Withdrawal Rates, Failure Rates, Asset Allocation to Equities

[Midpack]

With the 'what to do when you've won the game' and 'why risk anything on equities' discussions lately, I plug-n-chugged numbers through FIRECALC to satisfy my own curiousity (attempting to emulate an article by Wade Pfau). No surprises re: WRs of 4% or above, maybe re: lower WRs? Beyond that I won't attempt to draw conclusions (this time), though there may be insights here regarding:

• portfolio duration vs % equity
• effective withdrawal rates required to avoid depleting principal
• what AA has been safest even at low WR's
• the potential upside equities offer at low WR's without bringing on failure

to name a few.

It's worth remembering this is all worst case based on past history, in the spirit of giving greater weight to fear of portfolio failure over more upside than most of us could ever hope to spend (aka marginal utility).

I expect the audience for this is small, might be only one...

Notes:
- I used FIRECALC (so using past data), 30 yrs starting with $1MM and spending per WRs shown.
- I used the Total Market Portfolio and simply varied the % equity allocation.
- I left Fixed Income as-is. Different Fixed Income strategies will definitely influence the results, but will they fundamentally change them? IOW, is there an actual fixed income/no equity mix that has ever actually outperformed the optimal fixed income/with equity AA from 1871 thru present - even at (very) low WRs?

 

[Brett_Cameron]

Thanks, Midpack for sharing. It looks like you are having fun with this. The results do not surprise me. Too bad we can't predetermine the luck factor!

 

[rbmrtn]

I expect the audience for this is small, might be only one...

Not at all. I ran most of these some time ago, the investigate tab in firecalc is loads of fun. I also solved for the WR to give a given success probability. A small reduction in WR results in a nice improvement in the success rate.

 

[youbet]

Thanks Midpack.

I'm always glad to see some further demonstration of the fact that low or no equity asset allocations don't necessarily lead to safety. Your work shows that for a conservative 3% WR, a zero equity allocation = a 13% failure rate! Doesn't sound all that safe! Going to just 10% equities reduces the failure rate to 2%. And going to just 20% equities gives you a zero% failure rate.

I always enjoy looking at the FireCalc output graphs since they give you a general sense of the variability of your scenario. That is, different zero% failure scenarios can have wildly different outcomes depending on the time period, although none actually fail. Or, you can see the increasing variability of higher equity allocations. That is, a larger range of ending values despite fewer failures.

It's all fun to gaze over. I especially like the fact that no matter how you make your decisions, only the passage of time will tell what really happens regardless of the predicted probability! You pay your money, you take your chances! You need to do the prudent thing to support your own lifestyle choices and ability to tolerate risks and variability, but in the end, you have much less control than you think and it will be what it will be........

 

[mathjak107]

The problem with fixed income is the repeat reliability.

Like night follows day stock markets given enough time will always cycle through the business cycle .

Historically it has always been higher highs and higher lows.

Rates are not like that. We just finished a 36 year down trend in rates. We may never see rates this low again in our lifetime.

Betting on interest rates to follow a pattern is very different then betting the ranch on the business cycle.

 

[REattempt]

Thanks Midpack...

This is good stuff.

I also like this:

The Retire Early study on safe withdrawal rates - Millenniam Edition.

This is the retireearlyhomepage.com. The difference with this page and yours is that it looks at different "10s" and tells you the max WR to be 100% successful.

In my case I was looking at 40 years...3.54%

This is also close to what Jim Otars calculator and book say...3.2%

More interestingly, retireearlyhomepage says 77% is optimal stock. Otar says 40%. I haven't reconciled the two yet.

 

[Midpack]

I guess I will say reducing equity exposure when you get beyond FI is prudent and worth very serious consideration. But the case where 0% equity allocation has been safer or longer duration than some small 20-40% equity exposure hasn't happened at any WR I looked at since 1871 thru present. Anything is possible, but it would be unprecedented in the US at least?

Again, I'm not smart enough to conjure up all the fixed income/cash alternatives, there may be such a combination.

Even in the recent Swedroe video with the unfortunate far fetched example, at about 4:40 minutes in Larry mentioned a 30-40% equity allocation, seemingly common for folks near or in retirement who 'have won the game.'

 

[youbet]

I guess I will say reducing equity exposure when you get beyond FI is prudent and worth very serious consideration. But the case where 0% equity allocation has been safer or longer duration than some small 20-40% equity exposure hasn't happened at any WR I looked at since 1871 thru present.

Absolutely.

Even in the recent Swedroe video with the unfortunate far fetched example, at about 4:40 minutes in Larry mentioned a 30-40% equity allocation, seemingly common for folks near or in retirement who 'have won the game.'

Higher equity allocations for older folks are only dicey because of the limited time frame. Given the variability of a high equity portfolio, you might find youself living the last decade or more of your life during a downturn. While that likely won't lead to a failure (if your WR and AA are reasonable), it would be a period of heavy med drinking, as DAWG puts it.

I'm almost 7 yrs into FIRE, 65 yrs old and currently at an AA of approximately 54/46. My target is 50/50 but the current favorable market has me up in equities a bit. BTW, I count cash as fixed and my portfolio includes everything except the house and personal property. No side buckets of emergency cash or that sort of thing that are not included in the AA calculation.

 

[audreyh1]

More interestingly, retireearlyhomepage says 77% is optimal stock. Otar says 40%. I haven't reconciled the two yet.

If it's like the flattish domed graph the most studies show, the difference between 40% equities and 80% equities may be less than 1 or 2 tenths of a percent. Just speculation, but I was rather surprised at the precision given to the optimal equity % number when I suspect that the outcome difference is quite small.

 

[youbet]

If it's like the flattish domed graph the most studies who, the difference between 40% equities and 80% equities may be less than 1 or 2 tenths of a percent. Just speculation, but I was rather surprised at the precision given to the optimal equity % number when I suspect that the outcome difference is quite small.

Audrey,

When you refer to the outcome difference being small, are you referring to the failure rate or the average portfolio value as the outcome?

 

[audreyh1]

Audrey,

When you refer to the outcome difference being small, are you referring to the failure rate or the average portfolio value as the outcome?

The safe withdrawal rate.

 

[youbet]

The safe withdrawal rate.

Gotcha! Thanks.

 

[TOOLMAN]

It's worth remembering this is all worst case based on past history, in the spirit of giving greater weight to fear of portfolio failure over more upside than most of us could ever hope to spend (aka marginal utility).

Thanks for doing all the work.

I find different conclusions for AA when considering minimum risk only, and minimum risk + maximum return.

Using FIRCalc (142 yrs of data) in your chart it looks to me like minimum risk, when holding 30 yrs is 40% stocks

In "Stocks For The Long Run" [fig. 2-6] (200 years of data) when also using a 30 yr. holding the minimum risk is 75% stocks.

I'm not sure if it's the yrs of data used, black swan event(s), or how many hands the economist had.

In the above mentioned book, for maximum return + minimum risk; for a 30 year holding, [Table 2-2] shows; ultra conservative - use 71% stocks, conservative - use 90% stocks. I use 75% stocks as my best sleeping point.

It’s still not clear to me the ideal AA, however my plan over time is an average 3.5% WR, and AA of 75%-50%, and try not to play with the dials to much.

Thanks for the number crunching.

 

[ksr]

I expect the audience for this is small, might be only one...

I found this information very interesting, thanks!

 

[REWahoo]

If it's like the flattish domed graph the most studies show, the difference between 40% equities and 80% equities may be less than 1 or 2 tenths of a percent.

+1

 

[audreyh1]

Gotcha! Thanks.

Probably a negligible difference in the failure rate as well - but I'm just guessing based on a few of the graphs I've seen. [Oh - I see ReWahoo posted just such a graph ]

FWIW based on these types of studies, I've decided that I won't drop below 40% equities until I'm in my 70s (when I expect to be inside a 30 year survival window).

Ironically, since 2000, the lower equity allocations have been the winners given the tremendous outperform of bonds and the not so stellar performance of equities. Where are those cool charts showing asset allocation performance by decade?

I think this recent history is what is biasing investors so strongly towards low equity allocations.

 

[george76]

This is good stuff.

I also like this:

The Retire Early study on safe withdrawal rates - Millenniam Edition.

This is the retireearlyhomepage.com. The difference with this page and yours is that it looks at different "10s" and tells you the max WR to be 100% successful.

In my case I was looking at 40 years...3.54%

This is also close to what Jim Otars calculator and book say...3.2%

More interestingly, retireearlyhomepage says 77% is optimal stock. Otar says 40%. I haven't reconciled the two yet.

It looks the Retire Early study from retireearly.com was posted in 2000? Am I seeing that correctly? I wonder if their conclusions have changed at all since then with the equity downturn over the last decade?

 

[Midpack]

I guess I will say reducing equity exposure when you get beyond FI is prudent and worth very serious consideration. But the case where 0% equity allocation has been safer or longer duration than some small 20-40% equity exposure hasn't happened at any WR I looked at since 1871 thru present.

Absolutely.

Probably not important, but I don't know if that's an absolutely yes, or absolutely no...

 

[youbet]

Probably not important, but I don't know if that's an absolutely yes, or absolutely no...

Absolutely what you said......