When 60/40 just isn't appropriate

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Rick Ferri
Posts: 8201
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

When 60/40 just isn't appropriate

Post by Rick Ferri » Sat Feb 07, 2015 5:09 pm

Peter Bernstein wrote The 60/40 Solution in 2002. His seminal article laid out arguments for why 60% stocks and 40% bonds is the “ideal asset allocation” for long-term investors. He considered this allocation the “center of gravity” on a risk and return spectrum.

Bernstein’s observation is timeless advice for many investors, but not everyone. The 60/40 mix is a solid starting point for a discussion about asset allocation for investors who are accumulating assets for retirement. However, it may not be the right starting point for someone living off their savings because the returns can be too volatile.

Read The Center of Gravity for Retirees

Rick Ferri
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

User avatar
nisiprius
Advisory Board
Posts: 33643
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: When 60/40 just isn't appropriate

Post by nisiprius » Sat Feb 07, 2015 5:30 pm

Rick, I would really like to know the origins of "60/40." It was a "traditional" allocation long before Peter L. Bernstein's paper, and incidentally I do not agree that his paper presents any real rationale for 60/40. He asks the obvious question and then doesn't answer it!
Then why not 50/50, or even 40/60? The answer is in how markets work. Rational investors buy stocks only when they can expect to make enough extra in the stock market to compensate for the greater risks involved in owning stocks. This dynamic process of pricing stocks relative to less risky assets explains why, over the long run, stocks have returned more than bonds and why, therefore, more stocks than bonds makes good sense.
Even if you agree that this handwaving explanation explains why someone should have "more stocks than bonds," why 60/40 rather than 55/45 or 67/33 or 75/25?

It's not that I'm against 60/40 or anything, I just can't figure out where it comes from or in what way it's supposed to be optimum. I've been fooling around with efficient frontier charts based on the SBBI data back to 1926 for "large-company stocks," "long-term government bonds," "long-term corporate bonds," and "intermediate-term government bonds," and I can't find any time period for which 60/40 was optimum unless I do some [b]very very deliberate cherry-picking[b]. It doesn't matter whether or not I correct for inflation, by the way.

The results usually show the optimum to be at 50% stocks or less--sometimes much less. For example, if we consider 1926 to 1952, the year in which CREF was founded, I am seeing this. 35/65 was the optimum.

Image
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
whaleknives
Posts: 1210
Joined: Sun Jun 24, 2012 7:19 pm

Re: When 60/40 just isn't appropriate

Post by whaleknives » Sat Feb 07, 2015 5:49 pm

Here's another source, that recommended 60/40 for the "younger distributor:

    John Bogle, on the page opposite his "Simple Rule of Thumb", described a more general "Basic Asset Allocation Model (Stocks/Bonds)" of only four splits, from 80/20 for the younger accumulator to 50/50 for the older distributor. "For example, my highest recommended target allocation for stocks would be 80% for younger investors accumulating assets over a long time frame. My lowest target stock allocation, 50%, would apply to older investors in the distribution phase. These investors must give greater weight to the short-run consequences of their actions." (Bogle on Mutual Funds, 1994, pp. 238-239)

    This is more similar to Benjamin Graham's "never have less than 25% or more than 75%" in stocks (Wiki).
Last edited by whaleknives on Thu May 18, 2017 4:37 pm, edited 2 times in total.
"I'm an indexer. I own the market. And I'm happy." (John Bogle, "BusinessWeek", 8/17/07) ☕ Maritime signal flag W - Whiskey: "I require medical assistance."

Swampy
Posts: 721
Joined: Fri Aug 31, 2012 7:16 am

Re: When 60/40 just isn't appropriate

Post by Swampy » Sat Feb 07, 2015 5:55 pm

nisiprius wrote:

The results usually show the optimum to be at 50% stocks or less--sometimes much less. For example, if we consider 1926 to 1952, the year in which CREF was founded, I am seeing this. 35/65 was the optimum.


Personally, I am on the side of caution.

Attributed to Will Rogers:
"I am not so much concerned with the return on capital as I am with the return of capital."


"October. This is one of the peculiarly dangerous months to speculate in stocks. The others are July, January, September, April, November, May, March, June, December, August, and February."
-- Mark Twain
If you fail to plan, you plan to fail. | Failure is not an option. | If I have seen further, it is because I was carried on the shoulders of giants.

User avatar
jeffyscott
Posts: 6534
Joined: Tue Feb 27, 2007 9:12 am
Location: Wisconsin

Re: When 60/40 just isn't appropriate

Post by jeffyscott » Sat Feb 07, 2015 6:02 pm

Rick Ferri wrote:I propose the center of gravity for those who have accumulated enough for retirement to be 30% stocks and 70% bonds.


nisiprius wrote:For example, if we consider 1926 to 1952, the year in which CREF was founded, I am seeing this. 35/65 was the optimum.


Funny, not long ago it seemed like 50/50 was considered to be extremely conservative (for some reason). Nice to see some confirmation of my own choice to drop to around 35% stocks.
press on, regardless - John C. Bogle

User avatar
Peter Foley
Posts: 3968
Joined: Fri Nov 23, 2007 10:34 am
Location: Lake Wobegon

Re: When 60/40 just isn't appropriate

Post by Peter Foley » Sat Feb 07, 2015 6:07 pm

Rick

Interesting article. You provide, with data to support, a rational argument for retirees to consider 30/70 as the starting point for AA consideration while in the withdrawal phase.

A point I would make is that this is would apply to a subset of retirees; those dependent upon an approximate 4% withdrawal to fund their retirement. This group does not have much wiggle room and needs to invest conservatively so as to protect against running out of money.

Those who have more in savings and are only withdrawing at a 2% rate have more room to maneuver. While they may not have a need to take the risk associated with a 50/50 portfolio (for example), there is no reason for them not to so if they have the risk tolerance that permits them to do so. Please note that I cherry picked both number with this second group. A 50/50 AA with a 2% withdrawal rate is an indication that one's bond allocation is sufficient to cover one's retirement expenses.

jdb
Posts: 1193
Joined: Wed Dec 05, 2012 8:21 pm

Re: When 60/40 just isn't appropriate

Post by jdb » Sat Feb 07, 2015 6:36 pm

Thanks Rick. As a soon to be retiree with a 35-65 allocation it is good to see some support for the allocation amidst the seeming clamor for equity dominant portfolios. It just feels better to me, I agree with William Bernstein articles about why keep playing the game. Thanks again for your very helpful insights.
Last edited by jdb on Sun Feb 08, 2015 7:42 am, edited 1 time in total.

lack_ey
Posts: 5450
Joined: Wed Nov 19, 2014 11:55 pm

Re: When 60/40 just isn't appropriate

Post by lack_ey » Sat Feb 07, 2015 6:47 pm

For some second opinions, Vanguard target retirement funds hit 50/50 at retirement and continue down to 30/70 after 7 years into retirement. On the other hand, the TSP's lifecycle funds are significantly more conservative near the end and hit around 23/77 at retirement (71% of total in G fund). That said, the G fund is so good, skewing safe makes unique sense for it.

User avatar
alpenglow
Posts: 400
Joined: Tue May 31, 2011 12:02 pm

Re: When 60/40 just isn't appropriate

Post by alpenglow » Sat Feb 07, 2015 7:48 pm

Nisi,

Thanks for sharing your research. I've always been on the conservative side, particularly for a younger accumulator. What did you find to be optimum for large caps and intermediate term bonds? Thanks!

garlandwhizzer
Posts: 1523
Joined: Fri Aug 06, 2010 3:42 pm

Re: When 60/40 just isn't appropriate

Post by garlandwhizzer » Sat Feb 07, 2015 7:49 pm

I would like to suggest considering another point of view instead of 30/70, although I do think that Rick's logic is sound for many if not most retirees. First of all, the graphs look great and convincing for how successful this conservative portfolio is on backtesting. It's important to recall that backtesting is dominated by the greatest bond bull market in history 1982 to present. When that bull started Treasury 10 year bonds yielded 15% and 30 year Treasuries yielded about 14%. Currently the 10 year yields about 1.8% and the 30 year yields 2.5%. For todays bonds, be they 10s or 30s, returns are unlikely to be greater than inflation for their 10 or 30 year holding periods. Whereas over the holding period of 30 year bonds purchased in the early 1980s returns have outpaced inflation by about 10% and 10 years a bit less. The difference between bond returns going forward from here and what returns have been in the past 30 - 40 years is a very wide gulf indeed. So the nice graphs while accurately describing the past are unlikely to replicated in the future at least from the bond point of view. A more applicable historical period, more like today, was 1940 - 1980, a period that started with low yields and low inflation and ended up otherwise. Real bond returns during this 4 decade run were negative in inflation adjusted dollars. We don't know if our future will be like that dismal bond period, but if so and 70% of your portfolio is in bonds, you will not be a happy camper and will find little solice in these reassuring graphs.

One can argue that we also expect lower stock returns going forward from here for at least a decade or so. This is true. On the other hand stock returns are much harder to predict (wider dispersion of results) than are bond returns which have a 91% correlation to their yield according to Bogle. For what it's worth, essentially all prognosticators predict that stock returns for a balanced equity portfolio involving both US and international stocks will outperform high quality bonds over the next decade. It seems to me that in our future--the new normal low return low yield world, where longevity is increasing and expected returns are decreasing--the greatest objective risk is not stock volatility but running out of money late in life. Ten years of zero return in real inflation adjusted dollars by 70% of your portfolio (bonds) might reduce volatility risk but it adds risk to this unfortunate outcome.

The other point is that one allocation is not right for all. If one has a sufficiently large asset base (the ability to take risk) and the inclination to do so, I see no problem with a stock heavy portfolio as long as one has sufficient high quality bonds to cover all living expenses for a decade or more without selling any stocks whatsoever. It is unlikely but possible that the market even with dividends reinvested will decline for a decade or more, but again if the initial stock position is sufficiently large running out of money is not going to occur. If everything in the entire world collapses and goes to zero at one time, there would be trouble but who wants to live at that point anyway?

Garland Whizzer

User avatar
Garco
Posts: 626
Joined: Wed Jan 23, 2013 2:04 am

Re: When 60/40 just isn't appropriate

Post by Garco » Sat Feb 07, 2015 7:54 pm

@Nisiprius. Could it be that people find 60/40 attractive for some mystical reason? It's very close to the "golden ratio" -- https://en.wikipedia.org/wiki/Golden_ratio.

I have sometimes gravitated toward something like that in asset allocations of different types, e.g., between growth vs. value in my mid-cap stocks. Some secret force is at work. This ratio feeeels right.
Last edited by Garco on Sat Feb 07, 2015 7:58 pm, edited 1 time in total.

User avatar
nisiprius
Advisory Board
Posts: 33643
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: When 60/40 just isn't appropriate

Post by nisiprius » Sat Feb 07, 2015 7:57 pm

alpenglow wrote:Nisi,

Thanks for sharing your research. I've always been on the conservative side, particularly for a younger accumulator. What did you find to be optimum for large caps and intermediate term bonds? Thanks!
The list of caveats on this stuff is so long I don't really want to post details. Everything depends on endpoints, those silly MPT curves writhe around like worms when you change the endpoints. It's also sensitive to the assumed return of the riskless asset. I'm not totally clear on the relationships between the SBBI "long-term government bond," "long-term corporate bond," and "intermediate-term government bond" series data, and what you would see in real life in a mutual fund.

I'm interested in it intellectually because it is the supposed underpinning for so many things.

I just want to know "what is the supposed rationale for 60/40? Did someone calculate some numbers once, or is it just a meaningless tradition--someone pulled it out of the air and said 'sounds about right to me?'"
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.

Leeraar
Posts: 4109
Joined: Tue Dec 10, 2013 8:41 pm
Location: Nowhere

Re: When 60/40 just isn't appropriate

Post by Leeraar » Sat Feb 07, 2015 8:42 pm

Is there any real evidence that, within reasonable limits, Asset Allocation matters at all over the long run? Let's say, 20/80 to 80/20 in the modern era, 1950 to now. Or, even before?

I think Asset Allocation matters during the accumulation phase, mainly because it sets a course and helps you stick to it.

After you retire, in the spending phase, I think it is more or less irrelevant. Decide on your needed income (whatever your means are) and figure out how to get it, whether by liability matching (Bernstein) or assuring a floor (Zwecher, Otar). If your means are less, you are probably in SPIAs, bond or treasury ladders, pensions, and SS. If your means are more, you can have some degree of equities that are more related to your legacy than your need for income.

This data mining of the historical record is pointless and futile, if the aim is to construct strategies for going forward.

L.
You can get what you want, or you can just get old. (Billy Joel, "Vienna")

deci02
Posts: 194
Joined: Mon Apr 02, 2007 10:19 am

Re: When 60/40 just isn't appropriate

Post by deci02 » Sat Feb 07, 2015 8:49 pm

nisiprius wrote:
alpenglow wrote:Nisi,

Thanks for sharing your research. I've always been on the conservative side, particularly for a younger accumulator. What did you find to be optimum for large caps and intermediate term bonds? Thanks!
The list of caveats on this stuff is so long I don't really want to post details. Everything depends on endpoints, those silly MPT curves writhe around like worms when you change the endpoints. It's also sensitive to the assumed return of the riskless asset. I'm not totally clear on the relationships between the SBBI "long-term government bond," "long-term corporate bond," and "intermediate-term government bond" series data, and what you would see in real life in a mutual fund.

I'm interested in it intellectually because it is the supposed underpinning for so many things.

I just want to know "what is the supposed rationale for 60/40? Did someone calculate some numbers once, or is it just a meaningless tradition--someone pulled it out of the air and said 'sounds about right to me?'"


nisi, doesn’t necessarily address your rationale question but page 18 of this Sharpe study shows 60/40 as an average of total capitalization of stocks and bonds (the market portfolio) from 1976-2009.

Purpose of the paper, however, is addressing the variability of that ratio and an argument against rebalancing back to 60/40 but rather letting the market price action rebalance for you thus maintaining a consistent risk profile vis a vis the “market portfolio”.

Sharpe points out one can set up an allocation based on their own investor characteristics of need and risk aversion (so not necessarily 60/40 for every individual) but then just let the market price risk.

http://web.stanford.edu/~wfsharpe/aaap/wfsaaap.pdf

lack_ey
Posts: 5450
Joined: Wed Nov 19, 2014 11:55 pm

Re: When 60/40 just isn't appropriate

Post by lack_ey » Sat Feb 07, 2015 9:07 pm

Right, and even over that period 40/60 (for example) did better than 60/40 in risk-adjusted return and not much worse in nominal returns. Thanks for pricing this stuff for us, Mr. Market. As always, change the endpoints and...

Curiously enough, currently U.S. and global stock/bond splits by capitalization are just a smidge higher than 40/60.

itstoomuch
Posts: 4297
Joined: Mon Dec 15, 2014 12:17 pm

Re: When 60/40 just isn't appropriate

Post by itstoomuch » Sat Feb 07, 2015 9:10 pm

One of my favorite authors. In fact just checked out, "Against the Gods"
4 buckets: SS+pension;dfr'd GLWB VA & FI anntys, by time & $$ laddered; Discretionary; Rental. Do OK any 2 bkts. LTCi. Own, not asset. Tax 25%. Early SS. FundingRatio (FR) >1.1 Age 67/70

Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: When 60/40 just isn't appropriate

Post by Browser » Sat Feb 07, 2015 9:10 pm

"Optimum" on the efficient frontier means that risk-adjusted return increases up to the optimum point and then it begins to decline. I've done some back of the envelope figgering and have generally found, as as Nisi, that somewhere in the 30-40% range for equity allocation seems to be the "sweet spot". You need to have a sound rationale for moving beyond about 40% because every "unit" of additional risk past that point is rewarded by less-and-less expected return. By the time you hit 80% in equities, the risk-reward begins to go downhill precipitously. So, I too am baffled by the notion that 50% or 60% in equities is "optimal". It isn't. It's appropriate if you really, really need to generate higher expected returns or you want to err on the greed side of the equation. In my mind, most investors should start with 30-40% and work up or down from there based on how compelling their rationale is for doing so. Almost everyone should have at least 10% and almost no-one should hold more than 80%.
We don't know where we are, or where we're going -- but we're making good time.

User avatar
Taylor Larimore
Advisory Board
Posts: 25799
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

"The 60/40 Solution"

Post by Taylor Larimore » Sat Feb 07, 2015 9:16 pm

I just want to know "what is the supposed rationale for 60/40? Did someone calculate some numbers once, or is it just a meaningless tradition--someone pulled it out of the air and said 'sounds about right to me?'"

Nisiprius:

The "60/40 Solution by Peter Bernstein is a very well thought-out article by an authentic investment authority:

The 60/40 Solution

Best wishes.
Taylor
"Simplicity is the master key to financial success." -- Jack Bogle

Tamales
Posts: 1132
Joined: Sat Jul 05, 2014 10:47 am

Re: When 60/40 just isn't appropriate

Post by Tamales » Sat Feb 07, 2015 9:22 pm

There are a lot of different ways to read the tea leaves on this, but here's some data (with all the standard caveats) on the range of 100% to 0% total stock and total bond, from just before the '08 crash to today.

While we normally think about the percent maximum drawdown as indicating pain, look at the duration of that drawdown for the longevity of the pain (last 3 rows in the table). It holds pretty steady from 100% stocks down to 40% stocks but really drops off in going from 40% stocks to 30% stocks in this case. Also note the annualized return barely moves from 100% stocks down to 60% stocks, but both the max drawdown percent and the standard deviation both fell by 30+ percent over the same range.
Sharpe ratio gives little guidance since it is increasing all the way through 90% bonds.

Image

stlutz
Posts: 3854
Joined: Fri Jan 02, 2009 1:08 am

Re: When 60/40 just isn't appropriate

Post by stlutz » Sat Feb 07, 2015 10:05 pm

I would really like to know the origins of "60/40."


I tried googling this and didn't find an answer. What was interesting was how almost every link that came was negative about 60/40.

"Dangerous"
"Outdated"
"Undiversified"
"Not to be trusted"
"Doomed to fail"

User avatar
siamond
Posts: 3163
Joined: Mon May 28, 2012 5:50 am

Re: When 60/40 just isn't appropriate

Post by siamond » Sat Feb 07, 2015 11:50 pm

Ok, I don't get it... One would center his retirement strategy on the volatility (or lack thereof) of 5 years returns? Er, I do expect (I hope!) to be retired for much longer than that. And a 30/70 portfolio behaves especially poorly for 30 or 40 years of retirement - those not eager to do the math by themselves can go through a few runs of Firecalc or cFIREsim and see what happens, the failure rate dramatically increases as you go towards such high proportion of bonds. And this is with a past history where bond returns were significantly higher than current yields...

Seems that in order to avoid some short-term angst (reduced volatility), one would take huge long-term risks... Not terribly attractive to me.

pascalwager
Posts: 866
Joined: Mon Oct 31, 2011 8:36 pm

Re: When 60/40 just isn't appropriate

Post by pascalwager » Sun Feb 08, 2015 1:44 am

Bernstein makes a credible case for owning bonds, but not for any particular percentage allocation, including 40%.

RetiredinKaty
Posts: 112
Joined: Mon Sep 09, 2013 6:52 pm

Re: When 60/40 just isn't appropriate

Post by RetiredinKaty » Sun Feb 08, 2015 3:07 am

Mr. Ferri, thank you for sharing your current thinking on retirement allocation and for having the courage to evolve. At some point I think you should address how you have come to 30/70 as a more typical guideline with respect to your books on AA in which you suggest typical allocations of 50% stock for active retirees and 40% stock for mature retirees. Did the financial crisis, the advent of TIPS, or today’s high equity valuations influence your thinking?

For what it’s worth, I am a mid-60’s retiree and my allocation is 1/3 stock, 1/3 TIPS, and 1/3 Total Bond.

Regards.

User avatar
nisiprius
Advisory Board
Posts: 33643
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: When 60/40 just isn't appropriate

Post by nisiprius » Sun Feb 08, 2015 4:48 am

pascalwager wrote:Bernstein makes a credible case for owning bonds, but not for any particular percentage allocation, including 40%.
Exactly.
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
stemikger
Posts: 4400
Joined: Thu Apr 08, 2010 5:02 am

Re: When 60/40 just isn't appropriate

Post by stemikger » Sun Feb 08, 2015 7:53 am

I recently saw an interview on You Tube where John Bogle said he is increasingly nervous about Target Retirement Funds because they don't take social security into consideration. I know the Target Retirement Income Fund ultimately glides to 30/70 AA stocks/bonds. Isn't this too conservative if you include social security? Also, we are living longer and many would like to leave some money behind for family or charities. If that is one's way of thinking does 60/40 or 50/50 have a place for a permanent AA until death?
Stay the Course!! ~ Press on Regardless!!!

Tamales
Posts: 1132
Joined: Sat Jul 05, 2014 10:47 am

Re: When 60/40 just isn't appropriate

Post by Tamales » Sun Feb 08, 2015 10:34 am

siamond wrote:Ok, I don't get it... One would center his retirement strategy on the volatility (or lack thereof) of 5 years returns? Er, I do expect (I hope!) to be retired for much longer than that. And a 30/70 portfolio behaves especially poorly for 30 or 40 years of retirement - those not eager to do the math by themselves can go through a few runs of Firecalc or cFIREsim and see what happens, the failure rate dramatically increases as you go towards such high proportion of bonds. And this is with a past history where bond returns were significantly higher than current yields...

Seems that in order to avoid some short-term angst (reduced volatility), one would take huge long-term risks... Not terribly attractive to me.


There are a whole lot of variable amounts and time frames that you set in those calculators, which change the results far more than AA (especially annual expenses). I can get 0% failure for the full range from 100% stock/0% bond to 0% stock/100% bonds, for a set of assumptions that seem reasonable, so it's really hard to generalize that 30/70 or lighter is somehow inherently bad or have huge long term risks. It really depends on individual circumstances, but as you note it may not make any sense for some situations.

User avatar
packer16
Posts: 826
Joined: Sat Jan 04, 2014 2:28 pm

Re: When 60/40 just isn't appropriate

Post by packer16 » Sun Feb 08, 2015 10:37 am

Mr. Ferri and Mr. Bogle appear to disagree on the retiree "center of gravity". Mr. Bogle's is closer to 40/60 with a late retirement at 35/65 (based upon Bogle on Mutual Funds). Mr. Ferri appears to have more of an immunization strategy which could lead to a large amount of inflation risk over time. You could use TIPS to reduce the inflation risk but I wonder if that strategy would have a measurement problem as retirees inflation (having higher weight on health care and leisure) may be higher than overall inflation.

Packer
Buy cheap and something good might happen

Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: When 60/40 just isn't appropriate

Post by Browser » Sun Feb 08, 2015 10:42 am

As long as we're quoting Peter Bernstein, it's worth noting that his views about fixed allocation investment portfolios tilted heavily to equities changed a lot. Here are some of his comments from a 2003 interview, not terribly Bogleheadish:
I am suggesting that we have to begin by focusing on the meaning of the long run—think about it differently in the
post-bubble world. That means that our approach to investing’s fundamental problem, asset allocation, has to change.
The thrust of my argument is that we are going to have to learn to live without the crutch of things like policy portfolios—
because the conditions that justified their existence for so long have been shattered.

Well, the upshot of all those efforts is that all policy portfolios always allocate the most money to equities because, in the
long run, the return on equities has been higher than anything else.

You’re challenging that investment gospel?

Yes. I’m well aware that the case I made, that you have to throw this baby out with the bath water, is profoundly unsettling. My point is that we’ve reached a funny position where the long run doesn’t work. Where long run evidence doesn’t fit circumstances as they are today.

Which implies that the long run data are commonly misread on a number of scores. And that policy portfolios, if you will, are obsolete

That the long run, right now, is irrelevant, because “the old long run, she ain’t what she used to be.” And if that’s so, the implications are enormous. Consider: What if we can no longer be confident that stocks are the best place to be in the long run? Or what if nothing is? In other words, suppose we have to move around a lot more than we did in the past?

That’s why the traditional institutional approach, “I will structure my portfolio in this way and make variations on the theme,” won’t work. So what I’m suggesting is, throw it away. You have to be much more unstructured, opportunistic and ad hoc than you have been in the past

welling@weeden FEBRUARY 28, 2003
We don't know where we are, or where we're going -- but we're making good time.

tibbitts
Posts: 6739
Joined: Tue Feb 27, 2007 6:50 pm

Re: When 60/40 just isn't appropriate

Post by tibbitts » Sun Feb 08, 2015 11:06 am

siamond wrote:Ok, I don't get it... One would center his retirement strategy on the volatility (or lack thereof) of 5 years returns? Er, I do expect (I hope!) to be retired for much longer than that. And a 30/70 portfolio behaves especially poorly for 30 or 40 years of retirement - those not eager to do the math by themselves can go through a few runs of Firecalc or cFIREsim and see what happens, the failure rate dramatically increases as you go towards such high proportion of bonds. And this is with a past history where bond returns were significantly higher than current yields...

Seems that in order to avoid some short-term angst (reduced volatility), one would take huge long-term risks... Not terribly attractive to me.

We've become accustomed, at least in the US, to equities producing only "short-term angst", but there's nothing to say that we don't get worse equity returns than bond returns for the entire duration of a retirement - even from the disappointing real yield levels we have with bonds today.

User avatar
siamond
Posts: 3163
Joined: Mon May 28, 2012 5:50 am

Re: When 60/40 just isn't appropriate

Post by siamond » Sun Feb 08, 2015 11:30 am

Tamales wrote:
siamond wrote:Ok, I don't get it... One would center his retirement strategy on the volatility (or lack thereof) of 5 years returns? Er, I do expect (I hope!) to be retired for much longer than that. And a 30/70 portfolio behaves especially poorly for 30 or 40 years of retirement - those not eager to do the math by themselves can go through a few runs of Firecalc or cFIREsim and see what happens, the failure rate dramatically increases as you go towards such high proportion of bonds. And this is with a past history where bond returns were significantly higher than current yields...

Seems that in order to avoid some short-term angst (reduced volatility), one would take huge long-term risks... Not terribly attractive to me.


There are a whole lot of variable amounts and time frames that you set in those calculators, which change the results far more than AA (especially annual expenses). I can get 0% failure for the full range from 100% stock/0% bond to 0% stock/100% bonds, for a set of assumptions that seem reasonable, so it's really hard to generalize that 30/70 or lighter is somehow inherently bad or have huge long term risks. It really depends on individual circumstances, but as you note it may not make any sense for some situations.

I just ran the math with the default parameters, that's it. And I did the same with my own backtesting Excel spreadsheet many times. I mean, it's not rocket science, with historical bonds returns, if you extract more than a couple of points of your bonds-heavy portfolio for your living expenses, then you're slowly eating at it, and this kills you over a few decades unless your small portion of equities do really well. It's actually ironic in a way, for such strategy to work, you have to make a pretty unreasonable bet on the trajectory of equities...

Sure, if you can afford to take 2% a year, then I would agree with Rick's recommendation, but heck, then you can do pretty whatever you want with your AA... I still don't understand what Rick meant by focusing so much on the next 5 years...
Last edited by siamond on Sun Feb 08, 2015 11:37 am, edited 1 time in total.

User avatar
Rick Ferri
Posts: 8201
Joined: Mon Feb 26, 2007 11:40 am
Location: Vagabond, USA, Twitter: @Rick_Ferri
Contact:

Re: When 60/40 just isn't appropriate

Post by Rick Ferri » Sun Feb 08, 2015 11:36 am

RetiredinKaty wrote:Mr. Ferri, thank you for sharing your current thinking on retirement allocation and for having the courage to evolve. At some point I think you should address how you have come to 30/70 as a more typical guideline with respect to your books on AA in which you suggest typical allocations of 50% stock for active retirees and 40% stock for mature retirees. Did the financial crisis, the advent of TIPS, or today’s high equity valuations influence your thinking?

The difference between 60/40 for accumulators and 30/70 for retirees is the goal of the investor. The focus on 60/40 is as much "return desired" as it is "risk control" while the focus of 30/70 is more "risk control " than "return desired." Also, these are starting point, not absolutes. An accumulator can being their analysis at 60/40 and go to higher or lower to achieve a desired risk and return outcome while the 30/70 is purposefully designed at the lowest allocation to equity to be efficient because the goal of most retires is to have lower risk. That being said, the equity position can go higher and often does.

For example, not knowing anything else, I'd start a discussion with a 40 year old at 60/40, but it may end up at 40/60, 80/20 or someplace in-between. Not knowing anything else, I'd start a discussion with a 65 year old at 30/70, but the allocation to equity may well move higher based on their particular situation.

I hope that helps.

Rick Ferri
The views expressed by Rick Ferri are strictly his own as a private investor and author and do not reflect the views of any entity or other persons.

RetiredinKaty
Posts: 112
Joined: Mon Sep 09, 2013 6:52 pm

Re: When 60/40 just isn't appropriate

Post by RetiredinKaty » Sun Feb 08, 2015 11:43 am

Thanks for the response. I think my lower stock allocation fits well with my needs, but I can see that it may not be applicable to all.

Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: When 60/40 just isn't appropriate

Post by Browser » Sun Feb 08, 2015 12:02 pm

Unfortunately, retirees and others are motivated by low bond yields (and fear of rising rates) to shift to higher equity allocations. Magical thinking "Well, I know I can't earn much with bonds, stocks have historically done a lot better than bonds, so let's go heavier in stocks. Might as well take an uncertain risk there as opposed to the certain risk of not making any money with bonds." But this ignores the fact that low expected bond returns don't imply a relatively favorable risk premium for stocks. In fact, the current risk premium for stocks - particularly US Stocks - is near historical lows; which means you are accepting the high risk and volatility of stocks for a much lower expected reward. Not a great situation. When you are taking distributions from a portfolio, volatility is your enemy. Rick is right. A conservative stock allocation of no more than 30% is appropriate as a starting point for most retirees. If holding 70% in bonds gives you cramps, then boosting your stock allocation isn't the solution. A more prudent solution is to re-calculate your retirement spending budget or figure out another way to generate some income in retirement, or postpone retirement a bit if you have that option. We're all between a rock and a hard place, but it is what it is.
Last edited by Browser on Sun Feb 08, 2015 12:04 pm, edited 1 time in total.
We don't know where we are, or where we're going -- but we're making good time.

User avatar
Aptenodytes
Posts: 3740
Joined: Tue Feb 08, 2011 8:39 pm

Re: When 60/40 just isn't appropriate

Post by Aptenodytes » Sun Feb 08, 2015 12:03 pm

I get the logic of 30-70 at age 65. I see multiple strands of argument pointing that way. What I don't see people laying out clearly is what kind of ramp-down strategy makes the most sense. Going from 60% to 30% stocks in less than a decade requires some fast driving no matter how you cut it. In the absence of any clear alternative I am doing the unimaginative thing of cutting the stock percentage about 2.5 percentage points per year. Is there a smarter way to figure this out?

novicemoney
Posts: 247
Joined: Fri Aug 01, 2014 10:36 am

Re: When 60/40 just isn't appropriate

Post by novicemoney » Sun Feb 08, 2015 12:16 pm

siamond wrote:Ok, I don't get it... One would center his retirement strategy on the volatility (or lack thereof) of 5 years returns? Er, I do expect (I hope!) to be retired for much longer than that. And a 30/70 portfolio behaves especially poorly for 30 or 40 years of retirement - those not eager to do the math by themselves can go through a few runs of Firecalc or cFIREsim and see what happens, the failure rate dramatically increases as you go towards such high proportion of bonds. And this is with a past history where bond returns were significantly higher than current yields...

Seems that in order to avoid some short-term angst (reduced volatility), one would take huge long-term risks... Not terribly attractive to me.

I thought this too, but for us all the calculators validate a 30/70 AA at 95 to 100% success rate. Even if we use a conservative return rate of 3%. I am strictly an amateur in all this, but it seems that your retirement spending projections and SWR seem to be the bigger factors.

trasmuss
Posts: 220
Joined: Sun Oct 26, 2008 7:10 am

Re: When 60/40 just isn't appropriate

Post by trasmuss » Sun Feb 08, 2015 12:17 pm

When you used the term "center of gravity" I interpreted it as meaning that 30-70 would be the average of your recommendations for a retired investor and that there would be just as many recommendations below for equities as above. Your response below however indicates that 30-70 would be seen more as a minimum of equities. Which interpretation would most closely match your intent? Also, isn't this a difficult time to be looking at increasing portions of bonds when they are at such low yields? Thanks.

BigJohn
Posts: 1357
Joined: Wed Apr 02, 2014 11:27 pm

Re: When 60/40 just isn't appropriate

Post by BigJohn » Sun Feb 08, 2015 12:32 pm

stemikger wrote:Also, we are living longer and many would like to leave some money behind for family or charities. If that is one's way of thinking does 60/40 or 50/50 have a place for a permanent AA until death?

In my mind these two thoughts pull in the opposite direction. If the concern is living longer then you are probably withdrawing at 4% or close to it in which case I think the argument to be more conservative makes sense. If you have a desire to leave some money behind (and aren't planning to die early to do so) then you probably are below 4% in which case being more aggressive may be OK. However, if you think you can both protect yourself from living longer than planned and leave some money behind with a more aggressive allocation then you're really not protecting yourself for living longer. I think that was the whole point of Rick's article which I found quite useful.

Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: When 60/40 just isn't appropriate

Post by Browser » Sun Feb 08, 2015 12:38 pm

So, I compared a 30/70 to a 60/40 allocation (TSM + TBM), rebalanced annually, with a 4% real withdrawal rate from 2000-2014.

Portfolio___CAGR_____ IRR____Max Drawdown____Sharpe
30/70.......1.78%.......5.73%.......-7.57%.............0.67
60/40.......1.66%.......5.19%.......-20.39%............0.36

As you can see, the conservative allocation has actually turned out to be a better portfolio than the 60/40 portfolio for retirees taking distributions since 2000, despite holding half as much in equities and despite the massive runup in equities over the last 6 years. I believe it makes sense to "cherry pick" the starting date as 2000, since I think it makes sense for retirees to aftercast least favorable scenarios to provide a margin of safety in their planning. After all, you don't get any do-overs in retirement if things don't play out in your favor. The point is that having a portfolio with a small equity allocation doesn't necessarily mean it won't be able to provide the income stream that one with a higher (and riskier) equity allocation will.
We don't know where we are, or where we're going -- but we're making good time.

User avatar
nisiprius
Advisory Board
Posts: 33643
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: When 60/40 just isn't appropriate

Post by nisiprius » Sun Feb 08, 2015 12:38 pm

Rick, just curious. Michael Kitces and Wade Pfau have garnered a lot of ink from a proposal that stock allocation should dip to a low value during the period surrounding the start of retirement, but rises during retirement. Their center of gravity seems "a portfolio that starts at 30 percent in equities and finishes at 60 percent." What do you think about that?

(I don't want to be coy, I personally think it's a dubious idea--a subtle effect seen in backtesting that has gotten a lot of attention because it serves the purposes of the cult of equities...)
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
siamond
Posts: 3163
Joined: Mon May 28, 2012 5:50 am

Re: When 60/40 just isn't appropriate

Post by siamond » Sun Feb 08, 2015 12:51 pm

novicemoney wrote:
siamond wrote:Ok, I don't get it... One would center his retirement strategy on the volatility (or lack thereof) of 5 years returns? Er, I do expect (I hope!) to be retired for much longer than that. And a 30/70 portfolio behaves especially poorly for 30 or 40 years of retirement - those not eager to do the math by themselves can go through a few runs of Firecalc or cFIREsim and see what happens, the failure rate dramatically increases as you go towards such high proportion of bonds. And this is with a past history where bond returns were significantly higher than current yields...

Seems that in order to avoid some short-term angst (reduced volatility), one would take huge long-term risks... Not terribly attractive to me.

I thought this too, but for us all the calculators validate a 30/70 AA at 95 to 100% success rate. Even if we use a conservative return rate of 3%. I am strictly an amateur in all this, but it seems that your retirement spending projections and SWR seem to be the bigger factors.

This is a fair point. With such very conservative rate, this works ok for 30 years. But then it should have been spelled out in the article that such conservative assumption was made to make the long term add up. Not that many people can afford to retire at 3% of their portfolio... Plus good luck with that with the current yields, which will give a very likely poor sequence of returns for the coming decade. Stocks MAY provide a poor sequence of returns or maybe not, we just don't know that, but bonds yields have a pretty good track record at predicting the coming decade.

pkcrafter
Posts: 11764
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: When 60/40 just isn't appropriate

Post by pkcrafter » Sun Feb 08, 2015 1:09 pm

I think the 60/40 portfolio was at one time considered the "policy portfolio." That may have come from the actual invested market or the AA of institutional investors. It doesn't seen to hold any longer.

Article on the end of the 60/40 portfolio. The end, as least until something else comes along.

http://www.pionline.com/article/2012022 ... allocation

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

User avatar
Garco
Posts: 626
Joined: Wed Jan 23, 2013 2:04 am

Re: When 60/40 just isn't appropriate

Post by Garco » Sun Feb 08, 2015 1:10 pm

BigJohn wrote:
stemikger wrote:Also, we are living longer and many would like to leave some money behind for family or charities. If that is one's way of thinking does 60/40 or 50/50 have a place for a permanent AA until death?

In my mind these two thoughts pull in the opposite direction. If the concern is living longer then you are probably withdrawing at 4% or close to it in which case I think the argument to be more conservative makes sense. If you have a desire to leave some money behind (and aren't planning to die early to do so) then you probably are below 4% in which case being more aggressive may be OK. However, if you think you can both protect yourself from living longer than planned and leave some money behind with a more aggressive allocation then you're really not protecting yourself for living longer. I think that was the whole point of Rick's article which I found quite useful.

Left out of these calculations is that one way to protect yourself from "living longer" is to delay retirement. For someone like myself who is retiring at age 70, not only has that given me a chance to build a larger nest-egg, but to the extent that I model the future I think in terms of 25-year survival instead of the typical 30 years. Combine this with deferring SS til age 70, a moderate SWR, and a moderate AA (say 30-70, 40-60, or 50-50), and things look pretty decent.

User avatar
nisiprius
Advisory Board
Posts: 33643
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: When 60/40 just isn't appropriate

Post by nisiprius » Sun Feb 08, 2015 1:20 pm

pkcrafter wrote:...I think the 60/40 portfolio was at one time considered the "policy portfolio...."
Google suggests that you've hit on a key phrase, but I still can't quite find the answer. The phrase "policy portfolio" means one adopted as a policy, and apparently 60/40 was a popular "policy portfolio."

This poster says "The oldest documentation of it as a popular mix I can find is a book from 1986," but I am certain it is far, far older than that. I am pretty sure 60/40 was the target allocation for the Wellington Fund since.... ???????

There are just an amazing number of references that simply call it "traditional" or "classic" and say nothing more about it.

Anyone remember when the Wall Street Journal was publishing a regular series in which they polled money managers for their opinion on stock/bond allocation, and tracked their results and compared it to what WSJ called a "robot mix" of 60/40? That was back in the days when tactical asset allocation was all the rage and it was widely believed that any competent manager could beat a fixed 60/40 by gentle market timing.
Last edited by nisiprius on Sun Feb 08, 2015 1:25 pm, 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.

Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: When 60/40 just isn't appropriate

Post by Browser » Sun Feb 08, 2015 1:24 pm

Jim Otar makes the point that in retirement, three factors have the greatest effect on your fate:

1) Sequence of portfolio returns
2) Inflation
3) Spending rate

Asset allocation is only important in regard to how well it addresses these three risk factors.

-- Sequence of returns risk can be addressed by reducing equity holdings to a minimum, particularly early in retirement.

-- Inflation risk can be addressed by including a strong inflation hedge; but he argues that equities are not particularly effective for this purpose. TIPS are much better.

-- Spending rate should be conservative. When the withdrawal rate is greater than 3.5% the probability of depletion is non-optimal.

Interestingly, he also finds that a 30/70 stock-bond allocation fares the best in aftercasting during the "worst case" historical scenarios. For those who believe this is a prudent way to plan, once again 30/70 looks like the "center of gravity".
We don't know where we are, or where we're going -- but we're making good time.

User avatar
hornet96
Posts: 384
Joined: Sun Nov 25, 2012 6:45 pm

Re: When 60/40 just isn't appropriate

Post by hornet96 » Sun Feb 08, 2015 1:27 pm

Peter Bernstein wrote:
You’re challenging that investment gospel?
....
Consider: What if we can no longer be confident that stocks are the best place to be in the long run? Or what if nothing is? In other words, suppose we have to move around a lot more than we did in the past?

welling@weeden FEBRUARY 28, 2003


This statement from 2003 reeks of "prepper" fears, and implies to me that guns, gold, and canned food should receive a higher investment allocation. This statement was also made on the heels of the dot com crash, 9/11, the start of the wars overseas, etc.

The bottom line to me is that if one really thinks that stocks are doomed as a sound long-term investment choice going forward, then so is everything else as the economies and capital markets will themselves collapse. Absent this fear, I beleive the higher allocation to equities (e.g. 60/40) is based on the premise that stocks will (over the long run) outperform bonds based on the very structure of the capital markets, and thusly should receive more of the portfolio's money. (As for why 60/40 is better than 55/45, 51/49, etc, I don't know and agree with others here that the data is lacking).

Also, I think the relevant point being discussed here is whether lower expected returns from both stocks AND bonds in the near future (10 years?) can continue to support a withdrawal strategy that was based on past asset class performance, which will likely look different going forward. Which begs the question: how is the uncertainty faced (at this time) really any different than the uncertainties faced at any other point in time?

There are a number of factors that affect a sustainable withdrawal strategy - expected future returns being one, along with portfolio size, liquidity needs from the portfolio, expected annual expenses, expected annual inflation, time horizon, unique circumstances (charity, legacy), and so on. Thus, I personally don't beleive that a 30/70 allocation is a one-size-fits-all allocation either, but may be a starting point for the discussion - which I believe Rick noted in his response above.

Derek Tinnin
Posts: 861
Joined: Tue Jan 08, 2008 7:16 pm
Location: cincinnati, oh

Re: When 60/40 just isn't appropriate

Post by Derek Tinnin » Sun Feb 08, 2015 1:40 pm

What needs to be stressed is the dynamic nature of your stock/bond mix over time vs. what the mix might be at a single point in time. That's intuitive and backed by the evidence of what investors tend to do in real time as a natural response to their situation as well as the nature of probabilities.

The stock/bond ratio can be viewed as the accumulation mode and withdrawal mode being mirror images. If the surface of the mirror is the point in time where accumulation mode transitions to withdrawal mode, the stock/bond mix will be the same for both sides at that point in time. As you back away from the mirror, you go back in time, your reflection going forward in time, you will see that your lowest lifetime allocation to stocks is at the surface of the mirror. That transition/retirement date, the surface of the mirror, is at the bottom of one big U-shaped curve. Going back in time, you see your journey from high on the left leg of the U down to the bottom, as you go forward in time, you start your journey up the right leg...

This is all just Monte Carlo logic. The longer the time frame during accumulation mode, the probabilities favor higher stock allocations, the longer the withdrawal period, the probabilities favor lower stock allocations with the lowest stock allocation at the beginning of withdrawal mode...

Just as the impact of savings rates is highest when youngest, so is the impact of withdrawal rates when youngest. They both have a significant/permanent impact if they are "high" relative to your asset base, one being a positive the other being a negative. If you "survive" the first few years of retirement by being more conservative with both your allocation and your withdrawal rate, your subsequent allocation to stocks and withdrawal rate can drift higher over time.

Using Rick's 30/70 example, that might mean a 40 yr. old starts at 60/40, glides down to 30/70 at age 60, then glides back up to 60/40 by age 90, decreasing stocks by 1% per year from 30 to 60, increasing stocks by 1% per year from 60 to 90. If only it were that simple. :)

pkcrafter
Posts: 11764
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: When 60/40 just isn't appropriate

Post by pkcrafter » Sun Feb 08, 2015 1:50 pm

Nisiprius wrote:
There are just an amazing number of references that simply call it "traditional" or "classic" and say nothing more about it.

Yes, I spent quite a bit of time trying to run it down, but didn't find anything more. I could not even find what the overall average asset allocation is for the overall stock/bond market.

I certainly remember the 90's when wall street was telling retail investors to increase their equity allocation. It may have been that 60/40 was much more common before the surge. The odd thing that's going on now is investors are very willing to accept less return for the risk they take. Instead of saying no to lower returns and lower dividends, investors counter it with increased equity allocations to maintain their returns.


Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

peppers
Posts: 1268
Joined: Tue Oct 25, 2011 7:05 pm

Re: When 60/40 just isn't appropriate

Post by peppers » Sun Feb 08, 2015 1:58 pm

The Clash of the Cultures

p. 262 - 263 Wellington Fund

"During most of the Fund's first four decades of existence, the Fund's allocation of assets between equities - typically about 60 - 65 percent of resources - and corporate and government bonds - the remaining 35 - 40 percent varied around these ratios."
"..the cavalry ain't comin' kid, you're on your own..."

User avatar
1210sda
Posts: 1171
Joined: Wed Feb 28, 2007 8:31 am

Re: When 60/40 just isn't appropriate

Post by 1210sda » Sun Feb 08, 2015 2:06 pm

Tamales wrote:There are a lot of different ways to read the tea leaves on this, but here's some data (with all the standard caveats) on the range of 100% to 0% total stock and total bond, from just before the '08 crash to today.

While we normally think about the percent maximum drawdown as indicating pain, look at the duration of that drawdown for the longevity of the pain (last 3 rows in the table). It holds pretty steady from 100% stocks down to 40% stocks but really drops off in going from 40% stocks to 30% stocks in this case. Also note the annualized return barely moves from 100% stocks down to 60% stocks, but both the max drawdown percent and the standard deviation both fell by 30+ percent over the same range.
Sharpe ratio gives little guidance since it is increasing all the way through 90% bonds.

Image


Could you please explain the meaning and significance of "maximum draw down".
1210

scone
Posts: 1384
Joined: Wed Jul 11, 2012 4:46 pm

Re: When 60/40 just isn't appropriate

Post by scone » Sun Feb 08, 2015 2:21 pm

I think the 30/70 portfolio, with a 2.5% withdrawal rate, will go through hell and high water. It even does pretty well when interest rates are rising, as long as inflation isn't too bad. And of course, today one can use TIPS to help with inflation. For me, it's so much simpler to save more now, so that 2.5% is a good-sized sum, than to increase the stock portion and stress out about the markets. It's just not worth it to me.
"My bond allocation is the amount of money that I cannot afford to lose." -- Taylor Larimore

Post Reply