Insofar as housing is an investment, the question arises: what proportion of one's wealth should one put into housing? For many people, of course, the question is redundant. Some are compelled to have a zero allocation because they can't afford to buy whereas others - those who bought in London years ago and cannot leave the capital - have a huge weighting.
However, for the typical middle-aged person, housing is a massive investment. Researchers at the IFS have estimated that for the median 50-65 year-old, their housing equity accounts for almost 60 per cent of their net wealth. This is twice the proportion held by your correspondent. This poses the question: is housing so attractive relative to equities to justify the typical person having such a huge holding of it?
This question can't be settled by expected returns as these are quite similar. Since the end of 1985, house prices have risen by 5.6 per cent a year while the All-Share index has risen 6 per cent per year. Allowing for sampling error - perhaps the last 30 years have been unusually kind to shares and nasty for housing - we can't rule out the possibility that shares and house prices rise at the same long-run rate.
In fact, there's a good reason why they should do so. Back in 1957, the late Nicholas Kaldor proposed as a stylised fact that the shares of wages and profits in national income shouldn't change much over the long run; subsequent events have been consistent with this. This implies that if the ratio of house prices to wages and the ratio of share prices to profits have no upward or downward trend, then both share prices and house prices should rise at the same long-run rate. Granted, the two ratios can change. But some big reasons for them to do so - such as changes in interest rates or expected GDP growth - should affect both of them in the same direction.
Nor can we distinguish between shares and houses by looking at risk, in the sense of volatility. Granted, Halifax's house price index has been less volatile than the All-Share index since it began 30 years ago. But nobody owns the house price index; they own individual houses. And just as individual stocks are more volatile than the share index, so individual houses are more volatile than the house price index. It's likely, therefore, that the volatility of house prices is as great as that of a reasonably diversified equity portfolio.
If volatility and long-run returns can't adjudicate between the attractiveness of housing versus shares, what can? Plenty.
For one thing, volatility is not the same as uncertainty. House prices might be as volatile as shares, but many people feel they are less uncertain. Bricks and mortar are something you can see, and many feel more comfortable holding them than they do shares. Housing, then, appeals to those who are averse to ambiguity. One study in Spain has found that people who distrust others tend to hold more housing wealth and less equity wealth than those who are more trusting.
Whether this is a rational motive is, however, questionable. Researchers in Italy have found that people who are more financially literate tend to own less housing and more equities - consistent with the possibility that a preference for bricks and mortar is irrational.
A second difference between housing and equities is that one offers a consumption benefit. We might prefer to own rather than rent because it feels good to be a property owner or because we want to grow a garden. This, though, comes at a price. Shares pay dividends, whereas it costs money to keep a house. Owning housing rather than shares therefore has an opportunity cost; a #50,000 additional investment in housing means we miss out on almost #2,000 of dividends each year. And in the long run, dividends matter a lot.
There's a third difference, which argues for holding equities - liquidity risk. Selling a house is a slow, expensive and stressful process, whereas selling shares is not; personally, I try to stick to the rule of never doing anything that requires a lawyer. This is an especially important consideration if, like me, you are approaching retirement and so planning on running down one's wealth.
A cognitive bias magnifies this problem - the endowment effect. We tend to overvalue things merely because we own them. Anyone who has tried and failed to persuade an elderly relative to move into a smaller, more convenient property knows that this is especially true for housing. I fear, therefore, that people who regard their house as their pension are underestimating the sadness they will feel when they have to sell; remember that we are bad at predicting our future tastes.
There's a fourth consideration bearing upon the housing versus equity choice - the role of human capital, or labour income. In theory, risks to our labour income should point to housing having a small share of our wealth.
One reason for this is that most of us own a house near to where we work. But this means we're putting all our eggs into the same basket, as we're betting on local economic conditions. If a big local employer closes, we'd see our house price fall at the same time as we might lose our job.
A second reason is distribution risk. If Kaldor is proved wrong and income shifts from wages to profits, we'd see both our human capital and house price fall.
Both these risks suggest that people who own lots of human capital should have a low weighting in housing and a higher one in equities, as the latter better protect us from these two risks. This, however, is typically exactly what we don't see. We tend to invest heavily in housing when we are young - when we have most human capital - and accumulate shares later on, when we are old and our human capital is smaller and less risky.
There are, therefore, reasons to suspect that a big investment in housing might be a sub-optimal asset allocation.
How convincing are these? Put it this way. There are people who are 'asset rich but cash poor' - pensioners living in big houses. But the opposite condition, of being wealthy but under-housed, is almost unheard of; of course, many people do live in grotty accommodation but these are the young and poor rather than the rich. This suggests that the error of over-investing in housing is far more common than the opposite error, of under-investing.
And this in turn suggests that high house prices have a downside, of exacerbating a portfolio misallocation. But then, rationality and the UK housing market have only ever had the loosest of relationships.
