Market Power in Housing Construction

Part 1 of a Cantankerous 3 part series.

I’ve recently seen some bad takes that bothered me enough to do some grumpy old man posts about them.

The mythology about the causes of the 2008 crisis seeps deep into the cracks and crevices of the academy and the culture. You could be writing an article for a sociology journal, or a think piece for the Atlantic, or a script for a sitcom. In any of them, you could reference the glut of homes in 2008 or you could refer to the housing bust as an inevitable collapse of a lending bubble. You could reference it without citation and with full confidence. These myths are canon. You don’t need to present evidence to assert them as truths.

There is a widely held notion that the homebuilding industry has oligopolistic power, which I think is related to the misdiagnosis of 2008. The lesson builders learned in 2005 was that a highly cyclical marketplace doesn’t serve them, and so they have voluntarily metered their production in order to keep supply low, prices high, and cyclicality minimized. So the story goes.

The story goes that either they learned their lesson in 2008 and aren’t so quick to expand now, or capital markets were disciplined, or concentration of builders has increased so that they can engage in more oligopolistic forms of anti-competitive behavior. And, this generally leads to some anti-YIMBY conclusion that even if you make it legal to build homes, builders will choose not to meet demand. Or it leads to some anti-market conclusion that you need a big public housing budget to counteract the pro-cyclical behavior of the builders.

This came up previously in a debate between Ezra Klein and David Dayen. I wrote a post about their debate. Dayen cited a version of this idea. In part, he wrote, “Because we have limited public funding for housing in America, we are at the mercy of the free market. With no social-housing sector, whenever private developers pull back, affordability suffers. And when the housing bubble collapsed, for-profit homebuilders resisted risking capital on what was seen as a hazardous business.”

Unfortunately, Klein basically stipulated to this, writing, “He argues, for instance, that part of the housing crisis is insufficient construction in the aftermath of the Great Recession. True enough... Yes, housing manufacturers misjudged demand 10 years ago.”

No. It isn’t true enough. This subtly but substantially harms Klein’s very important cause. If anyone has access to Klein’s ear, please forward that post to him. I don’t know if he has seen it. If I may be so bold, Klein needs to see it.

I go into many points in that post.

Also, I will add that this is an especially incredible thing to believe right now, because US home construction markets have been in a very unusual state lately. Demand outpaced the capacity for supply channels to support homebuilder sales. And so homebuilders have been selling homes they couldn’t complete in a timely manner with excess margins while bidding up the price of inputs to try to complete the homes they have sold.

The cost of inputs has barely started to unwind and bloated inventories are still not fully normalized. That is not the behavior of an oligopoly. Stressing the communal capacity and driving up input prices in an effort to get rid of bloated inventory is quite the opposite of a market of producers purposefully metering sales to avoid cyclicality. Margins are temporarily high, and builders have metered sales because of bloated inventory that was waiting on inputs. So I suppose if you squint hard enough you can pretend that the builders aren’t clamoring for more sales where they can get the inputs to complete them.

Of course, part of what’s tricky here is that builders were scared after 2008. They were afraid of going through that again. And many of them think the problem in 2008 was that the market in 2005 was too cyclical. So, many of them might tell you this story about themselves, or about their competitors. But, if macroregulatory conditions had allowed sales to be quadruple what they were in 2010, there would have been no lack of builders lined up to supply them.

Here, let me just reiterate the basic narrative correction to the myth.

Here is a chart, from my paper, Build More Houses, of the annualized rate of housing starts, with the peak of each building cycle matched to time t=0. Before the 1980s, there were cyclical building cycles. Since the 1980s, the rate of new construction has been basically flat for 40 years, except for the huge, one-time drop after 2005.

During the 2000s, the tract home builders were in a boom. You can see from the Figure it didn’t add up to a boom in total building. Basically, we were increasingly building tract homes from large homebuilders instead of all the other types of homes that have been slowly regulated away - apartments, multiplexes, manufactured homes, and homes built by owners or contractors.

But, the country became taken with a moral panic about housing and building, and convinced itself that the orange line at t=0 had created an unprecedented supply glut which we had to purposefully stop. First that was accomplished with monetary policy, but the larger and permanent hits came from mortgage regulation. Builders lost more than half of their market. New single-family home sales are still half what they were in 2005, nearly 20 years later.

There is a paper that is a popular citation on this topic, from Luis Quintero at Johns Hopkins University. It claims a causal connection of more industry concentration leading to reduced construction and higher prices. Of course, I am vain enough to take a position, from time to time, against industry insiders myself, so I can’t really complain too loudly, but that paper seems like a good example of how it helps to just go talk to somebody. There is no way that anyone who has worked in the industry, or for that matter sat in on a builder’s quarterly conference call or read their SEC filings, could think that over the last decade, builders were holding back production oligopolistically.

One thing you might look for, if you thought that was happening, is that builders might be reducing the number of communities they were opening up in order to keep sales per community high, which boosts margins. But until the Covid boom, sales per community were persistently low in the industry. Rarely has an industry been so desperate for any marginally profitable revenue for so long.

To be frank, I think that paper could be useful as a statistical exercise. What is the weakness in the methods and robustness checks that could allow them to statistically confirm that consolidation caused lower production rather than the other way around? Are those methods being used in other papers in ways that might not be as dependable as those authors think? It’s sort of a “physicists prove bumblebees can’t fly” situation.

And, also, I think the ease with which people accept this framing, across ideologies, is an example of the ancient, deep seated prejudices we have about finance and commerce. Corporations will always be “the other”. The coordination that has emerged in modern economies is so diffused and complex that the institutions that direct that coordination can seem inhuman. This is sort of in the category of blaming rich foreigners for buying up the nice homes, or Latin American immigrants for bringing in drugs, or homeless people for creating their own problems. We - you and I - behave according to the constraints of the world. They behave according to their greed and shortcomings. The blame for low housing supply can be as directed by ingroup vs outgroup status as the blame for high housing demand.

There is a similar anti-corporate sentiment that I am becoming increasingly fearful of. There are now bills being submitted to enforce it and it is now common at any public hearing for speakers to complain that Wall Street investors and private equity are buying up homes and pricing out families.

They pay too much for homes while also raking in the bucks, which they accomplish by setting the rents too high. It’s a great example of a narrative that completely eschews natural constraints and assumes that all of the facts are driven by unrestrained bad motivations of the outgroup. When you and I are the highest bidder for a home, we’re not pricing other families out of homes. We are families. We’re the ingroup. And if we rented an investment property out, we would set the rent at a reasonable level, which of course would need to be high enough to cover our expenses and provide a reasonable return. And if we pocket a half million in capital gains when we sell our homes it’s because we and our neighbors worked hard to keep the neighborhood nice, unlike those nasty outsider corporations who buy up neighborhoods only to jack up the values out of pure greed.

They are the only marginal source of increased new housing under current laws, and if the populists ban them, Lord help the working class renters who have already suffered unprecedented rising rents.

Back to the builders. We killed their industry*, put millions out of work, bankrupted countless firms, large and small, and when the dust clears and we are still suffering the consequences of our communal errors, a plurality of us can lean back, take a knowing puff on our pipes, and say, “You know, I’ll bet they’re responsible for this.”

Don’t do this. If defending corporations as a group was fashionable, this rhetoric would be obviously offensive. But, more importantly, like all forms of prejudice, it tends to hurt the prejudiced observer because the conclusions that come from prejudice satisfy us to stop looking for the actual causes of our communal problems.

*You might reasonably be annoyed by that sort of language. “We did this” and “We did that”. You might ask - reasonably - “Who is this ‘we’?”. I give myself permission on this topic to use such language because the destruction of the American economy and housing finance in 2008 was the most popular political movement of my lifetime. In the summer of 2008, looser lending standards and monetary stimulus should have been implemented until housing starts rose, home prices stabilized, and Fannie and Freddie could reasonably assume that there won’t be billions of dollars of future collateral losses. That would have been a quite neutral and minimal set of expectations to aspire to. Please send me any examples of anyone making that argument at that time to an audience who responded with “Those sound like reasonable goals.” You can’t. There aren’t any. To a reasonable approximation, we did it.

Ironically, the subtitle of Atif Mian and Amir Sufi’s “House of Debt” is “How They (and You) Caused the Great Recession, and How We Can Prevent It from Happening Again”. The popularity of “House of Debt” as an explanation for the damages of the crisis is, ironically, at a meta level, a study in how we caused a recession.