Moving the market: A look at prices and impacts from trading

For large investors, it is not enough to know the current price of an asset to determine the best plan for buying or selling — the likely effect of the order itself must be taken into consideration. Although this concept may seem obvious and is certainly evident in actual trading data, the fact that price impact can increase trading costs is an important factor in how large trades are accomplished — particularly for crowded strategies. Since the 1990s, monitoring and controlling price impact has become an active area of research in quantitative finance, encompassing the size and timing of trades, cross-impacts and market liquidity. In a keynote talk at the BBQ (Bloomberg Quant) Seminar, Jean-Philippe Bouchaud, Chairman and Chief Scientist at CFM and professor at College de France, discussed some of the theory, practice and empirical findings that have emerged around price impact.

At the foundation of this work is the acknowledgment that prices, in general, exhibit non-Gaussian behavior, even though much of finance in based on Gaussian models. With that in mind, one of the central questions in constructing any large trade will be the slippage impact of that trading — how much will it cost and is the market liquid enough to accommodate it immediately? For many substantial orders, it is necessary to “slice and dice” the order — arranging for execution over several days rather than all at once. However, during this time, the price may move — in part due to natural fluctuations in the market and in part due to the impact of the trade itself. Whether the price moves a little or a lot depends on the quantity and the timing. For sophisticated traders working with good signals, it may be possible to predict and mitigate the move to some extent, but, clearly, execution risk can be painful if the price moves sharply against the trader.

During the 1980s, a number of authors and market participants began producing research on this phenomenon, including Almgren, BARRA, CFM and AQR. An empirical investigation of the price impact has shown that such impact appears to be governed by the square root law, with nonlinear impact and decay over time. What is most striking about the square root law is that is seems to be universal; it does not depend on whether the trades are market or limit orders, and it cuts across asset types — U.S. equities, international stocks and even Bitcoin appear to be subject to the same law. However, when assessing the impact, the analysis does depend on the assumption of a reasonable trading regime and that the trade is not too big or executed too quickly. “If you overwhelm the market,” says Dr. Bouchaud, “don’t expect the regime to hold.”

As a physicist, he points out, it is fascinating to observe the remarkable stability of results, across order types, tick sizes and treatments of data. The findings suggest that macro-liquidity matters, but micro-liquidity may not, and he and his firm, among others, have worked to develop models to explain the square root law. Taking a look at a global picture, they have developed a dynamic theory of latent liquidity — the Local Linear Order Book (LLOB) model (Donier, Bonart, Mastromatteo, Bouchaud) that offers a general reduced form model for price impact.

Other areas of research include co-impact, which happens when others trade simultaneously, and cross-impact, which pertains to how trading one stock impacts other untraded stocks. Building models inspired by these ideas will help develop an understanding of volatility and liquidity and how market stability and fragility evolve over time.

Given the varied array of market activities, from exchanges to auctions to dark pools, it is ever more important to estimate and understand the endogenous liquidity fluctuations in those markets. Models like LLOB offer useful insights and can be included in optimization algorithms to assist in navigating this terrain.

During the Q&A session, Bruno Dupire noted the long-standing question of correlation and causality, which engendered a good discussion; he also mentioned Walter Schachermayer’s work on dimensional analysis (which Dr. Schachermayer presented at the BBQ last year). Other participants asked aboutL the Volume-Weighted Open Price (VWOP) in relation to the research findings; whether ETFs and indexes exhibit the same impact effect; and how machine learning can fit into the analytical picture.

Lightning Talks

Following a short Q&A session, Bruno Dupire, the host of the event, kicked off a series of “lightning talks,” 5-minute presentations where industry experts, researchers and academics present a wide range of subjects to stimulate fresh thinking and interaction between various disciplines. Each talk examines a way that the industry is evolving and serves as an essential exploratory aspect of the Bloomberg Quant Seminar series.

In this session, Arun Verma of Bloomberg L.P. addressed the future of forecasting and the challenge of sourcing the truth; Bachir Taouli of Mount Sinai Hospital explained the value of radiology in diagnosing COVID-19; and Roza Galeeva of NYU Tandon examined the return to Bachelier, as the CME has announced a switch to that model in order to accommodate negative future options and strike quotes.

In addition, Haochen Liu of ETF Trace offered insights on identifying arbitrage in the ETF market using overbought and oversold signals; Benjamin Voyer of ESCP Business School and the London School of Economics provided a behavioral perspective on COVID-19 through the lens of consumer psychology; and Bryan Liang of Bloomberg L.P. presented work on options and elections, showing how events that bring uncertainty can affect implied volatility, risk reversals and the shape of volatility skew.

About the Bloomberg Quant seminar series

The Bloomberg Quant (BBQ) seminar series is held each month and covers a wide range of topics in quantitative finance. The BBQ seminar is offered in a virtual format and scheduled to accommodate participants from across EMEA and the Americas. Each session is chaired by Bruno Dupire, head of Quantitative Research at Bloomberg L.P., and features a keynote speaker presenting on his or her current research. This presentation is followed by several “lightning talks” of 5 minutes each in quick succession. This format gives the audience the opportunity to be exposed to a wider variety of topics. Sign up to receive invitations to future events in this series.