Let’s Talk Sense About Solving Our Middle-Housing Issues


Having toiled in the fields of housing economics and affordability for the past quarter century, I am encouraged to see that the issue is finally having its day in the public arena. It is also encouraging to see some progress on what has become known as “middle housing,” a concept I spent many years promoting.

What is discouraging, however, is the continued poor quality of the public dialog on the causes of our housing problems and, hence, the poor quality of the solutions on offer. Tapping into what I have learned over the years, I hope to shed some cold light on the underlying causes of our housing affordability problems and the hard time we have in articulating them. 

I begin with some principles that should guide any discussion of housing and affordability.

Single family and multi-family housing are two distinct topics, and must be treated separately. Land use plans count “units,” but no one lives in a “unit.” We live in a range of specific housing types, and if we are fortunate enough to have choices, we select the building type that meets our needs.

The great majority of housing can be categorized as either single family detached (ownership or rental), or multi-family (mostly apartments, some condominiums). These are not interchangeable in the market. In particular, it is exceedingly rare for an individual or family to substitute an apartment for a detached house, if the house is their preferred choice and they can afford one somewhere. If people cannot afford a house in their preferred location, they will substitute locations, not housing type. Hence the axiom, “drive to qualify.”

Two generational myths have grown up around this question. First, Baby Boomers were thought to want to move away from their big homes in the exurbs and into multi-family buildings in exciting urban centers. But Boomers are not moving to multi-family housing at any greater rate than previous generations. Second, Millennials were thought to want to live their lives in hip urban settings. As they age and form families, however, Millennials are moving to detached housing at the same rate as generations before them.

The cost and affordability problems of single family and multi-family housing are distinct, and the solutions for one are not applicable to the other. But there is a rhetorical whiplash. A discussion that starts out talking about the high cost of detached housing very quickly becomes a discussion about building more apartments around transit stations. The solution to expensive houses is to build more houses, not to build more apartments.

For a very long time, the housing stock of the region has been just about two-thirds single family and one-third multi-family. Multi-family housing in King County has surged in recent years, but only because of the massive influx of young, childless adults into Seattle and the Eastside. There is no secular shift in demand toward multi-family across the market. Furthermore, single-family housing is increasing its share in Pierce and Snohomish Counties as families look for affordable detached homes.

So, we need to continue to build large numbers of detached homes within reasonable commute distance of job centers. We need to pay attention to the policy problems we have placed in the way of that construction and not divert ourselves into discussions of multi-family.

High housing costs are the result of policy choices. High housing costs are not inevitable, in terms of either time or place. Housing was affordable nearly everywhere in the U.S. until about the mid-1980s. Then, housing costs began to shoot up in certain areas, but not in others. Actual construction costs—labor and materials—vary only slightly over time and place. What varies significantly is the cost of land and the cost of getting government permission to build things on that land. 

Housing costs are not a function of growth. Among the largest metro areas in the county, the correlation between population growth and housing price increases is exactly zero. There is a small correlation between housing price growth and GDP growth in those metro areas, but prosperity does not explain much of the differences across metro areas.

There is, however, a much stronger correlation between housing price growth and the level of land use regulation in metro areas. We can argue about the merits of those regulations, but we cannot kid ourselves: public policy is at the root of our housing cost problems. Other very successful metro areas have made different policy choices and have much more affordable housing.

Viable solutions must scale up to the size of the challenge. We are tens of thousands of homes short of our needs, and solutions must scale to that level to be worth considering. Clever ideas abound, but if they result in just a handful of homes, that is not helping much, and we should not spend time and energy on them.

We have experimented with co-housing, we have built houses out of shipping containers, we have land trusts. None of these kinds of efforts, interesting as they may be, have proven able to scale to the size of the problem, and none really gets at the core of the issues.

So why waste time and political capital on policy efforts that do not yield large numbers of homes? And why give political cover to leaders who will claim they have taken action on housing when that action is  not meaningful?

The homebuilding industry is rational, risk averse, and subject to constraints imposed by financing. Homebuilding entails lots of inherent risks, so we can’t expect builders to take on additional risks in pursuit of experimental housing forms. And, even if home builders want to take on more risk, their lenders will not let them. Homebuilding is financed with a lot of debt and equity capital, and the banks and investors who loan that money have a low appetite for risk.

There are always a handful of builders who seem to be able to break the rules and get away with it. They get a lot of attention, but they are the exception, not the rule. The market will move slowly toward new products, but proof of concept will need to come from solid demonstration projects and from innovations in large master-planned communities.

To get to the massive scale of building that we need, we must rely on typical builders operating within standard constraints. As national, publicly-traded builders come to dominate more of the local industry, we will see even more conventional thinking.

Subsidized housing is not a meaningful part of the solution. Here is another sacred cow. We have come to equate “affordable housing” exclusively with subsidies. This is wrong. Subsidized housing constitutes a very small share of the housing stock, and there is never going to be enough money to make it a larger share.

This is not to denigrate the fine work done by the housing authorities and the many non-profit providers in the region. But the inescapable reality is that there is not enough subsidy money to meet the housing needs of people who truly cannot support themselves (see the reaction to the draft plan of the King County Regional Homelessness Authority). People who have regular incomes should be able to find appropriate housing they can afford in the market, and scarce subsidy dollars should be reserved for those who can never operate in the housing market.

Beware the subsidy trap. If we expand subsidies without dealing with supply issues, we are courting economic folly by subsidizing demand in an environment of restricted supply. All this does is enrich the owners of the limited supply and squeeze out those who do not get the subsidies.

Providing down payment assistance for a very limited supply of low cost single family homes just increases the pool of potential buyers and allows sellers to raise prices. Those with Section 8 vouchers squeeze those who don’t get them out of low cost apartments. There is a limited supply of land zoned for multi-family, so subsidized developers can squeeze out commercial developers. Purchasing apartments in order to hold rents down just shifts the tenant mix.

Current subsidy programs will, of course, continue, but providing more subsidy dollars without increasing the supply of product in the market is pointless and often counterproductive.

“Affordable” almost always means old. Given high land prices, it is very difficult, and probably unadvisable, to build new construction, low-cost, market-rate housing. Inexpensive new starter homes and bare bones multiplexes are a thing of the past.

The biggest supply of unsubsidized affordable housing is found in two places. Many older walk-up apartments were built as middle-income buildings and gradually filtered down in the market as they aged and were never upgraded. Mobile home parks have always been a key source of affordability, especially older units. Mobile home prices and space rents have risen significantly in the past decade, but older manufactured homes on rental space still offer family-friendly affordability.

But both of these sources of market-rate affordability have stagnated. Few walk-ups are built in the area these days, and many of the older complexes are now owned by institutional investors who keep upgrading them. The only mobile home parks being built today are high-end retirement communities that are not accessible to low income families.

Housing affordability cannot be solved at the local government level. This principle has finally been recognized by states across the country. Cities are always creatures of state government, and legislatures are finally taking advantage of this fact to force cities to make housing-friendly zoning changes.

The incentive structures that local governments operate within make housing development a problem, especially within the context of the state Growth Management Act (GMA). The GMA requires local governments to act in politically irrational ways, and, as with homebuilders, we cannot predicate our plans on irrational behavior.

None of these principles is mysterious or groundbreaking. But they interrupt well-established narratives and cause discomfort among those who are invested in the planning and service delivery status quo. Those who want to keep doubling down on the current orthodoxy of housing planning and policy need to explain why, when we are a far more prosperous society than a couple of generations ago, we cannot house our citizens nearly as well as we did back then. 

Topics for later in this series: Why are single family homes so expensive?  Why are apartments so expensive? The difficult politics of housing. What the state Growth Management Act got wrong. Lastly, it doesn’t have to be this way.

Michael Luis
Michael Luis
Michael Luis is a public policy consultant who has been wrestling with housing, growth and economic development issues around Washington State for over 30 years. He is author of several books on local history and served as mayor of Medina.


  1. Bravo. This reality-based piece is a bracing shot of fresh air.

    It provides a welcome contrast to the simplistic rationales offered for the slew of “housing growth” bills in Olympia this session. Those bills are being pushed by advocates as the means of enabling development of even more of what has been built in Seattle recently to over-abundant levels: small units, in multifamily structures.

    I hope you write more for Post Alley. Lofty ambitions of Washington State lawmakers in the waning days of the Reagan era produced the transit-oriented development schemes that have controlled development throughout this region since. All the new construction stands as monuments to that unprecedented land use planning. What are the good, the bad, and the ugly outcomes of the Growth Management Act?

  2. Could you elaborate on this with data?

    “ Housing costs are not a function of growth. Among the largest metro areas in the county, the correlation between population growth and housing price increases is exactly zero. There is a small correlation between housing price growth and GDP growth in those metro areas, but prosperity does not explain much of the differences across metro areas.”

    It seems clear to me in our local environment that the influx of a mass of new high wage earners had a significant effect on the cost of housing. Do you think if Tech companies had not come to Seattle attracting tens of thousands of high wage employees that housing costs would be the same? (Having lost out to bidders with $100k escalation clauses my lived experience sees a distinct correlation.) During the fracking boom small towns in the Dakotas doubled in population and housing costs went into the stratosphere. There are many examples of population increase influencing cost of housing.

    One thing that puzzles me is how far and wide the increases in cost are, especially for rentals. If someone in Seattle wants to sell their home and retire to an affordable urban area in the Northwest, good luck. Median price of a home this spring in Wenatchee is $469k, the average is $649k.

    It will be interesting to see if the pandemic escalations completely or ever reset in the rental market in a way that makes a difference. I visited Knoxville last spring in Tennessee with a vision of living there part time. Rents had almost doubled in a year and a half, I checked recently and I don’t see that they have come down in any significant way. It used to be that places with lower standards of living i.e. most of the south were immune from the kind of expense, spikes of the coast and other desirable areas.

    How much of the increase in demand is population demographics and what in only 3-5 years is so different? From everything I am seeing we are approaching a population crisis going in the opposite direction that was predicted. Fewer and fewer women are having children. Japan, Scandinavia, and Italy are cautionary tales.

    I would love to see you unravel the hidden dynamics of the consolidation of the builder industry and the dominance of ownership by REIT’s.

    • Iskra:

      Among the 21 large metro areas that are tracked by the Case Shiller home price index, the correlation between price growth and population growth since 2000 is 0.0099. Pretty much zero. Quite a number of large, successful cities–mostly in the Sunbelt–have grown faster than Seattle but have had less home price increase. Now, I know that most people don’t want Seattle to look like Houston, but that is the point. We have made decisions about what we want to look like, and those decisions have had consequences for home prices. Ya can’t have it both ways.

      Sticking with Houston as an example, that city has more wealthy people than the Seattle area does, but manages to provide homes for not just them, but also for people of modest means.

      There will always be lags between the time a region becomes hot economically and when the housing market catches up (e.g. Nashville). But we have had over 30 years since the tech boom started (remember, those early tech workers had oodles of money from stock options, so we could see what was coming) to adjust our strategies, and we have not. Planning is still based on a 1980s economy dominated by Boeing and the Ports.

      As for your other questions, I will take them up in future articles. Thanks!

      Michael Luis

    • Right. An analysis based on a simple “growth” number, isn’t going to show the problem.

      Of course as population grows, housing grows along with it, but will it grow fast enough that supply is adequate at all times? When there’s a huge surge? Obviously not – no industry deregulation however generous is going to make that possible. It takes time for the industry to ramp up, and during that lag, there isn’t enough supply.

      And as you say, the economic impact is related to the income level of the people who arrive during that surge. A few years ago a realtor friend described it as “million dollar signing bonuses in their back pockets.” How’s that going to affect affordability?

      And I second the call for a look at investment activity, which seems likely to zero in on areas where prices are already escalating, and make it significantly worse. Housing went up 30% in the last decade, income levels 11%. It isn’t because the industry is over regulated.

      Yes, let’s talk sense – not more deregulatory nonsense. We need to enlarge the urban growth boundaries, because people snag all the land for horse pasture? Look up “King County Urban Growth Area” to see the boundaries – it’s a vast, often lightly settled area.

  3. If land and permitting are the housing cost drivers, isn’t it time for the legislature to review the Growth Management Act? And local governments to ask similarly hard questions.

    I agree that too often our State and local political leaders focus on “units.” And in my experience their focus is often driven by “planning departments” from where they may have gotten their start as “planning commissioners.”

    Case in point: “The 8(10,12) minute neighborhood” which maintains that with a large number of “units,” frequently subsidized and with limited parking, we can avoid building detached housing. “Unit dwellers” transportation needs will be met by walking or mass transit, thereby reducing carbon emissions.

    Maybe this is the future. Clearly, the “plan” is to change our life style. Where the “plan” falls short is that some of us want to have that “Growth” debate at both the State and local level before these decisions are made.

  4. Mr Luis,
    Where do you propose we should build tens of thousands (a few hundred thousand?) detached SF houses in our region? However you want to describe the region?

    • David:

      Begin by moving the urban growth boundary, especially in King County. When the UGBs were first established we were told they were not permanent, but a way to ensure “orderly and contiguous” development. Well, through the magic of revisionism, they have become permanent.

      The result is many square miles of five-acre tracts that provide McMansions, horsey ranches and hobby farms for the wealthy and privileged (not much viable agriculture at that scale). A quick look at the aerial photos of areas along the UGBs shows plenty of development opportunity that would not interfere with environmentally sensitive areas. The UGBs were drawn arbitrarily and should be moved.

      Yes, this will require infrastructure, but if we can afford $70 billion for light rail, we can afford to expand roads and utilities to these areas.

      As it is, people who work in King County and cannot afford to live there are moving north and south, and not into urban centers. King County can pat itself on the back for all its containment of urban growth, but, pre-pandemic over 300,000 people commuted into King County each day from adjacent counties. Might be a good and responsible thing for King County to do a bit more to house the people who work there.


      • ML,
        Thanks for the straight answer.

        Until & unless I knew exactly where you’d propose moving the boundary, to see exactly which areas at issue, I certainly couldn’t offer an opinion.

  5. As with blanket upzones, “moving the growth boundary” is a blunt hammer for a task that could be accomplished in other ways. I was at a lunch meeting of mostly builders and policy wonks what, —25?— years ago when Ron Sims was KC Executive and had just moved the growth boundary. I was the only woman who spoke up and the only person to call him on it. Only when I got back from the mike, sat down at my table and looked at the expressions on the faces around me did I realize every table was set to profit from the move.

    I spent part of my childhood in the Auburn-Enumclaw plateau on a pony farm. I remember the violation of the first subdivisions well. Suddenly cows and horses were the least popular residents, as the urban-transplant nose sniffed and complained. The same gentle density tactics that can be applied in the city can be applied, with nuance, in the suburbs. Do not build before infrastructure is in place. Cluster. Shrink lots modestly before upzoning to urban levels. Understand the quality of life that brought people to the border zones, respect and work with them. Notice that places like Enumclaw are running dry on groundwater. Retain environmental protections and design review. I look forward to your mapping of what moving the growth boundary might mean.

    Every neighborhood deserves respect and care. The West without the land is not the West.


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Comments Policy

Please be respectful. No personal attacks. Your comment should add something to the topic discussion or it will not be published. All comments are reviewed before being published. Comments are the opinions of their contributors and not those of Post alley or its editors.