Inflation is on the rise in many parts of the world, and that means interest rates likely will be too. Financial asset pricing models suggest that inflation can influence stocks and bonds similarly, resulting from a shared relationship with short-term interest rates. Therefore, some investors have begun to wonder: Will stock and bond returns start to move in tandem and, if so, what could that mean for diversification in a balanced portfolio?

To answer these questions, my colleagues and I have identified the factors that have historically driven stock and bond co-movements over time and have published our findings in The Stock/Bond Correlation: Increasing Amid Inflation, but Not a Regime Change. Chief among those drivers is inflation, and we found that it would take considerably more inflation than we're expecting for stocks and bonds to move together to a degree that would diminish the diversifying power of bonds in a balanced portfolio.

Why long-term investors maintain a balanced portfolio

It's important to understand why so many investors hold a balanced portfolio of stocks and bonds. Stocks serve as a portfolio's growth engine, the source of stronger expected returns in the majority of market environments. If they always outperformed bonds or otherwise had assured outcomes, however, investors would have little incentive to also hold bonds. Although stock prices historically have risen over time, their trajectory hasn't been straight. They've endured a lot of bumps—and several sharp contractions—along the way.

That's where bonds come in. Bonds typically have acted as ballast for a portfolio, with prices rising—or falling less sharply—during periods when stock prices are falling. That contrasting return pattern helps minimise losses to a portfolio's value compared with an all-stock portfolio. It helps investors adhere to a well-considered plan in a challenging return environment.

Correlations in context: Time matters

We use the term correlation to explain how stock and bond returns move in relation to one another. When returns generally move in the same direction, they are positively correlated; when they move in different directions, they are negatively correlated. The combination of negatively correlated assets will enhance diversification by smoothing the fluctuations in portfolio asset values through time. Lately, however, stock and bond returns have more frequently moved in the same direction and have even, at times, been positively correlated. But these positive correlations have happened for relatively brief periods. And, as it turns out, time matters.

Notes: Long-term stock/bond correlations were largely positive during much of the 1990s but have mostly been negative since about 2000. It is not uncommon for the correlation to turn positive over the shorter term, but this has not altered the longer-term negative relationship.
Sources: Vanguard, based on data from Refinitiv from January 1, 1990, through July 26, 2021. Data appear on chart only at the start of 1992 to reflect the end of the first 24-month rolling correlation.
Past performance is no guarantee of future returns.

As with any investment performance, looking solely at short periods will tell you only so much. Since 2000, stock/bond correlations have spiked into positive territory on numerous occasions. Correlations over the longer term, however, remained negative, and we expect this pattern to persist.

How much inflation would it take?

Our research identified the primary factors that have influenced stock and bond correlations from 1950 until today. Of these, long-term inflation has by far been the most important.

Because inflation moves stock and bond returns in the same direction, the question becomes: How much inflation would it take to move return correlations from negative to positive? The answer: a lot.

By our numbers, it would take an average 10-year rolling inflation of 3.5%. This is not an annual inflation rate; it's an average over 10 years. For context, to reach a 3% 10-year average any time soon—say, in the next five years—we would need to maintain an annual core inflation rate of 5.7%. In contrast, we expect core inflation in 2022 to be about 2.6%, which would move the 10-year trailing average to just 1.8%.

You can read more about our U.S. inflation outlook in our recent paper The Inflation Machine: What It Is and Where It's Going. The Federal Reserve, in its efforts to ensure price stability, targets 2% average annual inflation, far beneath the threshold that we believe would cause positive correlations of any meaningful duration. It's also well below inflation rates in the pre-2000 era, which from 1950 to 1999 averaged 5.3% and were associated with positive long-term stock/bond correlations.

Note: The figure shows Vanguard's projections for stock-bond correlations under four scenarios for 10-year inflation from April 2021 through December 2025.
Source: Vanguard.

Asset allocation, more than correlation, influences portfolio outcomes

What does this mean for the traditional 60% stock/40% bond portfolio? For investors who feel an itch to adjust their portfolios in preparation for a reversal in stock/bond correlations, we might say, "Not so fast." In the portfolio simulation environment that we tested, positive versus negative correlations affected measures of fluctuations in portfolio values, such as volatility and maximum drawdown, through time but had little impact on the range of long-term portfolio outcomes. What's more, we found that shifting a portfolio's asset allocation toward stocks—to 80% from 60%—led to a more prominent change in the portfolio's risk profile than did the portfolio's remaining 60/40 during a correlation regime change.

This aligns with something you may have heard us say before: Portfolio outcomes are primarily determined by investors' strategic asset allocations. And this is good news because, with proper planning, investors with balanced portfolios should be well-positioned to stay on course to meet their goals, instead of swerving to avoid bumps in the road.

Kevin DiCurio, CFA, Vanguard Capital Markets Model research team