Challenging the ontology of financial securities
June 29, 2008
I’ve just returned from “A Turn to Ontology?” a provocative workshop at the Said School of Business in Oxford. It was organized by the resident Science and Technology Studies group: Steve Woolgar, Javier Lezaun, Daniel Neyland and others. This group is by itself interesting: they study things such as focus groups, advertising — the non-finance counterpart of social studies of finance. (Last year, two alumni of this group, Elena Simakova and Catelijne Coopmans put together an interesting workshop covered here.) Given all this, I attended the workshop in search for new clues to think about Wall Street.
The main contention of the workshop, according to the organizers, is that “the essence and existence of entities are best understood as the temporary upshot of interconnecting relations.” The most circulated example came from Daniel Nyland and Steve Woolgar:
Recent work on mundane terror shows how ordinary objects in eg airport settings become transformed into objects requiring various apparatuses of regulation, monitoring and control (…) ordinary objects acquire an insecure ontology. A water bottle is transformed into a potential object of terror. (…) For example, when Dan and Steve travelled from Heathrow recently, they started by getting security cleared (and having various liquids confiscated in the process). Steve then purchased a terror free bottle of water in the departures lounge. But they managed to misread the signs (in the chaos of terminal 1) when walking to the departure gate and found themselves again the wrong (“dirty”) side of another security check. At this point the terror free bottle was transformed into
an object of potential terror and promptly confiscated.
The ontology of a water bottle, the example suggests, is not simply determined by the properties of the water, but by the social arrangements at the airport. Is any of this relevant to finance? I am not yet sure. I have already stumbled upon ontological issues in my research; what I’d like to do is to discuss it here and let the reader judge.
The setting was my ethnography of arbitrage. After two years of fieldwork at the equity derivatives trading room of pseudonymous International Securities with David Stark, I eventually concluded that modern arbitrageurs sustain above-normal returns by challenging the ontology of normal securities. In effect, the arbitrageurs that I followed, whether in statistical arbitrage, options arbitrage or merger arbitrage, devoted themselves to breaking up and isolating the properties of stocks to their own advantage.
How? Consider, for example, merger arbitrage. By buying the target company and shorting the acquirer, arbitrageurs cancel out their exposure to those factors that are common to the two– typically, the industry. So, for instance, when HP made a bid for Compaq and the traders played the trade, they were both short and long on the computer industry, thereby cutting their net exposure to the computer industry. They limited their exposure to the only factor they devoted their research to, the completion of the merger.
To see ontological dimension to this, think of the exposure created by a stock as a pie. Its overall value is determined by different qualities: its liquidity, volatility, mean reversion in the price, sector, etc. Think of these qualities as portions of the pie. Whereas buying a stock – let’s say, IBM – entails exposure to all the factors that determine its price, arbitrage entails limiting one’s exposure to just one. What arbitrageurs do, in other words, is to break up a whole into its constitutive parts.
This challenge to ontology came apart this summer in August 2007. As a remarkable paper by Amir Khandani and Andy Lo has argued, this approach at breaking up the qualities of a stock was subject to disruption. As multi-strategy funds lost money on their subprime bets, they winded down their positions in other, more liquid bets such as merger arbitrage. This seemingly inexplicable fall in stock price then led to further margin calls and, in turn, additional losses — adding up to a staggering nine percent in one day.
Recombining wholes into parts is not unique to finance. We see the same in the new high-tech Spanish cuisine. Star Spanish chef Ferran Adria isolates the individual qualities of food – their tact, aroma, color and texture — and then reshuffles them in the form of liquid paella, puffy rice grains or foamy coffee. Like Adria, equity arbitrageurs selectively strip their exposure of their unwanted qualities, then reshuffle them to their advantage.
These challenges to conventional financial ontology can also be found in fixed income. Indeed, it is arguably one reason for the current mortgage crisis. Securitization, or the aggregation of different mortgages into a single combined asset, creates a whole out of different parts. The operation first endows the mortgages with calculativeness – an average default rate, a mean return and a standard deviation. More recently, structured finance has also allowed bankers to transform bonds into different parts, or “tranches,” of different degrees of risk. (Incidentally, for outstanding treatment of the mortgage crisis, see the article by Donald MacKenzie.)
In this sense, securitization and structured finance is not so different from the secular movement towards processed food of the past century. Salami, sausages, spam, hamburgers and similar products make meat inexpensive by combining average and useless parts of an animal into acceptable food.
That this meddling with the ontology would entail unexpected challenges – and in the case of CDOs, financial catastrophe – is perhaps not that surprising. As we now know only too well, processed food opens up the danger for manipulation – for you don’t know what’s inside. The ontology workshop rightly reminded me to keep thinking about the ways in which resourceful traders turn apples into oranges, iron into gold — and mortgages into senior, “super-safe” tranches of a CDO.