Views on the integrity of global capital markets
07 April 2014

Debating Michael Lewis’ “Flash Boys”: High-Frequency Trading Not All Bad

Posted In: Market Structure

Kudos to Michael Lewis for bringing the complicated world of high-frequency trading to the public consciousness with publication of his book Flash Boys: A Wall Street Revolt . There’s no little irony in Lewis succeeding in inspiring a lot of interest in the technology-laden darker corners of the capital markets by focusing on some of the people involved and their back stories. It is an engaging narrative with many of the characteristics of good storytelling.

But like many storytellers, Lewis has to pick and choose among the many threads available in weaving his narrative, and the result is a rather broad-brush treatment of trading in the US capital markets. There’s more emphasis on the nefarious banks and brokers who allegedly step in front of client orders for their own benefit, and less attention to changes in market structure in the last decade that have undeniably contributed to tighter spreads and lower trading costs for investors.

As with any complex topic, it isn’t quite as easy as saying that Lewis got it all right or all wrong. There’s outsize attention to algorithmic trading that pings for customer interest and then races to capitalize on prices in other venues that are ever so slightly out of date. The combination of increasingly fragmented markets and aggressive deployment of technology (both hardware and software) allows for “slow market arbitrage” in which customer intentions signaled by an initial trade can be profitably exploited by intermediaries. While this isn’t market manipulation or “rigging the markets,” it is taking a slice of a trade that could have belonged to the investing customer and putting it into the pocket of the bank or broker. And while the system could be abused to engage in front-running of clients, the fact of its existence speaks more to arbitrage across multiple markets for orderly price formation.

The question for investors is whether there is value to this intermediation (through contributions of liquidity and market depth) and, if so, what that value is and how it compares to the money left on the table by the investor that is snatched up by the intermediary. Professional investors spend considerable time and resources on assessing their trading costs and effectiveness, and while there are challenges associated with trading horizons measured in milliseconds on venues that are by design opaque, the necessary cost-benefit analysis doesn’t seem out of reach. Perhaps the difficulty is in accounting for costs and benefits that go beyond trade execution that still inform the decision of who to trade with, including research and corporate access. Investors and those who advise them are paid to ask the right questions about portfolio companies, discern facts from often complex information, and make profitable decisions accordingly. Hardly the passive sheep waiting to be shorn, investors are actually rather well suited to exploring the consequences of who they choose to do business with to execute trades if they are only willing to expend the effort. They should, and in fact most have a fiduciary responsibility to do so.

Lewis spends a bit less time on some other important issues. The notion that application of technology in the market ecosystem is a patchwork of systems that are poorly understood and controlled by even those who own them is a disturbing indictment of the weak resilience of the markets. Flash crashes have thus far been disconcerting and damaging to investor trust, but the ingredients are there for a more catastrophic failure that would severely test investor confidence to say the least. We’ve recommended a variety of measures (including controls on access, circuit breakers, and/or harmonized trading halts across exchanges, and renewed focus on internal risk management controls over electronic strategies and algorithms, as well as sufficient capital to maintain complex systems) to make the system more durable over time.

There’s also some consideration in the book to the perils of complexity, and the potential to confound customers and regulators deliberately out of greedy intent. For example, the exotic order types that have emerged recently are difficult to relate to quality of execution. And the emergence of the “maker-taker” system of payment for orders adds a level of complexity and dimension of dysfunction that we question. These issues are worthy of further study as a component of current market structure, but we need not indict the entire system to disentangle these effects. We’d be wise to consider the risk of unintended consequences of regulatory responses (with the aftereffects of Reg NMS being an excellent example) and applaud generally the entrepreneurial instincts of market solutions that try to address customer needs. But customers need to make their priorities clear and be willing to put their buying power behind their choices, and we should be wary of the political clout of those market participants who have a stake in the status quo.

Meanwhile, let’s have honest conversations with the investors most likely to be jittery after the publicity blitz that accompanied release of the Lewis book. There are costs to investing, some of which are easy to identify and others that are not. There’s no turning back from the evolution of technology, and high-frequency trading is here to stay. It is a system that can be abused to the detriment of investors, but there is nothing fundamentally devious about it as a way to connect buyers and sellers, and the weight of evidence suggests that it has offered mildly positive benefits to investors. A terrific story shouldn’t tempt us to exclude those real benefits in consideration of the best ways to constrain wrongdoing.

If you liked this post, consider subscribing to Market Integrity Insights.

Photo credit:

About the Author(s)
Bob Dannhauser, CFA

Bob Dannhauser, CFA, is head of global private wealth management at CFA Institute.

20 thoughts on “Debating Michael Lewis’ “Flash Boys”: High-Frequency Trading Not All Bad”

  1. C Lance Durham says:

    If they find out one is buying in market A and then go and buy in market B, then there was already liquidity in market B.

    If there was already liquidity in market B such that they could buy in market B, then they are not adding liquidity in market B. Thus, the primary argument for them fails.

    Regarding other arguments such as falling spreads or depth, they are no comfort. If transactions go up 2X, but only because these guys are inserting themselves into the trade, that is not really improving the depth. If spreads go down by half, but they inserted themselves into the trade, then that is not really improving the spread. …Thus, these arguments fail as well.


  2. Bob Dannhauser, CFA says:

    @C Lance Durham – thanks for your comment. Liquidity attributable to HFT can indeed be illusory if all they are doing is intermediating, but it isn’t clear that that’s the case. Such traders may in fact be real sources of liquidity that narrow spreads, and academic study of this question seems to support this notion (see for example a nice summary with references to this by CFA charterholder Douglas Cumming in last week’s Financial Post). We’re especially interested in studying some of the other factors that affect liquidity in connection with HFT (including order routing methodologies and incentives for order flow.)

  3. I agree with the bulk of your conclusion which is that computerized trading has brought the cost of investing way down for investors. We wrote a blog post on this same topic for our investors and came to a very similar conclusion. I think where we disagree is, although small and rather irrelevant for a long-term investor, HFT programs (as explained by Michael Lewis and Brad Katsuyama) are front running trades. The skim they are taking is small, but the size of the skim doesn’t justify the actions. Computerized trading is great…computerized scalping should be illegal.

    1. James Tharin says:

      I think Fred Smoak is spot on. HFT is a major contributor to the lack of trust in the capital markets. I’m actually shocked and disgusted that CFA Institute isn’t taking the opposite stance on this and in an aggressive way. Was HFT really invented to to provide liquidity? Really?

  4. Fred Smoak says:

    They aren’t adding any liquidity to the markets if all they are doing is stepping in front of already-placed trades. You can look at their trading records and how often they post losses to see they aren’t risking their capital. Not impressed with this article.

  5. Bob,

    Are you saying that HFT was created in order to promote social welfare, liquidity and market stability? Give me a break! If you call the market moving away from me because of front running a good thing then you’re just playing spin doctor. HFT and FLASH crashes cause more mistrust in Wall Street by Main Street and that is never a positive for the industry to which I have dedicated my career. Your article is a blatant insult to intelligent people and probably self-serving.

  6. Marc Denoyer says:

    I agree with the bulk of your conclusion which is that computerized trading has brought the cost of investing way down for investors. Keep posting.

  7. Bob Dannhauser, CFA says:

    @JamesBTharin and @MarcDenoyer, thanks for your comments. Mr. Tharin, you might find this research ( interesting; just one study, but it looks specifically at the issue of prevalence of HFT front-running and finds no evidence of systematic front-running of non-HFT market participants. That isn’t to say it can’t – and doesn’t – happen, but we’d argue that HFT itself isn’t the problem.

  8. James Tharin says:

    Thanks for directing me to the study. After just reading the abstract it became obvious what the author’s point of view is. The abstract had seven or eight conclusions, all of which supported the merits of HFT. If you google the author, it won’t take long to figure out where his loyalties are. Everybody that seems to agree with the points that you raise appears to be making money on HFT. I have no problem with speed, but I do believe that front running is a violation of the code of ethics. I’m surprised that you are taking this position given your position at CFAI. One question that I do have is “how does HFT provide liquidity when a security is purchased thereby reducing supply and then quickly sold at a higher price?” That’s market manipulation 101. Anybody with half a brain knows that if the HFT traders are picking up pennies at enough of a rate to make enough money to justify the money being spent on the technology to do so then the profits are coming out of somebody’s pockets. If the profit isn’t coming from the muppets then where is it coming from?

    1. Bob Dannhauser, CFA says:

      Thanks for your comment. You may be interested in a forthcoming study that we will issue this summer, examining the stability and reliability of liquidity both before and after the fragmentation of markets that has characterized more recent years. We’ll use data on US, Dutch, French, and UK markets to develop some empirical analysis.

      1. James Tharin says:

        I guess HFT is going to be one of those things like politics and religion where we’ll just have to agree to disagree. But your reply just proved my point. You’ve drawn a conclusion from a study that has yet to be released. Is the study biased to support one side of the arguement? At the very least statements like the one you made have the effect of hurting public trust of the finance industry among the public.

        1. Bob Dannhauser, CFA says:

          Indeed, @JamesTharin, HFT does inspire very strong feelings amongst practitioners and investors alike. Kudos to Michael Lewis for making an often esoteric topic so interesting and engaging. Regarding the forthcoming market liquidity study, it will extend the current empirical research around this important topic. Our policy positions to date have been informed by our prior research, including many conversations with CFA Institute members with diverse perspectives. My comments did not anticipate the forthcoming study’s findings; I’m confident that the study design is not biased, but you can judge for yourself when it is released.

          1. Axel Apfelbacher, CFA says:

            @Bob – I’m highly interested in the results of that particular study. Has it been publiushed already? Not being a HFT trader myself, but having read multiple studies, stories etc on the emergence on algo trading et al, I am surprised to hear CFA Institute promoting this particular type of investing. The original idea of electronic marketplaces may have been the tightening of spreads, and this has successfully been accomplished. To use this market development as a round-about argument on the merits of algo trading and HFT for the investing community, despite such obviously illegitimate scams as special order types, dark pools hidden from both market oversight and market transparency, maker-taker fees etc., is surprising. I’d rather expect CFA Institute to promote an agenda that increases transparency in the markets and advocates better behaviour by market participants.

  9. Canada Bruce says:

    The HFT activity is analogous to the early 1900’s when racketeering bookies built telephone lines to the race tracks to shave the odds and scam their customers.

    1. Bob Dannhauser, CFA says:

      Thanks for your comment.

  10. neil crammond says:

    front running clients / retail orders either by arbitrage or similar contract is one of the main reasons for leaving open outcry !
    HFT has just replaced the tall trader front running hand signals . This pratice is easily erased if an exchange wants to restore fair and orderly markets . BUT please its an insult to say it provides liquidity

  11. James Tharin says:

    I would like to echo the comments of Axel. I’m also surprised that CFAI would take that postition on the HFT issue.

  12. Bob Dannhauser says:

    @AxelApfelbacher – thanks for your comment. Our study of the impact of HFT on market liquidity will be published towards the end of August, and will be available on We’ll also be blogging about it here with a link to the study. I appreciate your perspectives, similar to what we’ve heard from several other of your fellow CFA charterholders; but I’d respectfully suggest that the issues around dark pools, special order types, maker-taker regimes, and lots of other features of modern market microstructure aren’t at all obvious and merit further study, and we’ve devoted attention accordingly. We’re not “promoting” algo or HFT but rather are analyzing features of modern equity market structure to focus on benefits, costs, and risks to investors with a view to informing better policy decisions to reform markets. It is a set of polarizing issues to be sure, but I’ve come to appreciate that there is a lot of nuance to the issues at hand.

    1. Axel Apfelbacher, CFA says:

      @Bob Dannhauser Thanks for your additional comments. Looking forward to discussing the results of that study here. I understand that the emerging microstructures of the market may also have merit, so I’m interested in learning more through that study on the pros and cons associated with the ability of some market participants to influence stock prices for their own benefit as opposed to the benefit of the investing community. I guess we’ll have an interesting debate here starting end of August :-).

Leave a Reply

Your email address will not be published. Required fields are marked *

By continuing to use the site, you agree to the use of cookies. more information

The cookie settings on this website are set to "allow cookies" to give you the best browsing experience possible. If you continue to use this website without changing your cookie settings or you click "Accept" below then you are consenting to this.