A Better “Typical” Retirement Spending Model — Useful Finds #18

Hi from Stefan,

Welcome to the eighteenth issue of this biweekly newsletter.

Nothing here is sponsored or written for compensation.

Product update

I have revised the model for the “Typical” retirement income pattern, based on the latest research.

The earlier version was based on David Blanchett’s widely-cited 2014 paper that compared the average spending trajectory to a “smile” — spending declines in early to mid retirement, and then increases toward the end of life with greater health care needs.

Blanchett’s new research finds that the familiar late-life upturn largely disappears when looking at the median retiree rather than the average across retirees.

A minority of households incur very large late-life healthcare expenses, pulling up the average, while the median household’s real spending continues to decline. I’ve therefore revised The Best Third’s Typical spending pattern to reflect the median experience.

Bear in mind, though, that any individual household’s spending pattern is likely to vary considerably from the median, and vary from year to year.

If you ran a scenario with a Typical spending pattern in the past, you might want to rerun the scenario to see what it looks like under the revised assumptions.

Also, Fidelity has unlocked its platform to enable account linking through Plaid. If you have Fidelity accounts and would like the convenience of automatic balance updates and TIPS imports, visit https://app.thebestthird.com/linked after you login. Account-linking is read-only.

Retirement Tax Strategies

The Wall Street Journal on smart tax-savings strategies for retirees.

But with high uncertainty about both future tax rates and your future income, these strategies are something of a guessing game.

Answer to a reader question

Both Sanjay and David asked about a suitable amortization rate for a global, as opposed to U.S.-only, equity asset, wondering whether the lower historical long-term returns on global stocks would suggest a lower amortization rate.

My short answer: I don’t think investing globally, by itself, is a good reason to lower the amortization rate.

My research paper upon which the The Best Third equity withdrawal model is based looked at U.S.-only historical equity performance due to limited availability of historical international equity returns. But it is entirely reasonable for an investor to want global diversification.

The ideal amortization rate would be the future average real return on the underlying asset.

But that is unknowable, so The Best Third instead uses the long-term (1871-near present) geometric annualized mean real return on U.S. equities, which is 6.9%.

I’m not aware of any truly global historical index which goes back nearly that far, but here are some interesting data points:

For 1900-2020, Dimson Marsh and Staunton found that the cumulative real return on U.S. equities was 6.6%, while a multi-country international basket returned 4.5%. The U.S. index also had one of the lowest volatilities of any of the countries they looked at.

In a more recent period for which I have access to more complete data, 1993-present, $1 invested in the Russell 3000 in 1993 would today be worth $12.79 (real annual return of 8.0%, standard deviation of monthly returns 4.4%), while a similar $1 invested in the MSCI All World ex-US would today be worth $7.46 (real annual return of 3.6%, standard deviation of monthly returns 4.6%).

The correlation between the Russell and MSCI indexes over that period has been 0.84. The MSCI global index for that period had a 5.7% real annual return with standard deviation of 4.3%.

Of course, there have been multi-year periods where non-US stocks have out-performed US stocks. Historically, international diversification has not produced higher returns for U.S. investors over the long-term, and in the periods I’ve examined the reduction in portfolio volatility has been negligible.

Nobody can know whether a global stock portfolio will have higher risk-adjusted returns than a U.S. stock portfolio in the coming decades. There is some peace of mind in diversification in the sense of not putting all of your eggs in one basket. But if you believe that there is a good chance that a global portfolio will be a better investment than a U.S.-only portfolio, that suggests against using a lower amortization rate.

Ultimately, I wouldn’t worry too much about finding the “right” amortization rate.

The future average return is unknowable. A rate that’s too high shifts more withdrawals toward the earlier years; a rate that’s too low shifts them toward later years. But with any plausible rate, the amortization math protects the portfolio from premature depletion.

The Best Third’s 6.9% assumption is anchored in very long-term U.S. history rather than a forecast, which I think is more defensible than pretending I know what the next 30 years will bring.

I do plan at some point to add a feature that lets users set their own amortization rate for each scenario.

Warmly,
Stefan Sharkansky, PhD
Founder of The Best Third