Buy your home and kill a job

Andrew Oswald argues that all the usual explanations for unemployment are completely wrong

Economists have tried for a long time to understand why joblessness in the leading industrial countries has risen from around 2 per cent in the 1950s to approximately 10 per cent today. The standard explanations are that trade unions have become too powerful, unemployment benefits too generous, labour too highly taxed, the workers too inflexible. No doubt each of these explanations plays some role in the whole answer. The trouble is that the facts do not provide much support for these otherwise plausible ideas.

For example, those countries with the worst unemployment problems do not have especially strong unions: Spain, with 18.9 per cent unemployment (the highest in the OECD), has only 11 per cent of its workers in unions, while Sweden, with only 4.4 per cent unemployment, has more than 80 per cent in unions. The correlation between unemployment and payroll tax is just as weak: again, Sweden, with low unemployment, has high taxes on labour (37.8 per cent), while Ireland, with high unemployment (14.8 per cent) has low taxes (7.1 per cent). There is some evidence that countries that hand out high benefits have slightly greater unemployment; but the size of the effect is not large and it is not consistent. For example, the Netherlands, with relatively low unemployment, has relatively generous benefits.

The real answer, I believe, is staring us in the face, and yet we almost invariably ignore it. Consider the following. Spain, with the highest unemployment of the major industrial nations, has the highest rate of home ownership. Switzerland, with the lowest unemployment, has the lowest rate of home ownership. The US (hard though we may find it to believe now) had the highest unemployment in the late 1950s and early 1960s; at that time it had the highest rate of home ownership.

Those are startling facts but they are not statistical quirks. If you plot a graph of unemployment against home ownership in North America and western Europe in 1960, it shows a clear relationship. If you plot a similar graph for the 1990s, the relationship still exists. More home ownership, more unemployment. The relationship still holds even if you take into account other possible influences such as the proportion of women working, the age structure of the population and interest rates. Further, the relationship remains robust even within nations. Thus in the Swiss cantons and in the US states between 1970 and 1990 a growth in home ownership and a growth in unemployment are closely related.

Economists ought not to be so surprised. The underlying argument is that a country's "natural rate" of unemployment depends on the ease with which its people can move around to find jobs. Milton Friedman made exactly this point in his highly influential address to the American Economic Association in 1967. He did not mention the housing market, yet the figures showing the correlation between home ownership and unemployment were available when he made his address. Had he stumbled upon those figures and included them in his address, it is possible that European history of the past 30 years would have been very different.

In the 1950s it was the US that looked enviously at Europe's low unemployment and worried about whether it, too, should adopt a generous welfare state and strong trade unions. But it was Europe that copied the US - in its property market. Nearly all European nations (Switzerland is an exception) have, by conscious design, seen a strong growth in home ownership since the 1950s, while US home ownership (and its unemployment rate) has remained constant. Spain, for example, now has a home ownership rate of 80 per cent, compared with less than 40 per cent after the second world war. Governments continue to offer subsidies that persuade people to give up rented housing and become owner-occupiers. They also exempt owner-occupied houses from capital gains tax. Yet economies need to be adaptable. In an ever-changing world, workers have to move to find new jobs. Renting allows such mobility, enabling a square peg in, say, Zurich to drop into a square hole in Geneva.

I cannot claim that I fully understand the mechanisms behind the extraordinary correlation between low home ownership and low unemployment. But here are five likely links in the chain. First, selling a home and moving is expensive; that's why many homeowners who lose their jobs will commute long distances to find work. Owner-occupiers are less mobile than people who rent and therefore more vulnerable to economic downturns in their region.

Second, unemployed men and women cannot move into the right places. High home ownership levels block young people's ability to enter an area to find a job. Those without capital are at a particular disadvantage where ownership, rather than renting, is the dominant form of housing tenure.

Third, in an economy in which people are immobile, workers do jobs for which they are not ideally suited. This inefficiency is harmful to everyone: it raises costs of production and lowers real incomes. So prices have to be higher and real wages lower than in a more mobile society. Jobs get destroyed - or, more precisely, priced out of existence - by such inefficiencies.

Fourth, areas with high home ownership levels may deter entrepreneurs from setting up new operations. In the kind of Cotswold village where I live, with a few hundred owner-occupied houses, people will do everything they can, through planning laws and so on, to stop anybody setting up a factory or commercial enterprise. If we were all or mostly private renters, our stake in protecting the local environment would be much smaller and we would probably be more tolerant to incomers who bring jobs with them.

Fifth, homeowners commute much more than renters and over longer distances. This leads to transport congestion, which makes getting to work more costly and difficult for everyone. This in turn acts in the same way as higher unemployment benefits because it reduces the gain to be made from having a job. If it's expensive to get to work, the net effect is to increase, in the same way high dole payments do, the attraction of not working.

Finally, owner occupation persuades highly skilled people - who in the past would have moved to wherever work was most readily available - to settle for lower-skilled jobs near their homes. The result is that the unskilled or low-skilled people in their locality are bumped out of the labour market.

The conventional wisdom about unemployment - that it's the result of trade unions, over-generous benefits and so on - is based more on theory than on the weight of hard evidence. It is true that reverse causality is a possibility - that high levels of unemployment may lead to high levels of home ownership. But even apart from the inherent implausibility of such an argument, any analysis of the figures simply doesn't support it.

So my conclusion is that the housing market contains the solution to a puzzle that has defied economists and governments for a long time. We can put Europe back to work by reducing home ownership. That would require every chancellor and finance minister in the EU to alter his or her policies. But renting worked in the 1950s and it can work again.

The writer is professor of economics at Warwick University. His book "The Wage Curve", published by MIT Press, recently won Princeton University's Lester prize. This article is based on a paper given this month to the Chartered Institute of Housing