I wonder how many people who retired in Jan-Sept 21 (so those who have some time since quitting) have blindly continued to withdraw their inflation-adjusted 4%, exactly on schedule and without cuts, or bringing in additional income, or anything else.
Whether or not it proves to have been the "worst time to retire" or nearly worst, that's what really matters as far as individuals heading toward failure or not.
If DH and I had retired in that time frame, we'd likely be picking up a few thousand dollars (or perhaps much, much more) in additional income, and perhaps scaling down some plans.
I’ve yet to encounter anyone who has blindly followed a 4% withdraw strategy for thirty years with annual inflation increases, without earning any money, or taking more or less on a given year, or benefiting from a larger than expected windfall.
...
I hear this counter-argument every time and it completely misses the point. The goal of FI (measured by achieving your own SWR) is to have the freedom and independence to live within that SWR indefinitely. This gives you the freedom to never work again, or work in whatever way you see fit regardless of the income.
So while it may be true that folks can tighten their belt or go back to work, it misses the point that you make a decision about having achieved FI and it turns out that it wasn't FI. That is my definition of 'failure', not necessarily having zero dollars. During 2008/9, there was a lot of failure in the fact that many who had declared FIRE in 2007 suddenly found themselves worried about running out of money, changing their lifestyle, and questioning if their ER would last... Certainly seemed like a less 'free' lifestyle than working a little bit longer. The money that I stashed during 2008/9 in my 401k (as well as having peace of mind during that time) got me from quasi-FI in 2007 to full FI in 2009. Full FI would have been delayed significantly and some folks never get back to their FI number.
So yes, these models define 'failure' as running out of money and humans do not behave like models, but we also need to stop saying that we can just disregard the results because we'd never run out of money IRL. Running out of FI happens long before running out of money and is the thing we are actually trying to avoid when we run these models.
That's not the point though.
The point is that everyone knows that the 4% rule is a starting point, a rough guideline. People who can't easily cut their budget or can't easily being in more income should absolutely have other SORR contingencies.
The point is that, yes, the modeling is great for showing you why you should have contingencies, but it's not likely to indicate anything about actual, real life failure rates.
It comes back to the entire purpose of the modeling, which is that it helps you make decisions in the present based on a flawed but reasonable model of the potential future, it does not predict the future.
Pointing out what people do in reality doesn't invalidate the models, nor make them less useful, but it is important to recognize what they are actually useful for.
The main benefit of the models is to compare options by seeing the relative impact of variables.
So it's essentially meaningless saying that the theoretical retiree who saved exactly 25X their expenses and then quit, never modified their spending, and always spent 4% +estimated inflation, down to the penny, year after year would run out of money by X year.
But it's very useful to compare SORR mitigation strategies: saving to a lower steady WR, vs having a variable WR, having a lower WR and making it variable vs periodically earning some income, vs bond tent, vs cash reserve, which strategy or combination of strategies comes out ahead in an early retirement period like this??
There's no question people shouldn't treat the "safe withdrawal rate" as the "failsafe withdrawal rate," but I think that's been so well established it's just a given by this point.
I would love to see a bunch of analysis of which hedging approach is most effective in a period where the SORR risk is mostly driven by inflation.
Basically, if someone had retired in 21-22, what SORR plan would have worked out best?