CDS Markets Make The EU Too Big To Fail

Includes: AIG, BAC, C, EU, GE, GS, ITLY, JPM, MS, TLT
by: Chicago Long

For decades, we have supported the moral hazard of bailing out entities that are too big to fail, such as Long-Term Capital Management, the auto industry, GE and AIG. We now face this situation once again. The obvious reason for the EU bailout is to save it from collapse. In reality, however, we face a much deeper systematic problem that could wreak complete havoc on the global markets: the effect of credit default swaps on EU debt. And that's just the beginning.

Investors who hold EU sovereign debt from countries such as Spain, Italy and Greece frequently buy credit protection against default. They do this by entering into credit default swap (CDS) agreements with counterparties such as investment banks. The agreement states the counterparty who sells the credit protection owes a payout to the buyer should a "trigger event" (such as default) occur. Top U.S. underwriters include JP Morgan Chase, Morgan Stanley, Bank of America, Citigroup and Goldman Sachs.

To protect themselves against payout losses, the investment banks then offset some of the CDS with other counterparties to reduce their gross exposure to default. However, these counterparties tend to be European banks and they incur similar risk by entering into such hedges. Ultimately, this means everyone is buying from each other and not completely offsetting their positions (or if they are, they may be doing it assuming certain market correlations exist). What's worse, even if they were all perfectly hedged, swaps inherently have a large amount of counterparty default risk: If one party walks away from the contract, it leaves the other side in a lurch without recourse.

The problem of not hedging these transactions is mind-boggling. Let's draw an analogy using Long-Term Capital Management's on-the-run vs. off-the-run pair trading strategy using 30-year Treasury bonds, one of the safest instruments in the world. LTCM pair traded them by shorting the liquid on-the-run treasuries and buying the illiquid off-the-run treasuries, and consistently earned a small profit due to differences in liquidity preferences. The historical issuance price had a standard deviation of about 85 bps, but if the bonds were paired, the sigma of the trade was reduced to 3.5 bps - almost nothing.

While LTCM took both sides of each trade, its counterparties - 15 investment banks - chose not to hedge all of its trades with LTCM. In the ensuing market collapse, each bond lost well over 85 bps of risk (because of the 1998 Russian default crisis), times two sides (since the trades were not netted), or 35 million sigmas. Multiply that risk times 15 banks and you get one of the original "too big to fail" in the capital markets. Imagine what could happen with a trade that has a lot more volatility and over 20 participating banks worldwide. That is the dire situation we are in today.

This is likely why Greek debt holders agreed to (or more accurately, were forced to agree to) a restructuring haircut on their debt repayment…because a "voluntary" haircut skirts the trigger covenant of the CDS and avoids putting the entire banking system at risk. It's the same reason why Spain must be bailed out now…and others will be in the future.

Whether we agree or disagree with bailing out the EU, the implications for the bigger picture are far more frightening. Because the banks cannot afford to allow any EU country to default, it might lead these countries to act in ways they otherwise might not - because they know they will get away with it. In other words, there is a possibility that the EU countries may emulate the mistakes of those that have already been bailed out - so as to ensure they are bailed out, too.

Disclosure: I am long GE.