When Cities Switch To One-Winner Council Districts, Housing Growth Plummets

A new 11th-hour idea for rewriting the rules of Portland’s city government has several possible flaws, but here’s one: statistically speaking, it’d be likely to worsen the city’s housing shortage.

The proposal was publicly floated in a media interview three weeks ago by its loudest advocate, city Commissioner Mingus Mapps. Mapps’s idea, according to The Oregonian/OregonLive: to scrap the concept hammered out by a city-appointed citizen commission over the last year, a form of proportional representation, in favor of a winner-take-all system that would elect its entire council from smaller, one-winner districts.

A referendum that would implement proportional representation and other changes is on the November ballot in Cascadia’s third-largest city as measure 26-228. Mapps’s concept has yet to be fully hammered out and might yet change.

If Mapps’s proposal ends up as described by The Oregonian and wins support from other councilors and the public, Portland would be imitating a form of government that reduces housing growth in US cities by an average of 20 percent.

If Mapps’s proposal ends up as described by The Oregonian and wins support from other councilors and the public, Portland would be imitating a form of government that reduces housing growth in US cities by an average of 20 percent.

“The number of units permitted falls sharply . . . immediately upon the reform’s approval,” wrote Evan Mast, the author of an academic paper on this subject circulated in 2020 and published in May, “Warding Off Development: Local Control, Housing Supply, and NIMBYs.”

One Portland city councilor says he wants a council elected from single-winner wards  

Here’s the current political situation in Portland: In late 2020, its city council named 20 people to research and propose changes to Portland’s charter—essentially its city constitution. This summer (2022), after much testimony and debate, those 20 people (the Charter Commission) proposed some changes to how the city’s government operates and how voters elect its officials. By a 17–3 vote, they sent this package directly to Portland voters’ November ballots, bypassing any edits from City Council.

The new election system they proposed was modeled on a system that’s common around the world: proportional representation. It would divide the city into four relatively large districts, each of which would use multi-winner ranked choice voting to elect three commissioners in rough proportion to the various political opinions and interests that make up each district. If more than one-quarter of a district ranks a candidate highly, that candidate earns one of the three seats in that district.

During an informational presentation from the Charter Commission, three city council members announced that though they liked the proposed changes to how the city is run, they had concerns about that sort of election system. Then, on Aug. 30, one of those council members, Mingus Mapps, announced that if November’s ballot measure fails, he expects to propose his own, different election system, one more familiar to many Americans: eight smaller districts that would each elect just one person.

It’s that system—a council made up entirely of “winner-take-all” wards—that a growing body of research says tends to lead to fewer homes being built.

One-winner ward systems seem to cause housing slowdowns 

One recent academic paper with this finding is from Evan Mast, a housing economist at the University of Notre Dame. He looked at housing growth rates in 238 cities that recently made the sort of switch that Mapps is suddenly advocating for in Portland: from at-large to district elections. It turned out that these cities generally saw a sharp dip in new construction immediately after their change. Then, after about two years, they tended to rebound somewhat to a new, lower equilibrium of housing production. Cities that switched to single-winner districts settled at about 20 percent fewer new permitted units per resident than cities that continued using at-large elections.

Another relevant paper is by two political scientists. Michael Hankinson and Azya Magazinnik’s “The Supply–Equity Trade-off: The Effect of Spatial Representation on the Local Housing Supply” puts California cities under a similar microscope. They found that in cities required by state law to switch to districts, apartment building in particular fell by an average of 55 percent.

It’s possible that Hankinson and Magazinnik found a trend that’s unique to California. On the other hand, their finding was quite similar to Mast’s national one. And because the switches they looked at were mandatory, their test is more robust. We can be more confident that there’s no third factor that drove both the switch to a ward system and the building slowdown.

It seems hard to dispute that the new ward system was itself a major cause of the housing drop.

Hankinson and Magazinnik found other things, too. Housing fell more steeply in cities with more segregation, more racial diversity, and more historical overrepresentation on city council of the largest racial group. On a related note, they found that converting to districts changes the location of what new infill there is. A completely at-large council (like Portland’s current system) tends to concentrate more apartment construction in neighborhoods with more people of color, while district-based councils tend to spread out construction across the city. This is the “supply–equity trade-off” mentioned in their title: switching to districts tends to decrease the citywide supply of housing. It makes the housing pie smaller overall in number of units built, but it also distributes those smaller slices of pie more evenly across all districts.

Single-winner wards prioritize local problems over citywide solutions 

Why do ward systems seem to slow housing growth? A couple of possibilities come to mind.

One is the “ward boss” effect: when there’s a question of whether or not to allow more homes in a particular district, council members from single-member districts may tend to defer to the councilor in that specific district, essentially as a mutual benefit pact.

If you let me run my district, I’ll let you run yours.

In New York this practice is called “member deference.” In Chicago it’s “aldermanic privilege.” Los Angeles’s council “fiefdoms” make it “almost like 15 cities in one city,” a reformist councilor said in 2020. And because it becomes an unwritten rule, it’s very difficult to ban.

In some contexts, this sort of setup may trend toward actual corruption, which can be good for real estate insiders but bad for tenants, homebuyers, and taxpayers. When a single city councilor has discretionary power over a project, it may become more important for developers to have a friendly personal relationship with each councilor. In general, this only deepens the power of long-established real estate pros, which drives up prices by making it harder for new people with new ideas or less personal wealth to enter a market.

The other possibility is more structural. Additional homes may be annoying to some people who live nearby, but the benefits of building enough homes get spread throughout the city.

Hyperlocal costs, citywide benefits. It’s a lot like the politics of a homeless shelter, a waste transfer station, or a bus lane that would displace a parking lane. And when a council has many small districts, each with a single member attuned mostly to swing voters in their district, the citywide benefits of local change can get lost.

A bunch of single-winner wards seems to be a formula for a city that misses the forest for the trees.

How single-winner district reps nearly killed zoning reform in Charlotte 

The political differences between district and at-large councilors were on sharp display this year in Charlotte, North Carolina, a fast-growing city about the size of Portland.

Last month, Charlotte narrowly passed a progressive comprehensive plan that legalized up to four homes per lot in many low-density areas and emphasized consistent development rules rather than discretionary decisions. Voting for it were the mayor, every at-large councilor, and two representatives of close-in districts with more renters. Voting against it were the four other district representatives.

Meg Fencil, who pushed for the new comprehensive plan as director of impact for Sustain Charlotte, said the at-large members of Charlotte’s hybrid council were crucial to the housing reform.

“I think they’re able to take the bigger-picture, longer view of what’s happening at a citywide level,” she said. “They’re able to remind their colleagues that by allowing more density all across the city, it’s unlikely that any negative effects of that would be concentrated in one district, and that the positive effects of that would be spread across the city.”

But Portlanders don’t need to look so far for an example of an all-ward council being captured by anti-housing homeowners.

Just down Interstate 5, Eugene is one of relatively few cities in Oregon with almost exactly the council structure Mapps proposes. It is also home to maybe the most anti-housing politics in the state. Five years ago, after the state legalized backyard cottages, Eugene had to be dragged through the courts before its council would comply. This year, the council passed a pro-housing reform, only to see one of its staunchest advocates of infill housing and greener transportation recalled by her ward’s voters two weeks ago. The reason: she’d supported the study of new bus lanes in her district and prioritized homes over parking spaces.

Typical rents in Eugene recently exceeded those in Portland.

In Portland, single-winner districts wouldn’t tend toward diversity; proportional representation would 

Neither of the recent studies cited above looked at cities switching from at-large to multi-member districts, as Portland’s November ballot measure 26-228 proposes.

But if districts have multiple representatives with a variety of political perspectives, as Portland’s proportional representation system would be designed to deliver, it’s harder to imagine a “ward boss” effect. And with just four districts citywide, councilors would have to balance input from a wider variety of interest groups. Each of Portland’s four districts would be roughly the size of Cambridge, Massachusetts, whose nine-member council uses nonpartisan citywide proportional representation and has recently passed a string of pro-housing reforms.

Every city is unique, of course, and the Mast and Hankinson-Magazinnik papers were calculating averages. In Austin, housing advocate Dan Keshet said a 2014 switch from at-large to district elections has been good for pro-housing politics, because creating three majority-Hispanic districts gave Hispanic voters more structural power in government.

“After [2014], the Hispanic community elected three pro-development councilmembers,” Keshet said.

But that phenomenon—representation—is actually one of the key arguments in Portland for a proportional system and against a winner-take-all system like Mapps proposes. In part because Portland’s original brand of segregation was total exclusion of nonwhite residents, it wouldn’t be possible to draw districts where people of color are the majority even if the city tried. Unlike in Austin, a single-member district system in Portland wouldn’t tend to create a council that better represents voters of color. But a proportional representation system like the one on the November ballot could.

The proposed alternative could still change, if it moves forward at all 

Portland’s Commissioner Mapps says he’ll release more details about his rival concept by October 3. But even if voters were to reject November’s ballot measure, presumably hoping that a plan like Mapps’s were to come before them instead, he’d need two more council votes to send it to the ballot.

Portland tenants, homebuyers, and others who care about housing costs have every reason to watch these discussions carefully.

Which is to say: Mapps isn’t the only Portlander with influence over Plan B. Mapps’s council colleagues could reject his plan in favor of something else. They might insist on keeping today’s fully at-large council and focus only on other changes to city government. Or they could propose a hybrid council like Charlotte’s or Seattle’s, with a mix of at-large and district-based representatives.

Whatever happens, Portland tenants, homebuyers, and others who care about housing costs have every reason to watch these discussions carefully. There’s reason to think that an all-ward system would throw fuel on the fire of Portland’s deep housing shortage.