Can pursuing a dynamic portfolio strategy focusing on income replacement in retirement do better than some of the more traditional approaches, such as purchasing age-dependent target-date funds or sticking with a 60/40 equity/bond asset allocation?

The authors test different rule-based investment strategies using historical data and various simulations of how asset prices might move in the future.
The authors used historical data for the four-decade period from 1977 to 2017 for the following: monthly returns for the S&P 500 Index, the US Treasury yield curve, and US inflation as measured by the Consumer Price Index. The backtesting analyzed 10-year periods within the data.

The investment strategies compared included the following:

• A static portfolio of 60% equities and 40% bonds
• A static portfolio of 60% equities and 40% so-called SeLFIES (Standard-of-Living indexed, Forward-starting Income-only Securities) that promise a specified real, or inflation-adjusted, level of annual income for a predetermined period of time when retirement commences
• A dynamic target-date fund that allocates more to bonds as the individual gets older: The assumption is that the holdings are allocated 90% to equities initially. Over time, 0.5% of the portfolio gets switched to bonds each month during the decade analyzed. That should result in a 30% allocation of equities by the end of the period.
• A dynamic rule-based strategy known as GLIDeS (Goal-based, Lifetime Income–focused, Dynamic Strategy): This strategy involves a portfolio consisting of equities and SeLFIES. The level of equities held is dynamically altered according to the estimated funding status of the portfolio. If the portfolio assets could provide the desired income replacement in retirement, then the equity allocation is kept lower. If the assets are insufficient to reach the income goal, then the equity exposure is increased. The authors created two such GLIDeS portfolios with strategies that had aggressive and moderate goals for 100% and 70% replacement income, respectively.

The authors also used simulations to model the performance of the various strategies. In the base case, the authors assumed annual inflation of 2.5%, nominal annual equity returns of 6.6%, and a yield curve with interest rates of 1.5% for short-dated securities and 2.7% for long-dated ones. The simulations were run assuming a future 40-year time horizon.

The strategies tested using the simulation were similar to those previously discussed but not identical. They included the following:

• A static portfolio of 60% equities and 40% bonds
• Target-date funds with the percentage of equity decreasing over the time during which the assets would be invested
• A static portfolio of 60% equities and 40% SeLFIES
• Target-date funds using SeLFIES instead of bonds
• The GLIDeS approach for three separate goals of replacement incomes—50%, 70%, and 100%—corresponding to conservative, moderate, and aggressive GLIDeS goals, respectively

The authors also used simulations to test how the same strategies would perform in periods of ultra-low inflation and also high inflation.

In both the historical and the Monte Carlo simulations, the GLIDeS approach significantly outperformed the other strategies both in returns and risk. The authors note that in some instances, static strategies and target-date funds outperformed, but these were anomalies, which were the result of “chance and not by design.”

The alternative scenarios looking at the two dire economic outlooks of high inflation and low-to-zero inflation showed that investors would need to contribute more savings toward their retirement plans under any of the strategies tested. However, even under these extreme economic scenarios, the GLIDeS strategies would require lower savings than the others for investors to achieve the desired income replacement goals.

Of all the investment strategies analyzed, the use of a dynamic portfolio combining equities and SeLFIES as a safe asset with the goal of achieving a predetermined level of income replacement was the most successful. In other words, the returns gained were far higher than those of the other strategies, such as the age-related target-date portfolio or the portfolio with a fixed allocation of equities and SeLFIES.

This strategy can also be used to determine what annual savings are needed to achieve a given level of retirement income. The dynamic strategy required the lowest level of annual savings, compared with the alternatives.