The prospect of rising mortgage rates lends new importance to an old question: which is the better investment, housing or equities? I suspect the answer is the latter, but only just.
My chart provides context here. It shows that, except for the 1990s and early 2000s when shares outperformed housing before falling back during the tech crash, house prices haven’t moved much relative to equities since the early 1980s. In fact, the ratio of the two is the same now as it was in 1983.
Of course, this omits a lot of factors influencing the investment choice between the two. Housing has the advantage of a higher yield than equities; nationally, rental yields have usually been above dividend yields. And it’s easier to borrow to buy a house than equities. But, on the other hand, it’s easier to have a moderately diversified equity portfolio than a housing portfolio: you can easily buy a global equity tracker, but a house is stuck in a particular area. Equities don’t expose you to out-of-pocket costs such as insurance, heating bills, broken boilers or bad tenants. And you can put equities into tax shelters, which you cannot do for second houses. The weight you put upon these factors varies from person to person, so let’s put them aside and consider how the house price-All-Share index ratio can change.
There’s one thing that doesn’t much affect it – recessions. Looking at my chart alone it is hard to infer when the UK has been in recession. That’s because recessions cut both equity and house prices, without having much effect upon their relative prices.
Instead, one big influence upon this ratio is simple mean-reversion. When house prices have been high relative to equities they have subsequently fallen relative to them, and vice versa. The correlation between the ratio and subsequent three-yearly moves in it has been minus 0.54 since 1970.
With the All-Share index now low relative to house prices, this points to equities outperforming over the next three years. But given that the ratio isn’t hugely out-of-kilter, we cannot be very confident of this.
A second influence upon the ratio is interest rates. Since the Bank of England was given greater independence in 1997 there has been a slight positive correlation between three-year changes in five-year gilt yields and in the All-Share index relative to house prices: shares do better than houses when yields rise. To put this another way, leveraged assets such as housing do worse than others when interest rates rise.
This too might seem to favour equities over housing. But the correlation is small. And we cannot forecast the course of bond yields at all accurately. So, again, we must be cautious.
A third influence are profits and wages. Obviously, share prices are a claim upon (some) future profits. But house prices can be seen as a claim on future wages; how much people can afford in rent or mortgage payments is set by their wages. This means that actual and expected changes in the shares of wages and profits in GDP will change house prices relative to equities. Shares fell relative to house prices in the 1970s because wages fell relative to profits, a development investors feared might be permanent. When profits recovered relative to wages in the late 1970s and mid-1980s equities outperformed houses. And then equities fell relative to houses in the early 2000s as the profit share fell.
You might think this is now a reason for favouring houses, because a tight labour market might cause another squeeze on profits.
Don’t bet on it. For one thing, except in a few industries the labour market is not tight; we know this because real wages are falling. And for another thing, significant protracted moves in aggregate wages relative to profits are rare. The share of profits in GDP hasn’t changed much in the last 15 years and is in fact much the same as it was in the early 1970s. Back in 1957 Nicholas Kaldor proposed as a “stylised fact” that the shares of profits and wages in GDP were stationary over the long run. That has been true since. Which warns us not to base a view of the attractiveness of equities relative to houses upon forecasts of moves in those shares.
There is, however, one thing that does favour houses: political intervention. There are help-to-buy schemes for houses but not equities. And these might be extended if prices look like falling. House prices are supported by a 'Sunak put', but equities are not.
Against this, equities have the advantage of liquidity: they are easier to sell than houses, especially in hard times. In theory, house owners should be compensated for this liquidity risk with higher returns. In practice, this doesn’t seem to have been the case. This advantage of equities is especially great to the extent that it is possible to avoid the worst of bear markets by reducing exposure after prices fall below their 10-month average or after the yield curve inverts.
But there’s something else. For many of us, housing is not wealth. If we own only our own home, so what if its price rises? We can only profit from this by borrowing against its increased collateral or by selling and moving to a cheaper house. But that means consuming less housing, which makes us better off in the same sense we would be if we spent less on anything else. And even landlords aren’t necessarily better off when prices rise, if higher prices mean lower future rental yields.
On balance, then, I would rather be invested in equities now than in housing (which I am) – but there is not as much in it as one might think.
