Voice from the Past
Chapter 20

"S&Ls invested in everything from casinos to fast-food franchises, ski resorts, and windmill farms. Other new investments included junk bonds, arbitrage schemes, and derivative instruments." from An Examination of the Banking Crises of the 1980s and Early 1990s, Volume I, Chapter 4, The Savings and Loan Crisis and Its Relationship to Banking. .

My own experience with this crisis involved a remarkable example of a de-regulated savings and loan association. In 1982 and 1983 I worked as underwriter's counsel on an industrial development bond issue for an office building. The obligation of the borrower under the loan agreement that supported the issuer's revenue bonds was to be secured by a collateralized letter of credit. The issuer of the letter of credit was a Chicago suburban savings and loan association that had achieved considerable size by a recent merger and buying spree. It had never issued a letter of credit, and could not do so prior to de-regulaton. I doubt that the S&L's officers had any but the vaguest idea of what such an instrument was, even in its more traditional form for customary commercial purposes. Their lawyer certainly was not familiar with such documents. After failing to find him listed in a couple of directories, bond counsel was concerned enough about that lawyer to ask David Cholst, who was working with me (and complemented my recollection of this transaction), to check with the Illinois Attorney Registration and Disciplinary Commission to make sure that he was licensed to practice.

While you might think that neither an underwriter nor a rating agency would care for a bond issue secured by an instrument with which the issuer was unfamiliar, you should never underestimate the imagination and resourcefulness of underwriters. The transaction was structured so that the S&L's competence and experience (or lack thereof) didn't matter. This sort of transaction was devised to enable savings and loan associations, that had lots of low- interest home mortgages in their portfolios, to make a little money by pledging them as security for letters of credit. For this they would get appropriate fees. It was believed that this practice would comply with the arbitrage regulations because the pledged investments would bear a lower interest rate than the bonds being secured. Thus the yield on the pledged mortgages would not exceed that of the bonds and violate the invested sinking fund regulations. The theory sounded grand.

To get a satisfactory rating on the bonds, it was necessary to over-collateralize the letter of credit. The rating agency would require something like 115% collateralization if the collateral were U.S. government bonds, and well over that if the collateral were single-family mortgages. However the S&L we were dealing with had lots of mortgages that were granted when rates were low, so it appeared that this degree of over-collateralization would not be a problem. The deal went forward.

Then the problems surfaced. The S&L had acquired most of its mortgages by buying them from other saving and loan associations or by taking over the assets of other associations in mergers and buy-outs. The question arose: what, for arbitrage calculation purposes, is the yield on a low-rate mortgage that the current holder bought well below par? Does it make a difference if the mortgages were acquired by purchasing them directly or as assets of an organization that the S&L bought or merged with? How do you allocate the purchase price of a package of mortgages to any one mortgage or sub-group of them? David Cholst was getting his early education in municipal bond and arbitrage matters. I was thankful to have his background in mathematics available.

All these questions proved to be academic, however, because the S&L did not have adequate records to establish reliably the cost of enough mortgages to do the job. Or at least not in time for the closing.

The retired dentist who was president of the S&L nevertheless wanted to go ahead with the deal, perhaps as a matter of vanity. The issue could go forward if the letter of credit were collateralized by government bonds. The rating agency would permit a provision in the documents allowing the S&L to substitute appropriate mortgages (at the higher over- collateralization ratio) in the pledged collateral later, and the documents were so drafted, but with a condition to exclude mortgages bearing a yield in violation of the arbitrage regulations. This condition could be satisfied only with an opinion of counsel experienced in arbitrage matters. I recall wondering under what conditions I could give such an opinion. The S&L then bought a portfolio of government bonds bearing a much lower rate than an equal amount of mortgages would bear, and deposited them as collateral with the trustee under the bond issue.

The trustee had instructions to liquidate collateral if necessary to prevent default in the bonds. As a practical matter, the S&L was in a conduit position analogous to that of the issuer of the IDBs, except that it had liability if the collateral failed. The rating agency looked only to the collateral for security.

Later I heard that the S&L never tried to substitute mortgages for the governments in the collateral; and also that the user of the facility paid the bonds, so the letter of credit was never drawn on. Whatever fee the S&L got for writing the letter of credit was unlikely to have compensated it for purchasing low interest government bonds instead of buying higher rate single family mortgages. It was organized for, and its personnel had experience with, the latter. It was also one of the first to crumble when the crisis struck.

It seemed to me that giving this particular S&L the freedom of de-regulation was like giving a kite its freedom by cutting the string.

Manly W. Mumford