Policy Brief: The Local Control Con

Cities Will Not Fix Housing And Homelessness

“But Local Control”

I frequently encounter activists and electeds who extol the virtues of “local control” when it comes to decision-making, particularly on matters of housing and homelessness. 

But while local control is certainly a good fit in some cases, it is foolishly unfit in others, specifically, housing and homelessness. 

In the following I’ll explain why when it comes to issues like housing and homelessness, the parochialism of local-control makes us worse off. In so doing, I’ll make it clear why “we should have local control” is not remotely a good argument against state and federal efforts to preempt local action on housing and homelessness. 

Flailing Opportunism Dressed Up As Political Philosophy

One thing I think we need to admit is that few people operate with an actual reason for their favoritism toward a particular layer of government. 

In fact, I titled this “the local control con” because most arguments I’ve encountered about which layer of government should make decisions are unprincipled, opportunistic attempts to move fights to move favorable ground. 

As my friend Mason, the Mayor of a certain Seattle suburb says, “if someone says to me ‘I think that decision should be made at a different level of government’ it’s basically an admission that they don’t have a case for their position.” 

While there are actually good, principled arguments to be made in individual cases, they are almost never at play when people bring this topic up. 

That said, there are thoughtful ways to evaluate the question of where a problem should be addressed.

Here I’ll present one principled way to identify when local control is a bad idea and why Washington State should feel completely fine preempting local governments on issues of housing and homelessness. 

Gaming Out When Local Decision Making Sucks

Game theory is a branch of research dedicated to mathematically describing and understanding strategic interactions between “players,” who can stand in for individuals, companies, communities, and countries whose incentives are different from one another. 

It can be extremely useful for thinking about which level of government should address a problem. There is a category of games–called collective action problems–that, SURPRISE, require collective action to succeed. 

These “exist when individuals, acting rationally in pursuit of their self-interest, have incentives to make decisions that are harmful to the interests of others as well as, ultimately, the individual themselves.”

My rule is pretty simple: If leaving the issue at the local level keeps the collective action problem in place, it has to move to a higher level of government. 

Collective Action Problem? Preempt.

There are three big kinds of collective action problems–Prisoners’ Dilemmas, Tragedies of the Commons, and Free Riding. There are political problems that map onto all three–and when they do, preemption is appropriate. Otherwise, the collective action will remain in place and everyone will remain worse off. 

The relevant collective action problems for questions of housing and homelessness is The Free Rider problem. I’ll discuss the other two–Prisoner’s Dilemmas and Tragedies of the Commons–in an appendix at the end.

Free Riding

Anyone will recognize the free-rider problem from group projects in school, where the team shares the same grade.

While in a small group, it is fairly easy to pressure and socially sanction one another to drive the needed shared behaviors. (Let’s be honest, if you read this far, you probably just did the work for everyone to be sure to get an A).

But in bigger groups it gets much harder. If the group is big enough and the project large enough, it is difficult for any one person to significantly improve the outcome on his or her own, so each has an incentive to free ride on the others’ work. 

Still, these challenges can be overcome with the help of institutions. This is why most workers are better off in a union, part of why we have a single defense department instead of 50 state level versions, and it is why we have a (taxes required) fire department, rather than opt-in individualized insurance for fire-emergency response. 

In fact, collective action problems in general, and free rider problems in particular, are a huge part of why we even have a government.

Imagine if every year we just passed around a hat and asked people if they want to pay for public schools–each individual gets to decide whether she or he wants to opt in. That would be the end of the public schools! 

If the free-riding incentive remains in place, the desired system that makes everyone better off collapses. That is why we tie everyone’s fates together and require everyone to pay taxes, whether they have a kid in school or not. 

But collective action problems still bedevil public life. One of the reasons we don’t have a healthy electrical grid in the US, even though we’d all be better off if we simply said yes to grid upgrades, is individual jurisdictions love to mire everyone else in process, try to extract concessions from those building up the grid, or just NIMBY their way out of cooperation. 

True, we would all be better off if we had a better grid, but given the status quo, individual jurisdictions might just be better off than they are now if they get to keep their view and prevent those power lines or wind turbines from going in. And since they know everyone else is going to do this, and they aren’t likely to get their better grid–all the more reason to not cooperate!

Interestingly, this kind of collective action is at the heart of the statewide initiative I-2124. If it passes, people will be able to opt out of the long term care system in Washington, which will likely collapse a system that makes us collectively better off.  Vote no! 

The Housing Crisis

The Housing crisis is another classic free-rider problem. 

Individual cities, especially large ones, can make a significant dent in housing prices. Cities whose housing production keeps up with demand have lower rent burdens and price to income ratios and homelessness. 

But they are limited by the efforts of those around them. 

If prices relax, demand from the outside may increase, driving up in-migration, especially from the surrounding region. While demand isn’t infinite–there is certainly a lot of unmet demand for, say, housing in the central Puget Sound region. A city that steps out too far ahead may be inundated by a tsunami of change and dislocation overnight while everyone else sits on the sidelines. 

So most cities sit on the sidelines.

In fact, this feared outcome describes a kind of pattern we see within cities. Seattle’s urban villages or Bellevue’s downtown are great examples. When we upzone tiny slivers of land that are rich in transit access with good schools and jobs, they experience aggressive transformation overnight, with little change elsewhere. 

These few neighborhood transformations are not enough to sop up demand, but are big enough to experience the jarring changes and sometimes displacement that come from them. (Note, I actually like the changes in my neighborhood, but I understand that radical transformation is too much for most people, and in cases of displacement, can be genuinely harmful.)

Given that many people do not wish to see their communities similarly transformed overnight, they vigorously resist legalizing change in their neighborhood. 

But of course if we legalized housing statewide, most neighborhoods would only experience gradual change. Washington State has grown at roughly 1.2% a year, and King County at 1.3%. With broad legalization of housing growth, most of us would experience very little jarring change and much of the housing crisis would be fixed.

But instead, we are stuck with the incentive to free ride–people would rather let someone else deal with the growth, and someone else’s neighborhood change. 

(By the way, I’m not addressing the racist and classist elephant in the room - many people resist housing changes in their neighborhood because they do not want certain groups of people to live near them. That is obviously morally repugnant and as such, illegitimate–and is yet another reason for state preemption of local control.)

This is why the state has started to step in. 

While NIMBYs have yelled “local control” - local control has yielded decades of underbuilding, a massive drop in home-buying power for a couple generations, and astronomical rates of homelessness. And in fact these cities have, in part, been rationally responding to their incentive to free ride.

This can only mean the local level is the wrong place to solve it. 

If engaging at the local level leaves a collective action problem in place, and makes free-riding into an offer most places can’t refuse, it’s time to move the problem up a level. 

This is why the state stepping in to address missing ADUs and middle housing was so important, even if only a meaningful couple of first steps. It’s why the transit-oriented development bill that will eventually pass, and the reform of design review are so important. And since none of these will be enough to address the housing shortage, it’s why much more needs to be done. This includes things like permitting and code reform, broad apartment legalization and parking freedom, and accountability mechanisms for jurisdictions that try to free ride.

(This is also why the federal government needs to get much more involved, since these problems are not just regional, but inter-regional and national.)

Homelessness

While shortages of housing create higher rates of homelessness, addressing the state of an already homeless person (especially when there is housing scarcity, see above!) is a slightly different problem. 

Once people are homeless, they are more likely to become victimized, experience behavioral health crises, and need services. 

Here we see the free rider problem back up at full throttle. 

Despite the work of propagandists who want you to believe otherwise, most of those who are homeless in Seattle come from the area–this is similar for most larger cities relative to their metro areas. The bigger cities offer more options for work (yes, many people without homes work), opportunities to blend in, are often a bit kinder to those in need, and, most important, the efficient concentration of services.

But this breeds resentment among, say, Seattle residents who pay much more in taxes per capita providing homelessness services than do residents of Bellevue, even though Bellevue may actually generate more homelessness per capita than Seattle does. Whatever the case, the load is severely imbalanced.

Still, Seattle is only willing to go so far and fork out so much to handle the region’s problems, because others are already free riding. The King County Regional Homelessness Authority was, in part, conceived of to tackle this problem. Homelessness is a regional problem, the thinking rightly went, and so it should be addressed regionally. 

But, now that you understand collective action problems, you should see that the KCRHA was doomed to fail–because participation was voluntary. 

This left the incentive to free ride in place. 

Seattle tried to lead the way, and King County chipped in a good bit, which at least sort of collectively addressed the issue (since all of King County pays King County taxes). But many other cities did very, very little to address what they conveniently saw as a “Seattle issue” - and they had a strong incentive to keep telling themselves this sweet little story and failed to really fork out–they free rode on Seattle’s efforts.

The KCRHA would have had a much better shot if (beside not having 3 boards) it had taxing authority and the authority to place services despite local objections and then spread this out equally. Then our fates would have been bound together, we would have shared in the burdens and benefits–which would have allowed for a taxation rate that might meet the moment–and we’d all have been better off.

This is why, despite grueling effort by directors, board members, staff and electeds, to get some suburban governments to actually chip in–the regional homeless authority as it was contrived–whatever its administrative problems–was always going to under-deliver. 

(Note, this isn’t necessarily because of a failure of imagination, by the way–the state has not created the legal means for setting up a true regional body for this problem).

The temptation to free ride will remain as long as the collective action problem is in place. To address homelessness services most effectively, the State must fund large scale solutions, or a regional entity like KCRHA (preferably one that covers all of Central Puget Sound) should be empowered to tax and decide on the placement of affordable housing, shelter, and services. At the very least, King County should be empowered to tax more aggressively to address the issue. If we’re not all in this together, we are likely to continue to fail. 

Local Governments Can Go Farther

This doesn’t mean there is no place for action for local governments when it comes to collective action issues. Even if the state steps in, local governments may have reason or feel the need to go above and beyond the requirements. 

The point is that the state (or federal government) needs to set a solid minimum, a floor, that none can fall below, in order to prevent the negative outcomes associated with free riding.

Seattle will benefit from more compact neighborhoods than Bremerton, and the state is unlikely to iron all this out in setting zoning minimums. Seattle also houses the biggest job and talent center for nearly 1000 miles in every direction–so it has more room for levying taxes without putting business at risk than does Sedro Wooley or Longview. Regional variability is fine with a strong floor in place.

But none of this is an excuse to say that housing should be left to local control, or that locals know best what they need. Local governments have been responding to their incentive and free-riding when it comes to housing and homelessness for decades, and it has made us all worse off. 

One has to look no farther than Seattle’s unimaginative, flaccid excuse for a comprehensive plan from the Harrell administration. Unless we happen to get really bold leaders in place at the right time in a lot of places for a few decades–not a great bet!--we’re never going to get this done.

The state should boldly and confidently step in and address housing and homelessness and should not be deterred by demands for more local control. 

Local control got us into this mess and it is structured in a way that it will not get us out.

Appendix A: Other Times to Preempt

I want to be clear that I am not suggesting that the only time for preemption is collective action problems. Securing people’s rights for example, requires that local governments have no say.

I’m merely arguing here that when local control leaves a collective action problem in place, that is one case where we should preempt. 

Appendix B: Two Other Collective Action Problems

The Prisoners’ Dilemma

Game Theory’s most familiar collective action problem is the prisoner's dilemma. 

Imagine the police arrest two men who are partners in crime and put them in separate cells with no way to communicate. Each could be charged with a serious crime leading to ten years in prison–but the cops don’t have enough evidence to convict. However, they can prove a lesser charge and put each away for two years.

So the cops offer each of them the same deal. If one of the prisoners confesses to the serious crime and betrays the other, he will be pardoned, while his accomplice will serve ten years instead of two. 

As a pair, both would be better off if both stay silent. But as individuals, each has a strong incentive to betray his buddy. And in fact, each realizes his partner has this same incentive and so has an even stronger incentive to hurry up and speak up first. 

A classic example of a prisoner’s dilemma is real life is an arms race. Both countries would be better off with mutual disarmament, but both could gain an existential edge by betraying the other. Both know the other has this incentive to gain an existential edge, which increases the downside risk of not-betraying the other country. Left alone, the likelihood is the continuation of the arms race. (A reliable, shared regime of inspections can avert this).

You see this in the case of police pay and bonuses in the context of the national police shortage. While departments would be better off if they all agreed to pay similar rates (perhaps adjusted for cost of living), coordinating this is very difficult, and every department has an incentive to outbid the others. 

This (and some silly ideas about performative virtue signaling by the Seattle City Council) is a big part of why Seattle now pays the highest salary in the state and offers a hefty $50k signing bonus.

Tragedies of the Commons 

The second big type of collective action problem are tragedies of the commons. In tragedies of the commons, everyone is better off if a common resource is well managed and preserved. (Think of a grazing pasture, a salmon run, or a climate-stable biosphere). 

But each individual is better off grabbing as much as he (we’ll assume it’s men again, make what you want of that) can while he can, especially since he knows everyone else has the same incentive. This is why we over-grazed, still sometimes over-fish, and brazenly continue to heat up the climate. 

It is also why we have built up elaborate regulatory regimes, and riparian and property rights to manage natural resources. In the case of the climate, we are desperately trying to move the needle at the national and supra-national level with coordinated activities like the Paris Climate accord–because most of the players know that we are in a desperate race against time and face the wrong individual incentives.