What is a safe withdrawal rate in retirement?

Personal finance personality Dave Ramsey recently caused a stir in the financial planning community by claiming that clients should be able to withdraw 8% of their portfolios in the first year of retirement and adjust for inflation each year thereafter without being at risk of depleting their investments.

This bold claim flies in the face of more commonly cited retirement spending rules, like the 4% rule. Safe withdrawal rates — and the 4% rule in particular — have lots of problems, but they are commonly used by investors and recommended in many popular personal finance books. So, it’s worth addressing Ramsey’s claims.

Financial planner William Bengen ignited the safe withdrawal rate literature in 1994 when he wrote the research paper “Determining Withdrawal Rates Using Historical Data.” The result of Bengen’s research was a safe spending rate for a 30-year period, which was determined to be about 4% in his data. This finding was based on historical data for U.S. stocks and intermediate term Treasuries.

Ramsey’s logic is that “good” mutual funds return around 12% per year, while U.S. inflation has averaged around 4% for the last 80 years. This leaves 8% for the retiree to spend. This logic is flawed; retirement spending math simply does not work that way.

To start, “good” mutual funds are hard to find before the fact, so the idea that you can easily pick a market-beating fund is unsupported. The best performing funds historically do not tend to go on to be the best performing funds in the future.

The Standard and Poor’s SPIVA U.S. Persistence Scorecard illustrates the point. With a starting sample of 527 top-quartile funds, exactly 0% of them remained top quartile over five years. Similar data are available for Canada, though the sample is much smaller.

The case is similar with country returns. Ramsey references the returns of the S&P 500 as being a little below 12%, but, again, the phenomenal historical performance of the S&P 500 does not tell us much about expected returns.

The 2023 paper “Is The United States A Lucky Survivor: A Hierarchical Bayesian Approach” finds that the realized historical risk premium on U.S. stocks exceeds the expected premium by 2%. This premium is approximately equally split between contributions from luck, where cash flows ended up being higher than expected due to disasters that did not materialize, and learning, where investors lowered their required return on U.S. stocks over time as catastrophes did not happen, driving up the stocks’ valuations.

With U.S. valuations where they are now, there is a strong argument that expected returns for U.S. stocks are far below their historical returns. Earning high returns in the future is not as easy as picking the best performing country of the past.

Since we can’t pick winning funds or winning countries before the fact, a more sensible approach to planning for retirement is using the return experiences of countries around the world to gain an understanding of possible retirement outcomes.

This has been done in at least two papers: “The Safe Withdrawal Rate: Evidence from a Broad Sample of Developed Markets” (2023) and “An International Perspective on Safe Withdrawal Rates from Retirement Savings: The Demise of the 4% Rule?” (2010). The broad conclusion of both is that even 4% is too high to be safe for most retirees; a number around 3% — or lower — could be considered safe.

The other problem with Ramsey’s 8% spending claim is that even if we could find a fund that returns 12% on average, it would not sustain an 8% withdrawal rate because stock returns are volatile. A 12% average return will consist of big ups and downs, and inflation similarly goes through higher and lower periods.

Constant inflation-adjusted spending during consecutive down years, or years of high inflation, can deplete a portfolio quickly. A mutual fund that returned about 12% per year since 1935, the American Funds Investment Company of America, helps make the point.

The fund returned an impressive 11.73% annualized from 1934 through October 2023. Using its historical performance to test an 8% withdrawal in the first year followed by annual inflation adjustments over a 30-year period — a constant real $80,000 in spending — yields a success rate of a little more than 50%. This means that in around half of simulated trials, the client ran out of money before the end of the 30-year withdrawal period — not exactly safe.

At a more palatable 5% failure rate, the safe withdrawal rate for this fund would be about 4.6%, slightly higher than the 4% finding from Bengen, which makes sense since we are using a fund that we know, after the fact, has slightly outperformed the U.S. market.

In any case, the reality is that most clients will be best served by some form of variable spending strategy that responds to year-to-year changes in portfolio values and expected returns, rather than the type of constant inflation-adjusted spending suggested by safe withdrawal rate research.

For clients who may have heard Ramsey argue that 8% is a safe spending rate for retirees, the answer is, in no uncertain terms, no, it’s not.