A Dynamic Strategy that Outperforms Vanguard’s “LifeStrategy” Fund
Advisor Perspectives welcomes guest contributions. The views presented here do not necessarily represent those of Advisor Perspectives.
The founder of Vanguard, Jack Bogle, says that over the next decade a conservative portfolio of bonds will only return about 3% a year and stocks about 4% a year. However, returns can be improved with a dynamic asset-allocation strategy that adjusts stock- and bond-fund holdings in a retirement account according to the market climate.
I compare the performance of the Vanguard LifeStrategy Moderate Growth Fund (VSMGX), which holds static investments of 60% equity and 40% bond funds, with a model holding identical assets, but which switches to 100% bond funds during equity down-market periods. I call this the Dynamic MAC 60:40 Model.
For a $1 million initial VSMGX investment in January 1999, one could have withdrawn on average a real (inflation-adjusted) $2,420 per month and still keep the original investment intact to November 2017. The monthly withdrawal amount can be more than doubled by adjusting the investment according to market climate, as indicated by a 50-200 day moving-average-crossover system of the S&P 500. This would have allowed an average withdrawal of a real $5,480 per month for almost 19 years, while still retaining a real $1,000,000 in the retirement account at the end.
I backtested my model using the on-line simulation platform Portfolio 123, which also provides extended price data for ETFs prior to their inception dates, and economic and financial data. The model assumptions are:
- Slippage1 and trading costs were ignored, and closing prices were assumed for the simulation.
- Trades occur on the first business day of a week based on signals from the market climate indicator.
- A 40-day minimum holding period for a position was specified so that mutual funds can also be used instead of ETFs.
- Rebalancing to nominal allocation weighting was assumed to occur every 52 weeks, at the end of a calendar year. Rebalancing transactions, which amount to less than 5% of portfolio value, are excluded.
The start date of the backtest was 1/2/1999, because this is the earliest start date for a simulation on Portfolio 123.