Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.
The problem, as I see it, with academic research like this is that it emphasizes the theoretical and ignores what has actually occurred. Had these same researchers performed the same analysis 40 years ago would they have come to a different conclusion? I doubt it (not when your data base extends back to 1890). And if someone had based their withdrawal strategy on this conclusion back in 1982 that person might well have lived penuriously and died leaving behind not a small but an enormous fortune. So I'll go on the record and call this bunk. Ask me in 30 years which of us was right.
The problem with the 4% rule is that what sticks in most people mind is the 4% and not the 50% equity allocation it is based on. Lots of retirees, myself included, do not allocate 50% to stocks and as my 4th year of retirement grinds to an end I have not regrets that I don’t qualify for the 4% plan. Just my opinion.
The 1.9% safe withdrawal rate (SWR) referenced in the article is much lower than some others suggest. M* suggested a 3.3% SWR for a 50% stock/50% bond portfolio. Bill Bengen believes retirees can safely withdraw 4%-plus from their portfolios unless we get in a severe inflationary environment. Michael Kitces replicated Bill Bengen's original 1994 study with a broader dataset and concluded that a 4% - 4.5% SWR was feasible.
Like so much else in financial commentary various actors pick and choose whatever they need to "prove" whatever point that they're trying to make. Nothing new here.
Some of these comments make me wonder if the poster read the MW article (Hulbert is a smart and prudent cookie, in my long experience of reading him) , much less the original paper, downloadable here: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4227132 , and hugely sobering if the case.
I agree with your observation. So many posters on this board seem to assert American financial exceptionalism - why invest overseas when the US market is superior? Just go along with Bogle and Buffett. Yet no objection has been raised to using (lower) international performance figures in the analysis.
Hulbert surprised me with his choice of an adjective. "Infamous" 4% rule? A rate that will go down in infamy? Notorious might have been better. While it too carries a negative connotation, one can also achieve a positive measure of notoriety.
"...And that’s assuming you have a $1 million retirement portfolio. According to the most recent analysis by Vanguard, only 15% of retirement accounts at Vanguard are worth even $250,000. And according to an analysis of Federal Reserve data by the Boston College Center for Retirement Research, only 12% of workers have any retirement account in the first place..."
How many of those Vanguard defined contribution accounts are held by people who have only worked for their employer a few years? How many of those participants have other DC accounts at former employers or have additional retirement savings in IRAs?
Vanguard estimates that a typical participant should target a total contribution rate of 12% to 15%, including both employee and employer contributions. Forty-seven percent of participants had total employee and employer contribution rates that met those thresholds or reached the statutory contribution limit.
If one were to include IRA contributions, the percentage of participating employees meeting the recommended thresholds would be higher.
------ Hulbert grossly misread the BC study, or I did. That study contains only a single graph with a 12% figure in it. That is the rate of coverage by both defined contribution and defined benefit plans. The study says that 73% of workers were covered by DC plans alone in 2019.
As far as actually participating (having accounts) in DC plans is concerned, the BLS reports that in private industry (in 2021), 68% of workers had access to retirement plans. 75% of those participated. In other words, a majority of workers in private industry had retirement accounts.
Government workers? Even better. 92% had access to retirement plans, and 89% of those participated. over 4/5 of government workers had retirement accounts.
-------- A statistic from the Vanguard plan that I find informative is that only 2-3% of participating employees with wages under $100K max out (Figure 49). The figure would be even lower if we included the 1/3 of employees who aren't offered a plan, or the 1/4 of those who have that option but don't participate.
For all the consternation about limitations on contribution amounts, it's almost exclusively the higher salaried employees who would benefit from increasing the limit.
Explanation of “SWR” from the prospectus of QREARX
“Payments” reflect an assumed investment return of 4%. If the investment performance of the Account is constantly equal to the assumed investment return of 4% in a given year, a contract owner’s income payment in the following year would not change. If investment performance is 10% or 3% in a given year, income payments in the following year would increase by approximately 6% or decrease by approximately 1%, respectively.
@bee, TIAA & CREF use 4% AIR (Assumed Interest Rate) to calculate lifetime payouts for VA units. Other insurance co may use 2.5% or 3% AIR. This is not related to SWR (Safe Withdrawal Rate).
AIR becomes the hurdle rate. If the returns exceed AIR, annuitized VA NAV (and payout) goes up; if less, NAV (and payout) is lower.
Thanks for the further explanation of AIR. My point is that SWR (which is type of payout...that hopefully last a lifetime) should have in it's SWR methodology considerations for performance ( in both up and down markets).
"SWR (which is type of payout...that hopefully last a lifetime) should have in it's SWR methodology considerations for performance ( in both up and down markets)."
I'm surely not math gifted but how could that could be done? How could a Safe Withdrawal Rate methodology possibly predict what types of market conditions, either good or bad, might exist in any given time span?
Yes, one can construct a bracket of simulations that cover a range of overall market conditions in a given number-of-years time frame, and in fact I did exactly that over a span of fifty years prior to retirement. That helps one to see what range of asset mixes would be required to insure a decent retirement over a given length of time.
But that's a different animal than a projection that yields a number that guarantees an optimal withdrawal percentage over an entire retirement time span. If such a projection erred substantially in predicting poor market conditions, the retirement pot would be underfunded. If it erred substantially in predicting unusually good market conditions, then the retirement scheme would provide significantly less income than possible. "Safe", yes, but not terribly efficient.
Seems to me that there's no such thing as "one ring to rule them all" in financial projections over a long period of time.
My thought is to consider DSWR = “Dynamic” Safe Withdrawal Rate”
Simply, - you set a SWR…say 4% - you “dynamically” adjust your SWR based on your yearly portfolio performance. So, if you set SWR to 4%, but the market (more importantly… your portfolio) drops 25%…you “dynamically reduce your SWR by 25% thus making your DSWR 3% that year.
It seems to me that the fundamental concept of an SWR is to establish a constant rate that will be used during retirement, come what may.
If we're going to be adjusting the SWR up and down depending on market conditions (which makes perfect sense to me), then we don't actually have an "SWR", but rather a "DSWR", as bee suggests.
Which takes us back to the original Market Watch article, which as far as I'm concerned, is mostly baloney. As I said above, various actors, needing to fill up the space required for their commentary, pick and choose whatever they need to "prove" whatever point that they're trying to make. As long as the required space is filled, all is OK. These guys don't get paid to be right- they get paid to fill up space.
"Simple" was suppose to be 'set the SWR, increase by inflation and forget-about-it'. Hopefully you would have used some probability algorithm like Monte Carlo to give yourself some reassurance of success rates. Taking a 25% cut in pay because your portfolio is down takes the simple out of it, to me. Nice if you can afford it but if you can afford it you probably didn't need the 4% to begin with.
Not saying it's a bad idea or it's not prudent (I think it is), but adding dynamics is not the original intent of the 4% rule.
With an annuity, one gives up capital and potential capital appreciation in exchange for a steady (usually fixed) income stream for life. A SWR provides an alternative.
@Old_Joe, my understanding of "SWR as a percentage" is that it's not based on a fixed dollar amount (plus inflation). It's a percentage based on the yearly value of the portfolio balance. @MikeM, it is not a "set it and forget it approach". When your capital takes a hit, so does your SWR calculation (for that year's withdrawal).
The challenge is to simultaneously withdraw capital at a rate that is equal to or lower than your capital appreciates. Here's an example using FBALX and Portfolio Visualizer. A retiree starts withdrawals from a $100,000 portfolio in FBALX. The retire takes 4% SWR. In year one she withdraws 4% of her year end balance or $4,359. In 2008, the portfolio sustains a loss of 31.31% and as a result 2008's year end balance falls to $68,994. So, this year's 4% SWR is $2,875.
Over time, things improve. This is the challenge newly minted retirees are facing today. Some will choose a 5% fix annuity (set it and forget it). But inflation will eat away each year at their $5,000 fixed income payments.
In my 2007 - 2021 FBALX SWR scenario (see above), the retiree started with a SWR of $4,359 in their 1st year of withdrawals. By 2021, 4% SWR was $8,138. More importantly, the initial investment of $100,000 in 2007 is now worth $152.053 (even after yearly 4% SWR withdrawals). That would equate to $103,604 (adjusting for inflation). Capital preservation along with income seems as good or better than an annuity.
In 2022, FBALX will be down for the year (-22% right now) so the retiree should be prepared to receive 22% less than last year's $8,138 or $6,347. Still better than the fixed annuity payment of $5,000. That is the nature of SWR, it adjusts (it's dynamic) yearly...both up and down.
@Derf makes a good point. We retirees have no choice about the RMD percentage mandated by the IRS for our chronological age. We can’t reduce it, but we can withdraw more than required. Any SWR needs to account for this factor.
RMD are taken for tax purposes. The government wants to collect deferred taxes on the deferred arrangement of certain retirement accounts. Our budget should budget for these tax payments. Starting SWR prior to RMD may actually help lower RMD. Roth conversions early in retirement might also help lower RMDs. If SWRs come from tax deferred accounts they are a component of RMD. If SWR withdrawals are lower than RMDs, the remaining RMD dollars (after taxes are paid) could be contributed to a Roth IRA (if you or your spouse have work income).
Thanks David M. for the links. Note, for retirees the paper quotes a higher SWR, 2.26% - “ For a retired couple willing to accept a 5% chance of financial ruin, the real withdrawal rate of 2.26% for today’s retirees drops to 2.02% for today’s young adults and to 1.95% for today’s newborns.” Ever the (financial/economic) pessimist, I anticipate the markets will not be as generous in the future as they have been in the past.
The massive “Triumph of the Optimists” points out how bad the markets were in countries that lost major wars (Italy comes to mind). I would need to dig further but on the surface, I question whether some of the 38 countries should have been excluded. (https://www.goodreads.com/book/show/243456)
@bee : You said , "Starting SWR prior to RMD may actually help lower RMD ." So true ! When I retired I took what I thought I needed to live thru the year from IRA. Why, because I didn't start SS until 70. Also ran a few "Montys" to see some what ifs. Also the 15 or 20 times income to long term saving came into play. For those that can't control their spending in Retirement, will probably see a sharp drop in their accounts. Inheritance would also help the heavy spenders or winning the lottery. No such luck here !
As said before, different strokes for different folks, Derf P.S. Good luck to all with your withdrawals, most here won't have a problem.
@BenWP, … Starting SWR prior to RMD may actually help lower RMD. Roth conversions early in retirement might also help lower RMDs. If SWRs come from tax deferred accounts they are a component of RMD. If SWR withdrawals are lower than RMDs, the remaining RMD dollars (after taxes are paid) could be contributed to a Roth IRA (if you or your spouse have work income).
This has been a point of friction in my household. I want to do Roth conversions while my wife thinks it is crazy to pay taxes today that will otherwise be due in 12 years. I tell her the tax will be higher then. She is unfazed.
Comments
M* suggested a 3.3% SWR for a 50% stock/50% bond portfolio.
Bill Bengen believes retirees can safely withdraw 4%-plus from their portfolios
unless we get in a severe inflationary environment.
Michael Kitces replicated Bill Bengen's original 1994 study with a broader dataset
and concluded that a 4% - 4.5% SWR was feasible.
Hulbert surprised me with his choice of an adjective. "Infamous" 4% rule? A rate that will go down in infamy? Notorious might have been better. While it too carries a negative connotation, one can also achieve a positive measure of notoriety.
frightening.
If one were to include IRA contributions, the percentage of participating employees meeting the recommended thresholds would be higher.
------
Hulbert grossly misread the BC study, or I did. That study contains only a single graph with a 12% figure in it. That is the rate of coverage by both defined contribution and defined benefit plans. The study says that 73% of workers were covered by DC plans alone in 2019.
As far as actually participating (having accounts) in DC plans is concerned, the BLS reports that in private industry (in 2021), 68% of workers had access to retirement plans. 75% of those participated. In other words, a majority of workers in private industry had retirement accounts.
Government workers? Even better. 92% had access to retirement plans, and 89% of those participated. over 4/5 of government workers had retirement accounts.
https://www.bls.gov/opub/ted/2021/68-percent-of-private-industry-workers-had-access-to-retirement-plans-in-2021.htm
--------
A statistic from the Vanguard plan that I find informative is that only 2-3% of participating employees with wages under $100K max out (Figure 49). The figure would be even lower if we included the 1/3 of employees who aren't offered a plan, or the 1/4 of those who have that option but don't participate.
For all the consternation about limitations on contribution amounts, it's almost exclusively the higher salaried employees who would benefit from increasing the limit.
AIR becomes the hurdle rate. If the returns exceed AIR, annuitized VA NAV (and payout) goes up; if less, NAV (and payout) is lower.
Thanks for the further explanation of AIR. My point is that SWR (which is type of payout...that hopefully last a lifetime) should have in it's SWR methodology considerations for performance ( in both up and down markets).
I'm surely not math gifted but how could that could be done? How could a Safe Withdrawal Rate methodology possibly predict what types of market conditions, either good or bad, might exist in any given time span?
Yes, one can construct a bracket of simulations that cover a range of overall market conditions in a given number-of-years time frame, and in fact I did exactly that over a span of fifty years prior to retirement. That helps one to see what range of asset mixes would be required to insure a decent retirement over a given length of time.
But that's a different animal than a projection that yields a number that guarantees an optimal withdrawal percentage over an entire retirement time span. If such a projection erred substantially in predicting poor market conditions, the retirement pot would be underfunded. If it erred substantially in predicting unusually good market conditions, then the retirement scheme would provide significantly less income than possible. "Safe", yes, but not terribly efficient.
Seems to me that there's no such thing as "one ring to rule them all" in financial projections over a long period of time.
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4227132
My thought is to consider DSWR = “Dynamic” Safe Withdrawal Rate”
Simply,
- you set a SWR…say 4%
- you “dynamically” adjust your SWR based on your yearly portfolio performance. So, if you set SWR to 4%, but the market (more importantly… your portfolio) drops 25%…you “dynamically reduce your SWR by 25% thus making your DSWR 3% that year.
If we're going to be adjusting the SWR up and down depending on market conditions (which makes perfect sense to me), then we don't actually have an "SWR", but rather a "DSWR", as bee suggests.
Which takes us back to the original Market Watch article, which as far as I'm concerned, is mostly baloney. As I said above, various actors, needing to fill up the space required for their commentary, pick and choose whatever they need to "prove" whatever point that they're trying to make. As long as the required space is filled, all is OK. These guys don't get paid to be right- they get paid to fill up space.
Not saying it's a bad idea or it's not prudent (I think it is), but adding dynamics is not the original intent of the 4% rule.
@Old_Joe, my understanding of "SWR as a percentage" is that it's not based on a fixed dollar amount (plus inflation). It's a percentage based on the yearly value of the portfolio balance. @MikeM, it is not a "set it and forget it approach". When your capital takes a hit, so does your SWR calculation (for that year's withdrawal).
The challenge is to simultaneously withdraw capital at a rate that is equal to or lower than your capital appreciates. Here's an example using FBALX and Portfolio Visualizer. A retiree starts withdrawals from a $100,000 portfolio in FBALX. The retire takes 4% SWR. In year one she withdraws 4% of her year end balance or $4,359. In 2008, the portfolio sustains a loss of 31.31% and as a result 2008's year end balance falls to $68,994. So, this year's 4% SWR is $2,875.
Over time, things improve. This is the challenge newly minted retirees are facing today. Some will choose a 5% fix annuity (set it and forget it). But inflation will eat away each year at their $5,000 fixed income payments.
In my 2007 - 2021 FBALX SWR scenario (see above), the retiree started with a SWR of $4,359 in their 1st year of withdrawals. By 2021, 4% SWR was $8,138. More importantly, the initial investment of $100,000 in 2007 is now worth $152.053 (even after yearly 4% SWR withdrawals). That would equate to $103,604 (adjusting for inflation). Capital preservation along with income seems as good or better than an annuity.
In 2022, FBALX will be down for the year (-22% right now) so the retiree should be prepared to receive 22% less than last year's $8,138 or $6,347. Still better than the fixed annuity payment of $5,000. That is the nature of SWR, it adjusts (it's dynamic) yearly...both up and down.
References:
Using Fidelity Balanced Fund for SWR
RMD are taken for tax purposes. The government wants to collect deferred taxes on the deferred arrangement of certain retirement accounts. Our budget should budget for these tax payments. Starting SWR prior to RMD may actually help lower RMD. Roth conversions early in retirement might also help lower RMDs. If SWRs come from tax deferred accounts they are a component of RMD. If SWR withdrawals are lower than RMDs, the remaining RMD dollars (after taxes are paid) could be contributed to a Roth IRA (if you or your spouse have work income).
I Do Not Need My IRA RMD. Can I Put It in a Roth IRA?
Note, for retirees the paper quotes a higher SWR, 2.26% - “ For a retired couple willing to accept a 5% chance of financial ruin, the real withdrawal rate of 2.26% for today’s retirees drops to 2.02% for today’s young adults and to 1.95% for today’s newborns.” Ever the (financial/economic) pessimist, I anticipate the markets will not be as generous in the future as they have been in the past.
The massive “Triumph of the Optimists” points out how bad the markets were in countries that lost major wars (Italy comes to mind). I would need to dig further but on the surface, I question whether some of the 38 countries should have been excluded.
(https://www.goodreads.com/book/show/243456)
When I retired I took what I thought I needed to live thru the year from IRA. Why, because I didn't start SS until 70. Also ran a few "Montys" to see some what ifs. Also the 15 or 20 times income to long term saving came into play.
For those that can't control their spending in Retirement, will probably see a sharp drop in their accounts. Inheritance would also help the heavy spenders or winning the lottery. No such luck here !
As said before, different strokes for different folks, Derf
P.S. Good luck to all with your
withdrawals, most here won't
have a problem.