Remember me

Register  |   Lost password?


StatAlgo Logo

Flash Boys and Investors

Wed, 18 Jun 2014 11:56:06 GMT

How does technology impact the electronic exchange of assets, and how does this advancement impact the individual investor?

I spend a few minutes discussing high-frequency trading, then consider it's impact on the typical individual investor, and finally imagine where electronic trading is moving in the long run (hint: towards greater efficiency).

In general, my hope is that the public should become more informed about market efficiency and the benefits of diversified portfolios, rather than focusing on market structure issues. On the other hand, market structure has a major impact on institutional investors. Large institutional investors need to be very aware of how markets are structured, including how intermediaries play a role, and their executions should always be considered as part of their overall performance. As such, while the general consensus is that transaction costs are lower now than historically, I could see real value in more transparency around the implicit transaction costs incurred by very large orders.


  1. Flash Boys (in brief)
  2. The Individual Investor and the HFT
  3. Efficient Markets
  4. Technology and Exchanges

Flash Boys (in brief)

I have thus far resisted writing anything about Michael Lewis's book "Flash Boys", in part because I typically don't want to waste time on timely issues (instead focusing on topics that are of long-term interest), although it has been a real exercise in self-constraint because there have been few books published in my lifetime that have been as troubling to me (Piketty's recent "Capital in the Twenty-First Century" is another recent example of flawed economic thinking).

I heard well before "Flash Boys" was published that Lewis was working on a book about high-frequency trading (HFT), and I was very excited; I loved all his other books. Then, a month or so before it became available, I saw something that made my jaw drop: the book was about IEX. This surprised me not because IEX is bad, but it is tiny and insignificant. There were many other exciting things to discuss on the topic of HFT: why would Lewis focus on IEX?

A few general observations about the book:

  • It is not an objective account of reality. Lewis began writing the book having already decided in his mind that the markets are rigged. There is very little evidence that he made an effort to research for the book, short of talking to the IEX people.
  • The book itself and subsequent promotion has been an exercise in sensationalism. When Lewis, Brad Katsuyama, and Bill O'Brien (President of BATS) squared off on CNBC, O'Brien started the conversation by declaring "shame on both of you for falsely accusing literally thousands of people and possibly scaring millions of investors in an effort to promote a business model." There is an immense amount of truth in this sentiment. I am sure that both Lewis and Katsuyama believe what they're saying to be true and they are looking to improve the markets, but there is no doubt that using the "rigged" word and emphasizing how the little guy is getting hurt has been unconscionable.
  • For the most part, the "news" around Flash Boys is that Michael Lewis wrote a book. In other words, he has written an engaging story, although in this case, he has not uncovered anything that has not already been discussed at length by financial journalists such as Scott Patterson.
  • One of my favorite parts of the book, which really highlights how little the author and the IEX protagonists understand market structure, was towards the end when IEX is reviewing the first trades that were executed in their dark pool. They were surprised to see very tiny orders coming through, which they concluded might be part of a conspiracy to make them look bad by competitors. This hypothesis, of course, completely ignores the fact that algorithmic trading uses small trade units in order to reduce market impact following Almgren and Chriss "Optimal Execution of Portfolio Transactions" and all common sense which dictates that you want to hide your orders.

All that being said, I am not going to spend time on a lengthy review of the book because I expect it to ultimately gather dust as a missed opportunity (and spawned an industry of confused journalism, giving a major platform to the conspiracy theorists, such as this recent example from the Guardian), and many others have already gone through this exercise. Some highlights:

I don't know if IEX will succeed. But the fantastic thing about the US Equity market is that IEX exists. I hope that many other innovators try to launch exchanges to solve any market structure problems.

The Individual Investor and the HFT

What is the impact of HFT on the average investor?

Let us start answering this question by saying that the average investor cannot and should not try to compete directly with HFT. The average investor could never compete with the floor trader before there was electronic trading, and will never be able to compete with sophisticated investors in the long run. Even if you remove things like co-location, that won't remove the competition to be faster (people will compete for real estate). To suggest that a grandmother in Kansas should be engaging in intraday trading from E-Trade alone is nothing short of unethical, but then to further suppose that the fool who chooses to engage in this kind of trading should be able to compete with large institutions running major infrastructure and employing teams of quants and developers is akin to living permanently in the world of The Ingenious Gentleman Don Quixote.

HFT's play two major roles in electronic markets:

  • Liquidity provision
  • Short-term price discovery

There has never really been a role of the average investor in providing these two functions: these are services that allow markets to operate efficiently.

So, assuming that we hold that the average investor is an investor (i.e. is holding assets for long periods of time, making an investment), then what is the impact of HFT on the average investor? There are several:

  1. The average investor does not compete directly with HFT. When they enter a trade through their broker, the overwhelming majority of these orders are internalized.
  2. Bid/ask spreads are generally tighter now than they ever have been, and that is a cost reduction for small investors (large investors are more complicated).
  3. High frequency traders make markets more efficient (nicely summarized by Matt Levine, who is probably the most informed and entertaining writer on this subject), which is a good thing for small investors. When prices do not reflect all available information rapidly, then trades can be made erroneously.
  4. Many investors use investment vehicles such as mutual funds, hedge funds, and ETF's to gain exposures, or hold investments in the form of pension funds and endowments. The managers of these instruments are themselves institutional investors, and they do interact directly with HFT to source liquidity.

So the primary impact of HFT (or any intermediary/liquidity provider) on the small investor is through exposure to institutional investments. These institutions have a responsibility to fully understand their executions. They are often required by law to seek best execution, which can have a broad mandate: "Under MiFID Article 21, a firm must take all reasonable steps to obtain the best possible result, taking into account price, costs, speed, likelihood of execution and settlement, size, nature or any other consideration relevant to the execution of the order" (from FSA, particularly true in Europe under MiFID and in some Asian countries). Dealers in the US are also required to seek best execution, and Regulation NMS (2007) in the US also sought to strictly define a best price.

There are clearly issues that can be addressed in the current market structure, including questions around overall complexity (including order types), practices in dark pools, payment for order flow, and potential issues with broker routing (e.g. to capture maker/taker fees). I hope that bad legislation doesn't result from all the recent attention, and that the SEC and CFTC can continue to take a thoughtful approach to regulatory reform.

Lastly, I should point out that there is a vast amount of academic literature on the topic of high-frequency trading. The SEC recently started to summarize this literature. Another recent literature review can be found in "What Do We Know About High-Frequency Trading" by Charles Jones.

Efficient Markets

Taking a step back, what is this whole discussion all about? Why has there been such a huge reaction to Michael Lewis's book and subsequent flood of journalistic confusion? The cynic would say that this is simply an attempt at what journalists (and anyone competing for attention) love to do: create a panic in order to generate sales. But regardless of the underlying cause, the fact is that this emphasis by the media on HFT is completely misplaced.

Investing is hard. This should always be a starting point when considering whether to make a trade.

The ideas of market efficiency are most closely related to Eugene Fama. Paul Samuelson famously described the markets as micro efficient and macro inefficient, and this viewpoint has subsequently been supported by Robert Shiller. Whether you believe that markets are efficient in the EMH sense or not, it is very difficult to ignore the mountain of evidence that supports John Bogle's argument that passive investing beats average active investing over the long-term, and any "outperformance" is subsumed by the manager and fees. Samuelson made the same argument, that there may be skilled managers who earn alpha, but that they would likely not offers their services to the public.

In my view, the recent emphasis on HFT and the many arguments debating the minutia of the Efficient Markets Hypothesis are all part of the ongoing drag on the average investor. We should be emphasizing the benefits of diversified index portfolios that benefit from risk premia, including global equities and bonds, as well as most exotic factors such as value, momentum, and carry.

For the average, thoughtful individual investor out there who might be reading this post, please stop reading "Flash Boys" and start reading books that will actually help you. I would strongly encourage reading these instead:

Then, if you still have time left over, some of my other favorite popular books:

Technology and Exchanges

So given everything, we should step back and ask: how should markets function? I see the role for markets as essentially being the same now as they always have been:

  1. Markets efficiently aggregate information. The price of an asset should accurately reflect its value. At the most basic level, we can think of this in terms of supply and demand: any supply-side or demand-side manipulation can interfere with this critical role, causing prices to become inaccurate conveyors of information, which is a serious issue for the functioning of society.
  2. Market structure should support maximum liquidity. OTC markets generally favor large investors who are well-connected (see Darrell Duffie's excellent "Dark Markets: Asset Pricing and Information Transmission in Over-the-Counter Markets" for a recent review of this literature). The current US electronic market structure (of FIFO order books) arguably favors small investors (who do not require more liquidity than is available at the top-level of the order book). Some large institutions try to minimize their market impact by trading off the exchanges (e.g. in Dark Pools), but these venues may not serving the large institutions for executing block trades (as can be seen by the average trade size on these venues).

At the recent CFTC Technology Advisory Committee (TAC) Meeting, Chuck Vice from the ICE exchange made the following remark about HFT: "You can't put the toothpaste back in the tube." This is true: we should not be trying to roll back the technological improvements that have been made over the last 20 years, but should build on top of them.

Put all existing the regulatory issues aside and consider the future of exchange.

  1. Markets will become increasingly efficient. This means that prices should increasingly reflect their fair value at all times. Instruments which trade infrequently have inherently more uncertainty around their value. As information is incorporated more rapidly into prices through technology, this results in improved efficiency.
  2. Immediacy continues to have a value, which means that there will likely always be a role for intermediaries. Liquidity is a function of being able to transact at a price and size (i.e. with minimal price impact), but is also measured over time.
  3. Exchange will probably always be organized along a spectrum of liquidity. There is a difference between buying a house or trading a share of Apple.
  4. Risk will be more dispersed. The movement of swaps onto SEF's (and ultimately into central limit order books for standardized swaps) and through central clearing is a clear indication of the trajectory that all electronic assets will move: shared risk .
  5. Credit plays an expanding role. Derivatives that trade on margin have grown substantially over the last century, especially with the growth of swaps from 1981 until now.
  6. Costs will be lower for everyone. This will happen in a few different ways. Bid/ask spreads and commissions will decrease further, as will other fees. This will happen because of competition, the increasing role of automation, and additional sources of liquidity (including additional sources for hedging, which provide a very important form of liquidity transfer).

I, for one, am very optimistic about the continued improvements in financial markets (Robert Shiller may be the clearest proponent of this viewpoint, for instance, in "Finance and the Good Society"). There has been a remarkable amount of progress made in risk sharing over the past century, and much of this innovation is directly related to the capital ideas which have been codified in academia. I hope that we can continue to see progress in financial innovation, increased transparency and openness, more literacy and prudence for all investors in the future. Technology will play a critical role in the future of capital markets.

, , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,