Monday, May 14, 2012

Wall Street Complexity

There is much blame to go around for JPMorgan Chase's two billion dollar loss last week but part of that blame came back to us. In a New York Times web piece, How Moore’s Law Affects Wall St. Trading, Quentin Hardy argues
Faster, cheaper computing makes it possible to create more and better models for calculating cash movements, which can be turned into trading instruments. Areas like leasing, mortgages and project finance have exploded – as has the entire financial derivatives market — thanks to cheap computing...
Soon, it becomes nearly impossible to say what is going on where, and you get events like the 1998 blow-up at Long Term Capital Management, the creation and destruction of the subprime mortgage market in 2008 and perhaps even the “flash crash” in 2010. JPMorgan’s loss seems to be the latest in that series.
I've argued the dangers of reducing computational friction before. But here computational complexity comes in a different way. A derivative is just a function of current and future security prices. But a derivative complex enough can have a behavior that even its creator cannot understand. The Clay Math Institute offers a million dollars to settle "P v NP" but it cost Chase two billion.

9 comments:

  1. Hello

    Thanks for the post. I think you will enjoy this paper Maymin (NYU) (to be published in Algorithmic Finance, found aspects intriguing ("programming the market" - an original physical oracle - at least to me) : http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1773169


    Abstract:
    I prove that if markets are efficient, meaning current prices fully reflect all information available in past prices, then P = NP, meaning every computational problem whose solution can be verified in polynomial time can also be solved in polynomial time. I also prove the converse by showing how we can “program” the market to solve NP-complete problems. Since P probably does not equal NP, markets are probably not efficient. Specifically, markets become increasingly inefficient as the time series lengthens or becomes more frequent. An illustration by way of partitioning the excess returns to momentum strategies based on data availability confirms this prediction

    Have a great day

    Daniel

    ReplyDelete
  2. The failure of LTCM was not really due to computerized trading, nor was the subprime mortgage market. The flash crash was a minor event. I don't think we've hearing anything about JPMorgan's latest (tiny) loss that would suggest computerized trading was involved, much less any relation to PvNP.

    ReplyDelete
  3. I remember this paper showing that pricing certain financial derivatives are NP-complete http://www.cs.princeton.edu/~rongge/derivative.pdf

    ReplyDelete
  4. The NY Times article just sounds sensationalist, and I don't understand how any of it relates to computational complexity.

    The article merely says this: "those hi-tech Wall Street guys do some complex stuff using computers. When they have better computers, they can do even more complex stuff, which they may actually understand even less."

    This is true regardless of the state of TCS, or Moore's Law, because the statement is technically about human cognitive abilities in appreciating the consequences of complex schemes.

    Minor nitpick: Moore's Law is about chip density, not speedup in computation time. Also since not everybody throws away their 1 year old servers for brand new ones every year, a lot of people would disagree with you if you told them they should have 4x more computing power than they had in 2010.

    ReplyDelete
  5. It's nice to connect trading to computational complexity, but I suspect that their recent losses have a simpler explanation, related instead to the "heads I win, tails the taxpayers lose" incentives created by our oversight-free bailouts.

    ReplyDelete
  6. I think computational complexity is relevant to Wall Street the way "bandwidth trading" was to Enron or electronic slot machines are to casinos: helpful, but not crucial, in separating the marks from their money.

    ReplyDelete
  7. Foster Boondoggle8:09 PM, May 15, 2012

    Calvin Trillin explained clearly what happens when some smart guys get hired on Wall Street.

    http://www.nytimes.com/2009/10/14/opinion/14trillin.html

    ReplyDelete
  8. I have to caution against relying on NYT for financial information (except Dealbook, which is usually fine). They make a lot of serious errors and have a clear anti-corporate bias. If you want to learn about what happened at JPM, start with this article and follow the links: http://dealbreaker.com/2012/05/the-tale-of-a-whale-of-a-fail/

    This NYT article looked like scare-mongering to me, with little insight or facts.

    That said, it's been suggested that the JPM loss may have been partly due to a problematic risk model. At the least, if the risk numbers were correct, senior management would have been more concerned with the trade. However, this was NOT a computer-driven trade. It was surely designed and overseen by humans.

    Contrary to popular belief, it doesn't require a computer to lose a ton of money. Large trading losses have occurred throughout history, with or without computers. More typically, it is overconfident humans with not enough oversight who are to blame, as seems to be the case here.

    Although this article wasn't it, I feel like there is something interesting to say about the increasing amount of information we have to deal with. Increased transparency in the markets surely improves efficiency, but it comes at a cost of people processing and analyzing all that information. Moreover, there is, in some sense, greater information inequality, because not all market participants have the computational resources to actually process all the information they have access to. It would be very interesting to see a deeper analysis of the tradeoffs involved with the increased informational and computational resources of recent times.

    ReplyDelete
  9. I understand you try to promote Complexity Theory and its relevance to everyday problems, but it is not always beneficial to it. The problem of creating complex derivatives doesn't seem to have much to do with computational power. The financial sector has become a big casino, there is not that much of difference between gamblers and bankers these days, and the system rewards huge risk takers as long as they win, and obviously they cannot win all the time. More regulations are needed as president says but it will not solve the problem for a long time, bankers will find a way to get around them in a few years.

    ReplyDelete