Valuation Matters When Choosing a Safe Withdrawal Rate

[Animorph]

Video/transcript at M* with Kitces about how market valuation maters when calculating your SWR. Just a few numbers, but he's talking 4.5% with high valuations and 5.5% with low valuations, with the expectation that you will probably be able to raise your standard of living later in retirement if the worst case doesn't materialize. Not sure if it requires premium membership to view.

Valuation Matters When Choosing a Safe Withdrawal Rate

"So how I ultimately boiled that down in the research was to find if, in essence, that 4.5% withdrawal rate is a great starting point when you're actually dealing with a bad-valuation environment. But when you're dealing with merely average valuations, just things that aren't real bad, you should be talking about a number that's more like 5% than 4.5%. And if you're actually in a favorable-valuation environment, the number you should be talking about is more like 5.5%, and in fact there is a good chance you're going to be able to raise your spending further from there because you tend to get enormous bull-market runs that start in really low-valuation environment."

 

[ziggy29]

I think a lot of folks accept that conceptually. But I think many of us find financial planning easiest and least fretful when we take emotions and guesswork out of it. For example, we may have a target AA and rigid "trigger points" for when we rebalance -- so we take guesswork out of it as well as decisions made by fear or greed. Similarly, instead of trying to "guess" at a safe withdrawal rate based on current valuations, we may just accept 3-4% and stick with it.

I think it's pretty intuitively obvious that a higher withdrawal rate would seem sustainable in (say) 1982 or 2009 than in 1999 or 2007. But just how much? It's pretty subjective, and I think a lot of folks would rather work with a number carved in stone (which may have to err on the side of being too conservative to work in virtually all situations, including retirement in 1929 and 1966).

 

[obgyn65]

I am not sure I agree with a 5% SWR with merely average valuations. It sounds very optimistic IMHO.

But when you're dealing with merely average valuations, just things that aren't real bad, you should be talking about a number that's more like 5% than 4.5%.

 

[Gone4Good]

I am not sure I agree with a 5% SWR with merely average valuations. It sounds very optimistic IMHO.

As does 4.5% with 'high valuations.' It's pretty clear someone who retired in 2000 drawing 4.5% real is in deep trouble.

I'm certainly in favor of incorporating valuation opinions in my investment decissions. What's not clear to me, though, is how you roll something like that out to the masses. Unless I've missed it, nobody has developed a thermometer for valuation that everyone agrees with.

So the revevant question becomes, are markets currently 'overvalued', 'undervalued' or 'just right'? Unless you're pretty confident that you know, you probably default back to the most conservative withdrawal rate that funds your retirement . . . <=4%.

 

[chinaco]

Thanks for the link.


What I heard...

He says that the bad times... for example, approaching the Great Depression (high valuation) and the subsequent large number of years it took to recover caused the problem.

I am wondering if we are looking at one of those periods right now..... 10 years to recover. We just had a huge financial crisis.

That is the problem, no one can predict the future. But one might be able to manage to it. Which seems to be his message.

If one has a large amount of discretionary spending (above their base lifestyle)... there may be some cushion to mitigate the risk.

 

[MichaelB]

Valuations matter. What measure of valuation should be used to make this determination?

 

[MasterBlaster]

This topic has been studied a fair amount.

here is but one paper on predicting SWR normalized by PE10 valuation, dividend yield, and treasury bond yield:

Can We Predict the Sustainable Withdrawal Rate for New Retirees?

The following chart is the result.

 

[Independent]

Interesting chart. The fit between the blue and green is pretty impressive.

I see that his formula gives a "Predicted MWR" under 2% for recent years. That must be related to low earnings.

 

[Gone4Good]

I see that his formula gives a "Predicted MWR" under 2% for recent years. That must be related to low earnings.

I think this quote says it all . . .

there are important qualifications for these predictions. Most importantly, they depend on out-of-sample estimates, as the circumstances of the past 15 years have not been witnessed before.

We've had periods where PE-10 was greater. We've had periods where 10-yr rates were about where they are now. And we've had periods where dividend yields were lower.

But never have we had a period where PE-10, bond yields and stock yields were all as ugly as they are currently.

 

[MichaelB]

This topic has been studied a fair amount.

here is but one paper on predicting SWR normalized by PE10 valuation, dividend yield, and treasury bond yield:

Can We Predict the Sustainable Withdrawal Rate for New Retirees?

Right. Crestmont Research has done lots of research on starting PE and portfolio survival as well. This paper Destitute At 80: Retiring In Secular Cycles here is an example and points to a similar conclusion. They have other interesting papers as well.

 

[unclemick]

Sans the math I use unclemick's unified general theory of chickenheartedness: sometimes written as - place finger in belly button and review ending balance of 'your' portfolio. Pick the coming years SWR based on the 'belly button quiver'. Handgrenade wise roughly between 2 to 6% range, 4% benchmark, give or take.

Only since ER in 1993 so it needs more data so I plan to live longer.

heh heh heh - tongue in cheek but very close to what I've actually been doing these last 18 years.

 

[Koolau]

I've never been quite as scientific about it as unclemick, but, in essence, the "finger in the wind" approach seems to have served me reasonably well when it comes to SWR's. I used the "rigorous" version for the last few accumulation (aka planning) years before ER, but since then, it's been "play it by ear", listen to your fears, go with your gut, blah, blah, blah. 4% SWR? Give me a break!

Belt and suspenders, baby. Belt and suspenders.

YMMV.

 

[Telly]

Ya know, the more I read on this thread, the more I think of he-whose-name-shall-not-be-mentioned (aka *****).
Maybe we should invite him back here for a review [-]essay[/-] [-]doctoral thesis[/-] [-]treatise[/-] [-]monograph[/-] [-]economic version of Tolstoy's War and Peace[/-] [-]tome including all written words in every language of every civilization[/-].

I'm game

 

[bbbamI]

If one has a large amount of discretionary spending (above their base lifestyle)... there may be some cushion to mitigate the risk.

Our discretionary expenses are approx 20% of our budget.

I was told 4% was the right SWR...so I chose 3%. It has been said you need 25 times your annual expenses (if you are 65) in order to retire. We're 53/57 and have 38 times our annual expenses.

So I feel comfortable with our WR as I feel our cushions are fluffy enough.

btw....do those cushions make my butt look big?

 

[Nords]

Unless I've missed it, nobody has developed a thermometer for valuation that everyone agrees with.

Ya know, the more I read on this thread, the more I think of he-whose-name-shall-not-be-mentioned (aka *****).
Maybe we should invite him back here for a review [-]essay[/-] [-]doctoral thesis[/-] [-]treatise[/-] [-]monograph[/-] [-]economic version of Tolstoy's War and Peace[/-] [-]tome including all written words in every language of every civilization[/-].

As Telly's pointed out, you didn't miss anything.

I'm not even going to link to his book or his website for fear that he'd have one of his people start posting a defense of the "truth"...

 

[Animorph]

I don't think we should overly complicate something so inherently unpredictable as a future SWR, but I think it does point out that if the market is not close to a peak when you are checking SWR with FIRECalc then the results will be somewhat conservative. "Somewhat" because the data is yearly and doesn't hit market peaks exactly. So it is always checking to see what happens if the Great Depression starts tomorrow. If you were checking with it at the bottom of 2009's market you were daisy-chaining two gigantic market dips together and seeing if you could survive that. So I'm not too concerned when my WR looks high during market dips as long as it looks conservative during the peaks.

 

[jdw_fire]

I think a lot of folks accept that conceptually. But I think many of us find financial planning easiest and least fretful when we take emotions and guesswork out of it. For example, we may have a target AA and rigid "trigger points" for when we rebalance -- so we take guesswork out of it as well as decisions made by fear or greed. Similarly, instead of trying to "guess" at a safe withdrawal rate based on current valuations, we may just accept 3-4% and stick with it.

I think it's pretty intuitively obvious that a higher withdrawal rate would seem sustainable in (say) 1982 or 2009 than in 1999 or 2007. But just how much? It's pretty subjective, and I think a lot of folks would rather work with a number carved in stone (which may have to err on the side of being too conservative to work in virtually all situations, including retirement in 1929 and 1966).

I am not sure I agree with a 5% SWR with merely average valuations. It sounds very optimistic IMHO.

As does 4.5% with 'high valuations.' It's pretty clear someone who retired in 2000 drawing 4.5% real is in deep trouble.

I'm certainly in favor of incorporating valuation opinions in my investment decissions. What's not clear to me, though, is how you roll something like that out to the masses. Unless I've missed it, nobody has developed a thermometer for valuation that everyone agrees with.

So the revevant question becomes, are markets currently 'overvalued', 'undervalued' or 'just right'? Unless you're pretty confident that you know, you probably default back to the most conservative withdrawal rate that funds your retirement . . . <=4%.

Valuations matter. What measure of valuation should be used to make this determination?

he explains it more completely (and backs it up with data) in his paper, not just gut feel (nor all that subjective).