Interesting Questions Raised by Goldman Case
Click here for a printer-friendly version of this article
The Securities and Exchange Commission’s (SEC) civil case against Goldman Sachs has raised a number of fascinating legal, ethical, and political questions. The case also highlights a number of open questions for finance theory and the future of the derivatives transactions at the center of the case. Let’s look at four in particular:
1. Hindsight bias – The first issue raised is to what extent people’s views of the case are impacted by “hindsight bias.” Since the late 1980s, this psychological term has been applied to finance decision theory to demonstrate that realized events can seem more probable in retrospect than they really were at the time. The finance literature is full of papers describing how this “cognitive bias” makes the world seem more predictable in retrospect, which makes us less able to reflect objectively on past events and, possibly, leads us to more imprudent risk-taking in the future.
In the Goldman case, the hindsight bias stems from the enormity of the housing collapse as well as the financial returns of Paulson & Co. Since 2006, house prices in much of the country have fallen by 40%. As of the first quarter of 2010, more than 15% of all mortgages were either past due or in foreclosure. And this doesn’t include mortgages that have already been terminated and are now owned (or have been sold at auction) by the lender. Given these facts, it’s difficult to think back to a time when few people believed that even the worst subprime mortgages would default at a 15% rate. Prime mortgages defaulted at a 0.24% rate in 2006, meaning a 15% default rate would require someone to believe a subprime mortgage was more than 60-times more likely to default.
Moreover, the decline in house prices has so permeated our minds that it is hard to think back to a time when people argued that national house prices could stop increasing but they would never fall. Fed Chairman Alan Greenspan cautioned in July 2005 – 11 months before the peak of the bubble – that the Fed “cannot rule out declines in home prices, especially in some local markets.” This suggests that some people argued, apparently convincingly in some cases, that individual markets like Las Vegas and Phoenix wouldn’t even see price declines.
When one hears that John Paulson, billionaire credit fund manager, was hand-selecting which subprime mortgage bonds were to be included in a deal he intended to short, it reeks of foul play. But the process would not have seemed stacked in favor of Paulson at the time it was completed, given his anonymity in 2006 and the prevailing optimism about housing prices.
2. Cheapest to deliver – The second issue raised by the case is the concept of “cheapest to deliver.” In derivatives transactions like forwards and futures, the short position generally requires the sale or delivery of a reference security. But the contract usually allows for any number of securities to be delivered as long as they meet the criteria specified in the contract. As a result, the “short” uses complex models and software packages to determine which of the acceptable securities would be the least expensive to deliver to the long position. This can not only result in considerable cost savings, but actually offers the potential for arbitrage if an acceptable security can be purchased in the market at a discount to its delivery price.
The Goldman case is a variant of the cheapest to deliver phenomenon. ACA was willing to pay a certain price to buy exposure to mortgage assets. The cheapest mortgage assets to deliver (or reference, in this case) would be those worth the least, based on their intrinsic riskiness. While cheapest to deliver doesn’t really matter when it comes to Treasury bond futures because the differences across deliverable securities are minute, this problem looms large in markets with greater heterogeneity in the quality underlying assets.
It is not clear to what extent derivatives markets that depend on physical delivery can deal with this issue and continue to exist. Consider a consumer who offers $8,000 for a 2002 Ford Explorer. While that might be the Blue Book value, there is sufficient variance in the value of 2002 Explorers available in 2010 that few would be willing to enter into a contract to take delivery of any 2002 Explorer for $8,000 because of the likelihood that the “short” would seek out an Explorer with a checkered maintenance history, pay less than $8,000, and pocket the difference. It’s not clear that the same problem won’t arise for derivatives dealers.
3. Information Asymmetry and Caveat Emptor – The sort of foresight ascribed to economic actors in the “rational expectations” school spawned thinking on the ways that investors’ knowledge could actually result in market failure. For example, managers have inside information about the firm that outside investors do not possess. When a firm issues equity, the outside investor is likely to wonder whether this is because the manager thinks the firm is overvalued. This presumption could result in a risk premium that lowers the price of the shares that could actually result in the cancellation of otherwise profitable projects. Information asymmetries could also lead to credit rationing in cases where a lender believes a firm’s willingness to pay a higher interest rate is a signal about the investment’s riskiness rather than its inherent value.
The bottom line is that sophistication does not translate to perfect information. At the heart of the Goldman case is to what extent ACA and others were aware of the selection bias involved in structuring the pool of reference mortgage bonds. Rather than a random sample, the referenced mortgage bonds were selected specifically because of their undesirable pool-level qualities like high LTV and low FICO scores. It will be up to a judge to determine precisely what was known and what was legal. The point is that many transactions rely on outside investors speculating about the extent to which insiders are exploiting their informational advantage.
As the asymmetric information literature demonstrates, if sophisticated investors are led to assume that the insider is offering a security solely for the purpose of profiting from private information, the market will break down. This point was driven home in the health care debate, where coverage mandates were defended as means to guard against adverse selection. When sophisticated parties assume the only reason you want health insurance is because you know you you’re sick, the volume of coverage will be much lower than with perfect information. The same idea applies to financial markets when insurance is replaced with securities and coverage is replaced with capital accumulation.
4. Efficient markets – Finally, it is worth noting the extent to which the Goldman case confirms one’s preordained view of “market efficiency.” To some, the efficient markets hypothesis means that markets are always “right,” as they reflect all available information. To others, the hypothesis – whether weak, semi-strong, or strong – relates to the ability to beat the market on a risk-adjusted basis. Some argue that the price of housing and volume of subprime mortgage bonds issued in 2006 was an example of inefficient markets. Yet, this view overlooks the amount of risk Paulson & Co. and other investors had to assume to bet on the market’s eventual decline.
Closer analysis reveals that “bubbles” are often entirely consistent with an inability to beat the market. An investor aware that Japanese land and stock prices were unsustainable in 1987 would have likely gone bankrupt from shorting them before seeing his prediction proven true in the crash that soon followed. Similarly, the 2004 run-up in house prices was followed by another eighteen months of gains, leading to margin calls and huge losses for anyone “short” housing during this period. Bill Gates predicted Kodak’s fall due to digital technology in 1991 – yet the stock increased by 200% over the next ten years before crashing.
Nobody believes that markets are always right, but this has little bearing on the ongoing efficient markets debate. That is not surprising given that the same data are able to prove both that prices are “wrong” and that investors cannot beat the market on a risk-adjusted basis.