so people are really complaining that are “rule” (4%) that has been backtested for 30 yr periods fails to perform in a 10 year time horizon picked in the worst possible moment? Well hallo, of course it will fail if tested outside design.
Let’s try to look at it more realistically. Let’s observe the period between 31. May 1994 and 20. Dec 2017. This period is given automatically by the MSCI index pdf file. MSCI USA and MSCI Switzerland (expressed in USD) are almost identical: MSCI USA returned 9.31% p.a., MSCI Switzerland expressed in USD returned 9.52 % p.a. (gross return incl dividends). So the two countries, expressed with the same currency, have an almost identical return. We can compare the two then, as they are highly correlated.
Total gross annual return MSCI Switzerland index (in CHF, including dividends, no inflation adjusted): 7.14 % p.a.
Total Switzerland inflation in the same period: 13.7 % over 23 years, roughly 0.5% p.a.
Total inflation adjusted return for MSCI Switerland over 23 year: 6.64 % p.a.
Now USA:
Total gross annual return MSCI USA index (in USD): 9.31% p.a.
Total USA inflation over the same period: 64.7% p.a., roughly 2.3% p.a.
So total USA annual return inflation adjusted: 7.0%
CH 6.64% vs USA 7.0% so a wash basically, or a slight advantage for US. Over time horizon of 20+ years currency exchange don’t really matter.
The problem of the chosen time period is critical. And of course, of course currency exchange add a layer of volatility on top of your stocks volatility! So it’s even better to invest for growth in USD since more volatility usually means more returns (and higher risk).
But nobody said that in retirement you need to have all the money you are going to spend in foreign currency. That would not be wise. So the drawdown strategy in retirement is forcibly different than the accumulating strategy.