Market Momentum

Theory and Practice

Stephen Satchell, Andrew Grant

• Publisher: Wiley
• Year: 2020
• Print ISBN: 9781119599326
• eBook ISBN: 9781119599371
• Edition: 1
• Topic: Business
• Subtopic: Financial Engineering
• Index: Business

A one-of-a-kind reference guide covering the behavioral and statistical explanations for market momentum and the implementation of momentum trading strategies

Market Momentum: Theory and Practice is a thorough, how-to reference guide for a full range of financial professionals and students. It examines the behavioral and statistical causes of market momentum while also exploring the practical side of implementing related strategies.

The phenomenon of momentum in finance occurs when past high returns are followed by subsequent high returns, and past low returns are followed by subsequent low returns. Market Momentum provides a detailed introduction to the financial topic, while examining existing literature. Recent academic and practitioner research is included, offering a more up-to-date perspective.

What type of book is Market Momentum and how does it serve a range of readers' interests and needs?

• A holistic market momentum guide for industry professionals, asset managers, risk managers, firm managers, plus hedge fund and commodity trading advisors
• Advanced text to help graduate students in finance, economics, and mathematics further develop their funds management skills
• Useful resource for financial practitioners who want to implement momentum trading strategies
• Reference book providing behavioral and statistical explanations for market momentum

Due to claims that the phenomenon of momentum goes against the Efficient Markets Hypothesis, behavioral economists have studied the topic in-depth. However, many books published on the subject are written to provide advice on how to make money. In contrast, Market Momentum offers a comprehensive approach to the topic, which makes it a valuable resource for both investment professionals and higher-level finance students.

The contributors address momentum theory and practice, while also offering trading strategies that practitioners can study.

CHAPTER 1
Behavioural Finance and Momentum

ANDREW GRANT

1.1 INTRODUCTION

Behavioural finance aims to provide explanations for observed market phenomena outside the neo-classical view of financial markets. Proponents of efficient markets may argue that behaviourally biased traders will not affect prices in equilibrium, as they will make incorrect decisions and be driven out by rational arbitrageurs. Indeed, the ‘no free lunch’ or ‘economic’ approach to market efficiency suggests that, by and large, it should be difficult to profit from a simple trading strategy – such as momentum – without access to superior information. Under an alternative definition of market efficiency (e.g. Fama, 1970) asset prices should reflect fundamental value. A limits-to-arbitrage (Shleifer and Vishny, 1997) viewpoint contends that asset prices may not necessarily trade at fundamental levels because rational arbitrageurs face constraints that make it costly or risky to correct mispricing.

To provide an example, equity analysts play a crucial role in information production, and stocks without sufficient focus from market participants may not be correctly valued. Similarly, stocks with other trading frictions, such as low liquidity, high levels of idiosyncratic volatility, low levels of institutional investor ownership (making short-selling difficult) and high levels of valuation uncertainty (growth or technology stocks) are likely to experience stronger trading frictions. Indeed, much of the evidence shows that momentum is particularly prevalent among stocks that are subject to these trading frictions.

Other chapters in this book seek to explain momentum as a purely statistical phenomenon, for example, arising from autocorrelation, or statistical properties of sorted portfolios. In this chapter, we take a contrasting view, seeking to understand why the return processes may hold these properties to begin with. Behavioural finance helps provide insight into momentum returns by considering psychological explanations, including overreaction, underreaction, slow information diffusion, anchoring and sentiment. A common theme among these phenomena is that mispricing – or a temporary deviation from fundamental value – spurs momentum.

The approach taken in this chapter is not that momentum is solely driven by behavioural biases, but that a consideration of investor psychology may help to partially explain the prevalence of momentum profits or may be used to enhance the returns of a momentum strategy. A strict definition of what is categorised as ‘behavioural finance’ is not imposed, but I will consider issues that appear to be underpinned by either non-traditional investor preferences or beliefs, retail vs. institutional investors, or market-wide sentiment.

The behavioural approach mainly gained traction with the advent of the three-factor model of Fama and French (1996), and the noted inability of beta, firm size, and the value effect to explain short-term momentum returns. Following this, three highly influential models utilising various aspects of investor psychology were developed by Barberis, Shleifer and Vishny (1998), Daniel, Hirshleifer and Subrahmanyam (1998), and Hong and Stein (1999). These models have formed the basis for many empirical tests of momentum from the behavioural finance perspective. This chapter first examines the failure of risk-based explanations, then considers some of the predictions from the theoretical behavioural models, and then explores some empirical tests relating to these models.

One of the key insights of the Hong and Stein (1999) model is that slow information diffusion among market participants can lead to momentum. Empirical work by Hong, Lim and Stein (2000) has demonstrated momentum is particularly prevalent in small stocks, for example, supporting this notion. Chen and Lu (2017) use option markets, which are likely to be the choice of trading venue for informed traders, to infer the speed of information diffusion. They find that momentum is more pronounced in stocks that exhibit large changes in option implied volatility (where information is likely incorporated in the option but not stock market) exhibit stronger momentum than those that exhibit small changes in implied volatility.

The return pattern or price path taken by a stock, including large price movements – or an absence of large price movements – has also been shown to lead to continuation (momentum in returns), as investors may not efficiently impound information into stock prices. Recent evidence (Atilgan et al., 2020) suggests this is particularly an issue for downward price movements, where investors appear to underreact to potential risks of further price declines.

Closely related to momentum, the 52-Week High effect, as first documented by George and Hwang (2004), states that firms trading near the highest point over the previous year tend to continue upwards. Grinblatt and Han (2005) suggested that stocks near the 52-Week high are in the domain of gains for investors with prospect-theoretic preferences. Preference to sell winners (also known as the disposition effect) induces uninformed selling pressure by such investors, which consequently may lead to return continuation, supporting the contention of underreaction or delayed reaction driving momentum.

The issue of who ‘creates’ momentum can also be considered a behavioural issue. Unlike most other asset pricing anomalies, institutional investors appear to trade ‘with’ momentum strategies (Edelen, Ince and Kadlec, 2016). The counterparties, therefore are likely to be individual investors, selling out winners before price increases (as per the predictions of Grinblatt and Han, 2005), and similarly buying losers. While in the short-term, Kaniel, Saar and Titman (2008) argue that individuals are compensated for providing liquidity, at longer horizons, they tend to underperform institutional investors.

The final issue that is addressed in this chapter relates to investor sentiment. Recent attempts to operationalise behavioural finance have led to the construction of ‘top-down’ sentiment indices (e.g. Baker and Wurgler, 2006). The main idea of a sentiment index is to capture excessive mispricing (either over- or under-valuation), by combining factors such as IPO first-day returns that likely indicate excessive optimism or pessimism. Stocks that are difficult to value or difficult to arbitrage are most likely to load positively on a sentiment index. Momentum strategies appear to be mainly profitable during periods of high sentiment, but not during periods of low sentiment (e.g. Stambaugh, Yu and Yan, 2012; Antoniou, Doukas and Subrahmanyam, 2013). This is arguably driven by investor preferences but may also be driven by liquidity.

1.2 THE FAILURE OF RISK-BASED EXPLANATIONS

The inability to explain the returns to a momentum strategy was described as the ‘prime embarrassment’ of the three-factor model by Fama and French (1996, p. 75). After all, the value (High book-to-market value Minus Low book-to market value [HML]) factor in a three-factor model predicts that losing stocks will outperform winning stocks, consistent with long-term reversals but not with short-term momentum. Later, Carhart (1997) added the UMD (Up minus Down) factor to assess whether a portfolio's returns are consistent with those of a momentum strategy.

Other researchers have attempted to provide a risk-based explanation for momentum returns. Chordia and Shivakumar (2002), Cooper, Gutierrez and Hameed (2004) and Stivers and Sun (2010) all argue that momentum profits are strong during macroeconomic expansions but largely non-existent during recessions. Chapter 15 of this book discusses issues of momentum across the business cycle in more detail. Daniel and Moskowitz (2016) demonstrate that momentum performed particularly poorly during the recovery period from the financial crisis and argue that negative skewness in momentum returns (small positive returns most of the time, but occasional large crashes) is consistent with risk-based explanation. While I will not argue that these viewpoints are invalid, the fact that momentum has been a profitable trading strategy fairly consistently indicates that alternative (i.e. behavioural) explanations may be necessary to complete the picture.

1.3 BEHAVIOURAL MODELS OF MOMENTUM

Several key studies (e.g. Jegadeesh and Titman, 1993; Fama and French, 1996; Grundy and Martin, 2001; Griffin, Ji, and Martin, 2003) failed to find a satisfactory risk-based explanation for momentum profits, which led to the development of behavioural-based explanations. Three studies are considered seminal works in this area; Barberis et al. (1998, henceforth BSV), Daniel et al. (1998, henceforth DHS), and Hong and...

Table of contents

• Cover
• Table of Contents
• Abstracts
• Contributors
• Introduction
• CHAPTER 1: Behavioural Finance and Momentum
• CHAPTER 2: A Taxonomy of Momentum Strategies
• CHAPTER 3: Demystifying Time-Series Momentum Strategies: Volatility Estimators, Trading Rules and Pairwise Correlations
• CHAPTER 4: Risk and Return of Momentum in Developed Equity Markets
• CHAPTER 5: Momentum Across Asset Classes
• CHAPTER 6: Momentum in Momentum ETFs
• CHAPTER 7: CTA Momentum
• CHAPTER 8: Overreaction and Faint Praise – Short-Term Momentum in Contemporary Art
• CHAPTER 9: Volatility-Managed Momentum
• CHAPTER 10: Theoretical Analysis of the Fama-French Portfolios
• CHAPTER 11: Exploiting the Countercyclical Properties of Momentum and other Factor Premia – A Cross-Country Perspective
• CHAPTER 12: Time-Series Variation in Factor Premia: The Influence of the Business Cycle
• CHAPTER 13: Where Goes Momentum?
• CHAPTER 14: Time-Series Momentum in Credit: Machine Learning Approach
• CHAPTER 15: Momentum and Business Cycles
• CHAPTER 16: Momentum as a Fundamental Risk Factor
• CHAPTER 17: Momentum, Value and Carry Commodity Factors for Multi-Asset Portfolios
• Index
• End User License Agreement