September 16, 2024

US equity market betas – why they matter and how they are changing

Robin Marshall

M.A., M. Phil, Director, Global Investment Research

Mark Barnes

PhD, Head of Global Investment Research, Americas, Global Investment Research

Indhu Raghavan

CFA, Manager, Global Investment Research

Equity betas are not static but respond to changes in the benchmark, changes in the economic and regulatory environment, and economic shocks. This paper reviews the intuition of betas using Russell 1000 data and discusses some notable changes in industry betas driven both by structural change and potentially transitory changes. The authors highlight beta changes related to several market themes, including the green transition, the AI revolution, and Covid. Investors are reminded that they should monitor industry betas to assess the relative risks of their portfolios, and that the persistence of beta changes is likely tied to the persistence of the economic shocks causing the change. 

Key takeaways:

  • The beta of an investment security, like a stock or group of stocks, measures its correlation and volatility of returns relative to the market, or broader benchmark, and is an important characterisation of the security’s risk relative to the benchmark
  • Betas are not static but respond to changes in the benchmark, changes in the economic and regulatory environment, and economic shocks
  • In recent years, there have been notable changes in the beta of industries and sub-sectors driven both by potentially transitory changes (Covid supply shock, oil price shocks) and by structural shifts in the economy and regulation (investment in AI technologies, the green transition, the lasting legacies of Covid and the GFC)
  • Investors should be mindful of these shifts when thinking about the cyclical/defensive characteristics of their portfolios and assessing where investment opportunities may lie in the context of the economic and market cycle
  • It is difficult to determine whether changes in industry and sector betas are temporary or permanent, but investors should consider the persistence of economic shocks in forming expectations of beta

Points of differentiation:

  • The paper questions assumptions about the stability of industry/sector cyclicality using a data-based approach
  • The paper puts empirical changes in beta in the context of recent economic events and macro themes
  • By revealing recent changes in betas, the paper also helps investors construct more resilient portfolios

What does our research mean for investors

This paper enables investors to review their understanding of betas using Russell 1000 index data and the ICB classification of industries. Specifically, it helps them understand how betas can change in the context of economic events and macro themes, and underscores the importance for investors of continually monitoring and updating their views and expectations of portfolio risk characteristics.

  • The cyclicality of a group of stocks, as measured by their beta to the broad market, gives us an idea of the direction and degree of their co-movement with the market. Are stocks in a certain industry more cyclical, moving in sync with the market to some extent and benefiting from broad market rallies, or are they more defensive, moving differently from the market, thereby buffering portfolios in the event of a broad downturn? It is an important question when thinking about the risk characteristics of a portfolio.

    Investors may be lulled into thinking that these co-movement characteristics are set in stone, and indeed, we frequently talk about cyclical and defensive industries as if these are common knowledge. However, the cyclicality of stocks in a particular industry or sector is not static. 

  • It is important to know why betas have changed, since it may indicate whether these changes are likely to be temporary or permanent. Is it because of underlying structural changes in the economy, or because of one-off shocks, like Covid or the Global Financial Crisis (GFC), or perhaps a combination of the two? We assess these issues in this paper, drawing on Russell data from the last twelve years.