Structural innovation in state variables and expected stock returns

Abstract

Purpose

The purpose of this study is to investigate whether the surprise components of systematic risk, which are useful in forecasting future investment opportunities, help explain the cross-section of average returns associated with portfolios sorted on size, book-to-market and accruals. This study also aims to examine the mispricing attributes of the size, value and accrual effects by investigating the relative economic relevance of aggregate risk factors, which are related to exogenous shocks in state variables, in the cross-sectional returns of triple-sorted portfolios.

Design/methodology/approach

This study uses innovations of systematic risk, which affect the cash flows and risk-adjusted discount rates of all firms in an economy and determines the expected returns of portfolios based on firm characteristics. This study uses independent sorts based on size, book-to-market and total accruals – all of which are measured at the firm level – and construct three-dimensional test portfolios. For unobserved innovations, this study estimates a triangular structural vector autoregressive system and obtain the exogenous innovations in state variables. The author uses Fama-MacBeth two-pass cross-sectional regressions and examines whether the structural innovations explain a significant part of the cross-sectional variation in the average returns of the test portfolios.

Findings

This study finds that variations in expected returns of testing assets are determined by differences in the underlying assets’ exposure to systematic risk innovation. The empirical evidence also shows that exogenous innovation in Fama-French (FF) risk factors leaves out important cross-sectional information about expected returns, and additionally, the FF-factor betas have lower cross-sectional power than the proxy for innovation betas. The cross-sectional differences in the test portfolios’ sensitivity to instruments such as the short-term Treasury bill rate and term spread survive the presence of FF-factor betas.

Originality/value

In contrast to the existing literature, this study uses structural innovations that are uncorrelated and thus exogenous in nature. The author creates test portfolios that display a wide range of average returns and are unlikely to show spurious variability in risk exposures. Unlike the existing research, where size, value and accrual anomalies have been analyzed in isolation, this study examine these pricing patterns jointly, focusing on the possible contributing role of structural innovation in economy-wide predictor variables. To the best of the author’s knowledge, this paper is the first attempt to link the sensitivity of portfolios sorted on size, book-to-market and accruals to exogenous structural innovation.

Citation

Simlai, P.E. (2022), "Structural innovation in state variables and expected stock returns", Managerial Finance, Vol. 48 No. 2, pp. 289-312. https://doi.org/10.1108/MF-02-2020-0076