaware of the irony that they are the ones who are making stereotypical, unwarranted assumptions about the characteristics of those who will buy the new houses in the first place.

Outside activists who come into a neighborhood and invoke the bogeyman of “gentrification” are denying opportunity to nearby residents who could benefit from the new housing, discouraging private investment in places that need it, and serving to further urban decline in the community.

A recent article in The Economist describes this dynamic well:

Those who bemoan segregation and gentrification simultaneously risk contradiction…[Gentrification] boosts racial and economic integration. It can dilute the concentration of poverty—which a mountain of economic and sociological literature has linked to all manner of poor outcomes…Gentrification steers cash into deprived neighbourhoods and brings people into depopulated areas through market forces, all without the necessity of governmental intervention.

There is little empirical evidence that gentrification has led to the displacement of people living in the urban neighborhoods of the Rust Belt.  On the other hand, there is incontrovertible proof that thousands of middle-class residents are displaced by urban decline every year in these cities.  A shrinking city, with a declining tax base, that is getting poorer will help no one – the poor least of all.

Alan Mallach, in his new book The Divided Cityexplains, at length, why the changes occurring in a handful of gentrifying neighborhoods in Rust Belt cities, are on balance, good for these places, and also makes the point that these positive changes are dwarfed by the economic and social decline that is happening elsewhere in these cities.  We need more gentrification in the Rust Belt (if you insist on calling it that), not less.

More evidence of declining rents in Portland

Zillows data shows Portland rents have dropped 3.5 percent in the past year

A couple of weeks ago, we published the latest data from on the decline in rents in the Portland metropolitan area.  Their benchmark series for one bedroom apartments showed a year-over-year decline of 3 percent in Portland.

While we have a high regard for’s methodology (unlike some other sources), it’s always useful to look at a wide range of indicators data to verify what’s happening in the market. Accordingly, we’ve taken a look at the Zillow Rent Index for multi-family housing for the City of Portland. Zillow tracks rents and home values across the nation, and produces monthly estimates of current market prices. Their methodology is specifically designed to avoid the composition effects that can bias other estimates of current rental rates. Here’s what their latest data show:

Year-over-year (from June 2017 to June 2018) apartment rents in the city of Portland, measured by the Zillow Rent Index declined 3.5 percent. That’s very close to’s estimate of a 3.1 percent decline for one-bedroom apartments.


According to Zillow’s historical time series, Portland apartment rents peaked in April 2016, and have declined a cumulative 9 percent since that time. The median rent peaked at about $1,660 but has since declined about $150 to $1,509. Zillow’s data confirms that Portland experienced double- digit rent increases in 2016, but the decline since then is consistent with economic theory (a growing supply of new apartments has blunted rent increases) and our observations (a flowering of “For Rent” signs in Portland neighborhoods).

If you really care about housing affordability, the Zillow data (and the remarkably similar data) are what you ought to be paying attention to:  they show the rents Portlanders are paying in the current marketplace. These two independent data sources show that Portland’s housing is becoming more affordable. As we’ve explained at City Observatory, the declines we’re now seeing have a lot to do with the supply of new apartments finally starting to catch up with demand. If we want to see further improvement in housing affordability in Portland, we really ought to be looking for ways to encourage further increases in supply.


We disagree with the Washington Post about housing economics

Contrary to what you think you may have read in last week’s Washington Post, rental housing markets at all levels still conform to the laws of supply and demand

Monday’s Washington Post ran a provocative headline: “In expensive cities, rents fall for the rich–but rise for the poor.”  Citing data from one of the nation’s most authoritative real estate analytics firms, Zillow, it claims not only that rents are falling for the rich while rising for the poor, but also implies that building new apartments (chiefly at the high end) hasn’t done anything to help affordability.

It’s a classic “rich get richer, poor get poorer” story.

But in our view, it’s not quite right.

We follow these same data sources and housing markets closely, and what the article claims doesn’t square with our observations. That’s particularly true for the Portland market, which the Washington Post uses as a signal example of its claims.

As we explain below, in our view the story’s key claims can’t be supported by the data presented, and it presents a distorted and incomplete view of what’s happening in rental markets around the country.  When examined closely, it’s clear that increased supply–even at the high end of the market–is bringing down rental inflation in all segments of the rental marketplace.

For technical reasons,  the Zillow data series presented in this story really can’t be used to convincingly answer the questions posed.  For example, the “three tier” data relied upon in the Washington Post story don’t include any of the nation’s apartments; and in fact they’re made up primarily of estimates of how much single family homes would rent for, if they were rented (which most of them aren’t). In addition, the three tier data is constructed in a way that is subject to composition effects (the composition of each tier changes over time as new housing is built, and this can bias upward estimates of inflation in the lowest tier if housing is added primarily in upper tiers.)

Looking more closely at Portland, the city profiled in the Washington Post story, we find that there’s actually a very strong correlation between changes in rents in the high, medium and low tiers of the market. (Conversations with Zillow staff suggest that the connections between the high and low end of the market are actually especially strong in Portland). What that means is that rent inflation has closely followed the same path in all market segments. The Post story makes much of the fact that rents for the least expensive housing tier rose 42 percent from 2011 to 2018; but doesn’t tell readers than in the same time period rents for the middle tier rose 40 percent, and rents for the highest tier rose 33 percent.  Rents have gone up for everyone.

More importantly, as new housing supply has come on line in the past two years, rental inflation has declined sharply in every segment of the Portland housing market, according to Zillow’s data (again, the series that leaves out apartments).  As we reported earlier at City Observatory, Zillow’s separate apartment data (not stratified by price tier) show a 3.5 percent decline in apartment rents in the City of Portland since June 2017.

Taken as a whole, Zillow’s data shows housing markets, particularly Portland’s working in much the way that standard economic theory would suggest.  As new supply comes on line, rental inflation abates. And while increases may initially be stronger and more pronounced in one segment (like high priced dwellings), the prices and rents in different segments of the market tend to move in parallel to one another.  And the observation that rental inflation falls first and fastest in the market segment in which new units are being completed is fully consistent with economic theory.

In our view, the Washington Post’s story would have been much more accurate if it had said the following:

  • Estimated rents for housing (excluding apartments) are up across the board in Portland since 2011, and may have risen somewhat more for low and middle price tiers than for the highest price tier.
  • Monthly year-over-year price changes in the high, middle and low tiers of the Portland housing market are strongly correlated, suggesting that prices changes in one segment of the market affect other segments and/or that all segments are affected by common factors.
  • Over the past two years, rental price inflation in Portland has decelerated sharply in all segments of the housing market as large amounts of new housing, especially apartments, have been completed.
  • City of Portland apartment rents have now declined 3.5 percent over the past 12 months.

We’re grateful to Washington Post reporter Jeffrey Stein for answering our questions about his story, and to Zillow’s Matt Kreamer and Aaron Terrazas for engaging in a detailed discussion of the construction and interpretation of Zillow’s data (which we regard as some of the best available). The opinions expressed here are ours, not theirs.

The limits of Zillow data:  “All Homes/Three Tiers” omits apartments

Zillow gathers and publishes lots of data about housing, but the particular data series that the Washington Post used for its analysis leaves out estimates of the rents for apartments. As a result, may be invalid to make claims based on this data about whether rents have increased, decreased or stayed the same for the poor.  Here are the wonky details.

The Washington Post article  relies on a Zillow data series called the “ZRI All Homes 3 Tier” estimates. A little explanation is in order. If you’ve visited the Zillow website, you know that the company publishes, and regularly updates its estimates of the current market value of nearly all the single family homes and condominiums in the US.  It calls these its “Zestimates.” Based on an analysis of the home’s characteristics and the sales prices of similar properties in the area, Zillow’s model estimates how much that house would sell for this month. The critical input into the Zillow model is data on publicly recorded sales of single family homes, condominiums and cooperatives.

To track different levels of the “for sale” market in each geographic area it covers, Zillow divides these properties up into three equal groups, called “Tiers”.  They are ranked by price, with the most expensive third in the top tier, the middle third in the second tier, and the lowest priced homes in the bottom tier.  The critical thing about the three tiers is that they are based on those properties (single family homes, condos and coops) for which individual unit sales transaction data are available.  This data series, somewhat misleadingly labeled the “All Homes” data does not include apartments. See the methodology section of the Zillow website.

In addition to the home value Zestimate, Zillow has estimated how much these same houses would rent for, if instead of being offered for sale, were offered for rent.  These “Rent Zestimates” are prepared for all houses, including owner-occupied ones that aren’t listed for sale.  So “All Homes, 3 Tier” ZRI estimates reported in this series do not include the nation’s for-rent apartments.   What the three tier rental data show is variation only within the single family, condo and coop portion of the market, and not the rents of multifamily apartments.  In addition, a majority of the homes in these three tiers are owner-occupied, not rented, so the three tier data don’t reflect what renters are actually paying.

This is problematic because it is exactly this three-tier All Homes data that Jeff Stein is relying on to draw conclusions about the variation in trends by income groups.

Poorer city residents have experienced significant rent increases over the past several years. In Portland, average rents for the poor have risen from about $1,100 to $1,600 — or by more than 40 percent — since 2011.

As we’ve noted, the cited data don’t support this claim because they leave out all multifamily apartments. Also:  Its misleading to cite this 40 percent increase without saying what the estimates were for the other tiers for the same time period.  Between 2011 and 2018, Portland All Homes” ZRI in the lower tier rose 44 percent, in the middle tier rose 42 percent and in the highest tier rose 33 percent.  If you take that longer term view, the data (even ignoring its limitations) don’t support the headline claim that rents went up for the poor and down for the rich.  They went up for everyone.  (In addition, as we explain below, there’s one other technical problem that confounds using the three tier data for making claims about relative price inflation: they are subject to composition effects as the homes in each group change over time).

Rents move in tandem across all tiers

A central premise of the article is that cities like Portland have futilely pursued building higher income apartments which have essentially no effect on the rents paid by low income renters.  Despite its limitations, let’s take a closer look at the ZRI index for the three tiers to see what it shows.  The following chart shows the “All Housing” data for the Portland Metropolitan area.

This chart shows the year-over-year monthly monthly inflation rate based on ZRI data for each of the three tiers for the period 2011 through June 2018.  As you can see, these lines move very closely together, with similar seasonal and cyclical patterns. The high, middle and low tiers all record both peaks and troughs at very nearly the same time.  The highest tier appears to be somewhat more volatile than the others, the highest tier has higher inflation in the peaks and lower inflation in the troughs.  Statistically, the correlation between the monthly year-over-year price changes between the highest and lowest tiers is .86, meaning that they are either closely inter-related or determined by common factors. These data do not square with the article’s headline claim that rents are increasing for low tier houses and decreasing for high tier houses. From mid-2015 onward, rental inflation is heading down across the board.  (It’s possible to cherry-pick one or two months of data at the point where low tier inflation dips below zero to claim that in that time period rents went down in the low tier and up in the other tiers, but that ignores the strong parallel downward trend in all three tiers.)

The key point, though, for the purposes of this article is whether a decline the the estimated rents of higher end homes was associated with a decline in the estimated rents of lower end homes.  And that’s exactly what the data show:  Since the peak in 2015, the year-over-year rental increase for the highest tier has fallen from 15 percent to -5 percent, and for the lowest tier it has fallen from 10 percent to 2 percent.

Finally, although we believe these data are too incomplete to make such a determination, the greater volatility of higher tier properties (especially their steeper decline) is actually consistent with economic theory:  If we’re adding units at the high end of the market, they will be closer substitutes for other high end properties than they are for lower end properties, and so the price effects of adding supply will be felt first and most in the highest tier.  But as the strong parallel trajectory of these three tiers shows, adding supply in the high tier drives down rent inflation in the highest tier, and also in the middle and lowest tiers.  Far from disproving the economic prediction, these data (though limited) are consistent with it.

More Supply = Lower Rent Inflation

There’s a very simple way to make sense of what’s going on in Portland’s housing market.  Look at how many new apartments are built and what subsequently happens to rents.  In the wake of the Great Recession, apartment building in Portland fell off a cliff. Metro Portland had been adding 4,000 to 5,000 apartments per year, and that fell to just a 1,000 annually in 2009 and 2010.

(Barry Apartment Report, Spring 2018)

Portland’s economy rebounded quickly, but apartment construction did not.  As the building permit data show, it took several years for things to get going (and typically there’s a year or more between the time a permit is issued and a project is ready to rent).

In 2015, Portland experienced very low vacancy rates and double digit rent increases.  But since then, as more units have been built, that pressure has eased. And its apparent that rents are declining.  Zillow’s data on apartment rents have shown negative year over year changes for more than a year.

From 2011 through 2014, as supply dried up, rents went up, peaking at double digit levels in 2015.  When supply finally caught up, rent inflation declined, and over the past year, rents went down. As we’ve frequently stressed, these things play out with a lag.  During the recession, when new apartment construction slowed to a trickle, it took a while for vacant apartments to fill up and put pressure on rents. Similarly, once rents spiked, it took a while for new units to come on-line and exert downward pressure on rents.  (Editor’s note: the data in this chart are for the City of Portland).

A technical note:  The composition effect

Interpreting the Zillow All-Homes 3-tier rental data is complicated by a subtle problem called the composition effect. Zillow creates its three tiers by dividing its all homes database into three equal parts, and then computing the average prices for each part. The problem arises because the number of units in the database, and in each tier changes over time. If we add units primarily in one tier and not the others, some of the shift we observe in prices among tiers is due to the change in the composition of the tiers, rather than a change in the prices of individual houses.  This is particularly a problem based on the observation made in the Washington Post story that much new housing has been added in the top tier of the market. Adding more homes to the top tier causes a shift in the composition of the lower tiers, which, independent of any price changes tends to push the average of lower tiers upward.  The problem for the Washington Post story is that we don’t have a good way to know how much the composition effect is driving the observed changes in prices among different tiers.

Editor’s note: The original version of this post was changed to correct a word transposition in the first paragraph.


Philadelphia’s urban policy harmonic convergence

Philly’s University City: The urban challenge in a nutshell

The knowledge economy . . . tax breaks . . . NIMBYism . . . gentrification . . . Amazon’s HQ2 . . . high speed rail . . . university economic development?  All this in one location.

Many conversations about the nation’s urban challenges address individual issues as if they were separate and distinct from one another (zoning, transportation, housing affordability, economic development, etc). But in reality, a careful analysis will show that these things are often inextricably interrelated.  There’s one neighborhood in Philadelphia that brings all these strands together. This particular story has almost all of our favorite urban themes wrapped in a single bundle, from urban revitalization and anchor institutions, to gentrifying neighborhoods, to NIMBYism, to town-gown conflicts, to convoluted local development approval processes and tax breaks with just a hint of Amazon HQ2 thrown in for good measure.

PlanPhilly’s Jim Saksa has a long, but worthwhile read exploring the redevelopment of University City, the area near the city’s 30th Street station and adjacent to the Drexel University and University of Pennsylvania.  His article “As Drexel transforms University City, communities nearby prepare for gentrification” is available both in text and as a public radio podcast, courtesy of NPR station WHYY.

Big plans are afoot for redeveloping the area, and they raise, in a geographic nutshell, many of the current controversies in urban policy. Brandywine Realty is proposing “Schuylkill Yards”-a $3.5 billion (yes, with a “b”) redevelopment of 14 acres, to include millions of square feet of office space. The project has immediate access to the Northeast Corridor, is directly across the river from Center City Philadelphia. The site is short-listed as a potential location for Amazon’s HQ2, should it come to Philadelphia. However you like to characterize urban development strategies (tied to great transit, redeveloping brownfields, eds and meds, innovation district) you’ll find it here.

You’ll want to read the entire article, but here’s a short litany:

Growing enrollments, especially at Drexel, are bringing lots more young adults to the area, with housing demand spilling over into adjacent neighborhoods. The universities have built more student housing, but its expensive, and students looking for more affordable options are bidding for apartments in nearby neighborhoods.

The neighbors, concerned about rising rents, home prices and displacement, have persuaded the city (with the university’s support) to downzone many of these neighborhoods. While perhaps good for incumbent homeowners, this step further constrains the supply of housing and is actually likely to further push up land prices for those parcels eligible for greater density.

Philadelphia, with its famously arcane system for building and development approvals is likely partly responsible for driving up the cost of development. Local builders complain, but acknowledge that the opaqueness of Philadelphia’s process is a profitable competitive advantage over outsiders. One of the developer’s explains:

In fact, Sweeney argues that Philadelphia’s difficult building regime even makes Brandywine money, because it’s both an investor and a developer, and the city’s idiosyncrasies keep competitors away.

“So, while you might be frustrated as a developer, as an investor, you look for markets where it’s harder to build.”

Tax policy also figures prominently in area development plans. The entire site in question has been designated one of Pennsylvania’s “Keystone Opportunity Zones” which grant generous state and local tax breaks to new job-creating investment.

Urban redevelopment is never simple, and there are lots of moving and interacting parts to consider. This close look at how these difference forces are playing out in Philadelphia’s University City provides some keen insight in to these relationships.

(Hat tip to Jon Geeting for flagging this article on twitter, and pointing out the salient bits).

This post has been revised to correctly spell Jim Saksa’s name, and credit WHYY’s PlanPhilly for the original story and broadcast.

IoT: The Irrelevance of Thingies

People and social interaction, not technology, is the key to the future of cities

Smart city afficianado’s are agog at the prospects that the Internet of Things will create vast new markets for technology that will disrupt and displace cities. Color us skeptical; our experience with technology so far–and its been rapid and sweeping–is that it has accentuated the advantages of urban living and made cities more vital and important. From the standpoint of urban living, one should regard “IoT” as “the irrelevance of thingies.”

Smart cities will be bathed in beams of light! (TechCrunch)

It’s been more than two decades since Frances Cairncross published his book “The Death of Distance” that prophesied that the advance of computing and communication technologies would eliminate the importance of “being there” and erase the need to live in expensive, congested cities. (It goes down, along with Francis Fukuyama’s “End of History” and Kevin Hassett’s “Dow 36,000” as one of the demonstrably least accurate book titles of that decade.)

Back in the 1990s, when the Internet was new, there was a widely repeated and widely accepted view of the effect of technology on cities and residential location. The idea was “the death of distance”–that thanks to the Internet and overnight shipping services and mobile communications, we could all simply decamp to our preferred bucolic hamlets or scenic mountaintops or beaches, and virtually phone it in.

And these predictions were made in an era of dial-up modems, analog cell-phones (the smart phone hadn’t been invented, and while Amazon was still making most of its money cannibalizing bookstore sales). We were all going to become “lone-eagles”, tipping the balance of power away from cities and heralding a new age of rural economic development. Here’s a typical take from 1996, courtesy of the Spokane Spokesman Review:

Freed from urban office buildings by faxes, modems and express mail, lone eagles are seen by economic development experts as a new key to bolstering local economies, including those of rural areas that have been stagnant for much of the century.

Faxes?  How quaint.  Since then, of course we’ve added gigabit Internet and essentially free web conferencing and a wealth of disruptive apps. But despite steady improvements in technology, pervasive deployment and steadily declining costs, none of these things have come to pass. If anything, economic activity has become even more concentrated. Collectively, a decade after the Great Recession, the nation’s non-metropolitan areas have yet to recover to the level of employment they experienced in 2008; meanwhile metro areas, especially large ones with vibrant urban cores, are flourishing.

The economic data put the lie to the claim that cities are obsolete.  One of our favorite charts from Oregon economist Josh Lehner points point that larger metropolitan areas have outstripped smaller ones and rural areas have continued to decline in this tech-based era.

But what’s more than this is the growing premium that people pay to live in center of cities. Economists call this the urban rent gradient: the price of housing is more expensive in the center of regions, which are generally the most convenient and accessible to jobs, amenities and services. Over the past two decades, the urban rent gradient has steadily grown steeper: people now may more to live in the center of cities than ever before.

Edlund, Machaco, & Siviatchi (2015)

The importance of this trends was identified by University of Chicago economist Robert Lucas writing in the late 1980s. His words are even truer today that they were then:

If we postulate only the usual list of economic forces, cities should fly apart. The theory of production contains nothing to hold a city together. A city is simply a collection of factors of production – capital, people and land – and land is always far cheaper outside cities than inside. Why don’t capital and people move outside, combining themselves with cheaper land and thereby increasing profits?  . . . What can people be paying Manhattan or downtown Chicago rents for, if not for being near other people?

The expanding power and falling price of computation and electronic communication has made these things not more relevant, but less relevant to location decisions. Because you can get essentially all these things anywhere, they make no difference to where you locate. What’s ubiquitous is is irrelevant to location decisions.

The growing ease and low cost of communication has, paradoxically made everything else relatively more important in location decisions. What’s scarce is time and the opportunities for face-to-face interaction.

Both in production and consumption, proximity are more highly valued now than ever. Economic activity is increasingly concentrating in a few large cities, because they are so adept at quickly creating new ideas by exploiting the relative ease of assembling highly productive teams of smart people. Cities too offer unparalleled sets of consumption choices close at hand. From street food, to live music, to art and events, being in a big city gives you more to choose from, more conveniently located and cheaper than you can get it anywhere else. Plus cities let you stumble on the fun: discovering things and experiences that you didn’t even know existed.

The “death of distance” illusion is being repeated today with similar claims about the impending disruption from “the Internet of Things.” On a municipal scale this manifests with the cacophony of visions for tech-driven smart cities. Supposedly attaching sensors to everything from cars, to streetlights, to water meters is going to produce a quantum leap in city efficiency.

Far from provoking a shift to the suburbs and a decline of cities, the advent of improved computer and communication technologies has help accelerate the revival of urban work and living. In recent comments to the Urban Land Institute’s European meeting in the Netherlands, Harvard Economist Ed Glaeser explains.

“So why didn’t computers kill cities?” Glaeser asks. It’s a fair question: remote working has never been more possible, and productivity has never been higher. So why are more people flocking to urban areas than ever? “Cities are about exchanging ideas,” he says. The proximity of people to other people sparks ideas in a way that is impossible in remote areas.

As Philip Longman wrote at Politico a few years back:

. . . in the centers of tech innovation, . . . the trend has been toward even greater geographic concentration, as Silicon Valley venture capital firms such as the storied Kleiner Perkins Caufield & Byers have set up offices in downtown San Francisco, closer to the action. Apparently, there is no app that will bridge the gap. To seal the deal, you must be in the room, literally, just like some tycoon from the age of the robber barons.

As technology becomes cheaper and more commonplace, it ceases to be the determining factor in shaping the location of economic activity. All the other attributes of place, especially human capital, social interaction and quality of life–the kinds of things that are hardest to mimic or replace with technology, become even more valuable. To be sure, the Internet of Things may disrupt some industries and promote some greater efficiency, but the arc of change is moving inexorably to the city.

Do rich neighbors make low income people unhappy?

Lower income households are happier in higher income neighborhoods

How does your neighbor’s income affect your happiness? Do you feel worse off if you have less income than most of your neighbors? The “Keeping up with the Joneses” theory of happiness would suggest that if people live in a neighborhood with people who have more income than they do that they will be less happy than if their neighbors have a similar income.

One indisputable fact about gentrification is that it brings inequality into sharper focus: low income people suddenly find that they have higher income neighbors. Anecdotally, we’re told that this increases social and cultural friction, and we might think, lowers the well-being of long time, lower income residents. While that’s plausible, and certainly true for some individuals, does having high income neighbors generally make lower income people less happy?

Economists Abel Brodeur and Sarah Flèche have a new paper that looks at the self-reported happiness of residents of different neighborhoods to assess how neighborhood income affects well being. This work is part of the burgeoning field of happiness research that uses survey data to tease out the connections between a range of socioeconomic factors, behaviors and experiences and life satisfaction. Their study uses data from the Center for Disease Control’s Behavior Risk Factors Surveillance Survey (BRFSS), looks at variations in reported well-being at the zip code level.

The study finds that regardless of household income level, people who live in higher income neighborhoods are, on balance, happier than otherwise similar people who live in low income neighborhoods. This result holds after controlling for other factors that have been widely shown to affect happiness (age, income, educational attainment, etc).

And the effect is sizable: an increase in the average income of one’s neighbors has about one-quarter the positive effect on self-reported well-being as an equivalent increase in one’s own income.  Put another way: an increase in your neighborhood’s average income of 40% has roughly the same impact on your self-reported well being as an increase of 10% in your own household’s income.

The effect is actually stronger for households with lower incomes. For low income households, an increase in average neighborhood income increases self-reported well-being by about one-half of the positive effect of an increase on one’s own income.

There’s no question that neighborhood change often produces cultural conflict, as the habits and expectations of newer residents clash with those of long-time residents. But acknowledging that fact does nothing to answer the question of whether, on net, people are happier if their neighbors are better off. (Admittedly, this study is cross-section rather than looking at change over time, so it may not fully answer this question).

So why don’t people with richer neighbors feel worse, as the “Keeping up with the Joneses” theory predicts? Brodeur and Flèche find that it isn’t the rich neighbors that influence people’s well-being directly, but rather that overall neighborhood amenities are higher:

. . . richer ZIP code neighbors have a positive effect on residents’ well-being not because they are rich per se but because they bring public goods and amenities that are valuable

On balance, the positive effects of living in a nicer neighborhood (less crime, lower unemployment, more job opportunities and so on) appear to more than offset any sense of relative deprivation from having higher income neighbors.

This study provides some indirect support for the importance of neighborhood level economic integration, especially in terms of its benefits for the subjective well-being of poor households. It’s broadly consistent with the findings of Jacob Vigdor’s study of gentrifying neighborhoods that the improvements in the quality of the neighborhood more than compensated for the higher rents that residents had to pay.

Neighborhood change is always hard. For many neighborhood residents, an influx of new population will affect their sense of well-being, and for some these effects will be negative. But these data suggest that on balance, an influx of higher income people into a low income neighborhood, aka gentrification, is likely to be associated with increased levels of self-reported well being for lower income residents.

Abel Brodeur (Department of Economics, University of Ottawa, Ottawa, ON) ; Sarah Flèche (Aix-Marseille School of Economics, Marseille, France), Neighbors’ Income, Public Goods and Well-Being, December, 2017.



The Week Observed, August 31, 2018

What City Observatory did this week

1. If you want less displacement, build more housing. A common refrain at planning commission meetings around the country is that cities ought to block new housing as a way of insulating neighborhoods from change and displacement. The irony of this position is that making it harder to build new housing in the face of increased demand for urban living means that rents are bound to go up, and displacement is actually likely to accelerate. While giving low income neighborhoods the same kind of NIMBY powers that have enabled higher income neighborhoods the ability to block new housing seems like a viscerally satisfying equity move, in reality its counterproductive.

2. Does having wealthier neighbors make low income people unhappy? There’s a burgeoning field in economics called happiness research, which looks at the statistical correlations between self-reported life satisfaction and a wide range of personal, social and economic characteristics. A new study looks specifically at the connection between one’s neighborhood income and happiness. It finds that people who’s neighbors have higher incomes are on average happier. This effect is largest for low income households: the level of neighborhood income has about half as much power in explaining their happiness as does their own level of income. It’s a strong indicator that mixed income neighborhoods improve well-being–especially for low income households.

3. About that Federal Reserve study on housing supply and affordability. A few weeks back, there was considerable media attention to a study from two economists that suggested that in some instances, building more housing didn’t seem to have much effect on lowering rents. But a closer look at the study shows that its findings have at best, limited policy relevance.  The study looks only at a small increase in housing in higher income neighborhoods, and critically, assumes that any increase in housing supply is exactly offset by an increase in population. This assumption in effect precludes finding any positive effect from growing housing faster than population–which is the most important channel by which supply reduces rents. We summarize our analysis, as well as comments from Brookings economist Jenny Schuetz.

Must read

1. Why do we continued to be surprised when neighborhoods gentrify? Our good friend Daniel Kay Hertz has a commentary at Belt Magazine on the surprising persistence of common gentrification narratives. Sure, those young people are moving to sketchy urban neighborhoods today, but as they age, they’re going to move to the suburbs like everyone else. In a preview of his forthcoming book “The Battle for Lincoln Park,” Hertz shows that exactly that same story was told in the 1970s. While gentrification often is portrayed as a recent and perhaps temporary phenomenon, it’s actual a perennial aspect of the dynamic urban landscape.

2. More evidence of rent declines. ApartmentList tracks apartment rents in cities around the nation and reports year-over-year rent declines in 23 of the nation’s top 100 markets. They estimate Portland rents are down 2.6 percent in the last twelve months, in line with estimates from other sources, including Zillow.  The key reason for the decline: a surge in new apartment completions in the past year.

3. Is Scootergeddon overblown? NYC’s BikeSnob travels to Portland to fact check all the hype about e-scooters, and . . . is not displeased. There’ve’ been wild tales of how e-scooters from companies like Bird, Lime and Skip have instantly become menace to travel and public health, and a scourge of the public realm. Eben Weiss, aka NYC’s acerbic BikeSnob, traveled to Portland, which has had e-scooters for a little over a month now. His first hand reporting found none of the much-repeated tales of mayhem. The biggest problem was finding a fully charged scooter late in the day. His verdict:

. . . having now experienced shared electric scooters I can’t imagine not being thrilled by the potential inherent in a swift, cheap, and efficient form of transport that’s compatible with any wardrobe. They seem about as controversial to me now as escalators.

New Knowledge

Brown and Black.  Neighborhood change in south Los Angeles.  The USC Dornsife Center has a fascinating and comprehensive look at the last several decades of demographic change in south Los Angeles. The study, entitled “Roots|Raíces: Latino Engagement, Place Identities, and Shared Futures in South Los Angeles” looks at the steady transition of these neighborhoods from primarily African-American to increasingly Latino. The shift is vividly illustrated in this series of maps; the top row shows 1970; the bottom, 2010; green tracts are predominantly African-American; brown tracts are predominantly Latino.

In addition to a solid statistical analysis, there’s great qualitative insight on the way in which the neighborhoods have addressed cultural conflicts and constructed new neighborhood identities, that, for the most part, embrace the area’s diversity.

In the News

Writing in Planetizen, Todd Litman takes a look at the literature on housing filtering; how, as housing ages, it tends to depreciate in value and be occupied by successively lower income households, and cites our recent analysis of the role of new housing in reducing displacement.

City Observatory director Joe Cortright is quoted in Willamette Week’s story, “Portland is in the midst of a hotel-building spree. Can visitors keep up?”  It’s all about supply and demand; as the article relates “hotels are subject to the same market forces that caused Oregon cannabis prices to plummet and Portland apartment developers to offer two months’ free rent to prospective tenants.”


The Week Observed, August 24, 2018

What City Observatory did this week

1. Philadelphia’s urban policy harmonic convergence. The proposal to build a multi-billion dollar expansion of University City adjacent to Drexel University and Philadelphia’s Center City brings a host of urban issues to the fore all in one small location. The city is making a big bet on knowledge-based economic development (hoping to possibly attract Amazon’s HQ2).  Its bumping up against a host of well-known urban problems, from a growing demand from young adults for urban housing, to concerns about displacement. Many of these challenges are closely inter-related, but seldom so dramatically in such a small space.

2. IoT: The irrelevance of thingies. Plenty of pixels are being spilled over the prospect of Smart Cities, and the idea that new technologies will disrupt or replace key elements of urban living. We’re skeptical. The evidence from the past two decades of technological change has been that better communication and faster computing have simply made proximity and personal and social interaction even more valuable than before. The much heralded “death of distance” hasn’t happened, and as we’ve documented at City Observatory, people, especially well-educated young people are increasingly concentrating in cities–and are paying a premium for the opportunity to do so. That makes it unlikely in our view that the next round or two of technological innovation will put a dent in the increasing economic and social importance of cities.

Smart cities will be filled with bright lights!

Must read

1. California’s Neo-Feudal property tax system. In California, thanks to Proposition 13, the tax you pay depends on when you–or your parents–bought their home. Prop. 13, limits assessment increases on property to 2 percent per year, and reassesses to market rates only when a house is sold; and parents can pass this tax break on to their kids. A new story from the Los Angeles Times reporters Liam Dillon and Ben Poston shows how the children of the wealthy (including, for example, heirs of Lloyd Bridges and Dom DeLuise) get enormous tax discounts on houses they inherited (and now operate as rental properties). More than 60 percent of these inherited homes are either second residences or are rented out. In all, this tax break costs Los Angeles County local governments more than $280 million a year in lost revenue, according to the state’s Legislative Analyst. It’s hard to think of a tax policy that’s more ageist, more racist, and more anti-immigrant than charging different tax rates to people based on whether they (or their family) were able to afford a home in California in 1978.

2. The Bi-Partisan Appeal of NIMBYism.  The New York Times’ Emily Badger digs deep into the politics and financial incentives that make NIMBYism such a powerful and nearly universal force in land use planning across the country. Americans may be deeply divided on other subjects, but Democrats and Republicans alike–especially homeowning ones–share an abiding commitment to making sure that development generally, and things like apartments and more intensive land uses, get built in somebody else’s backyard. Badger has a typically thorough and clear summary of the academic literature, mostly from political scientists, demonstrating the connection between homeownership, voting, and support for restrictive local land use controls.

3.  The perversity of “Levels of Service.” Writing at Strong Towns, Tulsa’s Sarah Kobos looks at the application of the traffic engineering profession’s “Level of Service” measurement to area roads and sidewalks. The widely used LOS measure ranks roads on how empty they are: the less-used the better, because it means cars can drive fast. Optimizing the road system for LOS means chronically over-investing in road capacity, and producing environments that are spectacularly hostile for other road users, especially pedestrians and cyclists. Sarah explains how highway engineers weirdly want to use the same metrics to plan for pedestrians (nearly empty sidewalks are rated “A”) and shows how intersections optimized for cars endanger everyone else.

New Knowledge

How easy credit fueled speculation during the housing bubble. Atif Mian and Amir Sufi have a new paper looking back at causes of last decades housing bubble. They track the growth of “private label securities” (PLS) the securitized home loans than dominated mortgage lending from 2002 to 2006. There was considerable geographic variation in the growth of lending by the non-bank lenders who issued PLS-backed loans, and Mian and Sufi show that the metro areas that saw the biggest increases in this lending had many of the hallmarks of speculation: a high proportion of loans went to borrowers with multiple mortgages.  The worst hit markets–like Las Vegas and Phoenix–saw a big increase in prices and a surge in home construction, and experienced an even larger and more sustained collapse in the housing bust. Credit-fueled speculation in this highly elastic markets allowed an irrational few to temporarily push the market higher.

Atif Mian and Amir Sufi, “Credit Supply and Housing Speculation,” National Bureau of Economic Research, Workpaper #24823, July 2018.

In the News

Curbed’s Alissa Walker has an in-depth story “Why US cities need more multi-racial, mixed income neighborhoods,”  describing the findings of our recent City Observatory report “America’s Most Diverse, Mixed Income Neighborhoods.”

Willamette Week quoted City Observatory director Joe Cortright in its story about declining rents in Portland. Landlords are offering free months of rent and even Amazon gift cards to new tenants.


The Week Observed, August 17, 2018

What City Observatory did this week

1. We disagree with the Washington Post on housing economics. Two weeks ago, the Washington Post published an article claiming that rents were going down for higher income renters but increasing for lower income renters. That didn’t square with our reading of the data, so we took a closer look.  It turns out that the data used in the Post’s story don’t include apartments. But perhaps more importantly, the data they use (generated by Zillow, and estimating how much single family homes, condos and coops would sell for if they were rented, even though most aren’t) show that rents in different income tiers tend to move in tandem over time. In Portland, the city highlighted in the Post article, rents in the highest and lowest tiers were actually strongly correlated. That’s a strong sign that what happens in one part of the market ultimately affects other parts of the market. And although the article implied that the construction of new, higher tier housing had no effect on rents, data for Portland show that rent inflation declined sharply in all tiers when new apartment construction surged after 2015.

2. Is St. Louis gentrifying? In a guest commentary for City Observatory, the University of Missouri at St. Louis’s Todd Swanstrom take a close look at the use of the word gentrification to characterize neighborhood change in St. Louis. It’s important to keep in mind that over the past several decades, the city of St. Louis has lost a majority of its population. That leaves considerable room for urban neighborhoods to add population without displacing anyone. It seems like gentrification has become “the g-word”–a general purpose epithet to express displeasure with inequality.

3. How can we save the heartland? Much of rural America, and many smaller cities and metros are either economically lagging or in decline. A new book from James and Deborah Fallows, based on their observations from barnstorming small places in their Cessna, points to strategies that have worked in some places. Bloomberg’s Noah Smith highlights a couple of the best approaches: build on a local university, and encourage innovation. While that can work in some places, we question whether it can work in enough places, much less everywhere to reverse rural decline. And please save us from invocations of that tired chestnut “public private partnerships.” Our most powerful rural policies probably turn out to be social security and medicare, which keep an aging rural population in place.

Must read

1. Philadelphia Playstreets. If you’re a kid from a middle class family, maybe you get to spend your summers at camp, or at summer school. But kids living in low income urban neighborhoods have fewer opportunities. Philadelphia has a great program, “Play Streets” that blocks off city streets for use as supervised play spaces for neighborhood kids. It cleverly combines this with support from the school lunch program to provide meals (and attract children). It’s a reminder that city streets can do much more than serve as conduits or storage space for cars; used wisely they can actually contribute to the civic commons.

Play Street in Philadelphia (Philadelphia Inquirer)

2. More Philadelphia Streets:  Block Parties. Summer is a time for block parties in the city. It’s another classic example of how we can expand the civic realm and build community. In general Philadelphia makes it pretty easy: you need a neighborhood petition with 30 signatures and a $25 fee and you can close your local street for a day-long event. But in some neighborhoods, there are concerns that block parties may trigger confrontations that lead to violence, leading Philadelphia police to ban block party permits in some places (chiefly near  locations where there’s been a recent shooting, or where the police have concerns about traffic effects). Police department discretion to approve or deny permits raises some concerns: it may have a chilling effect on the willingness of some neighborhood residents to hold a block party at all. The Philadelphia Inquirer’s Julia Terruso talks about what’s happening on the streets.

3. Where the city is outstripping its suburbs: Seattle. The Seattle Times Mike Rosenberg has another terrific analysis of the local housing market, this time looking at the city vs. suburb pattern of housing construction in Seattle. For decades, most new homes (apartments and single family) have been built in the suburban portion of Seattle’s King County.  But in this decade, that’s gone into reverse: suburban building is down sharply, and city building is up, with the result that more new homes have been built in the city of Seattle than those suburbs.

There’s a lot to unpack here. A big part of the explanation is the huge demand for urban living in downtown Seattle, thanks both to the presence of jobs (Amazon, et al) and one of the nation’s premiere urban environments. The decline of the suburbs undoubtedly reflects some combination younger adults disenchantment with suburban lifestyles, and suburban restrictions on new housing, especially apartments. What’s striking is that even though rents are up dramatically in Seattle, so far, development in its suburbs has done little to meet the demand for additional housing.

New Knowledge

A detailed database on urban violence. Patrick Sharkey has a new, with an on-line database compiling crime statistics from the nation’s largest cities. Sharkey also has a new report based on the center’s data, entitled “Are U.S. Cities Getting More or Less Violent?”  Spoiler alert: US cities are far safer today than they were one or two decades ago. While there have been some short-term fluctuations in murder rates, the unequivocal long-term trend has been downward. This map shows that only a relative handful of cities, including Baltimore and St. Louis. have experienced an increased murder rate since the crime-peak of the 1990s.  On the map below, green represents a decline in the murder rate through 2017; orange represents an increase.

The report looks at both long and short term tends in murder rates across cities. While the long term trend has been down, the report confirms that murder rates moved upward between 2014 and 2017, but then have trended downward in the last 12-month period for which data are available. As the report concludes:

. . . the recent increase in violence does not compare in scale or breadth to the fall of violence that has taken pace since 1991. The recent rise of violence has barely made a dent in the long-term crime decline.

In the News

GreenBiz republished our essay “The Limits of Data-Driven Planning,” pointing up the car-oriented status-quo bias of reliance on big data.

The Waterloo-Cedar Falls Courier cited data from our report Less in Common, about how little interaction we have with our neighbors, in a story describing how neighbors called ‘911’ in response to an 11-year old girl’s lemonade stand.

Strongtowns republished our commentary on New York City’s wrong-headed decision to cap the number of Uber and Lyft vehicles in the city.


The Week Observed, August 10, 2018

What City Observatory did this week

1. Jason Segedy on gentrification. This week we feature a guest column from Akron planning director Jason Segedy. You can’t build new housing in any existing neighborhood, it seems, without some invoking the specter of gentrification. Writing from the perspective of an economically lagging rust-belt city, Segedy makes the case that if we’re really serious about turning around city economies and promoting greater integration, we ought to be welcoming new housing investment.

2. More evidence of declining rents in Portland. A couple of weeks back, we flagged data from showing that rents in Portland had declined about 3 percent in the past 12 months. We decided to double check this against Zillow’s estimates of average apartment rents in the city of Portland:  they show very nearly the same result:  since June of 2017, average rents have declined about 3.5 percent.  By Zillow’s reckoning, average apartment rents in Portland are down about 9 percent from the peak levels recorded in 2016.  It’s a clear sign that the added supply of new apartments built in the city is putting downward pressure on rents.

3. New York City headed the wrong way with ride-hail cap. New York’s City Council is poised to adopt a package of measures that will impose a cap on the number of ride-hailed vehicles in the city.  It’s based in part on the assumption that ride-hailing services like Lyft and Uber are responsible for the city’s growing traffic congestion, and in effect, emulates the medallion system for taxis that’s been in place for decades. But in our view–and that of other experts like Charles Komanoff and Robin Chase–the real problem is not the ride-hailing services, but  un- and under-priced city streets. Limiting the number of ride-hailed vehicles won’t reduce congestion, but will distract from the solution that is most needed and most effective: establishing congestion pricing that applies to all vehicles.

Must read

1. Blaming human error.  The case for the safety benefits of autonomous vehicles hinges on the oft-repeated claim that human error is responsible for 94 percent of traffic fatalities. Schimon Schaff  of Frontier Group argues that this is at best a superficial reading of the data. Take for example the fact that fully a quarter of fatalities are due to excessive vehicle speed; that’s not so much an error as an intentional choice. And it’s a problem that could be resolved with something a good deal less than level 5 vehicle automation (for example, installing GPS controlled speed-governors that prevented cars for exceeding posted speed limits). It’s also likely that while automation will eliminate many human errors, it will likely produce unique new computer errors when software and hardware are less perfect than even humans in coping with complex and unanticipated situations. We shouldn’t assume that automated vehicles are a technical fix for all our road safety problems

2. Chicago’s Great Black Exodus.  Writing in the Chicago Sun Times Alden Loury explores the reasons behind the dramatic and continuing decline in the Windy City’s African-American population.  The city’s black population declined by 180,000 in the past decade. Chicago’s black population peaked at 1.2 million in 1980, and is now projected to decline to less than 700,000 by 2030.  Chicago was, famously, the biggest destination for those in the “Great Migration” in the first half of the 20th Century. But now, that migration has gone into reverse, with African Americans moving both to Chicago’s suburbs, and to other metropolitan areas. To Loury, the reasons for this out-migration are clear: “Chicago’s dramatic black population loss is the result of our segregation and the racism that drives segregation, and it calls for a far more urgent pursuit of equity and inclusion.”

3. Rethinking Housing policy to benefit homeowners and renters.  The Brookings Institution’s Jenny Schuetz shares her thoughts on a new direction for federal housing policy. Reviewing recent proposals for Senators Kamala Harris (for a federal rent subsidy) and Corey Booker (for similar plan, coupled with exhorting local governments to upzone residential areas), Schuetz summarizes how we can re-balance federal policies to address affordability for both renters and homeowners.  She calls for reining in tax provisions like the mortgage interest deduction and capital gains exclusions (which chiefly benefit high income homeowners) and expanding programs like housing vouchers (that reach only about a fifth of those eligible), as well as requiring local governments to treat proposals for multi-family housing no more restrictively than they do those for single family housing.

New Knowledge

City size and entrepreneurship. Net new business formation in the US economy has been declining for decades. It’s a signal that maybe we’re not doing a good job creating the innovative new businesses that propel job and wage growth. A new study from the Kansas City Federal Reserve Bank takes a close look at numbers from the Business Dynamics Statistics, which tracks the opening and closing of new businesses. The key measure the study focuses on is the “turnover” rate:  the rate at which new businesses replace older ones. The trend has been downward, but when broken out by metro area size, the data show that larger metro areas are consistently out-performing smaller ones. In addition, in the latest period for which data are available (2013 to 2014), business turnover in the largest metro areas has actually accelerated. Metro areas with more than a million population now have higher business turnover than medium sized metros, a reversal of the pattern that existed prior to the Great Recession. (Hat tip to Issi Romem).

Jason P. Brown, The Widening Divide in Business Turnover between Large and Small Urban Areas, Economic Review (Federal Reserve Bank of Kansas City)

In the News

The Memphis Business Journal reported on the relative paucity of racially/ethnically diverse, mixed income neighborhoods in that city, as identified in our report “America’s most diverse  income neighborhoods.”

StrongTowns shared our proposal that if scooters are asked to pay cities $1 a day for the sue of city streets that the fee charged to cars ought to be ten times or more higher (which it isn’t now).

Planetizen also echoed our arguments about scooternomics in an article entitled “Who should pay for streets?”

What Iowa City Can Learn from Portland.” Gazette columnist Adam Sullivan says “I never thought I would have to say this, but let’s make Iowa City more like Portland, Oregon.” He says so after reading our analysis of how apartment rents in Portland have fallen by 3 percent as more new apartments have been completed.


The Week Observed, August 3, 2018

What City Observatory did this week

1.  Your summertime must read: Alan Mallach’s Divided City. We have a review of this newly released book, which we think every urbanist ought to read. Although written primarily from the perspective of lagging Rust Belt cities, Mallach’s book has a lot to say to urbanists through the nation. He argues strongly that our focus needs to be on the continuing (and worsening) problems of concentrated poverty. Particularly in the Rust Belt, gentrification is rare, and far from being malign, is actually one key to reviving cities and widening opportunity.

2. What E-scooters tell us about road pricing. E-scooters are rolling out in cities around the country, and local governments are wrestling with tough questions about how they ought to be regulated. In Portland, Bird scooters is paying the city $1 per scooter, per day to cover costs of city-provided infrastructure and to build bike lanes and promote safety. That got us thinking: if a 25-pound, 20 mile per hour electric scooter pays the city a dollar a day for using streets, what would be a proportionate charge on on a two-ton, 60 mile per hour fossil-fuel vehicle? Whether you look at size and space requirements, road damage, pollution, or danger to public health, cars ought to be paying a lot more than a dollar a day to use city streets; yet in most cases they pay almost nothing. If we charged cars anything approaching the costs they create, we’d have a lot less driving, pollution, traffic deaths, and we’d have better cities. If scooters help us come to that realization, they’ll have really transformed urban transportation.

Must read

1. Blame ride-hailed vehicles for traffic congestion. Bruce Schaller, former New York Transportation Commissioner has a new report summarizing much of the research on ride-hailing, including his own. His report: Automobility: Lyft, Uber and the Future of Cities, argues that the ride-hailed vehicles are overwhelming city streets, and supplanting transit ridership, rather than reducing private car ownership. Schaller also points out that there’s been little reduction car ownership-in most cities the number of registered vehicles is still growing as fast as population.

2. Don’t blame ride-hailed vehicles for traffic congestion.  Its the unpriced streets, not the ride-hailed vehicles, says Robin Chase. Chase, co-founder of car-sharing giant Zip-Car, argues that its fundamentally wrong-headed to blame ride-hailed vehicles. for traffic congestion. Last time we checked, a ride-hailed car takes up exactly the same amount of street space as a privately operated motor vehicle. It will take a while before the prevalence ride-hailing prompts a significant number of households to give up their own privately owned vehicles. Rather than demonizing Uber and Lyft, Chase urges that we work to implement fair user fees across all modes to encourage more efficient use of our street.

3. Gabriel Metcalf’s Exit Interview.  For much of the past two decades, Gabriel Metcalf has been at the helm of SPUR, the San Francisco-based urban advocacy group. As we all know, the Bay Area has been the center of the maelstrom of economic change, innovating new technology, and with the problems of housing and transportation coming at them at a speed and scale un-seen anywhere else. Metcalf is shortly headed to Sydney, Australia, to lead an urban non-profit there. He sat down with the San Jose Mercury News, to share a few parting thoughts.  The whole interview is worth a read, but we were particularly taken with Gabriel’s short explanation of why we ought to care about cities:

Cities are a vessel for holding human difference. That’s what a city is. And that essential purpose of holding human difference becomes a platform for a lot of other really interesting things. Cities end up fostering creativity of all kinds because they bring so many different kinds of people together. That shows up in political movements, it shows up in artistic movements, and it shows up in economic innovation, as well. And, it also turns out cities are incredibly ecologically efficient.

New Knowledge

Americans spend less time traveling now than in the 1990s. For all the hand-wringing talk you hear about rising congestion and lengthening commutes–and the utterly phony estimates of what this costs us–the most comprehensive measure of how we spend our most precious resource–time–shows that Americans are spending dramatically less time traveling now than they did two decades ago. The State Smart Transportation Initiative reports on research looking at data gathered as part of the American Time Use Survey, which asks a statistically representative sample of Americans to keep a daily diary of their activities. The reports show that the amount of time spent traveling peaked in the 1990s at 10 hours per week, and has declined more than 15 percent since then, to less than 8.5 hours per week.

The trend is influenced by a wide variety of factors, including a slowing rate of job growth relative to population, and more time spent at leisure (which often takes place at or near home). But a key factor may have been the growth and subsequent slowing of highway construction. As SSTI’s Chris McCahill reports, “The authors note that rather than reducing congestion, massive highway construction from 1960 to 2000 likely motivated this sprawling growth pattern.”  Bottom line: this is more powerful evidence of induced demand. Building roads prompts us to drive further and travel more, rather than saving us time. And if we’re concerned about the amount of time Americans spend traveling, building more roads will only make that problem worse, not better.

In the News

Joe Cortright’s observations on the economic contributions of Portland’s hundreds of food carts were featured in the Portland Business Journal (paywall).

CityLab cited our analysis of the location trends of young, well-educated adults in its article “Do Millennials Prefer Cities or Suburbs? Maybe Both.”

Writing at The American Conservative, Matt Robare quoted City Observatory’s look at Houston’s Katy Freeway, widened to 23-lanes at a cost of billions, and now even more congested than ever.