Under what conditions is a 50/50 portfolio worse than any other?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

My preference has always been for a portfolio that doesn't guess at the outcome of the markets - an all seasons type portfolio. I know that back-tests can show you what hasn't ever worked, but it can't show you what will work, and I have no rational basis for believing that a 60/40 portfolio going forward will out-perform a 30/70 portfolio over my investment period. I've looked at the Golden Butterfly, All Seasons, and Permanent Portfolios, and while interesting, I do worry that they are essentially back-tested to perfection, and may have nothing to do with the future.

As such, the only thing I can think of is to just go 50/50. 50% stocks (VTI+VXUS), 50% bonds. The bonds are there for ballast and hopefully some anti-correlation, not income, so they are treasuries only, split 50/50 between long and short, The long and short are split between nominals and inflation-protected.

Assuming this meets my risk tolerances (it does, need = 45%, want = 30%, can = 70%) and investment objectives (it probably does, assuming we don't have 3% returns and 5% inflation for the next 56 years straight):

a) Is there anything obviously wrong with this approach, something glaringly obvious in my own blind spot, and;

b) Is there anything that can be done to improve it, without guessing at the outcomes? For instance, inflation-protected only on the short end and nominals on the long end of the bonds? Or adding 5% gold, or something else mechanical and obvious?

I figure that if you go 50/50 you can only ever be half wrong, but is this itself wrong? Can a 50/50 actually be much worse than picking an overweight to bonds or stocks?
60% AVGE | 20 Year TIPS LMP | 5% Cash
Northern Flicker
Posts: 15363
Joined: Fri Apr 10, 2015 12:29 am

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Northern Flicker »

It is hard to find fault with a portfolio consisting of 50% stocks/25% treasuries/25% TIPS assuming you can tolerate that level of risk, and do not need to take more risk to meet your objectives.
JBTX
Posts: 11227
Joined: Wed Jul 26, 2017 12:46 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by JBTX »

I would think a 50/50 portfolio would always be better than an overweight of one and worse than an overweight of the other.

You could get scenarios where both are negative. Perhaps a period of sharply rising interest rates.
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

JBTX wrote: Sun Nov 28, 2021 2:48 am I would think a 50/50 portfolio would always be better than an overweight of one and worse than an overweight of the other.

You could get scenarios where both are negative. Perhaps a period of sharply rising interest rates.
This is the problem. I've used Simba to look at 1940-1980 and 1966-1996 as two particularly difficult times, and 50/50 does well enough in both of them, whereas to optimize for one or the other you'd need to overweight stocks or bonds. So it seems to be good enough for any era (stays at or above the 4% SWR consistently).
60% AVGE | 20 Year TIPS LMP | 5% Cash
Dude2
Posts: 1780
Joined: Fri Jun 08, 2007 3:40 pm
Location: FL

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Dude2 »

It may be true that the greater the percentage of bonds, the greater the portfolio is exposed to inflation risk. Stocks and bonds both may be negatively affected due to inflationary economies, but bonds have no mechanism of recovery. Once the money has been eaten away, it's gone. This isn't implying that stocks will recover but that they have the possibility of doing so. Therefore, a big takeaway for people is to include TIPS more as a function of bond percentage. Very recently nobody believed unexpected inflation could happen -- that the Fed couldn't even scare up the level of inflation they were aiming for. This may be very short lived and go back into that mode -- in which case nobody is going to care again. Lots of irrational exuberance on the TIPS side of late. I'm mostly talking about rational exuberance for TIPS based on high bond percentage. Arguments can be made that TIPS are the more perfect bond, and that isn't recency bias talking. We can go down the rabbit hole of that discussion. There are plenty of threads to examine already.

In general your concepts of holding short and long and worrying about short with TIPS etc. -- that all seems like too much to me. Have you seen FIPDX, i.e. something akin to the total bond market of TIPS? I try to use that because I believe in holding total markets and let the market decide proportions for me. That being said, TIPS are a small percentage of the total bond market, i.e. less than 10% of Treasuries, around 4% of TBM, possibly less than 2% of total world bond market. Also, the market knows nothing about my personal need for the money which should be a factor in the duration I select.

Anyway, those facts do not free us from the need for inflation protection with high bond percentages. A TIPS liability matching portfolio can be a nice way to solve this problem, and also the mechanics of it force people to buy and hold to maturity, guaranteeing face value of the bond in case deflation rears its head. LMP addresses the problem we are actually trying to solve, i.e. sufficient income to last us the rest of our lives, not just an arbitrary pot of money. Taking risk is generally necessary to get to where we have any hope of reaching a solution. In addition, what pot size is enough to weather all storms -- it doesn't solve the problem per se.

All of this is theoretical. Social Security is inflation adjusted. Your income at work is hopefully inflation adjusted. The point is that the more bonds you have, the more inflation risk, and the less reward potential. That is a scary position to be in. Stocks hide many ills inside of themselves and have historically compensated investors for their risks. Is 50% of them enough? I've been 50/50 since 2008, and things are fine. There is I think always in investing a luck factor, no matter what plan you pick and stick to. Figuring out a way to earn more and spend less is a good approach; otherwise, the 60/40 portfolio is often the go-to answer. Splitting hairs between 40/60, 50/50, and 60/40 has lead most to determine it didn't matter much. Even those with 100/0 are often beat into submission to recognize that 80/20 is pretty much the same. Nobody knows the answer, and this is a what helps you sleep at night kind of thing. 2008 obviously shook me to the core (and this was not that long after the dot com bubble). That was my investment lifetime. People older would have told me about similar crises that came before. If you can't handle the heat, lower the burner on stocks until you can. That's what drove me to 50/50.
Last edited by Dude2 on Sun Nov 28, 2021 6:02 am, edited 1 time in total.
Then ’tis like the breath of an unfee’d lawyer.
sycamore
Posts: 6360
Joined: Tue May 08, 2018 12:06 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by sycamore »

Regarding the thread question "Under what conditions is a 50/50 portfolio worse than any other?" are you literally asking if there are conditions where 50/50 does worse than all the other portfolio allocations in 100/0, 99/1, .. 1/99, 0/100? It seems to me one helpful answer to that question is: no -- 50/50 can and will do just as badly as other portfolios in the range 55/45 to 55/45 (or pick whatever range you like). Differences of a few percentage points aren't very meaningful. You can feel good knowing that 50/50 may do poorly but you'll have lots of company.

Regarding inflation protection, consider using I Bonds for part of the portfolio. They are limited to $10k per person per year, plus another $5k on a tax return. For many people that's a good chunk of spending protection. Maybe the rest of one's spending will be covered (or mostly so) with inflation-adjusted Social Security benefits. With I Bonds, the inflation adjustment never goes negative so you always have a non-negative real return (potentially zero, but not negative). Whereas with TIPS (or TIPS funds) the adjustment can go negative and you can lock in a negative real return.
mmcmonster
Posts: 618
Joined: Sun Jan 12, 2014 12:18 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by mmcmonster »

My concern about 50/50 portfolios (25% treasuries, 25% TIPS) is that the 60/40 was devised in a period where treasuries had much better returns than current and much better returns than current forecasts for the future.

If you are not dependent on income from the bonds half of your portfolio, that's totally fine... but I don't see a 4% SWR with 60/40 (let alone 50/50) as something we should target. It may work out, but I'm not confident.
loukycpa
Posts: 735
Joined: Wed Aug 05, 2020 9:52 am

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by loukycpa »

A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
"The safe assumption for an investor is that over the next hundred years, the currency is going to zero." - Charlie Munger
UpperNwGuy
Posts: 9479
Joined: Sun Oct 08, 2017 7:16 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by UpperNwGuy »

Phyneas wrote: Sun Nov 28, 2021 4:15 am
JBTX wrote: Sun Nov 28, 2021 2:48 am I would think a 50/50 portfolio would always be better than an overweight of one and worse than an overweight of the other.

You could get scenarios where both are negative. Perhaps a period of sharply rising interest rates.
This is the problem. I've used Simba to look at 1940-1980 and 1966-1996 as two particularly difficult times, and 50/50 does well enough in both of them, whereas to optimize for one or the other you'd need to overweight stocks or bonds. So it seems to be good enough for any era (stays at or above the 4% SWR consistently).
I don't think you're using the word "overweight" correctly in this thread. When we overweight something, we are deviating from a generally-accepted equilibrium, such as market capitalization. A 50/50 portfolio is a purely arbitrary asset allocation, so 60/40 and 40/60 are not overweighted.
Blue456
Posts: 2152
Joined: Tue Jun 04, 2019 5:46 am

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Blue456 »

loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
That really depends. Dr. Bernstein recommends 50:50 for young person until first prolonged recession is experienced. He makes a case that higher allocation can lead to panic, stock sale and permanent loss during prolonged and severe downturn such as 2008 or Dotcom. So there is a behavioral aspect of being always half right and a loss that is limited to only half of one portfolio.
User avatar
nisiprius
Advisory Board
Posts: 52211
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: Under what conditions is a 50/50 portfolio worse than any other?

Post by nisiprius »

50% stocks (VTI+VXUS), 50% bonds. The bonds are there for ballast and hopefully some anti-correlation, not income, so they are treasuries only, split 50/50 between long and short, The long and short are split between nominals and inflation-protected.
1) The portfolio is fine. I like it because, well, it's not wildly different from mine. Being old, retired, and conservative/risk-averse/fraidycat I have less than 50% in stocks. Always having been worried about inflation I have more than 50% of my bonds in TIPS and I bonds. Both my nominal bond fund (Total Bond) and TIPS fund (Vanguard Inflation-Protected Securities) have a mix of terms--long, short, and intermediate--and I haven't and won't bother to figure out what the percentages are.

I don't know what you mean by "worse than any other." I think that for many purposes, your portfolio is as good as any normal, traditional, centrist retirement savings portfolio. Many will argue that yours is worse than their favored portfolio but I don't think that's what you're asking. I think there's a swarm of portfolios that are all "good," of which you simply cannot prove that any of them is any better than any other. Thirty years from now you will know which did best, but you still won't know if it was "the best portfolio" or whether it was just the luck and the breaks of that time period.

John C. Bogle wrote
Successful investing involves doing just a few things right and avoiding serious mistakes.
I can't think of a plausible scenario in which your portfolio would do so much worse than any major target-date fund that you would call it "a serious mistake."

2) 50/50 is fine, but your reasoning for 50/50 is flawed. The problem is that all equal splits depend on how you've chosen to define the boundaries and categories you're splitting between. The assumptions you're hoping to avoid, cutting the Gordian knot with a 50/50 sword, just get smuggled in when you ask "splitting between what." You're kidding yourself when you think you've achieved some objective goal with an "equal" split between subjectively defined categories. We can see this in your own example. Why not ⅓ each stocks, nominal bonds, and TIPS? Why not ⅕ each stocks, long nominal, short nominal, long TIPS, and short TIPS? Why not a tenth each large-cap growth, large-cap blend, large-cap value, mid-cap growth, mid-cap blend, mid-cap value, small-cap growth, small-cap blend, small-cap value, and bonds? Why not a fifth each stocks, bonds, rental income property, commodity futures, and collectible gold coins? Oh, sure, I personally think stocks and bonds are almost natural categories (but preferred stocks are bond-like and junk bonds are stock-like and are REITs "just" stocks?)

By the way, "I don't know so I'll put equal amounts in each" has a name, "naïve diversification."

3) I think cap-weighting is a much more "natural" division than equal-weighting. However, I'm not so sure that it applies if the assets in question don't trade freely and frictionlessly against each other within a single unified market. Anyway, look up the thread on Bill Sharpe's preferred portfolio for arguments that the natural, prediction-free, viewpoint-free division is according to the total capitalization of stocks and bonds. Fortunately for your view, that turns out not to be wildly different from 50/50. See postings by others downthread for recent values.

4) I can cite three major authorities who have said something nice about 50/50 naïve diversification between stocks and bonds. (I'll leave it as an exercise to look them up if you don't know who these people were).

Benjamin Graham:
We have suggested as a fundamental guiding rule that the investor should never have less than 25% or more than 75% of his funds in common stocks, with a consequent inverse range of between 75% and 25% in bonds. There is an implication here that the standard division should be an equal one, or 50–50, between the two major investment mediums.
Harry Markowitz:
I should have computed the historical co-variances of the asset classes and drawn an efficient frontier. Instead, I visualized my grief if the stock market went way up and I wasn’t in it–or if it went way down and I was completely in it. My intention was to minimize my future regret. So I split my contributions 50/50 between bonds and equities.
John C. Bogle:
There are an infinite number of strategies worse than this one: Commit, over a period of a few years, half of your assets to a stock index fund and half to a bond index fund. Ignore interim fluctuations in their net asset values. Hold your positions for as long as you live, subject only to infrequent and marginal adjustments as your circumstances change. When there are multiple solutions to a problem, choose the simplest one.
5) It's reasonable to expect very low correlation between stocks and Treasury securities. This doesn't come solely from pragmatic observation, but also from the different fundamental nature of equity and debt.

6) There's spirited debate about this but I, for one, think it is totally unsound to rely, in any important way, on anticorrelation. It drives me bananas that you so often hear short sound-bite "of course, take-it-for-granted" asides saying that the two "have" negative correlation. The correct statement is that they "have had" negative correlation for twenty years. Following thirty-five years of positive correlation, which followed a period of negative correlation, and so on. My sound bite is "zero correlation with big fluctuations." Of course the sophisticates think they know the reasons for those fluctuations and think they make predictions based on them. You should keep the long-term picture of stock-bond correlation in your mind just the way you keep the long-term picture of stock performance in your mind:

Image

7) You need to define much more carefully what you mean by "better" and "worse."

8) The difference between 60/40 and 50/50 is negligible. That's why Vanguard thinks they cover the range adequately with only four LifeStrategy funds, allocated 80/20, 60/40, 40/60, and 20/80.

9) Obviously in any period of time when stocks are outperforming bonds, which is usually but not always, 50/50 will underperform portfolios with more stocks, and outperform portfolios with less stocks. However, if you use almost any measure that takes risk into account in any way, the picture changes. Oddly in view of the cultish adherence to 60/40, over most long periods of time, the highest risk-adjusted return as measured by the Sharpe ratio has been much lower than either 60/40 or 50/50. It has been more in the ballpark of 25/75. That is one of the fundamental bases of "risk parity" strategies. It is way more complicated than that, and involves belief systems about specific asset classes, and I don't personally like it, and the real-world performance of "risk parity" funds has included stumbles so bad that AQR abandoned the strategy in their own main risk parity fund... but anyway.

10) Bogleheads place a lot of weight on Bogle's exhortation, "Stay the course." You need to have enough conviction in your portfolio to stick to it. Virtually all presentations of strategies involve backtesting that assumes investors who did stay the course. You may not get the expected results from a backtested portfolio that you follow, but you sure as heck won't get the expected results if you don't follow. A long-term strategy that's revised frequently isn't a long-term strategy.
Last edited by nisiprius on Sun Nov 28, 2021 2:10 pm, edited 2 times 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.
loukycpa
Posts: 735
Joined: Wed Aug 05, 2020 9:52 am

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by loukycpa »

Blue456 wrote: Sun Nov 28, 2021 6:33 am
loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
That really depends. Dr. Bernstein recommends 50:50 for young person until first prolonged recession is experienced. He makes a case that higher allocation can lead to panic, stock sale and permanent loss during prolonged and severe downturn such as 2008 or Dotcom. So there is a behavioral aspect of being always half right and a loss that is limited to only half of one portfolio.
Ok sure if you want to consider the behavioral aspect. If you have a high proclivity to buy high and sell low then any substantial allocation to equities could be a problem. This person should really be steered toward a target date fund.

From an optimal portfolio perspective, with negative real interest rates today on bonds, if 50/50 turns out to be the optimum allocation for a young accumulator over the next 30 years it is going to be rough sledding.
"The safe assumption for an investor is that over the next hundred years, the currency is going to zero." - Charlie Munger
User avatar
nisiprius
Advisory Board
Posts: 52211
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: Under what conditions is a 50/50 portfolio worse than any other?

Post by nisiprius »

Backtesting is what it is, but at least it's a reality check. I'm going to make a guess and test it. I'm going to compare your portfolio, using the funds and ETFs that come to mind--PIMCO LTPZ and STPZ for TIPS.

I'm going to compare it to straight 60/40. I think you're talking US-only. I think I'll compare it to two 60/40 funds, the Vanguard Balanced Index Fund which is equivalent to 60% Total Stock, 40% Total Bond, and the Vanguard LifeStrategy Moderate Growth, which has been close to 60/40 and has had a more complicated portfolio and history. Wait, I'll add a third. Which Fidelity Freedom fund is currently about 60/40? Well, Fidelity Freedom 2030 is 61% stocks, let's use it.

My expectation is that there will not be huge differences between them, and that yours won't be either the lowest or the highest in average return for the whole period.

I'll post this however it turns out. Time period turns out to be limited by the PIMCO long-term TIPS ETF, LTPZ.

Source

Image

There you go. A reality check. Your portfolio isn't anything bizarre or crazy, it landed kind of in the middle. Beat Fidelity's target-date fund, didn't do as well as the pure Balanced Index fund, and almost perfectly matched the Vanguard 60/40 LifeStrategy fund. The main thing is that your portfolio made about 10% per year, and the spread on that chart is like 10%±1%.

Note that your portfolio beat all the others in risk-adjusted return as measured by the Sharpe and Sortino ratios, hey, impressive.

You can say correctly that this is not a very typical time period. You can also go nuts trying to figure out how to attribute the differences--my guess but I'm not going to waste time on it is that by cap weight you might be heavier on short-term bonds than either a total bond or total TIPS fund and that might have both smoothed out the 2019 and 2020 potholes and dragged down return, just a guess, too lazy to try to figure it out.
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
bertilak
Posts: 10725
Joined: Tue Aug 02, 2011 5:23 pm
Location: East of the Pecos, West of the Mississippi

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by bertilak »

Phyneas wrote: Sun Nov 28, 2021 1:15 am As such, the only thing I can think of is to just go 50/50. 50% stocks (VTI+VXUS), 50% bonds.
That had been my take, for the reasons you give, which I expressed as "Whatever happens, half of my (decently sized) portfolio was invested with the best allocation possible under the circumstances (all stock or all bond) and the other half was also a decent investment."

Then I started to realize that I was in a pretty safe position:
  • Large enough portfolio
  • Low expenses (no mortgage, conservative lifestyle)
  • Steady income (pension, social security) covering all those expenses
I started to think that 50/50 was probably more conservative than need be. So I moved to 60/40.

Next, I realized that whatever safety my portfolio had was in the SIZE of my bond holdings, not their percentage of the whole, so I decided to let my equity portion grow and not rebalance into bonds. I am now at 65/35 and happy with that.
May neither drought nor rain nor blizzard disturb the joy juice in your gizzard. -- Squire Omar Barker (aka S.O.B.), the Cowboy Poet
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

sycamore wrote: Sun Nov 28, 2021 6:02 am Regarding the thread question "Under what conditions is a 50/50 portfolio worse than any other?" are you literally asking if there are conditions where 50/50 does worse than all the other portfolio allocations in 100/0, 99/1, .. 1/99, 0/100? It seems to me one helpful answer to that question is: no -- 50/50 can and will do just as badly as other portfolios in the range 55/45 to 55/45 (or pick whatever range you like). Differences of a few percentage points aren't very meaningful. You can feel good knowing that 50/50 may do poorly but you'll have lots of company.
Yes, that's what I was looking for. I've noticed in Simba that ratios are less important than the constituent parts, and there can be surprising outcomes from mixing seemingly similar components in different proportions, situations where certain relationships of bond durations actually don't mix well together or something, so I was wondering if there was something that I had missed in just settling on 50/50, but that's good to know that it should have any outlier properties relative to its near surroundings.
60% AVGE | 20 Year TIPS LMP | 5% Cash
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

mmcmonster wrote: Sun Nov 28, 2021 6:10 am My concern about 50/50 portfolios (25% treasuries, 25% TIPS) is that the 60/40 was devised in a period where treasuries had much better returns than current and much better returns than current forecasts for the future.

If you are not dependent on income from the bonds half of your portfolio, that's totally fine... but I don't see a 4% SWR with 60/40 (let alone 50/50) as something we should target. It may work out, but I'm not confident.
You feel that a higher allocation to equities is necessary? I have the ability to go up to 70% equities before my SWR threshold is threatened in a big and prolonged downturn - something like that?

Regardless of the equity level, do you think that 50/50 TIPS/nominal is a sensible long-term position for bonds? I know that if you're going higher equities you're supposed to go longer bonds, but I'm not comfortable holding most of my bonds at a long duration under the present very-1940'ish situation.
60% AVGE | 20 Year TIPS LMP | 5% Cash
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
I am semi-retired at 37. I've started a new business with uneven income, but I'm at around 60x at the moment, so growth isn't the main priority, but running out of money over a potentially long retirement is the real concern.
60% AVGE | 20 Year TIPS LMP | 5% Cash
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

UpperNwGuy wrote: Sun Nov 28, 2021 6:30 am
Phyneas wrote: Sun Nov 28, 2021 4:15 am
JBTX wrote: Sun Nov 28, 2021 2:48 am I would think a 50/50 portfolio would always be better than an overweight of one and worse than an overweight of the other.

You could get scenarios where both are negative. Perhaps a period of sharply rising interest rates.
This is the problem. I've used Simba to look at 1940-1980 and 1966-1996 as two particularly difficult times, and 50/50 does well enough in both of them, whereas to optimize for one or the other you'd need to overweight stocks or bonds. So it seems to be good enough for any era (stays at or above the 4% SWR consistently).
I don't think you're using the word "overweight" correctly in this thread. When we overweight something, we are deviating from a generally-accepted equilibrium, such as market capitalization. A 50/50 portfolio is a purely arbitrary asset allocation, so 60/40 and 40/60 are not overweighted.
I should have clarified that, I meant overweight relative to my risk/baseline preference. What do you think of the basic premise though?
60% AVGE | 20 Year TIPS LMP | 5% Cash
KlangFool
Posts: 31525
Joined: Sat Oct 11, 2008 12:35 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by KlangFool »

Phyneas wrote: Sun Nov 28, 2021 10:37 am
loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
I am semi-retired at 37. I've started a new business with uneven income, but I'm at around 60x at the moment, so growth isn't the main priority, but running out of money over a potentially long retirement is the real concern.
60X with or without the business income?

KlangFool
30% VWENX | 16% VFWAX/VTIAX | 14.5% VTSAX | 19.5% VBTLX | 10% VSIAX/VTMSX/VSMAX | 10% VSIGX| 30% Wellington 50% 3-funds 20% Mini-Larry
theorist
Posts: 1329
Joined: Sat Sep 28, 2019 11:39 am

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by theorist »

nisiprius wrote: Sun Nov 28, 2021 8:09 am Backtesting is what it is, but at least it's a reality check. I'm going to make a guess and test it. I'm going to compare your portfolio, using the funds and ETFs that come to mind--PIMCO LTPZ and STPZ for TIPS.

I'm going to compare it to straight 60/40. I think you're talking US-only. I think I'll compare it to two 60/40 funds, the Vanguard Balanced Index Fund which is equivalent to 60% Total Stock, 40% Total Bond, and the Vanguard LifeStrategy Moderate Growth, which has been close to 60/40 and has had a more complicated portfolio and history. Wait, I'll add a third. Which Fidelity Freedom fund is currently about 60/40? Well, Fidelity Freedom 2030 is 61% stocks, let's use it.

My expectation is that there will not be huge differences between them, and that yours won't be either the lowest or the highest in average return for the whole period.

I'll post this however it turns out. Time period turns out to be limited by the PIMCO long-term TIPS ETF, LTPZ.

Source

Image

There you go. A reality check. Your portfolio isn't anything bizarre or crazy, it landed kind of in the middle. Beat Fidelity's target-date fund, didn't do as well as the pure Balanced Index fund, and almost perfectly matched the Vanguard 60/40 LifeStrategy fund. The main thing is that your portfolio made about 10% per year, and the spread on that chart is like 10%±1%.

Note that your portfolio beat all the others in risk-adjusted return as measured by the Sharpe and Sortino ratios, hey, impressive.

You can say correctly that this is not a very typical time period. You can also go nuts trying to figure out how to attribute the differences--my guess but I'm not going to waste time on it is that by cap weight you might be heavier on short-term bonds than either a total bond or total TIPS fund and that might have both smoothed out the 2019 and 2020 potholes and dragged down return, just a guess, too lazy to try to figure it out.
There is one thing here that really surprises me. The Fidelity Freedom 2030 is now about 60/40, but in 2010 it would have been much heavier in stocks. (The ramp down from about 80% stocks for these funds starts about 20 years out from their date, and 15 years out they’re still in the 70s). So with appreciably more stocks — during a bull market — I’d have expected it to significantly outperform the 50/50. Maybe the international allocation held it back, but the 60/40 Vanguard lifestrategy — which had more international for most of the period — did about as well. So color me surprised.

Anyway my spouse has started to indicate she’d like to “fraidy cat” our allocation a bit, and seeing this, maybe I really have no grounds for objection…
UpperNwGuy
Posts: 9479
Joined: Sun Oct 08, 2017 7:16 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by UpperNwGuy »

Phyneas wrote: Sun Nov 28, 2021 10:39 am
UpperNwGuy wrote: Sun Nov 28, 2021 6:30 am
Phyneas wrote: Sun Nov 28, 2021 4:15 am
JBTX wrote: Sun Nov 28, 2021 2:48 am I would think a 50/50 portfolio would always be better than an overweight of one and worse than an overweight of the other.

You could get scenarios where both are negative. Perhaps a period of sharply rising interest rates.
This is the problem. I've used Simba to look at 1940-1980 and 1966-1996 as two particularly difficult times, and 50/50 does well enough in both of them, whereas to optimize for one or the other you'd need to overweight stocks or bonds. So it seems to be good enough for any era (stays at or above the 4% SWR consistently).
I don't think you're using the word "overweight" correctly in this thread. When we overweight something, we are deviating from a generally-accepted equilibrium, such as market capitalization. A 50/50 portfolio is a purely arbitrary asset allocation, so 60/40 and 40/60 are not overweighted.
I should have clarified that, I meant overweight relative to my risk/baseline preference. What do you think of the basic premise though?
I think 50/50 is a good portfolio for a conservative investor who is approaching retirement or has already retired. For anyone else, I think 50/50 is just too conservative, and you will have a portfolio that grows too slowly.
Broken Man 1999
Posts: 8626
Joined: Wed Apr 08, 2015 11:31 am
Location: West coast of Florida, near Champa Bay !

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Broken Man 1999 »

bertilak wrote: Sun Nov 28, 2021 10:11 am
Phyneas wrote: Sun Nov 28, 2021 1:15 am As such, the only thing I can think of is to just go 50/50. 50% stocks (VTI+VXUS), 50% bonds.
That had been my take, for the reasons you give, which I expressed as "Whatever happens, half of my (decently sized) portfolio was invested with the best allocation possible under the circumstances (all stock or all bond) and the other half was also a decent investment."

Then I started to realize that I was in a pretty safe position:
  • Large enough portfolio
  • Low expenses (no mortgage, conservative lifestyle)
  • Steady income (pension, social security) covering all those expenses
I started to think that 50/50 was probably more conservative than need be. So I moved to 60/40.

Next, I realized that whatever safety my portfolio had was in the SIZE of my bond holdings, not their percentage of the whole, so I decided to let my equity portion grow and not rebalance into bonds. I am now at 65/35 and happy with that.
Size does matter!

I started at 50/50, then after 5 years moved to 55% equity/45% bonds, after another 5 years (2025), I plan to move to a 60% equity/40% bonds and probably go to a balanced fund of some sort. Right now I'm at 58% equity/42% bond. Before last week I was bumping up against my 5% rebalancing band, but like you, our existing bond holdings are large. I am struggling with the issue of buying more bonds the next time my rebalancing band is reached.

I might follow your action, foregoing buying additional bonds (save I-bonds with grandchildren as beneficiaries). It helps that the majority of our bonds are treasury bonds, so no worries about inferior bonds being held.

While our outside income streams are only SS benefits, they do cover about half of our regular expenses, and our portfolio distributions are very modest.

Broken Man 1999
“If I cannot drink Bourbon and smoke cigars in Heaven then I shall not go." - Mark Twain
User avatar
Forester
Posts: 2400
Joined: Sat Jan 19, 2019 1:50 pm
Location: UK

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Forester »

Phyneas wrote: Sun Nov 28, 2021 1:15 am
b) Is there anything that can be done to improve it, without guessing at the outcomes? For instance, inflation-protected only on the short end and nominals on the long end of the bonds? Or adding 5% gold, or something else mechanical and obvious?
I would add TIPS, gold and small cap stocks.

30% Global cap weight stocks
20% Global small cap
20% total bond
20% TIPS
10% gold

Ultimately this isn't so different to your simpler portfolio but I believe the returns will be more consistent by adding small cap and gold.
Amateur Self-Taught Senior Macro Strategist
sandan
Posts: 629
Joined: Wed Apr 03, 2013 12:48 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by sandan »

Thank you Nisiprius. Your write-up is very well organized and covers a lot of hot points.
ScubaHogg
Posts: 3572
Joined: Sun Nov 06, 2011 2:02 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by ScubaHogg »

Phyneas wrote: Sun Nov 28, 2021 1:15 am I've looked at the Golden Butterfly, All Seasons, and Permanent Portfolios, and while interesting, I do worry that they are essentially back-tested to perfection, and may have nothing to do with the future.
Without endorsing I’ll just say the the PP was published in ~1982. So it’s had a 38 year out of sample return to compare to its pre-1982 returns. In my mind that’s not really “back tested to perfection”

https://www.portfoliovisualizer.com/bac ... tion4_1=25
“Conventional Treasury rates are risk free only in the sense that they guarantee nominal principal. But their real rate of return is uncertain until after the fact.” -Risk Less and Prosper
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

KlangFool wrote: Sun Nov 28, 2021 10:42 am
Phyneas wrote: Sun Nov 28, 2021 10:37 am
loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
I am semi-retired at 37. I've started a new business with uneven income, but I'm at around 60x at the moment, so growth isn't the main priority, but running out of money over a potentially long retirement is the real concern.
60X with or without the business income?

KlangFool
Without. I'm self-employed in software design and general inventions, so the income is extremely extremely unreliable from the business. I'm investing assuming I may never make any money from the business, so what I already have has to last.
60% AVGE | 20 Year TIPS LMP | 5% Cash
stan1
Posts: 14246
Joined: Mon Oct 08, 2007 4:35 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by stan1 »

Phyneas wrote: Sun Nov 28, 2021 10:37 am
loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
I am semi-retired at 37. I've started a new business with uneven income, but I'm at around 60x at the moment, so growth isn't the main priority, but running out of money over a potentially long retirement is the real concern.
It would have been helpful to include this information in your first post. You are very much in an outlier situation, general advice and rules of thumb do not apply to a 37 year old who owns a business and has 60x expenses.

I think you are fine at 50/50, especially if you continue to maintain professional network connections and skills that could get you a higher paying job with health care benefits or a consulting gig if you wanted to do so. Your mitigation of risk is to earn more income.
sandan
Posts: 629
Joined: Wed Apr 03, 2013 12:48 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by sandan »

Blue456 wrote: Sun Nov 28, 2021 6:33 am
loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
That really depends. Dr. Bernstein recommends 50:50 for young person until first prolonged recession is experienced. He makes a case that higher allocation can lead to panic, stock sale and permanent loss during prolonged and severe downturn such as 2008 or Dotcom. So there is a behavioral aspect of being always half right and a loss that is limited to only half of one portfolio.
I agree, but I think its more biological/physical than behavioral (where biology can be driving behaviors). Organisms and materials exposed to large environmental changes don't do well. Personally, I think people are more sensitive than the cars parked in their garage.
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

UpperNwGuy wrote: Sun Nov 28, 2021 10:55 am
Phyneas wrote: Sun Nov 28, 2021 10:39 am
UpperNwGuy wrote: Sun Nov 28, 2021 6:30 am
Phyneas wrote: Sun Nov 28, 2021 4:15 am
JBTX wrote: Sun Nov 28, 2021 2:48 am I would think a 50/50 portfolio would always be better than an overweight of one and worse than an overweight of the other.

You could get scenarios where both are negative. Perhaps a period of sharply rising interest rates.
This is the problem. I've used Simba to look at 1940-1980 and 1966-1996 as two particularly difficult times, and 50/50 does well enough in both of them, whereas to optimize for one or the other you'd need to overweight stocks or bonds. So it seems to be good enough for any era (stays at or above the 4% SWR consistently).
I don't think you're using the word "overweight" correctly in this thread. When we overweight something, we are deviating from a generally-accepted equilibrium, such as market capitalization. A 50/50 portfolio is a purely arbitrary asset allocation, so 60/40 and 40/60 are not overweighted.
I should have clarified that, I meant overweight relative to my risk/baseline preference. What do you think of the basic premise though?
I think 50/50 is a good portfolio for a conservative investor who is approaching retirement or has already retired. For anyone else, I think 50/50 is just too conservative, and you will have a portfolio that grows too slowly.
I am both conservative and already retired-ish, but I'm on the younger side (37), so the concern about a longer retirement is there with a too-small equity allocation.
60% AVGE | 20 Year TIPS LMP | 5% Cash
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

ScubaHogg wrote: Sun Nov 28, 2021 11:17 am
Phyneas wrote: Sun Nov 28, 2021 1:15 am I've looked at the Golden Butterfly, All Seasons, and Permanent Portfolios, and while interesting, I do worry that they are essentially back-tested to perfection, and may have nothing to do with the future.
Without endorsing I’ll just say the the PP was published in ~1982. So it’s had a 38 year out of sample return to compare to its pre-1982 returns. In my mind that’s not really “back tested to perfection”

https://www.portfoliovisualizer.com/bac ... tion4_1=25
You're right, that's a fair point about the PP, and it has stood up well since 1982 as well.
60% AVGE | 20 Year TIPS LMP | 5% Cash
stan1
Posts: 14246
Joined: Mon Oct 08, 2007 4:35 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by stan1 »

stan1 wrote: Sun Nov 28, 2021 11:24 am
Phyneas wrote: Sun Nov 28, 2021 10:37 am
loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
I am semi-retired at 37. I've started a new business with uneven income, but I'm at around 60x at the moment, so growth isn't the main priority, but running out of money over a potentially long retirement is the real concern.
It would have been helpful to include this information in your first post. You are very much in an outlier situation, general advice and rules of thumb do not apply to a 37 year old who owns a business and has 60x expenses.

I think you are fine at 50/50, especially if you continue to maintain professional network connections and skills that could get you a higher paying job with health care benefits or a consulting gig if you wanted to do so. Your mitigation of risk is to earn more income.
Adding that $6M with $100K/year expenses is quite different than $600K with $10K expenses.
Again you are in a very personal situation and that's why it is called "personal finance". Generic advice and rules of thumb like an article on SWRs don't apply to you.

If you want some possibly better advice I'd recommend posting your portfolio using the forum template.
viewtopic.php?t=6212
Last edited by stan1 on Sun Nov 28, 2021 11:45 am, edited 1 time in total.
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Marseille07 »

Phyneas wrote: Sun Nov 28, 2021 1:15 am a) Is there anything obviously wrong with this approach, something glaringly obvious in my own blind spot, and;

b) Is there anything that can be done to improve it, without guessing at the outcomes? For instance, inflation-protected only on the short end and nominals on the long end of the bonds? Or adding 5% gold, or something else mechanical and obvious?

I figure that if you go 50/50 you can only ever be half wrong, but is this itself wrong? Can a 50/50 actually be much worse than picking an overweight to bonds or stocks?
There's nothing wrong with 50/50 but it's impossible to answer without knowing your withdrawal plans and what not.

If you walk at the age of 37 and do 4% SWR constant-dollar, 50/50 is probably not great because your equities portion is too small for a long retirement horizon.
User avatar
nisiprius
Advisory Board
Posts: 52211
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: Under what conditions is a 50/50 portfolio worse than any other?

Post by nisiprius »

ScubaHogg wrote: Sun Nov 28, 2021 11:17 am
Phyneas wrote: Sun Nov 28, 2021 1:15 am I've looked at the Golden Butterfly, All Seasons, and Permanent Portfolios, and while interesting, I do worry that they are essentially back-tested to perfection, and may have nothing to do with the future.
Without endorsing I’ll just say the the PP was published in ~1982. So it’s had a 38 year out of sample return to compare to its pre-1982 returns. In my mind that’s not really “back tested to perfection”

https://www.portfoliovisualizer.com/bac ... tion4_1=25
Except that is NOT the Permanent Portfolio that was published in 1982. The one that was published in 1982 was embodied in the Permanent Portfolio mutual fund, PRPFX, and managed by real people making real investments using real money. Harry Browne personally was one of the advisors. Yes, of course, it has an expense ratio, currently 0.83% but higher in past.

Still, PRPFX is the real out-of-sample test.

I don't know how to plot it together with an "Asset Class" backtest in PortfolioVisualizer, I guess I can actually superimpose the two charts graphically. I tinted the mutual fund reddish.

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.
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

Marseille07 wrote: Sun Nov 28, 2021 11:44 am
Phyneas wrote: Sun Nov 28, 2021 1:15 am a) Is there anything obviously wrong with this approach, something glaringly obvious in my own blind spot, and;

b) Is there anything that can be done to improve it, without guessing at the outcomes? For instance, inflation-protected only on the short end and nominals on the long end of the bonds? Or adding 5% gold, or something else mechanical and obvious?

I figure that if you go 50/50 you can only ever be half wrong, but is this itself wrong? Can a 50/50 actually be much worse than picking an overweight to bonds or stocks?
There's nothing wrong with 50/50 but it's impossible to answer without knowing your withdrawal plans and what not.

If you walk at the age of 37 and do 4% SWR constant-dollar, 50/50 is probably not great because your equities portion is too small for a long retirement horizon.
I am worried about that. I'm unmarried, no kids, no major health problems yet (though I do anticipate them), LCOL area and lifestyle. I think 50% equities will be just enough, but I know I'd sleep better in the long-run (not the short-run) with a higher allocation. I just have a tough time talking myself into it as I'm conservative with money by nature. But I do know the risk of not having enough is more important than the risk of feeling bad in a big down-turn.
60% AVGE | 20 Year TIPS LMP | 5% Cash
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

nisiprius wrote: Sun Nov 28, 2021 11:48 am
ScubaHogg wrote: Sun Nov 28, 2021 11:17 am
Phyneas wrote: Sun Nov 28, 2021 1:15 am I've looked at the Golden Butterfly, All Seasons, and Permanent Portfolios, and while interesting, I do worry that they are essentially back-tested to perfection, and may have nothing to do with the future.
Without endorsing I’ll just say the the PP was published in ~1982. So it’s had a 38 year out of sample return to compare to its pre-1982 returns. In my mind that’s not really “back tested to perfection”

https://www.portfoliovisualizer.com/bac ... tion4_1=25
Except that is NOT the Permanent Portfolio that was published in 1982. The one that was published in 1982 was embodied in the Permanent Portfolio mutual fund, PRPFX, and managed by real people making real investments using real money. Harry Browne personally was one of the advisors. Yes, of course, it has an expense ratio, currently 0.83% but higher in past.

Still, PRPFX is the real out-of-sample test.

I don't know how to plot it together with an "Asset Class" backtest in PortfolioVisualizer, I guess I can actually superimpose the two charts graphically. I tinted the mutual fund reddish.

Image
Doesn't PRPFX stray from the PP allocations though? It's talked about quite a bit on gyroscopicinvesting, how it doesn't reflect the PP as Harry Browne designed it.
60% AVGE | 20 Year TIPS LMP | 5% Cash
Marseille07
Posts: 16054
Joined: Fri Nov 06, 2020 12:41 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Marseille07 »

Phyneas wrote: Sun Nov 28, 2021 11:50 am I am worried about that. I'm unmarried, no kids, no major health problems yet (though I do anticipate them), LCOL area and lifestyle. I think 50% equities will be just enough, but I know I'd sleep better in the long-run (not the short-run) with a higher allocation. I just have a tough time talking myself into it as I'm conservative with money by nature. But I do know the risk of not having enough is more important than the risk of feeling bad in a big down-turn.
OK. I mean if you are 60x with plenty of buffer already included then 50/50 should be just fine, since your withdrawal rate would be something like 1.7%.
Mr.RegPark
Posts: 149
Joined: Sat Nov 14, 2020 3:01 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Mr.RegPark »

After perusing these posts, I see what an outlier I am
Just turned 66, and am at 80/20, don’t rebalance, and will likely exit this planet at 90/10. The portfolio size/ expenses ratio has to influence appropriate allocation however- my bond ladder alone is about 35x, so i guess I can afford the added risk
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by dbr »

I am struggling to imagine under what conditions or for what purpose a 50/50 allocation is worse than both more aggressive and less aggressive portfolios at the same time.
dbr
Posts: 46181
Joined: Sun Mar 04, 2007 8:50 am

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by dbr »

I am struggling to imagine under what conditions or for what purpose a 50/50 allocation is worse than both more aggressive and less aggressive portfolios at the same time.
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

dbr wrote: Sun Nov 28, 2021 12:26 pm I am struggling to imagine under what conditions or for what purpose a 50/50 allocation is worse than both more aggressive and less aggressive portfolios at the same time.
It does sound a bit silly when you say it like that, but i had to ask because I always miss something obvious whenever I design things, and I didn't want to miss something obvious here.
60% AVGE | 20 Year TIPS LMP | 5% Cash
DB2
Posts: 1396
Joined: Thu Jan 17, 2019 9:07 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by DB2 »

I think what has made 50/50 work pretty well is, the relatively low volatility which makes it a lot easier for one to stay the course. The drawdowns in 2008 and 2020 were not that bad at 50/50. Compare this to someone who is 100% equity or even 80/20 who cannot stomach major drawdowns (when push comes to shove) and starts market timing unsuccessfully.
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

Marseille07 wrote: Sun Nov 28, 2021 11:59 am
Phyneas wrote: Sun Nov 28, 2021 11:50 am I am worried about that. I'm unmarried, no kids, no major health problems yet (though I do anticipate them), LCOL area and lifestyle. I think 50% equities will be just enough, but I know I'd sleep better in the long-run (not the short-run) with a higher allocation. I just have a tough time talking myself into it as I'm conservative with money by nature. But I do know the risk of not having enough is more important than the risk of feeling bad in a big down-turn.
OK. I mean if you are 60x with plenty of buffer already included then 50/50 should be just fine, since your withdrawal rate would be something like 1.7%.
I ran the numbers initially through an inflation-adjusted growth and expenses chart with long-term 3.5% inflation and 3% real growth and it came out to about 47/53 to hit 0 in the terminal year, so it is near the redline if you assume return worse than anything we've seen in the US over a 50 year span. I tried to design it for the worst plausible case scenario (excluding the collapse of the dollar or something), so that part went into it as well.
60% AVGE | 20 Year TIPS LMP | 5% Cash
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

DB2 wrote: Sun Nov 28, 2021 12:31 pm I think what has made 50/50 work pretty well is, the relatively low volatility which makes it a lot easier for one to stay the course. The drawdowns in 2008 and 2020 were not that bad at 50/50. Compare this to someone who is 100% equity or even 80/20 who cannot stomach major drawdowns (when push comes to shove) and starts market timing unsuccessfully.
Staying in that 45/55->55/45 window seemed to provide the sweet spot in terms of performance vs draw-downs over various periods, which is a big selling point for me, because I'm that kind of investor you mention :)
60% AVGE | 20 Year TIPS LMP | 5% Cash
DB2
Posts: 1396
Joined: Thu Jan 17, 2019 9:07 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by DB2 »

Phyneas wrote: Sun Nov 28, 2021 12:41 pm
DB2 wrote: Sun Nov 28, 2021 12:31 pm I think what has made 50/50 work pretty well is, the relatively low volatility which makes it a lot easier for one to stay the course. The drawdowns in 2008 and 2020 were not that bad at 50/50. Compare this to someone who is 100% equity or even 80/20 who cannot stomach major drawdowns (when push comes to shove) and starts market timing unsuccessfully.
Staying in that 45/55->55/45 window seemed to provide the sweet spot in terms of performance vs draw-downs over various periods, which is a big selling point for me, because I'm that kind of investor you mention :)
I'm not much different. :happy But all you have to do is go back to late 2008 and early 2009 posts, and so many people realizing they truly overestimated their risk tolerance. Many sold off.
mmcmonster
Posts: 618
Joined: Sun Jan 12, 2014 12:18 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by mmcmonster »

Phyneas wrote: Sun Nov 28, 2021 10:34 am
mmcmonster wrote: Sun Nov 28, 2021 6:10 am My concern about 50/50 portfolios (25% treasuries, 25% TIPS) is that the 60/40 was devised in a period where treasuries had much better returns than current and much better returns than current forecasts for the future.

If you are not dependent on income from the bonds half of your portfolio, that's totally fine... but I don't see a 4% SWR with 60/40 (let alone 50/50) as something we should target. It may work out, but I'm not confident.
You feel that a higher allocation to equities is necessary? I have the ability to go up to 70% equities before my SWR threshold is threatened in a big and prolonged downturn - something like that?

Regardless of the equity level, do you think that 50/50 TIPS/nominal is a sensible long-term position for bonds? I know that if you're going higher equities you're supposed to go longer bonds, but I'm not comfortable holding most of my bonds at a long duration under the present very-1940'ish situation.
I think that if you're worried about the allocation of the Bonds in your portfolio, you're in a good place. :D

Don't get me wrong. I've been messing with my bond allocation all year. I currently have 20% state muni, 44% total muni, 17% Total Bond, 7% treasury, and 12% TIPS.

I'm going to increase my TIPS to ~30-40% by next Spring. But I figure this is something I'll do as I'm in the process of winning the game.

The exact allocation of my Bonds doesn't matter too much, so long as overall it's something that can blunt any drops in my total portfolio when the Stocks portion takes a dive.
User avatar
Beensabu
Posts: 5657
Joined: Sun Aug 14, 2016 3:22 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Beensabu »

loukycpa wrote: Sun Nov 28, 2021 7:58 am
Blue456 wrote: Sun Nov 28, 2021 6:33 am
loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
That really depends. Dr. Bernstein recommends 50:50 for young person until first prolonged recession is experienced. He makes a case that higher allocation can lead to panic, stock sale and permanent loss during prolonged and severe downturn such as 2008 or Dotcom. So there is a behavioral aspect of being always half right and a loss that is limited to only half of one portfolio.
Ok sure if you want to consider the behavioral aspect. If you have a high proclivity to buy high and sell low then any substantial allocation to equities could be a problem. This person should really be steered toward a target date fund.

From an optimal portfolio perspective, with negative real interest rates today on bonds, if 50/50 turns out to be the optimum allocation for a young accumulator over the next 30 years it is going to be rough sledding.
Actually, consistently contributing at 50/50 and not rebalancing has had a CAGR very close to a 100/0 portfolio while cutting max drawdown by 1/3 and decreasing time to recovery. It was like having a 66/34 portfolio and rebalancing annually. The equity allocation rises over the course of bull markets and resets during crashes. And all you have to do is set automatic contributions to 50/50. It's kind of an interesting method for those accumulators who know they're prone to behavioral error: contribute at a more moderate AA than you would normally and forget about rebalancing on either end because the market will do it for you.
"The only thing that makes life possible is permanent, intolerable uncertainty; not knowing what comes next." ~Ursula LeGuin
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

nisiprius wrote: Sun Nov 28, 2021 7:17 am
50% stocks (VTI+VXUS), 50% bonds. The bonds are there for ballast and hopefully some anti-correlation, not income, so they are treasuries only, split 50/50 between long and short, The long and short are split between nominals and inflation-protected.
1) The portfolio is fine. I like it because, well, it's not wildly different from mine. Being old, retired, and conservative/risk-averse/fraidycat I have less than 50% in stocks. Always having been worried about inflation I have more than 50% of my bonds in TIPS and I bonds. Both my nominal bond fund (Total Bond) and TIPS fund (Vanguard Inflation-Protected Securities) have a mix of terms--long, short, and intermediate--and I haven't and won't bother to figure out what the percentages are.

I don't know what you mean by "worse than any other." I think that for many purposes, your portfolio is as good as any normal, traditional, centrist retirement savings portfolio. Many will argue that yours is worse than their favored portfolio but I don't think that's what you're asking. I think there's a swarm of portfolios that are all "good," of which you simply cannot prove that any of them is any better than any other. Thirty years from now you will know which did best, but you still won't know if it was "the best portfolio" or whether it was just the luck and the breaks of that time period.

John C. Bogle wrote
Successful investing involves doing just a few things right and avoiding serious mistakes.
I can't think of a plausible scenario in which your portfolio would do so much worse than any major target-date fund that you would call it "a serious mistake."

2) 50/50 is fine, but your reasoning for 50/50 is flawed. The problem is that all equal splits depend on how you've chosen to define the boundaries and categories you're splitting between. The assumptions you're hoping to avoid, cutting the Gordian knot with a 50/50 sword, just get smuggled in when you ask "splitting between what." You're kidding yourself when you think you've achieved some objective goal with an "equal" split between subjectively defined categories. We can see this in your own example. Why not ⅓ each stocks, nominal bonds, and TIPS? Why not ⅕ each stocks, long nominal, short nominal, long TIPS, and short TIPS? Why not a tenth each large-cap growth, large-cap blend, large-cap value, mid-cap growth, mid-cap blend, mid-cap value, small-cap growth, small-cap blend, small-cap value, and bonds? Why not a fifth each stocks, bonds, rental income property, commodity futures, and collectible gold coins? Oh, sure, I personally think stocks and bonds are almost natural categories (but preferred stocks are bond-like and junk bonds are stock-like and are REITs "just" stocks?)

By the way, "I don't know so I'll put equal amounts in each" has a name, "naïve diversification."

3) I think cap-weighting is a much more "natural" division than equal-weighting. However, I'm not so sure that it applies if the assets in question don't trade freely and frictionlessly against each other within a single unified market. Anyway, look up the thread on Bill Sharpe's preferred portfolio for arguments that the natural, prediction-free, viewpoint-free division is according to the total capitalization of stocks and bonds. Fortunately for your view, that turns out not to be wildly different from 50/50. Something close to 40/60, I think.

4) I can cite three major authorities who have said something nice about 50/50 naïve diversification between stocks and bonds. (I'll leave it as an exercise to look them up if you don't know who these people were).

Benjamin Graham:
We have suggested as a fundamental guiding rule that the investor should never have less than 25% or more than 75% of his funds in common stocks, with a consequent inverse range of between 75% and 25% in bonds. There is an implication here that the standard division should be an equal one, or 50–50, between the two major investment mediums.
Harry Markowitz:
I should have computed the historical co-variances of the asset classes and drawn an efficient frontier. Instead, I visualized my grief if the stock market went way up and I wasn’t in it–or if it went way down and I was completely in it. My intention was to minimize my future regret. So I split my contributions 50/50 between bonds and equities.
John C. Bogle:
There are an infinite number of strategies worse than this one: Commit, over a period of a few years, half of your assets to a stock index fund and half to a bond index fund. Ignore interim fluctuations in their net asset values. Hold your positions for as long as you live, subject only to infrequent and marginal adjustments as your circumstances change. When there are multiple solutions to a problem, choose the simplest one.
5) It's reasonable to expect very low correlation between stocks and Treasury securities. This doesn't come solely from pragmatic observation, but also from the different fundamental nature of equity and debt.

6) There's spirited debate about this but I, for one, think it is totally unsound to rely, in any important way, on anticorrelation. It drives me bananas that you so often hear short sound-bite "of course, take-it-for-granted" asides saying that the two "have" negative correlation. The correct statement is that they "have had" negative correlation for twenty years. Following thirty-five years of positive correlation, which followed a period of negative correlation, and so on. My sound bite is "zero correlation with big fluctuations." Of course the sophisticates think they know the reasons for those fluctuations and think they make predictions based on them. You should keep the long-term picture of stock-bond correlation in your mind just the way you keep the long-term picture of stock performance in your mind:

Image

7) You need to define much more carefully what you mean by "better" and "worse."

8) The difference between 60/40 and 50/50 is negligible. That's why Vanguard thinks they cover the range adequately with only four LifeStrategy funds, allocated 80/20, 60/40, 40/60, and 20/80.

9) Obviously in any period of time when stocks are outperforming bonds, which is usually but not always, 50/50 will underperform portfolios with more stocks, and outperform portfolios with less stocks. However, if you use almost any measure that takes risk into account in any way, the picture changes. Oddly in view of the cultish adherence to 60/40, over most long periods of time, the highest risk-adjusted return as measured by the Sharpe ratio has been much lower than either 60/40 or 50/50. It has been more in the ballpark of 25/75. That is one of the fundamental bases of "risk parity" strategies. It is way more complicated than that, and involves belief systems about specific asset classes, and I don't personally like it, and the real-world performance of "risk parity" funds has included stumbles so bad that AQR abandoned the strategy in their own main risk parity fund... but anyway.

10) Bogleheads place a lot of weight on Bogle's exhortation, "Stay the course." You need to have enough conviction in your portfolio to stick to it. Virtually all presentations of strategies involve backtesting that assumes investors who did stay the course. You may not get the expected results from a backtested portfolio that you follow, but you sure as heck won't get the expected results if you don't follow. A long-term strategy that's revised frequently isn't a long-term strategy.
A lot to respond to here - first of all, thank you for the really detailed and thoughtful response:

1. Thirty years from now I want to say that I avoided any real huge mistakes, and that I'm still as comfortable going forwards as I am today. I'm not trying to increase my portfolio as much as humanly possible, but nor do I want to risk running out of money later on. I just want to avoid a really bad portfolio design that either motivates me to behavioral mistakes, or has some inherent flaw in it. If this has neither quality, and it has enough growth in it for the long-term while not being more risky than I'm comfortable with, then it's the right one for me. So far as I can tell, 50/50 accomplishes all of that for my circumstances.

2. I talked about this a bit in my reply to your other post, but I'm not so much seeking to split my chances of being right, as knowing that I'm not guessing at all, and thus won't be tempted to ever second-guess myself (literally).

3. I remember reading through at least one thread on here that deals with the Global Market Portfolio, which is along the same lines as what you're discussing. I think there is also a live link somewhere on bogleheads.org that shows the current allocation of it on a Google Sheet I think - ah yes, here it is. Thank you for the thread link for the Sharpe portfolio though, I'll take a look through it, you always learn something from other portfolio alternatives.

4. I know all three quotes by heart :)

5/6. I agree, and it's one of the major reasons why I stayed away from the All Seasons Portfolio or any version of risk parity - you're taking a 100% concentration risk on a single strategy to make your portfolio work. It's also why I stayed away from all LTTs. You never know when the correlations will go against you, even marginally, hence the preference for at least half in STTs. I've read Bernstein's take that if your worried about inflation vs deflation, inflation is far more likely, so you should be all STTs, and I don't necessarily disagree with it, but again, I don't want to guess either way. I also noticed in Simba that TBM, despite holding a lot of corporates and so on, tends to do pretty well in every environment, which reinforces the idea, at least to me, that diversification in bonds, even if done imprecisely, is just as important as in stocks, even where correlations don't appear to always be on your side.

7. I struggled with how exactly to describe it. I'm looking for outliers in how the components work together, something obvious to others that I've missed about how long and short bonds work together, or nominal bonds and TIPS. The concern is a false sense of safety in splitting the difference between Portfolio A that does well in Scenario 1, and Portfolio B that does well in Scenario 2, and yet the 50/50 does well in neither of them for some reason that isn't obvious at first sight. For instance, maybe a 30/70 portfolio is always better than a 50/50 for some reason I wouldn't think of. I didn't think it was a problem, and I still don't, but I wanted to ask just in-case.

8. I'd disagree with this a bit on the basis that I've noticed in Simba, for whatever it's worth, that tipping over the edge of allocations can result in significant differences in performance past thresholds for some reason. Usually it happens at the very edges (a 0/100 vs 10/90 portfolio), but I've seen it a few times going from a 20/80 to a 30/70 or a 50/50 to a 60/40. But it may not really be an issue over longer time periods, I was looking at very specific and in some cases short-er spans.

9. It's really interesting how often that 25/75 or 30/70 portfolio comes up the winner - I noticed that too in Simba. I didn't feel it would work for me though on account of the large bond holding and the consequent lower real growth. I'm looking at a longer retirement than most, so I can't afford taking any risks running out of money early, especially right at the 'usual' retirement age when I couldn't go back to work even if I had to. I should have mentioned in my original post that i am 37 though, that was an oversight that would have provided a lot more context to my question.

10. This is a paramount concern for me, the moreso because I already have experience with strategy-churn and I know how unproductive it is.
60% AVGE | 20 Year TIPS LMP | 5% Cash
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

DB2 wrote: Sun Nov 28, 2021 12:43 pm
Phyneas wrote: Sun Nov 28, 2021 12:41 pm
DB2 wrote: Sun Nov 28, 2021 12:31 pm I think what has made 50/50 work pretty well is, the relatively low volatility which makes it a lot easier for one to stay the course. The drawdowns in 2008 and 2020 were not that bad at 50/50. Compare this to someone who is 100% equity or even 80/20 who cannot stomach major drawdowns (when push comes to shove) and starts market timing unsuccessfully.
Staying in that 45/55->55/45 window seemed to provide the sweet spot in terms of performance vs draw-downs over various periods, which is a big selling point for me, because I'm that kind of investor you mention :)
I'm not much different. :happy But all you have to do is go back to late 2008 and early 2009 posts, and so many people realizing they truly overestimated their risk tolerance. Many sold off.
I really do wonder about that. The conditions are such (I think this was talked about quite a bit in the Hi I'm High Inflation and Jeremy Siegel posts) that almost everyone is being put between a rock and hard place, and a lot of people are opting for more equity exposure than they are comfortable with because TINA. It's not the first time people have put more into equities than they ought to have (for different reasons), but that doesn't make it any better. I know I'm giving up returns, but I also know I'd lose more money in the long run burning capital making a mistake during a crash. And if the markets keep going up, I still have enough in the game to benefit from it, so those are the trade-offs, and you just live with it :)
60% AVGE | 20 Year TIPS LMP | 5% Cash
djm2001
Posts: 382
Joined: Fri May 29, 2020 6:23 am

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by djm2001 »

nisiprius wrote: Sun Nov 28, 2021 7:17 am 3) I think cap-weighting is a much more "natural" division than equal-weighting. However, I'm not so sure that it applies if the assets in question don't trade freely and frictionlessly against each other within a single unified market. Anyway, look up the thread on Bill Sharpe's preferred portfolio for arguments that the natural, prediction-free, viewpoint-free division is according to the total capitalization of stocks and bonds. Fortunately for your view, that turns out not to be wildly different from 50/50. Something close to 40/60, I think.
It does Sharpe's portfolio a disservice to advertise it as 40/60 when in reality it's roughly 60/40. Sharpe's methodology is outlined here.

Other than that, great post.
AA = global stocks & bonds @ market weight (~60/40); EF = i-bonds; WR = -PMT(1%, 100-age, 1, 0, 1)
Topic Author
Phyneas
Posts: 336
Joined: Tue Apr 27, 2021 9:10 pm

Re: Under what conditions is a 50/50 portfolio worse than any other?

Post by Phyneas »

Beensabu wrote: Sun Nov 28, 2021 1:13 pm
loukycpa wrote: Sun Nov 28, 2021 7:58 am
Blue456 wrote: Sun Nov 28, 2021 6:33 am
loukycpa wrote: Sun Nov 28, 2021 6:29 am A younger person in the accumulation phase would not be well served by this portfolio. For a retired person sure.
That really depends. Dr. Bernstein recommends 50:50 for young person until first prolonged recession is experienced. He makes a case that higher allocation can lead to panic, stock sale and permanent loss during prolonged and severe downturn such as 2008 or Dotcom. So there is a behavioral aspect of being always half right and a loss that is limited to only half of one portfolio.
Ok sure if you want to consider the behavioral aspect. If you have a high proclivity to buy high and sell low then any substantial allocation to equities could be a problem. This person should really be steered toward a target date fund.

From an optimal portfolio perspective, with negative real interest rates today on bonds, if 50/50 turns out to be the optimum allocation for a young accumulator over the next 30 years it is going to be rough sledding.
Actually, consistently contributing at 50/50 and not rebalancing has had a CAGR very close to a 100/0 portfolio while cutting max drawdown by 1/3 and decreasing time to recovery. It was like having a 66/34 portfolio and rebalancing annually. The equity allocation rises over the course of bull markets and resets during crashes. And all you have to do is set automatic contributions to 50/50. It's kind of an interesting method for those accumulators who know they're prone to behavioral error: contribute at a more moderate AA than you would normally and forget about rebalancing on either end because the market will do it for you.
I had no idea about that - the no re-balancing part, that's really weird, and interesting, that it works out that way, but that definitely is a big point in it's favor - thank you!
60% AVGE | 20 Year TIPS LMP | 5% Cash
Post Reply