How High Frequency Trading Affect the Market Quality
High frequency trading (HFT) is a recent phenomenon that caught the attention of numerous investors, financial institutions, and regulators. By 2010, HFT has accounted for more than 70% of dollar trading volume in the US capital market (Zhang, 2010), and was also growing rapidly in popularity in other parts of the world. In recent years, many debates have been started over the role played by HFT in the modern-day financial markets. Critics have indicated that computer glitches and complex trading strategies are major factors that drive many small investors out of the market (Wall Street Journal, 2012).
Some studies have shown that HFT is, in fact, beneficial to the markets. Brogaard (2010) analyzed the impact HFT has on US equities market and finds that high frequency traders add to price discovery, provide best bid offer quotes, and do not seem to increase volatility. Research conducted on algorithmic trading by Hendershott and Riordan (2009) also indicates that HFT improves liquidity and price discovery in the stock market. However, academic research on this subject is still limited in numbers. Hence, we have decided to delve further into this subject area by exploring more on the market qualities surrounding the May 6 flash crash, with a focus on DJIA and its component stocks.
Market qualities, including stock prices volatility, trading volume, bid-ask spreads, and depths, are heavily influenced by the flow of new information into the financial markets. High frequency companies are building on their own proprietary models and algorithms by tapping on the low-latency technology (Golub, 2011). This has allowed the high frequency traders to engage in high-speed trading and capture fleeting price differences between trading securities.
This paper studies HFT from different angles. First, we look at the changes in market qualities before, during, and after the May 6 “Flash Crash” in the components stocks and ETF. Following which, we identify the sectors that are more affected by the “Flash Crash” and the individual stocks that were either leading or lagging behind the movements in the various market quality indicators. Lastly, we draw implications on the “Flash Crash” based on our analyses and findings, and how this information would be useful to investors.
The per-minute volatility increased across the market after the “Flash Crash”. By comparing the average volatility of 30 DJIA component stocks in the 5-day period before and after the event (Figure 1 & 2), it is clearly shown that the post-event 5-day average volatility is higher (refer to position of the line “Stock POST”). Table 3 also provides statistical evidence for the rising volatility by the high percentage of