New Wade Pfau article of great interest

[kevink]

In the article linked to below, Pfau makes what seems to me to be a very important point about the difference between the allocation in Bengen's pioneering work that led to the 4% SWR idea and that used in the subsequent Trinity study.

Historial SWR's increase considerably when IT Treasuries are used, rather than corporate bonds, due to the flight to safety effect. We saw this writ large in 2008 of course, but the article makes it clear it's a general principle worth knowing about.

Safe Withdrawal Rates for Retirement and the Trinity Study

 

[Options]

Yes. Saw it this morning and took another look at his much touted "dashboard".

Is now a good time to retire? Retirement Researcher

As of 4/1/15, he advises that 2.29% would be the new 4% for a conservative couple. Note his assumptions use a .5% adminitrative fees. He recently stated for those with lower fees (such as my .09% fees at VG), one could add as much as .25% back into that 2.29% figure. So we're back to 2.5% is roughly the new 4% for a conservative couple over 30 years.

I've yet to see anyone other than Bernstein as conservative as Pfau. His caution is well taken, but his assumptions still amount to predictions IMO. Not that I don't highly respect his work, just keeping an open mind on these things.

 

[kevink]

Options you're onto something of vital importance, it seems to me, in this ultra low-return environment: cutting investing costs to the bone. Mr. Bogle is calling for 2% from bonds and 5% from stocks for the next 20 years, so Pfau and Bernsteing are in good company in calling for dialed-down expectations.

Worth reminding ourselves of is that all of the number Pfau and his predecessors talk about are adjusted for inflation. Personally I work on living on 4% of actual (no adjustment) at all times, and have found that as an ER inflation need not affect me as thoroughly as it does someone working full time.

 

[FUEGO]

Options you're onto something of vital importance, it seems to me, in this ultra low-return environment: cutting investing costs to the bone. Mr. Bogle is calling for 2% from bonds and 5% from stocks for the next 20 years, so Pfau and Bernsteing are in good company in calling for dialed-down expectations.

Was that real or nominal returns? I think I've seen similar real returns from a recent Vanguard forecast.

 

[clifp]

Was that real or nominal returns? I think I've seen similar real returns from a recent Vanguard forecast.

Based on this article it appears to be nominal return.

I think Bogle is more likely to be right than wrong.

Current yield is excellent predictor of the 5-10 year yield of bonds.

When the next correction will be and how deep it will be is pretty much anybody guess, but I think Jack is far from predicting doom or gloom.

 

[Jack_Pine]

Options you're onto something of vital importance, it seems to me, in this ultra low-return environment: cutting investing costs to the bone. Mr. Bogle is calling for 2% from bonds and 5% from stocks for the next 20 years, so Pfau and Bernsteing are in good company in calling for dialed-down expectations.

Worth reminding ourselves of is that all of the number Pfau and his predecessors talk about are adjusted for inflation. Personally I work on living on 4% of actual (no adjustment) at all times, and have found that as an ER inflation need not affect me as thoroughly as it does someone working full time.

Nicely said. This summarizes my current thinking as well. Given the fact that I don't intend to adjust for inflation, I may start closer to 5%. Will see what the environment is like in a year or two when we pull the plug.

 

[utrecht]

I don't believe that Bogle, Pfau, or anyone else can predict stock market returns for the next 20 years.

 

[OrcasIslandBound]

It's really LT treasuries that are pretty good at having negative correlation with stocks. LT corporate bonds pay a higher dividend, but have a significant default rate during serious market downturns causing them to potentially have a positive correlation with stocks. However, if the Fed needs to raise interest rates to fight inflation, the LT treasuries become positively correlated with stocks also. This means both will go down simultaneously when interest rates are pulled up to fight inflation.

My strategy is to use a portion of long term treasury bonds to take advantage of the fact that usually they are negatively correlated with stocks.

Sent from my Nexus 4 using Early Retirement Forum mobile app

 

[OrcasIslandBound]

My family's goal is to have enough return with a minimum of volatility. We're at 45 %stock index ETFs and 55% bonds. We hope for about a 6% return overall and plan on a 3.5% safe withdrawal before SS, then 2.5% thereafter. We'll retire in 2 years at 59.

Sent from my Nexus 4 using Early Retirement Forum mobile app

 

[Texas Proud]

I don't believe that Bogle, Pfau, or anyone else can predict stock market returns for the next 20 years.

Actually they all can and all have....


The question is if their prediction is accurate or not....

 

[Spanky]

"4% rule is not safe… 1.8% is the new SWR .... " so what's next based on the assumption/prediction of significantly lower returns of stock and bonds?

 

[Spanky]

Actually they all can and all have....


The question is if their prediction is accurate or not....

agreed - Anyone can! It's only a question of accuracy or certainty.

 

[kevink]

True enough Texas Proud, but I appreciate Bogle and Pfau's pessimism-cum-realism and would rather be pleasantly surprised using their numbers as a basis for my expectations.

OrcasIslandBound (great name by the way - Orcas Island is one of my favorite places on earth!) you sound like a kindred spirit in terms of risk/return appetite. According to Paul Merriman's risk tables a 40:60 bond:stock allocation has a worst one year return of .23%, so you'd be taking on a bit more than that with your "conservative" 55:45 (I'm in the same boat), unless, as you say, you do a bond "barbell" of half 30 year Treasuries and the rest short term, a la Harry Browne's Permanent Portfolio.

I'll post a link to the historical risk:return matrix of the PP to a plain vanilla stock:bond portfolio below, but as with all of the other approaches based on backtesting (which is to say all approaches!) it seems to me several things that have given the PP such a great run are unlikely to be repeated (e.g. the one-time event of opening up the gold market in the 1970's). William Bernstein, in his recent book "Deep Risk: How History Informs Portfolio Design" does a great job of showing the limitations of the PP and other defensive "bunker" approaches, but (getting back to the Pfau article) I don't think he'd have any problem with having most if not all of one's bond allocation in IT Treasuries.

Worst 3 year PP performance ever? - Page 2

 

[marko]

"4% rule is not safe… 1.8% is the new SWR .... " so what's next based on the assumption/prediction of significantly lower returns of stock and bonds?

So if you've been using 4% for the past 5-10 years...should you now switch to 2% or so?

Might some folks have to go back to w*rk after being led astray by the same 'experts'?!

2% is the new 4%? In 5 years will it be 1%?

Are all these dire predictions based on foreseeable market performance or only based upon the current environment?

I'd hate to die really rich if these guys are wrong.

 

[W2R]

So if you've been using 4% for the past 5-10 years...should you now switch to 2% or so?

Might some folks have to go back to w*rk after being led astray by the same 'experts'?!

2% is the new 4%? In 5 years will it be 1%?

Are all these dire predictions based on foreseeable market performance or only based upon the current environment?

I'd hate to die really rich if these guys are wrong.

There are no guarantees in life, as I'm sure you know. We can read what all the so-called experts think, and then make our own decisions. That's really all we can do.

Personally I think I am very conservative financially, but 1% seems ridiculous even to me. I could be wrong! But right now I see nothing wrong with 3% or more.

When reading various predictions, it helps to check into the business interests of those doing the predicting. For example, I am sure the annuity pushers would tend to recommend a lower withdrawal rate from investments, and assert that an annuity is the only sure way to secure an income stream in retirement.

 

[utrecht]

I don't believe that Bogle, Pfau, or anyone else can correctly predict stock market returns for the next 20 years.

Fixed my own post

 

[photoguy]

The question is if their prediction is accurate or not....

This is basically untestable within our lifespan. They are generally making a forecast about the "expected return" which is a probability weighted average. So if Pfau says the expected return was 5% and the actual return was 7%, it doesn't mean he was necessarily wrong (or right).

 

[Texas Proud]

I just do not see how they can say that the new normal is so low....

Say you get zero real return... and we are talking about their time frame which is 30 years.... then a SWR is 3.33%..... 100/30.... sure, you will have zero when you are done, but that is not a failure in their terms....

Even they say that there should be a positive real return... so it should be higher than that...

Now, you do have the sequence or return problem to deal with... but I do not see them saying that is the problem they are trying to solve with the new rate...

 

[GrayHare]

However, if the Fed needs to raise interest rates to fight inflation, the LT treasuries become positively correlated with stocks also. This means both will go down simultaneously when interest rates are pulled up to fight inflation.

Historically this has been true only when the 10 year rate is over 5%. The reverse has been true under 5%.

 

[Cut-Throat]

"4% rule is not safe… 1.8% is the new SWR .... " so what's next based on the assumption/prediction of significantly lower returns of stock and bonds?

The problem is not "What is the SWR today"....The problem is using an SWR as a withdrawal method. (Which no one ever has anyway, so I fail to see the interest in this topic) The SWR was a 'Rule of Thumb'. If you use a Withdrawal Method that is Variable based on Portfolio Balance, you really don't have to care.

Say you start taking a 5% of Portfolio Balance and your Portfolio drops in Half due to market meltdown, you're now effectively taking 2.5% of your Portfolio. So, why start your spending with a 'worst case' Withdrawal, if you may not need to?

 

[bmcgonig]

I just do not see how they can say that the new normal is so low....

Say you get zero real return... and we are talking about their time frame which is 30 years.... then a SWR is 3.33%..... 100/30.... sure, you will have zero when you are done, but that is not a failure in their terms....

Even they say that there should be a positive real return... so it should be higher than that...

Now, you do have the sequence or return problem to deal with... but I do not see them saying that is the problem they are trying to solve with the new rate...

I think that Pfau IS saying that. His simulations are using a lower future bond rate, taking a while to normalize, and thus getting a lower SWR if I read it correctly.

The low predictions are due to the combination of the lousy interest rate in the bond market, and the high PE in the stock market. Bernanke is said to have said that interest rates won't normalize in his lifetime. That's ammunition for Pfau's prediction. Others, like Meb Faber point to high CAPEs at present as indicative of low future stock returns. I.e., when CAPE is in the top decile, stock returns will be in the lowest decile, or something to that effect. Probably similar thinking to Bogle etc. They may be just predictions but they seem to have some foundation.


Sent from my iPad using Early Retirement Forum

 

[Options]

I just do not see how they can say that the new normal is so low....

Say you get zero real return... and we are talking about their time frame which is 30 years.... then a SWR is 3.33%..... 100/30.... sure, you will have zero when you are done, but that is not a failure in their terms....

Even they say that there should be a positive real return... so it should be higher than that...

Now, you do have the sequence or return problem to deal with... but I do not see them saying that is the problem they are trying to solve with the new rate...

Again, Bernstein has said that 2% is "bulletproof" and 3% is "probably ok". As I said, I have seen no one as conservative as Pfau (if anyone has, please chime in).

It's also worth noting the 2.5% inflation adjusted (with reduced fees considered) WR example I gave above from Pfau's work is for a 25% equities PF. His table shows that higher equity allocations create a higher WR rate (2.78% for a 50/50 PF, throw in .25% for those with much lower fees, and you're at 3%). Also observe the higher WR rates using the Guyten decision rules. Point is, it's not all that dire.

Something else he shows is that one can have higher WR rates with bonds and annuities--a case for annuities as a plan B or C, perhaps. Then again, his work is sponsored by the insurance industry so make your own assumptions (personally that doesn't bother me as I've seen good arguments discussed elsewhere for delayed (vs. deferred) SPIA's as an alternative to the cat food/cardboard box scenario in later years).

IMO, it's worth remembering the upside to participation in the bond/equity markets, which Pfau aknowledges and specifically notes that his WR's do not show that upside. In my experience, ESPlanner does a brilliant job of showing what that upside could be, as well as the the % of possible time each scenario could occur (using MC simulations). It's for this reason I'm not too worried about low starting WR rates. Probabilities are they could be adjusted upwards if one manages to avoid the most worst case PF scenarios).

Finally, I agree with the above that noone, but noone, can predict the future (as beautifully pointed out by either Charlie Munger or Buffet in Berkshire's most recent annual letter). Pfua's numbers are based on expected returns, and it's good to remember that. In the markets, anything can happen, has happened, and will happen. Only on that much we can count on. The variation I've seen on Sir John Templeton's quote comes to mind: "“The four most expensive (he uses dangerous") words in investing are 'This time it's different.'”

 

[kevink]

I think that Pfau IS saying that. His simulations are using a lower future bond rate, taking a while to normalize, and thus getting a lower SWR if I read it correctly.

The low predictions are due to the combination of the lousy interest rate in the bond market, and the high PE in the stock market. Bernanke is said to have said that interest rates won't normalize in his lifetime. That's ammunition for Pfau's prediction. Others, like Meb Faber point to high CAPEs at present as indicative of low future stock returns. I.e., when CAPE is in the top decile, stock returns will be in the lowest decile, or something to that effect. Probably similar thinking to Bogle etc. They may be just predictions but they seem to have some foundation.


Sent from my iPad using Early Retirement Forum

Thanks for the great summary bmcgonig. Your point about CAPE is echoed in Bill Bernstein's most recent writing as well as that of Rick Ferri, Larry Swedroe, Jack Bogle and Paul Meriman, among others.

Bernstein's recommendations in particular ring true to me (for those in or near retirement, of course, since his advice to millenials is very different): high quality bonds, signficant international equity exposure with a value tilt, perhaps a smattering of natural resources and commodity stocks.

Pfau certainly is a proponent of annuitizing at least part of one's porfolio at the right time, but I haven't read anything by him yet that has made me think he's so biased in that regard that it skews his recommendations. I may well be wrong.

 

[audreyh1]

I just do not see how they can say that the new normal is so low....

Say you get zero real return... and we are talking about their time frame which is 30 years.... then a SWR is 3.33%..... 100/30.... sure, you will have zero when you are done, but that is not a failure in their terms....

Even they say that there should be a positive real return... so it should be higher than that...

Now, you do have the sequence or return problem to deal with... but I do not see them saying that is the problem they are trying to solve with the new rate...

Me too - 2.3% as an SWR strikes me as absurd.

Oh, I see that's for a 25% equities portfolio - never mind then.....

 

[Spanky]

This is basically untestable within our lifespan. They are generally making a forecast about the "expected return" which is a probability weighted average. So if Pfau says the expected return was 5% and the actual return was 7%, it doesn't mean he was necessarily wrong (or right).

Here's an answer: