CEPR - Center for Economic and Policy Research


En Español

Em Português

Other Languages

Home Publications Blogs Beat the Press Why Better Housing Policy Would Not Fill the Demand Gap

Why Better Housing Policy Would Not Fill the Demand Gap

Monday, 29 July 2013 05:24

Inspired by Noah Smith's tweets, I thought I would give a quick response to Brad DeLong's post arguing that better housing policy offers a quick way to fill much of the gap created by the collapse of the bubble. Brad has two contentions. First that years of very low building has led to huge pent-up demand for new housing units and second that if underwater homeowners could refinance their homes then we would see much more consumption.

Taking these in turn, Brad uses a trend line for housing construction to say that netting out the oversupply of the bubble years and the undersupply of more recent years, that we are way below trend. I would question the accuracy of the trend since the aging of the population might suggest a sharply lower rate of construction. (Also the increasing share of income going to medical care logically implies a decreasing share to everything else -- presumably housing would be affected by this.)

Anyhow, we have direct data on the extent of over or underbuilding, the vacancy data compiled by the Census. This shows that vacancy rates are down from the peaks reached in 2009-2010, but still well above pre-bubble levels. That doesn't sound like a market with lots of pent-up demand.

As far as the consumption story from allowing underwater homeowners to refinance and write-down debt, a little arithmetic would quickly destroy the illusions here. There are roughly 10 million underwater homeowners. Suppose they all refinanced tomorrow. How much more would they then consume?

The median income for homeowners is around $70,000. (I'm assuming that the average income for underwater homeowners is close to the median, meaning that it's below the overall average.) Suppose that refinancing and coming above water allowed them to increase their annual consumption by $5,000 each on average, which would be a huge increase for a family with an income of $70k.

This would imply an increase in annual consumption of $50 billion. If we assume a multiplier of 1.5 that will add $75 billion, or a bit less than 0.5 percentage points, to annual GDP. This would be helpful, but not exactly a game-changer.

Long and short, we absolutely should have done more to help underwater homeowners as a matter of fairness, as I advocated from the beginning of the crisis. But it is unrealistic to imagine that this would have hugely altered the course of the recession.

Comments (7)Add Comment
Total Vacancy Rate, Not Census' Vacancy Rate
written by LSTB, July 29, 2013 6:06
Table 8 doesn't separate the number of units rented or sold, but Dean's right that the total vacancy rate is still sky high.

I think a good chunk of it is being held off the market by the banks, which isn't included in the official vacancy rate measure.
written by skeptonomist, July 29, 2013 4:11
This particular Census table raises some questions.

The vacancy rate has been increasing fairly steadily at least since 1980, as shown by LSTB, and we are pretty much on the long-term trend after a bulge 2005-2009. If you accept that this trend of increasing vacancies is correct and is not harmful in itself, then in fact the bubble and crash really had little net effect on vacancies, contrary to what Dean and others have been saying. Why should we think the current rate is high, if it's on trend? But why have vacancy rates been increasing since 1980?

Total housing units per capita show a sharp increase up to about 1988, then are roughly constant. Taken at face value the data do not show any big increase during the bubble. However the table is wonky. The big drop in 2002 can't be real (were millions of houses destroyed?) - it is apparently the result of some kind of data revision. Thus it is hard to tell if the increase after that took the total to a new high or not (I suspect it did). I hope there are some better data on housing units.

written by skeptonomist, July 29, 2013 4:20
Graph of total housing units per capita:

written by urban legend, July 29, 2013 10:52
I have wondered whether Brad DeLong's trend line hasn't been affected by some wishful thinking -- i.e., picking data points that help fit the theory. The vacancy rate and the relatively small percentage of consumers represented by homeowners who could be helped seem to be major qualifications to the benefits. Call it a necessary but insufficient condition.
dejavu all over again....
written by pete, July 30, 2013 1:21
I was wondering when someone would quip something to the effect of (in the spirit of Krugman August 2002):
"I guess the only thing we can do now is have a housing bubble"
too late...
written by pete, July 30, 2013 9:38
Housing prices are approaching 2002 levels when Baker and Schiller said they were in a bubble...De Long is on the old Krugman bandwagon...hopefully Bernake will slow this bubble soon.
written by http://www.bands4you.com/blogs/post/6492, August 05, 2013 10:54

Write comment

(Only one link allowed per comment)

This content has been locked. You can no longer post any comments.


Support this blog, donate
Combined Federal Campaign #79613

About Beat the Press

Dean Baker is co-director of the Center for Economic and Policy Research in Washington, D.C. He is the author of several books, his latest being The End of Loser Liberalism: Making Markets Progressive. Read more about Dean.