I am sure that almost everyone has seen at least some of the coverage about the SEC’s civil action against the investment bank, Goldman Sachs, and the related Congressional hearings. The focus of the SEC’s complaint is a deal, apparently part of a group with the internal code name “Abacus”, which was based on synthetic Collateralized Debt Obligation based on a pool of sub-prime mortgages.
I am not a lawyer, and certainly no expert on the relevant parts of securities law. That Goldman apparently set up a “sweetheart” deal for one client that involved sticking other clients with a very dubious investment was certainly sleazy, but not necessarily illegal. (It is unfortunately not all that uncommon on today’s Wall Street, either.) Similarly, the idea of a firm like Goldman making a “bet” against the US housing market may be unappealing to you, but that in itself is part of how the market works.
However, there are two aspects of the deal that I think really do represent problems that need to be addressed by clarifying and strengthening the rules of the game. The first concerns the use of debt ratings, by firms like Standard & Poors, Moody’s, and Fitch. These companies got their start, many years ago, rating corporate debt. That was a fairly transparent process, not least because the ratings were largely based on corporate financial statements, which were published. The rating agencies, in those early days, were primarily acting as information gathering and collating agents. Today, the agencies still rate corporate debt, but they also give credit ratings to deals like Goldman’s Abacus. The rating agencies’ fees are paid by the issuer, creating an obvious conflict of interest. Today, some of the portions of the issue, originally AAA rated, have been downgraded to junk status. This matters because institutional investors, like pension funds and insurance companies, often have statutory or policy restrictions on what type of debt they can buy — restrictions based on the agency ratings. This makes absolutely no sense if the agencies are paid or not at the pleasure of the investment banks.
There is something more fundamental that is troubling about this kind of deal, though. When I worked in the investment business, starting back in the early 1970s, we told people that what we did, in choosing equity investments, had a useful social function: the efficient allocation of capital to businesses, as determined by market participants. That is, essentially, why we have a stock market. But the Goldman trades were different: they did not result in any incremental investment in any productive economic activity. They were simply a vehicle through which Goldman and its favored clients could make a large bet on what would happen to the housing market. Of course, in the end, someone would win and someone else would lose, but in terms of contribution to the real economy, they might as well have made the bet on the roulette wheel in Las Vegas. I suppose you might argue that the process contributes to price discovery in some way; but then, that is surely an argument for making all such trades public and transparent.
That gambling transaction might have been acceptable, though distasteful, back in the days when Goldman Sachs was a partnership, and the partners were betting with their own money (although their actual bets back then were a lot more conservative), but Goldman is now not only a public company, but a bank holding company, with access to the Federal Reserve as a lender of last resort. I can think of no good reason that the government should be subsidizing a special casino for high rollers, especially when the gamblers keep their winnings, but share their losses with the taxpayer.