An analysis digs into the differences between holdings-based and risk factor-based allocation methodologies
When it comes to multi-asset class portfolios, investors have traditionally based their rebalancing targets on the weights in holdings of the asset classes themselves. But is there another way to produce a more stable allocation?
A new note published by MSCI explores what happens when a portfolio is built and rebalanced using a holdings-based approach, and compares outcomes with an alternative risk factor-based allocation methodology. The analysis was performed by Andrea Amato, associate for Fixed Income and Multi-Asset Class Research, and Chenlu Zhou, vice president with the Multi-Asset Class Factor Research Team.
The two started with a hypothetical holdings-based allocation with 30% exposure to global equities, 20% to U.S. Treasury, 10% to inflation-linked bonds (TIPS), and 40% to investment-grade corporate bonds. From Q1 2011 to Q1 2019, they followed the portfolio and rebalanced it to the target weights on a quarterly basis. Over that period, they also noted the risk contributions of equity, interest-rate, inflation, and credit factors.
“[W]ith the equity allocation at just 30%, the risk contribution from equity averaged about 70% throughout the period,” they said. They also noted that factor risk contributions fluctuated significantly. In one significant instance, a bull market had pushed the equity risk contribution down to a decade-low level of 40% by the start of 2018, while the interest-rate factor risk contribution rose; but in the aftermath of the February 2018 market sell-off, the equity factor risk contribution shot up to 80%.
Amato and Zhou then created a second portfolio, this time setting allocations based on a “target risk contribution” from each factor. Factor risk contributions add up to total portfolio risk, which allows for a “risk budget” — rather than a capital budget — to be assigned to each factor.
“Intuitively, if investors expected that factor returns justified the amount of risk each factor contributes to the total portfolio risk, they could have inversely budgeted the risk contributions based on their expected factor returns,” they said. Assuming that expected returns remain relatively stable, an investor could aim to maintain a stable risk-contribution profile.
Following the factor-based approach over the same time period as the holdings-based portfolio, the two allowed asset weights to fluctuate in response to a changing factor regime, thus achieving a stable risk-contribution profile. They acknowledged that dynamic portfolio composition could lead to a markedly higher portfolio turnover rate, adding that certain constraints could be put in place to mitigate such an effect.
“When it comes to risk and returns, the results were mixed,” they said. While the factor-based allocation displayed more consistent total risk over the period, each portfolio outperformed in different periods, with the factor-based approach achieving higher risk-adjusted returns in 2013 and later 2017, while the holdings-based approach outperformed in 2015 and 2016.
“The holdings-based approach was effectively a buy-losers, sell-winners strategy, which was advantageous when returns exhibited mean reversion — as the equity market did in 2016,” they explained. “The factor-based approach, on the other hand, allocated more weight to an asset when the relevant factor volatility decreased. It outperformed while the low volatility lasted, as happened during late 2017’s persistent equity bull market.”
