Hat tip to daytrip for sending in the MSM’s version of Analysts Gone Wild.
It is a two-prong battle, How Many, and How Long.
The current leader of the How Many race is Laurie Goodman, who in this article predicted that 10.4 million mortgages will eventually be foreclosed unless something is done. Rick Sharga, VP of RealtyTrac (a company that sells foreclosure subscriptions!) was the previous leader and must be ticked – he should have a new forecast before long. I still have my money on him.
A recent survey of ivory-tower goofs has taken the lead on How Long:
From cnbc.com:
Home prices are unlikely to recover before 2020 and mortgage defaults will persist for years, says a survey of bank risk managers out Friday.
The survey conducted by the Professional Risk Managers’ International Association for FICO, found that 49 percent of respondents do not expect housing prices to rise back to 2007 levels for another nine years. Only 21 percent of respondents said they would.
The findings, which authors called “a decidedly pessimistic outlook”, are a sharp reversal from cautious optimism the survey respondents expressed late last year and in early 2011.
In addition, 73 percent of surveyed bankers say they expect mortgage defaults to remain elevated for at least another five years. And 46 percent believe mortgage delinquencies will increase over the next six months.
Only 15 percent of respondents expect mortgage delinquencies to decline during that period.
“While the housing sector will almost certainly gain strength during the next nine years, many bankers clearly believe prices will remain depressed for half a generation,” said Andrew Jennings, chief analytics officer at FICO.
Bankers concerns spread beyond the housing market.
A large number of respondents says they also expect to see an uptick in delinquencies on auto loans, credit cards and student loans.
Small businesses are expected to continue face a challenging credit environment. More than one-third of respondents forecast an increase in delinquencies on small business loans.
Bankers also appear to be pessimistic about recovery in consumer spending, with 64 percent of respondents expecting credit card usage to remain below pre-recession levels for at least five more years.
Half of the respondents expect credit card balances to increase over the next six months, due to higher spending by some households and smaller monthly payments by others.
Sooo – the bankers say the housing will be bad until 2020.
Aren’t these the same bankers that made all the NINJa loans?
So now their advice is trustworthy?
I trust my own advice and I see the big picture.
Patrick Pulatie at LFI Analytics has noticed a huge discrepancy in the number of reported foreclosures by varying organizations. Via Email Pulatie writes …
Much of my off time is spent in reviewing the reports of other entities regarding the foreclosure crisis, correlating data and trying to forecast what to expect in the future.
Frankly, I am at the point where I am wondering whether anyone has a true reading on what is happening.
Office of the Comptroller of the Currency
At the end of the 2nd Quarter, the OCC reports:
It Monitored 32.7 million first mortgages out of 53 million total, representing 62% of all first mortgages.
4.9% or 1.57 million were more than 60 days late, or at least 1 month late while in Bankruptcy.
5.5 million are late by over 30 days, or in the foreclosure process.
1.3 million in some stage of foreclosure.
88% of mortgages were current. (This would mean that 3.944 million were not current of the 32.7 million being monitored.)
800,000 REO’s
If 1.57 million loans are more than 60 days late and 5.5 million are late by over 30 days, then 3.8 million must be between 30 days and 60 days late.
Even worse, this only represents 62% of all first mortgages.
Amherst
Amherst’s Laurie Goodman comes up with different numbers.
Of 55 million mortgages, 4.5 million are non-performing or delinquent. (Non-performing would be those that are 30 days late or more.)
10.6 million “at risk” of foreclosure.
400,000 REO’s (compare to OCC 800,000)
Corelogic
Corelogic does not match any numbers except REOs of OCC but not Amherst
Current Residential Shadow Inventory is 1.6 million homes.
Of the 1.6 million shadow inventory, 770,000 loans are “seriously delinquent”.
430,000 are in some stage of foreclosure. (Compare to the OCC saying 1.3 million are in foreclosure.
800,000 REOs
Rick Sharga – formerly of RealtyTrac/Moodys
Sharga says:
800,000 are in foreclosure.
1.5 million are delinquent. (Does this include the foreclosure loans?)
LPS
This is what LPS says:
8.13% total loan delinquency rate.
4.11% of loans in the foreclosure process.
12.44% total delinquent or in foreclosure.
2.38 million loans less than 90 days delinquent
1.87 million loans 90+ days delinquent.
2.15 million loans in foreclosure process.
Total of 6.40 million loans delinquent or in foreclosure in August.
Why does the OCC, which looks at 62% of the loans tend to have numbers that if extrapolated over the total amount of loans, would suggest much higher numbers than all other entities?
I have to ask: “Does anyone really know what is going on with foreclosures?”
Data Sources for Figures Cited Above
Corelogic Report
LPS August Report
Rick Sharga, Realty Trac
Jim Saccacio, RealtyTrac CEO
OCC
Amherst – Laurie Goodman
Addendum:
As a comment to this post, Pulatie added …
OCC is likely getting their information from the GSE’s, since that would be easy to track. I “assume” this because they only follow 62% of the loans, which is about the size of the market that the GSE’s held during the peak of the boom.
Either LPS or Corelogic gets their foreclosure numbers from 2200 counties in the US. However, there are 3143 counties in the US. Likely, they use electronic reports, but if almost 1000 counties are missed, then what good is the reporting?
RealtyTrac uses county data. But, RealtyTrac is actually Moody’s. And with Mark Zandi, who knows what the hell that they doing and coming up with.
Amherst Holdings appears to be fairly accurate, based upon what information that they actually reveal. The do consider some outside influences that the other companies do not consider. Laurie Goodman appears quite knowledgeable when she reports to Congress in her testimony..
Amherst actually provides a “best case” scenario, and a “reasonable” scenario, at least in regards to their projections. However, “reasonable” appears to be at least 20% below what “worst case” would suggest at the moment. At this time, I think that “worst case” is more important than all others.
A potential source for delinquencies may be the Credit Reporting Agencies. Each lender would detail in their monthly tapes who was delinquent, and by how many months. Then, it would be a matter of actually determining what one considers seriously delinquent. (This is especially important to know, but in my view, since once a loan goes 30 days late, it is a 90% likelihood of default and foreclosure occurring.)
How they are determining “underwater” loans, I haven’t got a clue. Perhaps they are trying to incorporate Automated Appraisal Date sources. Or they take the original loan to value as reported in monthly reports on lending, and then apply general decreases in value in geographic areas to arrive at “underwater” numbers.
I believe that there is a tendency for the different companies to want to under report what is happening. This would be in part to try and stabilize housing to some degree. If the public thought things were much worse, then it would only worsen the public perception