The Techno-social link at the heart of the crash…

March 4, 2007

We’ve been hearing a lot about the possibility that the latest crash was, at least partially, computer-led. One report that I saw, however, draws a slightly more complex picture, that is worth our interest as sociologists of financial markets. This taken from a Newsweek story:

“On Tuesday at around 2 p.m., the market’s extraordinarily heavy trading volume caused a delay in our trading system,” explains a spokesperson for Dow Jones & Co., the media company that manages the index of 30 blue-chip stocks. For 70 minutes, a slow data feed to the Dow Jones industrial average (DJIA) calculator meant that traders were working off a slightly outdated set of numbers. When the error was caught, the system was switched to a backup server that immediately readjusted the figures—sending numbers across the board into a free-fall plunge of 178 points in a single minute.

“When they put the backup system in, [the market] went kerplunk,” Alfred E. Goldman, chief market strategist at A.G. Edwards & Sons Inc., tells NEWSWEEK. “That just scared the heck out of a lot of folks. […]” [T]he eye-popping adjustment helped lead to the market’s biggest single-day drop in almost four years, a total decline of 416 points.

Of course, this is only part of the story, but we can see that at least in this case it was not ‘the computers’ that are to blame in the crash, but the dependence of market participants on the efficiency of the computerised order-routing system. In other words, the traders got used to an immediate connection between price-changing information and the prices they got from the system. So, when the system ‘jumped’ and prices dropped suddenly, the reaction by many was to sell.

9 Responses to “The Techno-social link at the heart of the crash…”

  1. cpptrader Says:

    I am surprised the mass media isn’t squarely placing the blame on computers – in today’s world of “the end is near” news-reporting, I would have suspected it. But, I guess to their credit, they are spreading it around. Still, you bring up a point that I think some people are overlooking – the computers did not cause the crash. The orders were still placed. Data feeds may have been slowed, execution may have been off, but sell orders (or a lack of buy orders) are required to move the market lower. When it boils down to its simplest nature, what made the market drop was traders selling.

  2. Peter Says:

    I think there’s something about electronic markets as assumedly WYSIWYG markets – Daniel and David’s finding about lag notwithstanding, there’s an assumption that the screens are the market. In open outcry, contrast, only the lazy or naive looked at boards and reported those prices as opposed to asking in the pit ‘what’s here?’

    Does this mean that computers caused or didn’t cause the crash? Too simple either way – the current socio-technical system makes it likely that system-wide errors will be linked to the technology, wouldn’t you think? But cause, (agreeing with the way cpptrader says ‘traders selling’) probably not. But traders can sell for reasons having to do with computers, so it’s not quite so straightforward.

    Incidentally, did you see MacKenzie’s Oxford lectures last week? How did it go?

  3. Yuval Says:

    Yes, spot on: this is the point I’m making. Computer on their own do not cause crashes for the simple reason that computer and human traders are never ‘on their own’ – there are always practice-based connections between humans and machines in the markets (indeed, like your example about the role of the boards in open outcry trading).

    Now, just to sharpen the point: I do not think that market participants believe that what the computers show is ‘the market’. Nevertheless, more and more cases tend to show that participants believe (rightly or wrongly) that computerised systems present information they do not have, or aspects of the information of which they are not aware. So, when there is a big and sudden discrepancy between what the traders get from the trading floor and what they see on the screen as prices, they tend to be risk averse and limit their exposure to the market. This general reaction is translated many times to selling, which may exacerbate price drops and the result may be a crash.
    In such a scenario, have the computers caused the crash? Of course not. Have the traders caused the crash? No; they were merely reacting to conflicting messages that contributed to a growing uncertainty about ‘where the market was’, and what they did was simply to follow the age old saying: “when in doubt, kick the ball out”.

    Unfortunately, I could not attend Donald’s Clarendon lectures. I heard from colleagues that it went very well.

  4. tinaguenther Says:

    Unfortunately, I was not glued to the screens last week when the black week in international trading began and I the daily press I received contributed little to my enlightenment on this matter, but is it possible

    (1) that terms such as ‘technical’, ‘computer driven’, or as my newspaper stated ‘correction’ refer to the fact that people can be calmed down to a certain degree and one can sort of ‘play it down’ using a language which refers to the ‘technical’ as opposed to economic developments and social facts

    (2) since Yuval stressed that the ‘social’ behind the ‘technical’ needs to be revealed that sociologists should get better familiar with programming code since, here, important linkeages might be found?

  5. Yuval Says:

    I agree whole-heartedly, Tina. Sociologist who aim to analyse computerised environments should have some familiarity with the concepts, if not the nitty-gritty, of computers’ operation.
    Of course, there are many examples for sociological studies about computers. I have a bias here, but see for example this book

  6. Yuval Says:

    The book I am refering to is this:
    Mechanizing Proof: Computing, Risk, and Trust
    By Donald A. MacKenzie

  7. tinaguenther Says:

    ah, okay, thanks for the pointer, Yuval!

  8. […] insight with regard to financial crashes: if we want to improve our understanding about the techno-social connections that occasionally play part in market crashes, it can be useful to look for social intents of various kinds (commercial intents, in the case of […]

  9. Smealaype Says:

    продам Форд-Фокус 2008 года за 200 тр. торг возможет. срочно!!!
    +7 960 200 9209

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: