The recent SEC round-table on ‘Promoting Stability in Today’s Markets’ was held on October 2nd. It had all the earmarks of being yet another ‘government job’:
A) Round up the usual suspects
B) Bring out the big commissioners
C) Everyone reads from prepared statements
Well, to my surprise – the SEC actually did a fine job by including a fine woman by the name of Dr. Nancy Leveson of MIT. At first glance, I was starting to roll my eyes & expecting the usual ‘I’m from academia and here is how things should work’ speech, she delivered a performance that rivaled Jeff Goldblum in Jurassic Park – setting the stage for the remaining 5 1/2 hours of testimony and discussion of how to make the markets better and more stable.
All laudable goals, but given that most of the participants (exchanges, high frequency traders & market makers) are either conflicted, incentivised or actively profited from the most recent Knight Trading Snafu, ultimately doomed to failure by the denial / hubris and conflicts of the participants.
But back to my story – she proceeded to review her long & impressive career / resume in the world of computer science – expounding the lessons learned from working with everything from NASA, the aviation industry, nuclear power, medicine and software – and proceeded to impress even this jaded veteran of Wall St. with her insight and powerful statements and indictments on the design, development and testing of software.
I guess the reason this resonated with me so much has to do with the general, rather than the specific nature of her comments, all of which have been formed from working in every industry except for finance. Sitting back and thinking about what she said provided me with a pretty nice framework for forming an opinion of not only what should be done, but also the basis for my prediction as to what I think will happen.
Her first point – ‘All Software has bugs – there is no such thing as ‘perfect software’
Her second point – Keep things simple – humans have a tendency to make complex as opposed to simple programs, thus increasing the probability of unexpected behavior.
Her final point – Keeping public trust is paramount – the best way to create a ‘risk culture’ is to remind those involved of the ramifications resulting from errors – very effective, especially when it could mean the death of an entire industry, company or most importantly, their jobs.
She brought up excellent examples such as NASA, the Aviation Industry and Nuclear Power, all of which try and limit the amount of software dependencies on ‘mission critical’ components, and the very public reaction, fallout and analysis which results if something goes wrong.
Taking these points to heart – I have decided that in terms of electronic markets we now have the following situation
I – The Equity Marketplace is far too complex for its own good. The over 2000 order types alone, added to the ever growing number of exchanges, dark pools, crossing networks and the like have made the entire ecosystem far too complicated. Add to this the requirements of Regulation NMS, which mandates the markets to be electronically linked in the name of ‘customer protection’, has lead us to today’s market where it takes an extremely complex algorithm (s) to simply place and properly route an order.
II – Each exchange, trading firm or hedge fund has entirely too much faith in their own approach, platform, development and testing abilities for their own, much less the markets, good health. The Industry Hubris is such that all the leading players continue to believe in their own superiority, reinforced by profits, and believe that they are uniquely placed to be able and harness the most complex and daunting problems. Controls and such are for the lesser, weaker & more pedestrian players. As such, I am more certain than ever that we are due for yet another major ‘incident’ that will roil our markets, further reducing the public trust in our markets.
Given that the balance of the 5 1/2 hours of testimony eventually degraded into a discussion over ‘ISO 3000 Standards, Full end-to-end testing, and a length discussion over the need, design and operation of a ‘kill switch’ for markets – I left feeling even more pessimistic about the industry’s ability to heal itself.
The smartest guy I know, an industry legend if there ever was one, had a favorite saying: ‘People are Stupid’ – Which he seemed to use in response to a variety of questions about this or that. I am beginning to see what he meant – the more complex markets get, the more likely it is for it to all end in tears….