Scott Sumner takes note of a paper that contradicts the “Inside Job” theory of the housing/finance crisis. Here’s the twelve bullet point summary:
Fact 1: Resets of adjustable rate mortgages did not cause the foreclosure crisis.
Fact 2: No mortgage was “designed to fail.â€
Fact 3: There was little innovation in mortgage markets in the 2000s.
Fact 4: Government policy toward the mortgage market did not change much from 1990 to 2005.
Fact 5: The originate-to-distribute model was not new.
Fact 6: MBSs, CDOs, and other “complex financial products†had been widely used for decades.
Fact 7: Mortgage investors had lots of information.
Fact 8: Investors understood the risks.
Fact 9: Investors were optimistic about home prices.
Fact 10: Mortgage market insiders were the biggest losers.
Fact 11: Mortgage market outsiders were the biggest winners.
Fact 12: Top-rated bonds backed by mortgages did not turn out to be “toxic.†Top-rated bonds in collateralized debt obligations (CDOs) did.
Read the whole thing. I don’t know what caused the housing/finance crisis. I’m actually more interested in why prices rose so high so fast than in why they fell and came pretty close to taking the whole economy with them. The fall was implicit in the rise.
It was also a bit startling that one of the key innovations of Dodd-Frank — that originator’s maintain skin in the game — would have reduced their losses had they been in effect. It still may not be a bad idea, since it may be possible that had the rule been in effect, the originators would have initiated fewer mortgages (or more likely charged a higher interest rate to cover the risk). But we are paying the price of regulatory uncertainty because banks are not clear about their clawback exposure, so they would rather refi existing loans.
Skimming the report, with particular attention to Fact 2, I find myself back to where I started: There should be a minimum down-payment on home purchases. Twenty percent is probably ideal, but would probably hurt the housing market too much. Perhaps ten percent.
I’m actually more interested in why prices rose so high so fast than in why they fell and came pretty close to taking the whole economy with them.
…..and, that’s happening once again, during the present market.
Prices are competitively increasing, generating multiple offers which often are over that ‘asking’ price. IMO, much of the behavior seen before the ’08 crash is simply repeating itself.
One of the major differences, though, is that many of these jacked-up sales are ‘cash’ sales. I was also told the other day that rich Russians are now getting into the American housing market. Basically, the foreign influx of money into American housing is definitely a large part of today’s housing sales.
Oh, boy. When I have a little more time to comment.
I guess this comment probably isn’t very clear:
“[Dodd-Frank] would have reduced their losses had they been in effect.”
Dodd-Frank requires the originators to retain 5 percent of the credit risk. Had that requirement existed in 2005-2006, that retained credit risk would have been minuscule compared to the losses the market imposed on the banks.
PD- True, but only if they knew that ahead of time. If they know they are retaining assets, they have to be more careful.
The paper just addresses the question of whether they were all a bunch of crooks, or whether they all believed their models showing prices could not drop. The paper supports the latter conclusion. In a way I find that much more troubling. With about half of their loans being no-doc loans, how could they not expect them to fail?
Steve
steve, the next paragraph in the report addresses that as well:
“In addition, many analysts have argued that if the managers of ï¬nancial institutions had had their own money at stake, they would have been more careful (Rajan 2010, p. 164–165). But the losses suffered by Jimmy Cayne and Richard Fuld, the CEOs of Bear Stearns and Lehman Brothers, dwarf by an order of magnitude any clawback provision contemplated so far. And further down the organization chart, Lehman staff owned nearly a third of the company, so many managers obviously had signiï¬cant skin in the game as well (Sorkin 2010, p. 294)”
As to the low-doc and no-doc loans, the chart makes it appear that the failure rates were not bad at least through 2004, and if they charged a higher rate of interest for that risk, or required a greater deposit, etc., they might have been more profitable. Also, low-doc and no-doc loans were 33% of Fannie-Mae’s insured mortgages in 1990.
PD
re: Dodd Frank
No chance. Just no chance.
I can’t comment now. Busy. But don’t fool yourself.
I lived in Florida at the time and in my area demand began to seriously outstrip supply especially as speculators got into the market. When homes were priced based on comparables, the price was inevitably “low” as there was often a bidding war to get the property. So agents began to see the comparable price as floor because prices were rising so fast. This naturally tended to reinforce price rises. Of course, as places sold and became comparables for other homes, even that “floor” price rapidly rose. It was a big feedback loop.
Just as an example, we bought our house in 2003 for $150k. Over the next couple of years we put about 70k into it including a garage and other improvements thinking we would be in Florida for a while. Turned out we had to move earlier than we expected and we put our house on the market right at the peak of the bubble. The listing agent’s “low” price based on comparables was $495k. That sounded a bit crazy to me, but I trusted the expert and we listed it that “low” price thinking we’d get a quick sale. Unfortunately for us, the bubble was popping as it went on the market and we were on the back side of the price wave for the next year. There were lots of people looking, but we had no offers that year despite us dropping the price by $200k. It was hard crash. We gave up, financially stretched by the costs of keeping an empty house in another state, and rented it for a year. It went back on the market for $220k and we got an offer for that amount. The buyer’s bank appraiser did what all the appraisers were doing at the time, and put the appraisal for less than the agreed sale price. The buyers didn’t have the difference in cash, so there went another $20k so the final sale price $200k, not including fees and commissions. According to Zillow, the home’s estimated value today is back around $150k.
@Andy, I’ve always thought the appraisals were not capable of identifying a bubble. If prices are rising faster than the “true” underlying value of the properties, then recent comparable sales are just as likely to reflect the bubble price. The appraiser’s opinion is essentially: “Yes, that’s what houses sell for right now.”
That’s probably important because the appraisal might be one of the few independent view points on the transaction. The buyer, the seller, the realtors and the banks all want the sale to go through. At least the appraisal can prevent crazy sales from going through. I don’t know what the appraisal standards are now.
PD,
After my experience I agree with you about appraisers.
After reading the paper, I agree with their thesis, but it contains some stunning crap.
Glass-Steagall established a firewall between traditional and investment banking.
Fact 1 Resets of adjustable-rate mortgages did not cause the foreclosure crisis:
Their graphs for ARM vintage vs defaults is based upon one month, and they only include one reset. There should be at least four quarters for each year, and the resets are every 6 months. I am skeptical of the interest rates.
Fact 2: No mortgage was “designed to failâ€
… Marketing products that do not work is usually a
bad business plan, even in the short run, whether one is producing mortgages or motorcycles. …
– p. 8
Apparently, they do not get out much – Chinese produced toys, dog food, drywall, etc.
Fact 5: The originate-to-distribute model was not new
… which would eventually render more than half of S&Ls insolvent …
– p. 12
The S&L Crisis does not rate an entire sentence. Bury it in a dependent clause, and it disappears down the memory hole.
Facts 6: MBSs, CDOs, and other “complex ï¬nancial products†had been widely used for decades
Smartphones were available years before the iPhone, but the iPhone changed the market dramatically. High tops were available long before Air Jordan, but they changed they market dramatically.
Fact 10: Mortgage market insiders were the biggest losers
Apparently, they have never heard of CDS’s & AIG.
Fact 12: Top-rated bonds backed by mortgages did not turn out to be “toxic.†Top-rated bonds in collateralized debt obligations (CDOs) did.
This is a really good explanation of the ratings process.
3.1 Explanations based on asymmetric information
A hustler, scammer, con man, player will make you think you have more information than him. Somebody needs to explain three-card Monte to them.
No, read it again.
They look at three vintages (January 2005, January 2006 and January 2007) for mortgages over several years. They look to see the mortgage default rates for the mortgages issued in those two months over several years. It is similar to doing a cross-sectional analysis.
Your description is highly misleading. In looking at the graphs for all three vintages I don’t see a “step change” before and after the resets. There are likely thousands if not tens of thousands of mortgages in each vintage. The graphs look at those mortgages over the course of 2005-2009 and also plots the cumulative default rate. After each reset, the cumulative graphs show no appreciable change in their rate of growth.
In other words, the authors conclusions appear to be true.
What is the point of this comment other than to be snarky. The authors claim is absolutely true. Originate-to-Distribute is far older than the housing crisis. Whatever you think about the S&L crisis or the authors think is irrelevant to this fact.
You are also selective quoting which is highly misleading here is a fuller and more accurate quote which changes 100% your implications:
In other words, S&Ls went insolvent because they did not use an Originate-to-Distribute (OTD) model.
And that is more than one sentence. In fact, my quote is 4 sentences, and in the paper S&Ls are given an entire paragraph.
@Steve Verdon
In other words, the authors conclusions appear to be true.
I do not have time to fully examine their numbers. They probably used January to make it easy to compare the vintages, but the graphs they use can be misleading for somebody who does not fully understand 2/28 mortgages. They do explain how they work, and therefore, I am not accusing them of being disingenuous.
There should be four months – January, April, July, October – for each year. If I remember correctly, the way the MBS’s work means the rates should be stable for one quarter. Each of these quarter can have different starting interest rates, and they can have different reset rates. January could have been the best month, the worst month, or the average month.
Once the resets start, the payment can change every six months depending upon the interest rate. It is easier to mark the initial reset, but it would be easy for somebody to think the payment at that point was constant.
One factor they left out was the recession starting in 2008, and probably earlier for they lower end workers. This factor would mean that some of the foreclosures were due to the recession not resets. The recession caused defaults for all types of borrowers, but the worst borrowers would be hit harder. The sub-prime FRM mortgage defaults could have been increased due to the recession, and if they were a better quality than the ARM mortgages, it would account for the default difference being 5%.
If I am reading the graph correctly, the cumulative defaults is a percentage of the total defaults. I think the actual number would be better. I am assuming the cumulative defaults is not a running total.
What is the point of this comment other than to be snarky. The authors claim is absolutely true. Originate-to-Distribute is far older than the housing crisis. Whatever you think about the S&L crisis or the authors think is irrelevant to this fact.
I deleted the snarky parts, but they were not directed at the authors. The “inside/outside” conspiracy theorists are also at the bigfoot and 9/11 conspiracy meetings. I never heard of the movie, and I would not spend a lot of time refuting it. Their paper addresses a very specific premise: The housing crisis can be explained through an Inside/Outside theory. Again, I mostly agree with them, but I do have a few objections to the way they present their argument.
You are also selective quoting which is highly misleading here is a fuller and more accurate quote which changes 100% your implications:
I indicated it had been ripped from the context by using an ellipsis. at both ends. I also provided the page number to allow anybody to quickly evaluate my usage. I considered using the entire paragraph, but I would actually need to quote most of Fact 5 & 6. These facts establish that various methods and products for mortgages existed prior to the 2008 housing crisis. I do not disagree, but it is irrelevant.
The OTD model was where they referenced a S&L problem, but I was not addressing the OTD model. I am not an expert, but it is my understanding that the S&L Crisis was primarily about fraud. They were “cooking the books” to cover their insolvency.
I should have gone into greater detail about my point, but it was late. The point is that these instruments existed and had a role in the S&L Crisis. The role was not the same as the 2008 Housing Crisis, and it was more of a supporting role.