Sunday, 29 November 2009 05:00

The Truth! The Truth? Banker's Can't Handle the Truth!!!

I have finally updated my Alt-A and Subprime delinquency, charge-off and loss data using FDIC, NY Fed, Corelogic, First American, and Bloomberg (among others) as sources. If you thought things looked bad last year or this spring, they are getting worse - with no reprieve for the 3rd quarter despite the extreme amounts of liquidity and capital thrown at the situation by central bankers and the US government. As soon as I started writing this piece, CNBC comes out with "US to Push Mortgage Lenders To Modify More Home Loans: The
US Treasury announced plans to push lenders to modify more loans after
the administration's $75 billion housing rescue plan, called Making
Home Affordable, fell short and foreclosures continued rising.

Hmmm... $75 billion is a lot of money. Mayhap the problem is that the banks know how useless pushing on a string is, or mayhap $75 billion is not enough to stem $304 billion (and counting) in Alt A and subprime losses that are still in the pipeline (see graphic below).

It gets worse though. Let's glance at the non-conforming loan losses that have already occurred in comparison to the SCAP projections that justified the return of TARP in many cases. Recovery rates had the illusion of increasing ever so slightly due to an increase in prices as illustrated by the Case Shiller index. I have expressed my doubts about this housing price recovery for several reasons, the least of which is the construction flaws in the index itself which fail to capture the nature of the transient price increases, namely the activity of short term investors and flippers (see On the Latest Housing Numbers). There are some areas that have witnessed some firming of pricing though, but that firmness is the result of the Fed and Treasury trying to blow another bubble within a bursting bubble and is more than outdone by the rampant deterioration in credit quality of loans that result in the dumping of foreclosures -> REOs -> short turnaround sales/flips (via investors, which are not captured by Case Shiller, hence the illusion of a firming market in the lower end of housing prices) all over the place.


Subprime delinquency, charge-off and foreclosure rates are still flying through the roof - with many other categories rushing to keep up. This is as I described from the beginning (2007) through the Asset Securitization Crisis series - there was an underwriting induced crisis and never a true "subprime crisis". As such, there is a very strong chance that many other loan categories may outstrip subprime loans in terms of aggregate losses. It hasn't happened yet, but the Alt-A category is hot on subprime's heels (see below). Construction and CRE will follow up the rear with unsecured consumer (ex. credit cards) and commercial loans fighting to get into the race.


Below, you see the loss trend as of October 2009. These are losses that have most likely NOT been claimed by the banks, and they are significant. In addition, the credit deterioration trend is climbing, not falling. If I am correct in my assumption on the validity of the Case Shiller index in capturing true inventory price depreciation across investor related sales and bank "hold outs", then prices will soon start dropping again, killing recovery rates and causing losses to spike even further.



The mainstream financial media has led many to believe that the "subprime crisis" has passed. FRB and FDIC data actually show that subrpime credit deterioration is increasing in the face of lowered interest rates through QE/dollar debasing and HAMP government efforts. This is also despite certain bank policies that mask delinquencies, such as lagging the time it takes to mark a loan delinquent. We found Wells Fargo doing this last year with its HELOC portfolio.



As you can see, every reprieve seen since the crisis started has been followed by a spike in delinquencies. I expect the same to occur for 2010.


As can be expected, ARMs sport more than twice the delinquency rate as fixed rate loans. The drop in rates has caused a leveling of ARM delinquencies, but it is clear that rates can't remain at zero forever, and the bulk of these loans are close to if not passed the underwater mark. Literally any move by interest rates in the direction of equilibrium (read as the cessation or failure of the Fed's direct intervention in the interest rate markets) will cause a flood of delinquencies and foreclosures that are bound to overwhelm the banks. This is an inevitable occurrence. It is not a matter of if, but a matter of when. The interesting issue is that all of the categories are at currently a level that scream solvency alert!


The Case Shiller index has shown price increases for the last two quarters. Despite my reservations about its accuracy, the increases have done nothing to stem the onslaught of credit deterioration. As a matter of fact, actual losses are increasing despite the increase (and its requisite rise in recovery values) due to the amount of charge-offs and the rate in whch they are occurring. If you lool closely at the chart, the credit quality deterioration actually picked up as Case Shiller was showing price improvement.

Below are the loss rates by state. Be aware that these are "LOSS RATES", not delinquencies or charge offs. This is what the actual losses will end up being (sans admin and legal costs, which will driver the losses higher). These loss rates are calculated as follows: Overall default rate - Recovery rate (Case Shiller - LTV) = Loss Rate

California, Florida and Nevada have been held up as loss "icons" by the media, but there are pockets of significant loss throughout the country. States such a Wisconson and Michigan are pushing 50% in subprime losses, yet still pale to the near 60% in losses from states such as Florida. I query, "How can a bank with 10% equity, and 90% leverage wither 30% to 60% losses on its loans and still be considered solvent?" For those not versed in these matters, imagine you sold 100 million dollars in stock to the public to start a bank, and borrowed 900 million dollars to write mortgages on $3 billion worth of houses. You consequently take a 50% loss on those mortgages - to the tune of $500 million. How much is that initial $100 million of stack worth, even assuming you can make $30 million dollars a year trading risky assets with inflated spreads? You are now totally insolvent and still owe your investors $300 to $400 million in order for them to break even. Now, imagine that your stock just shot up 300% in price because.... Well, just because. Tongue out.

There are many banks that literally specialize in underwriting these (among others) high risk loans in the hardest hit of the areas below. There is no way in hell they are carrying those loans anywhere near their actual value.

A good example of this is PNC Bank. They were running a 41% loss rate in Wisconsin (according to our birds-eye view calculations using FRBNY and FDIC data) in May, and now are exposed to a 45.8% loss rate. Florida is even worse! Many states have seen a 5% INCREASE in LOSS rates. Again, it is no wonder why they aren't returning their TARP monies.




Realize that the losses graphed above are mostly losses that have not been marked yet by the banks. They are quite significant. For a birds-eye view of the situation, simply apply the state loss metric to your favorite bank's portfolio, on a state by state basis.

This is PNC's granular break down using the government data from May of this year - before the roughly 5% jump in economic losses that we have calculated for the most recent quarter.


you can see, going through each major loan category in PNC's books
reveals a much LESS optimistic scenario than ANY portrayed in their
SCAP take home test results... If one were to factor in the more recent information, the SCAP test and the return of TARP look more like fodder for litigation than any true attempt at solving the banking problem.

Sources: FirstAmerican CoreLogic, LoanPerformance Data, U.S. Census Bureau, and Federal Reserve Bank of New York

(a) Statistics calculated on first-lien and active (includes REO) loans.
(b) Statistics calculated on first-lien, owner-occupied, active (includes REO) loans.
(c) 'Prepayment penalty in force' denotes that the loan age is less than the prepayment penalty term.
(d) Statistics calculated on first-lien, owner-occupied, active (includes REO), variable rate loans.

Let's also revisit some data from "You've Been Bamboozled, Hoodwinked and Lied To! Here's the Proof. What Are You Going to Do About It?"
with which we can compare the assumptions that justified the bank's
return of TARP monies with the actual unfolding of events.

In terms of unemployment, we are already breaching the worst case scenario projection of two years into the future.


As you can see, the major driver of
future bank credit losses has been woefully underestimated, and thus
the capital requirements of said banks have been woefully
underestimated, among other things.

Below is subscriber content that reveals what the
banks REALLY needed in terms of capital and cushions to whether the
true rate of losses and unemployment to come. You may subscribe here to access this content.

Goldman Sachs Stress Test Professional Goldman Sachs Stress Test Professional 2009-04-20 10:06:45 4.04 Mb

Goldman Sachs Stress Test Retail Goldman Sachs Stress Test Retail 2009-04-20 10:08:06 720.25 Kb

MS Simulated Government Stress Test MS Simulated Government Stress Test 2009-05-05 11:36:25 2.49 Mb

MS Stess Test Model Assumptions and Stress Test Valuation MS Stess Test Model Assumptions and Stress Test Valuation 2009-04-22 07:55:17 339.99 Kb

PNC SCAP Results recast using FDIC and NY Fed data - Pro PNC SCAP Results recast using FDIC and NY Fed data - Pro 2009-05-15 07:31:21 455.37 Kb

PNC SCAP Results recast using FDIC and NY Fed data - Retail PNC SCAP Results recast using FDIC and NY Fed data - Retail 2009-05-15 07:30:25 395.18 Kb

PNC Stress Test Pro PNC Stress Test Pro 2009-04-13 02:10:17 3.11 Mb

PNC Stress Test update - Professional PNC Stress Test update - Professional 2009-04-21 15:55:56 3.00 Mb

PNC Stress Test Retail PNC Stress Test Retail 2009-04-13 02:11:08 323.51 Kb

PNC Stress Test update - Retail PNC Stress Test update - Retail 2009-04-21 15:53:52 777.50 Kb

PNC stress test write up - public lite PNC stress test write up - public lite 2009-07-27 02:37:11 995.30 Kb

Sun Trust Banks Simulated Government Stress Test Sun Trust Banks Simulated Government Stress Test 2009-05-05 11:37:13 1016.17 Kb

JPM Public Excerpt of Forensic Analysis Subscription JPM Public Excerpt of Forensic Analysis Subscription 2009-09-22 14:33:53 1.51 Mb

Last modified on Sunday, 29 November 2009 05:00