Prices, rents, and costs
The NBR has pointed to an article to the Economist that shows house price to income and house price to rent ratios – pointing out that the very high house price to rent ratio can be used as an indicator that the return on housing is very low/house prices are heavily overvalued.
Now one criticism that people may raise is that the “quality” of the housing stock and the rental stock has changed – and so the relative prices/spending from income could indeed change. However, the Economist uses figures from Quotable Value New Zealand (as well as Stats NZ) – and so the quality of housing is in fact taken into account in these indices! [Note: It is not necessarily clear the categories are comparable – so this argument could still be used]
As a result, we could say that this is true – the return for an investor in the housing market seems pretty low.
But what else can we tell from all this? Relative to historic averages, the price to rent ratio is 68% higher, and the price to income ratio is 20% higher. So this implies that price/rent is 168% of its average, and price/income is 120% of its average … which tells us that rent/income is 71% of its long-run average. If we believe these figures, the rental cost … the cost of actually consuming a housing service relative to income … is very low! [Update: So this is consistent with rising living standards, and having to spend less on housing services – a nice foil to all the suggestions that housing costs have been eating into incomes!]
I’m not sure how much I trust these figures overall, price to rent ratios in this index has been rising constantly over the last 40 years implying that there may be a “quality adjustment” issue in the data to me.
However, if we do use these figures to say that house prices are too high – they also tell us that rents are too low. Any explanation we have needs to explain both of these parts of the data. [Update: This is not clear from the data – and is actually a misleading statement, so just ignore it. In truth, we need to ask how much of the adjustment will occur through rents and how much through prices]
This is an odd question. This is like saying “If we use earnings figures to say that stocks are overpriced, we need to also explain low earnings”.
We can simply say that both earnings (and rent) are correct, and in a long term model should determine the value of stocks (houses). The fact that stocks (houses) are priced out of kilter with long term ratios only says that the stock (or house) is overpriced, not anything about rents or earnings.
I would say that it isn’t clear to what degree either rent or prices are out of whack – but I do agree that p/rent should be the more constant long-term relationship. Note that I say that the return for an investor in the housing market seems pretty low above.
My initial goal was to look at rent to income to discuss how the cost of purchasing housing services has fallen – but somehow I ended up running into completely the wrong point. When I looked at my own data, the decline in the Economist index over the last 40 years appeared to be a lot steeper – which lead me to the point that I think there is a quality adjustment issue in the data, but that isn’t clear here. Also, the fact that p/rent has experienced an upward trend over the whole 40 year period is suspicious – as this is the very series we would expect to mean revert when adjusted for quality (although specific issues, such as population demographics, are likely to have a relatively long lasting albeit temporary impact)
Also re-reading it, it does sound like I am suggesting that r/i should revert to trend, which was unintentional. I will amend 😉
Rent/income being low is interesting but doesn’t need quite the same explanation that high house-price/rent ratios do.
Houses can be thought of as financial instruments that have a flow of payoffs (rent) or a durable good that gives housing services over time. Under the first interpretations, the price of a house is function of rent and a bunch of other stuff like the interest rate and time preference. Same too with the second interpretation, because you could get the same flow of housing services from renting a house. If we think the interest rate and time preference are pretty stable in the long run, there should probably be a stable long-term relationship between rent and house prices.
Rent and income has a different relationship. Their tied together by people’s preferences rather than by financial pricing (and the arbitrage implied). Houses might be luxury, normal or inferior goods at different levels of income and there might be unbalance productivity growth complicating things further. Thus there’s no real reason to think rent and income will have a long term relationship, just like any other price. Should the price of bananas to income ratio converge to a long run average? What about car prices?
“Houses might be luxury, normal or inferior goods at different levels of income and there might be unbalance productivity growth complicating things further. Thus there’s no real reason to think rent and income will have a long term relationship, just like any other price. Should the price of bananas to income ratio converge to a long run average? What about car prices?”
Indeed, this is a fair point. As living standards increase I would expect rents relative to income to fall – and in fact that is exactly what has occurred. My initial comments completely ignore this fact unfairly, so I will point to this 😉
Given this fact that are in fact two things we can pull out of the data:
1) In so far as the intrinsic value of owning vs renting is unchanged, the cost of housing services relative to incomes has declined – in this way NZ society has become more wealthy.
2) The recent sharp increase in the rent to income ratio does not appear to be consistent with a “shortage” of property in recent years.