Diversification is not always, automatically, a free lunch

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
Topic Author
nisiprius
Advisory Board
Posts: 39724
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

Diversification is not always, automatically, a free lunch

Post by nisiprius » Fri Jan 10, 2020 9:52 am

In another thread, someone asked why "BNDX doesn't get more love on this forum since almost all Bogleheads believe in the free lunch of diversification..." Of the two propositions, "there ain't no such thing as a free lunch" versus "diversification is the only free lunch in investing," if you force me to choose, I choose the first. But in any case...

Diversification is not automatically an improvement. And even if it is an improvement it is not automatically a meaningful improvement.

A good example of an asset class that has been a diversifier, yet did not improve traditional portfolios, is "commodities" (collateralized commodities futures) funds, over the time period when they have been available in the form of mutual funds and ETFs. Our specific example will be PCRIX, once widely recommended as a diversifier, suitable for inclusion in otherwise-traditional retirement savings portfolios.

As I'll show below, PCRIX (PIMCO Commodity Real Return Strategy) has behaved very differently from VTSAX (Vanguard Total Stock Market Index Fund). Since inception, it has had a good claim on being a diversifier, a truly distinct asset class, independent from stocks.

Over the lifetime of PCRIX their correlation has been 0.40. (That's in the range that people call "low" when they are advocating for diversifiers; for example, over that same time period, the Vanguard Real Estate fund had a correlation of 0.68 with stocks; Vanguard Total International, 0.88; and DFA US Small-cap Value Portfolio, 0.91). (Source)

There are other ways to measure diversification, but I doubt that anyone can find a measure by which PCRIX hasn't qualified as "a diversifier."

Yet, if we take VSMGX, Vanguard LifeStrategy Moderate as a good example of a "traditional" portfolio (Portfolio 1, blue) and compare it to the effect of adding PCRIX (Portfolio 2, red, 80% VSMGX, 20% PCRIX), over the life of PCRIX, the effect of adding PCRIX would have been to make the portfolio worse by almost every measure:

Source

Image

Lower return (CAGR). Higher volatility (as measured by standard deviation). True, better "best year." But worse "worst year," worse drawdown, and lower risk-adjusted return as measured by Sharpe and Sortino ratios.

Now, some will say "you deliberately chose stinker of a diversifier to make your point," and so I did. The point is, it was a diversifier, yet it did not improve the portfolio as a whole? Why not? The answer is obvious: PCRIX, since inception, has been such a lousy investment in itself.

Image

Improvement through diversification is a balancing act with two pans.

On the one side, low correlation means that the volatility of the whole will be less than the weighted average of the volatility of the parts. That's the "free lunch." You can get your free lunch in the form of reduced volatility. Alternatively, you can adjust the portfolio to hold more of the now-less-risky mix, and get it in the form of increased return.

But on the other hand, the annual return of a portfolio is always just the weighted average of the returns of its parts. If it is 2014 and if fund X goes up 32.00% and and fund Y goes down -14.00%, then a 50/50 mix of the two will go up by 16.00%-7.00% = 9.00%, exactly, and nothing about the correlations or standard deviations of the two funds can change it.

Dragging down the return by a small amount may be worth it if it also cuts volatility by a large amount. The Sharpe ratio measures that relationship, and as noted if we get an improvement in Sharpe ratio, we can turn that into improved return rather than reduced volatility if we want to.

To return to my main point, though, diversification is a balancing act with two sides of the scale.

The "magic" of MPT is that if things are right, an asset that is not so great in itself can improve a portfolio as a whole. This is true because it is just math, but "if things are right" can be a strong requirement. The idealized example is two assets that have about the same return and low correlation. If they have about the same return, then the low correlation is roughly cost-free way to reduce the volatility. If they don't have about the same return, then the effect of diversification is to reduce volatility, but the lower-return asset also reduces return.

It's a balancing act, the balancing scale is the Sharpe ratio (or other measure of risk-adjusted return). A pretty good asset with very low correlation makes a big improvement. An almost-equal-return asset with not low but not 1.00 correlation makes a small improvement. In theory, a low return asset can make an improvement if the correlation is low enough. In theory a zero-return asset can make an improvement, but the correlation must be not just low, but actually negative. In theory, an asset could lose money yet improve a portfolio if the correlation were sufficiently negative.

But it is always a balancing act, and if the diversifier has low returns, it will not automatically improve a portfolio--it has to be an extremely powerful diversifier.

Now, the problem is that none of the things that govern whether a diversifier improves a portfolio are any more predictable than return itself is. "Past performance is no sure guide to future results" is just as true of volatility and correlation as it is for return.

Many diversifiers lie in the "grey zone of endless debate" and the usual reason is that the balancing pans--between "diversifier" and "drag" have, in the past, been so close level--either way depending on choice of time period--that nobody can produce a truly compelling case to believe that in the future, the good effects of diversification will outweigh the bad effects of drag.

If someone point out some fund or asset class as being "a strong diversifier, with low correlation with stocks," don't let that lull you into thinking "so it must be good." You have to ask hard questions. It needs to be at least a fairly good investment in itself, and you need to believe that both "decent return" and "low correlation" will continue into the future.
Last edited by nisiprius on Fri Jan 10, 2020 7:18 pm, edited 3 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.

dbr
Posts: 31167
Joined: Sun Mar 04, 2007 9:50 am

Re: Diversification is not always, automatically, a free lunc

Post by dbr » Fri Jan 10, 2020 10:01 am

Agree with the above.

I think the summary is that a good diversifier must have the following properties:

1. Expected return similar to the rest of the portfolio, or at least not so small as to be a drag that cannot be overcome. This, by the way, excludes most so-called diversifiers.

2. Expected volatility similar to the rest of the portfolio. This also excludes most so-called diversifiers as the result is dilution of risk and return unless one finds a magic investment that has high return and little volatility.

3. Lack of correlation with the rest of the portfolio, not necessarily to be confused with negative correlation. An unfortunate effect is that at just the wrong time correlation may shift strongly positive.

User avatar
Topic Author
nisiprius
Advisory Board
Posts: 39724
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: Diversification is not always, automatically, a free lunc

Post by nisiprius » Fri Jan 10, 2020 10:03 am

I know I'm long-winded, dbr, so thank you very much for that summary.

I don't like the phrasing about "correlations turning negative" (not criticizing you, everyone seems to phrase it that way) because it completely confuses the relation between true underlying correlations and correlations between samples. If you are doing textbook stuff and statistics on independent normally-distributed variables, the sampling error between correlations of samples is huge. If you measure a certain correlation between samples, the confidence limits on the true correlation are very wide. It's really no different from saying "invest in XYZ because it has had high return" (often misphrased as "has high return!") and then adding "at just the wrong time, return may shift strongly negative!"

Of course we have no reason to believe that underlying correlations are stable, intrinsic properties, either. But I think it's more likely that supposed "changes" in correlation are just illustrations of sampling error.
Last edited by nisiprius on Fri Jan 10, 2020 10:14 am, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

rkhusky
Posts: 7825
Joined: Thu Aug 18, 2011 8:09 pm

Re: Diversification is not always, automatically, a free lunc

Post by rkhusky » Fri Jan 10, 2020 10:13 am

Since stock market returns have no known statistical distribution, there is no known underlying correlation, and all you have is sample correlation.

User avatar
Topic Author
nisiprius
Advisory Board
Posts: 39724
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: Diversification is not always, automatically, a free lunc

Post by nisiprius » Fri Jan 10, 2020 10:14 am

rkhusky wrote:
Fri Jan 10, 2020 10:13 am
Since stock market returns have no known statistical distribution, there is no known underlying correlation, and all you have is sample correlation.
OK.
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.

dbr
Posts: 31167
Joined: Sun Mar 04, 2007 9:50 am

Re: Diversification is not always, automatically, a free lunc

Post by dbr » Fri Jan 10, 2020 10:27 am

rkhusky wrote:
Fri Jan 10, 2020 10:13 am
Since stock market returns have no known statistical distribution, there is no known underlying correlation, and all you have is sample correlation.
Right. We are in this odd situation of using sample statistics, not just correlations but means and standard deviations, as if they were estimates of an actual distribution. That does not mean that the model is not useful, but it certainly means people have to be careful what they infer from the model.

columbia
Posts: 2137
Joined: Tue Aug 27, 2013 5:30 am

Re: Diversification is not always, automatically, a free lunc

Post by columbia » Fri Jan 10, 2020 10:27 am

Wellesley Income - for whatever reason - usually has been a better* diversifier than most “but think about the diversification aspects!” investments.

*Leading to higher returns, lower volatility, lower drawdowns and higher risk adjusted return.

dbr
Posts: 31167
Joined: Sun Mar 04, 2007 9:50 am

Re: Diversification is not always, automatically, a free lunc

Post by dbr » Fri Jan 10, 2020 10:31 am

nisiprius wrote:
Fri Jan 10, 2020 10:03 am
I know I'm long-winded, dbr, so thank you very much for that summary.

I don't like the phrasing about "correlations turning negative" (not criticizing you, everyone seems to phrase it that way) because it completely confuses the relation between true underlying correlations and correlations between samples. If you are doing textbook stuff and statistics on independent normally-distributed variables, the sampling error between correlations of samples is huge. If you measure a certain correlation between samples, the confidence limits on the true correlation are very wide. It's really no different from saying "invest in XYZ because it has had high return" (often misphrased as "has high return!") and then adding "at just the wrong time, return may shift strongly negative!"

Of course we have no reason to believe that underlying correlations are stable, intrinsic properties, either. But I think it's more likely that supposed "changes" in correlation are just illustrations of sampling error.
Indeed, the whole enterprise is messy. I would still contend examination of the evidence using a statistical framework as we do is useful as long as one does not go too far in setting expectations. There are lots of enterprises in the world where statistics is useful this way even though the actual processes are hardly understood. There is also plenty of misinformation from the same source.

User avatar
305pelusa
Posts: 1247
Joined: Fri Nov 16, 2018 10:20 pm

Re: Diversification is not always, automatically, a free lunc

Post by 305pelusa » Fri Jan 10, 2020 10:57 am

nisiprius wrote:
Fri Jan 10, 2020 9:52 am
In another thread, someone asked why "BNDX doesn't get more love on this forum since almost all Bogleheads believe in the free lunch of diversification..." Of the two propositions, "there ain't no such thing as a free lunch" versus "diversification is the only free lunch in investing," if you force me to choose, I choose the first. But in any case...

Diversification is not automatically an improvement. And even if it is an improvement it is not automatically a meaningful improvement.

A good example of an asset class that has been a diversifier, yet did not improve traditional portfolios, is "commodities" (collateralized commodities futures) funds, over the time period when they have been available in the form of mutual funds and ETFs. Our specific example will be PCRIX, once widely recommended as a diversifier, suitable for inclusion in otherwise-traditional retirement savings portfolios.

As I'll show below, PCRIX (PIMCO Commodity Real Return Strategy) has behaved very differently from VTSAX (Vanguard Total Stock Market Index Fund). Since inception, it has had a good claim on being a diversifier, a truly distinct asset class, independent from stocks.

Over the lifetime of PCRIX their correlation has been 0.40. (That's in the range that people call "low" when they are advocating for diversifiers; for example, over that same time period, the Vanguard Real Estate fund had a correlation of 0.68 with stocks; Vanguard Total International, 0.88; and DFA US Small-cap Value Portfolio, 0.91). (Source)

There are other ways to measure diversification, but I doubt that anyone can find a measure by which PCRIX hasn't qualified as "a diversifier."

Yet, if we take VSMGX, Vanguard LifeStrategy Moderate as a good example of a "traditional" portfolio (Portfolio 1, blue) and compare it to the effect of adding PCRIX (Portfolio 2, red, 80% VSMGX, 20% PCRIX), over the life of PCRIX, the effect of adding PCRIX would have been to make the portfolio worse by almost every measure:

Source

Image

Lower return (CAGR). Higher volatility (as measured by standard deviation). True, better "best year." But worse "worst year," worse drawdown, and lower risk-adjusted return as measured by Sharpe and Sortino ratios.

Now, some will say "you deliberately chose stinker of a diversifier to make your point," and so I did. The point is, it was a diversifier, yet it did not improve the portfolio as a whole? Why not? The answer is obvious: PCRIX, since inception, has been such a lousy investment in itself.

Image

Improvement through diversification is a balancing act with two pans.

On the one side, low correlation means that the volatility of the whole will be less than the weighted average of the volatility of the parts. That's the "free lunch." You can get your free lunch in the form of reduced volatility. Alternatively, you can adjust the portfolio to hold more of the now-less-risky mix, and get it in the form of increased return.

But on the other hand, the annual return of a portfolio is always just the weighted average of the returns of its parts. If it is 2014 and if fund X goes up 32.00% and and fund Y goes down -14.00%, then a 50/50 mix of the two will go up by 16.00%-7.00% = 9.00%, exactly, and nothing about the correlations or standard deviations of the two funds can change it.

Dragging down the return by a small amount may be worth it if it also cuts volatility by a large amount. The Sharpe ratio measures that relationship, and as noted if we get an improvement in Sharpe ratio, we can turn that into improved return rather than reduced volatility if we want to.

To return to my main point, though, diversification is a balancing act with two sides of the scale.

The "magic" of MPT is that if things are right, an asset that is not so great in itself can improve a portfolio as a whole. This is true because it is just math, but "if things are right" can be a strong requirement. The idealized example is two assets that have about the same return and low correlation. If they have about the same return, then the low correlation is roughly cost-free way to reduce the volatility. If they don't have about the same return, then the effect of diversification is to reduce volatility, but the lower-return asset also reduces return.

It's a balancing act, the balancing scale is the Sharpe ratio (or other measure of risk-adjusted return). A pretty good asset with very low correlation makes a big improvement. An almost-equal-return asset with not low but not 1.00 correlation makes a small improvement. In theory, a low return asset can make an improvement if the correlation is low enough. In theory a zero-return asset can make an improvement, but the correlation must be not just low, but actually negative. In theory, an asset could lose money yet improve a portfolio if the correlation were sufficiently negative.

But it is always a balancing act, and if the diversifier has low returns, it will not automatically improve a portfolio--it has to be an extremely powerful diversifier.

Now, the problem is that none of the things that govern whether a diversifier improves a portfolio are any more predictable than return itself is. "Past performance is no sure guide to future results" is just as true of volatility and correlation as it is for return.

Many diversifiers lie in the "grey zone of endless debate" and the usual reason is that the balancing pans--between "diversifier" and "drag" have, in the past, been so close level--either way depending on choice of time period--that nobody can produce a truly compelling case to believe that in the future, the good effects of diversification will outweigh the bad effects of drag.

If someone point out some fund or asset class as being "a strong diversifier, with low correlation with stocks," don't let that lull you into thinking "so it must be good." You have to ask hard questions. It needs to be at least a fairly good investment in itself, and you need to believe that both "decent return" and "low correlation" will continue into the future.
Nisprus you know better than this. The concept of diversification means having assets that zig while others zag and vice-versa. The point of commodities is that they would hopefully produce outsized returns during highly inflationary periods, precisely when bonds and stocks would perform poorly short term. They would also provide a hedge against event risk, which we haven’t really seen much over that time period.

You can’t just look at what happened. You need to think about the hundreds of parallel universes that could’ve happened, and measure diversification over all of those. Since we can’t do that, a very long backtest might be a reasonable substitute since it would hopefully contain many different conditions. A backtest of the previous 16 years, where we have had trouble seeing the desirable 2% inflation is not good enough. Not even close.

HEDGEFUNDIE
Posts: 3973
Joined: Sun Oct 22, 2017 2:06 pm

Re: Diversification is not always, automatically, a free lunc

Post by HEDGEFUNDIE » Fri Jan 10, 2020 11:17 am

305pelusa and nisi are both right.

Diversification is important because we don’t know what future conditions will come to pass. And so we should all hold some amount of asset classes (like commodities, and international stocks) that have performed terribly in the past, with the expectation that they will come through for us if and when the right conditions strike.

However, we should also try to forecast the likelihood of those future conditions, so as not to overprotect ourselves for unlikely possibilities and give up likely return.

I will repeat the analogy I have used in the past.

If you live in Miami, how many winter coats should you own? Probably not zero. Probably not ten.

Uncorrelated
Posts: 249
Joined: Sun Oct 13, 2019 3:16 pm

Re: Diversification is not always, automatically, a free lunc

Post by Uncorrelated » Fri Jan 10, 2020 11:30 am

A factor regression of PCRIX shows pretty poor results with a R^2 of only 25% and massive negative alpha (not statistically significant).

Every once in a while a thread pops up that attempts to use correlation as a proxy for diversification. As you show that is a really bad idea. Low correlation can be explained in two ways: high idiosyncratic risk or an actual independent source of return. It is only the latter that you are interested in.

I only accept diversification from the sources that I trust, such as bonds and the size and value factors.

User avatar
hdas
Posts: 1323
Joined: Thu Jun 11, 2015 8:24 am

Re: Diversification is not always, automatically, a free lunc

Post by hdas » Fri Jan 10, 2020 11:36 am

HEDGEFUNDIE wrote:
Fri Jan 10, 2020 11:17 am

However, we should also try to forecast the likelihood of those future conditions, so as not to overprotect ourselves for unlikely possibilities and give up likely return
This is very naive and worthless for the average investor. A solution has been invented and marketed, it’s called All Weather Portfolio. Leveraged to your individual taste. Cheers :greedy
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

Seasonal
Posts: 628
Joined: Sun May 21, 2017 1:49 pm

Re: Diversification is not always, automatically, a free lunc

Post by Seasonal » Fri Jan 10, 2020 11:37 am

The issues raised in this thread are why I believe the search for a portfolio engineered to be highly diversified is a mistake. If you believe markets are more or less efficient, a portfolio resembling the capitalization weighted market portfolio is more than good enough, unless you have a strong reason to believe that something else would be better for you.

If the market thought something would be better, the market portfolio would change to reflect that. What do you, fan of diversification, know that that market doesn't know?

If you'd like academic support for this view, look to Bill Sharpe (there's a thread on his preference for approximating the market portfolio, with TIPS weighted to manage risk) and Eugene Fama (the market portfolio is always on the efficient frontier), among others. Taylor has an extensive collection of quotes that support an approximation of simple cap weighted portfolios (the three fund portfolio is a nice example). Even fans of highly engineered portfolios, such as Larry Swedroe, write that you can't improve on it on a risk-adjusted basis.

User avatar
305pelusa
Posts: 1247
Joined: Fri Nov 16, 2018 10:20 pm

Re: Diversification is not always, automatically, a free lunc

Post by 305pelusa » Fri Jan 10, 2020 11:44 am

Uncorrelated wrote:
Fri Jan 10, 2020 11:30 am
A factor regression of PCRIX shows pretty poor results with a R^2 of only 25% and massive negative alpha (not statistically significant).

Every once in a while a thread pops up that attempts to use correlation as a proxy for diversification. As you show that is a really bad idea. Low correlation can be explained in two ways: high idiosyncratic risk or an actual independent source of return. It is only the latter that you are interested in.

I only accept diversification from the sources that I trust, such as bonds and the size and value factors.
Did you just use stock market factors to regress commodities ?

I don’t even know where to begin...

Uncorrelated
Posts: 249
Joined: Sun Oct 13, 2019 3:16 pm

Re: Diversification is not always, automatically, a free lunc

Post by Uncorrelated » Fri Jan 10, 2020 11:46 am

305pelusa wrote:
Fri Jan 10, 2020 11:44 am
Uncorrelated wrote:
Fri Jan 10, 2020 11:30 am
A factor regression of PCRIX shows pretty poor results with a R^2 of only 25% and massive negative alpha (not statistically significant).

Every once in a while a thread pops up that attempts to use correlation as a proxy for diversification. As you show that is a really bad idea. Low correlation can be explained in two ways: high idiosyncratic risk or an actual independent source of return. It is only the latter that you are interested in.

I only accept diversification from the sources that I trust, such as bonds and the size and value factors.
Did you just use stock market factors to regress commodities ?

I don’t even know where to begin...
I didn't notice it was a commodities fund instead of an equity fund. That is definitely a mistake :oops:

HEDGEFUNDIE
Posts: 3973
Joined: Sun Oct 22, 2017 2:06 pm

Re: Diversification is not always, automatically, a free lunc

Post by HEDGEFUNDIE » Fri Jan 10, 2020 12:22 pm

hdas wrote:
Fri Jan 10, 2020 11:36 am
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 11:17 am

However, we should also try to forecast the likelihood of those future conditions, so as not to overprotect ourselves for unlikely possibilities and give up likely return
This is very naive and worthless for the average investor. A solution has been invented and marketed, it’s called All Weather Portfolio. Leveraged to your individual taste. Cheers :greedy
And how has All Weather performed over the last 30 years?

klaus14
Posts: 328
Joined: Sun Nov 25, 2018 7:43 pm

Re: Diversification is not always, automatically, a free lunc

Post by klaus14 » Fri Jan 10, 2020 12:26 pm

HEDGEFUNDIE wrote:
Fri Jan 10, 2020 11:17 am
305pelusa and nisi are both right.

Diversification is important because we don’t know what future conditions will come to pass. And so we should all hold some amount of asset classes (like commodities, and international stocks) that have performed terribly in the past, with the expectation that they will come through for us if and when the right conditions strike.

However, we should also try to forecast the likelihood of those future conditions, so as not to overprotect ourselves for unlikely possibilities and give up likely return.

I will repeat the analogy I have used in the past.

If you live in Miami, how many winter coats should you own? Probably not zero. Probably not ten.
This is a very important point and illustrates the problem with all risk-parity type portfolios. If you are allocating 25% to inflation risk, you are allocating too much. Especially if you are still earning, you should probably allocate 0%. So no commodities or TIPS are needed. (I still buy gold and IBonds for other reasons though)
5% ACWV | 15% VFMF, 15% NTSX | 10% ISCF, 5% EFAV | 5% FNDE, 5% EMGF, 5% VEGBX, 5% LEMB | 15% EDV, 5% CD (5y) | 10% GLDM

User avatar
hdas
Posts: 1323
Joined: Thu Jun 11, 2015 8:24 am

Re: Diversification is not always, automatically, a free lunc

Post by hdas » Fri Jan 10, 2020 12:31 pm

HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:22 pm
hdas wrote:
Fri Jan 10, 2020 11:36 am
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 11:17 am

However, we should also try to forecast the likelihood of those future conditions, so as not to overprotect ourselves for unlikely possibilities and give up likely return
This is very naive and worthless for the average investor. A solution has been invented and marketed, it’s called All Weather Portfolio. Leveraged to your individual taste. Cheers :greedy
And how has All Weather performed over the last 30 years?
Jun-96 (inception) to Oct-18

CAGR 7.5%, STD 9.9%
Unlevered
Last year was around 20%

Cheers :greedy
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

HEDGEFUNDIE
Posts: 3973
Joined: Sun Oct 22, 2017 2:06 pm

Re: Diversification is not always, automatically, a free lunc

Post by HEDGEFUNDIE » Fri Jan 10, 2020 12:52 pm

hdas wrote:
Fri Jan 10, 2020 12:31 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:22 pm
hdas wrote:
Fri Jan 10, 2020 11:36 am
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 11:17 am

However, we should also try to forecast the likelihood of those future conditions, so as not to overprotect ourselves for unlikely possibilities and give up likely return
This is very naive and worthless for the average investor. A solution has been invented and marketed, it’s called All Weather Portfolio. Leveraged to your individual taste. Cheers :greedy
And how has All Weather performed over the last 30 years?
Jun-96 (inception) to Oct-18

CAGR 7.5%, STD 9.9%
Unlevered
Compared to
50% S&P 500 & 50% LTT:

CAGR 8.1%, STD 8.0%

I rest my case.

User avatar
hdas
Posts: 1323
Joined: Thu Jun 11, 2015 8:24 am

Re: Diversification is not always, automatically, a free lunc

Post by hdas » Fri Jan 10, 2020 1:08 pm

HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:52 pm
hdas wrote:
Fri Jan 10, 2020 12:31 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:22 pm
hdas wrote:
Fri Jan 10, 2020 11:36 am
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 11:17 am

However, we should also try to forecast the likelihood of those future conditions, so as not to overprotect ourselves for unlikely possibilities and give up likely return
This is very naive and worthless for the average investor. A solution has been invented and marketed, it’s called All Weather Portfolio. Leveraged to your individual taste. Cheers :greedy
And how has All Weather performed over the last 30 years?
Jun-96 (inception) to Oct-18

CAGR 7.5%, STD 9.9%
Unlevered
Compared to
50% S&P 500 & 50% LTT:

CAGR 8.1%, STD 8.0%

I rest my case.
Lol...this is typical of the BH malaise of late. Strawman comparison, Complain about 0.6 cagr for carrying inflation protection, analyze a strategy based on n=1....same as the OP does it here. Good luck in sales. Cheers :greedy
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

HEDGEFUNDIE
Posts: 3973
Joined: Sun Oct 22, 2017 2:06 pm

Re: Diversification is not always, automatically, a free lunc

Post by HEDGEFUNDIE » Fri Jan 10, 2020 1:30 pm

hdas wrote:
Fri Jan 10, 2020 1:08 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:52 pm
hdas wrote:
Fri Jan 10, 2020 12:31 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:22 pm
hdas wrote:
Fri Jan 10, 2020 11:36 am


This is very naive and worthless for the average investor. A solution has been invented and marketed, it’s called All Weather Portfolio. Leveraged to your individual taste. Cheers :greedy
And how has All Weather performed over the last 30 years?
Jun-96 (inception) to Oct-18

CAGR 7.5%, STD 9.9%
Unlevered
Compared to
50% S&P 500 & 50% LTT:

CAGR 8.1%, STD 8.0%

I rest my case.
Lol...this is typical of the BH malaise of late. Strawman comparison, Complain about 0.6 cagr for carrying inflation protection, analyze a strategy based on n=1....same as the OP does it here. Good luck in sales. Cheers :greedy
Given how many posts I’ve gotten on my Excellent Adventure threads, I’d say my sales skills are well-honed, wouldn’t you?

pepys
Posts: 117
Joined: Mon Nov 12, 2018 10:34 am

Re: Diversification is not always, automatically, a free lunc

Post by pepys » Fri Jan 10, 2020 2:18 pm

hdas wrote:
Fri Jan 10, 2020 1:08 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:52 pm
hdas wrote:
Fri Jan 10, 2020 12:31 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:22 pm
hdas wrote:
Fri Jan 10, 2020 11:36 am


This is very naive and worthless for the average investor. A solution has been invented and marketed, it’s called All Weather Portfolio. Leveraged to your individual taste. Cheers :greedy
And how has All Weather performed over the last 30 years?
Jun-96 (inception) to Oct-18

CAGR 7.5%, STD 9.9%
Unlevered
Compared to
50% S&P 500 & 50% LTT:

CAGR 8.1%, STD 8.0%

I rest my case.
Lol...this is typical of the BH malaise of late. Strawman comparison, Complain about 0.6 cagr for carrying inflation protection, analyze a strategy based on n=1....same as the OP does it here. Good luck in sales. Cheers :greedy
It's not a strawman comparison. Saying that it provides inflation protection when the entire point was about the likelihood of needing inflation protection is just missing the point.

(And this is all assuming the inflation protection even works well. Hopefully gold doesn't go down 70% real as it did during Brazil's high inflation from 1980-2000.)
"Give me enough leverage and a fund on which to place it, and I shall move the world"

User avatar
hdas
Posts: 1323
Joined: Thu Jun 11, 2015 8:24 am

Re: Diversification is not always, automatically, a free lunc

Post by hdas » Fri Jan 10, 2020 2:32 pm

HEDGEFUNDIE wrote:
Fri Jan 10, 2020 1:30 pm
hdas wrote:
Fri Jan 10, 2020 1:08 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:52 pm
hdas wrote:
Fri Jan 10, 2020 12:31 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:22 pm


And how has All Weather performed over the last 30 years?
Jun-96 (inception) to Oct-18

CAGR 7.5%, STD 9.9%
Unlevered
Compared to
50% S&P 500 & 50% LTT:

CAGR 8.1%, STD 8.0%

I rest my case.
Lol...this is typical of the BH malaise of late. Strawman comparison, Complain about 0.6 cagr for carrying inflation protection, analyze a strategy based on n=1....same as the OP does it here. Good luck in sales. Cheers :greedy
Given how many posts I’ve gotten on my Excellent Adventure threads, I’d say my sales skills are well-honed, wouldn’t you?
Absolutely, I meant it as...It is a good thing you work in sales. Cheers :greedy
"whenever there is a randomized way of doing something, then there is a nonrandomized way that delivers better performance but requires more thought" ET Jaynes

james22
Posts: 1488
Joined: Tue Aug 21, 2007 2:22 pm

Re: Diversification is not always, automatically, a free lunc

Post by james22 » Fri Jan 10, 2020 4:03 pm

klaus14 wrote:
Fri Jan 10, 2020 12:26 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 11:17 am
However, we should also try to forecast the likelihood of those future conditions, so as not to overprotect ourselves for unlikely possibilities and give up likely return.
This is a very important point and illustrates the problem with all risk-parity type portfolios. If you are allocating 25% to inflation risk, you are allocating too much. Especially if you are still earning, you should probably allocate 0%. So no commodities or TIPS are needed. (I still buy gold and IBonds for other reasons though).
Bernstein in Deep Risk thought only inflation worth insuring against:

risk I probability I consequence severity I insurance I insurance cost

inflation I +++ I ++ I commodities, producers, TIPS I +

deflation I + I ++ I LTT, STT I +++

User avatar
danielc
Posts: 889
Joined: Sun Dec 10, 2017 4:48 am
Location: Iowa, USA
Contact:

Re: Diversification is not always, automatically, a free lunc

Post by danielc » Fri Jan 10, 2020 6:28 pm

nisiprius wrote:
Fri Jan 10, 2020 9:52 am
Lower return (CAGR). Higher volatility (as measured by standard deviation). True, better "best year." But worse "worst year," worse drawdown, and lower risk-adjusted return as measured by Sharpe and Sortino ratios.

Now, some will say "you deliberately chose stinker of a diversifier to make your point," and so I did. The point is, it was a diversifier, yet it did not improve the portfolio as a whole? Why not? The answer is obvious: PCRIX, since inception, has been such a lousy investment in itself.
I agree with everything you said up to (and including) this. The only kind of diversification I'm interested in is the kind that improves the portfolio (probably measured by the Sharpe and Sortino ratios). And this is coming from someone who holds international bonds. The evidence from Vanguard tells me that international bonds won't make a big difference one way or another, but are marginally more likely to be a small positive, and I like holding everything. But I don't buy PCRIX or other commodities, or lottery tickets. I'm not interested in diversifying by buying lousy investments.

nisiprius wrote:
Fri Jan 10, 2020 9:52 am
But on the other hand, the annual return of a portfolio is always just the weighted average of the returns of its parts.
Are you sure this is true? No rebalancing bonus?
nisiprius wrote:
Fri Jan 10, 2020 9:52 am
To return to my main point, though, diversification is a balancing act with two sides of the scale.

The "magic" of MPT is that if things are right, an asset that is not so great in itself can improve a portfolio as a whole. This is true because it is just math, but "if things are right" can be a strong requirement. The idealized example is two assets that have about the same return and low correlation. If they have about the same return, then the low correlation is roughly cost-free way to reduce the volatility. If they don't have about the same return, then the effect of diversification is to reduce volatility, but the lower-return asset also reduces return.

It's a balancing act, the balancing scale is the Sharpe ratio (or other measure of risk-adjusted return). A pretty good asset with very low correlation makes a big improvement. An almost-equal-return asset with not low but not 1.00 correlation makes a small improvement. In theory, a low return asset can make an improvement if the correlation is low enough. In theory a zero-return asset can make an improvement, but the correlation must be not just low, but actually negative. In theory, an asset could lose money yet improve a portfolio if the correlation were sufficiently negative.

But it is always a balancing act, and if the diversifier has low returns, it will not automatically improve a portfolio--it has to be an extremely powerful diversifier.
+1
nisiprius wrote:
Fri Jan 10, 2020 9:52 am
Now, the problem is that none of the things that govern whether a diversifier improves a portfolio are any more predictable than return itself is. "Past performance is no sure guide to future results" is just as true of volatility and correlation as it is for return.
I'm not entirely sure that that is true. I feel relatively confident that gold will remain very uncorrelated with equities, but I haven't the faintest idea of what its return will be. I think it is can be easier to argue that two assets have entirely different risks. I have some emerging market bonds in my portfolio. The risk that Latin America will have more defaults than usual is not entirely uncorrelated with the stock market, but I think that it is easy to argue that there are a lot of risks in one asset that are not present in the other.

User avatar
305pelusa
Posts: 1247
Joined: Fri Nov 16, 2018 10:20 pm

Re: Diversification is not always, automatically, a free lunc

Post by 305pelusa » Fri Jan 10, 2020 6:37 pm

danielc wrote:
Fri Jan 10, 2020 6:28 pm

nisiprius wrote:
Fri Jan 10, 2020 9:52 am
But on the other hand, the annual return of a portfolio is always just the weighted average of the returns of its parts.
Are you sure this is true? No rebalancing bonus?
It’s the weighted sum precisely because you rebalance, making sure each asset contributes its exact allocation worth of results. In fact, using the weighted sum assumes you rebalance continuously.

User avatar
danielc
Posts: 889
Joined: Sun Dec 10, 2017 4:48 am
Location: Iowa, USA
Contact:

Re: Diversification is not always, automatically, a free lunc

Post by danielc » Fri Jan 10, 2020 6:56 pm

HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:22 pm
hdas wrote:
Fri Jan 10, 2020 11:36 am
This is very naive and worthless for the average investor. A solution has been invented and marketed, it’s called All Weather Portfolio. Leveraged to your individual taste. Cheers :greedy
And how has All Weather performed over the last 30 years?
Surprisingly well it seems. I checked with Portfolio Visualizer. I couldn't get a lot of history for the 7.5% commodities so I replaced that with 7.5% in "precious metals". Over the last 30 years that portfolio had a slightly higher return, lower volatility, better Sharpe ratio, and MUCH lower "max drawdowns" than a 60/40 US stock/IT bond portfolio.

pepys
Posts: 117
Joined: Mon Nov 12, 2018 10:34 am

Re: Diversification is not always, automatically, a free lunc

Post by pepys » Fri Jan 10, 2020 7:39 pm

danielc wrote:
Fri Jan 10, 2020 6:56 pm
HEDGEFUNDIE wrote:
Fri Jan 10, 2020 12:22 pm
hdas wrote:
Fri Jan 10, 2020 11:36 am
This is very naive and worthless for the average investor. A solution has been invented and marketed, it’s called All Weather Portfolio. Leveraged to your individual taste. Cheers :greedy
And how has All Weather performed over the last 30 years?
Surprisingly well it seems. I checked with Portfolio Visualizer. I couldn't get a lot of history for the 7.5% commodities so I replaced that with 7.5% in "precious metals". Over the last 30 years that portfolio had a slightly higher return, lower volatility, better Sharpe ratio, and MUCH lower "max drawdowns" than a 60/40 US stock/IT bond portfolio.
I might be doing something wrong, but I am not getting that same result: https://www.portfoliovisualizer.com/bac ... ion5_2=7.5

For mine, 60-40 over the last 30 years has had a slightly higher CAGR (0.33% higher). The rest of what you said matches up, due to the heavy allocation to LT bonds.
"Give me enough leverage and a fund on which to place it, and I shall move the world"

User avatar
Topic Author
nisiprius
Advisory Board
Posts: 39724
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: Diversification is not always, automatically, a free lunch

Post by nisiprius » Fri Jan 10, 2020 8:09 pm

I don't want to go too far down the rabbit-hole of "is there a rebalancing bonus" and "can rebalancing manufacture return out of pure volatility."

I will content myself with saying that in the actual example I showed above--a "traditional" portfolio represented by Vanguard LifeStrategy Moderate Growth, and one with 80% moderate growth, 20% PCRIX--PIMCO Commodity RealReturn Strategy--these are the effects of no rebalancing, annual rebalancing, monthly rebalancing, and using rebalancing bands, and in this particular case it made very little difference. The actual performance of PCRIX since inception has simply been poor enough to more than outweigh any diversification advantage, regardless of whether or not rebalancing was done, or what kind of rebalancing regime was followed.

While creating this posting, I had to keep rechecking which corresponded to which rebalancing regime, because the charts and numbers were so similar across the four regimes.

Annual rebalancing:

Image

No rebalancing:

Image

Monthly rebalancing:

Image

Rebalancing bands:

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
abuss368
Posts: 16981
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!
Contact:

Re: Diversification is not always, automatically, a free lunch

Post by abuss368 » Fri Jan 10, 2020 8:29 pm

That is interesting. Investors have always said diversification is the only free lunch.
John C. Bogle - Two Fund Portfolio: Total Stock & Total Bond. "Simplicity is the master key to financial success."

cashboy
Posts: 193
Joined: Tue Sep 11, 2018 5:03 pm
Location: USA

Re: Diversification is not always, automatically, a free lunch

Post by cashboy » Fri Jan 10, 2020 8:29 pm

nisiprius wrote:
Fri Jan 10, 2020 9:52 am

Diversification is not automatically an improvement. And even if it is an improvement it is not automatically a meaningful improvement.
a lot of wisdom in those statements
Three-Fund Portfolio: FSPSX - FXAIX - FXNAX (with slight tilt of CDs - CASH - Canned Beans - Rice - Bottled Water)

afan
Posts: 4465
Joined: Sun Jul 25, 2010 4:01 pm

Re: Diversification is not always, automatically, a free lunch

Post by afan » Fri Jan 10, 2020 8:37 pm

Intermediate term bonds are great diversifiers with correlation coefficients near zero.
But no one cares because you don't have to pay an adviser 1% AUM to pick 2% ER funds to get this magic low correlation.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

james22
Posts: 1488
Joined: Tue Aug 21, 2007 2:22 pm

Re: Diversification is not always, automatically, a free lunch

Post by james22 » Fri Jan 10, 2020 11:42 pm

Why did you not extend the bond duration when adding commodities (as recommended), nisiprius?

User avatar
JoMoney
Posts: 8136
Joined: Tue Jul 23, 2013 5:31 am

Re: Diversification is not always, automatically, a free lunch

Post by JoMoney » Sat Jan 11, 2020 12:08 am

Diversification is only a "free lunch" if you can't otherwise distinguish between the items.
If you have 20 investments that all have the same 'expected return' but with varying random unknowable chance, diversifying across all 20 of them is a 'free lunch'.
If you know that 15 of the 20 have a higher expected return, owning the lower returning 5 will almost certainly have a cost.
In most situations, the 15 higher returning investments would likely be expected to have higher risks... so bigger question across the 20 investments is can you pick the ones with higher expected "risk adjusted return" ... which is a much harder question that may not be as clear.

Leaving the returns aside though, there are many situations where we can assess the risk of a situation. There are clearly higher risks involved with junk bonds than U.S. Treasuries. If your primary concern was to guarantee you wouldn't lose money, "diversifying" your treasury bonds to include a junk bond portfolio will definitely increase your risk.

One of the often quoted quotes is,
"In the absence of clarity, diversification is the only logical strategy"
Which is accurate, as long as you don't have "clarity". If you know one thing has risks you're not willing to take, and another doesn't... there's nothing unclear about it. If they all look equally risky (or at least relative to there risk-adjusted-return), diversify.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
CULater
Posts: 2341
Joined: Sun Nov 13, 2016 10:59 am

Re: Diversification is not always, automatically, a free lunch

Post by CULater » Sat Jan 11, 2020 8:42 am

Words from Peter Bernstein regarding diversification:
And I view diversification not only as a survival strategy but as an aggressive strategy, because the next windfall might come from a surprising place. I want to make sure I’m exposed to it. Somebody once said that if you’re comfortable with everything you own, you’re not diversified. I think you should have a small allocation to gold, to foreign currency, to TIPS [Treasury Inflation-Protected Securities]. If you’re worried about the strength of the U.S. dollar, then gold would be a good thing to own. It’s a very bad thing to own when things are good.

Stocks don’t have to do well in the future because they did well in the past. In fact, the opposite may be more likely.
As you know, I have my doubts about the certainty so many investors feel about the long-run attractions of investing in stocks. We do not know what is going to happen over the long run, never have, never will
https://jasonzweig.com/a-long-chat-with ... ernstein/

The rationale for diversification is the fact we are all ignorant about investment returns, and we want to protect ourselves as much as possible from our own ignorance by avoiding overconcentration in just one type of investment. Will that result in a "free lunch?" Maybe. Will that result in lousier results than we might have gotten? Maybe.
That’s what diversification is for. It’s an explicit recognition of ignorance.

— Peter Bernstein
While diversification has been called the “only free lunch in investing”, it doesn’t eliminate the risk of losses. And diversification does require accepting the fact that parts of your portfolio will behave entirely differently than the portfolio itself and may underperform a broad market index (such as the S&P 500) for a very long time.

-- Larry Swedroe
https://www.evidenceinvestor.com/tag/peter-bernstein/
On the internet, nobody knows you're a dog.

garlandwhizzer
Posts: 2558
Joined: Fri Aug 06, 2010 3:42 pm

Re: Diversification is not always, automatically, a free lunch

Post by garlandwhizzer » Sat Jan 11, 2020 12:31 pm

nisi wrote:

If someone point out some fund or asset class as being "a strong diversifier, with low correlation with stocks," don't let that lull you into thinking "so it must be good." You have to ask hard questions. It needs to be at least a fairly good investment in itself, and you need to believe that both "decent return" and "low correlation" will continue into the future.
1+

I think that sums it up pretty well. Personally, I believe the current investor obsession with diversification is overdone. As nisi's graphs point out, widely diversified portfolios with commodities, alternates, option strategies, may not only underperform traditional bond/stock portfolios, but while doing so, may not actually decrease volatility or maximal drawdown. Its diversification benefit is time period dependent. Although commodity exposure worked well in the commodity super-cycle 1996 -2006, it has been a powerful loser since that time. The question is how much of a price do you pay for the diversification benefit. Over the last decade relative to a balanced portfolio IMO that price has been too high.

Frankly it doesn't bother me that stocks go up and down in cycles. In the long run in the US and the vast majority of other countries they go up which is sufficient reassurance for me. Diversification apart from broadly based US and INTL equity, is usually a downdraft during bull markets which turns out to be most of the time. In severe bear markets all risk assets tank, approaching a correlation of 1.0. Alternates do not reliably hold up in such conditions which is when you really need diversification to an equity portfolio. Real estate, limited partnerships, P2P lending, other alternates all tank together at such times. Quality bonds are usually the only refuge and that is where money flows to as if flows away from risk. In general making your portfolio more complicated beyond traditional broadly diversified stocks and quality bonds, adding more moving "non-correlated" parts does not automatically improve it IMO. Sometimes the opposite happens. The appeal of alternate diversifiers IMO is often more theoretical than real.

Garland Whizzer

User avatar
bogglizer
Posts: 271
Joined: Tue Aug 16, 2016 8:56 pm

Re: Diversification is not always, automatically, a free lunc

Post by bogglizer » Sat Jan 11, 2020 1:06 pm

rkhusky wrote:
Fri Jan 10, 2020 10:13 am
Since stock market returns have no known statistical distribution, there is no known underlying correlation, and all you have is sample correlation.
I don't think you meant what you wrote. Everything random has a statistical distribution, and we have tons of data for stocks. Perhaps what you mean is that stock distributions might not have a finite variance due to fat tails? Also, stocks are highly correlated, especially with each other, and many of these correlations are highly understood. When gold goes up, gold mining companies that haven't sold ahead also go up. If people stopped buying TVs, LG and Samsung would suffer together.
My intellectual superiority will shortly be demonstrated here by a pithy quotation.

rkhusky
Posts: 7825
Joined: Thu Aug 18, 2011 8:09 pm

Re: Diversification is not always, automatically, a free lunc

Post by rkhusky » Sat Jan 11, 2020 2:09 pm

bogglizer wrote:
Sat Jan 11, 2020 1:06 pm
Everything random has a statistical distribution,
I don't believe that is true. To have a statistical distribution, you should be able to write down an equation that describes the distribution and you should be able to make a plot of the distribution. If you have a stationary process, you should get essentially the same distribution for different samplings of the data. If you have a non-stationary process, you should be able to describe how the sample distributions change from one sample to another. If data has a statistical distribution, you should be able to predict the distribution characteristics of out-of-sample data sets from analysis of in-sample data sets.

From what I understand, if you compare different time periods of stock market returns, you will not be able to mathematically describe the relationship between the distribution characteristics of different data samples. Specifically, you will not be able to accurately predict the statistical characteristics of future stock market returns, based on the statistical characteristics of past stock market returns.

User avatar
bogglizer
Posts: 271
Joined: Tue Aug 16, 2016 8:56 pm

Re: Diversification is not always, automatically, a free lunc

Post by bogglizer » Sat Jan 11, 2020 2:44 pm

rkhusky wrote:
Sat Jan 11, 2020 2:09 pm
bogglizer wrote:
Sat Jan 11, 2020 1:06 pm
Everything random has a statistical distribution,
I don't believe that is true. To have a statistical distribution, you should be able to write down an equation that describes the distribution and you should be able to make a plot of the distribution. If you have a stationary process, you should get essentially the same distribution for different samplings of the data. If you have a non-stationary process, you should be able to describe how the sample distributions change from one sample to another. If data has a statistical distribution, you should be able to predict the distribution characteristics of out-of-sample data sets from analysis of in-sample data sets.

From what I understand, if you compare different time periods of stock market returns, you will not be able to mathematically describe the relationship between the distribution characteristics of different data samples. Specifically, you will not be able to accurately predict the statistical characteristics of future stock market returns, based on the statistical characteristics of past stock market returns.
Stationary is not a requirement to have a distribution. Nor that the distribution of data in the past will predict future behaviour. The values that Exxon-Mobile takes over the next year form its statistical distribution for the next year, even though we don't know what it is yet. In log-space, the variance may be infinite, because Exxon may go bankrupt.
My intellectual superiority will shortly be demonstrated here by a pithy quotation.

heyyou
Posts: 3595
Joined: Tue Feb 20, 2007 4:58 pm

Re: Diversification is not always, automatically, a free lunch

Post by heyyou » Sat Jan 11, 2020 2:56 pm

At some point (older ages), one may have enough assets-to-necessary-income ratio (low enough relative expenses), to just ride out the stock fluctuations by spending from the bond side, without any other supposedly diversifying assets.
Striving for perfection of diversification could blind some to the advantage of just tolerating stock price reductions with bond fund sales for income, as needed.

JBTX
Posts: 5714
Joined: Wed Jul 26, 2017 12:46 pm

Re: Diversification is not always, automatically, a free lunch

Post by JBTX » Sat Jan 11, 2020 5:09 pm

As a long suffering investor in PIMCO commodities real return, I have become more skeptical of funds in a commodity sector being able to produce significant long term real returns.

But putting that aside, most of the discussions basically assume our investing horizon will be some subset of the last 20 years. Nobody really discusses an inflation scenario, nobody really discusses a Japan like scenario. If you just assume away that inflation never returns, then there is little reason to invest in commodities, gold, REITS, TIPS, ETC. (and for the record I suspect the probability of inflation is really low any time soon). If you just assume away that the US will always do better than the rest of the world, or our economy will never suffer a country specific shock, then sure international stocks have limited value in a portfolio.

But even though I don't think either of those will happen, I don't assume they won't.

User avatar
grabiner
Advisory Board
Posts: 25615
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: Diversification is not always, automatically, a free lunch

Post by grabiner » Sat Jan 11, 2020 5:32 pm

nisiprius wrote:
Fri Jan 10, 2020 9:52 am
Now, some will say "you deliberately chose stinker of a diversifier to make your point," and so I did. The point is, it was a diversifier, yet it did not improve the portfolio as a whole? Why not? The answer is obvious: PCRIX, since inception, has been such a lousy investment in itself.

Image
What this graph shows is that PCRIX failed as a diversifier. A good diversifier for a stock portfolio may be volatile, but it should not have a huge loss at the same time the stock market does.

I don't know whether this should be expected to happen in the next market crash.
Wiki David Grabiner

Uncorrelated
Posts: 249
Joined: Sun Oct 13, 2019 3:16 pm

Re: Diversification is not always, automatically, a free lunch

Post by Uncorrelated » Sat Jan 11, 2020 5:42 pm

Even if you assume that inflation will be high in the future, I don't see how that helps the case for commodities. The correlation between real gold returns and inflation hints is very week at best (95% confidence interval (-.54, .66)). I don't have data for other assets.

The inflation angle gets played quite a lot here, but I have yet to see an evidence that inflation is relevant for any asset class besides bonds.

User avatar
305pelusa
Posts: 1247
Joined: Fri Nov 16, 2018 10:20 pm

Re: Diversification is not always, automatically, a free lunch

Post by 305pelusa » Sat Jan 11, 2020 6:01 pm

Uncorrelated wrote:
Sat Jan 11, 2020 5:42 pm
Even if you assume that inflation will be high in the future, I don't see how that helps the case for commodities. The correlation between real gold returns and inflation hints is very week at best (95% confidence interval (-.54, .66)). I don't have data for other assets.

The inflation angle gets played quite a lot here, but I have yet to see an evidence that inflation is relevant for any asset class besides bonds.
I agree that the link between inflation and gold has been pretty weak. And conceptually, I'm not really sure why it would have to be strong.

But a basket of commodities, conceptually at least, I would imagine has a strong relationship with inflation since that IS roughly what inflation is. Something like BCI includes oil, food (soybean/pork/meat/corn/etc), and various types of metals. And how does high inflation affect the consumer? Why with high gas prices, food prices and so on. Again, roughly speaking at least.

That's fairly handwavy but the point is the same. High inflation is when the prices of the goods/services you consume go up. So if you buy a product that has high exposure to lots of different goods, it would be reasonable to expect some hedging IMO.

As for past data supporting it, some writers like Swedroe have written about how gold is pretty poor hedge for inflation, yet commodities actually do have a reasonable correlation to inflation:
https://www.etf.com/sections/index-inve ... -inflation

I don't invest in commodities (it's hardly an "investment" in my book) but I buy the logic behind it. It's just not convincing enough for me rn.

rkhusky
Posts: 7825
Joined: Thu Aug 18, 2011 8:09 pm

Re: Diversification is not always, automatically, a free lunc

Post by rkhusky » Sat Jan 11, 2020 7:51 pm

bogglizer wrote:
Sat Jan 11, 2020 2:44 pm
Stationary is not a requirement to have a distribution. Nor that the distribution of data in the past will predict future behaviour. The values that Exxon-Mobile takes over the next year form its statistical distribution for the next year, even though we don't know what it is yet. In log-space, the variance may be infinite, because Exxon may go bankrupt.
I never said stationarity was required; note that I mentioned non-stationary distributions. But if you can't predict the future distribution, then you don't really know the true distribution; all you have is an approximation to the true distribution, which could be completely different from the true distribution.

User avatar
bogglizer
Posts: 271
Joined: Tue Aug 16, 2016 8:56 pm

Re: Diversification is not always, automatically, a free lunc

Post by bogglizer » Sat Jan 11, 2020 9:24 pm

rkhusky wrote:
Sat Jan 11, 2020 7:51 pm
bogglizer wrote:
Sat Jan 11, 2020 2:44 pm
Stationary is not a requirement to have a distribution. Nor that the distribution of data in the past will predict future behaviour. The values that Exxon-Mobile takes over the next year form its statistical distribution for the next year, even though we don't know what it is yet. In log-space, the variance may be infinite, because Exxon may go bankrupt.
I never said stationarity was required; note that I mentioned non-stationary distributions. But if you can't predict the future distribution, then you don't really know the true distribution; all you have is an approximation to the true distribution, which could be completely different from the true distribution.
You can never predict the future distribution, which is no more than the values it will take. That's what random is.
My intellectual superiority will shortly be demonstrated here by a pithy quotation.

rkhusky
Posts: 7825
Joined: Thu Aug 18, 2011 8:09 pm

Re: Diversification is not always, automatically, a free lunc

Post by rkhusky » Sat Jan 11, 2020 9:43 pm

bogglizer wrote:
Sat Jan 11, 2020 9:24 pm
rkhusky wrote:
Sat Jan 11, 2020 7:51 pm
bogglizer wrote:
Sat Jan 11, 2020 2:44 pm
Stationary is not a requirement to have a distribution. Nor that the distribution of data in the past will predict future behaviour. The values that Exxon-Mobile takes over the next year form its statistical distribution for the next year, even though we don't know what it is yet. In log-space, the variance may be infinite, because Exxon may go bankrupt.
I never said stationarity was required; note that I mentioned non-stationary distributions. But if you can't predict the future distribution, then you don't really know the true distribution; all you have is an approximation to the true distribution, which could be completely different from the true distribution.
You can never predict the future distribution, which is no more than the values it will take. That's what random is.
In physical phenomena, once you know the true distribution, it doesn't change. For example, if you have a process that follows a Gaussian distribution, then it will follow a Gaussian distribution in the future. If it is stationary, the Gaussian parameters won't change. If it is non-stationary, then you will know how the parameters change over time. You won't know how it will perform in the future, since it is random, but you will know the probabilities of different scenarios.

With stocks, not only do you not know what will happen, but you don't even know the probabilities for different events occurring. Because you don't know how the distributions will change in the future.

This is the problem with applying statistics to stock investing.

User avatar
nedsaid
Posts: 12964
Joined: Fri Nov 23, 2012 12:33 pm

Re: Diversification is not always, automatically, a free lunch

Post by nedsaid » Sat Jan 11, 2020 10:03 pm

Part of the problem here is the fight the last war problem. Larry Swedroe is old enough to remember the inflation from the 1970's and I am too. I am an inflation hawk from way back. Larry's memories of the oil shocks and the bear market of 1973-74 are pretty well seared into his consciousness and this was reflected in earlier recommendation for Collateralized Commodity Futures and the Pimco Real Return fund. Somehow the Maginot Line designed for a repeat of World War I wasn't much of a defense against General Guderian's Panzer Divisions in World War II. What would have worked great in World War I didn't work so great in World War II. In Larry's case, then Federal Reserve Chief Paul Volcker slayed the inflation dragon and the dreaded 1970's era Stagflation never returned. In retrospect, Larry looks a bit like those French Generals and the designer of the Maginot Line.

So I am a bit like that, I have my REITs and TIPS locked and loaded and ready to fire at inflation when I see the whites of its eyes. So I stick with my old fashioned investments while my fellow Bogleheads have mostly moved on. Inflation seems pretty tame now and it has been for some time. Indeed, we were actually starting to worry about deflation. But by golly, I remember the 1970's. Inflation will rear its ugly head any minute now!
A fool and his money are good for business.

Seasonal
Posts: 628
Joined: Sun May 21, 2017 1:49 pm

Re: Diversification is not always, automatically, a free lunch

Post by Seasonal » Mon Jan 13, 2020 3:14 pm

nisiprius wrote:
Fri Jan 10, 2020 9:52 am
...
Over the lifetime of PCRIX their correlation has been 0.40. (That's in the range that people call "low" when they are advocating for diversifiers; for example, over that same time period, the Vanguard Real Estate fund had a correlation of 0.68 with stocks; Vanguard Total International, 0.88; and DFA US Small-cap Value Portfolio, 0.91). (Source)

There are other ways to measure diversification, but I doubt that anyone can find a measure by which PCRIX hasn't qualified as "a diversifier."...
It really depends on how you define diversifier. For example, if you interpret Fama's oft-repeated statement that the market portfolio is always on the efficient frontier to mean the market portfolio is completely diversified, then overweighting PCRIX would not increase diversification.

If you believe a portfolio can be both on the efficient frontier and can be improved by additional diversification, please explain your definitions.

rossington
Posts: 397
Joined: Fri Jun 07, 2019 2:00 am
Location: Florida

Re: Diversification is not always, automatically, a free lunch

Post by rossington » Tue Jan 14, 2020 4:49 am

nedsaid wrote:
Sat Jan 11, 2020 10:03 pm
Part of the problem here is the fight the last war problem. Larry Swedroe is old enough to remember the inflation from the 1970's and I am too. I am an inflation hawk from way back. Larry's memories of the oil shocks and the bear market of 1973-74 are pretty well seared into his consciousness and this was reflected in earlier recommendation for Collateralized Commodity Futures and the Pimco Real Return fund. Somehow the Maginot Line designed for a repeat of World War I wasn't much of a defense against General Guderian's Panzer Divisions in World War II. What would have worked great in World War I didn't work so great in World War II. In Larry's case, then Federal Reserve Chief Paul Volcker slayed the inflation dragon and the dreaded 1970's era Stagflation never returned. In retrospect, Larry looks a bit like those French Generals and the designer of the Maginot Line.

So I am a bit like that, I have my REITs and TIPS locked and loaded and ready to fire at inflation when I see the whites of its eyes. So I stick with my old fashioned investments while my fellow Bogleheads have mostly moved on. Inflation seems pretty tame now and it has been for some time. Indeed, we were actually starting to worry about deflation. But by golly, I remember the 1970's. Inflation will rear its ugly head any minute now!
But now we are energy independent as the largest producer of oil in the world.
Inflation will rear its ugly head any minute now!
Not from energy.... as was the primary inflation driver in the the 70's. Fracking, drilling, technology has replaced our "Maginot Line" and defeated the Mideast OPEC "Panzers".
I have my REITs and TIPS locked and loaded and ready to fire at inflation when I see the whites of its eyes.

REITS ("Quality" REIT companies) and TIPS (....maybe) seem fine to me.
The question is where is the inflation coming from anytime soon?
So I stick with my old fashioned investments while my fellow Bogleheads have mostly moved on.
To what? (Do you mean....just curious).
"Success is going from failure to failure without loss of enthusiasm." Winston Churchill.

Post Reply