The 52-week High Momentum Strategy and Economic Policy Uncertainty: Evidence from China

ABSTRACT

This is the first study to examine the 52-week high momentum strategy that takes economic policy uncertainty (EPU) into account. Empirically, we find significant 52-week high momentum in China, the second-largest stock market in the world, and our findings confirm the results of the US. We hypothesize that anchoring biases could explain the 52-week high momentum and generate significant momentum profits in low EPU periods. The empirical results show strong 52-week high momentum in low EPU periods; there is virtually no momentum when EPU is high, which supports our prediction. Further investigations show that there are no long-run reversals for the 52-week high momentum, and the negative impact of EPU on the 52-week high momentum decreases and eventually vanishes over the long term. All evidence supports the hypothesis that the 52-week high momentum is attributed to anchoring biases, especially when we consider EPU.

KEYWORDS:

• 52-week high
• momentum strategy
• anchoring biases
• economic policy uncertainty
• EPU

JEL:

• G12
• G14
• G23

Acknowledgments

We sincerely thank two anonymous reviewers and Narayan (the editor) for many constructive suggestions, which significantly enhance our paper’s exposition and scholarship. We would also like to thank Huacheng Zhang for bringing the 52-week high momentum strategy to the authors; Yuan Li for her kindly help on issues about 52-week high momentum; Ronghua Luo for many helpful comments; and Dongzhi Chen and Biao Yi for their kindly help with data.

Supplementary Material

Supplemental data for this article can be accessed on the publisher’s website.

Notes

1. We rewrite introduction at the suggestion of an anonymous reviewer.

2. Lou (2012) explains the momentum through a flow-based mechanism and argues that more capital will be attracted to the past winners’ funds, while the past losers’ funds are confronted with capital outflow or even redemption. Thus, investors often learn from a stock’s past performance based on this mechanism to adjust their investment decisions.

3. Baker, Bloom, and Davis (2016) constructed the Chinese EPU index, which is based on the scaled frequency count of articles about policy-related economic uncertainty in the South China Morning Post (hereafter, SCMP), which is Hong Kong’s leading English-language newspaper.

4. Ludvigson, Ma, and Ng (2019) argue that any examination of EPU effects ought to control for other uncertainties. At the suggestion of an anonymous reviewer, we use FSI as the proxy for a set of macro and financial-economic uncertainty variables. The FSI data for China can be downloaded from https://aric.adb.org/database/fsi.

5. We thank an anonymous reviewer for this constructive suggestion.

6. These no-momentum studies focus on JT momentum, which bet on stocks’ past returns to predict future returns.

7. The securities laws and regulations before 2000 include, but are not limited to Regulations on the management on the issuance and trading of stocks, Regulations on the information management of Shenzhen Stock Exchange, Rules for the management of traders in Shanghai Stock Exchange, and Interim provisions on securities and futures investment consulting management.

8. We thank an anonymous reviewer for the suggestion of better arrangement of Section 3.

9. We also compute the abnormal returns, and the results are similar to Table 1. To save space, we will provide them upon request.

10. We also test GH momentum strategies for out-of-sample period and sub-periods. The results are similar to Table 1. In addition, the results still hold when we consider the extreme returns and different holding periods. To save space, we will provide them upon request.

11. The arguments are suitable for Table 2.

12. We thank an anonymous reviewer for the suggestion of EPU in Equation (2)(2) $r_{\mathrm{t}}-r_{ft}=\alpha +{\beta }_{1}EP{U}_t+{\beta }_{2}MK{T}_t+{\beta }_{3}SM{B}_t+{\beta }_{4}HM{L}_t+{\epsilon }_t$(2) .

13. The results remain when we consider out-of-sample periods, sub-periods, extreme returns, and different holding periods. To save space, we will provide the results upon request.

14. We thank an anonymous reviewer for this insightful suggestion. The FSI data for China can be downloaded from https://aric.adb.org/database/fsi.

15. We first define each month (from t-K to t-1) as high (low) EPU if the EPU in the last month was in the top (bottom) 30% or 40%; Otherwise, the month is defined as moderate.

16. We thank an anonymous reviewer for this insightful suggestion.

17. Note that the regression of Equation (3)(3) $r_{K,t}={\alpha }_1{H}_t+{\alpha }_2{M}_t+{\alpha }_3{L}_t+{\epsilon }_{K,t}$(3) and simple summary statistics are equivalent. We thank an anonymous reviewer for pointing this out.

18. We thank an anonymous reviewer for this constructive suggestion.

19. To save space, we only provide relevant partial results in Tables 4–7.

20. We also estimate twelve cross-sectional regressions from Equation (6)(6) ${\bar{\beta }}_{3t}-{\bar{\beta }}_{4t}={\alpha }_1{H}_t+{\alpha }_2{M}_t+{\alpha }_3{L}_t+{\epsilon }_t$(6) when j = 2 to 13.

21. To save space, tables in this section are shown in the supplementary materials.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [71701171, 72073109, and 72071162] and Huaxi Futures Co., Ltd.