The “Safe Withdrawal Rate” May Be Changing

[TromboneAl]

The ever shrinking SWR must have a floor. After all, a 4% SWR requires to accumulate 25 x annual living expenses, so even with a long period of real returns at 0%, one's money should last at least 25 years,

That's a good point, but it's true only if the real returns have a floor. But real return could be less than zero. Also, even with a real return of zero, volatility might make the stash last less than 25 years.

 

[FIREd]

That's a good point, but it's true only if the real returns have a floor. But real return could be less than zero. Also, even with a real return of zero, volatility might make the stash last less than 25 years.

If you use TIPS, you do put a floor under real returns, since TIPS guarantee that your real return will be at least positive. And when using TIPS, as I explained above, volatility is irrelevant since you hold them to maturity. Now if you have a traditional stock/bonds portfolio, then you are right, nothing is guaranteed.

 

[FIREd]

My bold.

The world will end next week! Can you prove me wrong today?

Surely, as a man with a science degree, you do see a difference between your example and mine. My assertion (that lower real returns in the future would not support a 3.1% SWR) can be proven right or wrong mathematically. A simple excel spreadsheet for example will show you that, on average, you need an annual real return of at least 1% (with zero volatility) for a 3.1% SWR to be viable over a 40-year period.

That real return would have to be, on average, at least 2% per year for a 3.1% SWR to be viable over a 50-year period. If annual real returns are only 1% on average over the next 50-year period, then your SWR would have to drop to 2.5% for your portfolio to survive at least 50 years.

 

[TromboneAl]

Here's a summary of the article in four words:

Things are different now.

 

[Rich_by_the_Bay]

Here's a summary of the article in four words:

Things are different now.

Good summary.

But aren't things always different?

 

[youbet]

If real return is zero, this is equivalent to zero inflation with zero returns. Then if you draw out 1/25th of your money every year, it will last 25 years.

Peter

For those of us above the zero fed + state income tax level, adjustments must be made. A real pretax return of zero = a negative real return after taxes.

 

[rayvt]

I was not impressed by his article. It struck me as nothing but yet another "publish or perish" quality writing.

The question is always, "What can you do about it?" "What are your alternatives----and how many of them are real?"

If the future is going to be crappy with real returns much much lower than the historical returns, then we are in a world of hurt. But you know what? The financial/investment market isn't a world isolated to itself. If thngs are bad there, that's because things are bad in the entire economy. So there will be nowhere to hide. Lowering your SWR to 2% or 3% isn't going to keep you safe----it's just going to keep you going a little bit longer before you crash.

The primary thesis of his article is "Things are different now." But this is the biggest & most common economic fallacy there is. As long as human nature is human nature, then things are NEVER different now.

 

[audreyh1]

For those of us above the zero fed + state income tax level, adjustments must be made. A real pretax return of zero = a negative real return after taxes.

Your SWR includes taxes paid on your investments. So makes no difference to the SWR scenario.

Audrey

 

[youbet]

Surely, as a man with a science degree, you do see a difference between your example and mine. My assertion (that lower real returns in the future would not support a 3.1% SWR) can be proven right or wrong mathematically. A simple excel spreadsheet for example will show you that, on average, you need an annual real return of at least 1% (with zero volatility) for a 3.1% SWR to be viable over a 40-year period.

That real return would have to be, on average, at least 2% per year for a 3.1% SWR to be viable over a 50-year period. If annual real returns are only 1% on average over the next 50-year period, then your SWR would have to drop to 2.5% for your portfolio to survive at least 50 years.

Although I assume your math is correct, basing your analysis on "average" returns makes it meaningless. The sequence of annual returns over the period is key and must not be ignored.

The examples you give have never even come close to occuring historically and I'm betting never appear in the future either.

edit: you do mention "zero volatility" so apparently you're aware of the unlikelyhood of your numbers ever coming to bear in reality. But for others..... You can't work with average returns unless you're expecting future returns to be the same year after year after year. Sequence counts.

 

[youbet]

Your SWR includes taxes paid on your investments. So makes no difference to the SWR scenario.

Audrey

Ahhhh..... No. SWR is an estimate, based on historical investment return and inflation rates over a given time period, that does not include taxes.

Perhaps we're talking apples and oranges. My comment refered to real rates of return. In a zero real rate of return environment (returns = inflation), reducing the pre-tax return by the amount it is taxed (specific to you) will drive your apparent return negative.

 

[audreyh1]

Ahhhh..... No. SWR is an estimate, based on historical investment return and inflation rates over a given time period, that does not include taxes.

Perhaps we're talking apples and oranges. My comment refered to real rates of return. In a zero real rate of return environment (returns = inflation), reducing the pre-tax return by the amount it is taxed (specific to you) will drive your apparent return negative.

To stay within your selected SWR, all taxes on the portfolio or other tax expenses must come out of your SWR, otherwise you are actually withdrawing at a higher rate. Taxes are considered part of your expenses in all the SWR models. To say SWR does not include taxes is incorrect.

In a zero real return environment, your 4% withdrawal will still last you 25 years because you are paying your taxes out of that 4% withdrawal.

Audrey

 

[youbet]

To stay within your selected SWR, all taxes on the portfolio or other tax expenses must come out of your SWR, otherwise you are actually withdrawing at a higher rate. Taxes are considered part of your expenses in all the SWR models. To say SWR does not include taxes is incorrect.

In a zero real return environment, your 4% withdrawal will still last you 25 years because you are paying your taxes out of that 4% withdrawal.

Audrey

We seem to have two subjects here.

1. My original comments refered to income taxes reducing a zero real return to negative. This was without reference to SWR, which you added.

2. Taxes and SWR. The way I look at it Audrey, if I'm using a 4% SWR for my annual withdrawal, I withdraw $40k out of $1MM. I then include taxes as a budget item in spending the $40k.

In my thinking return (real or nominal) and SWR are NOT the same thing and I don't use them interchangably.

 

[Onward]

This is my route. If things go consistently poorly I guess you could end up pretty bare bones but what is the alternative - never retiring?

That seems to be the attitude of so many journalists these days. The best way to "master your retirement" is ... to keep working! Till you're 70! Till you're 80! Never mind that half of America is ready to blow their brains out every Sunday night.

This isn't really a comment on Scott Burns' article. He's maybe my favorite financial writer.

 

[audreyh1]

We seem to have two subjects here.

1. My original comments refered to income taxes reducing a zero real return to negative. This was without reference to SWR, which you added.

2. Taxes and SWR. The way I look at it Audrey, if I'm using a 4% SWR for my annual withdrawal, I withdraw $40k out of $1MM. I then include taxes as a budget item in spending the $40k.

In my thinking return (real or nominal) and SWR are NOT the same thing and I don't use them interchangably.

I assumed you were referring to SWR since SWR in low return environments is the subject of this thread and the zero real return example was referring to SWR.

Audrey

 

[youbet]

I assumed you were referring to SWR since SWR in low return environments is the subject of this thread and the zero real return example was referring to SWR.

Audrey

OK. No problem.

 

[walkinwood]

There's this thing called the Recency Effect.

I see no explanation from Scott on why dividends will remain low in the future relative to the past, why equities will appreciate less in the future relative to the past - other than to say that that's what has happened in the recent past.

 

[chinaco]

Well, If that happens we all will adjust.

 

[fritz]

Vanguard just recently held a webcast for Flagship members regarding withdrawal strategies for taxable investors. It discussed SWRs of 4~5% (although the success rate wasn't 100%) for stock/bond asset allocations of 50/50. I've attached the slides from that preso - pretty sure I can get the link to the rebroadcast. I prefer the percent of portfolio strategy where you shouldn't deplete your funds - Vanguard is going to publish a follow up paper on that strategy with a ceiling floor SWR.

Kumquat - don't worry about the end of the world - the Mayan's "cut it in stone" for 2012.....

 

[CaseInPoint]

Vanguard just recently held a webcast for Flagship members regarding withdrawal strategies for taxable investors. It discussed SWRs of 4~5% (although the success rate wasn't 100%) for stock/bond asset allocations of 50/50. I've attached the slides from that preso - pretty sure I can get the link to the rebroadcast. I prefer the percent of portfolio strategy where you shouldn't deplete your funds - Vanguard is going to publish a follow up paper on that strategy with a ceiling floor SWR.

Kumquat - don't worry about the end of the world - the Mayan's "cut it in stone" for 2012.....

Interesting presentation. I really should pay more attention to my VG email.

I'm curious, though, about slide #10 which says "No taxes—assumed to be paid from the $47,500." Why would there be an assumption of no taxes?

 

[CaseInPoint]

If you don't have strong real returns, then your SWR will have to plunge to 2.5% or even 2% in order to make it work.

That's probably true, and I really have to salute anyone who can happily survive on 2% of their portfolio.

 

[samclem]

Heretical comment follows.

I think it's useful to at least consider the possibility that the real anomaly has been the US equity performance for the 20th century. That's the data set that FIRECalc and most other simulations are based on, explicitly or implicitly. When we consider the US geopolitical preeminence and the economic boom that resulted, particularly since WW-II, it's quite easy to argue that this data set (this particular nation in these particular years) is really the likely anomaly. The same holds true, to a lesser extent, for the US business environment and conditions favoring US economic expansion over the last century--they were certainly unusually good when considered in the entire context of human history. So, when someone says that the future years are really going to be "different this time," we should ask if they mean "different from the recent very anomalous 'good years' in this one particular country" (7-10% PA total returns) or "different from what most developed countries in the modern age have experienced" (2-3% total returns)? Put another way--it's very possible that it really was "different this time" for the last century, and now we're going to experience the "normality" most countries have experienced.

By the same token, technological advances or even advances in our understanding of how economic systems function might result in even higher growth rates than those recently experienced. Or, an asteroid could hit us. We don't know

If a person is ever going to retire, they have to make some assumptions. But (IMO) anybody believing they can set up a fixed 40 year ironclad SWR based on history is fooling themselves. Anyone who comes away from FIRECALC believing there's a meaningful difference going forward between a portfolio allocation/withdrawal rate that generates an 80% historic survival rate and one that generates a 95% survival rate might be overestimating the power and precision of the tools and underlying data in forecasting the future.

What's the solution? I don't know. My particular answer has been:
- Continue to work and squirrel away money longer than "the data" says I need to. Quitting work now and possibly returning to the work force in 5-10 years is not a good answer for me for several reasons. But, I downshifted into work that is more psychologically sustainable.
- Plan to take a year-end percentage rather than a fixed percentage adjusted for inflation. I'm going to use 3.0 - 3.5% at first.
- Plan to be flexible. Monitor the buying power of my portfolio and make adjustments if it starts to slip. Adjustments would include lower annual withdrawals and taking more risk. If the buying power climbs, we'll raise withdrawals and/or reduce the risk we take in our portfolio ("risk" being broadly defined: due not only to volatility, but also from higher inflation, long-term reduced equity returns going forward, etc. Figuring out how to do this might be my new "career").

Sorry, no huge revelations in the foregoing observations. We're all in "measure with micrometer, mark it with a grease pencil, cut with an axe" territory, and what we're measuring is a blob of Jello. And the world going forward might be pudding.

 

[chinaco]

If one is concerned, they should consider taking a more conservative approach.

It would seem to me that a large part of the risk (being discussed) depends on how much discretionary spending is factored in versus non-discretionary (basic lifestyle). discretionary spending could flex a bit to absorb the impact of lower returns and likely preserve the basic lifestyle (comfortable and ability to meet basic needs).

DW and have enough cushion over and above our current spending level can flex.


IMO - One should be fairly confident about being able to provide for their basic lifestyle (for their expected lifespan) if they are considering ER. If not, perhaps the plan should be adjusted. They could w*rk a bit longer or w*rk partime. For many, it is easier to w*rk one more year than go back to w*rk after several years off. I know if I quite w*rking for 2-5 years, I will probably not be able to earn the same wage that I earn today. Plus, if a health problem occurs, I may not be able to earn at all.


Assuming one can fund their lifestyle (with a conservative estimate/projection)... I believe other factors can present far more risk. People have some level of control over their spending and can exercise it. Plus it is easy to understand. IMO - There is more risk in people making investment mistakes by unwittingly taking on too much risk or risks they did not understand.

 

[donheff]

Vanguard is going to publish a follow up paper on that strategy with a ceiling floor SWR.

I don't pay attention to my VG emails either. If you catch the ceiling/floor presentation followup here to remind those of us who are interested to check it out.

 

[fritz]

Interesting presentation. I really should pay more attention to my VG email.

I'm curious, though, about slide #10 which says "No taxes—assumed to be paid from the $47,500." Why would there be an assumption of no taxes?

I took it that they were only concerned with the rate of return on investments for the presentation, and did not account for any income tax scenarios (assumed that an individual's SWR from their investments includes what's appropriate for paying fed/state/local taxes).

 

[fritz]

I don't pay attention to my VG emails either. If you catch the ceiling/floor presentation followup here to remind those of us who are interested to check it out.

The next Flagship web cast is coming up on July 22nd about managing your portfolio in retirement (Retirement is only the beginning). I believe these are scheduled by invitation only, and not done for their general investor population due to limitations of allowing participants to interact with the presenters during the presentation.

If I manage to get the follow up white paper on percentage of portfolio withdrawals with a ceiling/floor scenario - will follow up. I don't always get (or pay attention to) all their correspondence either - if someone here participated in that webcast and should get it - would appreciate their follow up. Unfortunately, I was unable to successfully connect to the original webcast in March. It was redone in May (I was not on that invite list, as I was on the first one). Attached is the link to the May 27th presentation's replay (is still accessible for now).

https://admin.adobeconnect.com/_a45190094/p69664262