united kingdomen
The value of multiple momentum signals

The value of multiple momentum signals

18-07-2017 | Research

The concept of Momentum is not only a basic principle of physics. The ‘property or tendency of a moving object to continue moving’1 also applies to the world of finance. In a financial context, momentum refers to the observation that assets which, relative to their peers, have had the highest returns over the past six to 12 months, will continue to deliver above-average returns in the months to come. Due to its pervasiveness, momentum has been studied extensively by both the academic community, as well as practitioners.

  • Milan Vidojevic
    Milan
    Vidojevic
    Quantitative Researcher

Since the publication of Narasimhan Jegadeesh and Sheridan Titman’s seminal paper, back in 1993, which formally documented the phenomenon’s existence, researchers have been trying to understand whether the prevalence of momentum implies that markets are inefficient at processing information, or if the premium is reasonable compensation for bearing some systematic risk. In order to answer this question, one needs to link profits of momentum to either behavioral or risk based asset pricing theories. 

To illustrate the strength of the momentum effect, if you had invested one dollar in a passive, capitalization-weighted U.S. stock market index portfolio in January 1927, by the end of 2016, the yield would have been USD 4,538, while investing one dollar in the capitalization-weighted portfolio of past winners would have generated USD 473,517, and in a portfolio of past losers, just USD 111.2

No free lunch

Are momentum profits a free lunch? Definitely not. Momentum strategies can be quite risky and they require a very disciplined approach in order to be successful. For instance, conventional equity momentum strategies are known to exhibit significant dynamic exposures to systematic factors, also known as styles. In a bull market, for example, high beta stocks tend to outperform low beta ones. And when the trend ends, a long-short momentum strategy has a net positive exposure to the market factor. Such exposures can be devastating during style reversals. In 2009, when the market made a strong recovery, after a dismal financial crisis, the momentum factor lost 83%.

In 2001, Bruce Grundy and Spencer Martin3 showed that it is possible to dynamically hedge some of the strategy’s unwanted style exposures, which ex-post results in a less volatile strategy without a lower return. However, their strategy has been criticized, as it is hard to implement based only on the information available at the time of portfolio formation. “A few years later, Roberto Gutierrez and Christo Pirinsky4 and, building on their work, Robeco’s David Blitz, Joop Huij, and Martin Martens, proposed an alternative method to reduce these systematic style tilts by sorting stocks based on their past, firm-specific returns, dubbed ‘residual’, as opposed to total price returns5.

This residual momentum strategy exhibits only half the volatility of the conventional momentum, without a loss in return, thus doubling the Sharpe ratio of the strategy. But the question that remains is if conventional and residual momentum strategies are complements or substitutes.

Stay informed on Quant investing with monthly mail updates
Stay informed on Quant investing with monthly mail updates
Subscribe

Meet residual (idiosyncratic) Momentum

A new empirical study by David Blitz, Mathias Hanauer, and Milan Vidojevic6 addresses these questions. Using formal asset pricing tests, the authors find that conventional and residual (or ‘idiosyncratic’) momentum are distinct phenomena. In order to understand why momentum strategies tend to generate higher returns, the authors consider various prominent explanations for the momentum effect, that have been discussed in the academic literature, and test whether these can explain the profits of residual momentum.

The simple hypothesis is that if conventional and residual momentum strategies stem from the same sources, the observed dominance of residual momentum can be due to greater exposure to them. But the authors find that most of these explanations either fail to establish strong links between these theories and conventional momentum or only support the existence of much weaker links in the case of residual momentum.

Residual momentum can be used to distinguish between conventional momentum stocks with high and low long term returns

The only explanation for the existence of the residual momentum profits for which they found supporting evidence is the underreaction hypothesis. This theory postulates that information about a firm’s fundamentals diffuses gradually among investors, and consequently, prices adjust to it slowly. The authors show that residual momentum forecasts high long-term returns, even after accounting for conventional momentum and other asset pricing factors that are known predictors of long term stock returns.

On the other hand, conventional momentum predicts high returns, but only in the short-term, and the effect starts reversing as early as one year after portfolio formation − a pattern consistent with the overreaction hypothesis. In fact, residual momentum can even be used to distinguish between conventional momentum stocks with high and low long term returns.

Out-of-sample confirmation

This paper also validates the robust performance of residual momentum in several international stock markets. Based on over two decades of out-of-sample data from four investment universes − Europe, Japan, Asia Pacific ex-Japan and Emerging Markets − the authors find that residual momentum generates higher risk-adjusted returns than conventional momentum in all these regions. Moreover, the returns achieved with residual momentum strategies in these different markets cannot be explained by the market, size, value, and total return (conventional) momentum factors.

These results are particularly significant for Japan, where conventional momentum does not seem to function as a stand-alone phenomenon, which raised concerns over possible data mining issues. The authors found that the residual momentum of Japanese stocks included in the FTSE World Developed index generated a statistically significant return of 0.44% per month from December 1989 to December 2015.

Understanding where the different momentum premiums come from and how they are related to other proven factor premiums, such as low-volatility, value, and quality, helps us to decide whether combining them adds value on the portfolio level. The evidence that we present supports the claim that conventional and residual momentum stem from different market mechanisms, and as such are more useful when applied in a complementary fashion rather than as independent factors.

1 As defined by http://www.dictionary.com/browse/momentum
2 These are before costs, paper numbers. The data are obtained from the website of Professor Kenneth French. Winner portfolio is long in big and small capitalization-weighted past winners, and loser portfolio is long big and small past losers.
3 ‘Understanding the Nature of the Risks and the Source of the Rewards to Momentum Investing’, Bruce Grundy and Spencer Martin, The Review of Financial Studies, Vol. 14, No. 1 (Spring, 2001), pp. 29-78.
4 ‘Momentum, Reversal, and the Trading Behaviors of Institutions’, Roberto Gutierrez and Christo Pirinsky, 2006.
5 ‘Residual Momentum’, David Blitz, Joop Huj, and Martin Martens, Journal of Empirical Finance, Vol. 18, 2011.
6 ‘The Idiosyncratic Momentum Anomaly’, David Blitz, Mathias Hanauer, and Milan Vidojevic, 2017.

Disclaimer

Please read this important information before proceeding further. It contains legal and regulatory notices relevant to the information contained on this website.

The information contained in the Website is NOT FOR RETAIL CLIENTS - The information contained in the Website is solely intended for professional investors, defined as investors which (1) qualify as professional clients within the meaning of the Markets in Financial Instruments Directive (MiFID), (2) have requested to be treated as professional clients within the meaning of the MiFID or (3) are authorized to receive such information under any other applicable laws. The value of the investments may fluctuate. Past performance is no guarantee of future results. Investors may not get back the amount originally invested. Neither Robeco Institutional Asset Management B.V. nor any of its affiliates guarantees the performance or the future returns of any investments. If the currency in which the past performance is displayed differs from the currency of the country in which you reside, then you should be aware that due to exchange rate fluctuations the performance shown may increase or decrease if converted into your local currency.

In the UK, Robeco Institutional Asset Management B.V. (“ROBECO”) only markets its funds to institutional clients and professional investors. Private investors seeking information about ROBECO should visit our corporate website www.robeco.com or contact their financial adviser. ROBECO will not be liable for any damages or losses suffered by private investors accessing these areas.

In the UK, ROBECO Funds has marketing approval for the funds listed on this website, all of which are UCITS funds. ROBECO is authorized by the AFM and subject to limited regulation by the Financial Conduct Authority. Details about the extent of our regulation by the Financial Conduct Authority are available from us on request.

Many of the protections provided by the United Kingdom regulatory framework may not apply to investments in ROBECO Funds, including access to the Financial Services Compensation Scheme and the Financial Ombudsman Service. No representation, warranty or undertaking is given as to the accuracy or completeness of the information on this website.

If you are not an institutional client or professional investor you should therefore not proceed. By proceeding please note that we will be treating you as a professional client for regulatory purposes and you agree to be bound by our terms and conditions.

If you do not accept these terms and conditions, as well as the terms of use of the website, please do not continue to use or access any pages on this website.

I Disagree