Here we go again. Another pro-HFT academic research paper based on incomplete and subjective data is making the rounds. HFT proponents are touting the paper titled “High Frequency Trading and Price Discovery” written by Brogaard, Hendershott and Riordan as evidence that HFT provides real benefits to the market. They are pointing to lines in the paper that read:
“Overall HFTs increase the efficiency of prices by trading in the direction of permanent price changes and in the opposite direction of transitory pricing errors”
“Our results provide no direct evidence that HFTs contribute directly to market instability in prices.”
While the proponents are quick to cherry pick their favorite quotes, we decided to dig deeper and take a look at how the authors arrived at their conclusions. We found this paper to be one of the most egregious examples of biased research based on incomplete data that we have ever encountered. While we have come to expect that certain academics will be biased with their conclusions, we also expect that they will at least be analyzing an entire set of data to draw their conclusions. This latest paper falls woefully short of this expectation. Here is what we found:
1- The study only includes a sub-set of HFT firms. Excluded from their study are “those that also act as brokers for customers and engage in proprietary lower- frequency trading strategies, e.g., Goldman Sachs, Morgan Stanley, and other large integrated firms. HFTs who route their orders through these large integrated firms cannot be clearly identified so they are also excluded.”
2- Only trades that were executed on NASDAQ are analyzed. The authors state that “The data used in the study are from 2008-09 for 120 stocks traded on NASDAQ.” Since NASDAQ market share ranged from 20-30% during the sample data set period, this means that 70-80% of trades were excluded from this study.
3- The authors relied on NASDAQ’s subjective determination of which firms were considered HFT. The authors state “Firms are categorized as HFT based on NASDAQ’s knowledge of their customers and analysis of firms’ trading such as how often their net trading in a day crosses zero, their order duration, and their order to trade ratio.”
4- The authors use only a subset of volume to calculate limit order book imbalances. The paper states, “To test the hypothesis that HFTs use order book information to predict short- term subsequent price movements we calculate limit order book imbalances (LOBI) using the NBBO TAQ best bid and best offer size.” According to Eric Hunsader from Nanex, “The NBBO TAQ file only shows the size at the exchange listed as the best bid or offer. It does not show the aggregate sizes.” Therefore, any analysis of short term direction only using this subset of data is seriously flawed.
5- The study focuses on only one asset class. Many HFT strategies are based on multiple asset classes including futures and options. To make conclusions about price efficiency based on only one asset class is irresponsible and misleading.
While we can go further and debate many of the papers pro-HFT conclusions, we feel this is not necessary since their paper is based on flawed data. The authors even admitted “One of the difficulties in empirically studying HFTs is the availability of data identifying HFTs. Markets and regulators are the only sources of these and HFTs and other traders often oppose releasing identifying data.” Clearly, they knew the limitations of their data but decided anyway to inject their biased opinion into the HFT debate.
What makes this paper even more damaging is that it was submitted as a working paper for a European Central Bank fellowship program. This has led some to believe that the ECB supports and approves this paper. This is clearly not the case. The ECB clearly states that “the views expressed are those of the authors and do not necessarily reflect those of the ECB.” Hopefully, policy makers understand this and do not think that the ECB is supporting HFT.