Less pain, more gain: Volatility-adjusted residual momentum in international equity markets

ABSTRACT

We offer a new type of momentum strategy — the volatility-adjusted residual momentum (VARMOM) — which is based on average past residuals scaled with their volatility. We demonstrate its application for international asset allocation within 51 country indexes and 888 industry portfolios from developed and emerging markets. The VARMOM trading strategy notably outperforms and subsumes a standard momentum strategy, delivering Sharpe ratios that are two to three times higher. The VARMOM is particularly strong across portfolios characterised by high limits to arbitrage and following bull markets, supporting the behavioural explanation of momentum. The results are robust to alternative portfolio construction methods as well as the inclusion of trading costs and control variables. They are also valid for several subperiods and subsamples.

KEYWORDS:

• VARMOM
• residual momentum
• volatility-adjusted momentum
• country momentum
• industry momentum
• asset pricing
• international investment
• return predictability
• equity anomalies
• cross section of returns

JEL CODES:

• G12
• G15

Author’s note and acknowledgments

We thank the participants of the 7^th Scientific Conference ‘Modelling and forecasting national economy’ at the University of Gdansk (Poland, 2017) and the participants the Dubai Business School Research Seminar (UAE, 2017) for helpful comments that benefited the paper. This study is part of Project No. 2014/15/D/HS4/01235 financed by the National Science Centre of Poland. Correspondence concerning this article should be addressed to Adam Zaremba, Department of Investment and Capital Markets, Poznan University of Economics and Business, al. Niepodleglosci 10, 61-875 Poznan, Poland, e-mail: [email protected]

ORCID

Adam Zaremba http://orcid.org/0000-0001-5879-9431

Mehmet Umutlu http://orcid.org/0000-0003-1353-2922

Notes

1. Momentum returns display a negative skewness and a high kurtosis, pointing out a left-tail risk with occasional crushes (Barosso & Santa-Clara, 2015; Daniel & Moskowitz, 2016).

2. For more information on the ICB classification, visit www.icbenchmark.com

3. http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html (accessed 25 June 2016).

4. We require the full 60-month history to determine the parameters.

5. By excess return, we mean the raw return minus the return on a risk-free asset.

6. For robustness, we also checked alternative factor formation methods, including diversified value-weighted portfolios following Fama and French (1993), but this did not qualitatively change the results.

7. All of these variables have been tested for their predictive abilities over future returns on indexes: for market size, see Keppler and Traub (1993) and Keppler and Encinosa (2011); for long-run reversal, see Balvers and Wu (2006), Asness, Moskowitz, and Pedersen (2013), Malin and Bornholt (2013) and Bornholt, Gharaibeh, and Malin (2015); for beta, see Frazzini and Pedersen (2014); for idiosyncratic risk, see Bali and Cakici (2010) and Umutlu (2015); for EP ratio, see Macedo (1995) and Zaremba (2015).

8. The original study of Cooper et al. (2004) assumed a 36-month lookback window.

9. TED spread is calculated as the difference between the 3-month US$ LIBOR and the 3-month US benchmark T-Bill rate. All the data were obtained from Bloomberg. Furthermore, we are aware of the limitation of this study that all four proxies for limits to arbitrage are derived from US data. However, this is motivated in four ways: (a) this is consistent with the US dollar approach used in this study; (b) the US financial market is by far the largest and most important in the world; (c) the financial time series for the United States are readily available, which is not true for most emerging and many developed markets; and (d) the examined proxies based on US data and their few counterparts in other countries are strongly correlated. For example, the correlation coefficient between the Germany-based VDax and the US-based VIX during the last two decades was close to unity.

10. Naturally, this approach cannot be used directly as an asset allocation strategy because we do not know ex-ante the median values. Nonetheless, it provides a general picture of the behaviour of the momentum strategy.

11. The data are retrieved from the website of Jeffrey Wurgler: http://people.stern.nyu.edu/jwurgler/

12. All data are sourced from Bloomberg. For further information on CBCC, see: https://www.conference-board.org/data/consumerconfidence.cfm; for MICE: http://www.sca.isr.umich.edu/. Regarding, the US nature of our data, see footnote 7 regarding the limits to arbitrage.

13. For brevity, we do not report specific Sharpe ratios in Table 6. Nonetheless, it is important to note that for both industries and countries, both VARMOM strategies (ADJ-MOM_CAPM and ADJ-MOM_FF) displayed higher Sharpe ratios than MOM_RAW in each of the four investigated subperiods.