Why the Flash Crash Really Matters
nautil.us
nautil.us
I get that, and I get such systems can be subject to anomalous events (early fly by wire systems for dynamically unstable aircraft was a good test case to study). But such systems can also be 'calmed' by damping. Isn't the SEC in charge of adding such damping to our financial trading systems?
http://cdn.batstrading.com/resources/membership/BATS_US_Equi...
I think it would be great if the SEC believed that that was their mandate. Unfortunately, I think that is predicated on much more sophisticated and nuanced understand of the dynamics of the markets than regulators typically have.
The claim in the recent CFTC's Complaint that alleged market manipulator Navinder Sarao directly contributed to the crash is only one example of this. If one guy can cause a Flash Crash, there is a bigger problem with the structure of the markets.
http://www.sec.gov/News/Speech/Detail/Speech/1370541505819
I think it shows that the regulators (at least some of them) do want to have a better understanding of the dynamics of the US equity market, and that they are trying to build the tooling that will give them the right sorts of insights. Berman's suggestion that policy changes ought to be driven by data are, I think, something that most technologists would agree with. For whatever it's worth, Berman also has a physics Ph.D. from Princeton. So if your insinuation is just that the regulators are more lawyer than scientist I don't think that's completely true.
edit: Also wanted to add that there has been a lot written about the flash crash and I think this article is definitely one of the better ones.
I think there are two challenges to unpack. One, though I wasn't insinuating it, I could have been. I do believe that regulators are more lawyers than physicists. Berman is the exception rather than the rule.
Two, Berman, in particular, makes a fundamental error that I think is very easy to make. There's a difference between "complex" in the sense that something has a lot of parts, and interactively complex in the sense that parts of a system are fundamentally unknowable and it can experience wild and unexpected dynamics. I think Berman doesn't distinguish between those two types of systems (repeated analogies to cell phones give some indication of his thinking), and more generally, regulators don't understand the aggregate cost of complexity.
In my view, things like Midas are orthogonal to some deeper issues facing the markets. Regulators have created a quasi-competitive market that breeds this sort of interactive complexity. Then, when something goes wrong, they rely on punishment and enforcement actions [1] to target individual firms that have made "mistakes." This not only does not address root causes, it creates a culture of silence around technology risk issues within firms and across the industry. I've written more about this here: http://harvardkennedyschoolreview.com/preventing-crashes-les...
[1] See, e.g., http://www.sec.gov/litigation/admin/2013/34-70694.pdf and http://www.sec.gov/litigation/admin/2013/34-69655.pdf
In short, the damping mechanism in markets is that anyone who believes the price is wrong can come in, buy/sell/short and push markets back to the right place. Which is exactly why the flash crash fixed itself. I know a human trader who missed the entire flash crash due to an ill timed bathroom break.
[1] US markets will retroactively break "clearly erroneous trades".
But you are right that regulators (not sure if breaking trades is an exchange or SEC thing?) do reduce this natural damping factor.
The article doesn't seem to establish that the flash crash mattered at all in the sense of having any impact, positive or negative, on anyone not involved in the particular trades.
It's important not to understate the cost of failure in these markets. A firm might go bankrupt and have to lay off real people if caught on the wrong side of such an event.
I also think that there is a direct connection between events like the Flash Crash and things like Nasdaq's mishandling of the Facebook IPO, which, again, had real costs in terms of time and money. Both emerge, I would argue, from a similar flavor of complexity.
I'm struggling for an analogy to show that it matters. Maybe it's a little like Target's website going down. It's not Quality, in a Zen and the Art of Motorcycle Maintenance way, even if it's only down for a short time, and the consequences were "only lost orders." Compelling?
I think this was the point I was missing. So the goal of corrective regulation shouldn't be just to damp wild "flash crash" style transients per se, but to deal with unique anomalous events that commonly arise from the tightly coupled complexity of the market but have individually unpredictable causes and consequences?
Why?
If you mean the fund holding the securities, yes, they might go bankrupt. Risk of having that kind of "job".
If you mean the company whose shares are falling, I don't see how that affects them unless they just happened to be issuing their own stock that day (I don't think any of the companies in question were).
It is a particularly HN bias to think of the market as a place to go fund a company, but that is largely a by product for most people. For most people (including everyone with an index fund/etf) its a place for managing risk. That doesn't even take into account, that a big component of this story was about S&P futures. Which are all about risk.
So it is uninteresting (even if it were true, which i doubt) that microsecond time scales make the markets less about investing in companies, they already aren't about that.
At the micro level, the problem is uncertainty. The "true" or "fair" price of a security isn't something you can just calculate: it's uncertain and subject to change when new information is introduced into the system. So theoretically, when the price gets very far from the fair value, market makers should become more and more eager to trade, since they'll make a large profit when the price comes back. But in practice, they never know whether the person on the other side of the trade is just panic-selling and moving the price farther from equilibrium, or if they know something the market maker doesn't and are moving the price closer to a new equilibrium. As the price gets farther from its previous stable value, the probability that the counterparty is informed grows, and eventually no market maker is willing to trade and the market dissolves.
This failure mode is related to the "Market for Lemons" problem: asymmetric information causing a market breakdown (http://en.wikipedia.org/wiki/The_Market_for_Lemons).
God no. How many stocks even pay dividends any more? The ones that don't are, strictly, "worthless", so their global optimum price should be zero (I'm simplifying; if they come with voting rights then there's a non-zero value to that, but you get the point). It's a bunch of baseball cards at a swap meet.
The disconnect here is that there are agent effects. People make money selling and buying them, so they are incented to push the idea that selling and buying them is a good thing.