Would you challenge a financial adviser who put 90% of their 70-year-old client’s retirement portfolio in shares and the balance in government bonds?
If the answer is yes, be prepared for the rebuttal that the client is merely following legendary investor Warren Buffett’s advice.
In a letter from 2013, Buffett shared with a trustee his advice for the money he intends to leave his wife: ‘Put 10% of the cash in short-term government bonds and 90% in a very low-cost S&P 500 index fund.’
For comparison, the weightings in Vanguard’s – Buffett’s suggested provider – Target Retirement 2015 fund are currently 53% in bonds and 47% in stocks. This product is designed for those planning to retire between 2013 and 2017; Buffet’s wife turns 70 this year.
Put to the test
Could Buffett’s proposed allocation be suitable for retirees, though?
Javier Estrada, of the IESE Business School in Barcelona, tested the portfolio in a paper published in July.
He noted that retirement portfolios heavily skewed to shares have traditionally been viewed as risky because of their volatility. But for retirees, there are perhaps two greater risks: that the investor will outlive the portfolio; and that the portfolio’s value may not be maximized if passing on wealth is the objective.
Estrada drew his data from the Dimson, Marsh, and Staunton database of global investment returns, based on inflation-adjusted total returns. He studied rolling 30-year periods between 1900 and 2014, yielding a total of 86 overlapping periods from 1900 to 1929 and 1985 to 2014. Portfolios were rebalanced at the beginning of each year, and Estrada applied an initial withdrawal rate of 4% to reflect drawdown that was subsequently adjusted for inflation.
How did the 90/10 portfolio perform? For US stocks and bonds, it failed – meaning the portfolio was exhausted before the end of the 30 years – in 2.3% of the periods under review. A portfolio of 60% equities and 40% bonds did not fail in any of the periods, but the 90/10 allocation did have a lower failure rate than other weightings.
Turning to value maximisation, the 90/10 portfolio was more comprehensive. Its average return during the review period saw $1,000 grow to $2,638. In the worst 10% of periods, the $1,000 fell to an average of $219; in the best 10%, it became $6,695.
For the 60/40 portfolio, the average return was $1,267. In the worst 10% of periods, the $1,000 dropped to an average of $204; in the best 10%, it turned into $2,647.
Estrada also tweaked some of the inputs to test different scenarios. By employing a more dynamic rebalancing mechanism – not selling stocks in down markets – the 90/10 portfolio’s failure rate crept up to 3.5% but average and extreme returns were boosted substantially. And when Estrada reduced the drawdown rate from 4% to 3%, the 90/10 portfolio’s failure rate dropped to zero.
‘The evidence shows that an aggressive allocation does not necessarily have to be risky; it all depends on how risk is defined,’ Estrada said.