Book:

Financial Markets in Practice, From Post-Crisis Intermediation to FinTechs.

World Scientific Publisher with Lehalle, C. A.

The financial industry has been disrupted by three major innovations: First, Big data has played a key role in the development of the industry, then machine learning and nowadays artificial intelligence (AI) are both changing the way we are doing business. These innovations were rapidly adopted by the financial sector as it is the first-ranked sector in terms of technology consumption. Compared to other sectors, the amount of data handled is far bigger than in the rest of the industries. In heavy industry, for instance, the data are essentially generated from sensors deployed on the field. These sensors were constructed with a clear view of their purpose and used to asses a certain physical metric that gives insight on the state of production. What really matters is not the amount of data we can store but the quality of each metric that we dispose of, which is solely determined by the sensor’s quality. In contrast, the financial industry is a worldwide interconnected network where the smallest information in a tiny country could storm the entire system and put it in dire straits. Therefore, we need to keep track of a lot of information related to the real economy, asset classes, financial products, and interdependencies, so as to stay alert to potential hidden risks. Needless to mention the growing need for continuous now-casting and short-term forecasting of real economy for investment purposes which increased the complexity of managing non-structured data composed of numerical information, text (from web and news), and images (from satellites). All these implications make the financial industry the number one field of application for new technologies. In this chapter, we will focus on the impact of these new technologies on the financial system and how they changed its landscape. We will discuss the impact of FinTechs on improving customer relationships by personalizing the service to enhance their experience. We will also look further into the role of technology to better connect the financial system to the real economy.

Academic Research:

Stock market liquidity and the trading costs of asset pricing anomalies

Université Paris-Dauphine Research Paper, (3380239) (2019) With Briere, M., Lehalle, C. A., Nefedova, T.

Using a large database of the US institutional investors’ trades, this paper sheds new light on the question of anomalies-based portfolio transaction costs. We find that the real costs paid by large investors to implement the well-identified Fama-French anomalies (size, value, investment, and profitability) and Carhart momentum are significantly lower than documented in the previous studies. We show that the average investor pays an annual transaction cost of 16bps for size, 23bps for value, 31bps for investment and profitability, and 222bps for momentum. The five strategies generate statistically significant net returns after accounting for transaction costs of respectively 4.29%, 1.98%, 4.45%, 2.69%, and 2.86%. When the market impact is taken into account, transaction costs reduce substantially the profitability of the well-known anomalies for large portfolios, however, these anomalies remain profitable for average-size portfolios. The break-even capacities in terms of fund size are $184 billion for size, $38 billion for value, $17 billion for profitability, $14 billion for investment, and $410 million for momentum.

Modelling Transaction Costs When Trades May Be Crowded: A Bayesian Network Using Partially Observable Orders Imbalance

Machine Learning for Asset Management (2020) With Briere, M., Lehalle, C. A., Nefedova, T.

Using a large database of US institutional investors’ trades in the equity market, this paper explores the effect of simultaneous executions on trading costs. We design a Bayesian network modelling the inter-dependencies between investors’ transaction costs, stock characteristics (bid-ask spread, turnover, and volatility), meta-order attributes (side and size of the trade), and market pressure during execution, measured by the net order flow imbalance of investors meta-orders. Unlike standard machine learning algorithms, Bayesian networks are able to account for explicit inter-dependencies between variables. They also prove to be robust to missing values, as they are able to restore their most probable value given the state of the world. Order flow imbalance being only partially observable (on a subset of trades or with a delay), we show how to design a Bayesian network to infer its distribution and how to use this information to estimate transaction costs. Our model provides better predictions than standard (OLS) models. The forecasting error is smaller and decreases with the investors’ order size, as large orders are more informative on the aggregate order flow imbalance (R2 increases out-of-sample from -0.17% to 2.39% for the smallest to the largest decile of order size). Finally, we show that the accuracy of transaction cost forecasts depends heavily on stock volatility, with a coefficient of 0.78.

Liquidity Provision and Market-Making in Different Uncertainty Regimes: Evidence from the COVID-19 Market Crash

Université Paris-Dauphine Research Paper, (3815169) (2021) With Briere, M., Lehalle, C. A.

Kyle (1985) builds a pioneering and influential model, in which an insider observing private information submits an optimal order given the market-maker’s pricing rule, which is assumed a linear function of the aggregated order flow. We propose an extension to Kyle’s model where different types of uncertainty regimes exist and where the market maker estimates market uncertainty and uses it to set her price. The model implies that the elasticity of prices to liquidity demand will increase in high uncertainty regimes. We test the outcome of the model empirically by studying the price formation process during the COVID-19 pandemic crash. A period of agitation with important announcements having a major impact on financial markets, such as the state lockdown and the Fed’s fiscal response. We find that indeed the elasticity of prices to liquidity demand during the COVID-19 period has increased four times.

Broker Note:

Selecting the Best Price-Momentum for Your Investment Horizon

KeplerCheuvreux Dec 2016 Note With Besson, P.