Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
CULater
Posts: 1513
Joined: Sun Nov 13, 2016 10:59 am

Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CULater » Sun Aug 12, 2018 11:28 am

Study by Michael Kitces that illustrates the importance of real portfolio returns during the first decade of retirement for portfolio survival, and the relationship of valuation measures such as CAPE. Bottom line: this is not a great time to retire and begin portfolio drawdown unless real drawdowns can be kept relatively small. But who knows what they will need to withdraw? -- stuff happens. Happy sailing into the Golden Years!

it turns out that the true driver of sequence of return risk and safe withdrawal rates are the returns that the retiree earns over the first decade – and specifically, the real returns over the first decade, that provide an indication of whether the retirement portfolio will have produced enough real growth to keep up with inflation-adjusted spending for the rest of retirement. Fortunately, though, bad decades of returns are not entirely random, and instead can be reasonably predicted by long-term market valuation trends, providing retirees with at least a few tools to manage the dangers of sequence of return risk through adjusting asset allocation in retirement and setting a reasonable initial withdrawal rate in light of the market conditions that exist – and the potential for a bad decade of returns – when their retirement begins
the fact that 10-year equity returns are so predictive (even of the safe withdrawal rate from a 60/40 portfolio) is also why market valuation can be so effective as an indicator for retirement – because measures like Shiller P/E10 actually do a good job predicting real returns over a decade or more, which is the exact time horizon that matters for sequence of returns risk. In fact, as shown below (and as I’ve also shown in prior research on market valuation and safe withdrawal rates), if we look at the earnings yield of stocks using Shiller methodology (E10/P, or CAEP) and compare it to the 30-year SWR, the correlation is a remarkable 0.77! Market valuation and earnings yields at the start of retirement are remarkably predictive of 30-year safe withdrawal rates!
https://www.kitces.com/blog/understandi ... d-decades/
May you have the hindsight to know where you've been, The foresight to know where you're going, And the insight to know when you've gone too far. ~ Irish Blessing

michaeljc70
Posts: 3746
Joined: Thu Oct 15, 2015 3:53 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by michaeljc70 » Mon Aug 13, 2018 8:34 am

But, as I understand it, this doesn't change the floor for a SWR rate from his previous studies. A .77 correlation is not something I would take to the bank. What if you work 10 more years and valuations are still high?

heyyou
Posts: 3175
Joined: Tue Feb 20, 2007 4:58 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by heyyou » Mon Aug 13, 2018 11:13 am

Someone (Shiller?) noted that those with good returns during their accumulation years, should expect to have lower returns in their retirement years, and vice versa. Avoid spending methods based on only the retirement day asset value.

Spending a calculated percentage of the recent portfolio balance helps portfolio longevity, as does McClung's suggestion to adjust your initial WD % with respect to current valuations. He divides the historical CAPE and valuations (foreign and domestic) into quartiles, then averages those 1-4 numbers to see how the current number fits with history, changing the initial WD rate by a few tenths of a percent, to buffer what is expected, either more or less. A few tenths to, or from, 4% is a noticeable change.

dcabler
Posts: 663
Joined: Wed Feb 19, 2014 11:30 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by dcabler » Mon Aug 13, 2018 11:26 am

michaeljc70 wrote:
Mon Aug 13, 2018 8:34 am
But, as I understand it, this doesn't change the floor for a SWR rate from his previous studies. A .77 correlation is not something I would take to the bank. What if you work 10 more years and valuations are still high?
By then you'd have saved for 10 more years and would be 10 years closer to death, both of which work in your favor for portfolio survivability vs. not.

Topic seems to at least indirectly provide another reason to have a variable/flexible withdrawal plan.

bgyt
Posts: 64
Joined: Sat Jun 16, 2018 8:20 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by bgyt » Mon Aug 13, 2018 11:34 am

CULater wrote:
Sun Aug 12, 2018 11:28 am
Bottom line: this is not a great time to retire and begin portfolio drawdown unless real drawdowns can be kept relatively small
I would not draw such a broad conclusion from this excellent article (Thanks for posting).

More narrowly, if one’s retirement plan has them cutting it close to the SWR threshold with an AA majority in Equities then yes, you are wildly exposed to current market overvaluation and Sequence of Returns risk.

Yes, overvaluation means our portfolio are a lot fatter than normal. If we don’t account for the possibility of Reversion to the Mean, especially in the SoR window, then shame on us.
Last edited by bgyt on Mon Aug 13, 2018 1:31 pm, edited 1 time in total.

randomguy
Posts: 6504
Joined: Wed Sep 17, 2014 9:00 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by randomguy » Mon Aug 13, 2018 11:56 am

dcabler wrote:
Mon Aug 13, 2018 11:26 am
michaeljc70 wrote:
Mon Aug 13, 2018 8:34 am
But, as I understand it, this doesn't change the floor for a SWR rate from his previous studies. A .77 correlation is not something I would take to the bank. What if you work 10 more years and valuations are still high?
By then you'd have saved for 10 more years and would be 10 years closer to death, both of which work in your favor for portfolio survivability vs. not.

Topic seems to at least indirectly provide another reason to have a variable/flexible withdrawal plan.

You will also likely have 2x as much money if valuations are still high between investment returns and savings.

But lets look at the bottom chart
-~3% to +5% real over the first 10 years all have some chance of having SWRs between 4 and 5%. That is a pretty narrow range for a wide range of returns

As far as predicting results, CAPE10 is tough. When the mid 90s crowd with their 20+ (say 1994 with its 19.77 or even 1997 with it 28) high PE10s are likely to show better results. That 1997 guy is 20 years in with 737k million (real) with a 6% SWR. Sure his portfolio is down and the odds of making it the last 10 years aren't assured (i.e. he could buy a 10 year tip ladder at this point but you don't do that with SWR studies). And as always the earning in pe10 has not been constant over the past 100 years.

I find thee studies fascinating but I would never have enough faith to act on any of it. I would never do a 7% SWR in 1981 bases on low valuations or go to 3% in 2000 based on high valuations either. I am going to live in the 3.5%-4.5% range pretty much no matter what. You can decide if that makes me aggressive or conservative:)

CraigTester
Posts: 80
Joined: Wed Aug 08, 2018 6:34 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CraigTester » Mon Aug 13, 2018 12:29 pm

For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....

KlangFool
Posts: 10445
Joined: Sat Oct 11, 2008 12:35 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by KlangFool » Mon Aug 13, 2018 12:38 pm

CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
CraigTester,

<<For those who don't believe valuation matters, >>

1) I believe that is the wrong question. The correct question and conclusion are we have no idea what the sequence of real return will be when we start withdrawing. So, we need a strategy that works even if the sequence of real return is not favorable.

Kitces offered one. Aka, bond tent.

2) If the portfolio is 50 times the required expense or larger aka 2% SWR, any strategy will work.

3) So, this kind of articles only matters to folks on the margin. Aka, 4% SWR or 25 times the required expense.

KlangFool

CraigTester
Posts: 80
Joined: Wed Aug 08, 2018 6:34 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CraigTester » Mon Aug 13, 2018 1:03 pm

KlangFool wrote:
Mon Aug 13, 2018 12:38 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
CraigTester,

<<For those who don't believe valuation matters, >>

1) I believe that is the wrong question. The correct question and conclusion are we have no idea what the sequence of real return will be when we start withdrawing. So, we need a strategy that works even if the sequence of real return is not favorable.

Kitces offered one. Aka, bond tent.

2) If the portfolio is 50 times the required expense or larger aka 2% SWR, any strategy will work.

3) So, this kind of articles only matters to folks on the margin. Aka, 4% SWR or 25 times the required expense.

KlangFool
KlangFool - If you study the chart comparing SWR vs CAEP, you'll see that we actually do have a very good idea of what the sequence of real returns will be. That's kind of the whole point of the article. :happy

And its exactly because of this point, that we need a strategy to accommodate for very elevated valuations at present.

Now if you happen to have more than "enough", you may be able to afford to lose money in the market without reducing your spending level (e.g. the chart suggests 2% SWR is safe, even today)....

But just because your safe while losing money, doesn't necessarily mean that you're ok with losing money....

KlangFool
Posts: 10445
Joined: Sat Oct 11, 2008 12:35 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by KlangFool » Mon Aug 13, 2018 1:20 pm

CraigTester wrote:
Mon Aug 13, 2018 1:03 pm
KlangFool wrote:
Mon Aug 13, 2018 12:38 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
CraigTester,

<<For those who don't believe valuation matters, >>

1) I believe that is the wrong question. The correct question and conclusion are we have no idea what the sequence of real return will be when we start withdrawing. So, we need a strategy that works even if the sequence of real return is not favorable.

Kitces offered one. Aka, bond tent.

2) If the portfolio is 50 times the required expense or larger aka 2% SWR, any strategy will work.

3) So, this kind of articles only matters to folks on the margin. Aka, 4% SWR or 25 times the required expense.

KlangFool
KlangFool - If you study the chart comparing SWR vs CAEP, you'll see that we actually do have a very good idea of what the sequence of real returns will be. That's kind of the whole point of the article. :happy
CraigTester,

I do not believe that to be true. And, I do not count any of this type of forecast for my plan.

KlangFool

CraigTester
Posts: 80
Joined: Wed Aug 08, 2018 6:34 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CraigTester » Mon Aug 13, 2018 1:27 pm

KlangFool wrote:
Mon Aug 13, 2018 1:20 pm
CraigTester wrote:
Mon Aug 13, 2018 1:03 pm
KlangFool wrote:
Mon Aug 13, 2018 12:38 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
CraigTester,

<<For those who don't believe valuation matters, >>

1) I believe that is the wrong question. The correct question and conclusion are we have no idea what the sequence of real return will be when we start withdrawing. So, we need a strategy that works even if the sequence of real return is not favorable.

Kitces offered one. Aka, bond tent.

2) If the portfolio is 50 times the required expense or larger aka 2% SWR, any strategy will work.

3) So, this kind of articles only matters to folks on the margin. Aka, 4% SWR or 25 times the required expense.

KlangFool
KlangFool - If you study the chart comparing SWR vs CAEP, you'll see that we actually do have a very good idea of what the sequence of real returns will be. That's kind of the whole point of the article. :happy
CraigTester,

I do not believe that to be true. And, I do not count any of this type of forecast for my plan.

KlangFool
KlangFool - Like I said in my first sentence, "For those who don't believe valuation matters, this excellent article can just be discarded"

Random Walker
Posts: 3207
Joined: Fri Feb 23, 2007 8:21 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by Random Walker » Mon Aug 13, 2018 1:41 pm

Sequence of returns is a huge issue immediately preceding retirement and in early retirement. This, influenced by recent valuations, is perhaps the biggest reason I’ve diversified as much as I can across sources of return. I’ve decreased overall equity exposure, increased bond exposure, tilted equities heavily to SV, taken international equity exposure to 50%, added alternatives. I believe this will decrease the SD and maximal drawdown of the portfolio compared to more conventional TSM portfolios with similar expected return.

Dave

Random Walker
Posts: 3207
Joined: Fri Feb 23, 2007 8:21 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by Random Walker » Mon Aug 13, 2018 1:44 pm

Strongly recommend William Bernstein’s short e-book on lifecycle investing. He addresses this issue at the very end. If the markets have been generous, then it’s prudent for the investor getting close to retirement to take some definitive action.

Dave

randomguy
Posts: 6504
Joined: Wed Sep 17, 2014 9:00 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by randomguy » Mon Aug 13, 2018 1:59 pm

KlangFool wrote:
Mon Aug 13, 2018 12:38 pm


1) I believe that is the wrong question. The correct question and conclusion are we have no idea what the sequence of real return will be when we start withdrawing. So, we need a strategy that works even if the sequence of real return is not favorable.

Kitces offered one. Aka, bond tent.

2) If the portfolio is 50 times the required expense or larger aka 2% SWR, any strategy will work.

3) So, this kind of articles only matters to folks on the margin. Aka, 4% SWR or 25 times the required expense.

KlangFool
Have you seen any evidence that bond tents/rising equity/... noticeably change the outcome? Is there any case where the 1966 retire with a bond tent gets say 4.5% SWR instead of 4% out of their portfolio? They might sleep a lot better at night but that is a different issue

3) Of course. The person with 15x isn't interesting (they will always fail) and the 30x+ person isn't either (they always succeed). The only interesting cases to talk about are the ones around the edges where changes potentially matter. A scheme that lets someone spend 20% more or retire 3-4 years earlier would be highly valuable.

KlangFool
Posts: 10445
Joined: Sat Oct 11, 2008 12:35 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by KlangFool » Mon Aug 13, 2018 2:18 pm

randomguy wrote:
Mon Aug 13, 2018 1:59 pm
KlangFool wrote:
Mon Aug 13, 2018 12:38 pm


1) I believe that is the wrong question. The correct question and conclusion are we have no idea what the sequence of real return will be when we start withdrawing. So, we need a strategy that works even if the sequence of real return is not favorable.

Kitces offered one. Aka, bond tent.

2) If the portfolio is 50 times the required expense or larger aka 2% SWR, any strategy will work.

3) So, this kind of articles only matters to folks on the margin. Aka, 4% SWR or 25 times the required expense.

KlangFool
Have you seen any evidence that bond tents/rising equity/... noticeably change the outcome? Is there any case where the 1966 retire with a bond tent gets say 4.5% SWR instead of 4% out of their portfolio? They might sleep a lot better at night but that is a different issue

3) Of course. The person with 15x isn't interesting (they will always fail) and the 30x+ person isn't either (they always succeed). The only interesting cases to talk about are the ones around the edges where changes potentially matter. A scheme that lets someone spend 20% more or retire 3-4 years earlier would be highly valuable.
randomguy,

<<Have you seen any evidence that bond tents/rising equity/... noticeably change the outcome?>>

I am not that smart. I am not smart enough to do this analysis. I only know that I should keep about 5 years in fixed income now. Please see the following thread for my analysis of my own situation.

viewtopic.php?f=1&t=255683

KlangFool

david1082b
Posts: 371
Joined: Fri Jun 09, 2017 12:35 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by david1082b » Mon Aug 13, 2018 2:46 pm

CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
The only actual 30-year span that has existed for a starting CAPE of over 30 starts around August-September 1929 -- 30 year annualized return for August was 7.9% nominal and 6% real https://dqydj.com/sp-500-return-calculator/
http://www.multpl.com/shiller-pe/table?f=m

Starting in September 1929 gives 5.763% real annualized 30 year return. This is only a historical data-point of course, it says nothing about the next 30 years or what the 30 year returns are going to be from a start in 1997-1998 that had similar CAPE numbers to now. We'll just have to wait.

randomguy
Posts: 6504
Joined: Wed Sep 17, 2014 9:00 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by randomguy » Mon Aug 13, 2018 3:40 pm

KlangFool wrote:
Mon Aug 13, 2018 2:18 pm


randomguy,

<<Have you seen any evidence that bond tents/rising equity/... noticeably change the outcome?>>

I am not that smart. I am not smart enough to do this analysis. I only know that I should keep about 5 years in fixed income now. Please see the following thread for my analysis of my own situation.

viewtopic.php?f=1&t=255683

KlangFool
You don't give the rebalancing rules so it is hard to say what the outcome is. You would have to see how 1929 and 1966 treat your plan. The cases I have seen it makes pretty much zero difference for the 66 case(not as many people look at the 1929). The markets just don't stay down long enough
AND missing out on the rebound hurts just as much as missing out on the drops.

I have no doubt bond tents give more stable portfolios (a plus no doubt). I am not as convinced they give a higher SWR.

CULater
Posts: 1513
Joined: Sun Nov 13, 2016 10:59 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CULater » Mon Aug 13, 2018 4:31 pm

One thing I would NOT do if I were starting retirement is to employ some form of age-based allocation; e.g., "age in bonds." The problem with this strategy is that it exacerbates the impact of a poor sequence of returns. It does so, because the largest equity allocation is during the early part of retirement; furthermore, it can impair the ability to benefit from higher stock returns if stocks recover later in retirement because the equity allocation has been trimmed.

I've looked at the data several times and proved to myself that you're better off with a constant stock allocation that matches the average allocation over time of your age-based allocation formula, based on your assumed life expectancy. For example, if you held 35% stocks at age 65, declining to 5% in 30 years (assumed life expectancy of 95) your average annualized age-in-bonds stock allocation would be 20%. Just put 20% in stocks at age 65 and rebalance to maintain that allocation, unless you reach a point that you don't think you need even that much in stocks or you're so well off that you can afford to increase your allocation for the benefit of heirs. If 20% in stocks doesn't seem like enough, and you can tolerate the risk of a higher allocation, then use a higher allocation and just leave it there. The best strategy is usually the strategy that requires the least effort to follow, and constant allocation is that strategy.

The only way that an "age in bonds" strategy can actually help, compared to constant allocation, is if you are lucky enough to experience higher stock returns early in retirement (a favorable sequence of returns). But that's a gamble, not worth taking, especially now with stretched stock valuations. Just "set it and forget it."
May you have the hindsight to know where you've been, The foresight to know where you're going, And the insight to know when you've gone too far. ~ Irish Blessing

User avatar
HomerJ
Posts: 11846
Joined: Fri Jun 06, 2008 12:50 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by HomerJ » Mon Aug 13, 2018 5:07 pm

Random Walker wrote:
Mon Aug 13, 2018 1:44 pm
Strongly recommend William Bernstein’s short e-book on lifecycle investing. He addresses this issue at the very end. If the markets have been generous, then it’s prudent for the investor getting close to retirement to take some definitive action.

Dave
I suggest an investor getting close to retirement take some definitive action REGARDLESS of valuations.

Because the risk of a bad sequence of returns as you start retirement is NEVER zero.

Does that make sense to any of you?

You guys says be conservative as you enter reitrement if valuations are high.

I say be conservative as you enter retirement.

Am I so wrong?
The J stands for Jay

User avatar
willthrill81
Posts: 6170
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by willthrill81 » Mon Aug 13, 2018 6:26 pm

CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
I'm a 'valuation skeptic', but I would never say that "valuation doesn't matter." I certainly believe that they do. But valuations are (1) not reliably mean reverting, which virtually no one can empirically argue against, and (2) valuations have appeared to have, at best, only a moderate relationship with subsequent ~10 year returns (accounting for about 40% of the variation, leaving the other 60% 'unexplained'). In the context that Kitces is referring to, I think that valuations may have enough relevance to be actionable on the part of the individual investor. But if you note his research into historic valuations and subsequent SWRs, you'll note that he never found that valuations cast doubt on the '4% rule of thumb' for withdrawals. Even during periods of historically high valuations, 4% withdrawals always worked for 30 year retirements. If valuations were low, then the SWR went up significantly. But I doubt that anyone here would say "CAPE is at 10! I can have 6% withdrawals for the next 30 years without ever checking up on my portfolio's performance!"

So the moral of the story seems to be the same old shtick: 25x annual expenses is a safe, though not guaranteed, target portfolio size for retirees with a 30 year horizon.
“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

protagonist
Posts: 5437
Joined: Sun Dec 26, 2010 12:47 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by protagonist » Mon Aug 13, 2018 7:11 pm

KlangFool wrote:
Mon Aug 13, 2018 12:38 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
CraigTester,

<<For those who don't believe valuation matters, >>

1) I believe that is the wrong question. The correct question and conclusion are we have no idea what the sequence of real return will be when we start withdrawing. So, we need a strategy that works even if the sequence of real return is not favorable.

Kitces offered one. Aka, bond tent.

2) If the portfolio is 50 times the required expense or larger aka 2% SWR, any strategy will work.

3) So, this kind of articles only matters to folks on the margin. Aka, 4% SWR or 25 times the required expense.

KlangFool
Or if you trust yourself, just be reasonable about your spending habits and you will probably be ok. It seems to work for most of the retirees I know who never heard of SWR, Cape Valuation or sequence of real returns. A bit of common sense goes a long way, and may be the key to a much happier retirement than obsession over chaotic factors that one cannot predict or control.

That said, for those for whom control over their finances is a difficult issue, I do agree that a strict framework may be useful.

User avatar
willthrill81
Posts: 6170
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by willthrill81 » Mon Aug 13, 2018 7:22 pm

protagonist wrote:
Mon Aug 13, 2018 7:11 pm
KlangFool wrote:
Mon Aug 13, 2018 12:38 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
CraigTester,

<<For those who don't believe valuation matters, >>

1) I believe that is the wrong question. The correct question and conclusion are we have no idea what the sequence of real return will be when we start withdrawing. So, we need a strategy that works even if the sequence of real return is not favorable.

Kitces offered one. Aka, bond tent.

2) If the portfolio is 50 times the required expense or larger aka 2% SWR, any strategy will work.

3) So, this kind of articles only matters to folks on the margin. Aka, 4% SWR or 25 times the required expense.

KlangFool
Or if you trust yourself, just be reasonable about your spending habits and you will probably be ok. It seems to work for most of the retirees I know who never heard of SWR, Cape Valuation or sequence of real returns. A bit of common sense goes a long way, and may be the key to a much happier retirement than obsession over chaotic factors that one cannot predict or control.

That said, for those for whom control over their finances is a difficult issue, I do agree that a strict framework may be useful.
I largely agree, but the issues surrounding sequence of returns risk are not 'common sense' as evidenced by the many folks who said back in the 1990s that 7% withdrawals were reasonable; Dave Ramsay still espouses 8% withdrawals.
“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
Johnnie
Posts: 512
Joined: Sat May 28, 2016 3:18 pm
Location: Michigan

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by Johnnie » Mon Aug 13, 2018 8:31 pm

I have read here something like this: Variable distribution rate plans guarantee you'll never run out of money - but you might end up taking 5 percent (or whatever) out of $100 bucks.

It probably won't be that bad, but makes me wonder: What would a 5 percent variable distribution plan out of a $1 million 60/40 portfolio look like X years hence in a world of "CAEP = about 3% today?"
"I know nothing."

randomguy
Posts: 6504
Joined: Wed Sep 17, 2014 9:00 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by randomguy » Mon Aug 13, 2018 9:13 pm

I largely agree, but the issues surrounding sequence of returns risk are not 'common sense' as evidenced by the many folks who said back in the 1990s that 7% withdrawals were reasonable; Dave Ramsay still espouses 8% withdrawals.
[/quote]

That 7% was pretty reasonable. You have to remember this is preBengen and Peter Lynch was not talking about an inflation adjusted 7%. He was talking about taking 7k out of a 100k and watching the portfolio grow over time. I want to say his success rate was north of 90%. Bengen and Trinity have so altered the discussion of retirement spending it is hard to remember that they are pretty recent ways of framing the discussion

I will not guess about what DR is talking about. :)

CraigTester
Posts: 80
Joined: Wed Aug 08, 2018 6:34 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CraigTester » Mon Aug 13, 2018 9:16 pm

willthrill81 wrote:
Mon Aug 13, 2018 6:26 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
I'm a 'valuation skeptic', but I would never say that "valuation doesn't matter." I certainly believe that they do. But valuations are (1) not reliably mean reverting, which virtually no one can empirically argue against, and (2) valuations have appeared to have, at best, only a moderate relationship with subsequent ~10 year returns (accounting for about 40% of the variation, leaving the other 60% 'unexplained'). In the context that Kitces is referring to, I think that valuations may have enough relevance to be actionable on the part of the individual investor. But if you note his research into historic valuations and subsequent SWRs, you'll note that he never found that valuations cast doubt on the '4% rule of thumb' for withdrawals. Even during periods of historically high valuations, 4% withdrawals always worked for 30 year retirements. If valuations were low, then the SWR went up significantly. But I doubt that anyone here would say "CAPE is at 10! I can have 6% withdrawals for the next 30 years without ever checking up on my portfolio's performance!"

So the moral of the story seems to be the same old shtick: 25x annual expenses is a safe, though not guaranteed, target portfolio size for retirees with a 30 year horizon.
William Bengen (who did much of the work to coin the 4% rule we hear about today) really did nothing more than a "least common denominator" analysis.

In other words, based on all the retirement start dates available within his data range, he simply wanted to know who ran out of money first -- for a given withdrawal rate.

The unlucky winner happened to be the guy that retired in 1966.

The "4% rule" evolved from here to state that, as long as your not more unlucky than the 1966 guy, you'll be fine....

Understanding this, it really shouldn't be a surprise at all that Kitces analysis doesn't cast any doubt on the 4% rule. That was not really the point.

Instead, what Kitces is revealing for us is some pretty interesting data on where starting valuations happened to be when past retirees started their 30 year clock.

And what's interesting to note is that the same poor 1966 retiree who failed first, happened to also start his retirement at a moment when valuations were quite high.

He then goes on in his chart to plot where valuations were when all our historical retirees each began their 30 year run.

And the fascinating (though arguably intuitive) take-away is that Valuation matters.

In other words, if you start your 30 year retirement when valuations are high, you don't tend to have near as cushy of a ride as those who start their retirement when valuations are less stretched.

Again, if you choose to not believe that valuation matters, you can just ignore all of this historical "coincidence".

But for those willing to look at the empirical results, its extremely clear that valuation is the single biggest determinant of your future returns.

And as Kitces charts show, this simple truth extends to SWR's as well....

User avatar
HomerJ
Posts: 11846
Joined: Fri Jun 06, 2008 12:50 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by HomerJ » Mon Aug 13, 2018 9:30 pm

CraigTester wrote:
Mon Aug 13, 2018 9:16 pm
willthrill81 wrote:
Mon Aug 13, 2018 6:26 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
I'm a 'valuation skeptic', but I would never say that "valuation doesn't matter." I certainly believe that they do. But valuations are (1) not reliably mean reverting, which virtually no one can empirically argue against, and (2) valuations have appeared to have, at best, only a moderate relationship with subsequent ~10 year returns (accounting for about 40% of the variation, leaving the other 60% 'unexplained'). In the context that Kitces is referring to, I think that valuations may have enough relevance to be actionable on the part of the individual investor. But if you note his research into historic valuations and subsequent SWRs, you'll note that he never found that valuations cast doubt on the '4% rule of thumb' for withdrawals. Even during periods of historically high valuations, 4% withdrawals always worked for 30 year retirements. If valuations were low, then the SWR went up significantly. But I doubt that anyone here would say "CAPE is at 10! I can have 6% withdrawals for the next 30 years without ever checking up on my portfolio's performance!"

So the moral of the story seems to be the same old shtick: 25x annual expenses is a safe, though not guaranteed, target portfolio size for retirees with a 30 year horizon.
William Bengen (who did much of the work to coin the 4% rule we hear about today) really did nothing more than a "least common denominator" analysis.

In other words, based on all the retirement start dates available within his data range, he simply wanted to know who ran out of money first -- for a given withdrawal rate.

The unlucky winner happened to be the guy that retired in 1966.

The "4% rule" evolved from here to state that, as long as your not more unlucky than the 1966 guy, you'll be fine....

Understanding this, it really shouldn't be a surprise at all that Kitces analysis doesn't cast any doubt on the 4% rule. That was not really the point.

Instead, what Kitces is revealing for us is some pretty interesting data on where starting valuations happened to be when past retirees started their 30 year clock.

And what's interesting to note is that the same poor 1966 retiree who failed first, happened to also start his retirement at a moment when valuations were quite high.

He then goes on in his chart to plot where valuations were when all our historical retirees each began their 30 year run.

And the fascinating (though arguably intuitive) take-away is that Valuation matters.

In other words, if you start your 30 year retirement when valuations are high, you don't tend to have near as cushy of a ride as those who start their retirement when valuations are less stretched.

Again, if you choose to not believe that valuation matters, you can just ignore all of this historical "coincidence".

But for those willing to look at the empirical results, its extremely clear that valuation is the single biggest determinant of your future returns.

And as Kitces charts show, this simple truth extends to SWR's as well....
If you use 4%, valuations have not mattered at all.

Simple as that.
The J stands for Jay

protagonist
Posts: 5437
Joined: Sun Dec 26, 2010 12:47 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by protagonist » Mon Aug 13, 2018 9:37 pm

willthrill81 wrote:
Mon Aug 13, 2018 7:22 pm
protagonist wrote:
Mon Aug 13, 2018 7:11 pm
KlangFool wrote:
Mon Aug 13, 2018 12:38 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
CraigTester,

<<For those who don't believe valuation matters, >>

1) I believe that is the wrong question. The correct question and conclusion are we have no idea what the sequence of real return will be when we start withdrawing. So, we need a strategy that works even if the sequence of real return is not favorable.

Kitces offered one. Aka, bond tent.

2) If the portfolio is 50 times the required expense or larger aka 2% SWR, any strategy will work.

3) So, this kind of articles only matters to folks on the margin. Aka, 4% SWR or 25 times the required expense.

KlangFool
Or if you trust yourself, just be reasonable about your spending habits and you will probably be ok. It seems to work for most of the retirees I know who never heard of SWR, Cape Valuation or sequence of real returns. A bit of common sense goes a long way, and may be the key to a much happier retirement than obsession over chaotic factors that one cannot predict or control.

That said, for those for whom control over their finances is a difficult issue, I do agree that a strict framework may be useful.
I largely agree, but the issues surrounding sequence of returns risk are not 'common sense' as evidenced by the many folks who said back in the 1990s that 7% withdrawals were reasonable; Dave Ramsay still espouses 8% withdrawals.
That is a great argument against following bad advice (assuming the consumer knows enough to distinguish bad from good advice)....not a convincing argument against just using common sense regarding your spending habits (smile). After all, every "expert" has plenty of "evidence" to back up his claims. I'm sure Ramsay does as well.

randomguy
Posts: 6504
Joined: Wed Sep 17, 2014 9:00 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by randomguy » Mon Aug 13, 2018 9:41 pm

CraigTester wrote:
Mon Aug 13, 2018 9:16 pm

William Bengen (who did much of the work to coin the 4% rule we hear about today) really did nothing more than a "least common denominator" analysis.

In other words, based on all the retirement start dates available within his data range, he simply wanted to know who ran out of money first -- for a given withdrawal rate.

The unlucky winner happened to be the guy that retired in 1966.

The "4% rule" evolved from here to state that, as long as your not more unlucky than the 1966 guy, you'll be fine....

Understanding this, it really shouldn't be a surprise at all that Kitces analysis doesn't cast any doubt on the 4% rule. That was not really the point.

Instead, what Kitces is revealing for us is some pretty interesting data on where starting valuations happened to be when past retirees started their 30 year clock.

And what's interesting to note is that the same poor 1966 retiree who failed first, happened to also start his retirement at a moment when valuations were quite high.

He then goes on in his chart to plot where valuations were when all our historical retirees each began their 30 year run.

And the fascinating (though arguably intuitive) take-away is that Valuation matters.

In other words, if you start your 30 year retirement when valuations are high, you don't tend to have near as cushy of a ride as those who start their retirement when valuations are less stretched.

Again, if you choose to not believe that valuation matters, you can just ignore all of this historical "coincidence".

But for those willing to look at the empirical results, its extremely clear that valuation is the single biggest determinant of your future returns.

And as Kitces charts show, this simple truth extends to SWR's as well....

The problem is that there are very few periods with pe10s over 24 in history. Right now we have 1929 and 1966 and we are drawing our conclusions from them. Lets look at some data points
Assume a 50/50 portfolio
1966 PE10 was 24
1996 pe10 24.76 1.47 million dollars real dollars
1997 pe10 28.33 1.32 million
1998 pe10 32.8 1.05 million
2004 pe10 27.66 1.05 million
Obviously none of them are over yet but if you could sell and buy TIPs, you would succeed every time. In another 15 years PE10s it might turn out that PE10s>24 still result in 5% SWR half the time. Or maybe we are due for another really horrible decade (i.e. 0% real return from both bonds and stocks) and some of these fail.

The thing with valuations based on PE10 is that we are either in the biggest bubble ever(i.e. something like 29 of the last 30 years are above the historical average and the only reason the other isn't is because of the inflation of the average over the past 25 years) or something has changed which makes higher valuations now more reasonable than they were from 1880-1990. Good luck answering that question.

CraigTester
Posts: 80
Joined: Wed Aug 08, 2018 6:34 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CraigTester » Mon Aug 13, 2018 9:45 pm

HomerJ wrote:
Mon Aug 13, 2018 9:30 pm
CraigTester wrote:
Mon Aug 13, 2018 9:16 pm
willthrill81 wrote:
Mon Aug 13, 2018 6:26 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
I'm a 'valuation skeptic', but I would never say that "valuation doesn't matter." I certainly believe that they do. But valuations are (1) not reliably mean reverting, which virtually no one can empirically argue against, and (2) valuations have appeared to have, at best, only a moderate relationship with subsequent ~10 year returns (accounting for about 40% of the variation, leaving the other 60% 'unexplained'). In the context that Kitces is referring to, I think that valuations may have enough relevance to be actionable on the part of the individual investor. But if you note his research into historic valuations and subsequent SWRs, you'll note that he never found that valuations cast doubt on the '4% rule of thumb' for withdrawals. Even during periods of historically high valuations, 4% withdrawals always worked for 30 year retirements. If valuations were low, then the SWR went up significantly. But I doubt that anyone here would say "CAPE is at 10! I can have 6% withdrawals for the next 30 years without ever checking up on my portfolio's performance!"

So the moral of the story seems to be the same old shtick: 25x annual expenses is a safe, though not guaranteed, target portfolio size for retirees with a 30 year horizon.
William Bengen (who did much of the work to coin the 4% rule we hear about today) really did nothing more than a "least common denominator" analysis.

In other words, based on all the retirement start dates available within his data range, he simply wanted to know who ran out of money first -- for a given withdrawal rate.

The unlucky winner happened to be the guy that retired in 1966.

The "4% rule" evolved from here to state that, as long as your not more unlucky than the 1966 guy, you'll be fine....

Understanding this, it really shouldn't be a surprise at all that Kitces analysis doesn't cast any doubt on the 4% rule. That was not really the point.

Instead, what Kitces is revealing for us is some pretty interesting data on where starting valuations happened to be when past retirees started their 30 year clock.

And what's interesting to note is that the same poor 1966 retiree who failed first, happened to also start his retirement at a moment when valuations were quite high.

He then goes on in his chart to plot where valuations were when all our historical retirees each began their 30 year run.

And the fascinating (though arguably intuitive) take-away is that Valuation matters.

In other words, if you start your 30 year retirement when valuations are high, you don't tend to have near as cushy of a ride as those who start their retirement when valuations are less stretched.

Again, if you choose to not believe that valuation matters, you can just ignore all of this historical "coincidence".

But for those willing to look at the empirical results, its extremely clear that valuation is the single biggest determinant of your future returns.

And as Kitces charts show, this simple truth extends to SWR's as well....
If you use 4%, valuations have not mattered at all.

Simple as that.
I'm not sure the 1966 retiree would completely agree with you. He literally spent his last penny in his 30th year of retirement - after a whole series of very depressing conversations with his financial advisor along the way....

And to pour salt in his wounds, many of his buddies who retired at times when valuations were not so stretched, died with more than they started with....and never had a sleepless night...

CraigTester
Posts: 80
Joined: Wed Aug 08, 2018 6:34 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CraigTester » Mon Aug 13, 2018 10:10 pm

randomguy wrote:
Mon Aug 13, 2018 9:41 pm
CraigTester wrote:
Mon Aug 13, 2018 9:16 pm

William Bengen (who did much of the work to coin the 4% rule we hear about today) really did nothing more than a "least common denominator" analysis.

In other words, based on all the retirement start dates available within his data range, he simply wanted to know who ran out of money first -- for a given withdrawal rate.

The unlucky winner happened to be the guy that retired in 1966.

The "4% rule" evolved from here to state that, as long as your not more unlucky than the 1966 guy, you'll be fine....

Understanding this, it really shouldn't be a surprise at all that Kitces analysis doesn't cast any doubt on the 4% rule. That was not really the point.

Instead, what Kitces is revealing for us is some pretty interesting data on where starting valuations happened to be when past retirees started their 30 year clock.

And what's interesting to note is that the same poor 1966 retiree who failed first, happened to also start his retirement at a moment when valuations were quite high.

He then goes on in his chart to plot where valuations were when all our historical retirees each began their 30 year run.

And the fascinating (though arguably intuitive) take-away is that Valuation matters.

In other words, if you start your 30 year retirement when valuations are high, you don't tend to have near as cushy of a ride as those who start their retirement when valuations are less stretched.

Again, if you choose to not believe that valuation matters, you can just ignore all of this historical "coincidence".

But for those willing to look at the empirical results, its extremely clear that valuation is the single biggest determinant of your future returns.

And as Kitces charts show, this simple truth extends to SWR's as well....

The problem is that there are very few periods with pe10s over 24 in history. Right now we have 1929 and 1966 and we are drawing our conclusions from them. Lets look at some data points
Assume a 50/50 portfolio
1966 PE10 was 24
1996 pe10 24.76 1.47 million dollars real dollars
1997 pe10 28.33 1.32 million
1998 pe10 32.8 1.05 million
2004 pe10 27.66 1.05 million
Obviously none of them are over yet but if you could sell and buy TIPs, you would succeed every time. In another 15 years PE10s it might turn out that PE10s>24 still result in 5% SWR half the time. Or maybe we are due for another really horrible decade (i.e. 0% real return from both bonds and stocks) and some of these fail.

The thing with valuations based on PE10 is that we are either in the biggest bubble ever(i.e. something like 29 of the last 30 years are above the historical average and the only reason the other isn't is because of the inflation of the average over the past 25 years) or something has changed which makes higher valuations now more reasonable than they were from 1880-1990. Good luck answering that question.
I am new to this forum, but what I'm gathering in my first few days, is there has been a rash of criticism of PE10, market timing, etc, etc.

So now if anyone even mentions these terms, they are summarily taken to the wood shed...

But as a guy who hasn't been through these battles, here's how I very simply interpret the subject article of this thread.

If you retire when valuations are at historical highs, the one thing you know for sure is that you are a "high probability candidate" for being the next "least common denominator".

So why even put yourself at risk?

All we really "know" is that PE10 is presently at its second highest level in history.

And whenever we've reached these types of levels before, long term returns have been abysmal.

But perhaps this time will be different....

User avatar
willthrill81
Posts: 6170
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by willthrill81 » Mon Aug 13, 2018 10:13 pm

CraigTester wrote:
Mon Aug 13, 2018 9:45 pm
HomerJ wrote:
Mon Aug 13, 2018 9:30 pm
CraigTester wrote:
Mon Aug 13, 2018 9:16 pm
willthrill81 wrote:
Mon Aug 13, 2018 6:26 pm
CraigTester wrote:
Mon Aug 13, 2018 12:29 pm
For those who don't believe valuation matters, I suppose this excellent article can just be discarded.

But for those who are a little bit more philosophical about how recency bias from this 10 year bull market might cloud ones thinking, the chart comparing SWR vs CAEP, is outstanding.

CAEP = about 3% today.

This suggests anyone starting their 30 year run today, is going to be flirting with being a "least common denominator", if they stay the course.

Thank you for posting....
I'm a 'valuation skeptic', but I would never say that "valuation doesn't matter." I certainly believe that they do. But valuations are (1) not reliably mean reverting, which virtually no one can empirically argue against, and (2) valuations have appeared to have, at best, only a moderate relationship with subsequent ~10 year returns (accounting for about 40% of the variation, leaving the other 60% 'unexplained'). In the context that Kitces is referring to, I think that valuations may have enough relevance to be actionable on the part of the individual investor. But if you note his research into historic valuations and subsequent SWRs, you'll note that he never found that valuations cast doubt on the '4% rule of thumb' for withdrawals. Even during periods of historically high valuations, 4% withdrawals always worked for 30 year retirements. If valuations were low, then the SWR went up significantly. But I doubt that anyone here would say "CAPE is at 10! I can have 6% withdrawals for the next 30 years without ever checking up on my portfolio's performance!"

So the moral of the story seems to be the same old shtick: 25x annual expenses is a safe, though not guaranteed, target portfolio size for retirees with a 30 year horizon.
William Bengen (who did much of the work to coin the 4% rule we hear about today) really did nothing more than a "least common denominator" analysis.

In other words, based on all the retirement start dates available within his data range, he simply wanted to know who ran out of money first -- for a given withdrawal rate.

The unlucky winner happened to be the guy that retired in 1966.

The "4% rule" evolved from here to state that, as long as your not more unlucky than the 1966 guy, you'll be fine....

Understanding this, it really shouldn't be a surprise at all that Kitces analysis doesn't cast any doubt on the 4% rule. That was not really the point.

Instead, what Kitces is revealing for us is some pretty interesting data on where starting valuations happened to be when past retirees started their 30 year clock.

And what's interesting to note is that the same poor 1966 retiree who failed first, happened to also start his retirement at a moment when valuations were quite high.

He then goes on in his chart to plot where valuations were when all our historical retirees each began their 30 year run.

And the fascinating (though arguably intuitive) take-away is that Valuation matters.

In other words, if you start your 30 year retirement when valuations are high, you don't tend to have near as cushy of a ride as those who start their retirement when valuations are less stretched.

Again, if you choose to not believe that valuation matters, you can just ignore all of this historical "coincidence".

But for those willing to look at the empirical results, its extremely clear that valuation is the single biggest determinant of your future returns.

And as Kitces charts show, this simple truth extends to SWR's as well....
If you use 4%, valuations have not mattered at all.

Simple as that.
I'm not sure the 1966 retiree would completely agree with you. He literally spent his last penny in his 30th year of retirement - after a whole series of very depressing conversations with his financial advisor along the way....

And to pour salt in his wounds, many of his buddies who retired at times when valuations were not so stretched, died with more than they started with....and never had a sleepless night...
So you should be even more conservative than the worst period in over 100 years of market history? Some would say yes. I would personally say no, but that's just 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

CraigTester
Posts: 80
Joined: Wed Aug 08, 2018 6:34 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CraigTester » Mon Aug 13, 2018 10:23 pm

willthrill81 wrote:
Mon Aug 13, 2018 10:13 pm
CraigTester wrote:
Mon Aug 13, 2018 9:45 pm
HomerJ wrote:
Mon Aug 13, 2018 9:30 pm
CraigTester wrote:
Mon Aug 13, 2018 9:16 pm
willthrill81 wrote:
Mon Aug 13, 2018 6:26 pm


I'm a 'valuation skeptic', but I would never say that "valuation doesn't matter." I certainly believe that they do. But valuations are (1) not reliably mean reverting, which virtually no one can empirically argue against, and (2) valuations have appeared to have, at best, only a moderate relationship with subsequent ~10 year returns (accounting for about 40% of the variation, leaving the other 60% 'unexplained'). In the context that Kitces is referring to, I think that valuations may have enough relevance to be actionable on the part of the individual investor. But if you note his research into historic valuations and subsequent SWRs, you'll note that he never found that valuations cast doubt on the '4% rule of thumb' for withdrawals. Even during periods of historically high valuations, 4% withdrawals always worked for 30 year retirements. If valuations were low, then the SWR went up significantly. But I doubt that anyone here would say "CAPE is at 10! I can have 6% withdrawals for the next 30 years without ever checking up on my portfolio's performance!"

So the moral of the story seems to be the same old shtick: 25x annual expenses is a safe, though not guaranteed, target portfolio size for retirees with a 30 year horizon.
William Bengen (who did much of the work to coin the 4% rule we hear about today) really did nothing more than a "least common denominator" analysis.

In other words, based on all the retirement start dates available within his data range, he simply wanted to know who ran out of money first -- for a given withdrawal rate.

The unlucky winner happened to be the guy that retired in 1966.

The "4% rule" evolved from here to state that, as long as your not more unlucky than the 1966 guy, you'll be fine....

Understanding this, it really shouldn't be a surprise at all that Kitces analysis doesn't cast any doubt on the 4% rule. That was not really the point.

Instead, what Kitces is revealing for us is some pretty interesting data on where starting valuations happened to be when past retirees started their 30 year clock.

And what's interesting to note is that the same poor 1966 retiree who failed first, happened to also start his retirement at a moment when valuations were quite high.

He then goes on in his chart to plot where valuations were when all our historical retirees each began their 30 year run.

And the fascinating (though arguably intuitive) take-away is that Valuation matters.

In other words, if you start your 30 year retirement when valuations are high, you don't tend to have near as cushy of a ride as those who start their retirement when valuations are less stretched.

Again, if you choose to not believe that valuation matters, you can just ignore all of this historical "coincidence".

But for those willing to look at the empirical results, its extremely clear that valuation is the single biggest determinant of your future returns.

And as Kitces charts show, this simple truth extends to SWR's as well....
If you use 4%, valuations have not mattered at all.

Simple as that.
I'm not sure the 1966 retiree would completely agree with you. He literally spent his last penny in his 30th year of retirement - after a whole series of very depressing conversations with his financial advisor along the way....

And to pour salt in his wounds, many of his buddies who retired at times when valuations were not so stretched, died with more than they started with....and never had a sleepless night...
So you should be even more conservative than the worst period in over 100 years of market history? Some would say yes. I would personally say no, but that's just me.
No, you're getting caught in a mental trap by the way you framed your question. If valuations were not currently at their second highest levels in history, I'd happily sign up for a 30 year retirement plan, 50-50 stocks/bonds and a 4% SWR.

But I don't consider it excessively conservative to not want to use the same SWR that the 1966 retiree used, once I understood the market he retired in was less over-valued than today's.....

User avatar
HomerJ
Posts: 11846
Joined: Fri Jun 06, 2008 12:50 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by HomerJ » Mon Aug 13, 2018 10:23 pm

CraigTester wrote:
Mon Aug 13, 2018 10:10 pm
All we really "know" is that PE10 is presently at its second highest level in history.

And whenever we've reached these types of levels before, long term returns have been abysmal.
This is false.

1996 was the second highest level in history at the time, and long term returns have been 9% nominal, very close to historical norms.

Even 2000, the absolute highest valuations in U.S. history, have returned a bit over 6% nominal over the past 18 years. Most people here would be very happy if our FLOOR is 6% over the next 18 years.

What killed 1966 investors was not valuations, but inflation in the late 1970s.

And still 4% (okay 3.8%) worked even with the perfect storm of bad stock returns, rising interest rates hurting bond portfolios, AND double-digit inflation 12-14 years into your retirement.
Last edited by HomerJ on Mon Aug 13, 2018 10:34 pm, edited 2 times in total.
The J stands for Jay

User avatar
HomerJ
Posts: 11846
Joined: Fri Jun 06, 2008 12:50 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by HomerJ » Mon Aug 13, 2018 10:26 pm

CraigTester wrote:
Mon Aug 13, 2018 10:23 pm
No, you're getting caught in a mental trap by the way you framed your question. If valuations were not currently at their second highest levels in history, I'd happily sign up for a 30 year retirement plan, 50-50 stocks/bonds and a 4% SWR.

But I don't consider it excessively conservative to not want to use the same SWR that the 1966 retiree used, once I understood the market he retired in was less over-valued than today's.....
No, you don't understand. 4% is not for average times. 4% is for the WORST times.

So you saying "Hey, we're likely in for some bad times" doesn't mean anything, because 4% is already preparing for bad times.

4% withdrawals is like carrying around an umbrella every day. You joining this forum, and running up screaming "The chance of rain is much higher today than it was yesterday!" doesn't matter to me, BECAUSE I'M ALREADY CARRYING AN UMBRELLA.
The J stands for Jay

Random Walker
Posts: 3207
Joined: Fri Feb 23, 2007 8:21 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by Random Walker » Mon Aug 13, 2018 10:30 pm

I think it’s Bernstein or maybe Zwecher who refers to “Murphy’s Law Of Retirement”. An investor profits from a long bull market. As valuations rise, his portfolio swells. Seeing the larger than expected portfolio, the investor decides to retire. But he is retiring into a period of high valuations and low future expected returns.

Dave

CraigTester
Posts: 80
Joined: Wed Aug 08, 2018 6:34 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by CraigTester » Mon Aug 13, 2018 10:42 pm

HomerJ wrote:
Mon Aug 13, 2018 10:26 pm
CraigTester wrote:
Mon Aug 13, 2018 10:23 pm
No, you're getting caught in a mental trap by the way you framed your question. If valuations were not currently at their second highest levels in history, I'd happily sign up for a 30 year retirement plan, 50-50 stocks/bonds and a 4% SWR.

But I don't consider it excessively conservative to not want to use the same SWR that the 1966 retiree used, once I understood the market he retired in was less over-valued than today's.....
No, you don't understand. 4% is not for average times. 4% is for the WORST times.

So you saying "Hey, we're likely in for some bad times" doesn't mean anything, because 4% is already preparing for bad times.

4% withdrawals is like carrying around an umbrella every day. You joining this forum, and running up screaming "The chance of rain is much higher today than it was yesterday!" doesn't matter to me, BECAUSE I'M ALREADY CARRYING AN UMBRELLA.
Homer -- I'm never quite sure how to respond to you. I'd really like to help, but if your goal is to bounce the last check you write in year 30, that is for you to decide.

User avatar
HomerJ
Posts: 11846
Joined: Fri Jun 06, 2008 12:50 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by HomerJ » Mon Aug 13, 2018 10:43 pm

Random Walker wrote:
Mon Aug 13, 2018 10:30 pm
I think it’s Bernstein or maybe Zwecher who refers to “Murphy’s Law Of Retirement”. An investor profits from a long bull market. As valuations rise, his portfolio swells. Seeing the larger than expected portfolio, the investor decides to retire. But he is retiring into a period of high valuations and low future expected returns.

Dave
But if he reads this forum, he changes his Asset Allocation to be more conservative as he retires and moves money into bonds, LOCKING IN SOME OF THOSE GAINS.

Someone who saw their stock money double from 1996 to 2000, if they followed the standard advice on this forum, probably did just fine. Their portfolio swelled, they decided to retire, and moved from say 60/40 to 40/60, and got to keep most of that money. Actually ALL of it, since they could live off the bonds while waiting for the stocks to recover.

Not once did they have to look at valuations for this strategy to work.

The key variable is the fact that they retired. Regardless of valuations, they should have changed their allocation to be more conservative the moment they realized retirement was close or at hand.
The J stands for Jay

User avatar
HomerJ
Posts: 11846
Joined: Fri Jun 06, 2008 12:50 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by HomerJ » Mon Aug 13, 2018 10:50 pm

CraigTester wrote:
Mon Aug 13, 2018 10:42 pm
HomerJ wrote:
Mon Aug 13, 2018 10:26 pm
CraigTester wrote:
Mon Aug 13, 2018 10:23 pm
No, you're getting caught in a mental trap by the way you framed your question. If valuations were not currently at their second highest levels in history, I'd happily sign up for a 30 year retirement plan, 50-50 stocks/bonds and a 4% SWR.

But I don't consider it excessively conservative to not want to use the same SWR that the 1966 retiree used, once I understood the market he retired in was less over-valued than today's.....
No, you don't understand. 4% is not for average times. 4% is for the WORST times.

So you saying "Hey, we're likely in for some bad times" doesn't mean anything, because 4% is already preparing for bad times.

4% withdrawals is like carrying around an umbrella every day. You joining this forum, and running up screaming "The chance of rain is much higher today than it was yesterday!" doesn't matter to me, BECAUSE I'M ALREADY CARRYING AN UMBRELLA.
Homer -- I'm never quite sure how to respond to you. I'd really like to help, but if your goal is to bounce the last check you write in year 30, that is for you to decide.
I'm pretty sure you have little desire to help anyone. It's not normal to walk into a party and immediately start telling everyone there how stupid they are. But I'll be generous, and assume you actually want to have a conversation.

Nothing I said talks about bouncing a check in year 30.

You're new to this site, and haven't read many of the other conversations on this topic. 4% is not a iron-clad rule. No one here advocates anyone follow it blindly.

"Failure" of the 4% withdrawal plan means cutting back a bit 10-15 years into retirement. If we see the total portfolio value going down instead of up, we'd probably start taking 3 vacations a year instead of 4. We're not talking cat food living under a bridge. We're talking having to choose between Spain or Italy instead of going to both next year.
Last edited by HomerJ on Mon Aug 13, 2018 10:54 pm, edited 2 times in total.
The J stands for Jay

User avatar
willthrill81
Posts: 6170
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by willthrill81 » Mon Aug 13, 2018 10:51 pm

CraigTester wrote:
Mon Aug 13, 2018 10:42 pm
HomerJ wrote:
Mon Aug 13, 2018 10:26 pm
CraigTester wrote:
Mon Aug 13, 2018 10:23 pm
No, you're getting caught in a mental trap by the way you framed your question. If valuations were not currently at their second highest levels in history, I'd happily sign up for a 30 year retirement plan, 50-50 stocks/bonds and a 4% SWR.

But I don't consider it excessively conservative to not want to use the same SWR that the 1966 retiree used, once I understood the market he retired in was less over-valued than today's.....
No, you don't understand. 4% is not for average times. 4% is for the WORST times.

So you saying "Hey, we're likely in for some bad times" doesn't mean anything, because 4% is already preparing for bad times.

4% withdrawals is like carrying around an umbrella every day. You joining this forum, and running up screaming "The chance of rain is much higher today than it was yesterday!" doesn't matter to me, BECAUSE I'M ALREADY CARRYING AN UMBRELLA.
Homer -- I'm never quite sure how to respond to you. I'd really like to help, but if your goal is to bounce the last check you write in year 30, that is for you to decide.
By definition, that is how the safe withdrawal rate research defines success. If you want to have guaranteed money available at the end of your retirement, then maybe it should not be part of the portfolio that you are making withdrawals from, assuming you are taking the safe withdrawal rate approach.

You might be more interested in the perpetual withdrawal rate than the safe withdrawal rate.

And you're reading into the '4% rule of thumb' far too literally. I haven't even heard of a rumor of a retiree who actually followed it; virtually everyone reduces their withdrawals when the market has been hard on their portfolio.
“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: 1220
Joined: Sat Dec 03, 2016 10:22 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by marcopolo » Mon Aug 13, 2018 10:59 pm

CraigTester wrote:
Mon Aug 13, 2018 10:42 pm
HomerJ wrote:
Mon Aug 13, 2018 10:26 pm
CraigTester wrote:
Mon Aug 13, 2018 10:23 pm
No, you're getting caught in a mental trap by the way you framed your question. If valuations were not currently at their second highest levels in history, I'd happily sign up for a 30 year retirement plan, 50-50 stocks/bonds and a 4% SWR.

But I don't consider it excessively conservative to not want to use the same SWR that the 1966 retiree used, once I understood the market he retired in was less over-valued than today's.....
No, you don't understand. 4% is not for average times. 4% is for the WORST times.

So you saying "Hey, we're likely in for some bad times" doesn't mean anything, because 4% is already preparing for bad times.

4% withdrawals is like carrying around an umbrella every day. You joining this forum, and running up screaming "The chance of rain is much higher today than it was yesterday!" doesn't matter to me, BECAUSE I'M ALREADY CARRYING AN UMBRELLA.
Homer -- I'm never quite sure how to respond to you. I'd really like to help, but if your goal is to bounce the last check you write in year 30, that is for you to decide.
You could start by addressing some of the facts he has presented to counter your assertions.
For example, you have repeatedly said that when valuations are at their second highest levels in history, things end badly. HomerJ, has responded a couple of times showing you this statement is factually incorrect. Instead of thinking how your incorrect assertions might lead to poor conclusions, you simply keep repeating the same incorrect tidbit.

“What gets us into trouble is not what we don't know. It's what we know for sure that just ain't so.” -- Mark Twain
Once in a while you get shown the light, in the strangest of places if you look at it right.

AlohaJoe
Posts: 3851
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by AlohaJoe » Mon Aug 13, 2018 11:06 pm

willthrill81 wrote:
Mon Aug 13, 2018 6:26 pm
But I doubt that anyone here would say "CAPE is at 10! I can have 6% withdrawals for the next 30 years without ever checking up on my portfolio's performance!"
Along these lines, go back to 2009 when CAPE was actually below its mean & median and see if you can find a single post/blog/article from a person recommending people to retire and take out 4.5% "because valuations are low". It wasn't that long ago so it should be easy to find thousands of posts from valuation/CAPE believers advocating retirees crank up their SWR above 4%.......I mean, never mind that the New York Times was still running articles with headlines like "Steep Market Drops Highlight Despair Over Rescue Efforts" just go ahead and withdraw 4.5% now because CAPE.

User avatar
willthrill81
Posts: 6170
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by willthrill81 » Mon Aug 13, 2018 11:09 pm

AlohaJoe wrote:
Mon Aug 13, 2018 11:06 pm
willthrill81 wrote:
Mon Aug 13, 2018 6:26 pm
But I doubt that anyone here would say "CAPE is at 10! I can have 6% withdrawals for the next 30 years without ever checking up on my portfolio's performance!"
Along these lines, go back to 2009 when CAPE was actually below its mean & median and see if you can find a single post/blog/article from a person recommending people to retire and take out 4.5% "because valuations are low". It wasn't that long ago so it should be easy to find thousands of posts from valuation/CAPE believers advocating retirees crank up their SWR above 4%.......I mean, never mind that the New York Times was still running articles with headlines like "Steep Market Drops Highlight Despair Over Rescue Efforts" just go ahead and withdraw 4.5% now because CAPE.
:sharebeer
“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

stlutz
Posts: 4790
Joined: Fri Jan 02, 2009 1:08 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by stlutz » Mon Aug 13, 2018 11:16 pm

A point was made a above that kind of got glossed over. Things didn't go all that well for the 1966 retiree because of inflation. If you take a more global look at it, there are all sorts of scenarios where 4% didn't work. Most of those were driven by events, not valuations.

Changing asset allocations between stocks and bonds didn't help the 1966 retiree because nothing worked back then. And if it sort of worked it got taxed away. Changing from 60/40 to 40/60 doesn't help the retiree in Turkey right now.

I don't really have a solution for that problem. The early retiree can go back to work (albeit likely at a much lower salary than before he retired). You can't really do that when you're 80.

It's also worth keeping in mind that low valuations are an indication that the market perceives a lot of risk. Now, the market can be wrong, and sometimes risks just don't show up. But we talk an awful lot like it was guaranteed that the 2008 crisis would be handled in a way that turned out quite well for stock market investors. There were all sorts of different outcomes that could have happened that would have been much worse for Bogleheads. Low valuations in 2009 indicated a lot of opportunity; but they also indicated a lot of risk. "May you live in low valuation times" is not equivalent to saying "Merry Christmas".

In short, a low CAPE indicates a much wider range of possible outcomes, not that a 7% withdrawal rate is guaranteed to work.

It's bad to be reliant on investments at a time when there is a transition between high valuations and low valuations. But that type of transition is not driven by a mysterious force called "reversion to the mean"; it's driven by events.

Structuring a withdrawal approach that allows one to be successful in a wide variety of circumstances deserves more thought than it gets. I don't think that CAPE-based market timing is that answer, nor do I think that taking x% every year regardless of what happens to your portfolio is either (unless you buy an annuity, in which case it's by definition).

randomguy
Posts: 6504
Joined: Wed Sep 17, 2014 9:00 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by randomguy » Mon Aug 13, 2018 11:20 pm

CraigTester wrote:
Mon Aug 13, 2018 10:10 pm


I am new to this forum, but what I'm gathering in my first few days, is there has been a rash of criticism of PE10, market timing, etc, etc.

So now if anyone even mentions these terms, they are summarily taken to the wood shed...

But as a guy who hasn't been through these battles, here's how I very simply interpret the subject article of this thread.

If you retire when valuations are at historical highs, the one thing you know for sure is that you are a "high probability candidate" for being the next "least common denominator".

So why even put yourself at risk?

All we really "know" is that PE10 is presently at its second highest level in history.

And whenever we've reached these types of levels before, long term returns have been abysmal.

But perhaps this time will be different....
To reiterate what everyone else has said.

The returns after hitting 32 haven't been abysmal so far in US history
1998-2018 - 4.9 real cagr
2005-2018 6.37
Given the average since 1929 is just over 6 we aren't exactly talking about horrible returns. PE10s of 24 and 27 are much more dangerous. And 40. Hit that and historically you should sell 100% of the time:)

The thing with comparing PE10s is that you are comparing numbers that are only loosely related. We all know todays PE10s are inflated by 10-20% over pre2001 ones but does anyone know how a 2000 PE10 compares to a 1970 one or a 1950 one? And PE10s don't live in a vacuum. Things like interest rates should affect what you pay for a stock. How much you believe those normalizations make sense versus cooking the books is up to you:)

People on this board are always looking for a reason to be pessimistic. In 2009 nobody was talking about doing a 6% SWR because valuations were down at 15.:)

The question is when PE10s become actionable. If earnings are constant and the market doubles, I would cut back my stock allocation. If the market crashes 75% and valuations drop to single digits, I would buy more. Outside of the extremes I find it hard to justify acting. I remember well all the people in 2000-2 and 2008-11 who talked about going back to single digit PE10s and how selling everything now was the only reasonable play. So far they haven't been right.

User avatar
HomerJ
Posts: 11846
Joined: Fri Jun 06, 2008 12:50 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by HomerJ » Mon Aug 13, 2018 11:27 pm

AlohaJoe wrote:
Mon Aug 13, 2018 11:06 pm
willthrill81 wrote:
Mon Aug 13, 2018 6:26 pm
But I doubt that anyone here would say "CAPE is at 10! I can have 6% withdrawals for the next 30 years without ever checking up on my portfolio's performance!"
Along these lines, go back to 2009 when CAPE was actually below its mean & median and see if you can find a single post/blog/article from a person recommending people to retire and take out 4.5% "because valuations are low". It wasn't that long ago so it should be easy to find thousands of posts from valuation/CAPE believers advocating retirees crank up their SWR above 4%.......I mean, never mind that the New York Times was still running articles with headlines like "Steep Market Drops Highlight Despair Over Rescue Efforts" just go ahead and withdraw 4.5% now because CAPE.
CAPE was below it's mean and median for about 6 weeks in 2009. Maybe 8 weeks. At the very bottom of a recession that was nearly a full blown financial panic.

The rest of the past 26 years (since 1992), CAPE has been "high".
The J stands for Jay

AlohaJoe
Posts: 3851
Joined: Mon Nov 26, 2007 2:00 pm
Location: Saigon, Vietnam

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by AlohaJoe » Mon Aug 13, 2018 11:51 pm

HomerJ wrote:
Mon Aug 13, 2018 11:27 pm
CAPE was below it's mean and median for about 6 weeks in 2009. Maybe 8 weeks.
That's a bit of an exaggeration. Looking at Shiller's spreadsheet, CAPE10 was below 16.5 from October 2008 to July 2009 and it was below 15.5 from November 2008 to April 2009. So anywhere from 6-10 months depending on whether you're talking about mean or median. That seems like plenty of time for valuation-addicts to have cranked out hundreds of articles about how low valuations mean we should all leverage up to 120% stocks ;)
At the very bottom of a recession that was nearly a full blown financial panic.
Yes, that is my point. People that focus too much on CAPE often ignore the context. They also only seem to use CAPE as a negative ratchet. If CAPE was low I imagine they'd find some other excuse for why withdrawal rates should be below 4%. The main reason I wish CAPE would go low again: I want to see what the next set of excuses will be for Negative Nancies.

User avatar
willthrill81
Posts: 6170
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by willthrill81 » Tue Aug 14, 2018 12:36 am

randomguy wrote:
Mon Aug 13, 2018 11:20 pm
People on this board are always looking for a reason to be pessimistic.
Bogleheads generally like to call themselves 'conservative', but 'pessimistic' or 'looking for the cloud of every silver lining' would be nearer the mark.
randomguy wrote:
Mon Aug 13, 2018 11:20 pm
In 2009 nobody was talking about doing a 6% SWR because valuations were down at 15.:)
AlohaJoe wrote:
Mon Aug 13, 2018 11:51 pm
Looking at Shiller's spreadsheet, CAPE10 was below 16.5 from October 2008 to July 2009 and it was below 15.5 from November 2008 to April 2009. So anywhere from 6-10 months depending on whether you're talking about mean or median. That seems like plenty of time for valuation-addicts to have cranked out hundreds of articles about how low valuations mean we should all leverage up to 120% stocks ;)
This provides solid evidence, I think, that most CAPE proponents are only interested in using it as an excuse to be pessimistic and build even bigger portfolios. I might be reading too much into this, but I think that some people believe that a sub 3% withdrawal rate is something to brag about. :?
“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: 6504
Joined: Wed Sep 17, 2014 9:00 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by randomguy » Tue Aug 14, 2018 1:42 am

AlohaJoe wrote:
Mon Aug 13, 2018 11:51 pm
HomerJ wrote:
Mon Aug 13, 2018 11:27 pm
CAPE was below it's mean and median for about 6 weeks in 2009. Maybe 8 weeks.
That's a bit of an exaggeration. Looking at Shiller's spreadsheet, CAPE10 was below 16.5 from October 2008 to July 2009 and it was below 15.5 from November 2008 to April 2009. So anywhere from 6-10 months depending on whether you're talking about mean or median. That seems like plenty of time for valuation-addicts to have cranked out hundreds of articles about how low valuations mean we should all leverage up to 120% stocks ;)
Depends what median and mean you talking about. The one in 2009 or the one in 2018? The averages have been creeping up every year since I first heard about the number. At the current rate, in 50 years we will be talking about how PE10s of 30 are low:)

And the real question is how many people believe in PE10s enough to go invest in the countries that are on sale right now with low PE10s?:)

greenhill
Posts: 26
Joined: Sun Aug 06, 2017 4:27 am

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by greenhill » Tue Aug 14, 2018 1:59 am

CraigTester wrote:
Mon Aug 13, 2018 10:10 pm
randomguy wrote:
Mon Aug 13, 2018 9:41 pm
CraigTester wrote:
Mon Aug 13, 2018 9:16 pm

William Bengen (who did much of the work to coin the 4% rule we hear about today) really did nothing more than a "least common denominator" analysis.

In other words, based on all the retirement start dates available within his data range, he simply wanted to know who ran out of money first -- for a given withdrawal rate.

The unlucky winner happened to be the guy that retired in 1966.

The "4% rule" evolved from here to state that, as long as your not more unlucky than the 1966 guy, you'll be fine....

Understanding this, it really shouldn't be a surprise at all that Kitces analysis doesn't cast any doubt on the 4% rule. That was not really the point.

Instead, what Kitces is revealing for us is some pretty interesting data on where starting valuations happened to be when past retirees started their 30 year clock.

And what's interesting to note is that the same poor 1966 retiree who failed first, happened to also start his retirement at a moment when valuations were quite high.

He then goes on in his chart to plot where valuations were when all our historical retirees each began their 30 year run.

And the fascinating (though arguably intuitive) take-away is that Valuation matters.

In other words, if you start your 30 year retirement when valuations are high, you don't tend to have near as cushy of a ride as those who start their retirement when valuations are less stretched.

Again, if you choose to not believe that valuation matters, you can just ignore all of this historical "coincidence".

But for those willing to look at the empirical results, its extremely clear that valuation is the single biggest determinant of your future returns.

And as Kitces charts show, this simple truth extends to SWR's as well....

The problem is that there are very few periods with pe10s over 24 in history. Right now we have 1929 and 1966 and we are drawing our conclusions from them. Lets look at some data points
Assume a 50/50 portfolio
1966 PE10 was 24
1996 pe10 24.76 1.47 million dollars real dollars
1997 pe10 28.33 1.32 million
1998 pe10 32.8 1.05 million
2004 pe10 27.66 1.05 million
Obviously none of them are over yet but if you could sell and buy TIPs, you would succeed every time. In another 15 years PE10s it might turn out that PE10s>24 still result in 5% SWR half the time. Or maybe we are due for another really horrible decade (i.e. 0% real return from both bonds and stocks) and some of these fail.

The thing with valuations based on PE10 is that we are either in the biggest bubble ever(i.e. something like 29 of the last 30 years are above the historical average and the only reason the other isn't is because of the inflation of the average over the past 25 years) or something has changed which makes higher valuations now more reasonable than they were from 1880-1990. Good luck answering that question.
I am new to this forum, but what I'm gathering in my first few days, is there has been a rash of criticism of PE10, market timing, etc, etc.

So now if anyone even mentions these terms, they are summarily taken to the wood shed...

But as a guy who hasn't been through these battles, here's how I very simply interpret the subject article of this thread.

If you retire when valuations are at historical highs, the one thing you know for sure is that you are a "high probability candidate" for being the next "least common denominator".

So why even put yourself at risk?

All we really "know" is that PE10 is presently at its second highest level in history.

And whenever we've reached these types of levels before, long term returns have been abysmal.

But perhaps this time will be different....
Okay, so you are worried about the PE10. Am I concerned? Yes. Am I going to lower my stock allocation? No.

Treasuries was yielding near 5% back in 1966. About 6% in 1996. Just under 3% now. We are in a restaurant which sells hamburgers and French fries only. You are worried about hamburgers being more expensive than yesterday, but French fries are getting more expensive as well.

Bottom line: I will just save more, build a larger portfolio and lower my expectation. But I won’t change my strategy just because of PE10.

Random Walker
Posts: 3207
Joined: Fri Feb 23, 2007 8:21 pm

Re: Importance of sequence of real returns for SWR, CAPE valuation, and Luck

Post by Random Walker » Tue Aug 14, 2018 8:18 am

HomerJ wrote:
Mon Aug 13, 2018 10:43 pm
Random Walker wrote:
Mon Aug 13, 2018 10:30 pm
I think it’s Bernstein or maybe Zwecher who refers to “Murphy’s Law Of Retirement”. An investor profits from a long bull market. As valuations rise, his portfolio swells. Seeing the larger than expected portfolio, the investor decides to retire. But he is retiring into a period of high valuations and low future expected returns.

Dave
But if he reads this forum, he changes his Asset Allocation to be more conservative as he retires and moves money into bonds, LOCKING IN SOME OF THOSE GAINS.

Someone who saw their stock money double from 1996 to 2000, if they followed the standard advice on this forum, probably did just fine. Their portfolio swelled, they decided to retire, and moved from say 60/40 to 40/60, and got to keep most of that money. Actually ALL of it, since they could live off the bonds while waiting for the stocks to recover.

Not once did they have to look at valuations for this strategy to work.

The key variable is the fact that they retired. Regardless of valuations, they should have changed their allocation to be more conservative the moment they realized retirement was close or at hand.
Basically we’re on the same page, but valuations do have some information in them. Why not add that bit of information to the decision process you describe. Knowing that valuations are high, mean expected returns low, and whole dispersion of returns shifted left, the retiring investor might fine tune the decision we are talking about a bit.

Dave

Post Reply