I decided to make this post in response to the discussions in the comment thread of "Investment Advice in the MainStream Media: Hedge against Success???" For those don't know, the JPM discussion started due to this insolvency post: Re: JP Morgan, when I say insolvent, I really mean insolvent (a must read for anyone with economic interest in JP Morgan). WaMu, which was recently purchased by JPM, was the leader in Option ARM sales, and (from personal experience) has one of the most convoluted, inefficient and error prone 2nd lien underwriting processes out there. I explained the Option ARM dilemma in "The banking backdrop for 2009". To make a long story short:

Option ARMs to Reset Earlier than Expected

In 2009 and 2010, loans with 2004 and 2005 vintages would be recast. Besides these vintages, loans with negative amortization are expected to recast early. With more than 65% of borrowers electing to make Minimum Monthly Payment (reaching a staggering 85% for 2006 and 2007 vintages), loans which recast on account of negative amortization caps are expected to increase drastically.

The problem ahead: According to Fitch, of the nearly $200 bn of option ARMs outstanding, roughly $29 bn of loans are expected to recast by 2009. Of this $6.6 bn constitute 2004 vintage (that would be recast as a result of completion of the end of five-year term in 2009) and $23 bn constitute 2005 and 2006 vintage loans that would recast early due to the 110% balance cap limit.

Further an additional $67 bn is expected to recast in 2010 of which $37 bn belong to 2005 vintage (that would be recast as a result of completion of the end of five-year term in 2010) and the balance $30 bn consist of 2006 and 2007 vintage loans that would be recast early due to the 110% balance limit cap.

The potential average payment increase on the loans recast is 63%, representing an additional $1,053 due each month on top of the current average payment of $1,672. These large payment increases could cause delinquencies to increase, and increase dramatically, after the recast. The fact that only 65% of borrowers have elected (or are able) to make only minimum payments underscores the magnitude of the potential problem. The potential payment shock combined.

Other Blogs are reporting Reserve requiring downgrades: "Wednesday's downgrade of 2,446 classes of mixed RMBS caught traders off guard - even though it was viewed as an eventuality. While the market had largely priced in below-investment grade ratings to alt-A bonds following Moody's Investment Services' announcement that it would increase loss assumptions, the swiftness with which the rating agency acted has traders bracing for even more supply.

While the market was already trading bonds to these higher loss assumptions, the banks and insurance companies that own this paper are now going to have to hold more capital against these assets, and the increase given that these bonds are now junk [rated] is not a small matter," said one trader. . .

Moody's warned in a report last week that loss assumptions would be increased for RMBS and that downgrades could be expected. Moody's is projecting that alt-A deals originated in the second half of 2007 will experience 25.5% losses of original balance, compared to 23.9% of 1H07 deals, 22.1% for H206 deals and 17.1% for 1H06 deals. The rating agency in May expected average losses for 2006 and 2007 vintage deals to reach 11.2% and 14.7%, respectively.

Massive selling is not expected immediately though it is only a matter of time before a substantial portion of the downgraded bonds are put out to bid, a second trader said ..."

with the continuous deteriorating outlook for home prices and lack of refinancing opportunities could be a negative cause of concern for investors in Option ARM securities. Even more ominous, is pall cast upon the banks that hold these assets and are additionally exposed to other forms of consumer credit, ie. HELOCs, credit card debt and other unsecured loans (remember the links from the Asset Securitization Crisis above).

Alt-A

What bank has that"Other forms of consumer credit" exposure stated above? JP Morgan who doubled up on the exposure when it bought Washington Mutual, the Option ARM king.

We will get back to Option ARMs and what they will do to JPM in a later post. For now, let's realize that the JPM/WaMu combo has the highest concentration of 2nd lien loans (and Option ARMs) in the absolute worse housing areas in this country, CA, FL, and NV. Keep in mind that 2nd lien loans have diminished priority in terms of claims on assets, so if a house sells for 70% of its previous sales value, and there was a 70 LTV loan combined with a 80 CLTV (combined LTV) HELOC on top, the HELOC lender takes a 100% loss. Let me repeat that, "a 100% loss". How likely is this to happen JPM? Well, let's investigage a little further, shall we? JPM currently has these impaired loans from WaMu marked down 25%, and is currently not classifying them as non-performing even thought the mortgagees are not paying. Let's take a visual tour across the land to see if this 25% mark is realistic.

Well, now we know that there is a high concentration of 2nd lien loans on the books. A very high concentration, leveraged up nearly 200% (we're not even going to broach the topic of C&D and CRE loan risk).

Geographic and/or Product Concentration is a Bad Thing. WaMu has both

Do you see the two states that have been in the news the most lately have big spikes in my pretty little graph.

Now, if we drill down into those two big stalks we see above and observe who has the most concentrated exposure there, we see the following...

The JPM/WaMu combo has TWICE the California 2nd lien exposure as Countrywide, and we all know what happened to Countrywide! And in Florida, the condo capital of world, we have...

The JPM/WaMu combo is 2nd only to Suntrust. This is how I feel about Sun Trust: Sun Trust Forensic Analysis. Don't take this lightly, my triple digit returns didn't just pop out of thin air. I do my homework and can boast a pretty strong track record. Remember, concentration is a very risky thing in investing. You can hit a home-run short term, but long term the odds will catch up to you and hurt a lot. These banks rolled the die, and crapped out. The shareholders just don't know it yet. Below, I put a FICO score chart in just for the fun of it. FICO scores are a lot less relevant than many believe. They are still useful, but far from the be all and end all that they were marketed as.

Guess who has the 3rd lowest aggregate FICO out of the crew. Now, the bank at the lower and of the 2nd lien FICO graph has bought what is probably the most troubled large mortgage lender in the world in what is probably the worst macro environment in the last 7 decades. We can now see how that is ending. The 3rd lowest scorer is dragging down the bank who bought the most poisoned of the Investment banks, albeit with ample government assistance. Hmmmm... Was that smart???

As someone with direct experience in the residential lending arena can tell you, there is a big difference in the quality of loans written directly versus those sourced from a third party, ex. a broker. The reason, obviously, is that the brokers don't (didn't) have any skin in the game in terms of risk retention, and have significant incentive to "fudge" the numbers in order to push loans through (that is how they are paid). Then we have glorified brokers who post margin, got a warehouse credit line and call themselves "mortgage banks". These guys are just brokers with a fancy credit card that they parked loans on till they could sell them off to investors. The problem is, this forced minimum risk retention, but enough to drive 100's out of business when the market collapsed in 2007. Long story short, it is much more dangerous to rely on what you hope to be prudent underwriting from a brokered loan than from a direct channel loan. Amazingly enough, we had the exact same problem with brokers in the S&L crisis. I guess 1,200+ lending institution failures wasn't enough to teach a lesson that lasted more than 15 years. For more on this, see A comparison with the same during the S&L crisis.

Well, you see who has been relying on the brokers. "Nuff said. Don't be fooled by large asset bases and big brand names either. Bear Stearns was a big brand name, and Citigroup had the largest asset base out there.

So, all of this stuff ultimately leads to loans not getting paid. When loans don't get paid, foreclosure occurs. We can fairly easily map out what will happen to the JPM/WaMu/Bear Stearns combo when the foreclosures start hitting (which is right about now), using the widely followed Case-Shiller index. Let me warn you right now, though, these numbers will be highly optimistic due to the fashion in which the index is put together. It is an econometric marvel, yes, but it excludes so much economic reality as to be misleading in a fast trending market. See "A reminder concerning popular housing indices" for more on this topic.

Now, keep in mind that I am using optimistic numbers in these calculations and graphs, and then take a gander at the sharp drop in housing prices that have already occurred in the problem areas identified as JPM/WaMu/Bear Stearns hot spots. That is one hell of a roller coaster ride.

Now, in Re: JP Morgan, when I say insolvent, I really mean insolvent (a must read for anyone with economic interest in JP Morgan) I made clear (through the subscription content portion) that JPM has taken 25% marks on its WaMu credit impaired portfolio. Is that enough? Let's walk through the high concentration areas using this overly optimistic numbers that are 3 months old (the more recent numbers look worse since the housing price slide, foreclosure rate and unemployment is increasing in velocity and volume, not to mention the coming Alt-A debacle described earlier).

Assume the typical property in these hot spot areas had a 70% 1st lien mortgage (that means that the buyers put 30% down in cash - and we all know better than that!) and then somewhere along the line in 2007 to 2007 the homeowner took out a mere 10% home equity loan, which brought the CLTV (combined loan to value) ratio up to 805. These are very conservative numbers, for I know for a fact and through personal experience that these banks gave 90, 95, 100, and 110 LTV loans out like it was candy. Some loan packages were underwritten outright at those very high LTVs, while others had very optimistic, no SUPER OPTIMISTIC (if you know what I mean) appraisals that effectively put the true LTV way over 100%. It is an academic argument though, for as you can see, JPM/WaMu takes a total loss using the most conservative numbers imaginable.

As you can see, the marks on the HELOC's and 2nd lien loans should be 100%, not 25%. Even using the 10 and 20 city composite index, once you factor in the cost and expenses of foreclosure, holding and remarketing the property, you will have close to, if not greater than a 100% loss. If one were to factor those 100% marks into the entire portfolio, would you come up with an average of 25%? Subscribers should reexamine the sensitivity analysis that we did here JP Morgan Forensic Highlights 2009-01-06 19:18:08 133.34 Kb, here JP Morgan Q408 Quarterly opinion with sample trades - Professional & Institutional 2009-01-22 08:48:02 211.69 Kb (retail users, click here JP Morgan Q408 quarterly valuation opinion - Retail 2009-01-22 08:49:26 79.24 Kb) remaining cognizant that the base case valuation numbers and the call on JPM's insolvency was based on a mere 40% mark down. That is not taking into account the massive negative amortization and/or the expiration of teaser rates coming off of the Option ARMs described earlier. Let's take another look at that graph...

Starting right about now (February), the negative amortization recast should start to bring in a rash of Option ARM foreclosures as the housing values shoot downward, and the amounts owed on the houses actually shoot upwards. A nasty combination, indeed, and one in which JPM will suffer.

Now, let's take a look at the first liens that were written on what was the customary 10% down payment.

As you can see, JPM's 25% mark is rather aggressively optimistic for 1st lien loans, and that is using numbers from 3 months ago before the unemployment and foreclosure wave of the quarter hit. When combined with the extremely large 2nd lien loans, the mark should be more like 125% instead of 25% (I'm being a smart ass here, since it must be capped at 100%, but there is more to this 125% number, as we will see in a minute).

Now, these numbers are entrenched in the past, we are living in the present, and any additional marks are contingent upon future price movement. What does the future hold? Considerably higher marks, that's what.

Even using the highly optimistic Shiller index, we are in for a very pessimism generating fall, as can be seen from the graph in "When someone tells you they are seeking stabilization in the housing markets, show them this graph!"

I figure, we are about 50% through the housing price downfall cycle. That means there is a lot, and I mean a lot, more devaluation once it comes to residential housing prices. This doesn't even begin to scrape the surface of JPM's problems, either. Commercial real estate loans, consumer finance loans, credit cards, commercial leasing and business loans, leveraged loans, private equity loan - it doesn't look rosy going into a depression or severe recession. How many loans do you ask? Well, I pointed this out over a year ago...

1. More on the banking backdrop, we've never had so many loans!
2. The consumer finance sector risk is woefully unrecognized, and the US Federal reserve to the rescue
3. Reggie Middleton says don't believe Paulson: S&L crisis 2.0, bank failure redux

Before I go, I just want to let all know that as pessimistic as all this sounds, it is actually optimistic. If I were to be a realist, I would have mentioned that many of the mortgage assets that JPM bought/owns are highly leveraged derivative assets. These numbers in the article are referencing WHOLE loans, not loan derivatives. That means that the numbers mentioned here, both positive and negative, have to have a multiple applied to them. I believe JPM to be insolvent simply by examining the portfolio as if it contained all whole loans, which we all know it does not. If factor in the reality of the leverage, illiquidity, and murkiness in pricing of their leveraged mortgage derivatives.... Well, I think you guys are smart enough to figure out the rest.