The Securities Lending Fiasco

Most, if not all life insurance companies engage in securities lending to some degree.  AIG did it in a big way, involving almost all of their life subsidiaries.  When a life insurer lends out its bonds, they receive back safe liquid collateral equal to 100-102% of the par value of what they lent out.  Most companies leave well enough alone at that point.  After all, you still receive the income on the bonds you lent out, plus securities lending fees.  The borrower receives the income on his collateral, less securities lending fees.  The borrower sells the bonds he borrowed, hoping to buy them back cheaper.

So far, so good, but AIG added a wrinkle to the game.  The safe liquid collateral was a slack asset to them.  Why not replace it with equally safe and liquid assets that offered considerably more yield, like bonds backed by AAA-rated subprime or Alt-A mortgage collateral?  After all, AIG was already writing financial reinsurance through default swaps on such mortgages, why not add to a winning bet?

They did so in a big way:

It took an amazing amount of skill to lose 76% of the surplus of the affected life companies.  One company, American General L&A IC, lost more than double its surplus.  Wow.  Why did this turn out so wrong?  The assets were mismatched to the liabilities in two ways:  1) The mortgages had longer lives than the securities lending transactions.  Even if there were no credit issues, there was no way to assure that the mortgage bonds would be worth the same at the beginning and end of the transaction.

2) Though AAA-rated, they were not credit risk-free.  Non-prime mortgages were made to borrowers of lower quality.  Of their own, they wouldn’t be investment grade, much less AAA, without credit support.  That credit support came through subordination.  Other investors would take the first X% of losses before the AAA bondholders would take any losses.  That X-factor was set too low.  In order to maintain a AAA rating, the X-factor not only has to be high enough that losses don’t harm the AAA investors, it has to be high enough that other investors would think that it would be almost impossible for losses to harm the AAA investors.

As a result of the securities lending losses, and the troubles at AIGFP, the Fed and Treasury began the bailout of AIG.  (Look at the above table to see the amount pumped in and taken out of each subsidiary on net.)  Why did they indirectly bail out life insurance companies that they do not regulate including one that mainly serves foreigners (Alico), by bailing out the AIG holding company?

I can’t be totally certain here, but I suggest that all major state insurance regulators should send Ben Bernanke, Tim Geithner, and Hank Paulson some really nice gifts, because had AIG’s life companies failed, the state guaranty funds would have been hard pressed to come up with something north of $10 billion by surcharging the other insurance companies doing business in each state.  At a time like this, where many life insurers, particularly ones facing credit risks, and those having variable policies, where profitability has declined along with the stock market, the surcharges could have kicked additional life insurers over the edge, and who knows how big the cascade would have been.

(Note to corporate bond managers managing insurance money: this is why you don’t own insurance bonds in your neck of the industry.  The company you manage money for already has contingent credit exposure to all of their peers through the guaranty funds.)

AIGFP was the bigger issue, but the domestic life companies of AIG posed a separate, distinct issue that the US Government addressed, right or wrong.

This entry was posted on Wednesday, April 29th, 2009 at 9:55 am and is filed under Insurance, Real Estate and Mortgages, Speculation, public policy. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.