Equity Glide Paths - some new thoughts.

[Chuckanut]

It looks like new assumptions are changing the recent view on equity glide paths.

With a 4% withdrawal rate, he finds that declining equity glidepaths in retirement support higher probabilities of success than fixed equity glidepaths, which in turn supports higher probabilities of success than rising equity glidepaths. This finding contrasts with an earlier 2007 article he wrote

To Rise or Not To Rise: Stock Allocation During Retirement

 

[REWahoo]

The more things change...

 

[2017ish]

I am a big fan of Wade Pfau's research reports--as well as his openness when dealing with new information that may undercut his conclusions. Thanks for posting this (I was actually on my way to do so!).

 

[samclem]

Yes, it's great that he modifies his position when new information arrives.

The whole thing seems like drawing conclusions on far-from-solid underlying data (MC or historical). It's probably not appropriate to try to refine AAs down to 2 decimal places based on how we are doing this. Micrometer > grease pencil > axe.

I do find myself drawn to modifying allocations slightly as valuations change. I'd never hop out of equities entirely, but lightening up as they get to stratospheric PE10s seems logical and prudent. I'll have to look into that. And, he likes shorter-term bonds.

 

[2017ish]

I do find myself drawn to modifying allocations slightly as valuations change. I'd never hop out of equities entirely, but lightening up as they get to stratospheric PE10s seems logical and prudent. I'll have to look into that. And, he likes shorter-term bonds.

It has been a long time since I've read his publications, but I recall Benjamin Graham's thoughts on this: a 50/50 stock bond portfolio, but flex to 25/75 or 75/25 in light of market valuation. PE10 would seem to be a good ruler.

Going back to Pfau (and ignoring for now the ongoing rising glidepath research), his finds are generally consistent with the large number of posters here who are leery of a 4% withdrawal rate--and with the attractiveness of being flexible on spending to avoid the potential drain of true "SWR." My biggest take from him to date resulted in postponing ER to ensure that we can handle extremely bad times in the first 10 years. Sequence of returns plus current market valuation of both debt and equity....

 

[haha]

Why would anyone think that his new idea is any better, or better grounded in data, than his old idea? It's like hem lengths or lapel and tie widths in fashion. Pure manipulation, to sell something that appears new.


Ha

 

[galeno]

Our annual spending is 2.9% of port. Add in our TER = 0.9% and we're up to 3.8% of port.


If our 60/40 port declines by 30%, our gross AWR goes to 5.0% which, IMHO, is sustainable.

 

[clifp]

Yes, it's great that he modifies his position when new information arrives.

The whole thing seems like drawing conclusions on far-from-solid underlying data (MC or historical). It's probably not appropriate to try to refine AAs down to 2 decimal places based on how we are doing this. Micrometer > grease pencil > axe.

I do find myself drawn to modifying allocations slightly as valuations change. I'd never hop out of equities entirely, but lightening up as they get to stratospheric PE10s seems logical and prudent. I'll have to look into that. And, he likes shorter-term bonds.

I like Wade, but was not a fan of his use of MC in his early paper. Nice to see he is giving historical simulations a try also. As he says historically low interest rates, and high stock valuations are pretty unprecedented in American history.

The problem that I and many other face, is that as much as stock valuations scare me, bonds offer return free risk which terrify me. I can comfortable survive at withdrawal rates above 2%, but bonds don't even offer real rates above that.

So for me personally, I am looking at other asset classes. Angel investing is interesting but very high risk, Real Estate and especially for me Real Estate lending seems promising. I am even keeping an eye on things like Peer2Peer lending, but having been an early adopter, I'm cautious.

Anybody know what the ticker symbol is for beaver cheese futures?

 

[audreyh1]

I kept waiting for the specifics!

If I had started the rising glide path in 2000, that would have been great.

But now I'm stuck with a straight allocation, and there doesn't seem to be much benefit to gliding one way or the other.

 

[growing_older]

Either these new breakthroughs come from mining the data deeply and therefor are dependent on the future economic activity strongly resembling the past economic activity, or else someone makes a lot of assumptions about the ranges and correlations of various factors and anything predicted is highly dependent on the assumptions they made. It's like arguing how many angels fit on the head of a pin.

There's very little reason that economic futures will correspond to either of these scenarios so closely that they can be fine tuned. At best these yield general rules. At worst no one does any sensitivity analysis to see what if the future doesn't resemble the assumptions how far off does one set of actions go vs another set.

I'm not at all impressed that a declining glide path (or a fixed glide path) will give a few percentage points better results. I would be very concerned that one or the other of these might yield much worse results in some future circumstances not considered. Overly optimizing the general likely cases is not much of a gain, but becoming vulnerable to outliers might be a big deal.

 

[nun]

Sorry, someone's modeling retirement portfolio's and not including the possibility that bond yields might go up........how long do we think interest rates can stay this low. Surely everyone now holding bond funds is expecting interest rates to go up at some point and for those funds to claw their way back into positive territory on then back of increased yields.

Also I wonder what this means for Wade's paper that predicted greater success by replacing bonds with SPIAs? I speak as someone who just bought an annuity. There are just too many assumptions to take any of these models as anything other than indicators and interesting thought experiments. I think they can give insights into strategies, but must not be though of as roadmaps to success. Personally, I've been on a rising glide path for a while as I realize that most of my retirement income will come from rent, SS and a pension/annuity. My logic is that if I have income needs covered from guaranteed sources I can take more risk.

 

[Chuckanut]

. My logic is that if I have income needs covered from guaranteed sources I can take more risk.

Sorry, that's not sophisticated enough. You need to run it through a bevy of calculators using statistical regression analysis along with future assumptions based on historical Monte Carlo algorithms modified by analytic recursive analysis of the mean differentiation integral of the standard deviation mode based on the last 78 years of market backward trailing averages.

 

[adrift]

Why am I not surprised that the article ends like this:

But, David has produced an interesting article which suggests more research is needed on this topic.

Following Mr. Pfau's career path seems like a full time job.

 

[foxfirev5]

This thread points out our desire to cover all possibilities regarding the future. However it's the next factor which has been overlooked that will get ya.
No wonder I'm in OMY mode, this will never end. However I'm starting to see the light and will take the plunge regardless.

 

[bingybear]

Sorry, someone's modeling retirement portfolio's and not including the possibility that bond yields might go up........how long do we think interest rates can stay this low. Surely everyone now holding bond funds is expecting interest rates to go up at some point and for those funds to claw their way back into positive territory on then back of increased yields.

Also I wonder what this means for Wade's paper that predicted greater success by replacing bonds with SPIAs? I speak as someone who just bought an annuity. There are just too many assumptions to take any of these models as anything other than indicators and interesting thought experiments. I think they can give insights into strategies, but must not be though of as roadmaps to success. Personally, I've been on a rising glide path for a while as I realize that most of my retirement income will come from rent, SS and a pension/annuity. My logic is that if I have income needs covered from guaranteed sources I can take more risk.

I'm new to this game and FIRE.. just RE's this week. I first had to figure out what this term "glide path" meant... easy once defined... but had to look it up online.
Nun, it seems like you bought a SPIA, but think rates will go up. I would think one would wait to get a SPIA if one expected rates to rise based on my very limited knowledge.
I saw one of Wade's papers referenced that really concluded a FPIA was better (inflation indexed ... think it was assumed at 4%). But I really did not catch where he considered the added cost by reducing other parts of the portfolio. Getting a 4% inflation adjustment will cost a lot for the same initial payout.
I think one of the real questions with interest rates rising are how fast they will rise... slowly as happened after the great depression or quickly from more recent history.

 

[nun]

Sorry, that's not sophisticated enough. You need to run it through a bevy of calculators using statistical regression analysis along with future assumptions based on historical Monte Carlo algorithms modified by analytic recursive analysis of the mean differentiation integral of the standard deviation mode based on the last 78 years of market backward trailing averages.

Unfortunately any level of sophistication in the modeling is hostage to the assumptions made. Change an assumption and you get a different answer. So my modeling assumes conservative return. I don't have the data to set up Monte Carlo simulations so I just go with a pessimistic average annual return. If my numbers work with that I am good to go.

 

[nun]

I'm new to this game and FIRE.. just RE's this week. I first had to figure out what this term "glide path" meant... easy once defined... but had to look it up online.
Nun, it seems like you bought a SPIA, but think rates will go up. I would think one would wait to get a SPIA if one expected rates to rise based on my very limited knowledge.
I saw one of Wade's papers referenced that really concluded a FPIA was better (inflation indexed ... think it was assumed at 4%). But I really did not catch where he considered the added cost by reducing other parts of the portfolio. Getting a 4% inflation adjustment will cost a lot for the same initial payout.
I think one of the real questions with interest rates rising are how fast they will rise... slowly as happened after the great depression or quickly from more recent history.

I think the 4% you are referring to is the "Safe Withdrawal Rate". Pfau's paper actually concluded that non-COLA'd SPIAs won out over bonds and even COLAed annuities because the COLAed annuities were over priced. I now wonder if Pfau included rising bond yields in his simulations.....I sort of assume that he had as who would think that bond yields will stay low for the next 30 years. That argues for reading the papers in detail rather than skipping to the graphs and tables.

My recent annuity buy is a little special. It's not a commercial annuity it's buying into a state, COLAed, DB pension plan. It gives me a 7% withdrawal at age 55 which is a full 3% more than the 4% I'd be advised to take from the money if I was managing it myself. Also to match the pension income here are the rates of return I'd have to get on the money if I kept it outside the pension plan from 55 to various ages.

Age Interest rate%
60 -20 (although the balance of the account goes to my heirs)
65 -2.5
70 +3.7
75 +6.4
80 +7.7
84 +8.3 (this is my actuarial life span)

I am not a fan of giving up control, but the longevity insurance and the basic numbers looked too good to pass up. If you think an SPIA is appropriate for you in retirement you should not market time interest rates. If you are worried about interest rates they buy a fixed term SPIA for say 5 years.

The choice of your asset allocation going into retirement is going to be mostly governed by your feelings about risk and how to generate income. I expect most people will set a middle course and end up with something close to 50/50 after reading a few articles and inspecting their gut. At the extremes will be the 100% equity folks and the Zvi Bodie folks that have mostly TIPS ladders. Right now, if we ignore the money I just put into the pension I'm at 80/20 AA and I can see me staying there for the foreseeable future, so probably constant glide path rather than rising.

 

[Chuckanut]

Unfortunately any level of sophistication in the modeling is hostage to the assumptions made. Change an assumption and you get a different answer. So my modeling assumes conservative return. I don't have the data to set up Monte Carlo simulations so I just go with a pessimistic average annual return. If my numbers work with that I am good to go.

Thanks. I agree. You made the point much better than my attempt at humour did.

 

[nun]

Thanks. I agree. You made the point much better than my attempt at humour did.

Modeling has a place, but it must be kept in perspective and understood that it is not a recipe for success. It might give you an indication of the ingredients, but there's still a million ways to combine and cook them. Experts, and informed amateurs, can take things way too seriously and forget the stupidity of what they do.

 

[Independent]

Nun, it seems like you bought a SPIA, but think rates will go up. I would think one would wait to get a SPIA if one expected rates to rise based on my very limited knowledge.
.

I'd say it depends on what I'm selling to buy the SPIA. Suppose I intend to sell long term bonds. In that case, if interest rates go up, the premium for the SPIA should go down, but the market value of my LT bonds should also go down.

If the SPIA and my LT bond portfolio are identically sensitive to interest rates, then the possibility of interest rate changes shouldn't be an issue.

 

[bingybear]

I think the 4% you are referring to is the "Safe Withdrawal Rate". Pfau's paper actually concluded that non-COLA'd SPIAs won out over bonds and even COLAed annuities because the COLAed annuities were over priced.

My recent annuity buy is a little special. It's not a commercial annuity it's buying into a state, COLAed, DB pension plan. It gives me a 7% withdrawal at age 55 which is a full 3% more than the 4% I'd be advised to take from the money if I was managing it myself.

I would say you're SPIA is special. I have seen a number of these pension buy ins that looked very good. The consumer ones really look pretty bad to me at these interest rates... as you noted... where do we expect rates to go.
I looked back a Wade's paper.. quote below. I guess his conclusion is to buy a real FIA for longevity, but then notes (since this does not likely fun inheritance/estate plans) that your FA needs to help you adjust your plan "ok.. my interpretation".

The second question asked was: “Is it in the retiree’s best interest to fully mitigate against the four inflation-adjusted
retirement income risks?” This answer demands the retirement planner to use both the science and the art of sustainable
inflation-adjusted retirement planning to resolve the income conundrum confronting retiring baby boomers. Observing Pfau’s
analysis in Exhibit H and I, one can observe that a retiree may potentially align his retirement income to successfully mitigate
against the four major risks of sustainable inflation-adjusted retirement income through investing 100% of his retirement
assets into a state-of-the-art real FIA. This is the science

After being sold an annuity in my 20's and having my portfolio churned, I'm real skeptical of the whole financial industry.

 

[nun]

After being sold an annuity in my 20's and having my portfolio churned, I'm real skeptical of the whole financial industry.

Skepticism is good armour. Buying a variable or deferred annuity at that age is probably not the best move, too much complication and expense. Being a skeptic you should look at what you are buying and do it for the right reasons, not because someone says its a good move. In the late 1980s I contributed to a TIAA fixed deferred annuity not knowing anything about investing and I was lucky that at least I bought from a respectable company and fell into one of the "good" deferred fixed annuities with low fees; it's averaged 6% annual return since then and kept pace with my other fixed income with a lot less risk, but I was very lucky.

I'd avoid annuities unless you are retiring and want to insure for longevity or produce income and then I'd only annuitize up to 20% of my portfolio. The one exception would be TIAA-CREF as their retirement products are basically all structured as annuities for historical reasons. But they don't really act like annuities at all except for TIAA-Traditional and that's still a very particular product.

If you can derive your retirement income from guaranteed sources like pensions, SS, SPIAs then I'd feel far better about a rising glide path. It would take more guts than I have to go on a rising glide path where I was relying on 4% withdrawals from a portfolio totally exposed to stock and bond market fluctuations. I'd probably follow Bogle's advice to keep the percentage of fixed income equal to my age.

 

[nun]

It looks like new assumptions are changing the recent view on equity glide paths.

To Rise or Not To Rise: Stock Allocation During Retirement

I'm one year into ER and just took a step up in my equity percentage from 50% to 70% because I used stable value and some total bond index to buy an annuity, not much gliding, more of a leap to more equities. If I can now get 1% greater return on my remaining portfolio (6% rather than 5%) because I have a higher percentage of equities my spreadsheet shows the annuity with higher equites percentage portfolio overtaking the 50/50 portfolio at age 64, that's 10 years from now. My gut says I'm not taking on much more risk as I have basically swapped cash and bonds for the annuity....in fact do I have less risk?

 

[bingybear]

Well... I moved it over to a less costly annuity later. never really looked at the return ... looks like just under 8%. But most of the time was after conversion to less expensive custodian.

ER @ 53... expected WR 1.5%. I can tighten up in bad years easily to 1%... but in good years my take it higher (... "what else am I going to do with it"). I expect I should be fine with a 50/50.. 60/40...and probable would work with just cash.... but don't think that would be good.

When we get to SS (assuming taken @ 70 and similar to today).. We should easily be able to live on that. But it is nice to have options.

 

[nun]

Well... I moved it over to a less costly annuity later. never really looked at the return ... looks like just under 8%. But most of the time was after conversion to less expensive custodian.

ER @ 53... expected WR 1.5%. I can tighten up in bad years easily to 1%... but in good years my take it higher (... "what else am I going to do with it"). I expect I should be fine with a 50/50.. 60/40...and probable would work with just cash.... but don't think that would be good.

When we get to SS (assuming taken @ 70 and similar to today).. We should easily be able to live on that. But it is nice to have options.

That's close to my scenario. The pension buy in was a convenient way to produce income and get some longevity insurance. I was dubious about it as I have enough to fund my retirement without any issues, but if it also allows me to have a higher equity allocation and offers the chance to have a higher portfolio value by age 64 and the pension income too its sort of a slam dunk.....of course the issue with rising glide paths is also rising risk so the change of higher portfolio values must be tempered by increasing possibilities of failure.