Our time-varying portfolio and a dynamic economy

How our time-varying approach works

Typical portfolio construction theory seeks to maximize risk-adjusted returns. “You want a portfolio to get the most return per unit of risk that an investor is willing to take,” said Harshdeep Ahluwalia, a Vanguard senior investment strategist and member of the Vanguard portfolio construction team. “But the definition of risk and return typically changes. What is the expected return of the asset, and what is the volatility, and what is the correlation between them? Let’s form a portfolio with the best trade-off.”

The Vanguard Asset Allocation Model (VAAM) is the vehicle that optimizes the portfolio, mathematically configured to determine that best trade-off. It measures risk not in terms of volatility, as in traditional portfolio construction theory, but rather in terms of distribution of expected returns.

Inputs to VAAM come from the Vanguard Capital Markets Model^®, which projects the estimated interrelationship among risk factors and asset classes over time. The VCMM produces the 10-year asset-class return outlooks that Vanguard publishes regularly and that are based on the central tendency of the model’s results. (For more on how it works, see the “About the Vanguard Capital Markets Model” section at the end of the article.) VAAM, in assessing risk for its optimal time-varying asset allocations, looks beyond central tendency to the VCMM’s extremes—the expected outcomes at the 5^th and 95^th percentiles. Greater variation in expected outcomes at those extremes constitutes greater risk.

So for an investor with a 10-year goal, for example, “we’re looking at the end of the 10-year horizon and the dispersion of returns and weighing the trade-offs of different weights to different assets and identifying what provides the best return per unit of dispersion,” Ahluwalia said.

Here’s how our time-varying 60% stock/40% bond portfolio from a U.S. investor’s perspective has changed over the last two years.

The chart on the left shows a steady-state, or baseline, portfolio with stock allocations of 60% and bond allocations of 40%. In May 2020, early in the COVID-19 pandemic, stocks had sold off considerably, bringing valuations lower and sending expected 10-year returns higher. “The expected compensation increased for taking more equity risk compared with the steady-state portfolio’s equilibrium risk–return profile,” said Kevin DiCiurcio, who leads the VCMM research team. Such an environment increased the equity allocation to 75%, while allocations to bonds and their then historically low yields were reduced to 25%.

In May 2021, DiCiurcio noted, “interest rates were off their rock-bottom lows and the equity market had rallied. Compared with a year earlier, our forecasts had gone up for bonds and down for equities, so our optimal portfolio had taken off the equity overweight.”

And in May 2022, with rates having risen significantly and equities having sold off again, expected 10-year returns for both asset classes were improved. “It’s not as simple as saying equities have sold off; why not take a bit more equity risk? Because bonds have gotten so much more attractive,” DiCiurcio said. Rather, the concept of the “efficient frontier”— the set of portfolios expected to provide the greatest return across various risk appetites—comes into sharper focus.

“The risky-asset decision is a relative one between stocks and bonds,” he said. “VAAM is based on the actual forecast and tells us how much to reward an asset class in an allocation based on market conditions.”