Yes, you can diversify a market portfolio

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
jalbert
Posts: 3588
Joined: Fri Apr 10, 2015 12:29 am

Re: Yes, you can diversify a market portfolio

Post by jalbert » Sat Jun 09, 2018 10:21 pm

I find corrections a great time to unload any under-performers and trade into the best sectors at the given time.
If I understand what you are saying it is that you are buying them after they overperform, and selling after they underperform.

Yes, at any point in time some active funds will beat the market. It is usually about 20% of active funds in any given year, per the SPIVA scorecard. The problem is that there is no evidence that someone can pick the winners ahead of time.
Index fund investor since 1987.

pkcrafter
Posts: 13018
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Yes, you can diversify a market portfolio

Post by pkcrafter » Sat Jun 09, 2018 10:48 pm

Here is another article on the diversification ratio (DR) that may help readers understand what vineviz is promoting.

http://www.allaboutalpha.com/blog/2011/ ... portfolio/

Another article for perspective. Comparison of max diversification, minimum variance, tangent portfolio and equal-weighted.

https://www.tandfonline.com/doi/pdf/10. ... 18.1427533


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.

md&pharmacist
Posts: 262
Joined: Fri Mar 23, 2018 7:05 pm

Re: Yes, you can diversify a market portfolio

Post by md&pharmacist » Sat Jun 09, 2018 11:04 pm

jalbert wrote:
Sat Jun 09, 2018 10:21 pm
I find corrections a great time to unload any under-performers and trade into the best sectors at the given time.
If I understand what you are saying it is that you are buying them after they overperform, and selling after they underperform.

Yes, at any point in time some active funds will beat the market. It is usually about 20% of active funds in any given year, per the SPIVA scorecard. The problem is that there is no evidence that someone can pick the winners ahead of time.
Not sure what you mean by "there is no evidence that someone can pick the winners ahead of time"? Haven't the ones already in the winners picked the winners ahead of time?

My current portfolio of funds, YTD/1 YR, worst to best (%): -3/10, 2.5/11, 4.8/14 (Index Fund), 5.4/15, 5.3/22.5, 9.5/27, 14.5/22.5, 14.5/30, 18/53, 22.4/40.

The only reason I have any funds performing less than 20% 1 yr returns are for diversification purposes, which prove my point they don't perform as well. I can very consistently find funds that stay in the 1 yr 20-100% return range after purchase.

So when I way I want to exit under-performers, my under-performers still have a solid 10%, 11%, 14% 1 year return. Looking to get my portfolio to average a 40% annualized return. I am looking at some very health funds, some of which are outperforming my best funds.

jalbert
Posts: 3588
Joined: Fri Apr 10, 2015 12:29 am

Re: Yes, you can diversify a market portfolio

Post by jalbert » Sun Jun 10, 2018 12:44 am

Not sure what you mean by "there is no evidence that someone can pick the winners ahead of time"? Haven't the ones already in the winners picked the winners ahead of time?
20% of active funds outperform in a given year. I’ve seen no evidence that it is possible for someone to select which funds will be the winners a priori. If you are able to do that, you are a world class analyzer of fund managers.

It is possible to pick funds with a higher expected return than the market by taking more risk.
Index fund investor since 1987.

md&pharmacist
Posts: 262
Joined: Fri Mar 23, 2018 7:05 pm

Re: Yes, you can diversify a market portfolio

Post by md&pharmacist » Sun Jun 10, 2018 12:57 am

jalbert wrote:
Sun Jun 10, 2018 12:44 am
Not sure what you mean by "there is no evidence that someone can pick the winners ahead of time"? Haven't the ones already in the winners picked the winners ahead of time?
20% of active funds outperform in a given year. I’ve seen no evidence that it is possible for someone to select which funds will be the winners a priori. If you are able to do that, you are a world class analyzer of fund managers.

It is possible to pick funds with a higher expected return than the market by taking more risk.
I just started a string here where I will be posting my moves. If you follow it you will see I am either consistently extremely lucky or know how to do what possibly almost no one can. I can understand pessimism - that's why I started a string with specific purchase dates so you can follow the trade out 6 months, 1 year, 3 year, etc. You will see how it performs AFTER I buy it.

Generally yes, better returns come with higher risk although that risk can be mitigated by educated timing (buy/sell). That's where my trade dates are important.

User avatar
Eric
Posts: 701
Joined: Fri Feb 23, 2007 5:44 pm
Location: Texas

Re: Yes, you can diversify a market portfolio

Post by Eric » Sun Jun 10, 2018 3:30 am

For investors of a certain age, here's a real-world case where volatility was a poor measure of risk.

Years ago, Vanguard juiced the returns of its Total Bond Market fund through "corporate substitution" -- overweighting corporate bonds relative to the index. For a long time, this produced slightly higher returns for the same amount of risk as measured by volatility.

Then one year some corporate bonds blew up and the fund lagged its index by 200 basis points, which is huge for a bond fund.

I think of this whenever someone says they overweight small value stocks in their equity portfolio. This seems to me to be a variation on the corporate-substiution game. It may produce higher risk-adjusted returns for a long time, if risk is measured by volatility. But if small and value are real risk factors, in some sense other than just their impact on normal volatility, then this strategy is not optimal for the average investor (using "average" here to mean average by characteristics rather than average by skill).

Random Walker
Posts: 2997
Joined: Fri Feb 23, 2007 8:21 pm

Re: Yes, you can diversify a market portfolio

Post by Random Walker » Sun Jun 10, 2018 9:56 am

Eric wrote:
Sun Jun 10, 2018 3:30 am
I think of this whenever someone says they overweight small value stocks in their equity portfolio. This seems to me to be a variation on the corporate-substiution game. It may produce higher risk-adjusted returns for a long time, if risk is measured by volatility. But if small and value are real risk factors, in some sense other than just their impact on normal volatility, then this strategy is not optimal for the average investor (using "average" here to mean average by characteristics rather than average by skill).
But, as you point out, size and value are independent and unique risk factors from market beta. The average investor should want to be diversified, and diversifying across these weakly or uncorrelated sources of return should be productive. The return, volatility, max drawdown, Sharpe ratio, of the portfolio as a whole is what matters.

Dave

User avatar
nisiprius
Advisory Board
Posts: 36492
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: Yes, you can diversify a market portfolio

Post by nisiprius » Sun Jun 10, 2018 10:09 am

vineviz wrote:
Fri Jun 08, 2018 3:35 pm
...Your calculations are definitely incorrect...
I've double-checked them. I made one ludicrous but inconsequential error: in the definition of the diversification ratio I got the numerator and denominator swapped, because I was thinking in terms of "risk reduction." The ratio is sensibly defined so that a bigger ratio means more diversification. It doesn't matter because I was minimizing the reciprocal, and minimizing (1 / diversification ratio) is the same thing as maximizing the diversification.

I stand by my calculations, which are very close to those made by PortfolioVisualizer. They are not identical, but very close. The differences are probably due to a) my using annual numbers while PortfolioVisualizer is using monthly numbers, and b) I used 1985 through 2017 and PortfolioVisualizer only goes back to 1987.

I got 43/57 for tangency, 41/59 for maximum diversification, using annual data from 1985 through 2017.

PortfolioVisualizer, using slightly different data, is showing

41/59 for tangency ("maximize Sharpe ratio"), 40/60 for maximum diversification, using monthly data from 1987 through 2017.

I know you don't want to discuss this further in this thread but I did think I should check my math and acknowledge any mistake.

I also stand by my conceptual understanding that MPT/CAPM are looking for the point where a tangent drawn through the riskless asset touches the efficient frontier curve, and that maximum diversification is looking for the widest spread (measured as a ratio, not a distance) between the efficient frontier curve and a straight line jointing the two risky assets.
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
Eric
Posts: 701
Joined: Fri Feb 23, 2007 5:44 pm
Location: Texas

Re: Yes, you can diversify a market portfolio

Post by Eric » Sun Jun 10, 2018 12:20 pm

Random Walker wrote:
Sun Jun 10, 2018 9:56 am
But, as you point out, size and value are independent and unique risk factors from market beta.
You could say something similar in my bond market example: Corporate credit risk is an independent and unique risk factor from beta or term risk. Adding more corporate credit risk (relative to the market) appeared to increase risk-adjusted returns for a while, but only because volatility, Sharpe ratio, etc. did not properly reflect the true risk of the portfolio.

User avatar
vineviz
Posts: 1315
Joined: Tue May 15, 2018 1:55 pm

Re: Yes, you can diversify a market portfolio

Post by vineviz » Sun Jun 10, 2018 2:10 pm

nisiprius wrote:
Sun Jun 10, 2018 10:09 am
I've double-checked them. I made one ludicrous but inconsequential error: in the definition of the diversification ratio I got the numerator and denominator swapped, because I was thinking in terms of "risk reduction." The ratio is sensibly defined so that a bigger ratio means more diversification. It doesn't matter because I was minimizing the reciprocal, and minimizing (1 / diversification ratio) is the same thing as maximizing the diversification.
I apologize. I was thinking of another comment, and I should not have carelessly accused you of being in error.
nisiprius wrote:
Sun Jun 10, 2018 10:09 am
I also stand by my conceptual understanding that MPT/CAPM are looking for the point where a tangent drawn through the riskless asset touches the efficient frontier curve, and that maximum diversification is looking for the widest spread (measured as a ratio, not a distance) between the efficient frontier curve and a straight line jointing the two risky assets.
I think you can see, having done the calculations, that MDR is not utilizing expected returns in any way in its equations.

Your conceptual diagram is mapping the assets in two dimensional x,y space (expected return on one dimension, volatility on the other dimension), but MDR is working in just one dimension: volatility. So in the MDR case, there is no diagonal line because there is basically no y dimension.

If want to imagine MDR optimization graphically, you'd need to forget for a moment the idea that the two individual assets (call them A and B) exist as x,y points and instead imagine them as vertical lines (located at StdDev(A) and StdDev(B) respectively). Ultimately the MDR portfolio will exist on a vertical line intermediate between A and B.

To find its location, you find the weight of asset A as being the proportion of it inverse volatility to the sum of inverse volatilities for both assets. In other words w(A) = (1/StdDev(A))/((1/StdDev(A))+(1/StdDev(B)) . w(B)= 1-w(A). Use w(A) and w(B) to find the location of the vertical line representing the MDR portfolio, somewhere between line A and line B.

Now, if the assets are priced perfectly efficiently (such that their risk-adjusted returns are equal) then in a two-asset example his vertical line should intersect the efficient frontier exactly at the tangency point. If there are three or more assets, though, there is no guarantee the MDR portfolio will even BE on the efficient frontier. And if there are market inefficiencies, the MDR portfolio and the tangency portfolio can diverge (as in this illustration using SBBI small stocks and SBBI LT Govt bonds since 1930).

Image

And the specification of the Sharpe ratio portfolio is HIGHLY dependent on the accuracy and noisiness of the expected return estimation process once you start looking at more realistic mulit-asset portfolios. In historical data, observed returns and observed risk don't always conform precisely to CAPM so the two different approaches (mean-variance optimization and diversification ratio optimization) can indeed produce different portfolios. For instance:

Image

And neither one looks very much like the 60/40 portfolio that most investors might think of as "balanced".
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

jalbert
Posts: 3588
Joined: Fri Apr 10, 2015 12:29 am

Re: Yes, you can diversify a market portfolio

Post by jalbert » Sun Jun 10, 2018 2:32 pm

Generally yes, better returns come with higher risk although that risk can be mitigated by educated timing (buy/sell). That's where my trade dates are important.
Timing when to sell an active mutual fund before it starts underperforming is just as hard as selecting one to buy that will overperform. Very few investment professionals can beat the market that way.
Index fund investor since 1987.

User avatar
nisiprius
Advisory Board
Posts: 36492
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: Yes, you can diversify a market portfolio

Post by nisiprius » Sun Jun 10, 2018 3:46 pm

vineviz wrote:
Sun Jun 10, 2018 2:10 pm
...And neither one looks very much like the 60/40 portfolio that most investors might think of as "balanced"...
I have been trying for years to understand the origin of the "60/40" ratio. I have not been able to find any quantitative reasoning or theory behind it. It is not theory. It is just "wisdom" or "tradition."

The modern portfolio theory (MPT) optimum for most reasonably-long periods of time is closer to 40/60 than to 60/40.

No specifics for stocks and bonds are mentioned in the Harry Markowitz papers that are credited as the origins of MPT. However, the MPT optimum for large-cap stocks and long-term Treasury bonds from 1926-1952--the time period and asset classes most likely to have been used at that time--would have been 45/55, unlikely to have been "rounded" to 60/40.

I've occasionally posted on this and nobody's been able to come up with a satisfying explanation.

It's particularly mysterious to me, if we suppose it is just traditional wisdom, why it isn't 50/50. Which Benjamin Graham proposed, in The Intelligent Investor:
There is an implication here that the standard division should be an equal one, or 50–50, between the two major investment mediums. ... we can give the investor no reliable rules by which to reduce his common-stock holdings toward the 25% minimum and then rebuild them later to the 75% maximum. We can urge that in general the investor should not have more than one-half in equities unless he is has strong confidence in the soundness of his stock position and is sure that he could view of market decline of the 1969-70 type with equanimity.... We are thus led to put forward for most of our readers what may appear to be an oversimplified 50-50 formula...
Peter L. Bernstein wrote an article in 2002 entitled "The 60/40 Solution," in which he said
Once upon a time—that is, for many years before the great bull market of the ’90s—the most popular benchmark for portfolio asset allocation aimed at this goal was about 60 percent in stocks and 40 percent in fixed-income investments. After 1990, this seemingly stodgy arrangement largely went by the boards as stocks roared ahead, encouraging institutional and individual investors to become increasingly aggressive in their search for higher returns.... I propose restoring 60/40 to its rightful place as the center of gravity of asset allocation for long-term investors.
However, he gave no explanation of why 60/40 was "the most popular benchmark."

I wouldn't go so far as to say 60/40 is a straw man, because it was and continues to be the default "policy portfolio," but so far it continues to be a mystery to me where these numbers came from. It is certainly more "balanced" than 100% stocks, but I don't currently know of any metrics under which it is the point of balance. It does not appear to come out of MPT/CAPM, nor does it correspond to the "global market portfolio" (either for the world or for the United States alone).
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
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Yes, you can diversify a market portfolio

Post by SimpleGift » Sun Jun 10, 2018 4:23 pm

nisiprius wrote:
Sun Jun 10, 2018 3:46 pm
No specifics for stocks and bonds are mentioned in the Harry Markowitz papers that are credited as the origins of MPT.
It would certainly be better to have a definite reference by Mr. Markowitz about the origins of the 60/40 portfolio — but it's worth pointing out that his classic text, Portfolio Selection: Efficient Diversification of Investments, was first published by the Cowles Foundation in 1959. Soon thereafter in the 1960s, when investors started thinking about efficient portfolio selection, the average composition of the Global Market Portfolio was about 60/40 (chart below).
Not definitive evidence about the origins of the 60/40 portfolio, but perhaps "circumstantial evidence" — or just an historical coincidence.
Cordially, Todd

User avatar
vineviz
Posts: 1315
Joined: Tue May 15, 2018 1:55 pm

Re: Yes, you can diversify a market portfolio

Post by vineviz » Sun Jun 10, 2018 4:43 pm

SimpleGift wrote:
Sun Jun 10, 2018 4:23 pm
nisiprius wrote:
Sun Jun 10, 2018 3:46 pm
No specifics for stocks and bonds are mentioned in the Harry Markowitz papers that are credited as the origins of MPT.
It would certainly be better to have a definite reference by Mr. Markowitz about the origins of the 60/40 portfolio — but it's worth pointing out that his classic text, Portfolio Selection: Efficient Diversification of Investments, was first published by the Cowles Foundation in 1959. Soon thereafter in the 1960s, when investors started thinking about efficient portfolio selection, the average composition of the Global Market Portfolio was about 60/40 (chart below).
Not definitive evidence about the origins of the 60/40 portfolio, but perhaps "circumstantial evidence" — or just an historical coincidence.
I think that’s right.

Pre-Markowitz, many defined benefit plans held portfolios that were heavily invested in bonds. Sometimes entirely. As far as I can tell, moving to a market portfolio of 60% equity during the 1960s was a thing that pension funds did.

After ERISA in the 1970s, advisors needed a rule of thumb for their newest product for individuals, the IRA. And so investors got sold 60/40 portfolios.

By 2010, this was nowhere near a market portfolio (debt now dwarfs equity IIRC) but the rule of thumb was now canonical.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Yes, you can diversify a market portfolio

Post by SimpleGift » Sun Jun 10, 2018 5:32 pm

vineviz wrote:
Sun Jun 10, 2018 4:43 pm
I think that’s right.
One valid objection to this origin idea is that investors in the 1950s may not have known the composition of the global market portfolio — but from at least one source, it appears the U.S. market portfolio had a similar composition at that time:
Newberry Research wrote:The 60/40 portfolio appears to originate from Markowitz’s concept of the “market portfolio,” a theoretical market-capitalization-weighted portfolio of all investable assets, worldwide. In the 1950s, the U.S. equity market had a total capitalization of approximately $5T and the bond market had a total notional value of approximately $4T.
This would put the U.S. market portfolio in the 1950s at roughly 56% stocks/44% bonds — near enough to 60/40.
Cordially, Todd

User avatar
Taylor Larimore
Advisory Board
Posts: 27326
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

The 60/40 Portfolio is not for everyone

Post by Taylor Larimore » Sun Jun 10, 2018 5:57 pm

Bogleheads:

It is a serious mistake to believe that there is only one stock/bond allocation that is ideal for all of us. Each of us should base this very important decision on our goal(s), our time-frame, our risk-tolerance and our personal financial situation. We are all different.

Vanguard offers this Investor Questionnaire to help us decide.

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

columbia
Posts: 844
Joined: Tue Aug 27, 2013 5:30 am

Re: The 60/40 Portfolio is not for everyone

Post by columbia » Sun Jun 10, 2018 6:20 pm

Taylor Larimore wrote:
Sun Jun 10, 2018 5:57 pm
Bogleheads:

It is a serious mistake to believe that there is only one stock/bond allocation that is ideal for all of us. Each of us should base this very important decision on our goal(s), our time-frame, our risk-tolerance and our personal financial situation. We are all different.

Vanguard offers this Investor Questionnaire to help us decide.

Best wishes.
Taylor
Amen

At my current age, the average target date find would have me between 75-80% stocks. I just can’t stomach that and hover somewhere between 50-55%.

I sleep well at night.

User avatar
Phineas J. Whoopee
Posts: 7405
Joined: Sun Dec 18, 2011 6:18 pm

Re: Yes, you can diversify a market portfolio

Post by Phineas J. Whoopee » Sun Jun 10, 2018 9:09 pm

Is it absolutely unavoidable that we argue about the definition of the word diversify again?
PJW

Random Walker
Posts: 2997
Joined: Fri Feb 23, 2007 8:21 pm

Re: The 60/40 Portfolio is not for everyone

Post by Random Walker » Sun Jun 10, 2018 9:25 pm

Taylor Larimore wrote:
Sun Jun 10, 2018 5:57 pm
Bogleheads:

It is a serious mistake to believe that there is only one stock/bond allocation that is ideal for all of us. Each of us should base this very important decision on our goal(s), our time-frame, our risk-tolerance and our personal financial situation. We are all different.

Vanguard offers this Investor Questionnaire to help us decide.

Best wishes.
Taylor
+1, and our circumstances change over time. So for any individual, ability, willingness, and need can and will change over time. So the AA decision should be readdressed periodically. Not market timing, but looking at where one is relative to their goals and periodically adjusting if appropriate.

Dave

User avatar
spdoublebass
Posts: 454
Joined: Thu Apr 27, 2017 10:04 pm
Location: NY

Re: Yes, you can diversify a market portfolio

Post by spdoublebass » Sun Jun 10, 2018 9:28 pm

nisiprius wrote:
Sun Jun 10, 2018 3:46 pm
It does not appear to come out of MPT/CAPM, nor does it correspond to the "global market portfolio" (either for the world or for the United States alone).
I'm just a fly on the wall here reading and trying to follow along on this thread. Much is way over my head. Please do not take this comment as condescending in any way.

You don't think that the 60/40 that is often recommended and the Global Market Portfolio (that William Sharpe wrote about) has something in common? Even just coincidence? Sharpe Global portfolio is currently 57/43 at current market weights.

Again, I only ask this trying to learn.
I'm trying to think, but nothing happens

jalbert
Posts: 3588
Joined: Fri Apr 10, 2015 12:29 am

Re: Yes, you can diversify a market portfolio

Post by jalbert » Sun Jun 10, 2018 9:47 pm

Markowitz’s personal allocation:
Mr. Markowitz was then working at the RAND Corporation and trying to figure out how to allocate his retirement account. He knew what he should do: “I should have computed the historical co-variances of the asset classes and drawn an efficient frontier.” (That’s efficient-market talk for draining as much risk as possible out of his portfolio.)

But, he said, “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. So I split my contributions 50/50 between stocks and bonds.”
From: https://mobile.nytimes.com/2007/09/29/b ... ocera.html
Index fund investor since 1987.

Random Walker
Posts: 2997
Joined: Fri Feb 23, 2007 8:21 pm

Re: Yes, you can diversify a market portfolio

Post by Random Walker » Sun Jun 10, 2018 9:58 pm

jalbert wrote:
Sun Jun 10, 2018 9:47 pm
Markowitz’s personal allocation:
Mr. Markowitz was then working at the RAND Corporation and trying to figure out how to allocate his retirement account. He knew what he should do: “I should have computed the historical co-variances of the asset classes and drawn an efficient frontier.” (That’s efficient-market talk for draining as much risk as possible out of his portfolio.)

But, he said, “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. So I split my contributions 50/50 between stocks and bonds.”
From: https://mobile.nytimes.com/2007/09/29/b ... ocera.html
Great story! :-) a lot of wisdom on top of world class smarts!

Dave

aristotelian
Posts: 4609
Joined: Wed Jan 11, 2017 8:05 pm

Re: Yes, you can diversify a market portfolio

Post by aristotelian » Mon Jun 11, 2018 6:29 am

This is perhaps semantic, but OP is talking about selectively weighting assets thought to be non correlated. That is different than diversifying since market portfolio is by definition diversified. Personally I do weight long and intermediate Treasury on the bond side as well as international small cap and emerging markets on the stock side. In most circumstances, the portfolio should be less volatile overall even though the components are more risky individually. Still, the distinction is important because by weighing certain classes you are risking a black swan in which they correlate at the wrong time. The January correction is a good example of long bonds and stocks getting hammered simultaneously.

User avatar
nisiprius
Advisory Board
Posts: 36492
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: Yes, you can diversify a market portfolio

Post by nisiprius » Mon Jun 11, 2018 6:29 am

spdoublebass wrote:
Sun Jun 10, 2018 9:28 pm
nisiprius wrote:
Sun Jun 10, 2018 3:46 pm
It does not appear to come out of MPT/CAPM, nor does it correspond to the "global market portfolio" (either for the world or for the United States alone).
I'm just a fly on the wall here reading and trying to follow along on this thread. Much is way over my head. Please do not take this comment as condescending in any way.

You don't think that the 60/40 that is often recommended and the Global Market Portfolio (that William Sharpe wrote about) has something in common? Even just coincidence? Sharpe Global portfolio is currently 57/43 at current market weights.

Again, I only ask this trying to learn.
I am in much the same situation as you, spdoublebass, just for a longer time. First of all, what I personally believe is that nobody can make any reliable judgement on what the optimum overall stock allocation is going to be, going forward, to better than ±10%. Maybe not even to within ±10% but certainly no better than that. Benjamin Graham recommended fifty-fifty, that's what Markowitz has admitted to doing himself, and if 50/50 is reasonable, then I think anything in the range 50±10, i.e. 40/60 to 60/40 is reasonable. The more I learn, the less credence I give to attempts to determine fine tuning.

As you see from Simplegift's chart of Doeswijk & al's data, the global market portfolio, like all other interesting financial data, fluctuates all over the place.

But where are you seeing that the global market portfolio is currently 57/43, i.e. 57% stocks, 43% bonds? Can you provide a link?

I thought it was just about the reverse of that, as in the data Simplegift posted in his thread. I added a dotted line to his chart:

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
nisiprius
Advisory Board
Posts: 36492
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: Yes, you can diversify a market portfolio

Post by nisiprius » Mon Jun 11, 2018 6:54 am

The maximum diversification portfolio takes no account of the risk-free rate of return. I understand that it isn't trying to. But from the point of view of an investor making decisions in the real world, does it really make sense that you would ignore it? I personally remember back in 2007, I was doing some kind of rebalancing, I don't remember what, and I'd sold a largish bond mutual fund holding, so I'd just reduced my bond allocation (and thus increased my stock/bond ratio) and remembered thinking, casually, "hey, wait--if the money market settlement fund is earning 5%, why am I bothering with bonds at all?"

That is to say, when risk-free interest rates are competitive with bond returns, it seems reasonable just to take the sure thing, rather than gambling that the low-correlation independent movements of bonds against stocks are going to help more than they are going to hurt.

In the MPT/CAPM framework, if you can get higher risk-free return, the tangent point moves toward stocks. That is to say, if we wish to hold portfolio risk constant, and we look at all three traditional assets (stocks/bond/cash), if cash has a higher return, we shift out of bonds and into cash. In the extreme case, if cash is doing well enough, you don't use bonds at all--you use a stocks/cash portfolio instead of stocks/bonds. This seems like common sense. (See the three diagrams below, representing a highly exaggerated example).

In the "maximum diversification" framework, if I understand it, we continue to invest the same way, making no changes, regardless of the risk-free rate. Is that actually what advocates of the approach are advocating?

Actual data for Total Stock/Total Bond; over 1987-2017 inclusive, the actual risk-free rate was 3.2% and the tangent portfolio was 18/82.

However, if we had been able to get a risk-free return of 5.2%, the tangent portfolio changes to 44/56 (i.e. we allocate relatively more to stocks than bonds... while, of course, allocating more to cash if we wish to maintain the same risk level)

And if we had been able to get a risk-free return of 6%, it changes to 100/0. (That is, including cash as an asset and maintaining the same level, we shift entirely into cash in place of stocks).

I'm not saying that's sensible, or that I necessarily think it's better than sticking with the "most diversified portfolio" and ignoring times when the money market settlement account doing really well. A huge problem with MPT/CAPM is precisely that the tangent portfolio is very, very sensitive to small changes in the characteristics of the assets and the risk-free return.

Image
Image
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
spdoublebass
Posts: 454
Joined: Thu Apr 27, 2017 10:04 pm
Location: NY

Re: Yes, you can diversify a market portfolio

Post by spdoublebass » Mon Jun 11, 2018 9:16 am

nisiprius wrote:
Mon Jun 11, 2018 6:29 am

But where are you seeing that the global market portfolio is currently 57/43, i.e. 57% stocks, 43% bonds? Can you provide a link?
Thank you for the great reply. I gotta look over that link more closely.

What I was referring to was Bill Sharpe's market portfolio (Maybe I was getting the terms confused). From his work here: https://web.stanford.edu/~wfsharpe/RISMAT/RISMAT-7.pdf

Page 22 is when he starting breaking down the numbers. I track it just out of curiosity. So currently, it's roughly 31%US Stock, 26% EX US Stock, 22% US Bonds, 21% Int Bond.

Again, I was thinking you were referring to this when you mentioned the market portfolio.

Thanks again for your time. Great Thread.
I'm trying to think, but nothing happens

Random Walker
Posts: 2997
Joined: Fri Feb 23, 2007 8:21 pm

Re: Yes, you can diversify a market portfolio

Post by Random Walker » Mon Jun 11, 2018 9:29 am

aristotelian wrote:
Mon Jun 11, 2018 6:29 am
This is perhaps semantic, but OP is talking about selectively weighting assets thought to be non correlated. That is different than diversifying since market portfolio is by definition diversified. Personally I do weight long and intermediate Treasury on the bond side as well as international small cap and emerging markets on the stock side. In most circumstances, the portfolio should be less volatile overall even though the components are more risky individually. Still, the distinction is important because by weighing certain classes you are risking a black swan in which they correlate at the wrong time. The January correction is a good example of long bonds and stocks getting hammered simultaneously.
I don’t think it’s semantic. Just need to look at diversification differently, perhaps more broadly within an MPT framework. The most fundamental definition of diversification is “don’t put all your eggs in one basket”. Thus owning the entire stock market appears diversified. But the TSM investor is only exposed to one source of risk, market beta. If one looks at diversification a bit differently, “don’t put all your eggs in one risk basket”, then TSM seriously lacks diversification. Diversification is much more than number of stocks. It’s sources of uncorrelated risk and sources of uncorrelated return.

Dave

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Yes, you can diversify a market portfolio

Post by SimpleGift » Mon Jun 11, 2018 11:20 am

Random Walker wrote:
Mon Jun 11, 2018 9:29 am
The most fundamental definition of diversification is “don’t put all your eggs in one basket”. Thus owning the entire stock market appears diversified. But the TSM investor is only exposed to one source of risk, market beta. If one looks at diversification a bit differently, “don’t put all your eggs in one risk basket”, then TSM seriously lacks diversification.
Agreed. TSM is a great fund for market portfolio investors, but over its history, it's predominately been a large-growth fund. Using the Fama-French 2x3 definitions over the 1926-2011 period (from reconstructed, backdated data), the average allocation to large-growth stocks has been 51% (in purple, chart below).
Personally, while a portion of my risky assets have been allocated to TSM, I've tried to add other, less than perfectly correlated equity classes to balance the large-growth bias of the U.S. market portfolio. This has meant adding small caps, value, international, emerging markets, REITs — virtually whatever low-cost index funds Vanguard has come up with over the years!
Cordially, Todd

User avatar
JoMoney
Posts: 5582
Joined: Tue Jul 23, 2013 5:31 am

Re: Yes, you can diversify a market portfolio

Post by JoMoney » Mon Jun 11, 2018 11:42 am

Disagreed.
Unless you have better information than other market participants, deviating from the market consensus is just exposing yourself to idiosyncratic risk. Regardless if it's picking individual stocks, sectors, or styles. You're not bringing anything to the table, or doing anything deserving a "premium". You can roll the dice and hope being "different" turns out in a good way, but that's about the extent of it. The only thing more ridiculous then believing it will bring some "persistent" path to higher returns, is believing that everyone should be doing it (and still expecting it will lead to higher returns).
Falkenstein 2009, in Risk and Return in General: Theory and Evidence wrote:...It is possible that risk, however defined, is not positively related to return as an equilibrium in asset markets. This paper presents a survey of data across 20 different asset classes, and presents a model highlighting the assumptions consistent with no risk premium. The key is that when agents are concerned about relative wealth, risk taking is then deviating from the consensus or market portfolio. In this environment, all risk becomes like idiosyncratic risk in the standard model, avoidable so unpriced...
Warren Buffett wrote:...Risk premiums are mostly nonsense. The world isn't calculating risk premiums...
Benjamin Graham in The Intelligent Investor wrote:...It has been an old and sound principle that those who cannot afford to take risks should be content with a relatively low return on their invested funds. From this there has developed the general notion that the rate of return which the investor should aim for is more or less proportionate to the degree of risk he is ready to run. Our view is different. The rate of return sought should be dependent, rather, on the amount of intelligent effort the investor is willing and able to bring to bear on his task...
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

Random Walker
Posts: 2997
Joined: Fri Feb 23, 2007 8:21 pm

Re: Yes, you can diversify a market portfolio

Post by Random Walker » Mon Jun 11, 2018 12:11 pm

Putting all your eggs in market beta alone with TSM with current P/E ratio S&P 500 something like 26-30 sounds risky to me.

Dave

User avatar
JoMoney
Posts: 5582
Joined: Tue Jul 23, 2013 5:31 am

Re: Yes, you can diversify a market portfolio

Post by JoMoney » Mon Jun 11, 2018 12:20 pm

Random Walker wrote:
Mon Jun 11, 2018 12:11 pm
Putting all your eggs in market beta alone with TSM with current P/E ratio S&P 500 something like 26-30 sounds risky to me.

Dave
What is it that you think the P/E tells you? Look like small-caps like Russell 2000 PE is at almost 90... so ... higher anticipated growth ... ? ...
https://wsj.com/mdc/public/page/2_3021-peyield.html
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
SimpleGift
Posts: 3121
Joined: Tue Feb 08, 2011 3:45 pm
Location: Central Oregon

Re: Yes, you can diversify a market portfolio

Post by SimpleGift » Mon Jun 11, 2018 12:27 pm

JoMoney wrote:
Mon Jun 11, 2018 11:42 am
Unless you have better information than other market participants, deviating from the market consensus is just exposing yourself to idiosyncratic risk.
But portfolio theory (ala Markowitz) tells us that combining these less-than-perfectly-correlated risky assets of comparable expected returns together should produce better risk-adjusted portfolio results. While we don't really have enough historical data to prove the case, the empirical evidence that we do have from the past 5 decades does support the theory.

I do agree that, while generating higher risk-adjusted returns might be a nice benefit of a multi-asset portfolio, it's certainly not an essential financial goal for many individual investors. In most cases, the compounding returns of the simple market portfolio, held patiently and faithfully over decades, will prove more than adequate for success.
Last edited by SimpleGift on Mon Jun 11, 2018 12:34 pm, edited 1 time in total.
Cordially, Todd

User avatar
JoMoney
Posts: 5582
Joined: Tue Jul 23, 2013 5:31 am

Re: Yes, you can diversify a market portfolio

Post by JoMoney » Mon Jun 11, 2018 12:33 pm

^ While the math involved with MPT is sound, we have lots of empirical evidence in the real world (ala "Efficient Frontier" charts and there changes over different time periods), that neither projected guesstimates of analysts nor historical data can reliably provide the necessary inputs to make MPT work in practice... great theory, but garbage in garbage out...
https://en.wikipedia.org/wiki/Modern_po ... Criticisms
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
abuss368
Posts: 12738
Joined: Mon Aug 03, 2009 2:33 pm
Location: Where the water is warm, the drinks are cold, and I don't know the names of the players!

Re: Yes, you can diversify a market portfolio

Post by abuss368 » Mon Jun 11, 2018 1:55 pm

vineviz wrote:
Tue Jun 05, 2018 6:00 pm
Several times in recent weeks I've seen a reply on the Bogleheads forum that said "you can't diversify a market portfolio" or something along those lines.

This isn't true.
Hi vineviz -

I am not sure I follow. If an investor owns a market portfolio then by default are they are not diversified?

Best.
John C. Bogle: "You simply do not need to put your money into 8 different mutual funds!" | | Disclosure: Three Fund Portfolio + U.S. & International REITs

goblue100
Posts: 639
Joined: Sun Dec 01, 2013 10:31 am

Re: Yes, you can diversify a market portfolio

Post by goblue100 » Mon Jun 11, 2018 1:57 pm

JoMoney wrote:
Mon Jun 11, 2018 12:33 pm
^ While the math involved with MPT is sound, we have lots of empirical evidence in the real world (ala "Efficient Frontier" charts and there changes over different time periods), that neither projected guesstimates of analysts nor historical data can reliably provide the necessary inputs to make MPT work in practice... great theory, but garbage in garbage out...
https://en.wikipedia.org/wiki/Modern_po ... Criticisms
^To me this is the rub with most of this stuff. We debate endlessly about SWR's and asset allocation. I can tell there are some really smart people in this thread(exempting myself, of course), but once we end up in the great circle debate it seems like we always come back to the best place to be is somewhere between 40/60 and 60/40 and the lowest possible tolerable withdrawal rate. Unless someone can get in the time travel machine and bring us back the Callan Periodic chart of investment returns from 2028?
Some people are immune to good advice. - Saul Goodman

columbia
Posts: 844
Joined: Tue Aug 27, 2013 5:30 am

Re: Yes, you can diversify a market portfolio

Post by columbia » Mon Jun 11, 2018 2:43 pm

My understanding is that time travel is likely only possible for moving forward in time. :)

User avatar
JoMoney
Posts: 5582
Joined: Tue Jul 23, 2013 5:31 am

Re: Yes, you can diversify a market portfolio

Post by JoMoney » Mon Jun 11, 2018 2:54 pm

columbia wrote:
Mon Jun 11, 2018 2:43 pm
My understanding is that time travel is likely only possible for moving forward in time. :)
But you can re-write history, as long as nobody observed it.
https://youtu.be/8ORLN_KwAgs
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
vineviz
Posts: 1315
Joined: Tue May 15, 2018 1:55 pm

Re: Yes, you can diversify a market portfolio

Post by vineviz » Mon Jun 11, 2018 3:20 pm

abuss368 wrote:
Mon Jun 11, 2018 1:55 pm
I am not sure I follow. If an investor owns a market portfolio then by default are they are not diversified?
I think most people would say that the market portfolio is a diversified portfolio. Certainly it does, by definition, hold the greatest possible number of assets.

The question that started this thread was slightly different, though. That question was more akin to "is the market portfolio the MOST diversified portfolio possible."

As the countless comments have revealed, the answer to that question depends somewhat on how we define 'diversified'. Here's how I'd summarize the way I see the differing views:
  • Some people view diversification as spreading their investment capital over a large number of assets. For them, the market portfolio is as diversified as you can get.
  • Other people view diversification as taking no more or less risk than the market portfolio. This leads, via circular argument, to the conclusion that the market portfolio is as diversified as you can get.
  • Some people view diversification as as spreading their investment risk over as many uncorrelated sources of risk as possible. For them, the market portfolio can definitely be further diversified.
  • Still other people view diversification as minimizing the volatility of their portfolio. For them, the market portfolio can definitely be further diversified.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

aristotelian
Posts: 4609
Joined: Wed Jan 11, 2017 8:05 pm

Re: Yes, you can diversify a market portfolio

Post by aristotelian » Tue Jun 12, 2018 4:34 am

Random Walker wrote:
Mon Jun 11, 2018 9:29 am

I don’t think it’s semantic. Just need to look at diversification differently, perhaps more broadly within an MPT framework. The most fundamental definition of diversification is “don’t put all your eggs in one basket”. Thus owning the entire stock market appears diversified. But the TSM investor is only exposed to one source of risk, market beta. If one looks at diversification a bit differently, “don’t put all your eggs in one risk basket”, then TSM seriously lacks diversification. Diversification is much more than number of stocks. It’s sources of uncorrelated risk and sources of uncorrelated return.

Dave
Again, though, you are advocating something like "strategic concentration". I practice it myself. I do think it is important that we be clear that this is not "diversification" because you are taking some risk that past non-correlation doesn't hold up, and the asset classes you have weighted will crash at the same time.

Random Walker
Posts: 2997
Joined: Fri Feb 23, 2007 8:21 pm

Re: Yes, you can diversify a market portfolio

Post by Random Walker » Tue Jun 12, 2018 9:26 am

aristotelian wrote:
Tue Jun 12, 2018 4:34 am
Random Walker wrote:
Mon Jun 11, 2018 9:29 am

I don’t think it’s semantic. Just need to look at diversification differently, perhaps more broadly within an MPT framework. The most fundamental definition of diversification is “don’t put all your eggs in one basket”. Thus owning the entire stock market appears diversified. But the TSM investor is only exposed to one source of risk, market beta. If one looks at diversification a bit differently, “don’t put all your eggs in one risk basket”, then TSM seriously lacks diversification. Diversification is much more than number of stocks. It’s sources of uncorrelated risk and sources of uncorrelated return.

Dave
Again, though, you are advocating something like "strategic concentration". I practice it myself. I do think it is important that we be clear that this is not "diversification" because you are taking some risk that past non-correlation doesn't hold up, and the asset classes you have weighted will crash at the same time.
Getting at limits of my knowledge here, but a TSM fund will by definition have a market beta of 1, and no exposure to HmL or SmB. A typical SV fund will have a market beta of about 1 also, and HmL 0.3 and SmB 0.3. Which fund is more diversified across the factors market beta, SmB, HmL? So, yes “strategically concentrated” in a corner of the market, but actually more spread out across the unique and independent sources of risk/return.

Dave

User avatar
patrick013
Posts: 2331
Joined: Mon Jul 13, 2015 7:49 pm

Re: Yes, you can diversify a market portfolio

Post by patrick013 » Tue Jun 12, 2018 3:07 pm

vineviz wrote:
Mon Jun 11, 2018 3:20 pm

I think most people would say that the market portfolio is a diversified portfolio.
Certainly it does, by definition, hold the greatest possible number of assets.
Sometimes I think portfolios are over diversified. These numbers
are a little old and a little too high I think but still represents the
idea as evolved. A portfolio could be over-diversified and hold
lower quality stocks if too many stocks are indexed.
Image

So traditional portfolio management invests in a large number of companies
that appear to offer superior opportunities for growth of capital and from a
wide range of industries.

Further, the modern MPT structures everything with statistical measures.
But I think over and under diversification can still occur. Who knows ?
I don't mind increasing beta with mid and small cap indexes. Correlation
is high in a normal economy but so can be returns and risk/reward. Risk
taken is risk taken.
age in bonds, buy-and-hold, 10 year business cycle

Post Reply