How economists (should) think about the housing market

Professor Paul Cheshire of the LSE’s Spatial Economics Research Centre recently wrote a post seeking to rebut the evidence I’ve been presenting on whether inadequate housing supply is the cause of high house prices. However, the central case I have been making is left conspicuously unaddressed in his piece, which invokes house prices as indicative of a housing shortage.

This raises an interesting issue at the heart of this debate. There is an established body of economic theory in this area yet most housing commentators and policymakers, and some economists, appear to ignore it. It seems like we’ve forgotten what we knew about how the housing market (or indeed any asset market) works, and that this is at the root of a profound misdiagnosis of the housing crisis. In this blog I outline a simplified version of that theory in order to explain why housing shortage arguments based on house price data are flawed.

“No understanding of how housing markets work”?

Professor Cheshire writes that the housing shortage argument is based on “no understanding of how housing markets work”. But his central objection appears to be captured when he says: “A shortage can only be usefully defined in terms of the balance of supply and demand. Basic economics is enough to give at least a hint that if prices are persistently rising in the long term, as house prices have, supply is less than demand.” [Emphasis mine].

This is superficially persuasive. Supply and demand, innit?

But it fails utterly to deal with the essential insight from economic theory that lies behind my analysis: that housing costs, not house prices, are the market signal that tells us about the adequacy of the supply of and demand for places to live.* To put it another way, ‘basic’ housing economics is, in my view, clear that you cannot use rising house prices as prima facie evidence that there is a growing shortage of places to live, as Professor Cheshire does. So what does the economic theory say?

Standard housing theory framework

There are several papers on the economics of housing to choose from, mostly written in the 1980s, that all tell the same core story. One such foundational paper is MIT professor James Poterba’s 1984 QJE paper Tax subsidies to owner-occupied housing: and asset-market approach. The theoretical framework for housing costs that he (and others such as Dougherty and Van Order (1982)) sets out has subsequently been incorporated into inflation measurement by national statistical agencies around the world (including the ONS in developing CPIH). Poterba explains it as follows:

“The desired quantity of housing services depends on the real rental price of these services […] The flow supply of services is produced by the stock of housing structures. The stock of houses is fixed in the short run, so the equilibrium rental equates the demanded quantity of services with the existing service flow.”

In other words, rent is the ‘price’ that equalises the supply of and demand for places to live. The implication is clear: if housing supply is failing to keep up with the demand for homes, we would expect to see housing costs — as measured by rent — rising relative to household incomes.

The same conventional economic theory also explains why it is that even huge house price increases are entirely compatible with a growing surplus housing stock. Poterba again:

“Individuals consume housing services until the marginal value of these services equals their cost […] Homeowners equalize the marginal cost and marginal benefit of housing services” and as a consequence the cost of owning equals the rent.

To simplify radically, the cost of owning is captured by the property’s depreciation rate, plus the interest rate (paid on debt or foregone on the equity in the house), multiplied by the value of the property. Which is why I argued here that the cost of renting and owning tend to be equal. The simplified version of the relationship is therefore:

Rent = House price x (interest rate + depreciation rate)

Why does this matter?

So when we see house prices rocketing, as they have done in recent years, this body of theory tells us that there can be two causes:** either rent is increasing rapidly reflecting a worsening shortage of housing services, or interest rates are falling, lowering the cost of capital and pushing up house prices, even though there are plenty of places to live. Or it could be both. To underline the point: outside of bubble conditions or radical changes to the taxation of housing, the driver of house prices has to be at least one of these two fundamentals.

Consequently, to establish whether the cause of house price woe is (even partly) a shortage of homes, we need only look at what has happened to real rent. This is the central point, covered most comprehensively in this blog.

What we are seeing is a picture of falling rents both in real terms and relative to average household incomes. The fact that the surplus stock of houses over households has roughly doubled in 20 years merely corroborates the story told by these price signals. Meanwhile the long term fall in interest rates (and mortgage market liberalisation) makes the real rocket fuel of rising house prices abundantly clear, as Chris Dillow and Simon Wren-Lewis have also pointed out.

Popular consensus versus theoretical consensus

So the housing shortage consensus appears to be entirely at odds with the theoretical consensus on how we should think about the housing market given the available data.

Unsurprisingly therefore, credible arguments for a housing shortage based in appropriate data are thin on the ground (to see that, take this classic mistreatment of the numbers). But if it is possible to build such a case, the putative advocate has two options: one is to explain why the economic theory described here is wrong or incomplete in some way; the second is to identify problems with the (numerous) data sources that, within that theoretical framework, tell a clear and consistent story that a housing shortage is not the problem. I’ve not come across much of a case for either.

The alternative is to accept that the data we have clearly indicates that supply is not the reason why houses have become so pricey.

— — — — — — — — — — — — — — — — — — — — — — — — — — — — — —

*You can of course use house prices as an indicator of the supply and demand for housing assets, but what counts is whether we have the appropriate supply of housing to keep day-to-day housing costs in check. Chasing the housing asset market leads you to Ireland, Spain and the US circa 2006.

**In practice, if you want to get into the literature, other factors could influence the price — taxation, depreciation of housing, and expected capital gain (though the latter gets complicated). However these don’t affect (and indeed underline) the central role of rent in diagnosing supply adequacy, and the core fundamental drivers are as described above.