Thursday, July 16, 2009

John Y. Campbell, Stefano Giglio, and Parag Pathak estimate that Foreclosed houses sell at a 28 percent discount

The results imply two problems with thinking about house prices through the lens of the Case-Shiller Index.

In Southern California, somewhere in the neighborhood of 40 percent of sales are distressed sales. If lenders make decisions that are based on Case-Shiller, they will underestimate the value of transactions that are taking place in the absence of distress. Appraisers seem to be taking a Case-Shiller view of the world right now, and so deals are getting undone. There is reason to believe that when buyers are willing to place 20 percent down on a house, they actually believe the house is worth the offer price.

On the other hand, let's say we move to a world where only, say, 20 percent of sales are distressed. This will produced an observed increase in the index Case-Shiller Index of 6 to 7 percent--even if nothing is really changing about underlying house prices. This could lead markets to become too optimistic too quickly.


Michael Roberts said...

Interesting. I wonder if maybe the foreclosure discount is larger today than it is typically. And if so, does that not blur the line between the true underlying home price and the price of a typical foreclosure?

sasi said...

it sounds quite interesting..........
Boise real estate

Austin Kelly said...

I'd encourage you to have a look at Andy Leventis' paper at

Andy find that excluding distress sales knocks off 2 percentage per annum from the Cal price decline -20% instead of -22%. Enough to be worth noting, I guess, but not a very large impact.

Might also want to see the Harding, Rosenblatt, and Yao paper on foreclosure contagion.

They find that much, but not all, of what seems to be a contagion effect is realy just simultaneity - micro-areas that have declined more in price have more foreclosures, at least as much as micro areas that have more foreclosures decline more in price. But there still is a real contagion effect for close by properties.

Austin Kelly said...

I should add that Case-Schiller does not filter out REOs but does filter out "outliers" - so an REO with a 15% discount probably makes it into the data, while an REO with a 50% discount probably gets filtered out. That's one reason why the observed differences between an REO inclusive and an REO exclusive index are smaller than you might at first think.

Anonymous said...

Home prices are still way too high. The sooner they come down, the sooner the economy will recover.

Erica Smith said...

I want to make 1 guest post in your blog, if you permit me. The post contains 350 words above and totally unique as it is written by my content writers.

Please contact me at ericasmith568(at) soon. Moreover, I will place your link in one of my finance sites.