Follow up -- here is how I apply the thinking in my last post to an early retirement.
As I stated, I believe that the CPI slightly overstates inflation.
So I assume that I will "keep up" with the Jones' in early retirement, by keeping non-medical spending during early retirement constant with respect to CPI, and I track medical spending separately at higher inflation component.
This approach gives a fudge factor, because the CPI would be much lower without the medical component (well, about 0.5% lower, if memory serves), and if you believe like I do that the main CPI figure already overstates inflation, then this allows you to keep up with the neighbors, so to speak, whose standard of living is increasing. As I mentioned above, even if you are frugal LBYM, there are certain increasing expenses you cannot (nor do you want to) avoid in a society whose standard of living is increasing like the US. I figure this gives me at least half percent margin per year over a true fixed standard of living -- this is 16% after 30 years -- and maybe up to 1% per year -- this is 35% after 30 years.
On the medical part, I separate it out and increase it by about CPI + 8% or so until age 66. This accounts for my increasing age in retirement, as well. And I assume some extra costs over present-day medicare in retirment, more than someone would pay nowadays. (of course, if you get sick, your travel costs may decrease, etc.)
I have an escape valve in that if medical gets too high, I am outta here (USA).
So to summarize, for SWR, I just assume all expenses increasing at CPI, except medical insurance, which I assume at about CPI + 8%.
When I did all the math in a monte carlo engine for mid-40s retiree, this dropped SWR by about 0.4%.
If you have a larger margin at retirement and thus health costs are lower percentage of your initial withdrawal, you have much less to worry about than a barely marginal retiree, who will be affected disproportionately by medical inflation.
Kramer
ps: I am reading Robert Fogel's treatise (he is nobel laureate) on standard of living -- Escape from hunger and premature death: 1700-2100 -- and you can expect medical costs to continue to increase, not because of inflation, but because more and more treatments are coming on-line and this is a rational way to spend money. in fact, just today, annual government study came out and reported astonishing decrease in number of US deaths last year. life expectancy increased by 0.5 years in last 2 years. btw, at turn of 19th century, folks spent 1% of income on medical care. we live in an amazing time of astonishing equality -- used to be more than 10 years difference in lifespan between upper and middle classes.