It's not a secret that many pension fund, mutual fund and hedge fund managers are concerned about high-frequency traders (HFTs). While their concerns are many, perhaps the biggest uncertainty involves the actual extent of HFT participation in the markets, their identities and their intent.
While some claim that HFTs comprise 60-70% of all market participants, such numbers are seldom reached in reality. Scientific examinations find that HFTs still account for as little as 25% of all market activity in such frequently traded instruments as the S&P 500 E-mini futures (see Kirilenko et al., 2011). As Figure 1 shows, even in the very liquid S&P 500 ETF (NYSE: SPY), high-frequency traders on average account for just 20% of daily trades (Aldridge, Irene, U.S. Patent pending since 2012).
As shown in Figure 1, the average level of HFT participation in SPY remains remarkably stable: most days, 15-17% of all trades in SPY can be attributed to HFTs. At the same time, evidence of resource allocation to HFT suggests that the industry is growing at a rapid pace. A natural explanation reconciling the two observations exists: HFT has low barriers to entry, yet it can be extremely complex, requiring years of toiling with data to proficiently develop and deploy trading models.
Interestingly, HFT activity reaches its peaks with specific regularity. As Figure 1 shows, the intensity of HFT participation increases dramatically on the first day in the last months of every year, generating as much as 40% of all trading volume in the SPY.
What can cause such spikes in HFT activity? A potential answer is ironic, particularly from the perspective of large non-HFT hedge funds and pension fund managers. Many such managers seek to lock in their end-of-year performance figures by liquidating their portfolio holdings at the end of each year. Often, they do it only once a month, on the first or the last trading day of the month. The position reallocation often involves a large trade (block trade) in the given financial security. The HFT then enters the trading balance via the managers' execution brokers, who deploy HFT strategies to carefully disperse the large block trades in the daily market stream. While it is the executing brokers who actually deploy the HFT strategies, they do so on capital and with the explicit permission of the fund managers. As a result, it is the fund managers who bear the brunt of the fiduciary responsibility for such HFT activity, and the largest HFTs actually turn out to be the investment managers themselves.
What can fund managers do to alleviate the issue? The field of HFT has been evolving rapidly so staying on top of new developments is important. Learning about what is high-frequency trading and how it can be deployed is definitely the first step in the process. My new book, High-Frequency Trading: A Practical Guide to Algorithmic Strategies and Trading Systems, 2nd edition, published by Wiley & Sons (ISBN: 978-1118343500) discusses a variety of strategies used in execution, as well as provides a comprehensive overview of the industry. I also offer hands on training courses in the HFT strategies, available via http://www.HFTTraining.com. After all, wouldn't you like to know where your money is and how it is deployed?