Is the S&P 500 the best data set to base our 4% SWR on?

The 4% SWR rule is based, as far as I know, on a data set of the S&P 500 index from 1871 to 2006.

So, it means that if you retire and invest in stock of companies within that index, you should be OK with 4% SWR.

Now my rhetorical question is: isn't the S&P 500 an arbitrary choice? Why limit our retirement strategy to that one index?

Indeed, maybe there are other equity indexes, such as an index tracking stocks in the European or Asian market space for example, which, if we had a table of their monthly price data (since 1871 or a suitable broad enough data range), would lead to a higher than 4% SWR.

For the sake of argument, perhaps investing in the index of European stocks would be a better strategy than investing in S&P 500 (could be worse too of course, I do not know what the reality is, I have not done the analysis).

So my point is, why would the S&P 500 happen to be, by sheer coincidence, the best data set to base our SWR on?

Are we limiting our "life quality" with an SWR of 4%, just because the body of research work leading to that "magical 4%" seems to be based solely on analysis of US S&P stock charts?

Could we get a better than 4% SWR by analysing the data sets for other types of (equity) asset classes?

[dory36]

The research relates to the market as a whole, not the S&P 500. The original findings ("4% rule") were merely "if your investments do about as well as the market as a whole, then..." -- and not any preference for that as a choice of investments.

The overall market performance is available. Not many other subsets are, for a long period of time anyway.

ESRBob supplied a half dozen other classes which are available in Firecalc.

[retire@40]

Please don't tell me that Dorus is a combination of Dory and H*cus.

[Martha]

Originally Posted by retire@40

Please don't tell me that Dorus is a combination of Dory and H*cus.

If he was, matter and antimatter would collide and this forum would disappear.

[yakers]

Originally Posted by Martha

If he was, matter and antimatter would collide and this forum would disappear.

[Nords]

Originally Posted by retire@40

Please don't tell me that Dorus is a combination of Dory and H*cus.

Dory, did that "guest posting" feature get turned on again?

Or is "Dorus" actually signing up, posting, and then deleting his account? Didn't we turn that off too?

[modlair]

Dory, some time ago we had a great discussion of the nature of the data used in Firecalc. You provided me with a Yale link to a historical dataset of the S&P, on which Firecalc bases its conclusions.

You are no doubt aware of the recent challenges to the S&P wrt 1/N weighting coefficients vs capitalization. The question the guy in this thread asked about choice of index maybe warrants deeper thought in the context of the nature of the index itself.

I don't know if you have access to other more raw historical data than the pure S&P index. It occurs to me that the S&P is hugely variable from changes in divisors over the decades, with recent changes hugely influenced by currency fluctuation in that the larger companies are global and derive capitalization from overseas sales denominated in multiple currencies. I don't know if Seigal factored that into his 1/N work claiming 2-4% superiority over the raw S&P, but it would seem relevant.

Your previous points that the beginning years of a series are the most powerful may serve to minimize this globalization effect in that . . . it was less 30 yrs ago than now.

Anyway, it's a thought. The S&P500 index funds will probably soon see variants appear . . . i.e. S&P500 funds that are 1/N weighted among the stocks rather than using S&P's divisors.

[perinova]

I can easily guess that no Japanese has been able to retire successfully on 4% WR. Since the Nikkei is still not far from 50% of its peak 17 years ago.

This kind of bothers me and makes me a little quizzy about the 4% SWR. Can this kind of downturn happen here too? Why not?
Would it make sense to include in the data market results from other developped nations in Western Europe and Japan?

[kcowan]

Originally Posted by perinova

I can easily guess that no Japanese has been able to retire successfully on* 4% WR.* Since the Nikkei is still not far from 50% of its peak 17 years ago.* *

This kind of bothers me and makes me a little quizzy about the 4% SWR.* * Can this kind of downturn happen here too? Why not?*
Would it make sense to include in the data market results from other developped nations in Western Europe and Japan?

Of course we will have other recessions. That is why a statistical approach is taken. Many people who retired in 1999/2000 went back to work in 2002. Maybe just aprt-time. The key thing to remember is that no decision is a one-way street. We react to our new realities. Faced with a major recession, maybe we have to spend a few years in a motor home or mobile home. Not the best but an example of what might be a necessary adjustment.

Those with a very low risk tolerance should probably keep working. Many of my peers are still working and I retired 4 years ago. So far so good...

[wasdana]

Originally Posted by perinova

I can easily guess that no Japanese has been able to retire successfully on* 4% WR.* Since the Nikkei is still not far from 50% of its peak 17 years ago.* *

This kind of bothers me and makes me a little quizzy about the 4% SWR.* * Can this kind of downturn happen here too? Why not?*
Would it make sense to include in the data market results from other developped nations in Western Europe and Japan?

That's what puzzles me too.

The orginal research about the infamous "4%" was based on US market data. So a European or Japanese investor, who would likely invest more in local markets, should perhaps not be using the "4% rule". They should do their own analysis of what the SWR should be for their local environment. And, should they decide to invest in non-local markets anyway, then they have to take into account possible effects of currency exchange rates on their SWR.