Mortgage Credit and the Housing Bubble

My problem with this is that I think the mortgage glut and the house price bubble were well underway before 2004 -- say, starting about 1998.

The mortgage finance supply glut occurred because markets failed to price risk correctly due to the complexity and heterogeneity of the private-label mortgage-backed securities (MBS)

Wrong! They mispriced the risk because the Fed had distorted the price of loans, the interest rate. You can't expect the market to price things correctly when the Fed distorts the price of everything. The core of Hayek's thought on business cycles is that manipulating interest rates changes the prices of all goods and services relative to each other. CPI is unimportant compared to relative prices, especially the prices of consumer goods relative to capital goods, and the price of labor relative to sales. That's the gist of the Ricardo Effect.

The mortgage finance supply glut occurred because markets failed to price risk correctly due to the complexity and heterogeneity of the private-label mortgage-backed securities (MBS) that began to dominate the market in 2004.

I think it's important here to specify which markets.

As Russ Roberts points out his paper, bondholders of mortgage securities were made whole by the government. The mortgage market may have priced the risk correctly, incorporating the probability of a bail out.

Investors in these firms may not have priced the risk of not benefiting from a bail-out correctly.

Their site does not take comments so I am making my note here: there is a lot of merit in their thesis, but I am unclear the extent of the claim they are making. Regarding alternative explanations, they say those do not "fully explain" the crisis, which I read naturally to mean they have some explanatory value. Then you read the sentence above. Are they saying their thesis explains everything or that it is a partial cause? I cannot tell. I hope they will clarify. I tend to think events of this magnitude cannot be plausibly attributed to a single cause.

How can an asset bubble that started in 1997-98 be explained by securitization that "began to dominate the market in 2004"?

Seems like we should be looking for the foundations of this bubble in the early to mid 1990's.

If one eyeballs house price data, there appear to be two inflection points, one around 1997 and another around 2002-03. The first may have led to the second: loan losses were exceptionally low in between '97 and '02, which may have led lenders with short horizons to underestimate the risks in mortgage lending. This led to inadequate underwriting/risk pricing, which in turn pushed up the rate of house price growth even further. So the Levitin-Wachter view may have very well been at least part of the story.

I wish housing bubble obsessive compulsives (of which I am one) would focus on why leverage and foreclosures increased significantly more in California, Phoenix and Nevada than, for example, in the NYC metro area even though prices in both increased similarly from 2000-2007.

Also, I wish they would notice that housing sales volumes increased back to their near all time highs in California BEFORE Sept/2008 as prices were collapsing (and before the Feds created their "solutions").

Countrywide was a leader in California origination. The politics of this were as causal as any stylized set of monetary facts. The disconnect between the actual collateral (i.e., houses) and the RMBS and CDO's which financed them was also not very helpful.

But if "we" (i.e., the "Feds") had focused on de-constructing the latter, rather than fainting before the Cassandras crying to the roof tops about multiple trillion dollar "mark to market" losses, this would have been a crisis more like 1990 than 1930.

I've been a fan of the supply-side explanation for a while. It seems pretty clear that there was a large demand for collateralized securities and that created a lot of funding for mortgages, where a lot of relevant experiences and straightforward assets were.

private-label mortgage-backed securities (MBS) that began to dominate the market in 2004

The rise of private-label MBS exacerbated informational asymmetries between the financial institutions that intermediate mortgage finance and MBS investors. The result [emphasis added] was overinvestment in MBS that boosted the financial intermediaries' profits and enabled borrowers to bid up housing prices.

Based on that theory, I should be able to come up with an investment in nothing that no one understands and be able to sell billions of it to investors because of my informational asymmetry.... wow...

Spreads declined to historic lows in a host of credit instruments unrelated to mortgage finance, so I do not find this paper convincing at all.

Does anybody know what percentage of all defaulted dollars were in California? I saw one estimate that 66% of all housing value declines in the country in 2008 were in California, but whenever I Google for "dollars defaulted" I only get references to my own writings. Nobody else seems very interested.

There were indeed multiple inflection points in the housing bubble. If Bush had pulled the plug on the most absurd kind of zero downpayment mortgages as late as the day after Election Day in 2004 -- "Okay, we juiced the economy enough to get me re-elected, so let's turn off the steroids now" -- the disaster would have been much smaller. The worst loans were originated in 2005 through the first half of 2007. That's when Fannie and Freddie jumped in to exacerbate the already ongoing private stupidity.

Unfortunately, Bush was a true believer.