This page has been archived and commenting is disabled.
If a Bubble Bubble Bursts Off Balance Sheet, Will Anyone Be There to Hear It?: Pt 2 - JP Morgan
Since I write for a diverse audience, I will start this off with an
overview of securitization. If you are in the industry or are just a
smart ass dude, feel free to skip down to the JP Morgan specific
section of what may lay off balance sheet below. I also have a 30 part series on this Asset Securitization Crisis for those who are interested in my take on this from the beginning. It is a lot of reading, but it tells it like it is - at least from my perspective.
Overview
Securitization is still a very significant source of leverage and
opacity in the US and European economies, in spite of its predominant
role in the most recent global financial turbulence. It is a practice
where loans and other debt instruments are aggregated in a pool and
thereby used to issue new securities. Banks and financial institutions
started establishing Special Purpose Vehicles (SPV) and Qualifying
Special Purpose Entity (QSPE) under the FASB rules to securitized loans
and thereby reducing, from an accounting perspective (but more
accurately put), or transferring from an economic perspective,
financial risks on their balance sheets. Although these new founded
QSPE's were rated by rating agencies (Moody's, Fitch, S&P among the
few) prior to the issuance of securities, the underlying ratings failed
to capture the actual economic value of the underlying collateral.
Furthermore, the ratings established by the rating agencies are an
assurance of performance.
According to analysis conducted by Ann Rutledge of RR Consulting,
subordinate tranches have performed more predictably than senior
tranches since subordinate tranches had already built in high default
rates and prepayment assumptions. The mezzanine tranches during
downturn tends to be relatively stable. Quoting Ms. Rutledge (via her Institutional Risk Analytics interview),
"Paradoxically, because the underlying collateral experienced such high
rates of default and prepayments early on, the risky tranches now tend
to be very stable because they were only really ever worth 20 cents on
the dollar when the deals went out!". Further, since most of the deals
are not done on DCF analysis on the underlying collateral, they were
heavily mispriced at inception. According to Ann Rutledge if a risk is
analyzed correctly, most ABS deals will converge to "AAA" rating
regardless of the initial rating of the security. For instance,
securitization from Countrywide did not converged to AAA suggesting
that the deal was heavily mispriced at inception. For those interested
in hearing more about what Ms. Rutlege had to say, see "If a Bubble Bubble Bursts Off Balance Sheet, Will Anyone Be There to Hear It?" and scroll down to the second half.
JP Morgan securitization activities and QSPE exposure
JPMorgan securitizes and sells a variety of loans, including
residential mortgage, credit card, automobile, student, and commercial
loans through Special-Purpose Entities (SPEs) as a part of the
securitization process. These SPEs are structured to meet the
definition of a Qualifying Special-Purpose Entity (QSPE) and
accordingly, the assets and liabilities of securitization-related QSPEs
are not reflected on the Company's consolidated balance sheets.
As of June 30, 2009 total assets held by these QSPEs stood at $574.4 bn
while assets held by QSPEs with continuing involvement stood at $424.0
bn. Of the total assets held by QSPEs, prime mortgages, commercial
mortgages and credit card receivables formed majority of investment
portfolio with total assets of $211 bn, $160 bn and $102 bn,
respectively.
As of June 30, 2009 total interest held by JPM in these securitized
assets stood at $37.4 bn, or 39.5% of tangible equity with credit card
leading the pack with an 89% contribution. The table below details
QSPE's exposure for JPM as of June 30, 2009.
JPM provides servicing for mortgages and certain commercial lending
products on both a recourse and nonrecourse basis. In nonrecourse
servicing, the principal credit risk to the company is the cost of
temporary servicing advances of funds while in case of recourse
servicing, the servicer agrees to share the credit risk.
Although JPMs loan sale transactions have primarily been executed on a
nonrecourse basis, the company still has a substantial credit risk from
recourse transactions as well. As of December 31, 2008 the unpaid principal balance of loans sold with recourse was at $15.0 bn, or 15.8% of tangible equity
compared with only $557 mn as of December 31, 2007. You have read that
correctly, a 270% increase in full recourse risk in a year and a half.
The increase in loans sold with recourse basis was driven by none other than the Washington Mutual acquisition ((i,e) substantial portion of loans acquired from WaMu were principally on recourse loans). If you recall from the previous forensic reports:
JPM Public Excerpt of Forensic Analysis (free)
You will realize that JPM's loan portfolio is performing horribly, and
although they purchased the WaMu portfolio at a very deep discount
after restructuring, they are still under water and the deflationary
phase of the underlying is only 50% or so along its cycle. This is
surely a widely unrecognized problem for JPM!
Credit card securitizations
WMM Trust:
At
the time of the acquisition of the Washington Mutual banking
operations, the assets of the WMM Trust were comprised of Washington
Mutual subprime credit card receivables which had a much lower quality
of assets relative to that of JP Morgan's. In order to closely conform
WMM Trust's quality to the overall quality of a typical JPM credit card
securitization master trust, during 4Q08 JPM randomly removed $6.2 bn
of credit card loans from the WMM Trust and replaced them with $5.8 bn
of higher-quality receivables from the JPM's portfolio.
However,
during 2009 the credit quality of the WaMu portfolio deteriorated so
sharply that it could have led to an early amortization event unless
additional actions were taken. On May 19, 2009, JPM removed all
remaining credit card receivables originated by Washington Mutual.
Following this removal, the WMM Trust collateral was entirely comprised
of receivables originated by JPMorgan Chase. As a result of the actions
taken by the company, the assets and liabilities of the WMM Trust were
consolidated on the balance sheet of JPMorgan. Consequently, during the
2Q09 JPM had recorded additional assets with an initial fair value of
$6.0 bn, additional liabilities with an initial fair value of $6.1 bn,
and a pre-tax loss of approximately $64 mn. This was the VIE
consolidation referenced in the forensic research reports linked above.
During
2Q09 JPM securitized $12.5 bn of credit card receivables. The expected
loss rate built into these credit card receivables is at 8.7%. The
Company's assumption regarding expected loss rate build seems highly
optimistic, rosy and reminiscent of that of a Goldilocks fairy tale
considering the current charge-off rate for JPM's credit card loan
portfolio stands at 12.5%.
Of
the $23.2 billion securitized interests held by JPM, 25% is classified
as noninvestment grade. Interest in noninvestment grade securities form
6.0% of tangible equity while interest in investment grade securities
form 18.5% of tangible equity.
Next up is Bank of America, who reports later on this week. While I am sure they will try their best to be as accurate and responsible in thier reporting as humanly possible, I will post some stuff that I am sure their accountants and risk management departments may have inadvertantly forgot to include...
- advertisements -
Great research, Reggie. Will follow all your postings from now on. Do you have any data on UBS?
Reggie, do you also have fresh roasted coffee? Get the UBS report and the Coffee to Anon #99020 asap!
Today, five US banks, according to data in the just-released Federal Office of Comptroller of the Currency's Quarterly Report on Bank Trading and Derivatives Activity, hold 96% of all US bank derivatives positions in terms of nominal values, and an eye-popping 81% of the total net credit risk exposure in event of default.
The top three are, in declining order of importance: JPMorgan Chase, which holds a staggering $88 trillion in derivatives; Bank of America with $38 trillion, and Citibank with $32 trillion. Number four in the derivatives sweepstakes is Goldman Sachs, with a mere $30 trillion in derivatives; number five, the merged Wells Fargo-Wachovia Bank, drops dramatically in size to $5 trillion. Number six, Britain's HSBC Bank USA, has $3.7 trillion.
What Geithner does not want the public to understand, his "dirty little secret", is that the repeal of Glass-Steagall and the passage of the Commodity Futures Modernization Act in 2000 allowed the creation of a tiny handful of banks that would virtually monopolize key parts of the global "off-balance sheet" or OTC derivatives issuance.
http://www.atimes.com/atimes/Global_Economy/KD03Dj02.html
with the repeal of glass steagall , we now reap what we have allowed to be sown. citizenship is a terrible thing. unless the owners are continually watching, bad things happen and so it has this time again. so as a texan, i have to say rather sorrowfully, that the repeal of this act, was achieved with the robust help of one, former texas senator, phil gramm, who by the way,had his name mentioned numerous times etc back in the late 1980's during the savings and loan crisis. perhaps one day, he will get his just rewards, one can only hope. of course let us not forget his wonderful wife, cindy, who was on the board of directors for enron. the fruit doesn't fall far from the tree does it?
so this ticking time bomb is still in place. nothing can or will ever be done about it. only time will be the judge of this. the time is running out. can the taxpayer of the united states be forced to back stop trillions in real losses? i think not. but it is fall in amerika. time for football and hockey and the baseball playoffs. let us drink, for tomorrow we may die....
The data you used is from Quarter I, here is the data for Quarter two, GS jumps to the second
http://industry.bnet.com/financial-services/10003641/for-banks-derivativ...
http://i.bnet.com/blogs/derive4.png?tag=content;selector-perfector
The data you used is from Quarter I, here is the data for Quarter two, GS jumps to the second
http://industry.bnet.com/financial-services/10003641/for-banks-derivativ...
http://i.bnet.com/blogs/derive4.png?tag=content;selector-perfector
They will eventually fill the gap down in the mid 20's....
if you want best of breed, wait until they're not lying
anymore:)
I have discussed this extensively on my site, but I want to make it clear here as well. I actually think JPM is best of breed of the big banks. That being said, JPM is NOT in a good position in terms of credit quality of assets, risk and its balance sheet. What makes it best of breed is that its peers are that much worse. This does not make JPM good, just better. In addition, JPM is percieved as being in MUCH better condition than it is. This is the advantage to be had from name brand value, a government that aids and abets in creating reporting opacity, a market full of traders that move on momentum over fundamentals, and most telling - mass market financial media that fails to dig deep down from an analytical and investigative perspective and appears content to simply regurgitate sound bites, ex. green shoots, etc.
The real issue at hand here is the amount of risk concentration that these banks have. They say they hedge this and that, but when 5 banks hold 97% of the notional deriv dollars and they are all basically in the same businesses, how the hell can you hedge the risk. Nearly all risk is in the same tight circle of 5 counterparties. If one falls and it hedges with the other 4, the other 4 fall with it. We actually had less counterparty risk before Leh and BSC collapsed, unless you literally count Uncle Sam as a counterparty (which he is now, ask AIG, JPM and Goldman as Sam backstopped BSC and AIG).
Yes, thier swaps book is much larger than what is mentioned above, but realize that it is all cumulative and the stuff above is not trivial - even less so when added to all of the other issues. From this take into consideration that the MSM says JPM blew numbers out of the park with extreme earnings. They also blew chunks as extreme credit losses and reserve builds as well, and the earnings for banks are opaque now that the government has allowed them to fudge the numbers, so very few people actually have any grasp as to what the actual economic (as opposed to accounting) earnigns are. It appears as if the MSM is ignoring the balance sheet simply to focus on the income statement.
when 5 banks hold 97% of the notional deriv dollars and they are all basically in the same businesses, how the hell can you hedge the risk. Nearly all risk is in the same tight circle of 5 counterparties. If one falls and it hedges with the other 4, the other 4 fall with it.
I agree completely, and think this is self-evident, though I am sure there are some who would argue the point, especially wrt interest rate swaps. Notional value becomes real value in the event of default. Your previous piece on the bond insurers is informative. The risk in the derivatives is leverage, and any highly-leveraged large institution that crashes takes the whole system down with it.
No, this JUST isn't true.
Let's say you have a $1bn irs with a counterparty, and the next day put on a $1bn irs in the opposite direction.
You have tiny risk, but now you have $2bn notional.
People like you really make me sick.
Your ignorance is truly astounding.
No, because banks never do that.
Only self-deluding bankers think that they hedge their risks on a 1-1 basis. In the real world, there is only over-hedging and under-hedging. Inevitable, you can only hedge more or less than the original risk. Remember that even a notionally equal hedge minus the compensation to the bank that arranged the hedge is an under-hedge.
Under-hedging is fine. No-arbitrage, it costs a little money, but it smooths risk/return over time. Macro, it's beneficial because the theory suggests that smoothed risk creates calmer, more-liquid markets.
Over-hedging is not fine and what the notional numbers certainly seem to suggest is that banks now engage in overhedging. If overhedging is not fine for basically the reason that if overhedging worked then martingale betting systems would work. They don't. Overhedging only appears to mitigate the full risk (and give quite a little profit - important safety tip: hedging should never create much profit). In fact, what it mathematically must do - all else held equal - is build risk exponentially, albeit a small exponent on a very few basis points.
Even over-hedging would be bad but okay if successively larger hedges had to be paid in the order they were made or all hedges were magically netted out simulaneously. But as we saw with AIG and Goldman, efforts to do that quickly break down. The largest and most-knowledgeable counterparties are most likely to back out of the "agreement" among participants to settle hedges in an orderly way and allow hedging to continue as a crisis unfolds. What they will do instead is start demanding payment on the biggest, worst hedges first so that inevitably one counterparty in the group will find its marker called on a big hedge and nobody to lay the risk off onto. A systemic liquidity crisis is a virtual certainty in that event.
The saving grace of our financial system is that all else is not held equal. Overhedged counterparties are, in effect, bailed out by new participants coming into the system and building liquidity - until they don't. The usually come back, but if you keep playing that game you are going to lose. You can't win by martingaling your martingale, you just build risk ever-faster.
What does netting get you in a meltdown?
Tiny Risk.
U.K. Faced "Bank Runs, Riots" as RBS and HBOS Neared Collapse http://www.bloomberg.com/apps/news?pid=20601087& Royal Bank of Scotland Group Plc and HBOS were close to collapse, causing a chain reaction that could have ended with riots in U.K. cities, security analysts and economists said. Bank failures would have forced the government to cancel police leave and deploy troops as the breakdown of the financial payments system threatened the ability of utilities to provide essential services, said David Livingstone, a fellow at the Royal Institute for International Affairs in London, a former adviser to the government's Cobra crisis response committee. "You are talking about a situation with mass disorder and panic," the former Royal Navy officer said in an interview. There would be "riots, pandemonium, everyone fending for themselves."
Why wait for another meltdown... the big 5 have 97% of swaps outstanding.
Kick the alphabet soup crutches out from underneath these fuckers... nationalize and net these swaps to zero... then split the carcasses up into something not too big to fail.
Too Big to Fail = Too Big to Survive
As an added bonus it goes a long way towards solving regulatory and governmental capture...
NO THEY DON'T HAVE 97% OF SWAPS OUTSTANDING.
That shitty report didn't include EUROPEAN banks.
Your right... let's kick the crutches out from under those MF's as well. We could have missed a few if it wasn't for you... thanks!
Because there's no good reason to consider that both counterparties could fail, and simultaneously even. That would be truly unsettling.
Maybe he works for the IMF and is counting "the system as a whole": http://www.imf.org/external/pubs/ft/gfsr/2009/02/pdf/text.pdf
Caveats to the Application of Estimated Security Loss Rates to Bank Holdings
Our approach for estimating mark-to-market losses on securities includes only cash instruments, and thus does not account for potential leveraged exposures. As in other iterations, we assumed that derivatives exposures net out to zero for the system as a whole. We did not account for concentrations of counterparty risk.
What about counterparty risk?
Hey... guy said it's tiny... besides the Fed's got ya covered... until they don't.
Insurance works until you need it. Good thing the FED can Kill All Humans as it's escape clause.
Doesn't this all pale in comparison to JPM's trillions ($88 Trillion ?) in swaps (mostly interest rate sensitive I/R swaps) ???
bonddude...i believe the answer is yes.
i think mr. middleton has writtent he same to this effect but he can respond accordingly.
nice job Reggie...you do good work.
I don't know. The numbers you see for swaps seldom represents the real amount at risk. That's probably a notional amount, and besides, there would be netting involved and offsetting positions. There's probably a significant amount of risk there, to be sure. But I don't think it's captured in that number, and I don't know if the securitization issues necessarily pale in comparison.
nice piece, Reggie. my question is this: even if MOST of this crap blew up, it's still a managable proportion of JPM's equity, right?
Here are a few comments.
1) What do you expect when you give people 0% money for only a 3% charge?
2) What do you expect when you raise the rate to 22% on all current WaMu card accounts?
Given my experience with the WaMu accounts, here is my take. JPM takes over the accounts and raises the rates on all WaMu accounts while forcasting that some people will pay early and some will default. However, the good accounts are paying off the accounts and leaving. They will receive capital faster but they will be left with large default values and reduced return on capital on their remaining accounts. In addition, the good credit accounts will be expensive for JPM to win back. In other words, short term gain for long term pain.
Of course, this is my take. I only had two WaMu cards that were raise to 24%. Granted the debt was mine but I used it to pay off hospital bils. Funny, I was able to get rid of one account in one month and will get rid of the second account this month after three months. Somehow, I don't think that my business will come cheap for them in the future.
Also, I suggest this for everyone closing an account. Overpay it by $1 or $.50. Make them send you a check for the remainder. It is childish but it costs them money to send you a check. Also, if the check gets lost, they have to cancel the other check and resend it to you. I have a few $1 checks as I have closed accounts.
But these were supposed to be sold off and OTHER people in far off places were supposed to take the losses. They are simply unexported grenades that won't blow up until you ship them off and pull the pin?