The use and abuse of “the 4% rule”

I had a good time recently appearing on the Saverocity Observation Deck podcast with Noah from Money Metagame and host Joe Cheung.

One thing that I think got a little muddled in our conversation was my feelings about “the 4% rule,” which is much beloved by financial independence and early retirement enthusiasts. Rather than try to tell you I’m right and they’re wrong about financing early retirement, I think it might be more useful to isolate a few different factors that go into planning a stream of retirement income.

I don’t have anything against “the 4% rule” — as it actually exists

Back in the 90’s some erudite finance scholars got to work on a historical asset price dataset and determined that in the overwhelmingly majority of cases a retiree could withdraw 4% of their starting portfolio value as of their retirement date, adjust the withdrawal for consumer price inflation annually, and still have a positive portfolio value at the end of a retirement period ending 30 years later.

There are two key elements of this analysis to keep in mind: it’s based on historical data, and it describes a retirement period that ends 30 years after it begins.

To adopt such a rule to finance your retirement, you would therefore have to make two assumptions: the future will be like the past, and your retirement period will end 30 years after it begins.

I don’t have any clue how members of the early retirement community arrive at the conclusion that those two conditions are true. The first is a violation of the obligatory maxim that “past performance is not a guarantee of future results,” and the second is absurd in light of the very premise of early retirement: having more, in some cases many more, years of retirement than the 30 that traditional retirees have had (if they get lucky).

An important aside about Social Security retirement income

Most people are aware of the importance of adjusting prices and income for inflation: the general tendency under conditions of growth for prices to rise over time. A $1 avocado today doesn’t cost the same as a $1 avocado in 1970: it costs much, much less, since a dollar is worth much less in 2017 than it was worth in 1970 (no, I don’t have any idea how much an avocado cost in 1970).

That means to maintain the same absolute standard of living, retirement income needs to be adjusted for consumer price inflation, so that in retirement you’ll be able to afford the same standard of living you had upon retiring.

The Social Security Administration made the important observation that indexing retirement benefits to consumer price inflation was inadequate: a 95-year-old retiree with benefits indexed to consumer price inflation would today be able to afford the same standard of living as she had in 1987, at age 65. In other words, she wouldn’t be able to afford any of the conveniences of modern life, none of which existed in 1987!

Instead, Social Security retirement benefits are indexed to wage inflation, so that retirees don’t fall behind as productivity and output rises. Their benefits keep up with the rise in technology and productivity that the US economy provides.

An early retirement rule that successfully provides consumer price inflation-adjusted income will immiserate you in old age

The reason the above aside was necessary is that over a retirement stretching not 30 years, but 60 or more years, a rule that provides consumer price inflation-adjusted income will not maintain your relative standard of living. A simple way to illustrate this is using the Social Security Administration’s wage deflator. Earnings in 1990, just before a decade-long rise in productivity and earnings, are indexed at a rate of 2.29: each dollar earned in 1990 is worth $2.29 towards Social Security retirement benefit calculations in 2016.

Meanwhile, a dollar in 1990 was worth just $1.84 in 2016 according to the consumer price index. The difference? Wage growth tracks productivity growth, which usually exceeds consumer price inflation in a growing economy. Between 1990 and 2016, wage growth exceeded consumer price inflation by 24.5%. That means across the economy at large, wage-earners became 24.5% wealthier than those who successfully tracked consumer price inflation with their investment decisions.

At age 35, or 40, or 45, are you willing to lock in your absolute standard of living today? Are you willing to forego flying cars, neural implants, and vacations to Mars? The mass market innovations of tomorrow will remain affordable to people who participate in productivity growth; people whose income merely tracks consumer price inflation will fall further and further behind.

Equity investing gives you a chance to participate in productivity growth

Of course, the easiest way to participate in wage-inflation-adjusted standards of living is to earn a wage. But earning a wage, for obvious reasons, doesn’t feel like retirement, let alone early retirement.

The other way to participate in national, and even global, increases in productivity is to invest in the companies involved. Today, it’s fortunately possible to do so through low-cost indexed investment vehicles. While workers might participate in productivity growth both through increased wage income and the increased value of their investments, one out of two ain’t bad: owning a broadly diversified portfolio of stocks will at least allow you participate in one side of the global growth in output.

Across a broadly diversified portfolio, the dividends that companies pay out to shareholders will rise along with profits, which reflect the companies’ participation in overall economic growth. Think of this as the flip side of Henry Ford’s desire for his factory workers to be able to afford Model T’s: the more Model T’s the workers can afford, the fewer his shareholders can afford. While individual firms strike that balance in different ways (“This is frustrating. Labor is being paid first again. Shareholders get leftovers.”), across the entire economy all profits will be split between labor and capital.

A viable plan for early retirement requires goals, not rules

Of course, as Noah, Joe and I get into during the podcast (which I really do recommend if you want to hear us discuss these and other issues in more detail), many early retirement types also belong to a kind of loosely-organized ascetic community that treats as extravagant luxuries purchases which most Americans treat as staples. It’s more than possible that such people experience price inflation lower than the consumer price inflation which the federal government records.

But that’s a question that actually has to be answered before developing a plan for retiring radically early. A plan to simply save up 25 times your 2016 expenses and withdraw 4% per year, adjusted for consumer price inflation, will leave you relatively destitute in 30 years even if your early retirement goes exactly according to plan.