The Firecalc assumption is that you want your lifestyle to remain the same, even if inflation goes haywire and in years following a portfolio decline.
The study linked above makes some major assumptions about your willingness (and ability) to cut your spending
permanently in the face of inflation.
"The maximum annual inflationary increase is 6 percent." This means that even if inflation is over 6%, you can't increase your withdrawal by the inflation to pay your now-increased costs of living.
Is that a real issue? Well, my crystal ball is no good, but looking back, there were 25 years when you'd have had to tighten your belt. Inflation as they define it was nearly 14% in 1980, 12% in 1981, and 8.5% in 1982, the worst recent years.
(Geez -- who cares about inflation when it's 1-2%? You can easily absorb that -- but not when it's high enough to really matter.)
"Withdrawals increase from year to year in accordance with the inflation rule, except that there is no increase following a year where the portfolio's total return is negative."
Wow. This means that you would have to tighten your belt and get no "pay raise" despite inflation in a LOT of years. Like 2002, 2001, 1995, 1991, 1988, 1982, 1978, 1975, 1974, 1970, 1967,... all these and 39 other years were ones where their rule would keep you from increasing your withdrawal at all to match inflation and keep your buying ability constant. (FWIW, the average inflation rates for those years I just mentioned was 5.8%.)
Let's look at what those two rules would have done to us in the early 1980s.
Say we needed $40,000 to live on, and followed their model when retiring in 1979.
Inflation in 1980 was 13.91%, but we can only increase spending 6%, so our shortfall means we are living on $40,000 * 94% or $37,600 in constant dollars -- i.e., this is our spending ability based on our spending as of when we retired.
In 1981 inflation was "down" to 11.83%, and we are still capped at a 6% adjustment. The arithmetic puts our spending ability at $35,344.
In 1982, inflation dropped all the way to 8.39% -- but the portfolio lost money, so rule 2 kicks in, and says we can't take
any adjustment for inflation. Yikes -- our spending ability for 1982 is just $32,378 that year.
Retired just 3 years and we are already having to live on nearly 20% less than we planned!
There haven't been 6%+ inflation years since 1982 (although we've had several 5%+ years), so the cap doesn't affect us after that year. But 5 of the next 20 years -- one year in four, rule 2 got us. Every time that rule 2 kicks in, we take a pay cut in the amount of that year's inflation rate.
Oh yeah -- and in their definitions of rules one and two, they say
"There is no 'make-up' for a 'capped' inflation adjustment. " and
"There is no 'make-up' for a missed increase."
So those "pay cuts" are
permanent -- we can't ever make them up.
10 years after retirement, we are living not on $40,000 in spending power, but on $31,000. The 1990s were good to us, so by 22 years after we retire, we're only down to living on $27103 -- with no increase in spending ability ever, according to the model.
This scenario is our most recent period, not some ancient example from days of the great depression. I question a model that in the most recent 22 years would have our standard of living permanently cut by 32%...
_________________________________________________
So sure -- a 6% withdrawal rate can be "safe" if you are willing to only adjust your withdrawals to match cost of living increases when (a) your portfolio is up and (b) the inflation was fairly low, and absorb permanent pay cuts one year in four with no chance of regaining lost ground...
Ouch!