Knight Capital’s near failure was a stark reminder that a Wall Street firm can go bust in a matter of minutes.
Investors and regulators are once again up in arms, wondering how a firm could have such poor risk controls as to enable a software ‘glitch’ to cause havoc on the markets, particularly so soon after the debacles that hit BATS and Nasdaq as well as the 2010 Flash Crash.
But a New York Times article points out that the financial industry actually has a pretty good track record as far as failure goes.
Wall Street failures may be more high-profile than others and rile tempers more than the demise of Eastman Kodak or the bankruptcy of American Airlines, but the numbers speak for themselves. Last year, 86 publicly traded companies went bankrupt, according to bankruptcydata.com. Only 4.7 percent of these were in banking or finance, the NYT reports.
Further, Knight Capital’s near failure rattled markets and investors’ minds. But its $440 million software glitch was actually, wait for it, not that bad. (And not just compared to firms that have suffered more spectacular single-day losses due to glitches).
According to the New York Times article, Knight’s debacle was actually a success story for Wall Street as it showed that the industry could minimize and absorb its losses in days.
Regulators may be debating whether new rules are necessary to prevent another incident like the one that befell Knight Capital, but there are no rules that can completely, or should, eliminate risk, as it is an integral part of any successful financial system, the New York Times points out.
For others however, Knight’s near bankruptcy was no success story, despite the fact that the firm’s $440 million loss was absorbed and it is now running normally again.
“A success is something you are proud of,” says David Zinberg, principal, financial services & insurance – capital markets, Infosys, who describes the Knight fiasco as an “unqualified, and possibly preventable, failure.”
“The only upside here is the corrective action that it will inspire within the industry,” adds Zinberg, who previously worked in algorithmic trading product management at Lehman Brothers and Barclays Capital.
While we do not have enough information yet to make a fair judgment about what allowed the bug at Knight to get into production unnoticed, the event revealed what was likely a huge failure of risk management, Zinberg points out.
“CIOs across the industry are no doubt scrambling to find dangerous bugs lurking in their code before their own firms become headline news,” he says.
The possibility of such failure will always underlie complex electronic trading systems, Zinberg notes.
But testing, while imperfect, offers the best possibility of avoiding the kind of fiasco that had markets shaking on August 1st.
The Knight debacle can perhaps be best described as a company’s failure at its core competencies, which in this case is around technology-driven market operations.
As Sang Lee, founder of Aite Group, said when I interviewed him after news emerged of the Knight software failure: “If you’re a high-frequency trading firm and you created a scenario where your strategy is a software update and you don’t do enough testing, you’ll go out of business.”