The insurer has suffered $18 billion of losses in the last three quarters tied to guarantees it wrote on mortgage-linked derivatives. It ended June with $1.05 trillion of assets. Its failure would likely be larger than that of Lehman, which said it ended August with about $600 billion of assets.

What exactly were they paying their underwriting staff for? Insurance companies are supposed to understand risk, risk management. Did it not occur to them that financial instruments have inherently different characteristics to say chance involved in fire damage? One [fire] has a statistically normal distribution, financial assets based on insurance products, which AIG were selling, carry massive left sided tail risks, with zero compensating right side tail rewards.

With all the downgrades coming yesterday and today, any capital raising will be pretty much a miracle if it gets done. The leverage, again requires estimates of $70 Billion being required. In this current market…two hopes, Bob and none.

This credit cycle is claiming some really big “names.”

I shall use the debacle at Lloyds to highlight some similarities and differences within the Credit Default Swaps [CDS] market.

Lloyds was [is] concerned historically with marine insurance. The pertinent fact regarding marine insurance, at least in earlier times, was the all-or-nothing principal. The ship either made it to port, and the premium received could be booked as profit, or, there was total loss, and you paid out.

Lloyds organized on the basis of syndicates, backed by “Names” who pledged their assets to the paying of claims, proportionately to their share of premium written [for them] by the underwriter, who sitting in the “box” accepted or rejected business brought by brokers.

At some point, the model started to change. No longer did Lloyds specialise, they started to diversify into other lines of business. This in of itself, probably would not have been a critical error, save, that they started taking long-tailed business, that carried unlimited loss. In essence, writing long-tailed business, with uncapped [potential] losses is exactly the same as selling Call Options, the reward is limited to the premium received, but the losses are in theory infinite.

The answer to this problem resides [in theory] within the concept of “Reinsurance” and “Retrocession” where, I as the principal underwriter, limit my [potential] losses by purchasing Reinsurance [stoploss] with a portion of my received premium. The underwriter who accepts this risk, may, offset his risk by purchasing Retrocession [re-reinsurance]

There is no limit to how many times this can be done in theory. In practical terms, the initial premium payment is soon used up.

To cut a long story short, Lloyds, would reinsure risks written, within Lloyds of London, often with the same “Names” in a different “Syndicate” thus, rather than diversifying the risk, they concentrated the risk, compounded by shoddy underwriting, compounded by shoddy reserving, eventually, asbestosis claims killed them, being a long tailed risk, for which they assumed unlimited liability, for which adequate premiums were not charged, adequate reserves were not taken, being paid out as profits, thus bankrupting thousands of Lloyds names in the process.

Which brings us to CDS contracts.

CDS contracts are in essence insurance contracts that insure against the loss on Mortgage Backed Securities [amongst other financial instruments] paying out under terms of the contract.

These contracts are not standardised, thus, are open to all manner of challenges and interpretations, already holders have been caught out.

In a similar way, lets say a Hedge Fund wants to earn premium income, so they sell a CDS contract, but want to limit their risk, buying an offsetting CDS with part of their premium, they have in insurance terms, purchased Reinsurance.

Unfortunately, due to the lack of standardization in the contracts, they may not be covered, and this has already proven to be the case, a Hedge Fund believing their exposure to stand at $100K had to pay out $10M

Here is where we come to two major problems within the CDS market;
*Excessive leverage [monumentally]

The insurance industry, learnt, one would hope from the Lloyds example, that risks via reinsurance and retrocession really must be diversified, otherwise, you find in the event of claims, they overwhelm the ability to pay, thus triggering a systemic meltdown.

Have the Banks learnt this lesson? One would have to say probably not. They have proven once again that they are without doubt some of the greediest and most stupid people in business, but we’ll come back to this.

Under-reserving on the other hand, cannot be fully laid at their door. Money [profits] taken as reserves are tax deductible, thus, only minimum reserves are allowed to be taken. Therefore, should the bank engage in providing risky loans, they cannot adequately reserve against them, placing at risk their capital structures.

Of course, now the losses have been mounting, capital has become a critical issue, with many banks becoming insolvent, necessitating the Fed bailouts, interest cuts, Treasury swaps, etc.

With the CDS market at a notional $63T, and exposure estimated at say $3T, obviously should these fall payable, solvency for any bank simply evaporates.

This seemingly is the argument against the banks currently, and it has merit, which is why the Fed is willing to remove from the market any MBS, CDO, that has linked to it CDS exposure.

The MBS assets, are finite, not unlimited exposure, they probably could be termed “long tail” as the MBS may well have a potential 30yr lifespan, but for the Fed, this is not really a problem. That the risk is finite…viz cannot exceed 100% of value, the risk is quantifiable. Of course, ultimate losses will not run to 100%, the assets will have recoverable value.

By removing the “trigger” from the market and not allowing prices to enter the market, the CDS timebomb should be safely defused. In addition, CDS contracts have generally a 5yr lifespan, again, there is a limit to the risk exposure expressed in time.

Risk in MBS is also intimately tied to liquidity. The inability to close out positions quickly was what in part created the requirement for CDS cover. That the illiquidity of the underlying only magnified the movements in the derivatives, possibly was unplanned, as the standardisation of contracts seems to have initially caused problems, this however should be easily resolved now that it has been identified, but we shall see.

In conclusion, while CDS is certainly a bit of an unknown quantity, and the numbers involved are simply silly, it would seem that the magnitude of the problem has been significantly reduced by the Fed, who have deep enough pockets to absorb the underlying to maturity, thus eliminating the trigger on the derivative portion of the market. Unlike an unlimited long tail exposure, the risk is finite, quantifiable, and thus manageable. This being the case, the crisis will pass in time.

« Previous Page