Larry Swedroe: 3% is the new 4%

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
HomerJ
Posts: 13198
Joined: Fri Jun 06, 2008 12:50 pm

Re: Larry Swedroe: 3% is the new 4%

Post by HomerJ » Mon Sep 09, 2019 2:25 pm

305pelusa wrote:
Mon Sep 09, 2019 2:08 pm
I can trivially prove it for bonds. The real yields are currently lower than at any period the Trinity study looked it. This is a fact and indisputable. If you treat rates as random walks then, on average, the future returns of treasuries are unequivocally lower than past ones.
This doesn't appear to be a fact or indisputable.

Looking at past rates

https://www.multpl.com/10-year-treasury ... le/by-year

10-year rates on Sept 6 was 1.55%

It's true that is the lowest rate on the chart (although that chart only shows Jan 1st rates, so I don't know about the rates on other days of the year).

But those rates are nominal rates.

In the 1940s and 1950s, rates were nearly as low, mostly in the 2% range (1.95% in 1941), but inflation was higher most years.

So the real rates were lower for most of two decades than today.
The J stands for Jay

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Mon Sep 09, 2019 2:27 pm

305pelusa wrote:
Mon Sep 09, 2019 2:08 pm
The real yields are currently lower than at any period the Trinity study looked it. This is a fact and indisputable. If you treat rates as random walks then, on average, the future returns of treasuries are unequivocally lower than past ones.
But investors aren't limited to Treasuries today. We now have TIPS. As of today, real yields for 5, 10, and 30 year TIPS are .08, -.04, and .43, respectively. The real returns of bonds during the worst historic 15 year periods, which have been very strongly associated to the 30 year SWR rate, were -.15%. Beating that with TIPS shouldn't be too difficult.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

randomguy
Posts: 8089
Joined: Wed Sep 17, 2014 9:00 am

Re: Larry Swedroe: 3% is the new 4%

Post by randomguy » Mon Sep 09, 2019 2:28 pm

305pelusa wrote:
Mon Sep 09, 2019 2:08 pm


I can trivially prove it for bonds. The real yields are currently lower than at any period the Trinity study looked it. This is a fact and indisputable. If you treat rates as random walks then, on average, the future returns of treasuries are unequivocally lower than past ones.

Give a link to data?


In 1942, inflation was 11.35% and the 10 year was yielding 2.46% Those 1942 10 year investors got -2.9% real over the next 10 years. Today the 10 year tip yields -.03% of almost 3% better than those historical numbers. Obviously not quite an apples to apples but that seems a lot worse than today.

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Mon Sep 09, 2019 2:30 pm

EnjoyIt wrote:
Mon Sep 09, 2019 2:18 pm
HomerJ wrote:
Mon Sep 09, 2019 1:42 pm
305pelusa wrote:
Mon Sep 09, 2019 9:32 am
Mkay, fine. Then prove to me that today's returns, on average, are in fact not worse than the worst in the past. Because if you cannot, then you cannot claim 4% is fine based on already surviving the worst in the past.
I have no idea what you're saying here...

Today's returns are not worse than the worst in the past.

Tomorrow's returns may be though, sure.

But what are the odds? And what is the result?

If failure is bankruptcy and ruin, and starving under a bridge, then yes a 97% chance of success may not be good enough.

What if failure is 2 vacations a year instead of 4 vacations a year? Is 97% good enough then?

Imagine if the odds were presented this way.

80% chance your money continues to grow and you can spend more and more in retirement.
15% chance your money stays about even, but you never have to cut back.
4% chance you may have to cut back to 2 vacations a year instead of 4.
1% chance you may have to stop taking vacations all together.

Or you work an extra 10 years, your 10 healthiest years, your ten best years of the 0-30 you have left (note the possible zero).

And then you get it up to

90% chance your money continues to grow and you can spend more and more in retirement.
9% chance your money stays about even, but you never have to cut back.
1% chance you may have to stop taking vacations all together.

(You're never going to get rid of that 1%)

If you're 40 or 50, working the extra 10 years may be no great hardship, and the trade-off may make sense.
If you're 55 or 60, and you feel that creak in your bones, and you've already had 3 high school friends die, the trade-off may not be worth it.
This hits home. Watching people I know get cancer, debilitating strokes, get into critical car accidents, family and friends dying. Life sure is short and unpredictable.

Some people just love to work and it may make sense for them to keep doing so even if they may have a big heart attack on the way home from work or at some meeting never to experience retirement.

Personally, for me, 4% with the risk of cutting back on some vacation is fine by me.
When you take into account the likelihood that most ~65 year old retirees will not survive 30 years, 4% withdrawals look even more conservative for them. According to the SS Administration, 20% of 65 year old men won't survive to age 74.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
305pelusa
Posts: 558
Joined: Fri Nov 16, 2018 10:20 pm

Re: Larry Swedroe: 3% is the new 4%

Post by 305pelusa » Mon Sep 09, 2019 2:44 pm

randomguy wrote:
Mon Sep 09, 2019 2:28 pm
305pelusa wrote:
Mon Sep 09, 2019 2:08 pm


I can trivially prove it for bonds. The real yields are currently lower than at any period the Trinity study looked it. This is a fact and indisputable. If you treat rates as random walks then, on average, the future returns of treasuries are unequivocally lower than past ones.

Give a link to data?


In 1942, inflation was 11.35% and the 10 year was yielding 2.46% Those 1942 10 year investors got -2.9% real over the next 10 years. Today the 10 year tip yields -.03% of almost 3% better than those historical numbers. Obviously not quite an apples to apples but that seems a lot worse than today.
I mistyped that, my apologies. I was referring to 1966, since that was the year when 4% failed. My point was that real treasury bond yields are lower today than they were back in a period when 4% already failed.

My original quoted statement is hilariously wrong haha

BluesH
Posts: 40
Joined: Mon Nov 19, 2018 1:31 pm

Re: Larry Swedroe: 3% is the new 4%

Post by BluesH » Mon Sep 09, 2019 2:54 pm

HomerJ wrote:
Mon Sep 09, 2019 1:42 pm
If you're 40 or 50, working the extra 10 years may be no great hardship, and the trade-off may make sense.
If you're 55 or 60, and you feel that creak in your bones, and you've already had 3 high school friends die, the trade-off may not be worth it.
Ouch, this one hit home. I'm about to attend my 50 year HS reunion; so older than you're quoting above. I just logged onto the reunion site, and did a count of the "In Memorium" page. Counted 75 names. So yeah, mortality is an issue.

User avatar
HomerJ
Posts: 13198
Joined: Fri Jun 06, 2008 12:50 pm

Re: Larry Swedroe: 3% is the new 4%

Post by HomerJ » Mon Sep 09, 2019 2:55 pm

305pelusa wrote:
Mon Sep 09, 2019 2:44 pm
randomguy wrote:
Mon Sep 09, 2019 2:28 pm
305pelusa wrote:
Mon Sep 09, 2019 2:08 pm


I can trivially prove it for bonds. The real yields are currently lower than at any period the Trinity study looked it. This is a fact and indisputable. If you treat rates as random walks then, on average, the future returns of treasuries are unequivocally lower than past ones.

Give a link to data?


In 1942, inflation was 11.35% and the 10 year was yielding 2.46% Those 1942 10 year investors got -2.9% real over the next 10 years. Today the 10 year tip yields -.03% of almost 3% better than those historical numbers. Obviously not quite an apples to apples but that seems a lot worse than today.
I mistyped that, my apologies. I was referring to 1966, since that was the year when 4% failed. My point was that real treasury bond yields are lower today than they were back in a period when 4% already failed.

My original quoted statement is hilariously wrong haha
3.8% still worked in 1966. Someone retiring that year got hit with rising interest rates, 16 years of the stock market going nowhere, AND double-digit inflation halfway through retirement in the late 70s and early 80s.

And still 3.8% worked.

For someone to say 3%, that means that next 30 years will have be worse than that.

We'll have to have even lower bond returns (possible), 20+ years of the stock market going nowhere (possible), or even higher double-digit inflation (possible).

All that is possible. But all three have to happen again, AND at least one of them has to happen worse than last time. So 4% seems pretty safe to most of us BECAUSE of 1966. 1966 had a lot go wrong, and you could still pull 3.8%.

Yes, the next 30 years might be even worse. If it is, I'll cut my vacations down to 2 a year instead of 4 a year (or just take 4 cheaper trips instead of expensive trips). I don't like to travel that much anyway... it's more my wife's thing.
The J stands for Jay

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Mon Sep 09, 2019 3:04 pm

305pelusa wrote:
Mon Sep 09, 2019 2:44 pm
I mistyped that, my apologies. I was referring to 1966, since that was the year when 4% failed. My point was that real treasury bond yields are lower today than they were back in a period when 4% already failed.
But nominal yields don't matter. Real yields do.

The cumulative real return for intermediate-term Treasuries from 1966-1981 was -33.2%. That's extremely unlikely to happen today with TIPS.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
305pelusa
Posts: 558
Joined: Fri Nov 16, 2018 10:20 pm

Re: Larry Swedroe: 3% is the new 4%

Post by 305pelusa » Mon Sep 09, 2019 3:17 pm

willthrill81 wrote:
Mon Sep 09, 2019 3:04 pm
305pelusa wrote:
Mon Sep 09, 2019 2:44 pm
I mistyped that, my apologies. I was referring to 1966, since that was the year when 4% failed. My point was that real treasury bond yields are lower today than they were back in a period when 4% already failed.
But nominal yields don't matter. Real yields do.

The cumulative real return for intermediate-term Treasuries from 1966-1981 was -33.2%. That's extremely unlikely to happen today with TIPS.
The implication being that had TIPS existed then, 4% would've been successful?

I hadn't thought about that. It's a good point

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Mon Sep 09, 2019 3:21 pm

305pelusa wrote:
Mon Sep 09, 2019 3:17 pm
willthrill81 wrote:
Mon Sep 09, 2019 3:04 pm
305pelusa wrote:
Mon Sep 09, 2019 2:44 pm
I mistyped that, my apologies. I was referring to 1966, since that was the year when 4% failed. My point was that real treasury bond yields are lower today than they were back in a period when 4% already failed.
But nominal yields don't matter. Real yields do.

The cumulative real return for intermediate-term Treasuries from 1966-1981 was -33.2%. That's extremely unlikely to happen today with TIPS.
The implication being that had TIPS existed then, 4% would've been successful?

I hadn't thought about that. It's a good point
Yes, it would have. Simply matching CPI with one's fixed income (i.e. 0% real yield) would have bumped the 30 year SWR up to 4.0% for 1966 retirees.

The real annualized yield of ITT from 1966-1981 was -2.5%. TIPS are still far better than that. And compared to Treasuries right now, the 'inflation insurance' of TIPS right now is very cheap, arguably free.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

rapporteur
Posts: 105
Joined: Fri Sep 12, 2014 10:15 pm

Re: Larry Swedroe: 3% is the new 4%

Post by rapporteur » Mon Sep 09, 2019 3:27 pm

Another take on SWR, this time by Big Ern (Karsten Jeske). His blog includes over 20 articles dealing with retirement and SWRs but the one most relevant to this thread is Part 20, “More Thoughts on Equity Glidepaths” in which he presents his own views and calculations while also addressing the views of Pfau and Kitces.
https://earlyretirementnow.com/2017/09/ ... lidepaths/

While much of his writing is focussed on early retirement (FIRE) with 60-year horizons, Part 20 contains much data on 30-year retirement conditional on two regimes, CAPE > 20 and CAPE < 20.

As I read him the main takeaways for 30-year retirements are:
  • There is no benefit whatsoever in equity allocations below 60% for either CAPE regime. For CAPE < 20 a fixed 100% equity allocation has been optimal. For CAPE > 20 the sweet spot is in the 65 – 85% range for fixed equity allocations and similarly for the average equity holding for a rising glidepath.
  • For CAPE < 20 no glidepath is necessary or desirable – a fixed equity allocation suffices
  • For CAPE > 20 a rapidly rising equity glidepath in the first 8 – 16 years followed by an equity plateau thereafter helps slightly, producing SWRs perhaps 25 basis points above those from holding a comparable fixed equity allocation from the outset
  • If a rising equity glidepath is used it should be confined to the first third/half or so of a 30-year retirement to get past the sequence of returns risk and then held high thereafter. The average equity holding over the entire 30 year period is what should be compared to a fixed equity allocation.
  • Kitces and Pfau rising equity glidepaths (e.g., 30 ==> 70%) are far too slow/gradual to be effective - they result in a low average equity holding over the full 30-year period (e.g., 50%) rather than being confined to the “bad sequence of returns” window of the first 10-15 years. Moreover, Big Ern believes that the shortcomings of naïve Monte Carlo (no reversion, fixed correlations, etc.) lead to distortions compared to using the historical record which Big Ern favours.
  • Big Ern confines himself to US investments. Although I disagree with him on this point, it’s a great help in making comparisons to Bengen, Trinity, etc.
But read him and decide for yourself.

Regards,

EnjoyIt
Posts: 2479
Joined: Sun Dec 29, 2013 8:06 pm

Re: Larry Swedroe: 3% is the new 4%

Post by EnjoyIt » Mon Sep 09, 2019 3:35 pm

willthrill81 wrote:
Mon Sep 09, 2019 3:21 pm
305pelusa wrote:
Mon Sep 09, 2019 3:17 pm
willthrill81 wrote:
Mon Sep 09, 2019 3:04 pm
305pelusa wrote:
Mon Sep 09, 2019 2:44 pm
I mistyped that, my apologies. I was referring to 1966, since that was the year when 4% failed. My point was that real treasury bond yields are lower today than they were back in a period when 4% already failed.
But nominal yields don't matter. Real yields do.

The cumulative real return for intermediate-term Treasuries from 1966-1981 was -33.2%. That's extremely unlikely to happen today with TIPS.
The implication being that had TIPS existed then, 4% would've been successful?

I hadn't thought about that. It's a good point
Yes, it would have. Simply matching CPI with one's fixed income (i.e. 0% real yield) would have bumped the 30 year SWR up to 4.0% for 1966 retirees.

The real annualized yield of ITT from 1966-1981 was -2.5%. TIPS are still far better than that. And compared to Treasuries right now, the 'inflation insurance' of TIPS right now is very cheap, arguably free.
Will,
Just out of curiosity do you hold TIPS now? Do you plan to hold TIPS in retirement? If no to both, do you have a scenario that will have you transitioning to TIPS and what is it?

I have been evaluating for a few years now if adding TIPS to my investment policy is a good idea and have yet to come to a conclusion which means I do not hold TIPS at this time.

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Mon Sep 09, 2019 3:49 pm

EnjoyIt wrote:
Mon Sep 09, 2019 3:35 pm
Will,
Just out of curiosity do you hold TIPS now? Do you plan to hold TIPS in retirement? If no to both, do you have a scenario that will have you transitioning to TIPS and what is it?

I have been evaluating for a few years now if adding TIPS to my investment policy is a good idea and have yet to come to a conclusion which means I do not hold TIPS at this time.
I currently own no fixed income at all. You may not be aware of this, but I am a trend follower, and my strategy is outlined here. I am not sure when I will take on a permanent fixed income allocation, but almost certainly not sooner than 10 years before my planned retirement age of 55.

But if I did have a permanent fixed income allocation right now, it would be split, probably evenly, between a stable value fund in my 401k currently paying 3.0% nominal and 30 year TIPS. If I didn't have access to the stable value fund, I would instead buy 5 year CDs paying 3% or slightly higher. TBM is not very appealing to me.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

EnjoyIt
Posts: 2479
Joined: Sun Dec 29, 2013 8:06 pm

Re: Larry Swedroe: 3% is the new 4%

Post by EnjoyIt » Mon Sep 09, 2019 4:47 pm

willthrill81 wrote:
Mon Sep 09, 2019 3:49 pm
EnjoyIt wrote:
Mon Sep 09, 2019 3:35 pm
Will,
Just out of curiosity do you hold TIPS now? Do you plan to hold TIPS in retirement? If no to both, do you have a scenario that will have you transitioning to TIPS and what is it?

I have been evaluating for a few years now if adding TIPS to my investment policy is a good idea and have yet to come to a conclusion which means I do not hold TIPS at this time.
I currently own no fixed income at all. You may not be aware of this, but I am a trend follower, and my strategy is outlined here. I am not sure when I will take on a permanent fixed income allocation, but almost certainly not sooner than 10 years before my planned retirement age of 55.

But if I did have a permanent fixed income allocation right now, it would be split, probably evenly, between a stable value fund in my 401k currently paying 3.0% nominal and 30 year TIPS. If I didn't have access to the stable value fund, I would instead buy 5 year CDs paying 3% or slightly higher. TBM is not very appealing to me.
Thanks for the response. I am aware that you do trend following but honestly since it is not my cup of tea I have not followed your post.

Seasonal
Posts: 450
Joined: Sun May 21, 2017 1:49 pm

Re: Larry Swedroe: 3% is the new 4%

Post by Seasonal » Tue Sep 10, 2019 7:53 am

HomerJ wrote:
Mon Sep 09, 2019 12:48 pm
(Technically, something worse than the Great Depression, since 4% worked for the Great Depression, and one COULD have continued to take European river cruises)
There is no necessary connection between the economy and the market. It's quite possible for one to do well and the other poorly.

MathIsMyWayr
Posts: 764
Joined: Mon Mar 27, 2017 10:47 pm
Location: CA

Re: Larry Swedroe: 3% is the new 4%

Post by MathIsMyWayr » Tue Sep 10, 2019 8:07 am

Seasonal wrote:
Tue Sep 10, 2019 7:53 am
HomerJ wrote:
Mon Sep 09, 2019 12:48 pm
(Technically, something worse than the Great Depression, since 4% worked for the Great Depression, and one COULD have continued to take European river cruises)
There is no necessary connection between the economy and the market. It's quite possible for one to do well and the other poorly.
Only for a short term. Over a longer term, they are related with each other.

Topic Author
Random Walker
Posts: 4098
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: 3% is the new 4%

Post by Random Walker » Tue Sep 10, 2019 10:37 pm

Larry has an essay over on Advisor Perspectives discussing the challenges of retirement in a low interest rate and high equity valuation environment.

Dave

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Wed Sep 11, 2019 12:30 am

Random Walker wrote:
Tue Sep 10, 2019 10:37 pm
Larry has an essay over on Advisor Perspectives discussing the challenges of retirement in a low interest rate and high equity valuation environment.

Dave
Thanks for letting us know about that Dave. For those interested, the link to that piece is here.

Larry does a good job explaining why forward returns for both stocks and bonds are expected to be lower.
In fact, with the CAPE 10 at 28.5 (August 15, 2019), the expected real return to U.S. stocks is now just 3.5% (1/28.5). If we use the current spread between 10-year nominal Treasury securitiees [sic] and 10-year TIPS as our estimate of inflation, the result (1.5%) produces an expected nominal return to U.S. stocks of just 5.0% (3.5 +1.5). If, instead, we use the current 10-year inflation swap rate of about 1.8% (a cleaner estimate of expected inflation), we get an expected nominal return of 5.3% (3.5 + 1.8). If we combine that 5.3% expected return with the five-year Treasury note now yielding at about 1.4%, the expected nominal return of that 60/40 portfolio is now only about 3.8%, or less than 40% of the return earned in the prior 37 years. (International stocks have lower valuations and thus higher expected returns. Using data from AQR Capital Management, at the end of the second quarter, the CAPE 10 earnings yield for international developed markets was 5.3% [versus 3.3% for the U.S.] and 6.9% for emerging markets.)
However, he misses a key point, one that Michael Kitces discovered back in 2012.
Accordingly, while 30-year returns have little relationship to safe withdrawal rates, returns over the first 15 years of the retirement time horizon have a much stronger relationship; in fact, the 15-year real (inflation-adjusted) return of the portfolio actually has a whopping 0.91 correlation to the safe withdrawal rate, as shown in the graph below.
So the first 15 years' of returns have seemingly been highly predictive of the forward 30 year SWR.
The average real return on a 60/40 (re-)balanced portfolio associated with the worst safe withdrawal rate scenarios in history was a mere 0.86% average annual compound growth rate over the first 15 years of retirement.
Larry is claiming that the expected real return of an all U.S. 60/40 portfolio over the next 10 years is roughly 2%. I would argue that 10 year TIPS would be a much safer bet for someone relying on 4% withdrawals than nominal Treasuries in this situation since Larry is stating that 5 year Treasuries have a -.4% expected return, while 10 year TIPS' real return today is .06% (30 year is .47%). TIPS also remove the risk of unexpected inflation from the fixed income side of the portfolio. International diversification of the equities would also increase the real portfolio return, so with these two changes, the real return of a 60/40 would increase to 2.7%.

While 2.7% real is nothing to crow about, and that's just for the next 10, not 15, years, it's still not low enough to seriously threaten the 4% rule going forward. It's certainly possible that the 4% rule won't hold over the next 30 years, but 2.7% real is unlikely to do it. If it's looking like real returns will be 1% or lower over the next 10-15 years, then there will be a lot more room for concern, IMHO.

And I'm disappointed that Larry is still recommending Monte Carlo simulations without using any disclaimers to the widely known fact that they produce 'fat tails' unless they specifically incorporate mean reversion.

Definitely not his best work.

P.S.: The real return of an all U.S. 60/40 portfolio from 2000-2009 was a very meager .38%, and yet year 2000 retirees who strictly held to the 4% rule are in fine shape today.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

marcopolo
Posts: 2216
Joined: Sat Dec 03, 2016 10:22 am

Re: Larry Swedroe: 3% is the new 4%

Post by marcopolo » Wed Sep 11, 2019 12:36 am

willthrill81 wrote:
Wed Sep 11, 2019 12:30 am
Random Walker wrote:
Tue Sep 10, 2019 10:37 pm
Larry has an essay over on Advisor Perspectives discussing the challenges of retirement in a low interest rate and high equity valuation environment.

Dave
Thanks for letting us know about that Dave. For those interested, the link to that piece is here.

Larry does a good job explaining why forward returns for both stocks and bonds are expected to be lower.
In fact, with the CAPE 10 at 28.5 (August 15, 2019), the expected real return to U.S. stocks is now just 3.5% (1/28.5). If we use the current spread between 10-year nominal Treasury securitiees [sic] and 10-year TIPS as our estimate of inflation, the result (1.5%) produces an expected nominal return to U.S. stocks of just 5.0% (3.5 +1.5). If, instead, we use the current 10-year inflation swap rate of about 1.8% (a cleaner estimate of expected inflation), we get an expected nominal return of 5.3% (3.5 + 1.8). If we combine that 5.3% expected return with the five-year Treasury note now yielding at about 1.4%, the expected nominal return of that 60/40 portfolio is now only about 3.8%, or less than 40% of the return earned in the prior 37 years. (International stocks have lower valuations and thus higher expected returns. Using data from AQR Capital Management, at the end of the second quarter, the CAPE 10 earnings yield for international developed markets was 5.3% [versus 3.3% for the U.S.] and 6.9% for emerging markets.)
However, he misses a key point, one that Michael Kitces discovered back in 2012.
Accordingly, while 30-year returns have little relationship to safe withdrawal rates, returns over the first 15 years of the retirement time horizon have a much stronger relationship; in fact, the 15-year real (inflation-adjusted) return of the portfolio actually has a whopping 0.91 correlation to the safe withdrawal rate, as shown in the graph below.
So the first 15 years' of returns have seemingly been highly predictive of the forward 30 year SWR.
The average real return on a 60/40 (re-)balanced portfolio associated with the worst safe withdrawal rate scenarios in history was a mere 0.86% average annual compound growth rate over the first 15 years of retirement.
Larry is claiming that the expected real return of an all U.S. 60/40 portfolio over the next 10 years is roughly 2%. I would argue that 10 year TIPS would be a much safer bet for someone relying on 4% withdrawals than nominal Treasuries in this situation since Larry is stating that 5 year Treasuries have a -.4% expected return, while 10 year TIPS' real return today is .06% (30 year is .47%). TIPS also remove the risk of unexpected inflation from the fixed income side of the portfolio. International diversification of the equities would also increase the real portfolio return, so with these two changes, the real return of a 60/40 would increase to 2.7%.

While 2.7% real is nothing to crow about, and that's just for the next 10, not 15, years, it's still not low enough to seriously threaten the 4% rule going forward. It's certainly possible that the 4% rule won't hold over the next 30 years, but 2.7% real or even half that return is unlikely to do it.

P.S.: The real return of an all U.S. 60/40 portfolio from 2000-2009 was a very meager .38%, and yet year 2000 retirees who strictly held to the 4% rule are in fine shape today.
Valuations, as measured by CAPE, which everyone seems to love, has actually dropped over 10% since January 2018 (~33 then and ~29 now). Yet, it seems like the fear mongering just gets more dire. Why aren't we hearing that we can spend more than we could have last year thanks to improving valuations?
Once in a while you get shown the light, in the strangest of places if you look at it right.

stlutz
Posts: 5334
Joined: Fri Jan 02, 2009 1:08 am

Re: Larry Swedroe: 3% is the new 4%

Post by stlutz » Wed Sep 11, 2019 12:42 am

willthrill81 wrote:
Wed Sep 11, 2019 12:30 am

Larry is claiming that the expected real return of an all U.S. 60/40 portfolio over the next 10 years is roughly 2%. I would argue that 10 year TIPS would be a much safer bet for someone relying on 4% withdrawals than nominal Treasuries in this situation since Larry is stating that 5 year Treasuries have a -.4% expected return, while 10 year TIPS' real return today is .06% (30 year is .47%). TIPS also remove the risk of unexpected inflation from the fixed income side of the portfolio. International diversification of the equities would also increase the real portfolio return, so with these two changes, the real return of a 60/40 would increase to 2.7%.

While 2.7% real is nothing to crow about, and that's just for the next 10, not 15, years, it's still not low enough to seriously threaten the 4% rule going forward. It's certainly possible that the 4% rule won't hold over the next 30 years, but 2.7% real is unlikely to do it. If it's looking like real returns will be 1% or lower over the next 10-15 years, then there will be a lot more room for concern, IMHO.
Swedroe is arguing that 2% represents a midpoint of a range of reasonably possible returns, not the floor. He doesn't state what that range is, which is what isn't really helpful for this discussion.

Topic Author
Random Walker
Posts: 4098
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: 3% is the new 4%

Post by Random Walker » Wed Sep 11, 2019 7:44 am

willthrill81 wrote:
Wed Sep 11, 2019 12:30 am

So the first 15 years' of returns have seemingly been highly predictive of the forward 30 year SWR.
Sequence of returns is a huge issue in the years leading up to retirement and in early retirement. As time period shortens, the importance of diversification across uncorrelated sources of return increases. I’m not familiar with the Monte Carlo Simulation weaknesses WillThrill mentions above. But MCS does show improved likelihoods of success when risks are more evenly spread across factors.

Dave

YRT70
Posts: 308
Joined: Sat Apr 27, 2019 8:51 am

Re: Larry Swedroe: 3% is the new 4%

Post by YRT70 » Wed Sep 11, 2019 9:28 am

Random Walker wrote:
Tue Sep 10, 2019 10:37 pm
Larry has an essay over on Advisor Perspectives discussing the challenges of retirement in a low interest rate and high equity valuation environment.

Dave
Is this the one you meant?

https://www.advisorperspectives.com/art ... nt-plans-1

Topic Author
Random Walker
Posts: 4098
Joined: Fri Feb 23, 2007 8:21 pm

Re: Larry Swedroe: 3% is the new 4%

Post by Random Walker » Wed Sep 11, 2019 9:38 am

YRT70 wrote:
Wed Sep 11, 2019 9:28 am
Random Walker wrote:
Tue Sep 10, 2019 10:37 pm
Larry has an essay over on Advisor Perspectives discussing the challenges of retirement in a low interest rate and high equity valuation environment.

Dave
Is this the one you meant?

https://www.advisorperspectives.com/art ... nt-plans-1
Yes

Dave

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Wed Sep 11, 2019 9:40 am

Random Walker wrote:
Wed Sep 11, 2019 7:44 am
willthrill81 wrote:
Wed Sep 11, 2019 12:30 am

So the first 15 years' of returns have seemingly been highly predictive of the forward 30 year SWR.
Sequence of returns is a huge issue in the years leading up to retirement and in early retirement. As time period shortens, the importance of diversification across uncorrelated sources of return increases. I’m not familiar with the Monte Carlo Simulation weaknesses WillThrill mentions above. But MCS does show improved likelihoods of success when risks are more evenly spread across factors.

Dave
The problem with Monte Carlo simulations is that most of them have the assumption that annual returns are independent of each other (i.e. no mean reversion allowed for). Derek Tharp noted on Michael Kitces' blog a couple of years ago that this creates serious problems with the results of such analyses.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Wed Sep 11, 2019 9:42 am

stlutz wrote:
Wed Sep 11, 2019 12:42 am
willthrill81 wrote:
Wed Sep 11, 2019 12:30 am

Larry is claiming that the expected real return of an all U.S. 60/40 portfolio over the next 10 years is roughly 2%. I would argue that 10 year TIPS would be a much safer bet for someone relying on 4% withdrawals than nominal Treasuries in this situation since Larry is stating that 5 year Treasuries have a -.4% expected return, while 10 year TIPS' real return today is .06% (30 year is .47%). TIPS also remove the risk of unexpected inflation from the fixed income side of the portfolio. International diversification of the equities would also increase the real portfolio return, so with these two changes, the real return of a 60/40 would increase to 2.7%.

While 2.7% real is nothing to crow about, and that's just for the next 10, not 15, years, it's still not low enough to seriously threaten the 4% rule going forward. It's certainly possible that the 4% rule won't hold over the next 30 years, but 2.7% real is unlikely to do it. If it's looking like real returns will be 1% or lower over the next 10-15 years, then there will be a lot more room for concern, IMHO.
Swedroe is arguing that 2% represents a midpoint of a range of reasonably possible returns, not the floor. He doesn't state what that range is, which is what isn't really helpful for this discussion.
Correct, but using TIPS instead of 5 year Treasuries and diversifying equities internationally bumps up expected real returns to triple the return of the the historic scenarios that led to the 4% rule. As I noted above, 2000-2009 was a much worse return period than what Larry is predicting, yet the 4% seems very likely to hold up for 30 years for that cohort.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

larryswedroe
Posts: 15973
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Larry Swedroe: 3% is the new 4%

Post by larryswedroe » Wed Sep 11, 2019 10:21 am

will
Careful
Larry is forecasting only the MEAN return, to incorporate all potential outcomes you would have to subtract 8%, and no guarantee because historically it got all returns within that band (either side) the future will look the same--meaning it could be worse.
Also there was no guarantee would have come out of that bear market and recover---see Japan.
One should not use the worst case as the base, but one should at least consider very bad states as potential and have plan Bs to enact if needed. One should NOT use the mean return as the only scenario
I would add there is no guarantee of mean reversion of course--again see Japan. But yes lower valuations (not poor returns) predict higher future returns.
Finally would add 2008 was in bottom 5% of the MCS runs we were using. So not "unexpected."
larry

EnjoyIt
Posts: 2479
Joined: Sun Dec 29, 2013 8:06 pm

Re: Larry Swedroe: 3% is the new 4%

Post by EnjoyIt » Wed Sep 11, 2019 3:26 pm

larryswedroe wrote:
Wed Sep 11, 2019 10:21 am
will
Careful
Larry is forecasting only the MEAN return, to incorporate all potential outcomes you would have to subtract 8%, and no guarantee because historically it got all returns within that band (either side) the future will look the same--meaning it could be worse.
Also there was no guarantee would have come out of that bear market and recover---see Japan.
One should not use the worst case as the base, but one should at least consider very bad states as potential and have plan Bs to enact if needed. One should NOT use the mean return as the only scenario
I would add there is no guarantee of mean reversion of course--again see Japan. But yes lower valuations (not poor returns) predict higher future returns.
Finally would add 2008 was in bottom 5% of the MCS runs we were using. So not "unexpected."
larry
I hate how we point to Japan as what can happen, but kind of gloss over the few years leading up to the crash. It was one hell of a run wasn't it?

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Wed Sep 11, 2019 3:45 pm

larryswedroe wrote:
Wed Sep 11, 2019 10:21 am
will
Careful
Larry is forecasting only the MEAN return, to incorporate all potential outcomes you would have to subtract 8%, and no guarantee because historically it got all returns within that band (either side) the future will look the same--meaning it could be worse.
Also there was no guarantee would have come out of that bear market and recover---see Japan.
One should not use the worst case as the base, but one should at least consider very bad states as potential and have plan Bs to enact if needed. One should NOT use the mean return as the only scenario
I would add there is no guarantee of mean reversion of course--again see Japan. But yes lower valuations (not poor returns) predict higher future returns.
Finally would add 2008 was in bottom 5% of the MCS runs we were using. So not "unexpected."
larry
A quick look at valuations would show that the U.S. market isn't valued even remotely as highly as Japan's was before its crash, so we can safely ignore Japan, at least for the time being.

You keep mentioning "guarantee." I don't believe that I've ever heard anyone around here at least claim that 4% fixed real dollar withdrawals over 30 years are guaranteed to work. Of course really bad things could have occurred in the past, and they could still occur in the future. But the point is that the '4% rule' did work over some very difficult situations, much worse than the 2-3% real returns that are expected of balanced portfolios over the next decade.

Have you ever met anyone who actually implemented the '4% rule'? I certainly haven't. Virtually everyone makes adjustments as they go along. The real question then is whether 4% continues to be a good starting point for retirees with a 30 year horizon, who should understand that if returns are very poor and/or there is a very poor sequence of returns that they will need to reduce their withdrawals accordingly. The data seems to indicate to me that it still is.

While I doubt that I will convince you of anything, I really urge you to reconsider the use of MCS that don't incorporate some level of mean reversion. If they don't, the bottom (and top) 5-7% of scenarios generated are too apocalyptic to warrant real planning. I don't really know that any retirement income strategy could survive Great Depression Part 2 immediately followed by 1970's stagflation, the dot-com bust, and the global financial crisis, but these are exactly the sorts of scenarios that most of these MCS produce. They are certainly scenarios that Bernstein would just state that we should not plan for because we cannot.

Where I wholeheartedly agree with you is for retirees to have a plan B in place (and preferably a plan C and maybe even a plan D). For many, this could involve buying a SPIA with the bulk of what's left in one's portfolio if a very poor return/sequence of returns period shows up. Reverse mortgages are another possibility. Relocating to a lower cost of living area is another.
Last edited by willthrill81 on Wed Sep 11, 2019 3:47 pm, edited 2 times in total.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

randomguy
Posts: 8089
Joined: Wed Sep 17, 2014 9:00 am

Re: Larry Swedroe: 3% is the new 4%

Post by randomguy » Wed Sep 11, 2019 3:46 pm

EnjoyIt wrote:
Wed Sep 11, 2019 3:26 pm

I hate how we point to Japan as what can happen, but kind of gloss over the few years leading up to the crash. It was one hell of a run wasn't it?
Of course. Hasn't March 2009-now been a hell of a run also?:) Ok it should be pointed out that the valuation increase hasn't been much. The US went from PE10 ~15 to 30. Japan went from 20 to 80 in about 7 years. Thats a big run up. The great depression was 5->30 in a decade. We haven't had the huge PE10 expansion that screams huge bubble. Maybe a small one:)

The real take away from Japan is to be careful of single country risk. You see a lot of international countries with low SWR and the obvious point is that most of them were invaded/bombed. The second is that a lot of them are small economies. When you calculate the SWR of like Switzerland, it is like calculating the GDP of New York. You can have an economy that isn't very diversified.

visualguy
Posts: 1510
Joined: Thu Jan 30, 2014 1:32 am

Re: Larry Swedroe: 3% is the new 4%

Post by visualguy » Wed Sep 11, 2019 3:47 pm

EnjoyIt wrote:
Wed Sep 11, 2019 3:26 pm
larryswedroe wrote:
Wed Sep 11, 2019 10:21 am
will
Careful
Larry is forecasting only the MEAN return, to incorporate all potential outcomes you would have to subtract 8%, and no guarantee because historically it got all returns within that band (either side) the future will look the same--meaning it could be worse.
Also there was no guarantee would have come out of that bear market and recover---see Japan.
One should not use the worst case as the base, but one should at least consider very bad states as potential and have plan Bs to enact if needed. One should NOT use the mean return as the only scenario
I would add there is no guarantee of mean reversion of course--again see Japan. But yes lower valuations (not poor returns) predict higher future returns.
Finally would add 2008 was in bottom 5% of the MCS runs we were using. So not "unexpected."
larry
I hate how we point to Japan as what can happen, but kind of gloss over the few years leading up to the crash. It was one hell of a run wasn't it?
Not really. Japan has returned just 3.9% real annualized for 1969-2018, and that includes the bubble period you are talking about, but starts well before that, and ends well after that. The annualized real return for 2000-2018 was 0.8% which is awful.

larryswedroe
Posts: 15973
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: Larry Swedroe: 3% is the new 4%

Post by larryswedroe » Wed Sep 11, 2019 3:53 pm

Will
MCS isn't perfect of course and has that limitation. But don't know any better tool. And note it makes no sense to assume IMO mean reversion since we don't know what the mean is, and you only get higher expected returns if valuations fall, not stock prices. Example might be that Dems win all three seats of power and not only reverse corp income tax cut but impose even higher rates. Forgetting what might happen to valuations, but earnings would drop say 20-25%. So stock prices fall sharply but no mean reversion should be expected. And finally, one should only treat MCS output as ESTIMATE of odds of success, not actual odds. But IMO it's about best tool we have.

And of course no one should use one rule like 4% without adjusting. We run new MCS annually as valuations and rates change that can cause assumptions on returns and likelihood of success to change.

Best wishes
Larry

visualguy
Posts: 1510
Joined: Thu Jan 30, 2014 1:32 am

Re: Larry Swedroe: 3% is the new 4%

Post by visualguy » Wed Sep 11, 2019 3:56 pm

willthrill81 wrote:
Wed Sep 11, 2019 3:45 pm
larryswedroe wrote:
Wed Sep 11, 2019 10:21 am
will
Careful
Larry is forecasting only the MEAN return, to incorporate all potential outcomes you would have to subtract 8%, and no guarantee because historically it got all returns within that band (either side) the future will look the same--meaning it could be worse.
Also there was no guarantee would have come out of that bear market and recover---see Japan.
One should not use the worst case as the base, but one should at least consider very bad states as potential and have plan Bs to enact if needed. One should NOT use the mean return as the only scenario
I would add there is no guarantee of mean reversion of course--again see Japan. But yes lower valuations (not poor returns) predict higher future returns.
Finally would add 2008 was in bottom 5% of the MCS runs we were using. So not "unexpected."
larry
A quick look at valuations would show that the U.S. market isn't valued even remotely as highly as Japan's was before its crash, so we can safely ignore Japan, at least for the time being.

You keep mentioning "guarantee." I don't believe that I've ever heard anyone around here at least claim that 4% fixed real dollar withdrawals over 30 years are guaranteed to work. Of course really bad things could have occurred in the past, and they could still occur in the future. But the point is that the '4% rule' did work over some very difficult situations, much worse than the 2-3% real returns that are expected of balanced portfolios over the next decade.

Have you ever met anyone who actually implemented the '4% rule'? I certainly haven't. Virtually everyone makes adjustments as they go along. The real question then is whether 4% continues to be a good starting point for retirees with a 30 year horizon, who should understand that if returns are very poor and/or there is a very poor sequence of returns that they will need to reduce their withdrawals accordingly. The data seems to indicate to me that it still is.

While I doubt that I will convince you of anything, I really urge you to reconsider the use of MCS that don't incorporate some level of mean reversion. If they don't, the bottom (and top) 5-7% of scenarios generated are too apocalyptic to warrant real planning. I don't really know that any retirement income strategy could survive Great Depression Part 2 immediately followed by 1970's stagflation, the dot-com bust, and the global financial crisis, but these are exactly the sorts of scenarios that most of these MCS produce. They are certainly scenarios that Bernstein would just state that we should not plan for because we cannot.

Where I wholeheartedly agree with you is for retirees to have a plan B in place (and preferably a plan C and maybe even a plan D). For many, this could involve buying a SPIA with the bulk of what's left in one's portfolio if a very poor return/sequence of returns period shows up. Reverse mortgages are another possibility. Relocating to a lower cost of living area is another.
I think what you're saying is quite similar to what Larry is saying. Basically, don't rely on 4%. You advocate being ready for adjusting withdrawals lower if and when needed by reducing discretionary spending or moving to a lower cost area (which all assumes you have scope for reduction), and he's talking about a lower withdrawal rate. Not all that radically different.

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Wed Sep 11, 2019 4:02 pm

larryswedroe wrote:
Wed Sep 11, 2019 3:53 pm
Will
MCS isn't perfect of course and has that limitation. But don't know any better tool. And note it makes no sense to assume IMO mean reversion since we don't know what the mean is, and you only get higher expected returns if valuations fall, not stock prices. Example might be that Dems win all three seats of power and not only reverse corp income tax cut but impose even higher rates. Forgetting what might happen to valuations, but earnings would drop say 20-25%. So stock prices fall sharply but no mean reversion should be expected. And finally, one should only treat MCS output as ESTIMATE of odds of success, not actual odds. But IMO it's about best tool we have.
I personally see no need at all to estimate 'the odds of success' because that often assumes that the retiree won't make adjustments along the way. Using something like VPW or the time value of money formula, which I prefer, with estimates of forward returns has a 100% 'success rate' in the sense that it is impossible for it to prematurely deplete a portfolio (although the government defaulting on its bond obligations and nationalizing the stock market certainly would). It just means that users will have to adjust their spending in accordance with how their portfolio performs, which is what they should largely be doing anyway.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Wed Sep 11, 2019 4:04 pm

visualguy wrote:
Wed Sep 11, 2019 3:56 pm
I think what you're saying is quite similar to what Larry is saying. Basically, don't rely on 4%. You advocate being ready for adjusting withdrawals lower if and when needed by reducing discretionary spending or moving to a lower cost area (which all assumes you have scope for reduction), and he's talking about a lower withdrawal rate. Not all that radically different.
The difference is that I still believe that it is perfectly prudent to use 4% withdrawals as a starting point to subsequently adjust from. Larry apparently believes that 3% is the new starting point. On that point, I strongly disagree.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

visualguy
Posts: 1510
Joined: Thu Jan 30, 2014 1:32 am

Re: Larry Swedroe: 3% is the new 4%

Post by visualguy » Wed Sep 11, 2019 4:12 pm

willthrill81 wrote:
Wed Sep 11, 2019 4:04 pm
visualguy wrote:
Wed Sep 11, 2019 3:56 pm
I think what you're saying is quite similar to what Larry is saying. Basically, don't rely on 4%. You advocate being ready for adjusting withdrawals lower if and when needed by reducing discretionary spending or moving to a lower cost area (which all assumes you have scope for reduction), and he's talking about a lower withdrawal rate. Not all that radically different.
The difference is that I still believe that it is perfectly prudent to use 4% withdrawals as a starting point to subsequently adjust from. Larry apparently believes that 3% is the new starting point. On that point, I strongly disagree.
Seems like a mostly meaningless distinction if you withdraw based on portfolio performance... You could say the same thing about starting with 5% or 2%, etc. It doesn't mean much if that's not how you actually determine your withdrawals beyond the start...

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Wed Sep 11, 2019 4:18 pm

visualguy wrote:
Wed Sep 11, 2019 4:12 pm
willthrill81 wrote:
Wed Sep 11, 2019 4:04 pm
visualguy wrote:
Wed Sep 11, 2019 3:56 pm
I think what you're saying is quite similar to what Larry is saying. Basically, don't rely on 4%. You advocate being ready for adjusting withdrawals lower if and when needed by reducing discretionary spending or moving to a lower cost area (which all assumes you have scope for reduction), and he's talking about a lower withdrawal rate. Not all that radically different.
The difference is that I still believe that it is perfectly prudent to use 4% withdrawals as a starting point to subsequently adjust from. Larry apparently believes that 3% is the new starting point. On that point, I strongly disagree.
Seems like a mostly meaningless distinction if you withdraw based on portfolio performance... You could say the same thing about starting with 5% or 2%, etc. It doesn't mean much if that's not how you actually determine your withdrawals beyond the start...
On the contrary, it makes a big difference. Someone who had 25x (i.e. could implement 4% withdrawals) in the year 2000 and had a 20% savings rate with a 60/40 AA would have had to wait until 2012 before they reached 33.3x (i.e. 3% withdrawals). That 12 year difference would have been very meaningful to most would-be retirees.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

EnjoyIt
Posts: 2479
Joined: Sun Dec 29, 2013 8:06 pm

Re: Larry Swedroe: 3% is the new 4%

Post by EnjoyIt » Wed Sep 11, 2019 4:23 pm

visualguy wrote:
Wed Sep 11, 2019 3:47 pm
EnjoyIt wrote:
Wed Sep 11, 2019 3:26 pm
larryswedroe wrote:
Wed Sep 11, 2019 10:21 am
will
Careful
Larry is forecasting only the MEAN return, to incorporate all potential outcomes you would have to subtract 8%, and no guarantee because historically it got all returns within that band (either side) the future will look the same--meaning it could be worse.
Also there was no guarantee would have come out of that bear market and recover---see Japan.
One should not use the worst case as the base, but one should at least consider very bad states as potential and have plan Bs to enact if needed. One should NOT use the mean return as the only scenario
I would add there is no guarantee of mean reversion of course--again see Japan. But yes lower valuations (not poor returns) predict higher future returns.
Finally would add 2008 was in bottom 5% of the MCS runs we were using. So not "unexpected."
larry
I hate how we point to Japan as what can happen, but kind of gloss over the few years leading up to the crash. It was one hell of a run wasn't it?
Not really. Japan has returned just 3.9% real annualized for 1969-2018, and that includes the bubble period you are talking about, but starts well before that, and ends well after that. The annualized real return for 2000-2018 was 0.8% which is awful.
I have seen these figures, but may I ask, does that include dividends? Does that include any form of rebalancing on the way up and on the way down?

I think what Japan teaches us is two things.
1) It took a serious run up to create such high valuations prior to the drop
2) As Randomguy pointed out up above " be careful of single country risk."

visualguy
Posts: 1510
Joined: Thu Jan 30, 2014 1:32 am

Re: Larry Swedroe: 3% is the new 4%

Post by visualguy » Wed Sep 11, 2019 4:25 pm

willthrill81 wrote:
Wed Sep 11, 2019 4:18 pm
visualguy wrote:
Wed Sep 11, 2019 4:12 pm
willthrill81 wrote:
Wed Sep 11, 2019 4:04 pm
visualguy wrote:
Wed Sep 11, 2019 3:56 pm
I think what you're saying is quite similar to what Larry is saying. Basically, don't rely on 4%. You advocate being ready for adjusting withdrawals lower if and when needed by reducing discretionary spending or moving to a lower cost area (which all assumes you have scope for reduction), and he's talking about a lower withdrawal rate. Not all that radically different.
The difference is that I still believe that it is perfectly prudent to use 4% withdrawals as a starting point to subsequently adjust from. Larry apparently believes that 3% is the new starting point. On that point, I strongly disagree.
Seems like a mostly meaningless distinction if you withdraw based on portfolio performance... You could say the same thing about starting with 5% or 2%, etc. It doesn't mean much if that's not how you actually determine your withdrawals beyond the start...
On the contrary, it makes a big difference. Someone who had 25x (i.e. could implement 4% withdrawals) in the year 2000 and had a 20% savings rate with a 60/40 AA would have had to wait until 2012 before they reached 33.3x (i.e. 3% withdrawals). That 12 year difference would have been very meaningful to most would-be retirees.
Yes, but if they may need to adjust down sometimes to 3% (or even lower) as you propose, then what should they save? My point is that bumping up the multiple or bumping up the planned-for expenses (to allow for reductions) is pretty much the same thing.

visualguy
Posts: 1510
Joined: Thu Jan 30, 2014 1:32 am

Re: Larry Swedroe: 3% is the new 4%

Post by visualguy » Wed Sep 11, 2019 4:28 pm

EnjoyIt wrote:
Wed Sep 11, 2019 4:23 pm
I have seen these figures, but may I ask, does that include dividends? Does that include any form of rebalancing on the way up and on the way down?
Yes, it includes dividends. No re-balancing.

EnjoyIt
Posts: 2479
Joined: Sun Dec 29, 2013 8:06 pm

Re: Larry Swedroe: 3% is the new 4%

Post by EnjoyIt » Wed Sep 11, 2019 4:28 pm

visualguy wrote:
Wed Sep 11, 2019 4:25 pm
willthrill81 wrote:
Wed Sep 11, 2019 4:18 pm
visualguy wrote:
Wed Sep 11, 2019 4:12 pm
willthrill81 wrote:
Wed Sep 11, 2019 4:04 pm
visualguy wrote:
Wed Sep 11, 2019 3:56 pm
I think what you're saying is quite similar to what Larry is saying. Basically, don't rely on 4%. You advocate being ready for adjusting withdrawals lower if and when needed by reducing discretionary spending or moving to a lower cost area (which all assumes you have scope for reduction), and he's talking about a lower withdrawal rate. Not all that radically different.
The difference is that I still believe that it is perfectly prudent to use 4% withdrawals as a starting point to subsequently adjust from. Larry apparently believes that 3% is the new starting point. On that point, I strongly disagree.
Seems like a mostly meaningless distinction if you withdraw based on portfolio performance... You could say the same thing about starting with 5% or 2%, etc. It doesn't mean much if that's not how you actually determine your withdrawals beyond the start...
On the contrary, it makes a big difference. Someone who had 25x (i.e. could implement 4% withdrawals) in the year 2000 and had a 20% savings rate with a 60/40 AA would have had to wait until 2012 before they reached 33.3x (i.e. 3% withdrawals). That 12 year difference would have been very meaningful to most would-be retirees.
Yes, but if they may need to adjust down sometimes to 3% (or even lower) as you propose, then what should they save? My point is that bumping up the multiple or bumping up the planned-for expenses (to allow for reductions) is pretty much the same thing.
It is, except one is traveling 4 times a year and the other is sitting on their porch drinking iced tea. To each their own.

visualguy
Posts: 1510
Joined: Thu Jan 30, 2014 1:32 am

Re: Larry Swedroe: 3% is the new 4%

Post by visualguy » Wed Sep 11, 2019 4:35 pm

EnjoyIt wrote:
Wed Sep 11, 2019 4:28 pm
visualguy wrote:
Wed Sep 11, 2019 4:25 pm
willthrill81 wrote:
Wed Sep 11, 2019 4:18 pm
visualguy wrote:
Wed Sep 11, 2019 4:12 pm
willthrill81 wrote:
Wed Sep 11, 2019 4:04 pm


The difference is that I still believe that it is perfectly prudent to use 4% withdrawals as a starting point to subsequently adjust from. Larry apparently believes that 3% is the new starting point. On that point, I strongly disagree.
Seems like a mostly meaningless distinction if you withdraw based on portfolio performance... You could say the same thing about starting with 5% or 2%, etc. It doesn't mean much if that's not how you actually determine your withdrawals beyond the start...
On the contrary, it makes a big difference. Someone who had 25x (i.e. could implement 4% withdrawals) in the year 2000 and had a 20% savings rate with a 60/40 AA would have had to wait until 2012 before they reached 33.3x (i.e. 3% withdrawals). That 12 year difference would have been very meaningful to most would-be retirees.
Yes, but if they may need to adjust down sometimes to 3% (or even lower) as you propose, then what should they save? My point is that bumping up the multiple or bumping up the planned-for expenses (to allow for reductions) is pretty much the same thing.
It is, except one is traveling 4 times a year and the other is sitting on their porch drinking iced tea. To each their own.
They both have to work as many years and save as much - that's the problem. How the discretionary extra money (if available) is used or not used is a different issue.

EnjoyIt
Posts: 2479
Joined: Sun Dec 29, 2013 8:06 pm

Re: Larry Swedroe: 3% is the new 4%

Post by EnjoyIt » Wed Sep 11, 2019 4:51 pm

visualguy wrote:
Wed Sep 11, 2019 4:35 pm
EnjoyIt wrote:
Wed Sep 11, 2019 4:28 pm
visualguy wrote:
Wed Sep 11, 2019 4:25 pm
willthrill81 wrote:
Wed Sep 11, 2019 4:18 pm
visualguy wrote:
Wed Sep 11, 2019 4:12 pm


Seems like a mostly meaningless distinction if you withdraw based on portfolio performance... You could say the same thing about starting with 5% or 2%, etc. It doesn't mean much if that's not how you actually determine your withdrawals beyond the start...
On the contrary, it makes a big difference. Someone who had 25x (i.e. could implement 4% withdrawals) in the year 2000 and had a 20% savings rate with a 60/40 AA would have had to wait until 2012 before they reached 33.3x (i.e. 3% withdrawals). That 12 year difference would have been very meaningful to most would-be retirees.
Yes, but if they may need to adjust down sometimes to 3% (or even lower) as you propose, then what should they save? My point is that bumping up the multiple or bumping up the planned-for expenses (to allow for reductions) is pretty much the same thing.
It is, except one is traveling 4 times a year and the other is sitting on their porch drinking iced tea. To each their own.
They both have to work as many years and save as much - that's the problem. How the discretionary extra money (if available) is used or not used is a different issue.
Actually the 2% guy doesn't need to work as much. While the 4% guy is out galavanting the world and squandering all those dollars. The 2% guy doesn't spend money outside of rocking chairs and iced tea saving up for future iced teas.

It all depends where you start and what your goal is. If your scenarios are two people with the exact same income and age then I think it would look a little something like this:
a) Work 5-12 more years to get to a 3% starting withdrawal and take 4 trips a year no matter what.
b) Start at 4% withdrawals and take 4 trips a year without working longer but accept the chance that one may need to cut back on some of those vacations.
c) Don't travel at all and withdraw at 3% without working longer.
d) Some middle-ground in between

Everyone gets to make their choice. I am choosing b. I like to travel knowing full well that life is too short and unpredictable.

marcopolo
Posts: 2216
Joined: Sat Dec 03, 2016 10:22 am

Re: Larry Swedroe: 3% is the new 4%

Post by marcopolo » Wed Sep 11, 2019 5:06 pm

larryswedroe wrote:
Wed Sep 11, 2019 3:53 pm
Will
MCS isn't perfect of course and has that limitation. But don't know any better tool. And note it makes no sense to assume IMO mean reversion since we don't know what the mean is, and you only get higher expected returns if valuations fall, not stock prices. Example might be that Dems win all three seats of power and not only reverse corp income tax cut but impose even higher rates. Forgetting what might happen to valuations, but earnings would drop say 20-25%. So stock prices fall sharply but no mean reversion should be expected. And finally, one should only treat MCS output as ESTIMATE of odds of success, not actual odds. But IMO it's about best tool we have.

And of course no one should use one rule like 4% without adjusting. We run new MCS annually as valuations and rates change that can cause assumptions on returns and likelihood of success to change.

Best wishes
Larry
It seems odd to me to rely on MCS without some form of memory (reversion) which treats each year as an independent event, AND at the same time believe that high valuations lead to lower expected returns. Either past results influence future returns (in which case the MCS should take that into account), or they don't (in which case valuations should not affect future expected returns).
How do you reconcile believing BOTH to be the case simultaneously?
Once in a while you get shown the light, in the strangest of places if you look at it right.

visualguy
Posts: 1510
Joined: Thu Jan 30, 2014 1:32 am

Re: Larry Swedroe: 3% is the new 4%

Post by visualguy » Wed Sep 11, 2019 5:08 pm

EnjoyIt wrote:
Wed Sep 11, 2019 4:23 pm
I think what Japan teaches us is two things.
1) It took a serious run up to create such high valuations prior to the drop
2) As Randomguy pointed out up above " be careful of single country risk."
The Japanese stock market has done poorly in the long run even ignoring their bubble that crashed. Like I said, only 3.9% real from 1969 to 2018. The Japanese economy grew tremendously during the first couple of decades of that period, and even today it is the third largest economy. It has a proper stock market (unlike China). Still, even with all that, and for all the growth and development they've had from 1969 (good luck seeing that again in our lifetime), we got only 3.9% real.

The lesson goes beyond the bubble, and I think their bubble isn't the most interesting part of the lesson. Europe didn't have the same bubble, but it also stagnated significantly. Could the US join that club in the future?

EnjoyIt
Posts: 2479
Joined: Sun Dec 29, 2013 8:06 pm

Re: Larry Swedroe: 3% is the new 4%

Post by EnjoyIt » Wed Sep 11, 2019 5:23 pm

visualguy wrote:
Wed Sep 11, 2019 5:08 pm
EnjoyIt wrote:
Wed Sep 11, 2019 4:23 pm
I think what Japan teaches us is two things.
1) It took a serious run up to create such high valuations prior to the drop
2) As Randomguy pointed out up above " be careful of single country risk."
The Japanese stock market has done poorly in the long run even ignoring their bubble that crashed. Like I said, only 3.9% real from 1969 to 2018. The Japanese economy grew tremendously during the first couple of decades of that period, and even today it is the third largest economy. It has a proper stock market (unlike China). Still, even with all that, and for all the growth and development they've had from 1969 (good luck seeing that again in our lifetime), we got only 3.9% real.

The lesson goes beyond the bubble, and I think their bubble isn't the most interesting part of the lesson. Europe didn't have the same bubble, but it also stagnated significantly. Could the US join that club in the future?
No doubt, anything is possible.
Just curious, how would a Japanese 4% withdrawal investor do at 60/40 portfolio starting in 1969?
I say that because 3.9% real returns is pretty good and a 4% withdrawal portfolio would easily last 30 years if 3.9% consistently the average. The other way, if 3.9% is because of massive returns early in retirement and horrible returns late, then 3.9% was fantastic. The opposite is true if it came in reverse. We all know that sequence of returns matters.

visualguy
Posts: 1510
Joined: Thu Jan 30, 2014 1:32 am

Re: Larry Swedroe: 3% is the new 4%

Post by visualguy » Wed Sep 11, 2019 5:36 pm

EnjoyIt wrote:
Wed Sep 11, 2019 4:51 pm
Actually the 2% guy doesn't need to work as much. While the 4% guy is out galavanting the world and squandering all those dollars. The 2% guy doesn't spend money outside of rocking chairs and iced tea saving up for future iced teas.

It all depends where you start and what your goal is. If your scenarios are two people with the exact same income and age then I think it would look a little something like this:
a) Work 5-12 more years to get to a 3% starting withdrawal and take 4 trips a year no matter what.
b) Start at 4% withdrawals and take 4 trips a year without working longer but accept the chance that one may need to cut back on some of those vacations.
c) Don't travel at all and withdraw at 3% without working longer.
d) Some middle-ground in between

Everyone gets to make their choice. I am choosing b. I like to travel knowing full well that life is too short and unpredictable.
That's exactly my point: b and c are the same in terms of how long you have to work and how much you need to save.

There may be "excess" money in the portfolio in both cases, which is used for vacations in b and for increasing the portfolio growth in c, but whether you use it for this or that is an orthogonal issue - you worked the same number of years, and saved the same amount to reach your retirement "number".

User avatar
fortyofforty
Posts: 1585
Joined: Wed Mar 31, 2010 12:33 pm

Re: Larry Swedroe: 3% is the new 4%

Post by fortyofforty » Wed Sep 11, 2019 5:55 pm

notinuse wrote:
Sun Feb 03, 2019 10:29 am
zaboomafoozarg wrote:
Sun Feb 03, 2019 10:01 am
Personally, I wouldn't consider retiring with a 4% withdrawal rate. But I am shooting for an early retirement age and 40-50 years of retirement if I'm lucky.

Currently debating between 2.5% and 3% withdrawal, but haven't decided. I've got maybe 10 more years to figure it out.
That makes sense, since the original basis for the 4% guideline was a 30 year retirement. Not 40-50 years.
I would say that at no time in human history has anyone expected a retirement that lasted 40-50 years. Half a century living off working for what, about the same amount of time or less?
"In a time of universal deceit, telling the truth becomes a revolutionary act." - George Orwell | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Wed Sep 11, 2019 6:48 pm

visualguy wrote:
Wed Sep 11, 2019 4:35 pm
EnjoyIt wrote:
Wed Sep 11, 2019 4:28 pm
visualguy wrote:
Wed Sep 11, 2019 4:25 pm
willthrill81 wrote:
Wed Sep 11, 2019 4:18 pm
visualguy wrote:
Wed Sep 11, 2019 4:12 pm


Seems like a mostly meaningless distinction if you withdraw based on portfolio performance... You could say the same thing about starting with 5% or 2%, etc. It doesn't mean much if that's not how you actually determine your withdrawals beyond the start...
On the contrary, it makes a big difference. Someone who had 25x (i.e. could implement 4% withdrawals) in the year 2000 and had a 20% savings rate with a 60/40 AA would have had to wait until 2012 before they reached 33.3x (i.e. 3% withdrawals). That 12 year difference would have been very meaningful to most would-be retirees.
Yes, but if they may need to adjust down sometimes to 3% (or even lower) as you propose, then what should they save? My point is that bumping up the multiple or bumping up the planned-for expenses (to allow for reductions) is pretty much the same thing.
It is, except one is traveling 4 times a year and the other is sitting on their porch drinking iced tea. To each their own.
They both have to work as many years and save as much - that's the problem.
No, they don't. I just pointed out that a year 2000 retiree with 25x could have retired with 4% withdrawals then. Had they wanted to get to 33.3X and 3% withdrawals, they would have had to delay retirement for 12 more years.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

visualguy
Posts: 1510
Joined: Thu Jan 30, 2014 1:32 am

Re: Larry Swedroe: 3% is the new 4%

Post by visualguy » Wed Sep 11, 2019 7:00 pm

willthrill81 wrote:
Wed Sep 11, 2019 6:48 pm
visualguy wrote:
Wed Sep 11, 2019 4:35 pm
EnjoyIt wrote:
Wed Sep 11, 2019 4:28 pm
visualguy wrote:
Wed Sep 11, 2019 4:25 pm
willthrill81 wrote:
Wed Sep 11, 2019 4:18 pm


On the contrary, it makes a big difference. Someone who had 25x (i.e. could implement 4% withdrawals) in the year 2000 and had a 20% savings rate with a 60/40 AA would have had to wait until 2012 before they reached 33.3x (i.e. 3% withdrawals). That 12 year difference would have been very meaningful to most would-be retirees.
Yes, but if they may need to adjust down sometimes to 3% (or even lower) as you propose, then what should they save? My point is that bumping up the multiple or bumping up the planned-for expenses (to allow for reductions) is pretty much the same thing.
It is, except one is traveling 4 times a year and the other is sitting on their porch drinking iced tea. To each their own.
They both have to work as many years and save as much - that's the problem.
No, they don't. I just pointed out that a year 2000 retiree with 25x could have retired with 4% withdrawals then. Had they wanted to get to 33.3X and 3% withdrawals, they would have had to delay retirement for 12 more years.
The planned annual expenses aren't the same for these two cases. You aren't taking into account the needed room for adjustment that you assume implicitly. The first case (4%) needs to include a budget for 4 vacations a year (in our example, or other slack) that can be cut, while the second case (3%) doesn't.

User avatar
willthrill81
Posts: 12597
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Larry Swedroe: 3% is the new 4%

Post by willthrill81 » Wed Sep 11, 2019 7:11 pm

visualguy wrote:
Wed Sep 11, 2019 7:00 pm
willthrill81 wrote:
Wed Sep 11, 2019 6:48 pm
visualguy wrote:
Wed Sep 11, 2019 4:35 pm
EnjoyIt wrote:
Wed Sep 11, 2019 4:28 pm
visualguy wrote:
Wed Sep 11, 2019 4:25 pm


Yes, but if they may need to adjust down sometimes to 3% (or even lower) as you propose, then what should they save? My point is that bumping up the multiple or bumping up the planned-for expenses (to allow for reductions) is pretty much the same thing.
It is, except one is traveling 4 times a year and the other is sitting on their porch drinking iced tea. To each their own.
They both have to work as many years and save as much - that's the problem.
No, they don't. I just pointed out that a year 2000 retiree with 25x could have retired with 4% withdrawals then. Had they wanted to get to 33.3X and 3% withdrawals, they would have had to delay retirement for 12 more years.
The planned annual expenses aren't the same for these two cases. You aren't taking into account the needed room for adjustment that you assume implicitly. The first case (4%) needs to include a budget for 4 vacations a year (in our example, or other slack) that can be cut, while the second case (3%) doesn't.
I still don't see your point. What's wrong with starting your withdrawals at 4% and then cutting back to 3% if needed? Why take '3% medicine' if you're still healthy?
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

Post Reply