New Withdrawal Strategies from T. Rowe Price
New Withdrawal Strategies from T. Rowe Price
The Fall issue of the T. Rowe Price Report has an excellent article on Withdrawal Strategies. They present a Valuation Strategy and a Dynamic Strategy. See the article starting on page 15.
http://individual.troweprice.com/public ... Report.pdf
The Valuation Strategy has separate parameters for bonds and stocks. In the case of bonds, the initial withdrawal percentage is a function of the 10 year Treasury yield at retirement. A chart is provided showing the suggested withdrawal percentage as a function of Treasury yield. A higher initial Treasury yield translates to a higher withdrawal rate.
In the case of stocks, the initial withdrawal percentage is based on the 12 month trailing PE of the S&P 500 Index at retirement. The initial withdrawal percentage is higher for lower PE values.
The bond and stock withdrawal percentages can be combined for the typical balanced portfolio.
The Dynamic Withdrawal Strategy attempts to deal with market declines. The annual inflation adjustment is not taken following years of negative portfolio return.
Do these strategies look like a reasonable way to go? I don't recall seeing the valuation strategy based on initial bond yield and initial 12 month trailing PE on the S&P 500.
http://individual.troweprice.com/public ... Report.pdf
The Valuation Strategy has separate parameters for bonds and stocks. In the case of bonds, the initial withdrawal percentage is a function of the 10 year Treasury yield at retirement. A chart is provided showing the suggested withdrawal percentage as a function of Treasury yield. A higher initial Treasury yield translates to a higher withdrawal rate.
In the case of stocks, the initial withdrawal percentage is based on the 12 month trailing PE of the S&P 500 Index at retirement. The initial withdrawal percentage is higher for lower PE values.
The bond and stock withdrawal percentages can be combined for the typical balanced portfolio.
The Dynamic Withdrawal Strategy attempts to deal with market declines. The annual inflation adjustment is not taken following years of negative portfolio return.
Do these strategies look like a reasonable way to go? I don't recall seeing the valuation strategy based on initial bond yield and initial 12 month trailing PE on the S&P 500.
Enjoying the Outdoors
Re: New Withdrawal Strategies from T. Rowe Price
Interesting article and according to them, if I read it right, since stocks are quite high now, it's time to reduce my withdrawal rate to around 3.8 %.
How curious in that, that is what I was planning to do anyway.
How curious in that, that is what I was planning to do anyway.
- nisiprius
- Advisory Board
- Posts: 52214
- Joined: Thu Jul 26, 2007 9:33 am
- Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry
Re: New Withdrawal Strategies from T. Rowe Price
RANT ON
I detest dynamic or flexible withdrawal strategies because I cannot, for the life of me, understand how one can possibly compare one of them against another to determine which is better, or why any of them would be better than Taylor Larimore's withdrawal strategy.
Taylor's strategy is remarkably similar to what the authors of the original Trinity study said: "The word planning is emphasized because of the great uncertainties in the stock and bond markets. Mid-course corrections likely will be required, with the actual dollar amounts withdrawn adjusted downward or upward relative to the plan. The investor needs to keep in mind that selection of a withdrawal rate is not a matter of contract but rather a matter of planning."
All of these things seem theoretical and hypothetical, and there has been an amazing difference in the extremely precise results that are stated over time: from 7% in the mid 1990s, to 4% in the late 1990s and early 2000's, drifting more and more to 3% today.
It is notable that Vanguard, after launching the Managed Payout Growth & Distribution Fund with an initially targeted payout rate of 5%, has now, just six years later, reduced it to 4%. Oops, folks, we were overoptimistic by 25%, sorry about that.
I doubt that very many people have actually spent 25 years in retirement carefully following any single, formally-defined withdrawal strategy, or that there have been any sociological studies to find out how they've done.
It's all handwaving, seat-of-the-pants, shoot-from-the-hip stuff presented as financial science. At least the Trinity authors had the humility and the frankness to admit it. It's interesting that T. Rowe Price describes the 4% rule as "reliable" when the original authors of the rule said that it should NOT be considered reliable!
Understandably, the promulgators of such withdrawal strategies compete to see who can tell you how to "safely" withdraw the most, and tout the "good news" that they will give you permission to withdraw more than the other guy... but this seems like a race-toward-the-risky.
T. Rowe Price's idea is that "retirees could pursue an alternate strategy of not taking inflation increases in years following those in which their portfolios lost money." They say, blandly, "This can be a very powerful strategy. Giving up your relatively small, annual inflation increase every now and then is a small price to pay for that higher initial withdrawal rate," but who says the increase is always going to be "relatively small" and who says it is a "small price to pay?" And why is foregoing, let us say, 3% inflation increase (average historic rate) any less painful than a 3% decline in purchasing power for any other reason? Historically, it's not that unusual to have several bad years in a row, so you might be experiencing a cumulative effect of much more than 3%.
If it's as uncertain as I believe it to be, I don't see that there's much advantage in going with some codified system like that baked into Vanguard's Managed Payout funds, or these proposed T. Rowe Price systems over using the same intuition my wife and I have used for forty years to regulate our spending to fluctuating income and expenses.
RANT OFF
I detest dynamic or flexible withdrawal strategies because I cannot, for the life of me, understand how one can possibly compare one of them against another to determine which is better, or why any of them would be better than Taylor Larimore's withdrawal strategy.
Taylor's strategy is remarkably similar to what the authors of the original Trinity study said: "The word planning is emphasized because of the great uncertainties in the stock and bond markets. Mid-course corrections likely will be required, with the actual dollar amounts withdrawn adjusted downward or upward relative to the plan. The investor needs to keep in mind that selection of a withdrawal rate is not a matter of contract but rather a matter of planning."
All of these things seem theoretical and hypothetical, and there has been an amazing difference in the extremely precise results that are stated over time: from 7% in the mid 1990s, to 4% in the late 1990s and early 2000's, drifting more and more to 3% today.
It is notable that Vanguard, after launching the Managed Payout Growth & Distribution Fund with an initially targeted payout rate of 5%, has now, just six years later, reduced it to 4%. Oops, folks, we were overoptimistic by 25%, sorry about that.
I doubt that very many people have actually spent 25 years in retirement carefully following any single, formally-defined withdrawal strategy, or that there have been any sociological studies to find out how they've done.
It's all handwaving, seat-of-the-pants, shoot-from-the-hip stuff presented as financial science. At least the Trinity authors had the humility and the frankness to admit it. It's interesting that T. Rowe Price describes the 4% rule as "reliable" when the original authors of the rule said that it should NOT be considered reliable!
Understandably, the promulgators of such withdrawal strategies compete to see who can tell you how to "safely" withdraw the most, and tout the "good news" that they will give you permission to withdraw more than the other guy... but this seems like a race-toward-the-risky.
T. Rowe Price's idea is that "retirees could pursue an alternate strategy of not taking inflation increases in years following those in which their portfolios lost money." They say, blandly, "This can be a very powerful strategy. Giving up your relatively small, annual inflation increase every now and then is a small price to pay for that higher initial withdrawal rate," but who says the increase is always going to be "relatively small" and who says it is a "small price to pay?" And why is foregoing, let us say, 3% inflation increase (average historic rate) any less painful than a 3% decline in purchasing power for any other reason? Historically, it's not that unusual to have several bad years in a row, so you might be experiencing a cumulative effect of much more than 3%.
If it's as uncertain as I believe it to be, I don't see that there's much advantage in going with some codified system like that baked into Vanguard's Managed Payout funds, or these proposed T. Rowe Price systems over using the same intuition my wife and I have used for forty years to regulate our spending to fluctuating income and expenses.
RANT OFF
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
Re: New Withdrawal Strategies from T. Rowe Price
The TRowePrice article just codifies nisiprius's intuition. It will be helpful to the folks who like to work with numbers instead of with intuition. I predict the article will become the subject of future articles in the New York Times, USA Today, Wall Street Journal, blogosphere, etc.
But anybody who blindly trusts numbers should think about why they need or want to do that in the first place.
But anybody who blindly trusts numbers should think about why they need or want to do that in the first place.
Re: New Withdrawal Strategies from T. Rowe Price
One modest observation. Many retirees have tax-deferred (IRA) CD ladders in addition to 401K/IRA bond funds. The performance of a CD ladder over thirty years might well vary from the performance of the hypothetical "bond fund" used in the TRowePrice analysis. Moving right along, not all CD ladders are created equally.
That said, I found the article interesting. Each year, I tote up the amounts my wife and I have in equities, in bond funds, and in CDs. Using the withdrawal suggestions offered by TRP for equities and bond funds separately, it's just math to come up with a blended withdrawal amount for everything but the CDs. What I use for the CD ladder is a 30-year amortization, assuming an average yield (conservative) of 3%.
Add it up, compare to what you might need when you start withdrawals, then call it a day. Revisit in a year or so.
That said, I found the article interesting. Each year, I tote up the amounts my wife and I have in equities, in bond funds, and in CDs. Using the withdrawal suggestions offered by TRP for equities and bond funds separately, it's just math to come up with a blended withdrawal amount for everything but the CDs. What I use for the CD ladder is a 30-year amortization, assuming an average yield (conservative) of 3%.
Add it up, compare to what you might need when you start withdrawals, then call it a day. Revisit in a year or so.
Re: New Withdrawal Strategies from T. Rowe Price
Electron thanks for the link to the article.
I read the article. Nothing major is new. We have already heard about valuations from Michael Kitces and others. The dynamic strategy is similar to what Jonathan Guyton has proposed. Still, good stuff. More to digest and decide which is more applicable to individual retirees.
One thing I didn't like is the "monthly" rebalancing TRP used in their study.
1210
I read the article. Nothing major is new. We have already heard about valuations from Michael Kitces and others. The dynamic strategy is similar to what Jonathan Guyton has proposed. Still, good stuff. More to digest and decide which is more applicable to individual retirees.
One thing I didn't like is the "monthly" rebalancing TRP used in their study.
1210
Re: New Withdrawal Strategies from T. Rowe Price
Thanks all for your comments.
I agree that these strategies are not entirely new. T. Rowe Price has discussed the Dynamic Strategy before, but I don't recall their mention of the Valuation Strategy that looks at initial bond yields and initial PE ratio.
Hexdump - I'm not sure you need to reduce your withdrawal based on the recent rise in PE ratio. The article suggests basing your initial withdrawal at retirement on the PE at that time. I believe that would define a dollar amount that could then be increased each year to keep up with inflation. Thinking about this, I wonder how many people incorrectly implement the typical 4% withdrawal strategy. I believe these strategies define a specific dollar amount at the time of retirement that can then be increased based in inflation. The percentage withdrawal does not necessarily apply to the current account value each subsequent year.
One interesting thing about the PE Valuation Strategy is that a retirement account value would typically be lower if PE ratios have declined, or higher if PE ratios have risen. That should justify the higher initial percentage withdrawal in the lower PE case since stocks have more potential to rise in value in the future. One risk, however, might be that the PE ratio could continue lower for some period of time.
I agree that these strategies are not entirely new. T. Rowe Price has discussed the Dynamic Strategy before, but I don't recall their mention of the Valuation Strategy that looks at initial bond yields and initial PE ratio.
Hexdump - I'm not sure you need to reduce your withdrawal based on the recent rise in PE ratio. The article suggests basing your initial withdrawal at retirement on the PE at that time. I believe that would define a dollar amount that could then be increased each year to keep up with inflation. Thinking about this, I wonder how many people incorrectly implement the typical 4% withdrawal strategy. I believe these strategies define a specific dollar amount at the time of retirement that can then be increased based in inflation. The percentage withdrawal does not necessarily apply to the current account value each subsequent year.
One interesting thing about the PE Valuation Strategy is that a retirement account value would typically be lower if PE ratios have declined, or higher if PE ratios have risen. That should justify the higher initial percentage withdrawal in the lower PE case since stocks have more potential to rise in value in the future. One risk, however, might be that the PE ratio could continue lower for some period of time.
Enjoying the Outdoors
Re: New Withdrawal Strategies from T. Rowe Price
RANT ON
Some of the most experienced folks on this forum seem to favor an approach to withdrawal strategy which basically amounts to "I'll wing it based on current circumstances".
Fact is they may very well be able to do so, because their deep understanding of investments & financials will provide proper foundation to their intuition. Or simply because they put enough money aside (and established very disciplined habits about their spend) to afford to not have to think much about the whole thing.
But normal human beings like many of us (at least me!) do NOT necessarily have the same kind of deep financial background. Plus we're learning day over day while reading this board that being methodical and taking emotions out in the investment phase really pays off on the long run, and any emotion-based short-term decision is asking for troubles. So... an advice to 'wing it' year over year during the withdrawal phase (hence letting emotions free to roam) doesn't add up for me. No way I'll make such a bet on the fact that my intuition will work.
Yes, sure, just using a totally dry & theoretical formula disconnected from the reality of the day may not be the way to go either, but I'd rather use something methodical (e.g. Guyton-Klinger, Hebeler) by default (after playing a good deal with backtesting to convince myself that the method is sound), and only use some common sense (intuition?) to keep an eye on things (yes, granted, things may get out of control in a way that the dry formula has not foreseen).
RANT OFF
All this being said, the article doesn't strike me as terribly interesting advice. The PE math seems like a somewhat dubious indicator of market up & down (if it were reliable, some people would derive a successful investment strategy based on it, and this didn't happen, right?). So I wouldn't be comfortable using it as a key input to my withdrawal strategy. As to linking portfolio losses and forgoing the inflation adjustment, this is mixing apples and oranges, and never made much sense to me.
Seems to me that this entire topic of withdrawal strategies would indeed deserve much more thorough research and field calibration over long periods of time... So that we can all become a bit more methodical about it.
Some of the most experienced folks on this forum seem to favor an approach to withdrawal strategy which basically amounts to "I'll wing it based on current circumstances".
Fact is they may very well be able to do so, because their deep understanding of investments & financials will provide proper foundation to their intuition. Or simply because they put enough money aside (and established very disciplined habits about their spend) to afford to not have to think much about the whole thing.
But normal human beings like many of us (at least me!) do NOT necessarily have the same kind of deep financial background. Plus we're learning day over day while reading this board that being methodical and taking emotions out in the investment phase really pays off on the long run, and any emotion-based short-term decision is asking for troubles. So... an advice to 'wing it' year over year during the withdrawal phase (hence letting emotions free to roam) doesn't add up for me. No way I'll make such a bet on the fact that my intuition will work.
Yes, sure, just using a totally dry & theoretical formula disconnected from the reality of the day may not be the way to go either, but I'd rather use something methodical (e.g. Guyton-Klinger, Hebeler) by default (after playing a good deal with backtesting to convince myself that the method is sound), and only use some common sense (intuition?) to keep an eye on things (yes, granted, things may get out of control in a way that the dry formula has not foreseen).
RANT OFF
All this being said, the article doesn't strike me as terribly interesting advice. The PE math seems like a somewhat dubious indicator of market up & down (if it were reliable, some people would derive a successful investment strategy based on it, and this didn't happen, right?). So I wouldn't be comfortable using it as a key input to my withdrawal strategy. As to linking portfolio losses and forgoing the inflation adjustment, this is mixing apples and oranges, and never made much sense to me.
Seems to me that this entire topic of withdrawal strategies would indeed deserve much more thorough research and field calibration over long periods of time... So that we can all become a bit more methodical about it.
Re: New Withdrawal Strategies from T. Rowe Price
The good folks at TRP write:
The study focused on the “feasible
initial withdrawal rate” (FIWR): The
withdrawal rate for the first year of
retirement—subsequently adjusted for
inflation—that was sustained for 30
years in 90% of the rolling historical
periods examined.
Some of us retirees think it's more "feasible" to attend to the likes of Otar, Zwecher, Merton. Instead of FIWR, the plan starts with SSF--a solid safe foundation.
Then again: as long as your plan, method, intuition or whatever serves you as you move along the retirement path, that should be good enough.
If your whatever doesn't work out, or works just 90% of the time, then....
Lev
The study focused on the “feasible
initial withdrawal rate” (FIWR): The
withdrawal rate for the first year of
retirement—subsequently adjusted for
inflation—that was sustained for 30
years in 90% of the rolling historical
periods examined.
Some of us retirees think it's more "feasible" to attend to the likes of Otar, Zwecher, Merton. Instead of FIWR, the plan starts with SSF--a solid safe foundation.
Then again: as long as your plan, method, intuition or whatever serves you as you move along the retirement path, that should be good enough.
If your whatever doesn't work out, or works just 90% of the time, then....
Lev
Re: New Withdrawal Strategies from T. Rowe Price
What's missing from all of this is any assessment of needs and risks.
If you NEED 4%, go and buy an indexed (real) SPIA and you will get 4% for the rest of your life with virtually no risk.
Taylor Larimore says he has annuitized his required income, and is now gifting from the remainder during his life.
I agree with Nisiprius, these strategies that are tuned by back testing on historical data are more or less meaningless as strategies for individuals to adopt going forward.
Keith
If you NEED 4%, go and buy an indexed (real) SPIA and you will get 4% for the rest of your life with virtually no risk.
Taylor Larimore says he has annuitized his required income, and is now gifting from the remainder during his life.
I agree with Nisiprius, these strategies that are tuned by back testing on historical data are more or less meaningless as strategies for individuals to adopt going forward.
Keith
Déjà Vu is not a prediction
Re: New Withdrawal Strategies from T. Rowe Price
The T. Rowe Price Valuation Strategy does appear to be fit to past history, but the idea of basing withdrawal percentage on bond and stock valuation may still have some merit.
Consider John Bogle's simple projection models for both stocks and bonds that is presented in the book Common Sense on Mutual Funds.
The models project the return of each asset class for the decade ahead. In the case of bonds, the model uses the yield on a broad index such as the aggregate bond index.
In the case of stocks, the model starts with the dividend yield on the S&P 500 index and adds typical earnings growth and possible change in PE ratio.
This makes me wonder if dividend yield would have been a better parameter for the T. Rowe Price strategy rather than PE ratio. With dividend yield, you at least have a known dividend stream available at the start of the period.
Consider John Bogle's simple projection models for both stocks and bonds that is presented in the book Common Sense on Mutual Funds.
The models project the return of each asset class for the decade ahead. In the case of bonds, the model uses the yield on a broad index such as the aggregate bond index.
In the case of stocks, the model starts with the dividend yield on the S&P 500 index and adds typical earnings growth and possible change in PE ratio.
This makes me wonder if dividend yield would have been a better parameter for the T. Rowe Price strategy rather than PE ratio. With dividend yield, you at least have a known dividend stream available at the start of the period.
Enjoying the Outdoors