“You Are Too Impressed By the Small Number of Times That a High-Risk Withdrawal Rate Has Survived.”

Set forth below is the text of a comment that I recently posted to the discussion thread for another blog entry at this site:

No 30 year period has ever failed, Rob.  You can’t have it both ways.  You can’t say on one hand to look at the historical market and then say just the opposite when it doesn’t suit you.  If you doubt Greaney, you have to doubt Shiller……..and that means you are also wrong.

I agree that a 4 percent withdrawal has not failed in any 30-year time-period in the historical record.Yes, that’s history.

But, no, that doesn’t make a 4 percent withdrawal safe for a retirement beginning today. Throughout history, the most important factor affecting safety has always been the valuation level that applies on the day the retirement begins. Someone who believes that history is a good guide to what will happen in the future would say that any analytically valid calculation of the safe withdrawal rate must include consideration of the valuation level that applies on the day the retirement begins. Given that it always was a factor in the past, why wouldn’t it continue to be a factor in the future?

You are acting as if the concept of being safe and the concept of having survived are the same thing. They’re not.

Now — If the CAPE level had been 44 in every year of our stock market history from 1870 until today and 4 percent had always survived, then I would agree that it would be fair to say that 4 percent had shown itself to be safe. That’s a lot of data points. But that’s not the situation. We have only had one month when the CAPE hit that level.One data point is not enough to show that that withdrawal rate is safe at that valuation level. And that withdrawal rate barely survived on the earlier occasions when it was used starting from a valuation level that was super high. That shows that the valuation level matters when it comes to safety and should never be ignored.

You are too impressed by the small number of times that a high-risk withdrawal rate has survived. Saying that something is high-risk is not saying that it is certain to fail. It is saying that there is a chance that it might succeed and a chance that it might fail. There was a good chance that a 4 percent withdrawal for a retirement that began in 1929 would fail. It happened to succeed. That fact doesn’t retroactively make that withdrawal rate safe. It was high-risk in 1929 and it was still high-risk in 1959, when it had survived 30 years. Sometimes a high-risk withdrawal rate will survive. But a high-risk withdrawal rate will never retroactively become safe. Safety has to be determined at the time the retirement begins. In 1929, that withdrawal rate was high-risk because it was not known at that time whether the returns sequence that would come up would be a lucky one or an unlucky one.

All that Greaney showed is that it is POSSIBLE that a 4 percent withdrawal rate will work out starting from a super high valuation level. It is possible. The historical data shows that. It does not show that it is safe. It shows that it is high-risk.

Rob