Michael Kitces investment strategy -- opinions?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Enginerd
Posts: 10
Joined: Mon Feb 24, 2014 9:46 pm

Michael Kitces investment strategy -- opinions?

Post by Enginerd » Sun Oct 06, 2019 8:38 am

https://www.investmentnews.com/article/ ... trategy-is

Has anyone read this article? It's an interesting perspective. Interested in hearing your opinion.

User avatar
305pelusa
Posts: 1002
Joined: Fri Nov 16, 2018 10:20 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 305pelusa » Sun Oct 06, 2019 8:45 am

Enginerd wrote:
Sun Oct 06, 2019 8:38 am
https://www.investmentnews.com/article/ ... trategy-is

Has anyone read this article? It's an interesting perspective. Interested in hearing your opinion.
A rising equity glide path is the way to diversify your equity exposure temporally. It's very logical in my opinion.

User avatar
arcticpineapplecorp.
Posts: 4515
Joined: Tue Mar 06, 2012 9:22 pm

Re: Michael Kitces investment strategy -- opinions?

Post by arcticpineapplecorp. » Sun Oct 06, 2019 8:46 am

this is sorta old news. Kitces and Pfau's paper was published in 2013 I believe (source: https://papers.ssrn.com/sol3/papers.cfm ... id=2324930) and the bogleheads have discussed this. I typed in three different entries into the bogleheads search bar in the top right corner of the bogleheads site (kitces and pfau, rising equity glide path and bond tent) and found the following entries:

https://www.google.com/search?sitesearc ... s+and+pfau
https://www.google.com/search?sitesearc ... glide+path
https://www.google.com/search?sitesearc ... =bond+tent

I would recommend reading up on all those thoughts and responses and if you want to continue the discussion, feel free of course.

let the search bar be your friend.
"May you live as long as you want and never want as long as you live" -- Irish Blessing | "Invest we must" -- Jack Bogle

3funder
Posts: 1082
Joined: Sun Oct 15, 2017 9:35 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 3funder » Sun Oct 06, 2019 8:48 am

Kitces and Pfau are bright, thoughtful individuals. That said, this strategy isn't my cup of tea. I'm going with the following age-based glide path:

35: 85/15
40: 80/20
45: 75/25
50: 70/30
55: 65:35
60: 60/40
65: 55/45
70: 50/50

Higher bond allocation in 403(b) as I age to offset high stock allocation in Roth IRA and taxable account, neither of which I plan to touch unless I have to.
Last edited by 3funder on Sun Oct 06, 2019 9:01 am, edited 5 times in total.

User avatar
midareff
Posts: 6460
Joined: Mon Nov 29, 2010 10:43 am
Location: Biscayne Bay, South Florida

Re: Michael Kitces investment strategy -- opinions?

Post by midareff » Sun Oct 06, 2019 8:50 am

The interesting and implied portion of this strategy is that you would be withdrawing from bonds for the initial years of your retirement when the sequence of return risk is highest and the asset allocation flows upwards for equities from there.

workerbeeengineer
Posts: 154
Joined: Thu Jun 02, 2016 12:22 am

Re: Michael Kitces investment strategy -- opinions?

Post by workerbeeengineer » Sun Oct 06, 2019 8:53 am

I agree that now is a very challenging time to retire, given high valuations and low interest rates (which might possibly go negative in a few years). I'm a few years out, and am sitting at an AA of 50/50 stocks/bonds. I read Pfau's book, as well as McClung’s (I basically agree with his harvesting approach, but probably won't follow it rigorously). Will draw from the bonds and leave the stocks to (hopefully) appreciate. So I guess I'm good with Kitces recommending a rising glide path for stocks.
Last edited by workerbeeengineer on Mon Oct 07, 2019 10:09 pm, edited 1 time in total.

JBTX
Posts: 5536
Joined: Wed Jul 26, 2017 12:46 pm

Re: Michael Kitces investment strategy -- opinions?

Post by JBTX » Sun Oct 06, 2019 9:04 am

I am somewhat skeptical of the notion of rising equity glide paths. It is based upon historical data, which assumes the future will always fall into some sub set of the past. It also just measures the percent failure rate, vs quantifying how badly you failed.

Look at it this way. You are 70-80 years old, and have some amount of nestegg left over. Let's say your portfolio just went from $1 million to $600k in a crash. Perhaps you were pulling out $50k per year. You just went from 5% withdrawal rate to about 8%. The theory is that the markets will recover, because they always have, so don't worry, be happy! My assertion is that is not the way humans work. The future doesn't always follow the past, and if the market doesn't recover in due time (Japan) you may fail, and fail badly.

Using a failure rate isn't a perfect way to measure things. It may be better the slightly fail 5% of the time than spectacularly fail 2% of the time. The truth is you can always adjust your withdrawal rate to avoid that failure. With a higher stock allocation at an older age you will have greater portfolio volatility.

If I'm looking at this wrong I'd like somebody to point out the errors in my thinking. I don't argue with the historical math. My argument is it fails to take into account risk aversion properly.

User avatar
305pelusa
Posts: 1002
Joined: Fri Nov 16, 2018 10:20 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 305pelusa » Sun Oct 06, 2019 9:43 am

JBTX wrote:
Sun Oct 06, 2019 9:04 am
I am somewhat skeptical of the notion of rising equity glide paths. It is based upon historical data, which assumes the future will always fall into some sub set of the past.
Actually, a rising equity path strategy makes sense from a mathematical and logical perspective too. I have no clue as to its origin (maybe it did come by looking at past data, no clue), but it has serious theoretical backing.
JBTX wrote:
Sun Oct 06, 2019 9:04 am
Let's say your portfolio just went from $1 million to $600k in a crash. Perhaps you were pulling out $50k per year. You just went from 5% withdrawal rate to about 8%. The theory is that the markets will recover, because they always have, so don't worry, be happy!
The theory has absolutely nothing to do with the "market recovering" or some other "double-down" or "value" perspective (although I can see why it looks like that).
The rising equity glide path should be used by retirees with savings and social security. As you age and deplete your portfolio, your social security (an inflation-linked, perfectly safe stream of income) payments become a larger portion of your entire "portfolio". So as the portfolio depletes, it is as though more of your wealth is in safer assets.

Increasing your equity allocation in your slowly shrinking portfolio maintains your desired risk-return characteristics.
If you don't have social security, you wouldn't need a rising equity glide path.
JBTX wrote:
Sun Oct 06, 2019 9:04 am
My argument is it fails to take into account risk aversion properly.
It in fact is necessary to keep account for risk aversion properly. Imagine two retirees with identical risk aversion and desire for risk:
A: Has 500k in savings and SS benefits that might be worth maybe 500k in present day. He invests his portfolio 50/50 (for a total of 250k in stocks and 750k in safe assets).
B: Has 1MM in savings and no social security. He invests 25% in stocks, 75% in bonds (250k/750k)

After a decade, they deplete 30% of their assets. So B now has 175k in stocks and 525k in bonds (for his overall, desired 25/75 split). How should A invest to be logically identical and consistent with B, who has the same attitudes towards risk?
A should put 175kk in stocks and leave 525k in safe assets as well (his SS as well as 25k in bonds). Look at A's portfolio though: It is 87.5% stocks!

That sounds crazy! A must think the market must recover right? Not at all; all A has done is stay consistent with his risk aversion and appetite for risk just like B did. A's portfolio has approximately the same risk-reward characteristics of B.

If A did not increase his equity allocation, he would become more conservative than B, which must be incorrect because we assumed from the beginning they are identical otherwise so they should have the same asset allocation too.

dbr
Posts: 30798
Joined: Sun Mar 04, 2007 9:50 am

Re: Michael Kitces investment strategy -- opinions?

Post by dbr » Sun Oct 06, 2019 9:53 am

I think Kitces and Pfau and others who try to elucidate the mysteries of retirement financial planning are worth reading for the education.

I am far less convinced that there is a single strategy to be found here that is so helpful and so certain as to be singled out as a "must do." I think in many cases differences that may be real are also relatively small. It is very difficult to place bounds of uncertainty on projections that are inherently based on very uncertain assumptions.

One of the lessons about retirement income planning that always sticks in my mind is that you cannot asset allocate your way out of having the bad luck to spend too much money for a retirement started at the wrong time. It is also a lesson that the worst case is also very rare.

Another lesson is that life itself is far more variable than investments are variable, meaning common sense flexibility is more important than detailed engineering of investment strategy.

User avatar
Wildebeest
Posts: 1203
Joined: Fri Dec 27, 2013 2:36 pm

Re: Michael Kitces investment strategy -- opinions?

Post by Wildebeest » Sun Oct 06, 2019 9:57 am

My two cents: Fluff piece. To get you to use advisors.

Case in point: the last paragraph "A financial plan is going to be wrong the next day after you deliver that plan," Ms. O'Leary said. Retirees need regular checkups and forecasts that account for changing market assumptions. "The conversation really needs to happen on an ongoing basis."


I like Kitces and I like Pfau.

The Pfau quote: Retirement income expert Wade Pfau agreed that the strategy could be especially helpful for people retiring now, which could be one of the most difficult times to retire in history.

"We are in uncharted waters," Mr. Pfau said. "We've never had the high valuations and the low interest rates at the same time."

in my mind that would make the case for a liability matching portfolio more so than the rising equity glide path.
The Golden Rule: One should treat others as one would like others to treat oneself.

RadAudit
Posts: 3612
Joined: Mon May 26, 2008 10:20 am
Location: Second star on the right and straight on 'til morning

Re: Michael Kitces investment strategy -- opinions?

Post by RadAudit » Sun Oct 06, 2019 9:59 am

JBTX wrote:
Sun Oct 06, 2019 9:04 am
If I'm looking at this wrong I'd like somebody to point out the errors in my thinking.
Please do the same for me. But, I doubt that you are wrong.

I've added a small wrinkle to my withdrawal approach. Hate to call it a strategy. IIRC, McClung suggested a similar approach. (Apologies in advance to McClung if I got it wrong.) At the start of retirement, I chose an asset allocation. I withdraw from the bond side (less volatility) and rebalance IAW my IPS. Hopefully, that maintains a chosen risk level.

Additionally, I chose not to rebalance if doing so would drive the stock side below a level based on the amount I started with. In that case, I just continue to withdraw from the bond side until stocks recover sufficiently to permit rebalancing. If stocks don't recover quickly and bonds go to zero, then I'll take from the stock side. In my case, my bond side should allow for a number of years of withdrawals.

This results in that case with a rising equity portion in the portfolio and not a steady AA. Of course, reducing withdrawals is always an option.

And, if I get to the point where I can't calculate this stuff with a spreadsheet, I'll go to a single life-strategy fund and hope.

YMMV. Best of luck.

Apologies to 305pelusa. Read your post just before I hit the submit button. Now, I'll have to go back and re-think withdrawals. Ought to take about six months.
Last edited by RadAudit on Sun Oct 06, 2019 10:32 am, edited 2 times in total.
FI is the best revenge. LBYM. Invest the rest. Stay the course. - PS: The cavalry isn't coming, kids. You are on your own.

randomguy
Posts: 8402
Joined: Wed Sep 17, 2014 9:00 am

Re: Michael Kitces investment strategy -- opinions?

Post by randomguy » Sun Oct 06, 2019 10:16 am

305pelusa wrote:
Sun Oct 06, 2019 8:45 am
Enginerd wrote:
Sun Oct 06, 2019 8:38 am
https://www.investmentnews.com/article/ ... trategy-is

Has anyone read this article? It's an interesting perspective. Interested in hearing your opinion.
A rising equity glide path is the way to diversify your equity exposure temporally. It's very logical in my opinion.
Here is a chart from one of his papers
Image

notice how little difference there is between justing holding say 40/60 or 50/50 and doing this glide path? Do you think the model is accurate enough where 1% differences are more than just noise?

Now things like ending balances and fluctuations along the way also matter but I have seen nothing to suggest that going with a rising equity path or bond tent or any of these other schemes does a better job. They do a slightly different job. You have to decide if the complexity is worth it.

User avatar
nisiprius
Advisory Board
Posts: 39444
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: Michael Kitces investment strategy -- opinions?

Post by nisiprius » Sun Oct 06, 2019 10:19 am

Kitces is a thoughtful, knowledgeable expert. I don't happen to know offhand of any actual conflicts of interest (he isn't being paid by any company selling rising-glide-path target-date retirement funds, for example).

It's not well described as a "rising equity glide path" and Kitces himself now prefers to call it a "bond tent." The whole concept is complicated and the original paper was a fairly cursory view whose main conclusion was that it deserved further exploration. The essential idea is not "80-year-olds need more stocks," it is that "65-year-olds need less." Compared to traditional glide paths, the idea is which is a depression or notch in stock allocation, a lower stock allocation in the years close to retirement age. Or, if you invert it and look at bonds, a rise or peak in bonds, which can be called a "bond tent" to suggest a sheltering effect. The fact that it rises afterwards is a side-effect of being lower at about retirement age. As the original paper says, "in many scenarios, the optimal rising equity glidepaths still finish with little more in equities than the 60/40 portfolios that are commonly used by many planners anyway."

I find detailed comparisons of small differences in glide paths to be unconvincing. An Achilles heel of all of them is that they are extremely sensitive to assumed tolerable failure rate. Say that you can tolerate 10% and the studies will all show improvements with higher stock allocations. Say that you can only tolerate 2.8%--the chance of rolling a "2" on a pair of dice--and they allow astonishingly level results, almost independent of stock-bond allocation except at the extremes.

These studies do not take into account the effect of "wealth shocks" on health and wellbeing in retirees. Again, the original paper says
Notably, the clear caveat and concern of this approach is that it may also create concerns for seniors in their later years, who may not be comfortable from a risk tolerance perspective handling the greater equity exposures implied by this approach (even if the results would technically be optimal from the mathematical and markets perspective).
It is unusual for these studies even to mention this. The basic assumption is that until the actual day they run out of money,[/i] a retiree is unaffected by fluctuations, and feels just as happy with $700,000 as they felt with $1,000,000. In contrast, a small but growing body of studies suggests that wealth shocks actually have a direct effect on health and life expectancy.
Last edited by nisiprius on Sun Oct 06, 2019 11:35 am, edited 1 time in total.
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.

User avatar
305pelusa
Posts: 1002
Joined: Fri Nov 16, 2018 10:20 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 305pelusa » Sun Oct 06, 2019 10:35 am

randomguy wrote:
Sun Oct 06, 2019 10:16 am
305pelusa wrote:
Sun Oct 06, 2019 8:45 am
Enginerd wrote:
Sun Oct 06, 2019 8:38 am
https://www.investmentnews.com/article/ ... trategy-is

Has anyone read this article? It's an interesting perspective. Interested in hearing your opinion.
A rising equity glide path is the way to diversify your equity exposure temporally. It's very logical in my opinion.
Here is a chart from one of his papers
Image

notice how little difference there is between justing holding say 40/60 or 50/50 and doing this glide path? Do you think the model is accurate enough where 1% differences are more than just noise?

Now things like ending balances and fluctuations along the way also matter but I have seen nothing to suggest that going with a rising equity path or bond tent or any of these other schemes does a better job. They do a slightly different job. You have to decide if the complexity is worth it.
Fascinating. So if the strategy was very successful historically, you have posters who are skeptical because they think it only works if you assume the future will be like the past (i.e. data mining).

If it wasn't that successful historically, then other posters will ask "why bother? It didn't change things much."

My 2 cents: I don't know much about the historical performance, benefits, etc of the strategy. Nor do I care. As a BH, I want to keep my desired risk-return consistent with my portfolio via rebalancing. If SS is a large part of your wealth, it means overcompensating on the portfolio side by possibly being more aggressive thanks to the fungibility of money. This isn't even complicate to do; since you're in withdrawal, it just means withdrawing more from bonds than stocks accordingly.

If you think that's too complicated, not worth your time, data mined, etc. then do whatever else you think is correct. Just know that if you don't do the above (and follow a constant % allocation), then those without SS will end up being more aggressive than they intended and those with SS more conservative than they intended. I merely recommend you follow the above so you are as aggressive as you intended.

User avatar
convert949
Posts: 378
Joined: Thu May 07, 2009 8:33 am
Location: Fort Myers, FL

Re: Michael Kitces investment strategy -- opinions?

Post by convert949 » Sun Oct 06, 2019 10:40 am

Wildebeest wrote:
Sun Oct 06, 2019 9:57 am

I like Kitces and I like Pfau.

The Pfau quote: Retirement income expert Wade Pfau agreed that the strategy could be especially helpful for people retiring now, which could be one of the most difficult times to retire in history.

"We are in uncharted waters," Mr. Pfau said. "We've never had the high valuations and the low interest rates at the same time."

in my mind that would make the case for a liability matching portfolio more so than the rising equity glide path.
+1... My wrinkle on that is keeping enough in safe money with duration matched to expected need to get through the "high expense" portion of retirement (travel, etc. while you are still able) and count on SS, annuities, Pensions (if you have one, I do not) for your later years (basic expenses)... Equities become less critical and can be used for discretionary purchases in good years and legacy... I find it easier to manage than a true LMP.

User avatar
2pedals
Posts: 929
Joined: Wed Dec 31, 2014 12:31 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 2pedals » Sun Oct 06, 2019 11:25 am

I like the idea and I think it is good "general advice". However there are many things for your situation where this may not pass the "smell test" though. Do you have the ability, willingness and/or need for more equity risk and as you get older?
Last edited by 2pedals on Sun Oct 06, 2019 11:26 am, edited 1 time in total.

randomguy
Posts: 8402
Joined: Wed Sep 17, 2014 9:00 am

Re: Michael Kitces investment strategy -- opinions?

Post by randomguy » Sun Oct 06, 2019 11:26 am

305pelusa wrote:
Sun Oct 06, 2019 10:35 am

Fascinating. So if the strategy was very successful historically, you have posters who are skeptical because they think it only works if you assume the future will be like the past (i.e. data mining).

If it wasn't that successful historically, then other posters will ask "why bother? It didn't change things much."

My 2 cents: I don't know much about the historical performance, benefits, etc of the strategy. Nor do I care. As a BH, I want to keep my desired risk-return consistent with my portfolio via rebalancing. If SS is a large part of your wealth, it means overcompensating on the portfolio side by possibly being more aggressive thanks to the fungibility of money. This isn't even complicate to do; since you're in withdrawal, it just means withdrawing more from bonds than stocks accordingly.

If you think that's too complicated, not worth your time, data mined, etc. then do whatever else you think is correct. Just know that if you don't do the above (and follow a constant % allocation), then those without SS will end up being more aggressive than they intended and those with SS more conservative than they intended. I merely recommend you follow the above so you are as aggressive as you intended.
I am pretty sure that isn't historical data but it is instead monte carlo simulations seed from historical data. I will leave it up to you to decide if athe model is accurate enough to distinguish between 94.6% and 95.1% and even if it is accurate enough, do you care?

Now as far as risk, lets look at your example
A: Has 500k in savings and SS benefits that might be worth maybe 500k in present day. He invests his portfolio 50/50 (for a total of 250k in stocks and 750k in safe assets).
B: Has 1MM in savings and no social security. He invests 25% in stocks, 75% in bonds (250k/750k)

After a decade, they deplete 30% of their assets. So B now has 175k in stocks and 525k in bonds (for his overall, desired 25/75 split). How should A invest to be logically identical and consistent with B, who has the same attitudes towards risk?
A should put 175kk in stocks and leave 525k in safe assets as well (his SS as well as 25k in bonds). Look at A's portfolio though: It is 87.5% stocks!
A's SS isn't worth 500k at 75. It is worth say 350k. So A has 175k in stocks and 175k in bonds and has a 50% stocks in his portfolio. So no A doesn't need a rising equity glide path to maintain his risk exposure. Obviously this just works out because of the numbers. For most people they would need to get more conservative as the portfolio grows and SS shrinks.

If I wanted to target risk-return, I would do it directly and not through a rising equity glide path.

User avatar
JoMoney
Posts: 7786
Joined: Tue Jul 23, 2013 5:31 am

Re: Michael Kitces investment strategy -- opinions?

Post by JoMoney » Sun Oct 06, 2019 11:35 am

The article was behind a registration pay-wall that didn't seem to be worth finding a way around.
From what others have said, the gist seems to be the frequently discussed topic of a rising equity allocation in retirement.
My consideration would be, why not use a SPIA for the initial "bond" allocation at retirement? It should have the same expectation as a portfolio of bond withdrawals amortized over ones life expectancy, along with the guarantee that payments won't stop if you live longer.
After that liability matching floor is set by the SPIA the remainder can be in stocks to support a variable withdrawal rate. If stocks do well, the relative "allocation" will drift higher and support a higher variable withdrawal.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

dbr
Posts: 30798
Joined: Sun Mar 04, 2007 9:50 am

Re: Michael Kitces investment strategy -- opinions?

Post by dbr » Sun Oct 06, 2019 11:57 am

JoMoney wrote:
Sun Oct 06, 2019 11:35 am
The article was behind a registration pay-wall that didn't seem to be worth finding a way around.
From what others have said, the gist seems to be the frequently discussed topic of a rising equity allocation in retirement.
My consideration would be, why not use a SPIA for the initial "bond" allocation at retirement? It should have the same expectation as a portfolio of bond withdrawals amortized over ones life expectancy, along with the guarantee that payments won't stop if you live longer.
After that liability matching floor is set by the SPIA the remainder can be in stocks to support a variable withdrawal rate. If stocks do well, the relative "allocation" will drift higher and support a higher variable withdrawal.
I think one paper that Pfau or Kitces wrote suggested that a possible benefit is gained by combining an annuity with an all stock investment portfolio.

But I think one needs to be aware that these kinds of things are studies of how things might work if done one way or the other way. I think there is still a ways to go to arrive at a clear cut system for retirement investing that is known to be "the" answer. Again it has to do with how big the difference is and what kinds of assumptions go into the model.

User avatar
nedsaid
Posts: 12772
Joined: Fri Nov 23, 2012 12:33 pm

Re: Michael Kitces investment strategy -- opinions?

Post by nedsaid » Sun Oct 06, 2019 11:59 am

JoMoney wrote:
Sun Oct 06, 2019 11:35 am
The article was behind a registration pay-wall that didn't seem to be worth finding a way around.
From what others have said, the gist seems to be the frequently discussed topic of a rising equity allocation in retirement.
My consideration would be, why not use a SPIA for the initial "bond" allocation at retirement? It should have the same expectation as a portfolio of bond withdrawals amortized over ones life expectancy, along with the guarantee that payments won't stop if you live longer.
After that liability matching floor is set by the SPIA the remainder can be in stocks to support a variable withdrawal rate. If stocks do well, the relative "allocation" will drift higher and support a higher variable withdrawal.
Wow. Are you psychic or what? I was going to post that if we face uncertainties with high valuations and low interest rates, why not take a look at Single Premium Immediate Annuities? Great minds think alike.
A fool and his money are good for business.

dbr
Posts: 30798
Joined: Sun Mar 04, 2007 9:50 am

Re: Michael Kitces investment strategy -- opinions?

Post by dbr » Sun Oct 06, 2019 12:08 pm

Don't forget that when interest rates are down annuity payouts that can be obtained are down as well.

I don't think there is a magic carpet that rides above the bad luck of when you invest. But . . . it remains to be seen that our luck is really all that bad right now. It will take decades to find out for sure. Certainly current investors have benefited from the last ten years run-up in stocks, and that counts too.

JBTX
Posts: 5536
Joined: Wed Jul 26, 2017 12:46 pm

Re: Michael Kitces investment strategy -- opinions?

Post by JBTX » Sun Oct 06, 2019 12:10 pm

305pelusa wrote:
Sun Oct 06, 2019 9:43 am
JBTX wrote:
Sun Oct 06, 2019 9:04 am
I am somewhat skeptical of the notion of rising equity glide paths. It is based upon historical data, which assumes the future will always fall into some sub set of the past.
Actually, a rising equity path strategy makes sense from a mathematical and logical perspective too. I have no clue as to its origin (maybe it did come by looking at past data, no clue), but it has serious theoretical backing.
JBTX wrote:
Sun Oct 06, 2019 9:04 am
Let's say your portfolio just went from $1 million to $600k in a crash. Perhaps you were pulling out $50k per year. You just went from 5% withdrawal rate to about 8%. The theory is that the markets will recover, because they always have, so don't worry, be happy!
The theory has absolutely nothing to do with the "market recovering" or some other "double-down" or "value" perspective (although I can see why it looks like that).
The rising equity glide path should be used by retirees with savings and social security. As you age and deplete your portfolio, your social security (an inflation-linked, perfectly safe stream of income) payments become a larger portion of your entire "portfolio". So as the portfolio depletes, it is as though more of your wealth is in safer assets.

Increasing your equity allocation in your slowly shrinking portfolio maintains your desired risk-return characteristics.
If you don't have social security, you wouldn't need a rising equity glide path.
JBTX wrote:
Sun Oct 06, 2019 9:04 am
My argument is it fails to take into account risk aversion properly.
It in fact is necessary to keep account for risk aversion properly. Imagine two retirees with identical risk aversion and desire for risk:
A: Has 500k in savings and SS benefits that might be worth maybe 500k in present day. He invests his portfolio 50/50 (for a total of 250k in stocks and 750k in safe assets).
B: Has 1MM in savings and no social security. He invests 25% in stocks, 75% in bonds (250k/750k)

After a decade, they deplete 30% of their assets. So B now has 175k in stocks and 525k in bonds (for his overall, desired 25/75 split). How should A invest to be logically identical and consistent with B, who has the same attitudes towards risk?
A should put 175kk in stocks and leave 525k in safe assets as well (his SS as well as 25k in bonds). Look at A's portfolio though: It is 87.5% stocks!

That sounds crazy! A must think the market must recover right? Not at all; all A has done is stay consistent with his risk aversion and appetite for risk just like B did. A's portfolio has approximately the same risk-reward characteristics of B.

If A did not increase his equity allocation, he would become more conservative than B, which must be incorrect because we assumed from the beginning they are identical otherwise so they should have the same asset allocation too.
I'll give you the desirability of an increased bond allocation during a pre social security retirement phase. I think somebody called it liability matching. As to when you are drawing social security - when we talk about allocations, we often assume and ignore other income streams. It does make sense however that those income streams affect your portfolio composition.

However, if we are assuming long expense needs (not wants) exceed your social security income stream, then that incremental portion is what we are focusing on.

These studies are always backward looking. Looking backwards, over a prior 30-40 year period and what would have ended up with the most optimal outcome. But that is not how a risk averse investor invests. They invest based upon where they are now, and they continually reevaluate that going forward.

A backward approach may show that having a heavy stock allocation in old age is usually survivable. To the extent stocks may tank, chances are in prior years they rose more than average. It is an indirect way of market timing. So it might not seem unreasonable to have a high stock allocation or an unusually high and temporary withdrawal rate looking backward, when the probabilities are known, but at that point in time, not knowing the future, based upon uncertain probabilities going forward, a high stock allocation may not make sense for a risk averse investor.

To put it more simply, I can't imagine it ever makes sense to be 85 or 90 years old and have 70% stocks, if that is money you need. Now in terms of passing wealth to heirs, or if you have way more than you need, those are different discussions.

Also as to social security, it is an interesting wrinkle into the discussion, but these prior studies were just about withdrawal rates and glide paths. Social security was not part of the discussion, as far as I know.

dkturner
Posts: 1510
Joined: Sun Feb 25, 2007 7:58 pm

Re: Michael Kitces investment strategy -- opinions?

Post by dkturner » Sun Oct 06, 2019 12:15 pm

nedsaid wrote:
Sun Oct 06, 2019 11:59 am
JoMoney wrote:
Sun Oct 06, 2019 11:35 am
The article was behind a registration pay-wall that didn't seem to be worth finding a way around.
From what others have said, the gist seems to be the frequently discussed topic of a rising equity allocation in retirement.
My consideration would be, why not use a SPIA for the initial "bond" allocation at retirement? It should have the same expectation as a portfolio of bond withdrawals amortized over ones life expectancy, along with the guarantee that payments won't stop if you live longer.
After that liability matching floor is set by the SPIA the remainder can be in stocks to support a variable withdrawal rate. If stocks do well, the relative "allocation" will drift higher and support a higher variable withdrawal.
Wow. Are you psychic or what? I was going to post that if we face uncertainties with high valuations and low interest rates, why not take a look at Single Premium Immediate Annuities? Great minds think alike.
If you buy a single premium immediate annuity when interest rates are historically low aren’t you simply locking in a low returning investment?

User avatar
305pelusa
Posts: 1002
Joined: Fri Nov 16, 2018 10:20 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 305pelusa » Sun Oct 06, 2019 12:18 pm

randomguy wrote:
Sun Oct 06, 2019 11:26 am
305pelusa wrote:
Sun Oct 06, 2019 10:35 am

Fascinating. So if the strategy was very successful historically, you have posters who are skeptical because they think it only works if you assume the future will be like the past (i.e. data mining).

If it wasn't that successful historically, then other posters will ask "why bother? It didn't change things much."

My 2 cents: I don't know much about the historical performance, benefits, etc of the strategy. Nor do I care. As a BH, I want to keep my desired risk-return consistent with my portfolio via rebalancing. If SS is a large part of your wealth, it means overcompensating on the portfolio side by possibly being more aggressive thanks to the fungibility of money. This isn't even complicate to do; since you're in withdrawal, it just means withdrawing more from bonds than stocks accordingly.

If you think that's too complicated, not worth your time, data mined, etc. then do whatever else you think is correct. Just know that if you don't do the above (and follow a constant % allocation), then those without SS will end up being more aggressive than they intended and those with SS more conservative than they intended. I merely recommend you follow the above so you are as aggressive as you intended.
I am pretty sure that isn't historical data but it is instead monte carlo simulations seed from historical data. I will leave it up to you to decide if athe model is accurate enough to distinguish between 94.6% and 95.1% and even if it is accurate enough, do you care?

Now as far as risk, lets look at your example
A: Has 500k in savings and SS benefits that might be worth maybe 500k in present day. He invests his portfolio 50/50 (for a total of 250k in stocks and 750k in safe assets).
B: Has 1MM in savings and no social security. He invests 25% in stocks, 75% in bonds (250k/750k)

After a decade, they deplete 30% of their assets. So B now has 175k in stocks and 525k in bonds (for his overall, desired 25/75 split). How should A invest to be logically identical and consistent with B, who has the same attitudes towards risk?
A should put 175kk in stocks and leave 525k in safe assets as well (his SS as well as 25k in bonds). Look at A's portfolio though: It is 87.5% stocks!
A's SS isn't worth 500k at 75. It is worth say 350k. So A has 175k in stocks and 175k in bonds and has a 50% stocks in his portfolio. So no A doesn't need a rising equity glide path to maintain his risk exposure. Obviously this just works out because of the numbers. For most people they would need to get more conservative as the portfolio grows and SS shrinks.

If I wanted to target risk-return, I would do it directly and not through a rising equity glide path.
For MOST investors in this country, portfolio savings decrease much faster than SS. Most people heavily rely on their savings at the beginning of retirement and might rely almost entirely on SS by the end of retirement. This people need a rising equity path to target a consistent risk-reward.

For investors whose portfolio actually GROWS through retirement (and are not interested in the max SWR), which is in your example, then you'd do the opposite, I agree. I don't think this is the majority of Americans but if it is, then I agree the majority should have a falling equity glide path.

The only way you'd get a constant risk-reward with a constant % allocation would be if, as you said, the numbers happen to work out perfectly. They probably won't for most.

User avatar
nedsaid
Posts: 12772
Joined: Fri Nov 23, 2012 12:33 pm

Re: Michael Kitces investment strategy -- opinions?

Post by nedsaid » Sun Oct 06, 2019 12:23 pm

dkturner wrote:
Sun Oct 06, 2019 12:15 pm
nedsaid wrote:
Sun Oct 06, 2019 11:59 am
JoMoney wrote:
Sun Oct 06, 2019 11:35 am
The article was behind a registration pay-wall that didn't seem to be worth finding a way around.
From what others have said, the gist seems to be the frequently discussed topic of a rising equity allocation in retirement.
My consideration would be, why not use a SPIA for the initial "bond" allocation at retirement? It should have the same expectation as a portfolio of bond withdrawals amortized over ones life expectancy, along with the guarantee that payments won't stop if you live longer.
After that liability matching floor is set by the SPIA the remainder can be in stocks to support a variable withdrawal rate. If stocks do well, the relative "allocation" will drift higher and support a higher variable withdrawal.
Wow. Are you psychic or what? I was going to post that if we face uncertainties with high valuations and low interest rates, why not take a look at Single Premium Immediate Annuities? Great minds think alike.
If you buy a single premium immediate annuity when interest rates are historically low aren’t you simply locking in a low returning investment?
Yes, but you get regular payments and you get a 5% to 7% withdrawal rate compared to a 3% to 4% withdrawal rate from a portfolio. You are taking advantage of the mortality credits, that is from the people in annuities that die early. There are drawbacks no matter what you do, you pick the options that make most sense at the time.
A fool and his money are good for business.

User avatar
305pelusa
Posts: 1002
Joined: Fri Nov 16, 2018 10:20 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 305pelusa » Sun Oct 06, 2019 12:26 pm

JBTX wrote:
Sun Oct 06, 2019 12:10 pm
305pelusa wrote:
Sun Oct 06, 2019 9:43 am
JBTX wrote:
Sun Oct 06, 2019 9:04 am
I am somewhat skeptical of the notion of rising equity glide paths. It is based upon historical data, which assumes the future will always fall into some sub set of the past.
Actually, a rising equity path strategy makes sense from a mathematical and logical perspective too. I have no clue as to its origin (maybe it did come by looking at past data, no clue), but it has serious theoretical backing.
JBTX wrote:
Sun Oct 06, 2019 9:04 am
Let's say your portfolio just went from $1 million to $600k in a crash. Perhaps you were pulling out $50k per year. You just went from 5% withdrawal rate to about 8%. The theory is that the markets will recover, because they always have, so don't worry, be happy!
The theory has absolutely nothing to do with the "market recovering" or some other "double-down" or "value" perspective (although I can see why it looks like that).
The rising equity glide path should be used by retirees with savings and social security. As you age and deplete your portfolio, your social security (an inflation-linked, perfectly safe stream of income) payments become a larger portion of your entire "portfolio". So as the portfolio depletes, it is as though more of your wealth is in safer assets.

Increasing your equity allocation in your slowly shrinking portfolio maintains your desired risk-return characteristics.
If you don't have social security, you wouldn't need a rising equity glide path.
JBTX wrote:
Sun Oct 06, 2019 9:04 am
My argument is it fails to take into account risk aversion properly.
It in fact is necessary to keep account for risk aversion properly. Imagine two retirees with identical risk aversion and desire for risk:
A: Has 500k in savings and SS benefits that might be worth maybe 500k in present day. He invests his portfolio 50/50 (for a total of 250k in stocks and 750k in safe assets).
B: Has 1MM in savings and no social security. He invests 25% in stocks, 75% in bonds (250k/750k)

After a decade, they deplete 30% of their assets. So B now has 175k in stocks and 525k in bonds (for his overall, desired 25/75 split). How should A invest to be logically identical and consistent with B, who has the same attitudes towards risk?
A should put 175kk in stocks and leave 525k in safe assets as well (his SS as well as 25k in bonds). Look at A's portfolio though: It is 87.5% stocks!

That sounds crazy! A must think the market must recover right? Not at all; all A has done is stay consistent with his risk aversion and appetite for risk just like B did. A's portfolio has approximately the same risk-reward characteristics of B.

If A did not increase his equity allocation, he would become more conservative than B, which must be incorrect because we assumed from the beginning they are identical otherwise so they should have the same asset allocation too.
I'll give you the desirability of an increased bond allocation during a pre social security retirement phase. I think somebody called it liability matching. As to when you are drawing social security - when we talk about allocations, we often assume and ignore other income streams. It does make sense however that those income streams affect your portfolio composition.

However, if we are assuming long expense needs (not wants) exceed your social security income stream, then that incremental portion is what we are focusing on.

These studies are always backward looking. Looking backwards, over a prior 30-40 year period and what would have ended up with the most optimal outcome. But that is not how a risk averse investor invests. They invest based upon where they are now, and they continually reevaluate that going forward.

A backward approach may show that having a heavy stock allocation in old age is usually survivable. To the extent stocks may tank, chances are in prior years they rose more than average. It is an indirect way of market timing. So it might not seem unreasonable to have a high stock allocation or an unusually high and temporary withdrawal rate looking backward, when the probabilities are known, but at that point in time, not knowing the future, based upon uncertain probabilities going forward, a high stock allocation may not make sense for a risk averse investor.

To put it more simply, I can't imagine it ever makes sense to be 85 or 90 years old and have 70% stocks, if that is money you need. Now in terms of passing wealth to heirs, or if you have way more than you need, those are different discussions.

Also as to social security, it is an interesting wrinkle into the discussion, but these prior studies were just about withdrawal rates and glide paths. Social security was not part of the discussion, as far as I know.
It's perfectly acceptable for an 80 year old to be 80+% in stocks if all of their needs and more are covered by their SS. That's the whole point; as SS becomes an ever larger part of your portfolio (and source of covering liabilities) you can take more risk with savings.

If the discussion does not have SS, a rising equity glide path still might make sense from a temporal diversification perspective (where risk is minimized by keeping a constant real dollar value exposure to the market). This eliminates sequence of return risk. But the analysis becomes a little more complicated (and some uposters have developed withdrawal strategies that exploit this, will have to dig the thread later). Since most of us do have SS, then (provided we do want to deplete our portfolio with a SWR or VWR), then an equity glide path is necessary for constant risk-reward. It just becomes a question of how much to glide and when.

JBTX
Posts: 5536
Joined: Wed Jul 26, 2017 12:46 pm

Re: Michael Kitces investment strategy -- opinions?

Post by JBTX » Sun Oct 06, 2019 12:38 pm

305pelusa wrote:
Sun Oct 06, 2019 12:26 pm
JBTX wrote:
Sun Oct 06, 2019 12:10 pm
305pelusa wrote:
Sun Oct 06, 2019 9:43 am
JBTX wrote:
Sun Oct 06, 2019 9:04 am
I am somewhat skeptical of the notion of rising equity glide paths. It is based upon historical data, which assumes the future will always fall into some sub set of the past.
Actually, a rising equity path strategy makes sense from a mathematical and logical perspective too. I have no clue as to its origin (maybe it did come by looking at past data, no clue), but it has serious theoretical backing.
JBTX wrote:
Sun Oct 06, 2019 9:04 am
Let's say your portfolio just went from $1 million to $600k in a crash. Perhaps you were pulling out $50k per year. You just went from 5% withdrawal rate to about 8%. The theory is that the markets will recover, because they always have, so don't worry, be happy!
The theory has absolutely nothing to do with the "market recovering" or some other "double-down" or "value" perspective (although I can see why it looks like that).
The rising equity glide path should be used by retirees with savings and social security. As you age and deplete your portfolio, your social security (an inflation-linked, perfectly safe stream of income) payments become a larger portion of your entire "portfolio". So as the portfolio depletes, it is as though more of your wealth is in safer assets.

Increasing your equity allocation in your slowly shrinking portfolio maintains your desired risk-return characteristics.
If you don't have social security, you wouldn't need a rising equity glide path.
JBTX wrote:
Sun Oct 06, 2019 9:04 am
My argument is it fails to take into account risk aversion properly.
It in fact is necessary to keep account for risk aversion properly. Imagine two retirees with identical risk aversion and desire for risk:
A: Has 500k in savings and SS benefits that might be worth maybe 500k in present day. He invests his portfolio 50/50 (for a total of 250k in stocks and 750k in safe assets).
B: Has 1MM in savings and no social security. He invests 25% in stocks, 75% in bonds (250k/750k)

After a decade, they deplete 30% of their assets. So B now has 175k in stocks and 525k in bonds (for his overall, desired 25/75 split). How should A invest to be logically identical and consistent with B, who has the same attitudes towards risk?
A should put 175kk in stocks and leave 525k in safe assets as well (his SS as well as 25k in bonds). Look at A's portfolio though: It is 87.5% stocks!

That sounds crazy! A must think the market must recover right? Not at all; all A has done is stay consistent with his risk aversion and appetite for risk just like B did. A's portfolio has approximately the same risk-reward characteristics of B.

If A did not increase his equity allocation, he would become more conservative than B, which must be incorrect because we assumed from the beginning they are identical otherwise so they should have the same asset allocation too.
I'll give you the desirability of an increased bond allocation during a pre social security retirement phase. I think somebody called it liability matching. As to when you are drawing social security - when we talk about allocations, we often assume and ignore other income streams. It does make sense however that those income streams affect your portfolio composition.

However, if we are assuming long expense needs (not wants) exceed your social security income stream, then that incremental portion is what we are focusing on.

These studies are always backward looking. Looking backwards, over a prior 30-40 year period and what would have ended up with the most optimal outcome. But that is not how a risk averse investor invests. They invest based upon where they are now, and they continually reevaluate that going forward.

A backward approach may show that having a heavy stock allocation in old age is usually survivable. To the extent stocks may tank, chances are in prior years they rose more than average. It is an indirect way of market timing. So it might not seem unreasonable to have a high stock allocation or an unusually high and temporary withdrawal rate looking backward, when the probabilities are known, but at that point in time, not knowing the future, based upon uncertain probabilities going forward, a high stock allocation may not make sense for a risk averse investor.

To put it more simply, I can't imagine it ever makes sense to be 85 or 90 years old and have 70% stocks, if that is money you need. Now in terms of passing wealth to heirs, or if you have way more than you need, those are different discussions.

Also as to social security, it is an interesting wrinkle into the discussion, but these prior studies were just about withdrawal rates and glide paths. Social security was not part of the discussion, as far as I know.
It's perfectly acceptable for an 80 year old to be 80+% in stocks if all of their needs and more are covered by their SS. That's the whole point; as SS becomes an ever larger part of your portfolio (and source of covering liabilities) you can take more risk with savings.
Yes but that's not what we are talking about. Sure if you have more money than you'll ever need it doesn't matter. If all of your needs are covered by SS, then you are talking beyond retirement, or fun money where the consequences of "failure" are limited.
If the discussion does not have SS, a rising equity glide path still might make sense from a temporal diversification perspective (where risk is minimized by keeping a constant real dollar value exposure to the market). This eliminates sequence of return risk. But the analysis becomes a little more complicated (and some uposters have developed withdrawal strategies that exploit this, will have to dig the thread later). Since most of us do have SS, then (provided we do want to deplete our portfolio with a SWR or VWR), then an equity glide path is necessary for constant risk-reward. It just becomes a question of how much to glide and when.

User avatar
goingup
Posts: 3676
Joined: Tue Jan 26, 2010 1:02 pm

Re: Michael Kitces investment strategy -- opinions?

Post by goingup » Sun Oct 06, 2019 12:39 pm

Rising equity glide-path makes sense to me. I wouldn't be willing to start retirement at 30/70 however. We're starting at 60/40 and may drift from there.

User avatar
JoMoney
Posts: 7786
Joined: Tue Jul 23, 2013 5:31 am

Re: Michael Kitces investment strategy -- opinions?

Post by JoMoney » Sun Oct 06, 2019 12:39 pm

nedsaid wrote:
Sun Oct 06, 2019 12:23 pm
dkturner wrote:
Sun Oct 06, 2019 12:15 pm
nedsaid wrote:
Sun Oct 06, 2019 11:59 am
JoMoney wrote:
Sun Oct 06, 2019 11:35 am
The article was behind a registration pay-wall that didn't seem to be worth finding a way around.
From what others have said, the gist seems to be the frequently discussed topic of a rising equity allocation in retirement.
My consideration would be, why not use a SPIA for the initial "bond" allocation at retirement? It should have the same expectation as a portfolio of bond withdrawals amortized over ones life expectancy, along with the guarantee that payments won't stop if you live longer.
After that liability matching floor is set by the SPIA the remainder can be in stocks to support a variable withdrawal rate. If stocks do well, the relative "allocation" will drift higher and support a higher variable withdrawal.
Wow. Are you psychic or what? I was going to post that if we face uncertainties with high valuations and low interest rates, why not take a look at Single Premium Immediate Annuities? Great minds think alike.
If you buy a single premium immediate annuity when interest rates are historically low aren’t you simply locking in a low returning investment?
Yes, but you get regular payments and you get a 5% to 7% withdrawal rate compared to a 3% to 4% withdrawal rate from a portfolio. You are taking advantage of the mortality credits, that is from the people in annuities that die early. There are drawbacks no matter what you do, you pick the options that make most sense at the time.
If you buy bonds (or have a bond portfolio) you are already "locking in a low returning investment".
If interest rates rise, the bonds you bought (or already owned) at the lower rate will commensurately fall in value. Their expected return will simply be the rate at which they were bought at.
Since we're looking at withdrawal strategies, I don't think there should be any allusion that the bonds are going to be rolled over into higher returning bonds (and even that would rely on rising rates that may not occur). Maybe that's the expectation for some, but that implies the bond maturities were not matched to their expected need of the money, but were kept somewhat shorter term expecting to time the market/interest rates.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

livesoft
Posts: 68589
Joined: Thu Mar 01, 2007 8:00 pm

Re: Michael Kitces investment strategy -- opinions?

Post by livesoft » Sun Oct 06, 2019 12:44 pm

I've often felt that one is not going to live forever. Therefore as one grows older one has fewer and fewer years of their life remaining that they need to pay for. With fewer years to live, one actually needs less money to pay for it. That is, longevity risk is going down.

In the meantime, if one if following a typical investment allocation, then one's portfolio is growing and growing. And if one is following a typical withdrawal strategy, then one's portfolio is not being depleted on average. With all these investments hanging around one probably doesn't need all their money in bond funds.

Then one gets to the point where a 50% market decline won't change anything for them since they have so much money they are not spending it anyways. They are going to die and leave lots of money to heirs, charity, or their pets. One might as well not have so much in bond funds at that point.
Wiki This signature message sponsored by sscritic: Learn to fish.

User avatar
vineviz
Posts: 5391
Joined: Tue May 15, 2018 1:55 pm

Re: Michael Kitces investment strategy -- opinions?

Post by vineviz » Sun Oct 06, 2019 1:11 pm

nedsaid wrote:
Sun Oct 06, 2019 12:23 pm
dkturner wrote:
Sun Oct 06, 2019 12:15 pm
If you buy a single premium immediate annuity when interest rates are historically low aren’t you simply locking in a low returning investment?
Yes, but you get regular payments and you get a 5% to 7% withdrawal rate compared to a 3% to 4% withdrawal rate from a portfolio. You are taking advantage of the mortality credits, that is from the people in annuities that die early. There are drawbacks no matter what you do, you pick the options that make most sense at the time.
Very true.

The main challenge with the "locking in low bond yield" argument is actually knowing whether bond yields will be higher or lower in the future. As with most investment questions, you won't know the answer until it is too late (i.e. you're dead).

Imagine a version of yourself retiring in Augst of 2002 and trying to decide whether to buy a SPIA with 10-year Treasuries yielding just 4.26%, the lowest level in 37 years (since 1965)! Someone on this forum would have told you that you were "locking in yields at all-time lows", but ignoring the chants of the market timing folk would have been the right thing to do then just as it is now.

Also, leaving aside the market timing question, the relative benefit of mortality credits is higher for SPIAs when yields are lower which at least partially offsets the lower returns. In other words, as long as the SPIA is taking the place of the bond allocation in your portfolio then it's not obvious that it matters much what the current yield curve looks like.

But any investor who is at or approaching retirement age who is prone to worry about such things, annuitizing some portion of the bond allocation using a series of SPIAs over the course of the next 5 or 10 years probably wouldn't be completely unreasonable. And example of this might be buying an SPIA today using, say, 33% of the money you intend to annuitize with the other 67% in a TIPS ladder or TIPS mutual fund. In 2022 years, annuitize half of what remains and then annuitize the rest in 2025 or 2026. There are plenty of other strategies that would work as well, or maybe better, but this kind of "DCA into SPIA" approach might alleviate the nerves of commitment-phobic investors and would be simple to describe in an IPS and to implement.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
AerialWombat
Posts: 879
Joined: Tue May 29, 2018 1:07 pm

Re: Michael Kitces investment strategy -- opinions?

Post by AerialWombat » Sun Oct 06, 2019 1:12 pm

randomguy wrote:
Sun Oct 06, 2019 10:16 am
Here is a chart from one of his papers
Image
Soooooo.... If I start at 30/70, my AA, and stay there, I have a 94.3% historic chance of success at 4% withdrawal rate.

That’s what this is telling me?

That’s good enough for planning purposes. I see zero reason to add further complication.
“Life doesn’t come with a warranty.” -Michael LeBoeuf

randomguy
Posts: 8402
Joined: Wed Sep 17, 2014 9:00 am

Re: Michael Kitces investment strategy -- opinions?

Post by randomguy » Sun Oct 06, 2019 1:16 pm

305pelusa wrote:
Sun Oct 06, 2019 12:18 pm


For MOST investors in this country, portfolio savings decrease much faster than SS. Most people heavily rely on their savings at the beginning of retirement and might rely almost entirely on SS by the end of retirement. This people need a rising equity path to target a consistent risk-reward.

For investors whose portfolio actually GROWS through retirement (and are not interested in the max SWR), which is in your example, then you'd do the opposite, I agree. I don't think this is the majority of Americans but if it is, then I agree the majority should have a falling equity glide path.

The only way you'd get a constant risk-reward with a constant % allocation would be if, as you said, the numbers happen to work out perfectly. They probably won't for most.
The numbers I have seen have people with 80% of the preretirement savings (not sure if that is nominal or real), after 20+ years of retirement. 1/3rd have more money than they started with. I am pretty sure SS drops in value faster than that. Heck just think of what the death of a spouse does. Neither of these numbers is shocking. If you are down in the 5% SWR range, that is what you would expect.

I have no idea if constant-risk reward is a good idea or if factor in SS into those calculations make sense, but if you want to do that, it has nothing to do with rising equity glide path. You can get a rising or falling path depending on how things go.

randomguy
Posts: 8402
Joined: Wed Sep 17, 2014 9:00 am

Re: Michael Kitces investment strategy -- opinions?

Post by randomguy » Sun Oct 06, 2019 1:21 pm

AerialWombat wrote:
Sun Oct 06, 2019 1:12 pm

Soooooo.... If I start at 30/70, my AA, and stay there, I have a 94.3% historic chance of success at 4% withdrawal rate.

That’s what this is telling me?

That’s good enough for planning purposes. I see zero reason to add further complication.
Well if you up your stocks to 70% over time, you can up that rate to 95.1%:) Or you can look at these charts and go anything between 30/70 and 70/30 and variations in between don't make much of a difference and worrying about if you should be 40/60 or 35/65 isn't worth the time.

User avatar
billthecat
Posts: 437
Joined: Tue Jan 24, 2017 2:50 pm

Re: Michael Kitces investment strategy -- opinions?

Post by billthecat » Sun Oct 06, 2019 1:52 pm

goingup wrote:
Sun Oct 06, 2019 12:39 pm
Rising equity glide-path makes sense to me. I wouldn't be willing to start retirement at 30/70 however. We're starting at 60/40 and may drift from there.
Likewise, I'm at 60/40, but will be at 50/50 upon selling my home (if I put the proceeds into fixed income), which would be the same time as retiring. Then I would whittle bonds down as I withdraw. So this essentially creates a bond tent. I probably would let it drift up to 60/40, but not go beyond that. At that point I might even start withdrawing to let it drift back down to 50/50, or lower, over time.
We cannot direct the winds but we can adjust our sails.

User avatar
305pelusa
Posts: 1002
Joined: Fri Nov 16, 2018 10:20 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 305pelusa » Sun Oct 06, 2019 2:03 pm

randomguy wrote:
Sun Oct 06, 2019 1:16 pm
305pelusa wrote:
Sun Oct 06, 2019 12:18 pm


For MOST investors in this country, portfolio savings decrease much faster than SS. Most people heavily rely on their savings at the beginning of retirement and might rely almost entirely on SS by the end of retirement. This people need a rising equity path to target a consistent risk-reward.

For investors whose portfolio actually GROWS through retirement (and are not interested in the max SWR), which is in your example, then you'd do the opposite, I agree. I don't think this is the majority of Americans but if it is, then I agree the majority should have a falling equity glide path.

The only way you'd get a constant risk-reward with a constant % allocation would be if, as you said, the numbers happen to work out perfectly. They probably won't for most.
The numbers I have seen have people with 80% of the preretirement savings (not sure if that is nominal or real), after 20+ years of retirement. 1/3rd have more money than they started with. I am pretty sure SS drops in value faster than that. Heck just think of what the death of a spouse does. Neither of these numbers is shocking. If you are down in the 5% SWR range, that is what you would expect.

I have no idea if constant-risk reward is a good idea or if factor in SS into those calculations make sense, but if you want to do that, it has nothing to do with rising equity glide path. You can get a rising or falling path depending on how things go.
1) I don't know the numbers of retirees that would benefit from a rising or falling equity path. It literally depends on the numbers of the portfolio. Not even sure why this is worth arguing.
2) It's not that constant risk-reward is desirable or not. It's that your portfolio might have more OR less risk than you intended and that's never good. I want my portfolio to have the risk I intend, whether that's constant or otherwise.

If you intend to have a certain risk-reward of your portfolio, then as years go by and your portfolio grows/shrinks, in the presence of SS, it might be necessary to change the equity % to achieve that intended risk-reward. This is because SS does count as a source of wealth but is not included in your portfolio as an AA. So you might have to under/overcompensate accordingly in your savings I don't see how this is so difficult to understand.

Topic Author
Enginerd
Posts: 10
Joined: Mon Feb 24, 2014 9:46 pm

Re: Michael Kitces investment strategy -- opinions?

Post by Enginerd » Sun Oct 06, 2019 2:17 pm

arcticpineapplecorp. wrote:
Sun Oct 06, 2019 8:46 am
this is sorta old news. Kitces and Pfau's paper was published in 2013 I believe (source: https://papers.ssrn.com/sol3/papers.cfm ... id=2324930) and the bogleheads have discussed this. I typed in three different entries into the bogleheads search bar in the top right corner of the bogleheads site (kitces and pfau, rising equity glide path and bond tent) and found the following entries:

https://www.google.com/search?sitesearc ... s+and+pfau
https://www.google.com/search?sitesearc ... glide+path
https://www.google.com/search?sitesearc ... =bond+tent

I would recommend reading up on all those thoughts and responses and if you want to continue the discussion, feel free of course.

let the search bar be your friend.
Thank you for posting the prior discussions.

User avatar
nedsaid
Posts: 12772
Joined: Fri Nov 23, 2012 12:33 pm

Re: Michael Kitces investment strategy -- opinions?

Post by nedsaid » Sun Oct 06, 2019 3:28 pm

vineviz wrote:
Sun Oct 06, 2019 1:11 pm
nedsaid wrote:
Sun Oct 06, 2019 12:23 pm
dkturner wrote:
Sun Oct 06, 2019 12:15 pm
If you buy a single premium immediate annuity when interest rates are historically low aren’t you simply locking in a low returning investment?
Yes, but you get regular payments and you get a 5% to 7% withdrawal rate compared to a 3% to 4% withdrawal rate from a portfolio. You are taking advantage of the mortality credits, that is from the people in annuities that die early. There are drawbacks no matter what you do, you pick the options that make most sense at the time.
Very true.

The main challenge with the "locking in low bond yield" argument is actually knowing whether bond yields will be higher or lower in the future. As with most investment questions, you won't know the answer until it is too late (i.e. you're dead).

Imagine a version of yourself retiring in Augst of 2002 and trying to decide whether to buy a SPIA with 10-year Treasuries yielding just 4.26%, the lowest level in 37 years (since 1965)! Someone on this forum would have told you that you were "locking in yields at all-time lows", but ignoring the chants of the market timing folk would have been the right thing to do then just as it is now.

Also, leaving aside the market timing question, the relative benefit of mortality credits is higher for SPIAs when yields are lower which at least partially offsets the lower returns. In other words, as long as the SPIA is taking the place of the bond allocation in your portfolio then it's not obvious that it matters much what the current yield curve looks like.

But any investor who is at or approaching retirement age who is prone to worry about such things, annuitizing some portion of the bond allocation using a series of SPIAs over the course of the next 5 or 10 years probably wouldn't be completely unreasonable. And example of this might be buying an SPIA today using, say, 33% of the money you intend to annuitize with the other 67% in a TIPS ladder or TIPS mutual fund. In 2022 years, annuitize half of what remains and then annuitize the rest in 2025 or 2026. There are plenty of other strategies that would work as well, or maybe better, but this kind of "DCA into SPIA" approach might alleviate the nerves of commitment-phobic investors and would be simple to describe in an IPS and to implement.
Yes, an SPIA could be a proxy for bonds. Remember John Bogle's discussion of Social Security as a bond? For many people, the actuarial value of Social Security is probably $300,000 to $400,000, the idea is that you could consider that value of Social Security as sort of a "Shadow Bond" and take that into account in your asset allocation. This is one reason that Bogle said that most retirees were invested too conservatively and that a 65% stock/35% bond portfolio was good for most investors. So the higher percentage of expenses that are covered by guaranteed income from Social Security, Pensions, and Annuities; the more aggressively you can invest as a retiree.
A fool and his money are good for business.

User avatar
nedsaid
Posts: 12772
Joined: Fri Nov 23, 2012 12:33 pm

Re: Michael Kitces investment strategy -- opinions?

Post by nedsaid » Sun Oct 06, 2019 3:31 pm

JoMoney wrote:
Sun Oct 06, 2019 12:39 pm
nedsaid wrote:
Sun Oct 06, 2019 12:23 pm
dkturner wrote:
Sun Oct 06, 2019 12:15 pm
nedsaid wrote:
Sun Oct 06, 2019 11:59 am
JoMoney wrote:
Sun Oct 06, 2019 11:35 am
The article was behind a registration pay-wall that didn't seem to be worth finding a way around.
From what others have said, the gist seems to be the frequently discussed topic of a rising equity allocation in retirement.
My consideration would be, why not use a SPIA for the initial "bond" allocation at retirement? It should have the same expectation as a portfolio of bond withdrawals amortized over ones life expectancy, along with the guarantee that payments won't stop if you live longer.
After that liability matching floor is set by the SPIA the remainder can be in stocks to support a variable withdrawal rate. If stocks do well, the relative "allocation" will drift higher and support a higher variable withdrawal.
Wow. Are you psychic or what? I was going to post that if we face uncertainties with high valuations and low interest rates, why not take a look at Single Premium Immediate Annuities? Great minds think alike.
If you buy a single premium immediate annuity when interest rates are historically low aren’t you simply locking in a low returning investment?
Yes, but you get regular payments and you get a 5% to 7% withdrawal rate compared to a 3% to 4% withdrawal rate from a portfolio. You are taking advantage of the mortality credits, that is from the people in annuities that die early. There are drawbacks no matter what you do, you pick the options that make most sense at the time.
If you buy bonds (or have a bond portfolio) you are already "locking in a low returning investment".
If interest rates rise, the bonds you bought (or already owned) at the lower rate will commensurately fall in value. Their expected return will simply be the rate at which they were bought at.
Since we're looking at withdrawal strategies, I don't think there should be any allusion that the bonds are going to be rolled over into higher returning bonds (and even that would rely on rising rates that may not occur). Maybe that's the expectation for some, but that implies the bond maturities were not matched to their expected need of the money, but were kept somewhat shorter term expecting to time the market/interest rates.
Again great minds think alike. Vineviz made the comment that we don't know the future of interest rates. Maybe in 30 years, us old timers will fondly remember the "good old days" of 2% bond yields. Are negative interest rates in our future? My guess is not. But 2% rates might look awfully, awfully good someday just as we now pine for 6% interest rates.
A fool and his money are good for business.

Pilgprog
Posts: 29
Joined: Mon Aug 19, 2019 10:15 am
Location: Northern New Hampshire

Re: Michael Kitces investment strategy -- opinions?

Post by Pilgprog » Sun Oct 06, 2019 3:52 pm

Wm J. Bernstein put it well - “...in later years aggressive investing may place an otherwise secure retirement at risk. As one approaches [retirement and] has accumulated enough ... capital to safely retire, it makes little sense to put at risk the funds earmarked for retirement living expenses. (The Ages of the Investor: A Critical Look At Life-Cycle Investing (Investing For Adults Book 1) Kindle edition pg 46.) “

This is of course all the more true in the later years of retirement.

Allen

RadAudit
Posts: 3612
Joined: Mon May 26, 2008 10:20 am
Location: Second star on the right and straight on 'til morning

Re: Michael Kitces investment strategy -- opinions?

Post by RadAudit » Sun Oct 06, 2019 4:23 pm

305pelusa wrote:
Sun Oct 06, 2019 2:03 pm
If you intend to have a certain risk-reward of your portfolio, then as years go by and your portfolio grows/shrinks, in the presence of SS, it might be necessary to change the equity % to achieve that intended risk-reward. This is because SS does count as a source of wealth but is not included in your portfolio as an AA. So you might have to under/overcompensate accordingly in your savings I don't see how this is so difficult to understand.
Understand the concept? Yep. Got that. Calculate the changing risk-reward of the portfolio with age and SS? Now I have a problem. I retired at 63 with a 50 / 50 AA, and pension and SS. 9 years later, same thing. But, I'm now more conservative because because I'm older and SS and the portfolio have increased a bit.

How to I calculate the risk-reward of the portfolio then and now? Do you modify your SWR if your current risk-reward is lower than first calculated?
Last edited by RadAudit on Sun Oct 06, 2019 4:46 pm, edited 2 times in total.
FI is the best revenge. LBYM. Invest the rest. Stay the course. - PS: The cavalry isn't coming, kids. You are on your own.

BigJohn
Posts: 1801
Joined: Wed Apr 02, 2014 11:27 pm

Re: Michael Kitces investment strategy -- opinions?

Post by BigJohn » Sun Oct 06, 2019 4:43 pm

Wildebeest wrote:
Sun Oct 06, 2019 9:57 am
in my mind that would make the case for a liability matching portfolio more so than the rising equity glide path.
I think liability matching often results is a rising equity glide path so not sure there that different in practice. If you’re 65 and put 30 years in safe bonds and the rest in stocks you end up at a retirement start allocation. Let’s just use 40/60 as the example. Ten years later you’ve incurred 10 years of liability and have funded that from your safe bonds. The stocks sit there as either fun and/or emergency money. Unless you’ve had too much fun or a huge emergency, that side of the portfolio should have grown but let’s just say it’s flat after 10 years (still worth 40). Since you’ve funded most of you expense from bonds and only have 20 years of liability left the total there should be about 2/3 of the retirement start (from 60 down to 40). As a result, your probably somewhere around 50/50 allocation and have a rising equity glide path.

FWIW, I set my bond allocation when I retired early based on a liability matching concept. After five years I’ve had the right amount of fun and no emergency so my equity allocation has drifted up a bit just as I expected it would.

User avatar
305pelusa
Posts: 1002
Joined: Fri Nov 16, 2018 10:20 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 305pelusa » Sun Oct 06, 2019 5:05 pm

RadAudit wrote:
Sun Oct 06, 2019 4:23 pm
305pelusa wrote:
Sun Oct 06, 2019 2:03 pm
If you intend to have a certain risk-reward of your portfolio, then as years go by and your portfolio grows/shrinks, in the presence of SS, it might be necessary to change the equity % to achieve that intended risk-reward. This is because SS does count as a source of wealth but is not included in your portfolio as an AA. So you might have to under/overcompensate accordingly in your savings I don't see how this is so difficult to understand.
Understand the concept? Yep. Got that. Calculate the changing risk-reward of the portfolio with age and SS? Now I have a problem. I retired at 63 with a 50 / 50 AA, and pension and SS. 9 years later, same thing. But, I'm now more conservative because because I'm older and SS and the portfolio have increased a bit.

How to I calculate the risk-reward of the portfolio then and now? Do you modify your SWR if your current risk-reward is lower than first calculated?
The way to do it is to think of all of your wealth as your portfolio. You have stocks, bonds and cash. Your pension and SS count as well of course. You need to discount all of your expected pension and SS payments to the present (back when you were 63) to determine its present intrinsic value back then. Do the same exercise for today.

From there, you'd apply your desired AA to the entire thing. I can't do this exercise without more information but that tells you how to do so.

What we can say is the following: You say your portfolio has actually grown. Also, the present value of your pension and SS has clearly decreased (you have 9 years of fewer payments left). So the proportion of stocks as part of your total wealth has definitely increased this past 9 years if you've kept a 50/50 allocation. To know how much, you'd need to run the numbers. But we can at least say it has increased. Interestingly, you expressed a desire for a more conservative portfolio. So you're a prime example of why I believe this exercise could be relevant.

User avatar
2pedals
Posts: 929
Joined: Wed Dec 31, 2014 12:31 pm

Re: Michael Kitces investment strategy -- opinions?

Post by 2pedals » Sun Oct 06, 2019 8:46 pm

I like the prime harvesting rules (by McClung) or something similar to it as described as McClung-Smooth in the following article.

https://earlyretirementnow.com/2017/04/ ... harvesting

The key to me is if stocks do poorly right after retirement you want to give stocks a chance to perform before you sell. So in this case it makes sense to float your asset allocation, i.e. wait for the $ amount stocks to go up over the guardrail "say 120%" of your starting point at retirement before you sell. If the stocks do well right after retirement the opposite may apply so selling stocks early may be a good plan.

BH+
Posts: 22
Joined: Sat Aug 31, 2019 5:15 pm

Re: Michael Kitces investment strategy -- opinions?

Post by BH+ » Sun Oct 06, 2019 9:02 pm

At a very basic level, Kitces strategy is very appealing. One withdraws safe and sure assets for near term liabilities, and lets the magic of compounding do its work on stocks over longer time horizons. But even though there is less dispersion of returns for stocks over longer time horizons, there is still the possibility that stocks do not produce the desired returns on the tail end of retirement of an individual retiree.

I think a more balanced approach (such as 50/50) or McClung's prime harvesting may be superior in the sense that if the first decade of retirement coincides with greater stock returns, the retiree take advantage by spending or rebalancing from stocks.

User avatar
beyou
Posts: 2843
Joined: Sat Feb 27, 2010 3:57 pm
Location: Northeastern US

Re: Michael Kitces investment strategy -- opinions?

Post by beyou » Sun Oct 06, 2019 9:16 pm

In theory at 70 when taking SS, you both have more fixed income cf, and less longevity to plan. Makes sense except for fact that we have no idea if SS benefits may be reduced as has been feared, and no idea how long one may live past that point. Can be 30 years except 30 years horizon at the age of 30 is not the same as at 70, in terms of opportunity to have earned income.

As I approach retirement I am still on the same glidpath. When/if I get to SS, I suppose my health may factor in, and the state of the portfolio (factoring in past results, then current expenses, etc). If I ever feel I am investing for my kids, why not increase equity based on their age ? Until I have made such a mental leap, can’t see gliding back up unless I start market timing.

User avatar
firebirdparts
Posts: 346
Joined: Thu Jun 13, 2019 4:21 pm

Re: Michael Kitces investment strategy -- opinions?

Post by firebirdparts » Sun Oct 06, 2019 9:38 pm

I also like the prime harvesting strategy better than this. To me, the prime harvesting strategy is based on the fact that, 10 year into retirement, you would know whether you had good or bad sequence of returns.

This kitces idea is more or less a way of addressing the same risk blindly. It's kind of like the leveraged-when-young stock allocation in reverse. There is a range of times where the negative returns can uniquely hurt you and this range extends from retirement in both directions. Far away from retirement, in both directions, they don't really hurt you if all goes well otherwise.
A fool and your money are soon partners

BigJohn
Posts: 1801
Joined: Wed Apr 02, 2014 11:27 pm

Re: Michael Kitces investment strategy -- opinions?

Post by BigJohn » Mon Oct 07, 2019 7:31 am

If you look at comparisons of various strategies near the end of the prime harvesting paper it would seem that any variation on the theme of rising equity gives similar results and all four beat a fixed allocation. It's likely not statistically significant but the simple rising equity was slightly better than the other four but let's call it a tie. So, while the prime harvesting strategy may seem better, it's unclear that it will give better results.

I personally opt for the simplicity of just a simple glide path. Much easier to implement and stick with as we move through the inevitable cognitive decline that comes with age.

User avatar
firebirdparts
Posts: 346
Joined: Thu Jun 13, 2019 4:21 pm

Re: Michael Kitces investment strategy -- opinions?

Post by firebirdparts » Mon Oct 07, 2019 8:48 am

One of the arguments here against prime harvesting and the bucket strategy was exactly the potentially high equity holdings.
viewtopic.php?f=10&t=286370&start=50
viewtopic.php?f=10&t=192105&start=250

In the case of prime harvesting, that can only happen where you had a "bad" sequence of returns, and of course if you have 4% or 5% withdrawal rate, you can readily see that you could be 10 years down the road into a bear market before you get to a high equity holding. Not selling stocks in a protracted down market is how you get there. There has been some discussion about adding "guardrails" that would force you out of the strategy (selling stocks in a down market before you ran out of money).

Good discussion here:
viewtopic.php?f=10&t=195878

Which in fact includes a comment by rodc that kitces research did not support his conclusions, but they published it anyway. This may already be linked, I'm sorry.

And anyway, 50/50 plus or minus 10 works pretty well.
A fool and your money are soon partners

Post Reply