Year 2000 retirement: a 4% SWR reality check

Almost exactly a year ago I wrote a brief thing on Year 2000 retirees.

They faced two horrible crashes, in 2000 and 2008, at the start of their retirement. In 2010 Wade Pfau wrote that Year 2000-retirees were (at that point) on track for the worst retirement in US history.

Pfau was far from alone in his feelings. EarlyRetirementNow (ERN) has more recently written a detailed series on Safe Withdrawal Rates. One of the entries is on a Year 2000 retirement.

After looking at the current state of the portfolio of a Year 2000 retiree, he concludes:

If anything, the 2000–2016 episode was a worst-case scenario for early retirees. Quite the opposite of the “4% rule did OK” myth.

I think ERN has come to the wrong conclusion and I think he did it because he didn’t look deeply enough at the in-progress reality of other retirements. I’ve seen other people make a similar mistake: they believe that if a $1,000,000 portfolio has dropped to $600,000 after a decade then that means the 4% rule is doomed.

I think they reach that conclusion because they only really think about their portfolio after 16 years. They’ve never really looked at what other historical portfolios looked like after 16 years. And the reality is that plenty of retirements featured portfolios dropping quite low in value and then later rebounding.

That’s one of the features & flaws of safe withdrawal rates compared to variable withdrawal strategies. They are counting on the market rebounding. They are counting on that reversion to mean.

And when we look at a Year 2000 retirement portfolio…it definitely doesn’t look like “a worst-case scenario for retirees”.

This is what portfolios looked like at the year 16 mark for all retirements.

This is an apples-to-apples comparison. A Year 2000 retirement portfolio would have dropped to $638,000 inflation-adjusted dollars. It certainly isn’t all clear sailing. But we also see that plenty of other portfolios have dropped even lower…but eventually recovered enough that the 4% rule remained intact.

In fact, the Year 2000 retirement is “only” 22nd percentile. Sure, 78% of retirements were doing better at this point…but 22% were doing worse.

Here’s a snapshot of all the retirements that were doing worse than a Year 2000 retirement and their portfolio value after 16 years.

Some of them were doing way worse.

Things are complicated because when we do this kind of backtesting we almost always use inflation-adjusted dollars. But people rarely think that way in the real world. That 1906 portfolio with only $263,679 inflation-adjusted dollars? It actually had $591,633 nominal dollars. If you had started with $1,000,000 and still had $591,633 left you probably feel reasonably okay. At least, you’d feel a heckuva lot better than if your brokerage statement only said $263,679.

When we look at nominal portfolio figures, the Year 2000 — with its $890,398 — looks a bit worse. It drops from 22nd percentile to 15th percentile.

There are still plenty of years worse off.

As I said, this doesn’t mean the Year 2000 retirement is guaranteed to be fine. People are worried about low bond yields and high valuations. You could make an argument that future returns are muted and a Year 2000 retiree will never see the reversion to mean that saved those other portfolios. Another crash in the next year or two could change things dramatically. On the other hand…people have been saying that for five or six years at this point. Every year that goes by means the Year 2000 retiree can breathe a little easier.

Perhaps one reason why I disagree with ERN is that I think he’s mixing two arguments into one. On the one hand, he’s talking about the 4% rule for a Year 2000 retiree. As we’ve seen, there’s nothing especially notable about them. On the other hand, he’s also pushing back on claims from some in the early retirement community that a 4% constant dollar withdrawal rate is suitable for early retirement. I agree with him that it is not…but I don’t think you need to look at the Year 2000 to find counterexamples.

The retirement research literature has known for years that many portfolios from the late 1960s were exhausted (or nearly exhausted) after 30 or 32 years. They clearly couldn’t support a 4% safe withdrawal rate for 40 or 50 years, as early retirement would require.

I think another confounding factor is that when we make the shift from academic research to the real world our concept about safe withdrawal rates undergoes a subtle shift. The academic definition is just that the portfolio lasts 30 years. It doesn’t make any claims about the portfolio value at various points along the way. As we’ve seen, that often means some pretty steep drawdowns. In the real world, I think our definition is something more like “the portfolio never drops below 70% of its original value in the first 15 or 20 years of my retirement and then never drops below 40% of its value for the rest of my life”. Or something like that. I think people want that margin of safety.

What would the Super Safe Withdrawal Rate look like? The withdrawal rate required to ensure you always had at least, say, 50% of your portfolio left at any given time?

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