ES -- DX/CL -- isee -- cboe put/call -- specialist/public short ratio -- trinq -- trin -- aaii bull ratio -- abx -- cmbx -- cdx -- vxo p&f -- SPX volatility curve -- VIX:VXO skew -- commodity screen -- cot -- conference board

Wednesday, April 16, 2008


credit default swap market still growing

and at breakneck pace, per the financial times via yves smith.

The total volume of outstanding credit derivatives contracts stood at $62,200bn at the end of last year, up from $34,500bn a year earlier, the International Swaps and Derivatives Association will announce at its annual conference in Vienna today. This is 10 times the level of four years ago.

these are incredible figures -- $62tn is several multiples the size of the corporate debt market in total, and four times the gdp of the united states. but most shocking should be the underlined -- this is a highly leveraged speculative market which is completely untested at anything like its current size in an actual cycle of recession and accelerating real bankruptcies.

this -- and not housing, amazingly enough -- looks to this observer to be the true dark heart of the leverage boom that appears to be ending now. when this market is finally tested later this year, when corporate bankruptcies spike as the credit crisis overtakes weak balance sheets dependent on easy refinancing and rising asset prices, when a significant player beyond even the new expanded envelope of protection offered by the federal reserve goes bust and cannot pay -- only then will we really be made to realize just how incredibly excessive the credit boom that began in 1982 had become in its late stages.

as institutional risk analytics well said two days ago as part of an intelligent polemic against fair value accounting in a world of highly irrational markets:

For some months now, we've been pondering what happens to all of those net short credit default swap portfolios at dozens upon dozens of hedge funds that will be going out of business this year due to the Great Unwind. Hedge funds have no permanent capital, thus there are no assets available to support the defeasance of a book of net-short OTC derivatives positions should the fund be forced into involuntary liquidation.

In such a scenario, you can forget about netting; won't be nothing left to net, in or out of bankruptcy. And since the old habit of simply writing more CDS contracts is not available once the fund starts liquidating, we wonder if leading CDS dealers like JPMorgan (NYSE:JPM) won't be forced to take these trades back as hedge funds expire. What's the "fair value" of a book of short OTC derivative positions taken by a dealer in payment of other debts?

Indeed, if you think of BSC not as a broker dealer, but instead as a clearing customer of JPM, then the logic of the acquisition makes perfect sense. JPM could not let BSC go into Chapter 11 because doing so might have started a chain reaction among the OTC derivative counterparties of both firms.

Between JPM, BSC and BSC's customers there were three levels of leverage, making the ratio of Economic Capital to Tier One Risk Based Capital computed by The IRA Bank Monitor (4.7:1) for JPM at the top of the leverage pyramid seem entirely too generous! If you impute even a fraction of the downstream leverage residing with clearing customers to JPM, the giant bank's capital shortfall becomes alarming.

A bank holding company, after all, is thinly capitalized and in many ways was the precursor of the hedge fund model. On a parent-only basis, JPM's $314 billion asset balance sheet includes $200 billion representing investments in its subsidiary banks and non-bank units, supported by half as much equity and more than $200 billion in debt.

And remember that JPM's on-balance sheet capital does not even partially support the counterparty risk of its vast OTC derivatives businesses, thus the BSC acquisition was a "must do" deal for Mr. Dimon. Think of it this way: JPM is essentially an uncapitalized, $76 trillion OTC derivatives exchange with a $1.3 trillion asset bank appendage. By the way, we are working to include factors for OBS securitizations in the next iteration of our Economic Capital simulation in The IRA Bank Monitor.

But you understand that Fed officials still believe, even today, that the US markets are not over-leveraged.

when one begins to understand that some hedge funds have made a living over the last few years by doing little more than collecting premium from selling CDS and magnifying that into big steady returns by operating over huge leverage -- and that very probably some are still doing so, to judge by the reckless pace of expansion in the market -- it becomes obvious that, in a real recession with real bankruptcies, counterparty risk is not a potential hazard but an inescapable and imminent eventuality.

a month ago questions started to surface about jpmorgan chase in relation to its commerical banking exposure. but a far more central threat to it is its status at the single biggest player in credit default swaps. i don't know where it ends for the house of morgan when the CDS market finally is faced with deleveraging and forced liquidation. but i suspect its implied status as one of the banks simply "too big to fail" will be tested to the fullest.

Labels: ,

Thanks for your continued insightful commentary!

One thing I don't understand is how leverage is applied to the CDS market... is it that the party selling the swaps is borrowing money to meet whatever capital requirements are required to issue the swaps, so that they can issue more of them?

A more general question about fractional reserve lending that I haven't found an answer to: When a loan is made by a federal reserve bank, the loaned money is "created" (with reserve set aside). How are loans sold (whether securitized/packaged or not)? Is the selling price the value of the future interest payments PLUS the principle, or just the value of the future interest payments? When a borrower defaults on a loan that has been sold, whose money is "destroyed" to compensate? When a borrower pays down principle, is the principle "destroyed" still -- i.e., the effect of fractional reserve lending transfers to the buyer of the loan and the buyer is at risk for much more than the face value of the loan?

I can't find any articles discussing this, but I'll understand if you (or other readers) don't want to play the role of educator :)

------ ------- ------
hbl -- i'm happy to educate where i can!

i think that's a good general view of how leverage underlies CDS. a capital pool (eg hedge fund) uses its capital as collateral against a much larger loan, which enables it to support a large balance sheet. that balance sheet can be expanded to post collateral against CDS that are moving against the fund -- up to a point.

however, if the pool is faced with redemptions, it may be forced into rapid balance sheet contraction and indeed may not be able to post collateral. creditors may also pare or call lines to the pool, forcing sales into an illiquid market. a pool could also be overexposed to a rash of credit events and fail outright.

------ ------- ------
re: the fed -- currently, all loans being made by the fed are "sterilized" -- which is to say that when the bank extends a loan to a member bank, rather than creating money on the liability side the fed is offsetting the new asset on its balance sheet by selling other assets (eg treasuries).

the fed is also engaging in asset swaps -- that is, loaning treasuries in exchange for other, lower-quality assets like ABS. this too has no money-creating effect.

the result has been a steady monetary base.

if the fed gets into the inflationary business of creating cash -- "quantitative easing" -- it would simply make loans (ie increase assets) and, rather than sell other assets, increase that all-important liability of the fed, cash. this would have the side effect of driving the fed funds rate down to near-zero.

alternatively, it could expand its balance sheet by creating as a liability not cash but federal reserve bonds, which would be sold to the public or the treasury.

that hasn't happened yet. if it happened, the fed would likely sell them at a discount and redeem them at some future point at par (the difference being the return to the investor).

------ ------- ------
Thanks for the response!

I think you answered my first question, but unfortunately I think I misphrased my second question. I was referring to loans made by federal reserve chartered commercial banks -- i.e., the textbook scenario of how money is created via lending. And I'm trying to determine how the creation and destruction of money via that mechanism translates into a world where loans are routinely packaged and sold to other parties.

Your explanation of the Fed's monetary actions are a great concise summary but I think I already had a grasp of that side of things :)

------ ------- ------
... but in rereading your comment, hbl, i think you may not be talking about the fed but any member bank of the federal reserve system! different topic:

when a deposit-taking bank makes a loan, it effectively shifts cash (which was booked to balance the liability of the deposits) to accounts receivable. it can regain cash by then selling the loan.

loans are sold in all kinds of ways. sometimes they are sold whole -- sometimes stripped into principal-only and interest-only components -- sometimes pooled and tranched into ABS. the cash that can be received in exchange has everything to do with the marketplace, depending on perceptions of interest-rate risk, default risk, prepayment risk -- very complex, particularly in the case of mortgages.

when the borrower defaults, the creditor does effectively see his asset "destroyed", at least in part. if the loan is secured, he gets something tangible (a car or house) to try to redeem. if the loan has been sold, whomever is holding the principal-only slice is an injured party. if the loan has been pooled and tranched, it's whomever is next in line to take the loss.

if the borrower prepays rather than defaults, obviously the principal is repaid and there's no injury (it's the interest-only slice that's hurt exclusively, as there won't be any more interest payments).

in the abstract, whether or not money is destroyed in this instance is up to the now-repaid creditor -- if he makes/buys another loan, then no; if he uses his returned capital to pay down an obligation of his own (ie, reduce the size of his balance sheet by offsetting a liability with cash), then yes. this isn't optional in case of default, obviously -- the balance sheet is reduced, with the loss of the asset being paired with a loss of equity.

------ ------- ------
I very much appreciate the detailed response. I think I need to learn some accounting (beyond the very basics) to understand your explanation of the balance sheet mechanism fully, but I think I understand the answer to at least the broadest aspect of my question in that total money supply (via larger balance sheets) does stay expanded even as loans move around the system, until they are paid off or default (on aggregate).

Part of why I'm curious is I'm trying to form my own opinion/understanding of the feasibility of the fed preventing broad money supply (and consequently asset prices in general) from significantly contracting, even if it monetizes newly issued treasury or federal reserve debt. It seems like the answer should be knowable but yet there is so much disagreement out there on inflation/deflation/etc.

Sorry, I haven't used blogging software otherwise I'd try to help on the RSS question (from a previous post), as I am technical.

------ ------- ------
I'm trying to form my own opinion/understanding of the feasibility of the fed preventing broad money supply (and consequently asset prices in general) from significantly contracting, even if it monetizes newly issued treasury or federal reserve debt. It seems like the answer should be knowable but yet there is so much disagreement out there on inflation/deflation/etc.

it's the $64 question, to be sure. the other day i posted this, which i think gets close to the heart of things. it really does come down to reducing claims (that is, dollars of credit/debt) per base dollar (what the fed controls) to a less dangerous level as painlessly as possible.

the ratio contracted meaningfully from 1990-94 as the savings and loan bust played out. this was less a deleveraging than simply a stagnation of lending while monetary base grew beneath as a pretty healthy clip. that same dynamic, over a longer period, *could* be our fate this time too -- a "malaise".

but i obviously fear that the amount of debt -- particularly financial sector debt -- has gotten so large as to make deleveraging a more dangerous process, prone to financial earthquakes. moreover, i think we're restricted in how much we can expand monetary base vs gdp because, as an international debtor nation, a really aggressive devaluation may result in a dollar crisis and repudiation resulting in a severe "standard of living adjustment" (read: depression).

i do, though, think the muddle-through possibility is a strong case, and too much attention (including here) is paid to the less likely extremes.

------ ------- ------
I did read that post. It struck me as likely very intelligent and insightful (I have not seen any other sources or bloggers comment on it) but I have to admit I don't understand the importance of the ratio of debt to base dollars or how such a small increase in the monetary base in absolute terms (as opposed to the large 37% in percentage terms) can bring the financial system back to solvency with a likely 20-30% decline in housing prices plus other inevitable asset price declines. It seems like it would be trivial for the Fed to add that much to the monetary base by buying just $320 billion in assets (with congressional approval if needed), if that's all it took to solve our problems, and that the impact on the currency or price inflation wouldn't be too significant... but perhaps I don't fully understand the inflationary impact of increased base money supply versus increased broad money supply if the former has a higher impact. Or is it that this large an increase in base money supply would put short term rates uncomfortably near the zero bound? Perhaps if you do another post on this topic you could elaborate on the rationale behind this measure.

My intuition tells me muddle through is improbable based on the seemingly large gap over the last several decades between money creation (along with asset appreciation) and underlying productivity growth (the only truly sustainable driver of wealth creation over the long term) plus demographic changes, and that inflation may reverse and send us down a path like Japan's or worse, but I'm trying to understand the actual facts. Your blog is a great source of thoughtful commentary -- keep it up!

------ ------- ------
how such a small increase in the monetary base in absolute terms (as opposed to the large 37% in percentage terms) can bring the financial system back to solvency with a likely 20-30% decline in housing prices plus other inevitable asset price declines.

hbl, i would say that a 37% increase in money supply is a *big* deal. consider that, looking back, that's the move from 1998 to today -- ten years of money supply growth in what many would consider a regime of quite high (if officially suppressed) inflation (much of it due, though, to credit growth of course). and that only to get the ratio back to 40 in a period of credit stagnation, which is still an elevated level. such a period would probably see raw input prices rise dramatically in relative terms to leveraged asset classes, which would be punished by higher interest rates, higher input costs and slow economic activity -- but overall asset prices would hopefully rise as debt remained stagnant, giving a nominal deleveraging.

you'd also be explicitly signalling to a lot of dollar claimants that you were going to pay them back in significantly devalued currency -- which has its own dangers.

in 1990-94, inflation moderated as lending slowed but remained positive -- and as monetary base was expanded 56% over the four years. the system didn't actually deleverage in aggregate (though some parts of it, centered around savings and loans, did) -- therefore, national equity was not seriously consumed. but a deleveraging of sorts was effected. this was a difficult period for banking.

claims over base is really just an approximation of leverage in the system -- the higher the ratio, the less of a decline in asset prices is needed to wipe out equity.

that leverage must come off if chronic disasters are to be avoided. periods of asset price declines are survivable when leverage is modest. but when leverage is THIS high, a modest spark can be enough to consume all the equity in the financial system and leave little to support any debt at all.

that's the fear i share with you, i think. unlike the 1990s, we now have massive private leveraged entities both central to the system and outside the system beyond the fed's purview. i'm not at all sure that an orderly period of stagnation as inflation corrodes real debt loading is possible. investors in that sphere won't patiently wait out subpar profits in the way that capitalized banks can.

------ ------- ------
as luck would have it, yves smith links to a comparison paper, examinign today in light of the early 1990s.

------ ------- ------
The importance of reduced leverage makes sense.

I think I did not understand the significance of a 37% increase in base money because I was thinking it was changes in broad money supply rather than base money supply that would ultimately drive inflation (with a lag of course) -- probably in part because it seems like excessive credit expansion since the early 1980s has been a major contributor to the high but underreported price inflation appearing in recent years. $320 billion is only a ~3% increase on $11+ billion of M3, for example, verus the 37% increase relative to base money supply. I need to find some more good economics texts and fill in more of the gaps in my knowledge. Thanks for your patience. There is another question that has been bothering me for years but I'll spare you for now ;)

------ ------- ------

Post a Comment

Hide comments

This page is powered by Blogger. Isn't yours?