Working Papers

The Long and the Short of Housing Wealth and Consumption

Charles Calomiris, Stanley Longhofer and William Miles, on why there is no wealth effect on consumption from changes in house prices:

any decrease in house prices hurts only those who are net “long” in housing, that is, those who own more housing than they plan to consume. This might include, for instance, “empty-nesters” who are planning on selling their current houses and downsizing. On the other hand, the decline in home values helps those who are not yet homeowners but plan to buy. Most homeowners, however, are neither net long nor net short to any significant degree; they own roughly what they intend to consume in housing services. For these households, there should be no net wealth effect from house price change. And when one thinks about the economy as a whole (which is a combination of all three types of households) the aggregate change in net housing wealth in response to house price change should be nearly zero; changes in house prices should affect the distribution of net housing wealth, but have little effect on aggregate net housing wealth. Thus any effect from net housing wealth change on aggregate consumption spending should be similarly small.

Put differently, an increase in house prices raises the value of the typical homeowner’s asset, but such a price increase is also an equivalent increase in the cost of providing oneself housing consumption. In the aggregate, changes in house prices will have offsetting effects on value gain and costs of housing services, and leave nothing left over to spend on non-housing consumption.

The authors also debunk the work of Karl Case, John Quigley and Robert Shiller.

posted on 23 June 2009 by skirchner in Economics, Financial Markets, House Prices

(2) Comments | Permalink | Main

| More


That may be the case if the house price-rent ratio remains steady. But what if changes in prices do not (fully) flow through to rents as has been the case in Australia over the past 15 years? Under these circumstances, even those homeowners who are not ‘long’ housing benefit from higher prices as the imputed cost of ‘providing themselves housing consumption’ has not risen by as much as the price of their asset. So, does this argument rely on mean reversion on the price-rent ratio, and if so, I thought you were of the view that this was unlikely to happen?

Posted by .(JavaScript must be enabled to view this email address)  on  06/23  at  10:08 PM

The rental yield is loosely tied down by long-run relationships with other asset classes, but that is more of an arbitrage condition rather than a simple mean reversion process.  What I question is not so much the equilibrium relationship, but the assumption that all of the adjustment must come via prices.

Posted by skirchner  on  06/24  at  09:29 AM

Post a Comment

Commenting is not available in this channel entry.

Follow insteconomics on Twitter