Join our community of smart investors
OPINION

As unsafe as houses

As unsafe as houses
February 10, 2015
As unsafe as houses

Katharina Knoll, Moritz Schularik and Thomas Steger, three German economists, have compiled house price indices going back to the 19th century. These show that, between 1899 and 1938, UK house prices actually fell in real terms, by 1 per cent a year. This was not a quirk of the UK. The authors show that on average across 14 developed economies, house prices were flat in real terms between the late 19th century and mid-20th century. It is only since the 1950s that house prices have risen much.

If this long sample of history is representative of likely future returns, then housing is a poor investment. UK house prices have risen by only 0.4 per cent per year since 1899. That compares with a total return on equities (including dividends) of 5.6 per cent. Such a low return is not enough to compensate us for the risk of house prices falling and - worse still - of the risk that they'll fall during a recession when we lose our job and so most need wealth.

But is the pre-war experience really relevant? It's arguable. To see why house prices fell then, remember John Betjeman's film, Metroland, in which he showed how the opening of the Metropolitan railway led to the expansion of small villages such as Pinner and Northwick Park. This was a specific instance of a global phenomenon. The spread of railways allowed people to commute to work and so increased the supply of usable land. Houses in Harrow became substitutes for those in Camden Town, so prices of the latter fell as the supply of the former surged because the Metropolitan railway encouraged a boom in housebuilding.

Technical progress in the transport industry thus reduced house prices. It is only after this process stopped in the mid-50s that house prices exploded. (The boom might have been helped by the introduction of planning restrictions in 1947, but that doesn't explain why house prices showed a similar pattern to the UK's in most other developed economies.) As David Ricardo pointed out in the early 19th century, the price of land (and indeed most other things) depends upon a race between diminishing returns and technical progress; diminishing returns raises prices and technical progress reduces them. Between the 19th and mid-20th century, technical progress won. Since then, diminishing returns have.

Will this remain the case? Maybe not. HS2 could allow houses in Birmingham to substitute for London houses in the way that the Metropolitan line allowed houses in Harrow to do so. And if driverless cars reduce the discomfort of being stuck in traffic jams, they might encourage longer commutes and hence more substitutability between houses.

There has, though, been another form of technical progress - broadband. This has enabled your correspondent to work from home in Rutland. Just as the Metropolitan line allowed bank clerks to move from overpriced central London homes into more reasonably priced areas, so broadband has enabled me to do the same.

But, of course, I'm unusual. Broadband has not created a sufficiently big rise in homeworking to cause significant migration out of London. And it certainly hasn't triggered the housebuilding boom that the Metropolitan line did.

But this might just be a matter of time. As Andrew McAfee and Erik Brynjolfsson point out in The Second Machine Age, it can sometimes take decades for the applications of a new technology to be fully used. On the other hand, it could be that homeworking is only feasible for a minority of employees and so broadband will never have the revolutionary impact on house prices that the railways did - although the reasons for this might be psychological and political rather than technical.

Whatever the answer to this question is, there's a point here which property investors have overlooked. Investing in housing isn't just a bet upon incomes, interest rates and idiosyncratic local factors. It is also a bet on the pace and direction of technical change. While this might be reasonably predictable in the near term, it is much less so in the very long term. In this sense, longer-term property investors face a lot of unavoidable uncertainty.