FTSE Russell Insights

Multi-asset return correlations – a new regime, or an era of instability? 

Indrani De, CFA, PRA

Head, Global Investment Research, FTSE Russell

Mark Barnes, Ph D,

Head of Global Investment Research, Americas, Global Investment Research, FTSE Russell

Robin Marshall, M.A., M. Phil,

Director, Global Investment Research, FTSE Russell

Alex Nae, M.Sc

Quantitative Research Analyst, Global Investment Research, FTSE Russell
  • Evidence from FTSE Russell indices shows higher correlation of multi-asset returns persisted in the post-Covid period, despite disinflation…
  • …with confirmation of a structural break higher in 2021 for asset return correlations across bonds, equities and credit
  • It is too early to draw structural conclusions about the durability of this increase in the correlations of returns…
  • …but its persistence may be due to higher for longer short rates and inertia in core inflation.
  • The results indicate inflation is a strong driver of the higher correlation in returns, even if the relationship is non-linear, with non-linearity evident when inflation moves above Fed target levels.
  • This makes portfolio diversification benefits (particularly from sovereign bonds), based on the so-called 60/40 model – 60% equity weighting, 40% in bonds - more variable over time.

Introduction

In a short note last year[1], we pointed out the observed correlation of US stock and government bond returns had increased sharply in 2022-23, as the Federal Reserve raised interest rates in response to a sharp increase in inflation. This ended a long period, starting in the mid-1990s, in which the correlation of stock and government bond returns had been relatively stable, low and even negative (during the deflationary shocks of the GFC and Covid). Indeed, the evidence since 2021 confirms that relying on historical correlations to predict future correlations may give poor results. Figure 1 shows both how low the correlation between stock and bond returns was from the mid-1990s to 2020, and the extent of the recent increase, since Covid.

Figure 1: Correlation of 7-10 year Treasuries with Russell 1000 index 

Chart 1 shows Chart shows rolling 12 month correlation of daily returns of Russell 1000 index and FTSE Russell 7-10 year US Treasury index.

Source: FTSE Russell, data from January 1994 to April 2024. Please see the disclaimers for important legal disclosures.

Changes in key macro drivers may explain both the changes in the correlation of asset returns and in the underlying volatility of returns (which, in turn, drive the higher correlations). So, in a longer paper, published this month, we seek to identify the key drivers of correlations in returns, using FTSE Russell multi-asset index data since 2000. To assess the correlations in returns, we looked at the purest risk-off asset in US Treasuries, using the FTSE Russell 7-10 year Treasury index returns, and its correlation coefficient with US equity returns, using the Russell 1000 index.

Pre-Covid, low inflation meant growth & deflationary shocks drove low correlations 

In the Goldilocks era, and in the period after the GFC, structurally low inflation meant investors focussed on cyclical fluctuations in real growth and corporate earnings as the principal drivers of equity returns. Meanwhile, cyclical slowdowns and monetary policy easing by the Federal Reserve drove lower US Treasury yields (and superior bond returns). So equities outperformed during growth recoveries, while government bonds did best during recessions, or deflationary shocks, like the GFC and Covid, when investors switched into safe havens like US Treasuries, as central banks adopted zero interest rate policies and QE. Inflation upticks were generally mild, and driven by excess demand, rather than the supply-side energy shocks of the 1970s.This demand-led inflation has been described as “ good inflation”[2] , as opposed to the “ bad inflation “ of supply-side shocks like Covid or energy price spikes.

Since Covid, a perfect storm of higher inflation and short rates may have driven high return correlations

Evidence of the impact of higher inflation on asset return correlations since Covid, may be found in Figure 2. Of course, this does not establish a causal link to higher inflation and it may also be due to the sharp rise in short interest rates that followed. But it is consistent with longer term evidence showing “ higher stock-bond correlations tend to follow higher short rates and (to a lesser extent) higher inflation rates “[3] Intriguingly, the correlation of returns also remained high in 2023, when US inflation fell sharply, suggesting the relationship may not be one-directional with inflation, and could be non-linear.

Figure 2: US inflation and correlation of US 7-10 year Treasury and Russell 1000 returns.

Figure 3 shows he chart shows the relationship between US inflation and correlation of US Treasury 7-10 year. and Russell 1000 equity returns from Q1-2001 to Q1-2024.

Source: FTSE Russell and Refinitiv-Datastream. T Please see the disclaimers for important legal disclosures.

Finally, extending the analysis to other asset classes in fixed income delivers similar results, with correlations of returns again showing a structural break higher after Covid, as Figure 3 shows. The results reflect the risk characteristics of different asset classes, with risk-on high yield credit showing higher correlation to equity returns in all periods. In fact, HY is the only asset class that does not show a structural break higher in correlations since Covid.

Figure 3: Asset class correlation of US 7-10 yr Treasury and Russell 1000 returns.

Figure 3 shows he chart shows the relationship between US inflation and correlation of US Treasury 7-10 year. and Russell 1000 equity returns from Q1-2001 to Q1-2024.

Source: FTSE Russell and Refinitiv-Datastream. T Please see the disclaimers for important legal disclosures.

 

1. A marriage of inconvenience: the remarkable harmony between stocks and bonds”, LSEG, April 2023.

2. See “Inflation and Asset returns “, Anna Cieslak and Carolin Pflueger, Working paper No. 2023-34, Becker Friedman Institute, University of Chicago, March 2023.

3. The stock–bond correlation and macroeconomic conditions: One and a half centuries of evidence”, Jian Yang, Ying-gang Zhou, and Zijun Wang, Journal of Banking and Finance, April 2009

Read more about

Stay updated

Subscribe to an email recap from:

Disclaimer

© 2024 London Stock Exchange Group plc and its applicable group undertakings (“LSEG”). LSEG includes (1) FTSE International Limited (“FTSE”), (2) Frank Russell Company (“Russell”), (3) FTSE Global Debt Capital Markets Inc. and FTSE Global Debt Capital Markets Limited (together, “FTSE Canada”), (4) FTSE Fixed Income Europe Limited (“FTSE FI Europe”), (5) FTSE Fixed Income LLC (“FTSE FI”), (6) FTSE (Beijing) Consulting Limited (“WOFE”) (7) Refinitiv Benchmark Services (UK) Limited (“RBSL”), (8) Refinitiv Limited (“RL”) and (9) Beyond Ratings S.A.S. (“BR”). All rights reserved.

FTSE Russell® is a trading name of FTSE, Russell, FTSE Canada, FTSE FI, FTSE FI Europe, WOFE, RBSL, RL, and BR. “FTSE®”, “Russell®”, “FTSE Russell®”, “FTSE4Good®”, “ICB®”, “Refinitiv” , “Beyond Ratings®”, “WMR™” , “FR™” and all other trademarks and service marks used herein (whether registered or unregistered) are trademarks and/or service marks owned or licensed by the applicable member of LSEG or their respective licensors and are owned, or used under licence, by FTSE, Russell, FTSE Canada, FTSE FI, FTSE FI Europe, WOFE, RBSL, RL or BR. FTSE International Limited is authorised and regulated by the Financial Conduct Authority as a benchmark administrator. Refinitiv Benchmark Services (UK) Limited is authorised and regulated by the Financial Conduct Authority as a benchmark administrator.

All information is provided for information purposes only. All information and data contained in this publication is obtained by LSEG, from sources believed by it to be accurate and reliable. Because of the possibility of human and mechanical inaccuracy as well as other factors, however, such information and data is provided "as is" without warranty of any kind. No member of LSEG nor their respective directors, officers, employees, partners or licensors make any claim, prediction, warranty or representation whatsoever, expressly or impliedly, either as to the accuracy, timeliness, completeness, merchantability of any information or LSEG Products, or of results to be obtained from the use of LSEG products, including but not limited to indices, rates, data and analytics, or the fitness or suitability of the LSEG products for any particular purpose to which they might be put. The user of the information assumes the entire risk of any use it may make or permit to be made of the information.

No responsibility or liability can be accepted by any member of LSEG nor their respective directors, officers, employees, partners or licensors for (a) any loss or damage in whole or in part caused by, resulting from, or relating to any inaccuracy (negligent or otherwise) or other circumstance involved in procuring, collecting, compiling, interpreting, analysing, editing, transcribing, transmitting, communicating or delivering any such information or data or from use of this document or links to this document or (b) any direct, indirect, special, consequential or incidental damages whatsoever, even if any member of LSEG is advised in advance of the possibility of such damages, resulting from the use of, or inability to use, such information.

No member of LSEG nor their respective directors, officers, employees, partners or licensors provide investment advice and nothing in this document should be taken as constituting financial or investment advice. No member of LSEG nor their respective directors, officers, employees, partners or licensors make any representation regarding the advisability of investing in any asset or whether such investment creates any legal or compliance risks for the investor. A decision to invest in any such asset should not be made in reliance on any information herein. Indices and rates cannot be invested in directly. Inclusion of an asset in an index or rate is not a recommendation to buy, sell or hold that asset nor confirmation that any particular investor may lawfully buy, sell or hold the asset or an index or rate containing the asset. The general information contained in this publication should not be acted upon without obtaining specific legal, tax, and investment advice from a licensed professional.

Past performance is no guarantee of future results. Charts and graphs are provided for illustrative purposes only. Index and/or rate returns shown may not represent the results of the actual trading of investable assets. Certain returns shown may reflect back-tested performance. All performance presented prior to the index or rate inception date is back-tested performance. Back-tested performance is not actual performance, but is hypothetical. The back-test calculations are based on the same methodology that was in effect when the index or rate was officially launched. However, back-tested data may reflect the application of the index or rate methodology with the benefit of hindsight, and the historic calculations of an index or rate may change from month to month based on revisions to the underlying economic data used in the calculation of the index or rate.

This document may contain forward-looking assessments. These are based upon a number of assumptions concerning future conditions that ultimately may prove to be inaccurate. Such forward-looking assessments are subject to risks and uncertainties and may be affected by various factors that may cause actual results to differ materially. No member of LSEG nor their licensors assume any duty to and do not undertake to update forward-looking assessments.

No part of this information may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written permission of the applicable member of LSEG. Use and distribution of LSEG data requires a licence from LSEG and/or its licensors.