Another 2.8% SWR Article Quoting Wade Pfau

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by vineviz »

willthrill81 wrote: Tue Apr 13, 2021 2:56 pm
vineviz wrote: Tue Apr 13, 2021 12:36 pm
willthrill81 wrote: Tue Apr 13, 2021 11:53 am I was only using 2% likelihood of failure in each year for illustrative purposes. I'll be willing to bet that the likelihood of zero historic failures according to your model would be very low, even though that's what actually happened.
There have not been "zero historic failures": a 4% WR failed in the mid-1960s which, as I explained, is the only other time when the LIKELIHOOD of failure as greater than 0%.

That it hasn't failed since is not surprising since there was basically a zero probability that it would have.
The SWR was 3.8% in 1966, a failure on paper but barely one in reality.

You're saying that your model predicted a literal 0% chance of failure for every other year on record?
There was a measurable likelihood of failure for a couple of years in the 1960s (I think the peak was in 1963 at about 7%-10%).

But yes, the estimated likelihood of failure was literally 0.0% from 1968 to 2001. I didn't have the data I need to estimate pre-1960.
Last edited by vineviz on Tue Apr 13, 2021 3:57 pm, edited 2 times in total.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
marcopolo
Posts: 8446
Joined: Sat Dec 03, 2016 9:22 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by marcopolo »

willthrill81 wrote: Tue Apr 13, 2021 3:42 pm
vineviz wrote: Tue Apr 13, 2021 3:38 pm
willthrill81 wrote: Tue Apr 13, 2021 2:55 pm
vineviz wrote: Tue Apr 13, 2021 12:32 pm
willthrill81 wrote: Tue Apr 13, 2021 11:36 am My point is that I strongly suspect that vineviz's model is overly pessimistic, partly because it sounds like it doesn't take into account the fact that stock prices and expected returns tend to move against each other.
Such a suspicion would be unfounded. Adjusting for serial correlation of returns has a negligible impact in this kind of model, and virtually no impact in the left tails that are of primary concern for us.
Why?
For one thing, there just isn't much serial correlation in annual return data. Data sampled at annual intervals pass statistical tests for normality pretty much universally, unlike high-frequency data (e.g. daily, weekly, or sometime monthly) data.

For another thing, SWRs in the left-hand tail are effectively bounded in a non-linear fashion already. This is because as the growth rate of the portfolio drops, the SWR becomes dominated by the distribution of capital term (e.g. 1/n where n= number of years in decumulation) and the volatility term.

Third, even in the right-hand tail most of the "implausible sequences" end up by their very nature in the extreme ends of the tail (e.g. the top 1% or 2% of outcomes). If our analysis was relying on 98th or 99th percentile outcomes then we'd probably be worried enough about serial correlation to correct for it or at least check for its impact. But at the 90th or even 95th percentile the impact of correction is minuscule (maybe a 10-15 bps difference in indicated SWR). And at the 10th percentile, where we conventionally measure SWR, the impact is less than 2bps.
Thanks.

So you're saying that 2.8% is already a 10th percentile outcome in your model?
If one had faith that their model was accurate, and it said 10% chance of failure with a 2.8% withdrawal rate, why not start buying a series of annuties?
Once in a while you get shown the light, in the strangest of places if you look at it right.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by vineviz »

willthrill81 wrote: Tue Apr 13, 2021 3:42 pm So you're saying that 2.8% is already a 10th percentile outcome in your model?
2.8% is Pfau's number. I'd estimate the 10th percentile as being a little bit higher, at maybe 3.0% to 3.25%.

In my mind, given my preferences, a solid retirement plan would be one that starts at roughly a 4.5% withdrawal rate but which has enough spending flexibility to survive a 3% SWR if necessary.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by 000 »

HomerJ wrote: Tue Apr 13, 2021 11:34 am Sure, it's reasonable to expect low returns going forward for a few years, but low returns aren't enough to break 4%. We would need NEGATIVE returns for 25-30 years in everything.

How can someone accurately predict the chances of something that has never happened before in the U.S.?

I'm not saying it CAN'T happen... I'm saying no one can confidently model it accurately.
30 year TIPS have barely positive real yield now and shorter duration TIPS have negative real yield. If CPI is understated or taxes are in play, TIPS of any duration will have negative real returns. Assuming nominal treasuries are priced for the same expected inflation, they also have expected negative real return. So there's your negative real returns for 25-30 years on the bond side. Now, if the stock premium comes from risk over bonds, then all you need is for that risk to show up to get negative returns on stocks too. If you believe the stock premium is just something free that shows up (I don't think you actually do, HomerJ, based on your AA), then all I can say is good luck.

I don't know that we can predict the chances to a narrow error bound but certainly we can observe a bad starting position (bond yields have to matter for bond returns at least, that's why they're called fixed income) relative to the worst case on record, which vineviz has done above.
marcopolo
Posts: 8446
Joined: Sat Dec 03, 2016 9:22 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by marcopolo »

vineviz wrote: Tue Apr 13, 2021 3:57 pm
willthrill81 wrote: Tue Apr 13, 2021 3:42 pm So you're saying that 2.8% is already a 10th percentile outcome in your model?
2.8% is Pfau's number. I'd estimate the 10th percentile as being a little bit higher, at maybe 3.0% to 3.25%.

In my mind, given my preferences, a solid retirement plan would be one that starts at roughly a 4.5% withdrawal rate but which has enough spending flexibility to survive a 3% SWR if necessary.
That sounds very reasonable to me.
But, seems down right reckless given some of your analysis.
Once in a while you get shown the light, in the strangest of places if you look at it right.
Thesaints
Posts: 5108
Joined: Tue Jun 20, 2017 12:25 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Thesaints »

vineviz wrote: Tue Apr 13, 2021 12:43 pm You can't reconcile "ignoring inflation" with the concept of an SWR. The latter is always an inflation-adjusted withdrawal rate.

I'm only pushing back because there are some folks (not you, but other active participants) who seem to think that 3.33% amounts to some magical floor on SWR which can easily be achieved and therefore there is never a need to plan for the possibility of the SWR actually being below that.
I think we fully agree on this. My humorous reply was to be interpreted that if one can ignore inflation (and they can't, over a 30 year span), then everything is easy peasy: just put cash in a box and withdraw 3.33% yearly, which will last for 30 years at which point the box is empty.

Reality is much more complex: inflation can't be ignored, you are not sure that 30 years is a sufficiently long time span, and above all one wants to retire at the earliest possible date ! As such they are kind of forced to invest in risky assets and complication ensue with an estimated expected return, instead of a fixed, guaranteed, one and with a more or less large volatility of said return (and all the higher distribution momenta).
That's why optimally estimating one's retirement based on market returns is extremely difficult and, instead, a pension scheme is much more appropriate.
2pedals
Posts: 1988
Joined: Wed Dec 31, 2014 11:31 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by 2pedals »

vineviz wrote: Tue Apr 13, 2021 12:43 pm
Thesaints wrote: Tue Apr 13, 2021 12:03 pm
vineviz wrote: Tue Apr 13, 2021 11:46 am
Thesaints wrote: Tue Apr 13, 2021 10:34 am
2pedals wrote: Mon Apr 12, 2021 7:24 pm Can we have the final, definitive thread on safe withdrawal rates for the 2020's?
If you ignore inflation and assume no returns the 30-year SWR is trivial and mathematically exact: 3.33%.
Only if you ALSO ignore portfolio variance.

There is no way to generate a guaranteed 0% real return for 30 years with zero variance.
As I said, "ignoring inflation", you put cash in a safe box and withdraw 3.33% for 30 years. Guaranteed and we don't need to wait 30 years to say so, since variance and all higher momenta of the process are zero.

btw, the 2009-Present market returns are not due to mean reversion as much as they are to generous Fed.
You can't reconcile "ignoring inflation" with the concept of an SWR. The latter is always an inflation-adjusted withdrawal rate.

I'm only pushing back because there are some folks (not you, but other active participants) who seem to think that 3.33% amounts to some magical floor on SWR which can easily be achieved and therefore there is never a need to plan for the possibility of the SWR actually being below that.
Ahhh shucks I was rooting for Thesaints, I was hoping for something that is final and definitive. Something like 3.33% or 33x spending makes me happy, if not what is the magical number?
cheezit
Posts: 493
Joined: Sat Jul 14, 2018 7:28 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by cheezit »

vineviz wrote: Tue Apr 13, 2021 11:46 am
Thesaints wrote: Tue Apr 13, 2021 10:34 am
2pedals wrote: Mon Apr 12, 2021 7:24 pm Can we have the final, definitive thread on safe withdrawal rates for the 2020's?
If you ignore inflation and assume no returns the 30-year SWR is trivial and mathematically exact: 3.33%.
Only if you ALSO ignore portfolio variance.

There is no way to generate a guaranteed 0% real return for 30 years with zero variance.
If zero-coupon TIPS ("TIPS STRIPS"?) of appropriate durations existed and the yield curve were positive for durations between 0 and 30 years, one could build a ladder of either 30 or 360 STRIP TIPS (depending on whether you want monthly or annual income) to achieve this, right? Of course, neither of those conditions is presently true.
Thesaints
Posts: 5108
Joined: Tue Jun 20, 2017 12:25 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Thesaints »

cheezit wrote: Tue Apr 13, 2021 4:16 pm
vineviz wrote: Tue Apr 13, 2021 11:46 am
Thesaints wrote: Tue Apr 13, 2021 10:34 am
2pedals wrote: Mon Apr 12, 2021 7:24 pm Can we have the final, definitive thread on safe withdrawal rates for the 2020's?
If you ignore inflation and assume no returns the 30-year SWR is trivial and mathematically exact: 3.33%.
Only if you ALSO ignore portfolio variance.

There is no way to generate a guaranteed 0% real return for 30 years with zero variance.
If zero-coupon TIPS ("TIPS STRIPS"?) of appropriate durations existed and the yield curve were positive for durations between 0 and 30 years, one could build a ladder of either 30 or 360 STRIP TIPS (depending on whether you want monthly or annual income) to achieve this, right? Of course, neither of those conditions is presently true.
The real issue is that TIPS returns, even under "more normal" conditions, are much lower than expected returns from stock investments. One has to choose whether to delay retirement for many more years, until a TIPS ladder is sufficient, or roll the dice with a stock/bonds portfolio and retire sooner.
Topic Author
Leesbro63
Posts: 10640
Joined: Mon Nov 08, 2010 3:36 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Leesbro63 »

If zero-coupon TIPS ("TIPS STRIPS"?) of appropriate durations existed and the yield curve were positive for durations between 0 and 30 years, one could build a ladder of either 30 or 360 STRIP TIPS (depending on whether you want monthly or annual income) to achieve this, right? Of course, neither of those conditions is presently true.
Yet if the very inflation that TIPS are purchased to protect against (big inflation...1970's style or even Weimar Germany style) arrives, the "taxflation" problem will hit most but ROTH accounts.
jarjarM
Posts: 2511
Joined: Mon Jul 16, 2018 1:21 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by jarjarM »

Leesbro63 wrote: Tue Apr 13, 2021 4:27 pm
If zero-coupon TIPS ("TIPS STRIPS"?) of appropriate durations existed and the yield curve were positive for durations between 0 and 30 years, one could build a ladder of either 30 or 360 STRIP TIPS (depending on whether you want monthly or annual income) to achieve this, right? Of course, neither of those conditions is presently true.
Yet if the very inflation that TIPS are purchased to protect against (big inflation...1970's style or even Weimar Germany style) arrives, the "taxflation" problem will hit most but ROTH accounts.
I think if Weimar Germany style inflation arrives, Roth account will also be taxed. Let's hope that's not a scenario we all have to plan for.
Topic Author
Leesbro63
Posts: 10640
Joined: Mon Nov 08, 2010 3:36 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Leesbro63 »

jarjarM wrote: Tue Apr 13, 2021 4:55 pm
Leesbro63 wrote: Tue Apr 13, 2021 4:27 pm
If zero-coupon TIPS ("TIPS STRIPS"?) of appropriate durations existed and the yield curve were positive for durations between 0 and 30 years, one could build a ladder of either 30 or 360 STRIP TIPS (depending on whether you want monthly or annual income) to achieve this, right? Of course, neither of those conditions is presently true.
Yet if the very inflation that TIPS are purchased to protect against (big inflation...1970's style or even Weimar Germany style) arrives, the "taxflation" problem will hit most but ROTH accounts.
I think if Weimar Germany style inflation arrives, Roth account will also be taxed. Let's hope that's not a scenario we all have to plan for.
Agreed that would be catastrophic. Probably the bigger risk is like the inflation after WW1 and WW2 and Vietnam. Something like prices doubling in 5-7 years (just my guesstimate). Still horrible, but we'd probably muddle through.
jarjarM
Posts: 2511
Joined: Mon Jul 16, 2018 1:21 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by jarjarM »

Leesbro63 wrote: Tue Apr 13, 2021 5:11 pm
jarjarM wrote: Tue Apr 13, 2021 4:55 pm
Leesbro63 wrote: Tue Apr 13, 2021 4:27 pm
Yet if the very inflation that TIPS are purchased to protect against (big inflation...1970's style or even Weimar Germany style) arrives, the "taxflation" problem will hit most but ROTH accounts.
I think if Weimar Germany style inflation arrives, Roth account will also be taxed. Let's hope that's not a scenario we all have to plan for.
Agreed that would be catastrophic. Probably the bigger risk is like the inflation after WW1 and WW2 and Vietnam. Something like prices doubling in 5-7 years (just my guesstimate). Still horrible, but we'd probably muddle through.
I think 3.8% was okay (barely) for the late 1960s retiree who took a big hit for 1970s-1980s high inflation (10+% p.a.). So yeah, I think we could muddle through but won't be as pleasant of a lifestyle as most of us would hope. :beer
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by willthrill81 »

vineviz wrote: Tue Apr 13, 2021 3:57 pm
willthrill81 wrote: Tue Apr 13, 2021 3:42 pm So you're saying that 2.8% is already a 10th percentile outcome in your model?
2.8% is Pfau's number. I'd estimate the 10th percentile as being a little bit higher, at maybe 3.0% to 3.25%.

In my mind, given my preferences, a solid retirement plan would be one that starts at roughly a 4.5% withdrawal rate but which has enough spending flexibility to survive a 3% SWR if necessary.
That sounds plausible to me.
The Sensible Steward
afan
Posts: 8195
Joined: Sun Jul 25, 2010 4:01 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by afan »

NearlyRetired wrote: Tue Apr 13, 2021 1:18 pm I really don't understand this whole argument.

Surely the easiest way to calculate what you might need for retirement is to:
  • Calculate how much gross you would need a year to have a life-style you would like
  • Adjust in any given year if you have extra income/expenses for that year
  • For each year, inflate that figure - you have to guess this, but possibly use the governments inflation target, or higher if you wish to play safe
  • Do this for as many years as you wish the portfolio to last for
  • Add it all up and that is then (in real terms) how much your portfolio needs to be when you retire
So, you are guessing here, what future inflation will be like, and how long you are going to need your portfolio for, but as all of this is totally unknown, surely a guess is good enough - what else is there?

If you have at least this much in your portfolio, then "happy days" - if not, then you need to work longer/save more/re-evaluate your retirement spending. If I understand the SWR calculation, then you are baking in an investment growth to reduce your starting balance (which is why I think we have all the conversations about SWR).

You can put the core amount into cash if you wanted to (it is already potentially protected from inflation as this is baked in) and leave the rest invested .

There are still risks of course, for instance, you get the inflation figure wrong, or how long the portfolio needs to last for etc., but as no one really knows what the future will bring, so what. This has no relation to SWR etc, just basic maths (at least for me, anyways).

I really don't get the trying to use this rule - it just doesn't account for real life and the way people behave, spend etc. depending on investment performance, and you are using historical performance to try and guess what the future might look like.

What am I missing - why is it not this simple to calculate your starting portfolio balance?
To put it more simply "assume a zero real return, on average, throughout retirement." That is a conservative approach but there is nothing to guarantee that the real return will not be negative. You would still need to account for the risks of markets, your true expense pattern, inflation and taxes.

I model our retirement at 1% real return on the portfolio. I don't do a SWR, estimate, since we expect our after tax RMDs to be more than we will spend. The difference will be invested. In effect, RMDs will be "sustainable".
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by willthrill81 »

marcopolo wrote: Tue Apr 13, 2021 3:51 pm
willthrill81 wrote: Tue Apr 13, 2021 3:42 pm
vineviz wrote: Tue Apr 13, 2021 3:38 pm
willthrill81 wrote: Tue Apr 13, 2021 2:55 pm
vineviz wrote: Tue Apr 13, 2021 12:32 pm

Such a suspicion would be unfounded. Adjusting for serial correlation of returns has a negligible impact in this kind of model, and virtually no impact in the left tails that are of primary concern for us.
Why?
For one thing, there just isn't much serial correlation in annual return data. Data sampled at annual intervals pass statistical tests for normality pretty much universally, unlike high-frequency data (e.g. daily, weekly, or sometime monthly) data.

For another thing, SWRs in the left-hand tail are effectively bounded in a non-linear fashion already. This is because as the growth rate of the portfolio drops, the SWR becomes dominated by the distribution of capital term (e.g. 1/n where n= number of years in decumulation) and the volatility term.

Third, even in the right-hand tail most of the "implausible sequences" end up by their very nature in the extreme ends of the tail (e.g. the top 1% or 2% of outcomes). If our analysis was relying on 98th or 99th percentile outcomes then we'd probably be worried enough about serial correlation to correct for it or at least check for its impact. But at the 90th or even 95th percentile the impact of correction is minuscule (maybe a 10-15 bps difference in indicated SWR). And at the 10th percentile, where we conventionally measure SWR, the impact is less than 2bps.
Thanks.

So you're saying that 2.8% is already a 10th percentile outcome in your model?
If one had faith that their model was accurate, and it said 10% chance of failure with a 2.8% withdrawal rate, why not start buying a series of annuties?
Apart from the current state of the market, I believe that using at least a portion of one's fixed income holdings at the point of retirement to buy a SPIA with a 2-3% COLA that would, in conjunction with SS benefits and any other non-portfolio income, provide enough income to comfortably fund one's essential spending needs is perfectly logical. I also believe that using a portion of one's fixed income to buy TIPS and/or I-bonds that could be used to buy a SPIA when one reaches age 70-80, when payout rates are attractive, is also a logical possibility.
The Sensible Steward
User avatar
AerialWombat
Posts: 3107
Joined: Tue May 29, 2018 1:07 pm
Location: Cashtown, Cashylvania

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by AerialWombat »

deleted
Last edited by AerialWombat on Sat Feb 05, 2022 5:53 pm, edited 1 time in total.
This post is a work of fiction. Any similarity to real financial advice is purely coincidental.
heyyou
Posts: 4461
Joined: Tue Feb 20, 2007 3:58 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by heyyou »

The strawman in the presentation is the concept of setting a fixed spending amount on retirement day, then never deviating from that real amount, until portfolio depletion. As if someone would try to spend for 30 years without ever again looking at remaining portfolio value! 4% of retirement day assets (25 multiples of spending) is a good planning tool/goal, but is not a good permanent spending plan.

A better spending plan is to adapt annually to recent portfolio value as retirement proceeds--small adjustments every year by using some percentage of each recent portfolio value. As vaguely mentioned, the RMD spending method uses your age-based RMD percentage applied to each recent annual portfolio value, plus spending the interest and dividends.
FactualFran
Posts: 2777
Joined: Sat Feb 21, 2015 1:29 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by FactualFran »

willthrill81 wrote: Tue Apr 13, 2021 2:56 pm The SWR was 3.8% in 1966, a failure on paper but barely one in reality.
In what ways did the portfolio and the calculations of the SWR being 3.8% for 1966 differ from what Bengen used when determining that the SWR was 4%, as described in Determining Withdrawal Rates Using Historical Data?

That article contained the following paragraph.
Assuming a minimum requirement of 30 years of portfolio longevity, a first-year withdrawal of 4 percent [Figure l(B)], followed by inflation-adjusted withdrawals in subsequent years, should be safe. In no past case has it caused a portfolio to be exhausted before 33 years, and in most cases it will lead to portfolio lives of 50 years or longer. By comparison, a 4.25-percent first-year withdrawal could exhaust a portfolio in as little as 28 years, were past conditions to repeat themselves.
The figure was for a portfolio of 50% stocks and 50% intermediate-term government bonds. Note that for no starting year was the portfolio exhausted before 33 years when using an initial withdrawal rate of 4%.

At the time Bengen did the calculations, he had historical data up to only 1992. For later years he used averages for the returns and inflation. With 1966 as the starting year, Bengen used average return and inflation starting with year 28. However, using actual historical data that became available later, the number of years the portfolio lasted was one more, 34.

[Edit: corrected number of years portfolio lasted with actual data after 1992 in place of the averages used by Bengen.]
Last edited by FactualFran on Tue Apr 13, 2021 9:44 pm, edited 1 time in total.
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by 000 »

heyyou wrote: Tue Apr 13, 2021 9:10 pm The strawman in the presentation is the concept of setting a fixed spending amount on retirement day, then never deviating from that real amount, until portfolio depletion. As if someone would try to spend for 30 years without ever again looking at remaining portfolio value! 4% of retirement day assets (25 multiples of spending) is a good planning tool/goal, but is not a good permanent spending plan.
It is not a strawman. That is the literal definition of the SWR framework. I agree with you that it is not a good withdrawal plan.
luckyducky99
Posts: 415
Joined: Sun Dec 15, 2019 6:47 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by luckyducky99 »

heyyou wrote: Tue Apr 13, 2021 9:10 pm The strawman in the presentation is the concept of setting a fixed spending amount on retirement day, then never deviating from that real amount, until portfolio depletion. As if someone would try to spend for 30 years without ever again looking at remaining portfolio value! 4% of retirement day assets (25 multiples of spending) is a good planning tool/goal, but is not a good permanent spending plan.

A better spending plan is to adapt annually to recent portfolio value as retirement proceeds--small adjustments every year by using some percentage of each recent portfolio value. As vaguely mentioned, the RMD spending method uses your age-based RMD percentage applied to each recent annual portfolio value, plus spending the interest and dividends.
Nobody here thinks that anyone should use a fixed withdrawal rate. Flexibility is key, but there are too many conditionals and moving parts to be able to concisely, generically model that flexibility. The point of SWR estimates is that they are useful because they provide a simple model for understanding how much consumption a portfolio could support. The more flexible you can be, the more it can support.

It's a baseline -- an unrealistic one -- a massless rope over a frictionless pulley situation -- but understanding it is still useful as a first step towards understanding your own messier and more subtle situation, and it is a common starting point that most people can use.
Patzer
Posts: 701
Joined: Wed Jun 10, 2015 10:56 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Patzer »

vineviz wrote: Sun Apr 11, 2021 7:24 pm
geerhardusvos wrote: Sun Apr 11, 2021 6:37 pm
vineviz wrote: Sun Apr 11, 2021 2:37 pm
The model used at ERN is much more problematic than the one used by Pfau and most other professionals.
We’ve proved this wrong over and over again, vineviz....
You’re mistaken, I’m afraid.

Expecting real bond yields of -1% to provide the same SWR as real bond yields of +2% is a common error, but an error nonetheless.
Why do you think bond yields will not change? I was getting 2.8% on a 2 year CD 2.5 years ago.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by willthrill81 »

FactualFran wrote: Tue Apr 13, 2021 9:15 pm
willthrill81 wrote: Tue Apr 13, 2021 2:56 pm The SWR was 3.8% in 1966, a failure on paper but barely one in reality.
In what ways did the portfolio and the calculations of the SWR being 3.8% for 1966 differ from what Bengen used when determining that the SWR was 4%, as described in Determining Withdrawal Rates Using Historical Data?
The 1966 cohort didn't finish its 30 year period until 1995. Bengen's latest data at the time of his seminal publication in 1994 was from 1992.

I believe that the 3.8% number came from the Simba backtesting spreadsheet that many worked on collaboratively on this forum, but I'm not quite sure. I think that a 60/40 AA was used.

Kitces didn't find that the 4% rule has ever failed in history, but I don't recall offhand what allocation and specific assets he used to make that determination. The choice of something as seemingly mundane as 10 year bonds vs. corporate bonds vs. long-term Treasuries can impact the SWR for a specific cohort (i.e., 1966, now the defining year of the 4% rule) significantly. But in reality, whether the 30 year SWR for 1966 retirees was 3.8%, 4.1%, etc., doesn't really change anything for anyone.

While I don't know the details about Pfau's current 2.8% estimate, he has made assumptions in the past that were advantageous to annuities (e.g., assumed, not real, payouts) and disadvantageous to stock/bond portfolios (e.g., hefty expense ratios, hefty AUM fees) in a comparison of the two.
The Sensible Steward
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by vineviz »

willthrill81 wrote: Tue Apr 13, 2021 10:51 pm While I don't know the details about Pfau's current 2.8% estimate, he has made assumptions in the past that were advantageous to annuities (e.g., assumed, not real, payouts) and disadvantageous to stock/bond portfolios (e.g., hefty expense ratios, hefty AUM fees) in a comparison of the two.
This might be part of the reason my SWR estimates are little higher than his have been (maybe 0.5% higher).
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by vineviz »

Patzer wrote: Tue Apr 13, 2021 10:13 pm
Why do you think bond yields will not change? I was getting 2.8% on a 2 year CD 2.5 years ago.
Estimating the return from market timing is really hard, and the expected return of that behavior is negative anyway.

So for a 30 year retirement period, the real bond yield you'd want to use in forecasting would be from a bond with duration of roughly 15 years. And the best estimate for the realized return of a bond over its duration is the starting yield, since the starting yield curve already reflects market expectations for changes in future yields.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Young Boglehead
Posts: 321
Joined: Wed Jan 02, 2019 4:11 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Young Boglehead »

Considering most bogleheads are very conservative, I don’t get why “SWR” rates are so heavily discussed when we know ANY flexibility drastically increases your chances of success. I think it would be more prudent to make a more realistic withdrawal method the standard so that we can stop talking about silly things like a 2.8% withdrawal rate.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by vineviz »

Young Boglehead wrote: Wed Apr 14, 2021 7:32 am Considering most bogleheads are very conservative, I don’t get why “SWR” rates are so heavily discussed when we know ANY flexibility drastically increases your chances of success. I think it would be more prudent to make a more realistic withdrawal method the standard so that we can stop talking about silly things like a 2.8% withdrawal rate.
SWR is an important concept because of what it represents: the maximum safe level of retirement income that a retiree can expect.

Yes, having flexibility in spending increases the chance of success for a retiree. But that's a reflection of the investor's policy or strategy.

SWR is a measure of expected capacity, and without employing SOME measure of capacity there is no way to establish whether a particular withdrawal method is "realistic" or not. Having an accurate measure of retirement income capacity is essential for helping investors calibrate their plan. After all there are downsides to saving far more than is needed just like there are downsides to saving far less than is needed.

A retiree whose desired withdrawal rate is lower than the expected SWR can tune out discussions like this. But the more the desired withdrawal rate EXCEEDS the expected SWR, the more flexibility the investor needs to incorporate into their plan.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
CyclingDuo
Posts: 6009
Joined: Fri Jan 06, 2017 8:07 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by CyclingDuo »

willthrill81 wrote: Tue Apr 13, 2021 10:51 pmWhile I don't know the details about Pfau's current 2.8% estimate, he has made assumptions in the past that were advantageous to annuities (e.g., assumed, not real, payouts) and disadvantageous to stock/bond portfolios (e.g., hefty expense ratios, hefty AUM fees) in a comparison of the two.
Yes, that would be true regarding his premise of the Safety-First approach. He talks about income annuities on the podcast with Rick Ferri from last July:

Episode 024: Dr. Wade Pfau, host Rick Ferri
https://bogleheads.podbean.com/e/episod ... ick-ferri/

I found the interview and his presentation rather interesting.

CyclingDuo
"Save like a pessimist, invest like an optimist." - Morgan Housel | "Pick a bushel, save a peck!" - Grandpa
User avatar
mrspock
Posts: 2158
Joined: Tue Feb 13, 2018 1:49 am
Location: Vulcan

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by mrspock »

59Gibson wrote: Sun Apr 11, 2021 4:44 pm
David Jay wrote: Sun Apr 11, 2021 1:50 pm
HomerJ wrote: Sun Apr 11, 2021 1:34 pmWe can check Pfau from 10 years ago... He was comically wrong.
You mean like here? viewtopic.php?t=174030

His 2% SWR included 1% AUM and .67 ER. Effective SWR for DYI Boglehead would have been 3.6% without challenging any of Pfau's assumptions.
+1. This is very important. Nearly all of the swr doomsayers "this time is different, so 2-2.8%" are slinging annuities/and or their assumptions are so chock full of fees to drag the wd rate down to the 2s.
1% AUM lolz .... over my cold dead body. Nice of the WSJ to gloss over such an important point.
retiringwhen
Posts: 4743
Joined: Sat Jul 08, 2017 10:09 am
Location: New Jersey, USA

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by retiringwhen »

mrspock wrote: Wed Apr 14, 2021 10:01 am 1% AUM lolz .... over my cold dead body. Nice of the WSJ to gloss over such an important point.
It was not germaine the subject of the article at all so no shame on WSJ. At the same time, I would bet that the average AUM fee for WSJ readers is a lot closer to 1% than it is for Bogleheads.

I know of trusts paying 1% AUM fees to run a corporate bond ladder that could probably be bought much cheaper and track more closely to the goals of the trust if the trust bought an SPIA and fired the financial advisor. That is the kind of reader/investor Mr. Pfau is hoping to get to listen.
User avatar
Scott S
Posts: 1937
Joined: Mon Nov 24, 2008 2:28 am
Location: building my position

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Scott S »

Young Boglehead wrote: Wed Apr 14, 2021 7:32 am Considering most bogleheads are very conservative, I don’t get why “SWR” rates are so heavily discussed when we know ANY flexibility drastically increases your chances of success. I think it would be more prudent to make a more realistic withdrawal method the standard so that we can stop talking about silly things like a 2.8% withdrawal rate.
vineviz already did a great job answering the main question, but there is a wrinkle with flexible withdrawal plans that is also worth keeping in mind: Typically, they are designed to allow the retiree to withdraw more in "good years" so they aren't locked into spending a measly 4% while watching their portfolio value skyrocket. Unfortunately, this can put their portfolio at greater risk during the "bad years". ERN ran simulations for several flexible withdrawal plans for the beleaguered 1966 cohort, and as you can see, withdrawing 2.8% (or less) went from being "silly" to becoming a necessity for stretches of time: https://earlyretirementnow.com/2018/05/ ... s-reality/
"Old value investors never die, they just get their fix from rebalancing." -- vineviz
Thesaints
Posts: 5108
Joined: Tue Jun 20, 2017 12:25 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Thesaints »

A flexible spending plan consisting in withdrawing X% of the current assets value never fails for any value of X<100...
Patzer
Posts: 701
Joined: Wed Jun 10, 2015 10:56 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Patzer »

vineviz wrote: Wed Apr 14, 2021 7:28 am
Patzer wrote: Tue Apr 13, 2021 10:13 pm
Why do you think bond yields will not change? I was getting 2.8% on a 2 year CD 2.5 years ago.
Estimating the return from market timing is really hard, and the expected return of that behavior is negative anyway.

So for a 30 year retirement period, the real bond yield you'd want to use in forecasting would be from a bond with duration of roughly 15 years. And the best estimate for the realized return of a bond over its duration is the starting yield, since the starting yield curve already reflects market expectations for changes in future yields.
Pretending the rates will stay what they are right now seems completely wrong based on any historical data.
Why would I lock in low rates with long-term bonds?
I use short-term/liquid investments when yields are lower than inflation, and intermediate-term bonds when they are above it.
I would never buy a 30 year bond, because of inflation risk and Fixed income is supposed to be the low risk part of my portfolio.
I am getting 1% in my checking account right now, so why would I lock in a 5 year bond at that rate?
Finally, if you want to talk about a 30 year retirement period, the 30 year treasury is yielding 2.32%, which is not -1% real.
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by vineviz »

Patzer wrote: Wed Apr 14, 2021 4:44 pm Pretending the rates will stay what they are right now seems completely wrong based on any historical data.
There's no "pretending" going on: the best estimate of future bond returns is the starting yield. Total returns from bonds are as likely to be lower than that starting yield as to be higher.
Patzer wrote: Wed Apr 14, 2021 4:44 pm Why would I lock in low rates with long-term bonds?
Because unless you have a crystal ball or a time machine, there's no way to know whether today's rates are "low" or "high". If it wasn't obvious in March of 2000 that "locking in" a 6.4%. yield on 20-year Treasuries was a great deal, why is it so obvious today that locking in 2.3% on 20-year Treasuries is a bad deal?
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
User avatar
vineviz
Posts: 14921
Joined: Tue May 15, 2018 1:55 pm
Location: Baltimore, MD

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by vineviz »

Thesaints wrote: Wed Apr 14, 2021 3:30 pm A flexible spending plan consisting in withdrawing X% of the current assets value never fails for any value of X<100...
Maybe, but many people would view a 70%+ drop in retirement income a "failure" even though the portfolio balance never technically drops to $0.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch
Normchad
Posts: 5648
Joined: Thu Mar 03, 2011 6:20 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Normchad »

Scott S wrote: Wed Apr 14, 2021 10:26 am
Young Boglehead wrote: Wed Apr 14, 2021 7:32 am Considering most bogleheads are very conservative, I don’t get why “SWR” rates are so heavily discussed when we know ANY flexibility drastically increases your chances of success. I think it would be more prudent to make a more realistic withdrawal method the standard so that we can stop talking about silly things like a 2.8% withdrawal rate.
vineviz already did a great job answering the main question, but there is a wrinkle with flexible withdrawal plans that is also worth keeping in mind: Typically, they are designed to allow the retiree to withdraw more in "good years" so they aren't locked into spending a measly 4% while watching their portfolio value skyrocket. Unfortunately, this can put their portfolio at greater risk during the "bad years". ERN ran simulations for several flexible withdrawal plans for the beleaguered 1966 cohort, and as you can see, withdrawing 2.8% (or less) went from being "silly" to becoming a necessity for stretches of time: https://earlyretirementnow.com/2018/05/ ... s-reality/
I have to give a shout out to Scott S for his avatar, especially in these SWR threads.

I love that we constantly debate the 4% observation. The power of the debate, is that this is absolutely getting hammered into our consciousness. So when a planner comes along, and recommends 7%, 10%, or 12%, we will instinctively doubt their advice.

Safe is a loaded word in all of this. Going up and down the stairs is safe, but there is still a chance that I could fall down the stairs and die. In that same way, I think 4% is safe. It’s not guaranteed to never fail, but it’s safe, like using the stairs.
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by 000 »

Patzer wrote: Wed Apr 14, 2021 4:44 pm Why would I lock in low rates with long-term bonds?
Deflation insurance?
Thesaints
Posts: 5108
Joined: Tue Jun 20, 2017 12:25 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Thesaints »

vineviz wrote: Wed Apr 14, 2021 5:18 pm
Thesaints wrote: Wed Apr 14, 2021 3:30 pm A flexible spending plan consisting in withdrawing X% of the current assets value never fails for any value of X<100...
Maybe, but many people would view a 70%+ drop in retirement income a "failure" even though the portfolio balance never technically drops to $0.
Indeed! Therefore "equal withdrawals" throughout retirements is an objective condition, "Being flexible" much less so.
We can definitely tell if a x% SWR fails, when the retiree is forced to withdraw y<x at any point. How do we tell that a "soft SWR" failed ?
We can all agree that a 70% drop is a failure. How about a 30% drop ? What if it has to drop only for a couple of years, before originally planned withdrawal can resume ?
Last edited by Thesaints on Wed Apr 14, 2021 5:59 pm, edited 1 time in total.
HENRYGRUGER
Posts: 148
Joined: Mon Oct 12, 2020 11:32 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by HENRYGRUGER »

chipperd wrote: Mon Apr 12, 2021 5:10 am
Leesbro63 wrote: Sun Apr 11, 2021 9:50 am I hesitate to start yet another "4% won't work" SWR thread, but I saw this WSJ article and think anything quoting Wade Pfau will be of interest to Bogleheads:

https://www.wsj.com/articles/retirement ... 1617929631

"Wade Pfau, a professor at the American College of Financial Services, has run tests that support Mr. Bengen’s earlier research. But Prof. Pfau says someone retiring today and taking fixed inflation-adjusted withdrawals from a portfolio shouldn’t use the 4% rule; he calculates the safe number is just 2.8%, mostly because bond rates have tumbled."
Your hesitation is warranted, esp. given that this is, again, about a guy who is financially supported by the insurance industry and, not coincidently, tries to plant fear in solid retirement planning that don't have annuities as part of that planning process.
Bottom line, we all have our biases, but Pfau has a real conflict of interest. I would suggest conflicts of interest be noted by any poster quoting a financial professional as a disclaimer.
Disclaimer: I'm financially supported by self and as such, have a bias towards my own opinions.
BTW, the graduation rate from that American College of Financial Services is in the mid 30% range. What a rip off.
I have no interest in debating your position regarding the SWR, as you are entitled to you opinion. You are not, however, entitled to you own "truth.

Your statement regarding graduation rates from the American College for Financial Services is facially incorrect, at least partially.

Here are the facts:

Solomon S. Huebner School (Undergraduate Studies)

The Huebner School offers six professional designation programs for individuals working in the financial services industry and an educational program qualifying students to sit for CFP Board’s national certification exam. Students may take a single course, a full program, or multiple programs according to their individual preferences. Program completion rates reflect those student preferences as well as the rigorous nature of the coursework.

Huebner School designation programs range in scope from 3 to 8 courses. Aggregate designation completion rates vary by designation from 26.4 percent to 55.3 percent for students who entered The American College between 2012 and 2014.The overall completion rate for all Huebner School designation programs was 34.9 percent for the same period of time.

Between 2014 and 2019, the CFP® certification exam pass rate for students who have completed The American College’s CFP® education program exceeded the national average for all other providers by an average of 11.1 percentage points.

Irwin Graduate School (Graduate Studies)

Admission to The American College’s graduate degree programs requires advanced professional experience in addition to collegiate studies. Graduation rates reflect the expertise of students accepted into these programs as well as the rigorous nature of the coursework.

The Master of Science in Management (MSM) had an aggregate graduation rate of 95.1 percent for students who enrolled in the program between 2013 and 2017.

The Master of Science in Financial Services (MSFS) had an aggregate graduation rate of 62.5 percent for students who enrolled between 2008 and 2012.2

The completion rate for the Chartered Advisor in Philanthropy® (CAP®) was 65.9 percent for students who enrolled between 2012 and 2014.2

As of November 2019, The American College had awarded 13 PhDs.


And since we pride ourselves on this board in providing transparency, I am on the faculty at The American College, but unlike most academics, I spent 41 years "in the field" prior to joining the College, 12 years ago.

You may not respect Dr. Pfau's work, but members of Congress and his peers in the academy do.
Last edited by HENRYGRUGER on Thu Jul 29, 2021 6:00 pm, edited 1 time in total.
Patzer
Posts: 701
Joined: Wed Jun 10, 2015 10:56 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Patzer »

000 wrote: Wed Apr 14, 2021 5:29 pm
Patzer wrote: Wed Apr 14, 2021 4:44 pm Why would I lock in low rates with long-term bonds?
Deflation insurance?
Cash does the same thing, with no inflation side risk.
Intermediate bonds do the same thing, with less inflation side risk.
Patzer
Posts: 701
Joined: Wed Jun 10, 2015 10:56 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Patzer »

vineviz wrote: Wed Apr 14, 2021 5:14 pm
Patzer wrote: Wed Apr 14, 2021 4:44 pm Pretending the rates will stay what they are right now seems completely wrong based on any historical data.
There's no "pretending" going on: the best estimate of future bond returns is the starting yield. Total returns from bonds are as likely to be lower than that starting yield as to be higher.
Patzer wrote: Wed Apr 14, 2021 4:44 pm Why would I lock in low rates with long-term bonds?
Because unless you have a crystal ball or a time machine, there's no way to know whether today's rates are "low" or "high". If it wasn't obvious in March of 2000 that "locking in" a 6.4%. yield on 20-year Treasuries was a great deal, why is it so obvious today that locking in 2.3% on 20-year Treasuries is a bad deal?
Do you really not see the difference between a bond yield that can fall 6.4% and a bond yield that can fall 2.3% before hitting zero?
Bonds aren't going below 0 for any extended period of time and if they do, then I won't be holding bonds, so it doesn't matter.

I don't think the bond market is efficient. Last year in March, I could get 1% for cash, 1% for a 2 year CD, or .39% for a 5 year treasury. So, I could get 2% returns in 2 years with a CD or 2% returns in 5 years with a Treasury and have the option to reinvest after 2 years at a potentially higher rate.

If the bond market is efficient, than wouldn't it be safe to assume that long-term bond buyers are pricing in inflation to be lower than the total return of the bonds they are buying, or why would they invest for the long-term in those asset classes? It's one thing to move into an asset class in a panic, i.e. when 30 year bonds fell to .99% on March 9th, 2020, but we aren't talking about that.
I am talking about 30 year bonds trading in a 2.3-2.4% range consistently for the last month. If the market is efficient, and that rate is not likely go up, than it must mean the efficient market thinks that inflation will be below the current 30 year bond rate, or the rate would be rising.
If the market thinks inflation will be below that rate, then bonds do not and will not yield the -1% Real that you said they do.

If bonds do yield -1% real for a significant portion of the time like you suggested, then less people will be interested in bonds, and bond yields will rise to efficiently correct for the asset yielding a negative real amount, right?

So, the statement that people should calculate their SWR with an assumption of a -1% return for bonds seems off to me.
Thesaints
Posts: 5108
Joined: Tue Jun 20, 2017 12:25 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Thesaints »

Patzer wrote: Wed Apr 14, 2021 6:35 pm I don't think the bond market is efficient. Last year in March, I could get 1% for cash, 1% for a 2 year CD, or .39% for a 5 year treasury. So, I could get 2% returns in 2 years with a CD or 2% returns in 5 years with a Treasury and have the option to reinvest after 2 years at a potentially higher rate.
..or at a potentially lower one.
Patzer
Posts: 701
Joined: Wed Jun 10, 2015 10:56 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Patzer »

Thesaints wrote: Wed Apr 14, 2021 6:37 pm
Patzer wrote: Wed Apr 14, 2021 6:35 pm I don't think the bond market is efficient. Last year in March, I could get 1% for cash, 1% for a 2 year CD, or .39% for a 5 year treasury. So, I could get 2% returns in 2 years with a CD or 2% returns in 5 years with a Treasury and have the option to reinvest after 2 years at a potentially higher rate.
..or at a potentially lower one.
A 5 year treasury paid out .39% x 5 = after 5 years you got 1.95%.
A 2 year CD paid out 1% x 2 = 2% after 2 years.
So, as long as the rate was 0 or more, then the CD would win. Cash under the mattress for the next 3 years would still beat the Treasury.
If however rates were anything above 0%, the margin of the win for the CD would only have grown.
Thesaints
Posts: 5108
Joined: Tue Jun 20, 2017 12:25 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Thesaints »

Patzer wrote: Wed Apr 14, 2021 6:45 pm
Thesaints wrote: Wed Apr 14, 2021 6:37 pm
Patzer wrote: Wed Apr 14, 2021 6:35 pm I don't think the bond market is efficient. Last year in March, I could get 1% for cash, 1% for a 2 year CD, or .39% for a 5 year treasury. So, I could get 2% returns in 2 years with a CD or 2% returns in 5 years with a Treasury and have the option to reinvest after 2 years at a potentially higher rate.
..or at a potentially lower one.
A 5 year treasury paid out .39% x 5 = after 5 years you got 1.95%.
A 2 year CD paid out 1% x 2 = 2% after 2 years.
So, as long as the rate was 0 or more, then the CD would win. Cash under the mattress for the next 3 years would still beat the Treasury.
If however rates were anything above 0%, the margin of the win for the CD would only have grown.
Oh, yes. You are correct.
But the CD vs. treasuries issue has nothing to do with market efficiency. The market is efficiently pricing risk, instead, which is a lot higher for a CD above FDIC threshold than for treasuries.
What you saw is one of the very few small advantages that retail investors enjoy over the big players from time to time.
User avatar
HomerJ
Posts: 21282
Joined: Fri Jun 06, 2008 12:50 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by HomerJ »

vineviz wrote: Wed Apr 14, 2021 5:14 pmTotal returns from bonds are as likely to be lower than that starting yield as to be higher.
There you go assuming a normal distribution again. Even at the extremes. There are other variables involved when you get into negative return territories. Variables that you ignore, and could possibly break your model.

Every academic fails in the real world because they just wave away all those other variables "Well, assuming everything else stays the same, here's what my model shows"
"The best tools available to us are shovels, not scalpels. Don't get carried away." - vanBogle59
Thesaints
Posts: 5108
Joined: Tue Jun 20, 2017 12:25 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Thesaints »

HomerJ wrote: Wed Apr 14, 2021 7:57 pm
vineviz wrote: Wed Apr 14, 2021 5:14 pmTotal returns from bonds are as likely to be lower than that starting yield as to be higher.
There you go assuming a normal distribution again. Even at the extremes.
There are many distributions with zero expected value, which would give an equal cumulative probability of rates to be lower, or higher.
The gaussian is only one of them.
The issue with vineviz's statement is that when rates are near zero there is much less space downward, than there is upward (even accounting for slightly negative rates).
However, because of the size of public debt, the consequent interest outlays, and the stiffness of expenditures, we can be reasonably certain that large upward movements are as unlikely as large downward ones in all the advanced economies.
000
Posts: 8211
Joined: Thu Jul 23, 2020 12:04 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by 000 »

Patzer wrote: Wed Apr 14, 2021 6:09 pm
000 wrote: Wed Apr 14, 2021 5:29 pm
Patzer wrote: Wed Apr 14, 2021 4:44 pm Why would I lock in low rates with long-term bonds?
Deflation insurance?
Cash does the same thing, with no inflation side risk.
Intermediate bonds do the same thing, with less inflation side risk.
A long term deflationary episode might push interest rates and the general return on capital downward.
Patzer
Posts: 701
Joined: Wed Jun 10, 2015 10:56 am

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by Patzer »

000 wrote: Wed Apr 14, 2021 8:17 pm
Patzer wrote: Wed Apr 14, 2021 6:09 pm
000 wrote: Wed Apr 14, 2021 5:29 pm
Patzer wrote: Wed Apr 14, 2021 4:44 pm Why would I lock in low rates with long-term bonds?
Deflation insurance?
Cash does the same thing, with no inflation side risk.
Intermediate bonds do the same thing, with less inflation side risk.
A long term deflationary episode might push interest rates and the general return on capital downward.
Serious question, not being rhetorical. What conditions do you think it would take to make a truly long-term deflationary episode happen.
Even the great depression was only 4 years of deflation before inflation kicked back in, which cash would have been fine through.
Factor in the power of modern central bankers, and I have a hard time seeing them not just printing money until inflation started again.

Even if a disease wiped out half the population, suddenly doubling the amount of land per person available and several other resources, we might get a massive spike of deflation, that would then be followed by a return to inflation.

The only scenario I can really think of is if Bitcoin became the global currency and replaced local national currencies. Since Bitcoin is inherently deflationary (fixed supply of Bitcoins, but goods and services grow), if it was the world currency it would bring an unending era of deflation. But, long-term bonds would not protect against that, unless they were denominated in Bitcoins instead of dollars.
afan
Posts: 8195
Joined: Sun Jul 25, 2010 4:01 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by afan »

HENRYGRUGER wrote: Wed Apr 14, 2021 5:58 pm
BTW, the graduation rate from that American College of Financial Services is in the mid 30% range. What a rip off.
The overall completion rate for all Huebner School designation programs was 34.9 percent for the same period of time.

You may not respect Dr. Pfau's work, but members of Congress and his peers in the academy do.
1. The 34.9% seems well characterized as "mid 30% range". Or were you limiting your analysis to the other programs?
2. Being respected by members of Congress is hardly evidence of expertise. Not to mention that there is no way to know whether members of Congress respect him.
3. Evidence of respect among his peers in the academy would be the frequency with which they cite his work in their peer-reviewed academic publications and the quality of journals in which he publishes. Do you have data on those metrics? How often does he publish in the top economics or finance journals? Quarterly Journal of Economics? Journal of Financial Economics? Journal of Finance? Econometrica?
Last edited by afan on Wed Apr 14, 2021 9:23 pm, edited 2 times in total.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama
index2max
Posts: 514
Joined: Mon Jan 21, 2019 10:01 pm

Re: Another 2.8% SWR Article Quoting Wade Pfau

Post by index2max »

BernardShakey wrote: Sun Apr 11, 2021 12:23 pm
Leesbro63 wrote: Sun Apr 11, 2021 9:50 am I hesitate to start yet another "4% won't work" SWR thread, but I saw this WSJ article and think anything quoting Wade Pfau will be of interest to Bogleheads:

https://www.wsj.com/articles/retirement ... 1617929631

"Wade Pfau, a professor at the American College of Financial Services, has run tests that support Mr. Bengen’s earlier research. But Prof. Pfau says someone retiring today and taking fixed inflation-adjusted withdrawals from a portfolio shouldn’t use the 4% rule; he calculates the safe number is just 2.8%, mostly because bond rates have tumbled."
Wish I had a subscription to WSJ but I'm too cheap...
If you want to bypass the paywall, simply view the article using the website, archive.is, which has several domains such as archive.vn, archive.today etc.

Here's a version you can view that someone else requested a snapshot of.

https://archive.vn/bBkLa

I also recommend installing the add-on "bypass paywalls:

https://github.com/iamadamdev/bypass-paywalls-chrome
Post Reply