Safe Withdrawal Rates with a different measure of “inflation”

In a previous article I floated the idea that using CPI-U to measure “keeping up” during retirement might be the wrong thing, especially for (extremely) early retirees that are contemplating a retirement of half a century.

(I have another future article in mind on other problems with using CPI-U in backtesting retirement outcomes, as well.)

If we wanted to use something other than CPI-U…what would we use?

Let’s remind ourselves of why we want an alternative to CPI-U: we want to “keep up” with the expenditures of the average American. Yes, that is lifestyle inflation in a sense. It means getting a mobile phone, even though you retired in 1985. It means getting an automobile, even though you retired in 1920. It means we maintain our general socioeconomic class during that half century of retirement.

Instead of CPI-U we’re going to use the US Consumer Bundle instead. The US Consumer Bundle is easy to understand: “The value of the consumer bundle is the average annual expenditures of consumer units. Expenditures are for goods and services; also included are gifts and charitable contributions, as well as insurance premiums and pension contributions. The value of the consumer bundle is expressed in dollars; it is not corrected for inflation.”

This is all the money (directly) spent by consumers. Now that we know the average amount spent every year, we can calculate the Maximum Safe Withdrawal Rate necessary to “stay average”.

Here the lowest ones, assuming a 40-year retirement and a 60/40 portfolio…

1937    0.024753
1939 0.028628
1936 0.029474
1940 0.030197
1934 0.031209
1906 0.033219
1916 0.033839
1946 0.034076
1902 0.034194
1941 0.034411
1938 0.034597
1910 0.034746
1909 0.034766
1903 0.034949
1905 0.035047
1965 0.035230
1966 0.035264
1901 0.035275
1962 0.035421
1907 0.035667
1912 0.035905
1964 0.036297
1935 0.036600
1911 0.036728
1913 0.037021
1968 0.037517
1945 0.037777
1929 0.037790
1969 0.038305
1917 0.038365

Two things are noticeable:

  • The lowest numbers are substantially lower than the 3.25% or 4% (depending on what exact length of retirement you choose) you would get using CPI-U.
  • The terrible 1960s, while still not great, is nowhere to seen among the “worst” years. Instead the worst time period appears to be the mid-1930s to the late-1970s.

Remember this was a period of dramatically expanding American wealth. All of our stereotypes of idyllic American life come from this time period. Do you really want to be living a Depression-era lifestyle while literally is changing to something a lot closer to what we’d recognise as a modern lifestyle?

In 1945, as the war came to a close, inflation was 2.2%…but the Consumer Bundle went up 9.4%. The next year inflation was 18% but the Consumer Bundle went up 27%. The gap isn’t always that large but it is always there. After all, that’s what a “rising standard of living” means.

I think this is another reason to question the value of bonds in the portfolio of an early retiree. Yes, you might have volatility. Learn to live with it. The alternative is to risk a gradually reduced standard of living as your bonds erode over decades.

The lowest SWR with a 60/40 portfolio is 2.4% (in 1937). There are 2other years where the SWR is under 3%. The median SWR is just 3.9% and the 10% percentile SWR is 3.4%.

On the other hand, with a 100% equity portfolio things change quite a bit. The lowest SWR is 2.8% (0.4% higher). The median SWR is 4.7% (0.8% higher). The 10th percentile SWR is 3.7% (0.3% higher).

In every measurable way, 100% equities appear do better than anything else when we’re using a Consumer Bundle instead of CPI-U.

But it is also another reason to have reservations about Safe Withdrawal Rate strategies when thinking about retirement. Should a 1937 retiree really use a 2.8% withdrawal rate? This is what constant withdrawals look like compared to variable withdrawals:

Image for post
Image for post

Using constant withdrawals makes you cut spending preemptively. In this case, you cut spending for 20+ years, often by tens of thousands of dollars, just to save yourself from spending cuts in the last four or five years of retirement. (A time you may not even reach; it is likely you will die before then.)

Those SWRs from above — 2.4% for a 60/40 portfolio or 2.8% for a 100/0 portfolio — assume you spend it all by the end. But a lot of early retirees have dreams of maintaining their portfolio. That it is 50% or even 100% of the original value when they die.

But those are expensive things to do.

If you have a 60/40 portfolio and want a final portfolio value of 50% of the original (instead of 0%), the SWR drops from 2.4% to 1.7%. And if you want to maintain 100% of the original, the SWR drops from 2.4% to 1.05%.

If you have a 100/0 portfolio and want to maintain 50% of the portfolio, the SWR drops from 2.8% to 2.4%. To maintain 100% it drops from 2.8% to 2.06%.

Maintaining your portfolio value across half a century, while keeping up with a society-wide standard of living, appears to be an expensive luxury.

Written by

Learn how to enjoy early retirement in Vietnam. With charts and graphs.

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store