A big question is: where do house prices go from here? According to Halifax, house prices peaked in December 2007 and have fallen 17 per cent since then. Real house prices have fallen even further -- by around 27 per cent. Homeowners on trackers have done really well. Their payments fell sharply as interest rates fell to historically low levels after the Monetary Policy Committee (MPC) cut the Bank of England rate to 0.5 per cent. This has kept delinquencies down but it is unlikely to continue when interest rates rise. This will inevitably have a downward impact both on house prices themselves and, inevitably, on consumption also.

Based on house-price-to-earnings ratios (HPE), a measure of affordability, it does look as if house prices are unsustainable at current levels and hence still have quite a long way to fall.

Chart 1 (click here [2] for a bigger version) illustrates this, using data from the Halifax. The index stands at 4.45, compared with a peak of 5.81 in July 1987 and a long-run average from 1983 to 2000 -- prior to the house price boom -- of 3.64. The question is by how much. These numbers suggests that house prices have another 20 per cent or so to go, with the concern that, as has occurred in other house price corrections, there is a bigger overshooting before prices return to the long-run equilibrium. Interestingly, a comparison of gross rental yields, relative to a long-run average, also indicate that housing is at least 20 per cent overvalued.

But claims about the sustainability of HPEs come up against the counter-claim that low interest rates have made valuation metrics less useful as a guide to the sustainability or otherwise of prevailing house prices. Compelling new work by Paul Diggle from Capital Economics sheds some light on this issue. He argues that comparing house prices to equity prices, which should also have benefited from low interest rates, still suggests that house prices are about 15 per cent too high.

There are similarities, he suggests, between how equities and property "should" be priced. As a claim on a company's future earnings, the price of a share, he claims, should equal the present discounted value of the expected earnings to which it entitles the owner, with a suitable allowance for risk. An equivalent way of determining the "fair value" price for property is by using the present discounted value of the future stream of rental income, adjusted for risk and the costs of owning and maintaining property. So, Diggle argues, by lowering the rate at which future income is discounted, low interest rates should have benefited both asset classes. Even so, relative to a simple long-run average, the ratio between house prices and equity prices seems to suggest that either equities are around 15 per cent too cheap or housing is around 15 per cent too expensive. (See Chart 2 -- click here [3] for bigger version).

Given that the FTSE all share price/earnings ratio indicates that stock market valuations are very close to average historical levels, Diggle argues, there is little evidence for the former. The house-price-to-equities ratio seems to imply that house prices are higher than can be justified by low interest rates.

It is significant that the extent to which housing is overvalued on this new measure is similar to other measures, such as the HPE and rental values. The house-price-to-equities ratio adds to the case that a downward adjustment in prices is required. House prices look to be headed down.