Scott Sumner  

We should focus on building "unaffordable" housing

The Best of Econlib: 2017 (con... The Unintended Consequences of...

Matt Yglesias has a good article on housing affordability:

When people -- specifically market urbanists versus regulation fans -- argue about housing affordability on the internet it seems to me that the two groups are using the concept of "affordable" in different ways.

1. In one usage, the goal of improving affordability is to make it possible for more people to share in the economic dynamism of a growing, high-income city like Seattle.
2. In the other usage, the goal of improving affordability is to reduce (or slow the rise of) average rents in an economically dynamic, high-income city like Seattle.

These are both things that a reasonable person could be interested in. But since they are different things, different policies will impact them.

Thus if you allow developers to go into a low rise slum in Seattle and build lots of high-rise units in that neighborhood, then more people will be able to live there, but the rents will probably be higher. Both supply and demand will have shifted, but demand will shift even further.

In recent posts I've argued that efficiency is far more important than distribution, and housing provides an excellent example as to why this is the case.

Many left-wingers start with the false assumption that society needs to build more "affordable housing". In fact, in any well functioning society the vast majority of the new housing being built would be "unaffordable", that is, out of the price range of the median income.

To see why, consider the nature of economic progress. During the Middle Ages, most people lived in miserable hovels. Today, most people in America live in nice houses and apartments. This transition occurred because the new homes being built tended to be superior to the existing stock of homes, at any given point in time.

Moving to higher quality homes is an important part of economic progress. Because one year's worth of new construction is only about 1% of the existing stock of homes, it's difficult to rapidly upgrade the quality of our housing stock. But if we are to make any progress at all, it's essential for new homes to be of much higher quality (and hence more expensive) that the average of existing homes. New houses should be unaffordable to average people.

The key point here is that by far the most effective way of providing "affordable housing" for average people is to get upper middle class people to vacate their existing homes, to free them up for middle class people to move in.

If high rise luxury buildings are built in a (low rise) gentrifying Seattle neighborhood, then it will become more difficult for lower middle class people to live in that particular neighborhood. However it will also reduce the overall price of housing at an aggregate level, and thus help working class people in aggregate.

Many non-economists make the mistake of thinking about living standards in terms of money and affordability. The correct way to think about these issues is in terms of output. If you build it they will come. We can "afford" whatever we can build. If America built 100 million McMansions, then almost all of America could afford to live in a McMansion.

[Yes, this example oversimplifies things; some would be bought by foreigners. And there are maintenance issues. But these complications do not negate the basic point I am making here.]

In a recent comment section I've been discussing the question of why young people in America and other developed countries are not out in the streets protesting our restrictive housing policies. There seems to be a consensus that young people do not understand that these housing regulations explain why it's hard for them to find "affordable" houses. In the 1960s, young people marched in favor of civil rights and against the Vietnam War, because they correctly understood the issues at stake. Today the issues seem to be too complex for people to grasp. So they become apathetic, or else focus on simpler social issues such as political correctness.

As an analogy, once the "low hanging fruit" of obvious artistic possibilities was exhausted, artists turned to more "difficult" styles, which were too challenging for average people to understand. (I have a good vantage point to observe this phenomenon, as I am able to grasp difficult visual works of art, but much less so in music and poetry.)

Something similar seems to be happening in politics. In 1896, the issues at stake in monetary policy were pretty simple, and could be understood by average people such as farmers. Today monetary policy has improved so much that average people had trouble grasping what was at stake in 2008. Today's policy mistakes are subtler.

As the obvious outrages like Jim Crow laws and the Vietnam War get removed from the agenda, issues become more complex. This may explain why young people are so apathetic about issues such as restrictive zoning rules, even though you'd expect them to be demanding that California adopt Houston-style housing regulations. They don't even understand that there is a solution to the problems that they face.

Screen Shot 2017-12-27 at 10.50.01 AM.png

Comments and Sharing

COMMENTS (29 to date)
RSF writes:

What’s the picture of?

Is there any actual chance of less restrictive policies? Most informed people seem to agree and yet restrictions continue to grow.

Andrew_FL writes:

I seriously doubt-actually I flatly deny-that young people marched against Vietnam because they "understood what was at stake". What, in your opinion, was at stake that they were right about, that they would not have been equally right about if they were marching in 1942? No, the far more likely explanation is simply that they did not want to be drafted into a war in which they had a significant probability of dying. This may have reflected their personal stake in the issue but it was not "what was at stake" in the war, which was whether Vietnam was to be a Communist country.

I can't say I'm surprised you think the farmers understood the monetary policy issue in 1896, either, but they did not.

E. Harding writes:

"As the obvious outrages like Jim Crow laws and the Vietnam War get removed from the agenda,"
There was nothing obvious about either of these policies from the standpoint of the time, especially where they were happening (South Vietnam and the U.S. Deep South).

@Andrew_FL, women are always more dovish than men, young always more dovish than old (even if the war mostly negatively affects men).

"In 1896, the issues at stake in monetary policy were pretty simple, and could be understood by average people such as farmers. Today monetary policy has improved so much that average people had trouble grasping what was at stake in 2008."
No; the issues are just as hard (and just as easy) to understand in both periods.

There is an even simpler reason why new houses will tend to be more expensive: they are new.

(Technological improvement in housing is actually pretty fast, surprisingly fast, but just the fact that newly built is more expensive than used would be enough to to explain why new housing will tend to be more expensive.)


In the 1960s, young people also marched in favor of genocidal extreme-Left regimes.

Weir writes:

Ask Theresa May. Turns out there are plenty of sons and daughters planning on inheriting an aged parent's London home. That's their long-range plan: Someone's mum must die.

Doesn't that complicate things? When the children understand perfectly well that the best way to afford a place in London is for the government to keep things as they are, with pensioners sitting on millions of dollars in real estate and, at the same time, expropriating some random young chap's wages?

Kevin Erdmann writes:

You make good points, but I think Yglesias is overstating the conflict. You could increase the population of a given city by some extreme amount, like 50%, with only marginal displacement. Even in the extreme, for every displaced resident, 20 others would have much lower costs. This is more of a Bastiat-style PR problem than an actual conflict.

For instance, here is a story about a rundown McDonalds that San Francisco bought for $15 million (!). One article I saw on it claimed the lot could support 90 units! This is slated for affordable housing, so it won't make the neighborhood more expensive and it won't displace a single resident.

Here is the local neighborhood response. There will be years worth of haggling over a bunch of virtue signaling and obstruction. Eventually, negotiations will settle for it to be approved for some marginal number of units that will cost a million dollars each.

This nonsense happens all over the problem cities. It is what makes them different from other cities. The conflict Yglesias proposes isn't relevant to this problem, and addressing that conflict won't address this problem, because that conflict is a red herring. The problem is that those cities are held hostage by an over-arching anti-market sentiment. Only subsidized units that meet a litany of moralist demands are considered "units". Here is an article by a San Francisco supervisor who complains that market-rate units can't solve the supply problem, but you can see why, since he only considers "affordable" units to be "units", and by definition market-rate units aren't affordable.

This is wholesale intellectual bankruptcy. We know that nothing handled under this sort of regime can be supplied generously. This isn't how any functional distribution system can work.

Yet, even with all of the obstruction, still, the trickle of new supply can stabilize costs, even at today's building rates. So, we get articles like this one at the Financial Times, which, I kid you not, complains in the last two paragraphs, that oversupply in the luxury market has brought down property values, which has cut into developers' profits, making it harder for them to fulfill their commitments for affordable units. So "oversupply", by this logic, makes it harder to provide affordable housing. The closing sentence: “The private sector alone cannot deal with affordable housing. This way, it only gets built if the top end of the market is working.” So, we can only get affordable housing if we can maintain a luxury market at 4 times the unencumbered cost of building. Again, these mental gymnastics were made possible without even touching on the problem Yglesias mentions.

The problem is that if market values that are 4 times the cost of building is the problem, then "oversupply" is the solution, but even if you get past all the neighborhood group obstruction, the solution itself is considered to be destabilizing.

All of this is a problem without displacing a single resident. In fact, the supposed problem of "oversupply" is that even a little market expansion brings down rents and prices so much that they are considered disruptive even at these low building levels. If these cities actually approved a sustainable amount of new building, rents and prices would fall faster than any local developments could counter. Nobody's rent would rise. The solution would require bankrupting all the existing developers, lenders, and many homeowners. It is falling rents that prevents the solution, not rising rents. And, it all comes, fundamentally, from the problem of urban governance in an age of political overreach, where amenities which would be reasonably supplied in any marginally flexible market have been moved into a political framework. Yglesias' conflict is simply one out of a hundred potential political tools that would used to prevent distribution of supply. No compromise within that regime will solve it. If, somehow, market mechanisms could get a foot in the door in those cities, the problem would be solved so fast that it would bankrupt millions of asset holders.

This reminds me of the time I was defrauded. The fraud was never going to pay his debts. When he gave a reason why he couldn't pay, it was tempting to put effort into solving that problem. But, when that problem was solved, he simply had another reason not to pay. The only solution was to sue him so that he might be made to pay against his will. Solving Yglesias' problem is a fool's errand.

Mark writes:

In a closed system, there is no conflict at all between 1 and 2. If we assume pretty much everyone lives somewhere, then building high end units, once people move into them, increases supply of whatever kind of units the renters or buyers moved out of. Unless some equal number if high end units are being destroyed elsewhere, then somewhere along the line, some people moved out of low end or 'middle end' units to move into high end ones (either ones that were just built, or the ones formerly inhabited by the people who just moved into the ones that were just built). Ergo, building luxury units for which there is demand necessarily increases the supply of lower end units.

The only way anti-development progressives' reasoning begins to make sense is if one ignores the big picture and focuses only one one's own city: of the new high end units in Seattle are instead be expanding the supply of (and making more affordable) lower end units in suburbs of Seattle, or worse, Chicago or Des Moines (or where ever people move from to Seattle), then who cares? We only care about Seattle proper; poor people everywhere else don't matter. That seems to be the moral premise. Doesn't seem particularly virtuous to me. It also seems to be rooted in the same fallacy as nationalist mercantilism, only inverted: instead of quashing demand for foreign goods to subsidize domestic ones to benefit employers, they're trying to quash foreign demand for domestic (or, within-city) goods to keep prices down for the ostensible benefit of native consumers.

So, the way I see it, the only way one can justify opposition to developing high end units is if one doesn't believe rich people necessarily live somewhere; or that when you build a high end unit, rich residents just spawn ex nihilo to buy or rent them. Alas, the spontaneous generation of rich people doesn't seem like a persuasive theory.

Jaap Weel writes:

I wouldn't say that last thing about Californians not even understanding the problem. A lot of them do. There's a political battle to be had, but it's going on, and the people who are on your side in it are, by national standards, just as much on "the left" as the ones you are opposing. Literally just hours ago, we had a major statewide upzoning bill introduced. By a senator who's been building a name for himself with housing reform bills that have actually passed.

Scott Sumner writes:

RSF, Newport Rhode Island.

Andrew, I was only 13 years old in 1968, and yet even I knew what the Vietnam War was about trying to prevent the communists from taking over Vietnam.

Do you think young Americans suddenly became more afraid to fight between 1941 and 1968?

And I disagree on the 1896 question as well, deflation was hurting farmers.

Luis, Yes, it's true that some did not understand the nature of communism.

Weir, That's a false belief, but I agree that many young people think that way.

Kevin, Certainly one problem is of trivial importance compared to the other.

Mark, That's right.

Jaap, If that's the bill I've read about, it's not the solution.

Wayne Lusvardi writes:

The writer is right but for the wrong reasons.

The housing market is a "move up" market. When families sell their older homes and move up to newer housing they "create" more affordable housing for others. This is called neighborhood filtration.

Older housing often has obsolescent features (old heating-air conditioning systems, lack of insulation and energy efficient windows, insufficient electrical capacity, etc.) and are in older neighborhoods that lack transportation linkages and newer commercial retail centers.

So to create for "affordable" housing, build new, more expensive housing that ends up creating more affordable older housing at the bottom of the housing ladder.

Hazel Meade writes:

A sometimes overlooked component of this problem is that San Francisco makes it very hard to evict renters even if they aren't paying rent. So landlords will often keep properties unoccupied until they find a renter they trust. This incentivizes pricing units out of the range of poor people and building luxury units rather than mid-range units or low-cost units. It probably also combines with various civil rights laws - if you own low-priced rental units you HAVE to rent them to all comers, there's a limit on your ability to discriminate due to credit history. So as a result landlords just don't want to own low-cost units. If they do, they will renovate them to turn them into luxury units, so they can rent them to someone with a good rental history and a stable job.

Bob Graboyes writes:

RSF: Pretty sure the picture is Swannanoa, Augusta County, Virginia.

Jim writes:

Housing costs increase linearly for each story added up to about 40 stories so there is a ton of basically unused housing supply just sitting out there. Furthermore, without all the asine regulations, the city of houston is able to build 40 story buildings for $350/sq foot (or less). Meanwhile 1-2 story homes in downtown SF and NYC are selling for $1000 per square foot. It'll never happen, but if we could convert all these 2 story houses to 40 story condo buildings, we'd have a 300% increase in population in SF and NYC and housing prices would drop 70% (a good thing)

mariorossi writes:

While I mostly agree with everything that you say, I think it's interesting to think of situations in which the basic equilibrium you describe wouldn't work.

I think many people have a different fear when they look at new luxury housing: I think they fear luxury housing is a way to store wealth without nvesting it. Imagine that luxury property gets built without been lived in or rented out. That sounds impossible in many ways (why wouldn't you not rent it out if you can? that's just extra income), but I think that's a wide-spread fear.

And even more interestingly it seems to be something that actually happens. Or at least it happens in London. Luxury flats are not purchased with the intention to rent or live in them, but simply to own them. I guess you could blame the frictions in the rental market, but it does actually happen.

In a world in which people value crypto currencies at some $500 billion, is it that crazy that people want to build houses just to own them? When government bonds yield are negative, is investment in empty property necessarily crazy?

I don't think it happens enough to really invalidate the basic story: build enough of them and the housing situation is bound to get better at some point, but I think there is something wierd going on. The desire to store rather than invest wealth is visible everywhere I think.

It bothers me a lot, since I think it is dangerous. It almost makes me feel that risk capital which is what trouble drives productive investment has contracted significantly in teh last 10 years, even if we don't understand why...

I might be simply falling in the investment != saving fallacy, but it bother me.

Mr. Econotarian writes:

I saw this happen in Washington, DC. From 1990 to today, those living in urban single-family dwellings and suburban multi-family dwellings moved into suburban single-family dwellings (with better school systems), and younger people with higher skills and income moved back into the city living in urban single-family dwellings as well as new "luxury apartment" urban multi-family dwellings. As the higher skilled folks age, some seek new "luxury houses" in the exurbs or teardowns in the inner suburbs, depending on their taste for commuting and mass transit.

Mike S. writes:

RSF, it's the Breakers in Newport, Rhode Island, summer home of Cornelius Vanderbilt II.


Kevin writes:

Generally, I agree, but you are conveniently ignoring that new construction drives the prior residents out of that neighborhood. Your solution of just build more supply does not perfectly solve the first goal Matt proposes:

"In one usage, the goal of improving affordability is to make it possible for more people to share in the economic dynamism of a growing, high-income city like Seattle."

The construction of class A new supply forces prior low income residents to other parts of a city which may feature lower economic dynamism.

Per Kurowski writes:

It is always better to allow our young adults to afford homes than to give them affordable homes. Go and ask a bank how much it needs to hold in capital when giving a mortgage, and how much it needs to hold when lending to an entrepreneur

Dangerman writes:

There seems to be a deep implicit assumption about the durability of houses in there.

Namely, that the "move up the ladder of quality" approach only works if the newly built [high quality] housing stock is durable enough to last until the rich people move up to the next rung in the future.

But what if it's not?

What if we build $1mil mansions that only last 20 years, then tear them down to build new $2mil mansions?

Ben writes:

Imagine if the public had the same fetish for building affordable new cars that it has for affordable new housing. Imagine that we required car companies to make a certain number of new cars available at discount on a means-tested basis. Or reserved leased cars (in a limited amount of course) for existing lessees, with price increases tightly controlled by local political appointees. Or limited the number of cars that could be built, out of outrage over excess profits taken by car companies. Or simply capped the price of a new car.

The results would be absurd. Cars--new and used--would be in short supply. Existing owners and their heirs would cling to their car-control rights, while others would be priced out of the market for battered old clunkers.

Thank goodness that, in reality, nobody expects that everyone has the right to an "affordable" new car. People understand that car companies build for people with the means to drive new cars, and that those cars filter down over the years to the rest of the population through the used car market. True, some new cars (the Maybach or Lamborghini) may never become affordable, some people may buy more car than they need, and some people will end up with the clunkers. But people generally understand that the system works well for giving people the cars they need.

P Burgos writes:

Where did you get those numbers on building costs increasing linearly up to 40 stories? I'm not disputing the accuracy of the numbers, just curious as to where they come from.

John writes:

[Comment removed pending confirmation of email address. Email the to request restoring this comment and your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Scott Sumner writes:

Wayne, You said:

"The writer is right but for the wrong reasons."

Actually, that's exactly what I said.

Kevin, You said:

"Generally, I agree, but you are conveniently ignoring that new construction drives the prior residents out of that neighborhood."

Actually I didn't ignore that issue, please read my post again.

Daws writes:

I recently wrote an article in Seattle Weekly that focuses on development's significance for workers in Seattle's building trades. Trying all ángles, fellas

Picador writes:

[Comment removed. Please consult our comment policies and check your email for explanation.--Econlib Ed.]

Steven Drury writes:

That's "The Breakers", Cornelius Vanderbilt's summer home in Newport, Rhode Island. It was on my paper route when I was a kid. We lived in a medieval hovel about a mile away.

Kaz writes:

Forgetting the apathy point for a second, it's unclear what your policy proposal is. It seems to something akin to 'build more housing' or abolish restrictive barriers to new housing construction.

However, your response to Jaap above muddies the waters. The recently introduced California bill is the most radical liberalization of housing construction that I'm aware of. How is a radical upzoning of large parts of SF and other cities, near high quality transit, insufficient to allowing much more "unaffordable" housing?

Generally, I'm unfamiliar with your prior pieces on housing affordability. It appears that you're using monetary policy as your primary lense. I'm personally happy that more economists are more greatly considering the topic. With that said, it appears that much of the work laid down by other economists that have studied and written about this topic are unaddressed or ignored.

For considering the affordability question, I'd recommend looking to Glaeser's work. He breaks affordability into two parts:
-poverty, housing offered at production costs will always be unaffordable
-insufficient supply, overly restrictive housing regulations have led to market-based housing costing much greater than the production costs

References to some of Glaeser's work and book on the subject:

Jim writes:

@P Burgos

Here are a few links:

also from the book "truimph of the city" by Edward Glaeser

"In New York City, the price of building an additional square foot of living space on the top of a tall building is less than $400. Prices do rise substantially in ultratall buildings, say over fifty stories, but for ordinary skyscrapers, it doesn’t cost more than $500,000 to put up a nice, new 1,200-square-foot apartment. The land costs something, but in a forty-story building, a 1,200-square-foot unit is only using 30 square feet of Manhattan, less than a thousandth of an acre. At those heights, the land costs become pretty small. If there were no rules restricting new construction, then prices would eventually come down to somewhere near construction costs, about a half million dollars for a new apartment. That’s a lot more than the $200,000 that it costs to put up a nice 2,500-square-foot house in Houston but a lot less than the $1 million or more that such an apartment now costs in New York."

Tagore Smith writes:

I’ve always wondered about this myself, and not just in terms of the expensive cities like NYC and SF. It seems to me that during my lifetime there have been a couple of shifts.

In many respects we are far better off, across the board, than we were 40 years ago. But having the necessities does not seem to have become concomittantly easier. Indeed, it seems to have become more difficult at the bottom. And this is mainly a matter of the cost of housing.

I’m inclined to think that this is mainly because, regardless of what people say, they do not really want to live near people poorer than they are. That housing is restricted in more desirable areas is what makes them more desirable.


Return to top