Stop worrying about the 4% rule

[Retire-Canada]

Canada is modernizing its assisted dying laws, which I applaud. Given the high cost of health care in the very last years  I'm confident that process will continue and I plan to put into place legal instructions to avoid spending a prolonged period in care when my situation is not worth [in my opinion] extending my life.

If my portfolio won't support it I am okay with using Government funded care if I can no longer live at home yet don't meet the requirements of my assisted dying protocol.

In any case it seems a poor trade off to work extra years now to try and head of some potential scenario that will be an issue in my old age. I think a far more effective way to mitigate such concerns is to stop working after hitting something like a 4%WR, being flexible and keep my mind and body strong. If you give me a choice between more money or better health I'll take better health every time. Despite having a [relatively] high level of movement and low stress for a desk job it's pretty clear to me more full time work is not as beneficial to my health as more free time to pursue my interests...most of which involve significant exercise and low stress levels.

[Classical_Liberal]

Actually, it is supported by data from the US - https://www.sciencedaily.com/releases/2016/06/160615163036.htm - however, I was unaware that your medical expenses were also capped.

I absolutely agree with end of life care costs are very high as I work in the medical field and see anecdotal evidence of this daily.  This particular study is only over the last year of life. The kites article shows total spending is reduced later in life despite the increasing health care costs.  As people age other reduced costs offset the increases, albeit over a period longer than the last year.

[AdrianC]

Another article somewhat skeptical of the 4% rule:

https://earlyretirementnow.com/2017/01/18/the-ultimate-guide-to-safe-withdrawal-rates-part-6-a-2000-2016-case-study/

If you’ve been following our series on withdrawal rates (part 1 here) you have noticed that we’re quite skeptical about the 4% rule. That would be especially true for early retirees with a much longer horizon than the standard 30 years.

....

Summary

The 4% rule worked just fine during the Tech Bubble and Global Financial Crisis IF:
•You have a 30-year retirement horizon.
•You are comfortable depleting your money at the end of that horizon and/or significantly cutting your real withdrawal amounts.
•You had a relatively low equity portion (60% or less).
•You are not a passive investor but rather have the foresight to time long-term vs. short-term bonds. Specifically, you needed the ability (or dumb luck?) to implement the exact allocation that didn’t work in 1965/66 and avoid the allocation that did actually work quite beautifully in 1965/66.
•Did we miss any other qualifiers? Please let us know in the comments section!

[mathjak107]

i trust michael kitce's studies and looks at things far more than 90% of these financial writer jerks out there .

as michael found:

The 4% rule has been much maligned lately, as recent market woes of the past 15 years – from the tech crash of 2000 to the global financial crisis of 2008 – have pressured both market returns and the portfolios of retirees.

Yet a deeper look reveals that if a 2008 or even a 2000 retiree had been following the 4% rule since retirement, their portfolios would be no worse off than any of the other “terrible” historical market scenarios that created the 4% rule from retirement years like 1929, 1937, and 1966. To some extent, the portfolio of the modern retiree is buoyed by the (only) modest inflation that has been occurring in recent years, yet even after adjusting for inflation, today’s retirees are not doing any materially worse than other historical bad-market scenarios where the 4% rule worked.


https://www.kitces.com/blog/how-has-the-4-rule-held-up-since-the-tech-bubble-and-the-2008-financial-crisis/

[Gunny]

Another article somewhat skeptical of the 4% rule:

https://earlyretirementnow.com/2017/01/18/the-ultimate-guide-to-safe-withdrawal-rates-part-6-a-2000-2016-case-study/

If you’ve been following our series on withdrawal rates (part 1 here) you have noticed that we’re quite skeptical about the 4% rule. That would be especially true for early retirees with a much longer horizon than the standard 30 years.

....

Summary

The 4% rule worked just fine during the Tech Bubble and Global Financial Crisis IF:
•You have a 30-year retirement horizon.
•You are comfortable depleting your money at the end of that horizon and/or significantly cutting your real withdrawal amounts.
•You had a relatively low equity portion (60% or less).
•You are not a passive investor but rather have the foresight to time long-term vs. short-term bonds. Specifically, you needed the ability (or dumb luck?) to implement the exact allocation that didn’t work in 1965/66 and avoid the allocation that did actually work quite beautifully in 1965/66.
•Did we miss any other qualifiers? Please let us know in the comments section!

I've been reading this series with much interest.  Overall, the author is not saying the 4% rule is no longer viable, he simply states that if you are looking at a much longer time horizon and wish to leave an inheritance, then lower withdrawal rates provide higher chances of success.  I particularly like the article on withdrawal rates based on current market valuations.  His research is in line with what I've read on MadFientist.  His research is of interest to me since my DW will loose my pension if I precede her in death and she will have only SS and our stash to live on.  A little smaller WR now to provide her a higher amount later on would let me fall asleep faster at night.  This is an academic exercise at this juncture as I have yet to start tapping the stash. 

[AdrianC]

i trust michael kitce's studies and looks at things far more than 90% of these financial writer jerks out there .

Any comment on the criticism of Kites in the article?

The article claims a 2000 retiree is doing very badly with 40% in short-term treasuries, OK with 10-year treasuries. The author accuses Kites of hindsight bias when he uses the 10-year in 2000-onwards studies.

[mathjak107]

we can all cherry pick assets that work or don't work , same with time frames  . i only  go by what i show and my mix . hypothetical means nothing . through 2000 i held intermediate term bonds and equity's and did okay .

[Maschinist]

Another article somewhat skeptical of the 4% rule:

https://earlyretirementnow.com/2017/01/18/the-ultimate-guide-to-safe-withdrawal-rates-part-6-a-2000-2016-case-study/

If you’ve been following our series on withdrawal rates (part 1 here) you have noticed that we’re quite skeptical about the 4% rule. That would be especially true for early retirees with a much longer horizon than the standard 30 years.

....

Summary

The 4% rule worked just fine during the Tech Bubble and Global Financial Crisis IF:
•You have a 30-year retirement horizon.
•You are comfortable depleting your money at the end of that horizon and/or significantly cutting your real withdrawal amounts.
•You had a relatively low equity portion (60% or less).
•You are not a passive investor but rather have the foresight to time long-term vs. short-term bonds. Specifically, you needed the ability (or dumb luck?) to implement the exact allocation that didn’t work in 1965/66 and avoid the allocation that did actually work quite beautifully in 1965/66.
•Did we miss any other qualifiers? Please let us know in the comments section!

This is by far the best and most detailed study about the 4% rule that is available until now. Everybody who is dismissing this should take a break and read it from front to end.

- The KITCES study does not account for any inflation on the portfolio level. (which after a decade of more makes all the difference in optical portfolio valuation and with that also withdrawal rate)
- The ERN study does and on top of that is also looking for longer retirement periods aka what everybody here should be interested in. This part is the most important in my opinion and should open the eyes of early retires in their 30's-40's
https://earlyretirementnow.com/2016/12/21/the-ultimate-guide-to-safe-withdrawal-rates-part-3-equity-valuation/

Going from 4% to 3% withdrawal rate makes all the difference in the world at current market valuations and guarantees 50+ years peace of mind.

Im not connected in any way to the author but when you read the article you know that its pure gold.

[brooklynguy]

The KITCES study does not account for any inflation.

This is not true.  The Kitces study in question most certainly takes inflation into account.  The commentary in the Kitces article pertaining to the specific chart critiqued by the ERN article (which uses nominal figures) expressly discusses the fact that it is based on nominal (non-inflation-adjusted) dollars.  And it goes on to provide another chart displaying inflation-adjusted spending as a percentage of total portfolio value, which demonstrates that the year 2000 retiree (whose portfolio followed the parameters of Kitces' asset allocation and other assumptions) was faring better (in terms of inflation-adjusted withdrawal rate) at the 15-year mark than history's worst-case retiree (which the ERN article seems to have conveniently overlooked when it highlighted Kitces' use of a portfolio value chart based on nominal dollars).

[Maschinist]

The KITCES study does not account for any inflation.

This is not true.  The Kitces study in question most certainly takes inflation into account.  The commentary in the Kitces article pertaining to the specific chart critiqued by the ERN article (which uses nominal figures) expressly discusses the fact that it is based on nominal (non-inflation-adjusted) dollars.  And it goes on to provide another chart displaying inflation-adjusted spending as a percentage of total portfolio value, which demonstrates that the year 2000 retiree (whose portfolio followed the parameters of Kitces' asset allocation and other assumptions) was faring better (in terms of inflation-adjusted withdrawal rate) at the 15-year mark than history's worst-case retiree (which the ERN article seems to have conveniently overlooked when it highlighted Kitces' use of a portfolio value chart based on nominal dollars).

KITCES inflation is only calculated for the withdrawal but not for the portfolio value.
You can check the results easily with cfiresim. (when you check the period from 1966 onwards this flaw becomes even more obvious because of high inflation rates).

Starting year 2000 with a 60/40 portfolio with 0.1% fees leaves you with ~$700k adjusted now. (and not a nearly $1M like in the kitces study, because account value is not inflation adjusted (only the yearly withdrawal).
Everybody with a higher stock mix would have fared worse.
Currently the 10year bond yield is at ~2.5% instead of ~6% @ year 2000. The damage limiting component of bond yields is just not there today.

The ERN study is covering all that.

[dandarc]

The KITCES study does not account for any inflation.

This is not true.  The Kitces study in question most certainly takes inflation into account.  The commentary in the Kitces article pertaining to the specific chart critiqued by the ERN article (which uses nominal figures) expressly discusses the fact that it is based on nominal (non-inflation-adjusted) dollars.  And it goes on to provide another chart displaying inflation-adjusted spending as a percentage of total portfolio value, which demonstrates that the year 2000 retiree (whose portfolio followed the parameters of Kitces' asset allocation and other assumptions) was faring better (in terms of inflation-adjusted withdrawal rate) at the 15-year mark than history's worst-case retiree (which the ERN article seems to have conveniently overlooked when it highlighted Kitces' use of a portfolio value chart based on nominal dollars).

KITCES inflation is only calculated for the withdrawal but not for the portfolio value.
You can check the results easily with cfiresim. (when you check the period from 1966 onwards this flaw becomes even more obvious because of high inflation rates).
Starting year 2000 with a 60/40 portfolio with 0.1% fees leaves you with ~$700k adjusted now. (and not a nearly $1M like in the kitces study, because account value is not inflation adjusted (only the yearly withdrawal).
Everybody with a higher stock mix would have fared worse.
Currently the 10year bond yield is at ~2.5% instead of ~6% @ year 2000. The damage limiting component of bond yields is just not there today.

The ERN study is covering all that.

Why would you need to count inflation twice?

[brooklynguy]

KITCES inflation is only calculated for the withdrawal but not for the portfolio value.

Yes, the Kitces article recognizes this, discusses it, and accounts for it in the analysis.

Starting year 2000 with a 60/40 portfolio with 0.1% fees leaves you with ~$700k adjusted now.

Run the same numbers for the comparable period starting in 1966 and the remaining inflation-adjusted portfolio value is even lower, which was Kitces' point.

[Maschinist]

The KITCES study does not account for any inflation.

This is not true.  The Kitces study in question most certainly takes inflation into account.  The commentary in the Kitces article pertaining to the specific chart critiqued by the ERN article (which uses nominal figures) expressly discusses the fact that it is based on nominal (non-inflation-adjusted) dollars.  And it goes on to provide another chart displaying inflation-adjusted spending as a percentage of total portfolio value, which demonstrates that the year 2000 retiree (whose portfolio followed the parameters of Kitces' asset allocation and other assumptions) was faring better (in terms of inflation-adjusted withdrawal rate) at the 15-year mark than history's worst-case retiree (which the ERN article seems to have conveniently overlooked when it highlighted Kitces' use of a portfolio value chart based on nominal dollars).

KITCES inflation is only calculated for the withdrawal but not for the portfolio value.
You can check the results easily with cfiresim. (when you check the period from 1966 onwards this flaw becomes even more obvious because of high inflation rates).
Starting year 2000 with a 60/40 portfolio with 0.1% fees leaves you with ~$700k adjusted now. (and not a nearly $1M like in the kitces study, because account value is not inflation adjusted (only the yearly withdrawal).
Everybody with a higher stock mix would have fared worse.
Currently the 10year bond yield is at ~2.5% instead of ~6% @ year 2000. The damage limiting component of bond yields is just not there today.

The ERN study is covering all that.

Why would you need to count inflation twice?

Not counting twice but to ignore inflation on the portfolio size gives a complete wrong sense of safety for longer periods:

For example the 1966 cohort from the KITCES chart where you see nearly $1M in 1981 with 4% withdrawal rate are in reality only around $225k inflation adjusted! 1981 $
That means you have to withdraw more than 15% per year from this portfolio after only 15 years of retirement. So its failing before it reaches even 30 years.
When you take all historical years with Shiller Cape ~25 or higher and also take into account the currently non existing bond yield after inflation what do you expect? Its pure and simple math.
This is what you can see in the ERN study.

[mathjak107]

the 1965/1966  group had really good 30 year stock returns . over 10% .  however inflation , stock returns and bond returns killed them in  the first 15 years . it was really inflation that put the nail in the coffin .

[Classical_Liberal]

The KITCES study does not account for any inflation.

This is not true.  The Kitces study in question most certainly takes inflation into account.  The commentary in the Kitces article pertaining to the specific chart critiqued by the ERN article (which uses nominal figures) expressly discusses the fact that it is based on nominal (non-inflation-adjusted) dollars.  And it goes on to provide another chart displaying inflation-adjusted spending as a percentage of total portfolio value, which demonstrates that the year 2000 retiree (whose portfolio followed the parameters of Kitces' asset allocation and other assumptions) was faring better (in terms of inflation-adjusted withdrawal rate) at the 15-year mark than history's worst-case retiree (which the ERN article seems to have conveniently overlooked when it highlighted Kitces' use of a portfolio value chart based on nominal dollars).

KITCES inflation is only calculated for the withdrawal but not for the portfolio value.
You can check the results easily with cfiresim. (when you check the period from 1966 onwards this flaw becomes even more obvious because of high inflation rates).
Starting year 2000 with a 60/40 portfolio with 0.1% fees leaves you with ~$700k adjusted now. (and not a nearly $1M like in the kitces study, because account value is not inflation adjusted (only the yearly withdrawal).
Everybody with a higher stock mix would have fared worse.
Currently the 10year bond yield is at ~2.5% instead of ~6% @ year 2000. The damage limiting component of bond yields is just not there today.

The ERN study is covering all that.

Why would you need to count inflation twice?

Not counting twice but to ignore inflation on the portfolio size gives a complete wrong sense of safety for longer periods:

For example the 1966 cohort from the KITCES chart where you see nearly $1M in 1981 with 4% withdrawal rate are in reality only around $225k inflation adjusted! 1981 $
That means you have to withdraw more than 15% per year from this portfolio after only 15 years of retirement. So its failing before it reaches even 30 years.
When you take all historical years with Shiller Cape ~25 or higher and also take into account the currently non existing bond yield after inflation what do you expect? Its pure and simple math.
This is what you can see in the ERN study.

The very next graphic in the article shows the WR's after 15 years.  This is not at all deceiving.

[mathjak107]

here are the results of the 30 year time frames that failed .

they all had pretty good results  with a 60/40 mix

1907 stocks returned 7.77% -- bonds 4.250-- rebalanced portfolio 7.02- - inflation 1.64--

1929 stocks 8.19% - - bonds 1.74%-- rebalanced portfolio 6.28-- inflation 1.69--

1937 stocks 10.12 - - bonds 2.13 - rebalanced portfolio -- 7.24 inflation-- 2.82

1966 stocks 10.23 - -bonds 7.85 -- rebalanced portfolio 9.56- - inflation 5.38

for comparison the 140 year average's were:

stocks 8.39--bonds 2.85%--rebalanced portfolio 6.17% inflation 2.23%

so what made those time frames the worst ? what made them the worst is the fact in every single retirement time frame the outcome of that 30 year period was determined not by what happened over the 30 years but the entire outcome was decided in the first 15 years.

so lets look at the first 15 years in those time frames determined to be the worst we ever had.

1907--- stocks minus 1.47%---- bonds minus .39%-- rebalanced minus .70% ---inflation 1.64%

1929---stocks 1.07%---bonds 1.79%---rebalanced 2.29%--inflation 1.69%

1937---stocks -- 3.45%---bonds minus 3.07%-- rebalanced 1.23%--inflation 2.82%

1966-stocks minus .13%--bonds 1.08%--rebalanced .64%-- inflation 5.38%

[Maschinist]

...
so what made those time frames the worst ? what made them the worst is the fact in every single retirement time frame the outcome of that 30 year period was determined not by what happened over the 30 years but the entire outcome was decided in the first 15 years.
...

Yes that's correct. And on top they had in common, that in each case Shiller Cape was in the mid twenties (to low thirties in 1929).
That makes the ERN study so valuable. It separates cases with low and high Cape and makes probabilities for each group.

After reading, it gets obvious that the current statistical risk/failure rate, specially for early retires, is a good amount higher than Trinity study or others suggest.

[Eric]

...
so what made those time frames the worst ? what made them the worst is the fact in every single retirement time frame the outcome of that 30 year period was determined not by what happened over the 30 years but the entire outcome was decided in the first 15 years.
...

Yes that's correct. And on top they had in common, that in each case Shiller Cape was in the mid twenties (to low thirties in 1929).
That makes the ERN study so valuable. It separates cases with low and high Cape and makes probabilities for each group.

After reading, it gets obvious that the current statistical risk/failure rate, specially for early retires, is a good amount higher than Trinity study or others suggest.

CAPE is an okay comparison, but it's definitely not a be all end all number.  It fails to take into account changes in GAAP over time, changes in company attitudes towards dividends, and the underlying factors that drive investments (like interest rates).

I'm not completely dismissing it as a metric, but at the same time, expecting CAPE to simply revert back to its historical mean or assuming that a "high" CAPE is a signal of an impending crash is completely ignoring the context of the measure.

This writeup gets into these issues (deeply) if you want to better understand why CAPE is not a great comparison tool over time periods with different characteristics.

http://www.philosophicaleconomics.com/2013/12/shiller/

[Retire-Canada]

This writeup gets into these issues (deeply) if you want to better understand why CAPE is not a great comparison tool over time periods with different characteristics.

http://www.philosophicaleconomics.com/2013/12/shiller/

Thanks for posting. That was a good read.

[dandarc]

This writeup gets into these issues (deeply) if you want to better understand why CAPE is not a great comparison tool over time periods with different characteristics.

http://www.philosophicaleconomics.com/2013/12/shiller/

Thanks for posting. That was a good read.

+1 - I'd read about the accounting standards change's impact on CAPE, but it wasn't presented this clearly.

[brooklynguy]

This writeup gets into these issues (deeply) if you want to better understand why CAPE is not a great comparison tool over time periods with different characteristics.

http://www.philosophicaleconomics.com/2013/12/shiller/

Thanks for posting. That was a good read.

+1 - I'd read about the accounting standards change's impact on CAPE, but it wasn't presented this clearly.

Note that the Philosophical Economics author put out a few subsequent articles where he or she continued to refine his or her thinking on CAPE, including this one (in which he or she constructed a new-and-improved version of CAPE and ultimately concluded that future market returns are likely to be below the historical average):

http://www.philosophicaleconomics.com/2015/03/payout/

[Classical_Liberal]

http://www.philosophicaleconomics.com/2015/03/payout/

Super interesting, thankyou! While the entire article should be read, some interesting conclusions.

"We conclude with the question that all of this exists to answer: Is the market expensive? Yes, and returns are likely to be below the historical average, pulled down by a number of different mechanisms.  Should the market be expensive?  “Should” is not an appropriate word to use in markets.  What matters is that there are secular, sustainable forces behind the market’s expensiveness–to name a few: low real interest rates, a lack of alternative investment opportunities (TINA), aggressive policymaker support, and improved market efficiency yielding a reduced equity risk premium (difference between equity returns and fixed income returns).  Unlike in prior eras of history, the secret of “stocks for the long run” is now well known–thoroughly studied by academics all over the world, and seared into the brain of every investor that sets foot on Wall Street.  For this reason, absent extreme levels of cyclically-induced fear, investors simply aren’t going to foolishly sell equities at bargain prices when there’s nowhere else to go–as they did, for example, in the 1940s and 1950s, when they had limited history and limited studied knowledge on which to rely.

As for the future, the interest-rate-related forces that are pushing up on valuations will get pulled out from under the market if and when inflationary pressures tie the Fed’s hands–i.e., force the Fed to impose a higher real interest rate on the economy.  For all we know, that may never happen.  Similarly, on a cyclically-adjusted basis, the equity risk premium may never again return to what it was in prior periods, as secrets cannot be taken back."

Yes, the market may be inflated due to policies and interests rates.  If these factors change, lower returns will likely result in the medium term.  Not great for very recent early retirees, if it happens.  OTOH, reduced long term expected returns via a loss of equity premium (ie knowledge of buy and hold) may not necessarily be a bad thing for early retirees.  Since we all know that sequence of returns is far more important than mean real returns; reduced mean equity returns from loss of equity premium will equate to decrease length and breadth of market downturns.  This could actually increase portfolio survivability over the long run.

[AdrianC]

Not sure if this idea has been brought up before - get a quote on an annuity and see how it compares to your SWR assumption.

The idea was raised in a recent Motley Fool discussion. It factors in expected longevity and the current interest rate environment.

I got a quote for my wife (younger than me, female, longer life expectancy) for an SPIA with a CPI cost of living adjustment. It gives us an initial 2.7%.

From here:
http://www.retireearlyhomepage.com/annuity_costs.html

We learn that "hidden fees and costs can capture as much as 30% of the money you put into an annuity" (thank you  intercst).

Backing out the 30% in fees gives us an initial 2.7/0.7 = 3.86%.

Pretty close to 4%. And that's in our low interest rate environment. Quite interesting as a double check against our SWR assumptions.

[arebelspy]

That's a pretty good one Adrian because:
A) It takes into account your age (younger = more expensive annuity, because you have longer to live), and
B) The annuity companies will be conservative so they can ensure a profit.

[Ursus Major]

Not sure if this idea has been brought up before - get a quote on an annuity and see how it compares to your SWR assumption.

The idea was raised in a recent Motley Fool discussion. It factors in expected longevity and the current interest rate environment.

I got a quote for my wife (younger than me, female, longer life expectancy) for an SPIA with a CPI cost of living adjustment. It gives us an initial 2.7%.

From here:
http://www.retireearlyhomepage.com/annuity_costs.html

We learn that "hidden fees and costs can capture as much as 30% of the money you put into an annuity" (thank you  intercst).

Backing out the 30% in fees gives us an initial 2.7/0.7 = 3.86%.

Pretty close to 4%. And that's in our low interest rate environment. Quite interesting as a double check against our SWR assumptions.

So was your quote from one of those insurance companies that take out 30% in fees? If not, then you might be adding (or subtracting or dividing apples and oranges.

[Ursus Major]

That's a pretty good one Adrian because:
A) It takes into account your age (younger = more expensive annuity, because you have longer to live), and
B) The annuity companies will be conservative so they can ensure a profit.

It sure is an interesting data point, but I don't understand what it does prove for you or what the point is in regard to the 4% rule?

Sure you could hedge your longevity risk by using a few % of your portfolio to buy a deferred annuity. Then you know exactly how long your portfolio needs to last. I don't think though it says much about the SWR from a (presumably) predominately equity-based portfolio.

Perhaps you could explain your thinking on this a little more.

[arebelspy]

That's a pretty good one Adrian because:
A) It takes into account your age (younger = more expensive annuity, because you have longer to live), and
B) The annuity companies will be conservative so they can ensure a profit.

It sure is an interesting data point, but I don't understand what it does prove for you or what the point is in regard to the 4% rule?

Sure you could hedge your longevity risk by using a few % of your portfolio to buy a deferred annuity. Then you know exactly how long your portfolio needs to last. I don't think though it says much about the SWR from a (presumably) predominately equity-based portfolio.

Perhaps you could explain your thinking on this a little more.

It may give peace of mind to those worried about their money running out.

What it says is that major companies, with their actuaries and data, feel comfortable pricing it at that and making a profit, so they think the chances are pretty good.

It doesn't change anything in terms of your odds, but perhaps peace of mind.

[arebelspy]

So was your quote from one of those insurance companies that take out 30% in fees? If not, then you might be adding (or subtracting or dividing apples and oranges.

)

Which ones don't take out large fees and profit for themselves?

[Ursus Major]

That's a pretty good one Adrian because:
A) It takes into account your age (younger = more expensive annuity, because you have longer to live), and
B) The annuity companies will be conservative so they can ensure a profit.

It sure is an interesting data point, but I don't understand what it does prove for you or what the point is in regard to the 4% rule?

Sure you could hedge your longevity risk by using a few % of your portfolio to buy a deferred annuity. Then you know exactly how long your portfolio needs to last. I don't think though it says much about the SWR from a (presumably) predominately equity-based portfolio.

Perhaps you could explain your thinking on this a little more.

It may give peace of mind to those worried about their money running out.

What it says is that major companies, with their actuaries and data, feel comfortable pricing it at that and making a profit, so they think the chances are pretty good.

It doesn't change anything in terms of your odds, but perhaps peace of mind.

But on the downside it will may give those in the pre-FIRE stage a false sense of security. A double-edged sword...

[EscapeVelocity2020]

Sorry I'm late to the game, but I had this out with Nords.  I expressed my being disenfranchised about not having surety in my early retirement (by buying an annuity) as compared to his government sponsored retirement...  I'll try to find the link.  He basically said that the annuity is not a good proxy because he would never buy one, but didn't have any other similar measure.

[arebelspy]

He basically said that the annuity is not a good proxy because he would never buy one

I would be shocked if that was the crux of his argument, because him buying one is irrelevant to if it's a good comparison or not.  I think you're probably vastly oversimplifying his position.

There are problems with using it, but like I said above, it may make someone feel better about DIYing.

[Monkey Uncle]

That's a pretty good one Adrian because:
A) It takes into account your age (younger = more expensive annuity, because you have longer to live), and
B) The annuity companies will be conservative so they can ensure a profit.

It sure is an interesting data point, but I don't understand what it does prove for you or what the point is in regard to the 4% rule?

Sure you could hedge your longevity risk by using a few % of your portfolio to buy a deferred annuity. Then you know exactly how long your portfolio needs to last. I don't think though it says much about the SWR from a (presumably) predominately equity-based portfolio.

Perhaps you could explain your thinking on this a little more.

It may give peace of mind to those worried about their money running out.

What it says is that major companies, with their actuaries and data, feel comfortable pricing it at that and making a profit, so they think the chances are pretty good.

It doesn't change anything in terms of your odds, but perhaps peace of mind.

But you have to remember that the company probably issues thousands of annuity contracts.  They get to run many trials, so only a bare majority of them have to succeed for them to make a profit.  So they could handle, say, a 70% chance of success and still be profitable.  An individual only gets one trial, and so typically aims for a much higher chance of success, thereby reducing the SWR.

Just playing devil's advocate...I still think 4% is probably fine for a 30 year retirement.

[TomTX]

Not sure if this idea has been brought up before - get a quote on an annuity and see how it compares to your SWR assumption.

The idea was raised in a recent Motley Fool discussion. It factors in expected longevity and the current interest rate environment.

I got a quote for my wife (younger than me, female, longer life expectancy) for an SPIA with a CPI cost of living adjustment. It gives us an initial 2.7%.

From here:
http://www.retireearlyhomepage.com/annuity_costs.html

We learn that "hidden fees and costs can capture as much as 30% of the money you put into an annuity" (thank you  intercst).

Backing out the 30% in fees gives us an initial 2.7/0.7 = 3.86%.

Pretty close to 4%. And that's in our low interest rate environment. Quite interesting as a double check against our SWR assumptions.

If you can get an annuity that pays 3.86% with annual inflation adjustment from CPI and 30 years of expected life left - more power to you.

Then you just have to worry about the insurance company going under or otherwise failing to pay you.

[AdrianC]

I'm not in the least bit interested in buying an annuity. Just thought it an interesting idea. A feel-good double-check, if you like.

Good points raised above. I don't mistrust my own withdrawal rate so I'm not going to pursue the idea further. I already done goofed and worked too long :-)

[EscapeVelocity2020]

Here is the conversation I was referring to, I went about the annuity arguement from the other side (that it would cost me ~1.6M to 'guarantee' 40k/yr inflation adjusted, which was ~$600,000 more than SWR).

To be fair though, a $40k+ military pension makes it considerably easier to "rely" on the 4% rule than someone who will be 100% sustained from portfolio withdrawals. I would be careful presuming the risk tolerance for your situation is the same as someone where 100% of their annual spend is from investments.
...
You can run your own numbers here (https://investor.vanguard.com/annuity/fixed), but for a 42 year old getting a 'pension' with COLA, sole annuitant and lifetime immediate annuity worth $42k/yr starting April 2016, I'd have to pay $1,589,000 today.  Relying on 4% SWR I only have to pay $1,050,000... not to mention the many other benefits to investing it outside the annuity, so no, I don't call that a real option for ER.

Also, I have to mention, the annuity provider could conceivably go bankrupt in the 50 or so years I'm receiving benefits, so it's not even that good for diversification.

[EscapeVelocity2020]

He basically said that the annuity is not a good proxy because he would never buy one

I would be shocked if that was the crux of his argument, because him buying one is irrelevant to if it's a good comparison or not.  I think you're probably vastly oversimplifying his position.

ARS, you should look at his response and tell me how you would better summarize it.

[arebelspy]

He basically said that the annuity is not a good proxy because he would never buy one

I would be shocked if that was the crux of his argument, because him buying one is irrelevant to if it's a good comparison or not.  I think you're probably vastly oversimplifying his position.

ARS, you should look at his response and tell me how you would better summarize it.

You apparently failed to grok his point.

The key sentence:

This thread isn't about the military (or about any particular occupation) or about defined benefits pensions.  This is about asset allocation. 

Re-read the posts with that in mind.

[MDM]

He basically said that the annuity is not a good proxy because he would never buy one

I would be shocked if that was the crux of his argument, because him buying one is irrelevant to if it's a good comparison or not.  I think you're probably vastly oversimplifying his position.

ARS, you should look at his response and tell me how you would better summarize it.

Don't know about summarizing it, but I particularly liked "I occasionally hear "Yeah, I'd be able to retire too if I had a <insert high-risk career here> pension."  Very few of those commenters have the context of any part of the marathon but the finish line" and what followed.

[EscapeVelocity2020]

He basically said that the annuity is not a good proxy because he would never buy one

I would be shocked if that was the crux of his argument, because him buying one is irrelevant to if it's a good comparison or not.  I think you're probably vastly oversimplifying his position.

ARS, you should look at his response and tell me how you would better summarize it.

You apparently failed to grok his point.

The key sentence:

This thread isn't about the military (or about any particular occupation) or about defined benefits pensions.  This is about asset allocation. 

Re-read the posts with that in mind.

As usual, we can agree to disagree, I took my own message.  You seem to think spending 60% more to get the same thing (an annuity costs 60% more than a 4% SWR for the same income) then I guess it's fine to ignore the cost.

And it is worth reading his post in its entirety and not trying to summarize, so thanks for that.

[sol]

But you have to remember that the company probably issues thousands of annuity contracts.  They get to run many trials, so only a bare majority of them have to succeed for them to make a profit.  So they could handle, say, a 70% chance of success and still be profitable.  An individual only gets one trial, and so typically aims for a much higher chance of success, thereby reducing the SWR.

I think you might have misunderstood what the SWR literature is saying.  It's not that 95% of individual retirees with a 4% SWR successfully last 30 years, it's that 100% of people who retire in 95% of YEARS will successfully last 30 years.

Every single person in a given year will succeed or fail together.  There is no averaging effect for an insurance company to exploit, because if 2017 turns out to be one of the few years when a 4% SWR fails, then every single annuity contract they sell in 2017 will go bankrupt.

Most people don't seem to understand this important difference.  We're not each gambling that our personal portfolio will be in the 95% of successful cases like it's a random spin, we're gambling that our chosen retirement moment will not be among the 5% of worst years in world history to invest. 

[Retire-Canada]

Here is the conversation I was referring to, I went about the annuity arguement from the other side (that it would cost me ~1.6M to 'guarantee' 40k/yr inflation adjusted, which was ~$600,000 more than SWR).

Using  cFIREsim and extending the FIRE period to 40yrs I need to start with $1.5M to get to 100% success with 25/75 stocks/bonds. So your annuity costs an extra $100K for management and profit. I'm assuming the insurance company asset allocation is conservative. That could be wrong. They can also gamble with lower success rates to make more money.

When doing FIRE calculations having a bit of spending flexibility and being willing to take something less than 100% success against historical data can save you tons of money. It just depends how much "insurance" you need.

Using cFIREsim for 40yrs starting with $950K 75/25 stocks/bonds variable WR between $35K - $45K/yr gets you ~93% success. Median annual WR is over $40K/yr.

So you need to decide how much security you need and how much are you willing to pay for it. Personally I'm not even going to get to 4%WR as my time is too precious to me for that. I'm sure as heck not paying 60%+ more for an annuity.

[Retire-Canada]

Most people don't seem to understand this important difference.  We're not each gambling that our personal portfolio will be in the 95% of successful cases like it's a random spin, we're gambling that our chosen retirement moment will not be among the 5% of worst years in world history to invest.

Presumably the insurance company is selling annuities every year so they can average over the decades between years that are profitable and those they lose money on. If the cFIREsim math is right most starting years they'll end up with a big old pile of client money and in a few starting years they'll lose some.

We need a Vanguard investor owned annuity/pooled risk product that isn't trying to make a ton of profit for the FIRE community. If we shared the risk over the decades as well we should all be better off at the expense of not growing a massive fortune for most FIREers.

[The Happy Philosopher]

But you have to remember that the company probably issues thousands of annuity contracts.  They get to run many trials, so only a bare majority of them have to succeed for them to make a profit.  So they could handle, say, a 70% chance of success and still be profitable.  An individual only gets one trial, and so typically aims for a much higher chance of success, thereby reducing the SWR.

I think you might have misunderstood what the SWR literature is saying.  It's not that 95% of individual retirees with a 4% SWR successfully last 30 years, it's that 100% of people who retire in 95% of YEARS will successfully last 30 years.

Every single person in a given year with succeed or fail together.  There is no averaging effect for an insurance company to exploit, because if 2017 turns out to be one of the few years when a 4% SWR fails, then every single annuity contract they sell in 2017 will go bankrupt.

Most people don't seem to understand this important difference.  We're not each gambling that our personal portfolio will be in the 95% of successful cases like it's a random spin, we're gambling that our chosen retirement moment will not be among the 5% of worst years in world history to invest.

But an annuity contract cannot go bankrupt, it just becomes unprofitable. With the ability to pool risk the insurance company makes up for this with contracts that are more profitable than expected. Annuities are an interesting way to address retirement, but seem to be useless for most early retirees as the payout rate is so low. They are great for older people that want to simultaneously hedge sequence of return and longevity risk though.

[Retire-Canada]

Most people don't seem to understand this important difference.  We're not each gambling that our personal portfolio will be in the 95% of successful cases like it's a random spin, we're gambling that our chosen retirement moment will not be among the 5% of worst years in world history to invest.

One comforting thing about this reality is that if you do start in a bad FIRE year you'll have lots of support to get through it because all your other MMM forum buddies in the same cohort will be going through the same shit.

[sol]

We need a Vanguard investor owned annuity/pooled risk product that isn't trying to make a ton of profit for the FIRE community. If we shared the risk over the decades as well we should all be better off at the expense of not growing a massive fortune for most FIREers.

We call that annuity program "social security" and it already does exactly what you describe.

[Retire-Canada]

We call that annuity program "social security" and it already does exactly what you describe.

In Canada it's called CPP and maxes out at $11K/yr you pay for it directly. The Gov't gives you an additional ~$7K/yr paid by taxes. The problem is this starts at 65yrs old and you can't voluntarily pay more to get a bigger benefit.

What I'm thinking of would be totally self-funded at level you choose and for a duration you choose so you could retire early, but with pooled risk across starting years and not for profit so the cost isn't as high as the annuity stuff we are talking about.

[sol]

What I'm thinking of would be totally self-funded at level you choose and for a duration you choose so you could retire early, but with pooled risk across starting years and not for profit so the cost isn't as high as the annuity stuff we are talking about.

Done.  I just started a mustachian annuity fund open to all of you.  Send me your deposits, and I will guarantee you a fixed SWR on that money for the remainder of your natural life, regardless of duration or maximum dollar amounts.  Today I am offering an inflation-adjusted 1% per year SWR on your money.  PM me for account details.  All transactions must be made electronically, because I'm not mailing you any paper statements.

If anyone else would like to offer a rate higher than 1%, you are free to start your own mustachian insurance company.

[Retire-Canada]

What I'm thinking of would be totally self-funded at level you choose and for a duration you choose so you could retire early, but with pooled risk across starting years and not for profit so the cost isn't as high as the annuity stuff we are talking about.

Done.  I just started a mustachian annuity fund open to all of you.  Send me your deposits, and I will guarantee you a fixed SWR on that money for the remainder of your natural life, regardless of duration or maximum dollar amounts.  Today I am offering an inflation-adjusted 1% per year SWR on your money.  PM me for account details.  All transactions must be made electronically, because I'm not mailing you any paper statements.

If anyone else would like to offer a rate higher than 1%, you are free to start your own mustachian insurance company.

You failed at a key point I highlighted.

[sol]

You failed at a key point I highlighted.

Costs are determined by what the market will bear.  I've opened the bidding, you are free to do one better. 

If not, why not?  We're not asking for a free lunch, are we?

My product is significantly different from the fixed term annuities we've been talking about.  I'll open a retirement/lifetime-income account for your newborn baby and I will pay out to your family descendents forever.  I won't ask you about your health history.  I won't require ongoing premium payments.  I don't care what your tax rate is.  I'm not sure a comparable product is even currently available, so I don't know if you can criticize the plan based on cost.

[arebelspy]

1.1%, minimum 300k.

(Probably should take this to another thread, if it'll continue.)