Here's a statement of the obvious: The opinions expressed here are those of the participants, not those of the Mutual Fund Observer. We cannot vouch for the accuracy or appropriateness of any of it, though we do encourage civility and good humor.
You might want to visit and exercise the tools accessible on the PortfolioVisualizer website to secure evidence of diversification benefits. Here is the Link to the site:
That website offers several tools that will allow you to test the benefits as a function of time and as a function of portfolio construction. The benefits change depending on timeframe.
You might find the asset class correlations of some value. One shortfall of that option is the historical limits of the database. It doesn't go back very far in time.
You can postulate various portfolio constructions ( 3 at a time) and explore their returns over various timeframes using another site option. For some time periods diversification doesn't do much. For other periods and for longer timespans diversification delivers the goods.
Sorry for my brief and incomplete reply to your question. Today is a happy day for our family. We will celebrate my beautiful and intelligent wife and my 55th Wedding Anniversary with a seafood lunch at the Redondo Beach pier. By the way, her consistently wise decisions and persistent saving tactics were major contributors to our retirement portfolio.
As the old saying goes, it is always a good practice to learn lessons from your experiences, but it is an even better practice to learn from other folks’ mistakes and experiences. I do that by reading extensively.
I too have applied the PortfolioVisualizer toolkit to a variety of investment issues and options, but I have not done so exhaustively. My studies have been highly focused on my special needs. So my takeaways from that limited work are very generic.
Those takeaways take the form of broad and rather conventional investment axioms. For example, portfolio construction should be a top-down exercise, the 80/20 rule operates in the mutual fund domain, reversion-to-the-mean happens, simplicity conquers complexity, a diminishing returns rule applies, and, yes, diversification is possibly the marketplace’s only free lunch.
Actually, the excellent Periodic Table of Investment Returns that you referenced serves to demonstrate many of these same general lessons. Doing specific what-if scenarios on tools like those provided by the PortfolioVisualizer website add meat to the bones.
The historical data sets and the tools used to analyze that data demonstrate the limitations of Golden Investment rules. Investment Golden rules do exist, but they are punctured by frequent exceptions.
Almost every study ever rigorously completed concludes that a portfolio’s returns are mostly controlled by the basic asset allocation decision, and not specific fund selections. Index fund investing avoids the likely low-profit time sink of active fund study and selection decisions.
Most active fund managers fail to match their benchmarks (the 80/20 rule wins once again). With a few exceptions, a reversion-to-the-mean applies. Persistent top performance is rare. That’s true for fund managers and for major investment classes. Your Periodic Table reference provides an excellent illustration of annual reversals for categories.
A simple investment policy is less costly, less time intensive, and easier to monitor and execute than a complex one. As Albert Einstein said: “Everything should be made as simple as possible. But not simpler”. The challenge is to identify that “simpler” threshold.
The law of diminishing returns applies here. Harry Markowitz explored this investment dimension in his early 1950s studies. His Mean-Variance Modern Portfolio Theory model set the stage for the more recent work. Adding complexity (numbers) to a portfolio quickly falls victim to the diminishing returns phenomena. Markowitz demonstrated that effect for individual stock holdings.
The various Lazy Portfolios serve the same purpose for a mutual fund investor. Using the PortfolioVisualizer portfolio performance tool, the diminishing returns principle can be verified by any user of that fine tool. Outcomes are sensitive to the study timeframe.
I have done a few portfolio diversification calculations on the referenced website. It works in theory. It also works in practice. Anecdotally, the broad diversification that I implement in our family portfolio worked to our advantage in the 2000-2002 period market meltdown. Our portfolio delivered positive returns in that depressing period. That same diversification plan failed to completely protect us during the 2008 horrid year.
Nothing in the investment universe works universally each and every time.
But to return to the original question: What are the notable (actionable) benefits of diversification the last decade and more. In other words, reasons not to simplify, with age, into US LC indexlike holdings.
Sorry that I did not answer your diversification benefit question to your satisfaction. Sometimes it’s difficult to gauge exactly what an MFOer knows and what he needs to fill the gaps in his understanding. My second attempt will be hopefully more directly to the point. I hope my lunch wine will not compromise my response too much.
The primary benefit of diversification comes from the historical observation that various asset classes do not respond and behave identically to changing market conditions. The numerical representation of that observation is that correlation coefficients differ from a perfect value of One. And correlation coefficients are one factor in determining a portfolio’s standard deviation.
That fact is nicely captured in capsule form by the Periodic Table of Investment annual returns. That graphical representation demonstrates the random mobility within the ordering of annual returns; what is at the top of the totem pole one year is not the same asset the following year. Change happens.
Your referenced Periodic Table nicely demonstrates that mobility. But wait, that same Table also demonstrates that a combination of holdings, like a balanced set of funds significantly attenuates the annual fluctuations. That is a reduction in the portfolio’s overall standard deviation.
When assembling a portfolio, the primary benefit of adding all those various asset class holdings is to reduce portfolio volatility (standard deviation) without seriously compromising projected annual return. This comes directly from Harry Markowitz’s work. The tradeoff between risk, as measured by standard deviation, and portfolio returns always exists.
Mathematically, a reduction in standard deviation translates into a higher cumulative annual compound return since standard deviation operates to subtract from average annual return according to the approximate mathematical equation that annual compound return equals annual average return minus one-half the square of the portfolio’s standard deviation. Standard deviation always subtracts from average annual return taken over time. Compound annual return determines end wealth.
Emotionally, a reduction in standard deviation often translates into an investor “staying the course”. Wild annual returns are reduced and an investor is typically more comfortable with such a portfolio. He is less inclined to panic at the wrong time. In representative cases, a portfolio’s standard deviation can be shortened by a factor of 50% while only sacrificing a 10% projected annual return. Many investors favor such a tradeoff.
I hope this explanation satisfies your discomfort with the concept and benefits of portfolio diversification. I had assumed a level of understanding that was apparently not present. My error.
I also hope the wine has not distorted my reply too much.
But to first order the classes do correlate for the most part, depending on what one means and what the criteria for performance/outperformance are. No discomfort on my part (reading). My question was about actual $ benefits. I never know what level of understanding is present in me, apparent or invisible. I was asking a little more concretely than a general sermon on diversification = lower volatility. So I perhaps should have phrased it, Are the smoothing benefits worth doing? What have we gotten the last 10-20y by augmenting good-quality US large caps with the other classes? Going beyond sleep-at-night, will it likely prove worth it for 15-20y to put all noncash into NOBL (say), or into AOR (say)? I was looking for specific responses, in other words. My investing penchants reduce extreme volatility perforce (divs etc.). The bromide about stddev reduction = higher returns would of course indicate low-vol investing for all, yet one regularly reads such as this from a WSJ article last fall: In an analysis last year, Prof. Novy-Marx found that in a 45-year span ended in 2013, moderately volatile stocks performed fairly similarly to stocks with the lowest volatility—and significantly better than the most volatile stocks. (A key reason is that the most volatile stocks often are those of unprofitable companies classified as “small growth” stocks.) In large cap value, the study added, aggressive stocks actually outpaced more-defensive issues. So I was interested in that. Why ever bother with AOR or GLRBX or FPACX instead of betting the house on NOBL or FCNTX or PRBLX or, with a longer term, FLPSX? Why diversify further, given the recent past? Maybe these are unanswerable ponderings.
Aha! Thanks to your more detailed description of your original question, I believe I have finally had a Eureka Moment. I suspect that I can now provide a better reply to the thrust of your original post.
In a sense, you are asking if portfolio concentration in an anticipated outperforming equity class is superior in terms of expected returns over a more diversified equity positioning. The debate about concentration vs. diversification has raged for decades with both sides capable of referencing studies that favor their judgment.
Much depends on timeframe. A universal answer likely does not exist, although most investment gurus support some diversification. The law of diminishing returns kicks in rather early. One danger of diversification can be characterized as over-diversification which operates to negate the initial benefits from some diversification.
Regardless, my answer to your question is “Yes”, a portfolio for most investors would benefit from the more common diversification equity classes. Note I restricted my answer to most investors; smart investors like Warren Buffett are excluded from my generic reply.
My “Yes” response is based on a few simulations I completed using PortfolioVisualizer tools, specifically to address your question. To simplify the task, I postulated 4 all equity portfolios and used historical equity class (Index like) returns.
A small fraction of active fund managers will deliver better results, but to project which ones and by how much is hazardous duty. The random-like checkerboard pattern that the Periodic Returns Tables typically display and the DALBAR studies support my contention.
The 4 portfolios that I explored were: (1) the S&P 500 Index, (2) a 100% Large Cap Blend, (3) 80% Large Cap Blend, 10% Large Cap value, 10% Small Cap Value, and (4) 40% Large Cap Blend, 10% Large Cap Value, 10% Small Cap Value, 10% Mid-Cap Blend, 10% REIT, 10% Int’l Stocks, 10% Emerging Markets.
No attempt was made to optimize the Number (4) portfolio. The primary aim was to diversify the class holdings. The portfolio leanings towards Value products reflect the famous Fama-French findings.
I ran 4 timeframes, all ending in 2015. The 4 periods were 10, 15, 20, and 25 years long. Standard deviations differed for each period and for each portfolio.
My “Yes” answer is based on the superior returns produced by diversified portfolio Number (4). For all 4 timeframes, that diversified portfolio generated Compound Annual Growth Rates (CAGR) in excess of those recorded for the other portfolios.
Will that always be the case? Probably not. Unlike the laws that govern physics, investing has no universal laws. Exceptions to investment rules can always be discovered.
Tyvm. I think the answer is probably yes, always the case. Interesting that 4, the diverse one, prevailed for 10y, not just the longer periods. What approximately were the excess percentages or overages? Merriman should have some MJG portfolios, I say.
When you're done viewing this for long term performance, scroll up again, select the "risk" tab and when that loads, scroll down to view the upside/downside.
The category rank isn't displaying at the time of this post, but it is "1's and 2's" back 15 years.
I would need to see documented proof that very many average, retail investors have obtained a better than 7% annualized return from 10 years ago; which obviously contains the "market melt". Only the very smart or very patient, and/or the combo of.....
An amazing achievement for a 40/60 vehicle, yes, especially its first decades after 1970. Although I would not have been able to retire had I limited myself to it. (Nor, for that matter, to SPX.) And yeah, we are now right at the 8-9y breakpoint for US LC vs it. It does show "notable (actionable) benefits of diversification the last decade", unless you were lucky, and there was no way to choose luck's path: one could just as well have been in the highly recommended DODGX or WVALX or SSHFX (which it outperforms) as in the equally recommended FCNTX or TWEIX or JENSX (which outperform it).
Agree with you that Merriman's life work has been to try and maximize the right mix. Might check out his site particularly the withdrawal table scenarios. Interesting stuff.
I recently added it to my Ira after transferring it to Fidelity from merrill. They did not offer some funds I wanted to add to my ira, like VWINX and SFGIX so switched to Fido and bought them.
$50 now. edit: that's in general; this fund $75 as noted below, tnx.
I am split b/w ML and Fido; each has advantages and offerings. ML offers lots of completely free ETF and stock purchases w enough holdings, plus some fund deals Fido does not. E.g., if PRWCX were not closed, I could buy it NTF/NL.
'Course, if one's account is set with the brokerage feature (all accounts that qualify, should be); one may travel just about any direction seeking the "money grail".
@MJG Thanks as always for your thoughtful, detailed, and highly convincing comments. I noticed you didn't include any bonds in your model portfolios. Was that because this discussion is just about stocks, because your analysis is that bonds have no role in a long-term investor's portfolio?
@Old_joe: Normally VWINX costs $75 at Fidelity, but because I transferred monies to them, they are giving me some free trades. I asked for them before I transferred my ira. Vanguard generally does not participate in revenue sharing with brokerages so they are not ntf most places, with exceptions in some of their etfs.
Diversification is good under most circumstances, over-diversification can ruin that goodness, but on the other hand, concentration might offer benefits depending on the proclivities of an investor.
Some investors accumulate mutual funds like a housewife accumulates recipes. That’s especially true if the investor favors active fund management. That might not be a good idea. And that’s because of the 80/20 rule which loosely applies to active fund managers.
The accumulated data strongly supports the observation that only about 20% of fund managers deliver returns in excess of appropriate benchmarks. The remaining 80% fail that meaningful target. The managers filling the elite 20% pool constantly change and are difficult to identify in advance of their superior performance.
Monte Carlo methods are ideal tools to use in modeling the inherent uncertainties that are characteristic of this issue . This has been done several times. One earlier effort was completed by financial advisor Allan Roth. More recently Richard Ferri added his version of a Monte Carlo study.
Allan Roth studied active fund returns and measured their success against relevant benchmarks. He explored two dimensions of the problem: holding time and number of funds in a portfolio. His Monte Carlo simulation examined timeframes from 1 to 25 years and number of portfolio funds from 1 to 5 to 10.
Roth concluded that those funds that outdistanced their benchmarks eroded as a function of time and as a function of the number of active funds in the portfolio. After 25 years, only 1% of a 10 active fund portfolio outperformed its benchmark equivalent. At shorter holding periods of 10 years and 5 years, the odds of outperforming benchmarks improved to 6% and 9%, respectively.
These are dire results indeed. The odds definitely tilt against active fund management. As an investor, do you believe you can be in the prescient 1% cohort? That’s not me.
The key summary findings of the Ferri work are that (1) “Index funds have a higher probability of outperforming actively managed funds”. (2) “The probability of index fund portfolio outperformance increased when the time period was extended from 5 years to 15 years”, and (3) “The probability of index fund portfolio outperformance increased when two or more actively managed funds were held in each asset class.”
These conclusions reinforce the Roth results. If you want to integrate active fund management into your portfolios, these Monte Carlo simulations make a strong case to severely limit the number of those active fund managers. In this instance, the less, the better is good policy. So diversification in actively managed funds hurt end wealth.
Thank you for your patience. I have referenced these Monte Carlo studies in earlier postings. They warrant repeating.
MFOer Expatsp asked if I only included stocks in my analysis because I believed bonds had no place in a portfolio? No, not so! I merely limited my brief study to equities to simplify the analysis. My portfolio includes major fixed income components.
Old Joe & Slick - From the Fidelity website re: Transaction fees on mutual fund purchases
1. Transaction Fee:
A transaction fee is similar to a brokerage fee or commission which you pay when you buy or sell a stock. For some funds available through Fidelity you are required to pay a transaction fee. However, you will not pay a sales load on Transaction Fee (TF) funds. You will only be charged a transaction fee when you buy a FundsNetwork TF fund, not when you sell one. All other fees and expenses described in a fund's prospectus still apply. You can choose to buy or sell shares directly from the fund itself or its principal underwriter or distributor without paying a transaction fee to Fidelity.
Online Transaction Fees: $49.95 for most funds. Certain funds will have a transaction fee of $75. To identify any applicable transaction fees associated with the purchase of a given fund, please refer to the "Fees and Distributions" tab.
Fidelity Automated Service Telephone (FAST): 25% off representative-assisted rates, Maximum: $187.50, Minimum:$75 Representative-Assisted: 0.75% of principal, Maximum: $250, Minimum: $100 Automatic Investment: $5 per transaction, after the initial investment.
Slick is correct in this case, the fee will be $75. I checked because all of my previous buys of non-NTF funds have been $49.95.
What you could do Joe is open the account at Vanguard and then have the VWINX fund transferred to Fidelity. Additional "automatic" purchases would then only cost you $5 per purchase. If you intend to add to your fund with random purchases those would cost you $75 each however. Something about no free lunch.
In my last post I mentioned the yeoman work that Allan Roth performed about a decade ago. I failed to provide a specific reference to his work or his interpretation of it. I just located a very brief commentary that Roth made to the AARP community last year. He is still singing the same song, and solidly defends his earlier pathfinder study. Here is a Link to the article:
When making any financial decision, it’s always a good policy to “Play the odds when investing”. The Roth study definitely made a few assumptions and simplifications that not everyone would agree too. But that’s true for any modeling study. His work does properly capture the significant tradeoffs.
When comparing various Monte Carlo simulations, specific outcome numbers change depending on modeling details and timeframe dates, but the fundamental outcome trends and conclusions are consistent.
I recognize that some of you guys distrust Monte Carlo analysis feeling that it is too sophisticated mathematically. In reality it is none of that. It is conceptually simple. However, it can only be executed for the necessary thousands of random portfolio choices because of the speed of modern computers.
Many MFOers did spreadsheet portfolio analyses when making retirement decisions and didn’t fully realize they were doing a very limited (actually a too limited) Monte Carlo type analysis using estimates for yearly portfolio returns. Monte Carlo codes yield a more reliable base for that crucial retirement decision simply because it completes that same analysis thousands of times.
Comments
You might want to visit and exercise the tools accessible on the PortfolioVisualizer website to secure evidence of diversification benefits. Here is the Link to the site:
https://www.portfoliovisualizer.com
That website offers several tools that will allow you to test the benefits as a function of time and as a function of portfolio construction. The benefits change depending on timeframe.
You might find the asset class correlations of some value. One shortfall of that option is the historical limits of the database. It doesn't go back very far in time.
You can postulate various portfolio constructions ( 3 at a time) and explore their returns over various timeframes using another site option. For some time periods diversification doesn't do much. For other periods and for longer timespans diversification delivers the goods.
Please give the reference a fair look see.
Best Wishes.
Sorry for my brief and incomplete reply to your question. Today is a happy day for our family. We will celebrate my beautiful and intelligent wife and my 55th Wedding Anniversary with a seafood lunch at the Redondo Beach pier. By the way, her consistently wise decisions and persistent saving tactics were major contributors to our retirement portfolio.
As the old saying goes, it is always a good practice to learn lessons from your experiences, but it is an even better practice to learn from other folks’ mistakes and experiences. I do that by reading extensively.
I too have applied the PortfolioVisualizer toolkit to a variety of investment issues and options, but I have not done so exhaustively. My studies have been highly focused on my special needs. So my takeaways from that limited work are very generic.
Those takeaways take the form of broad and rather conventional investment axioms. For example, portfolio construction should be a top-down exercise, the 80/20 rule operates in the mutual fund domain, reversion-to-the-mean happens, simplicity conquers complexity, a diminishing returns rule applies, and, yes, diversification is possibly the marketplace’s only free lunch.
Actually, the excellent Periodic Table of Investment Returns that you referenced serves to demonstrate many of these same general lessons. Doing specific what-if scenarios on tools like those provided by the PortfolioVisualizer website add meat to the bones.
The historical data sets and the tools used to analyze that data demonstrate the limitations of Golden Investment rules. Investment Golden rules do exist, but they are punctured by frequent exceptions.
Almost every study ever rigorously completed concludes that a portfolio’s returns are mostly controlled by the basic asset allocation decision, and not specific fund selections. Index fund investing avoids the likely low-profit time sink of active fund study and selection decisions.
Most active fund managers fail to match their benchmarks (the 80/20 rule wins once again). With a few exceptions, a reversion-to-the-mean applies. Persistent top performance is rare. That’s true for fund managers and for major investment classes. Your Periodic Table reference provides an excellent illustration of annual reversals for categories.
A simple investment policy is less costly, less time intensive, and easier to monitor and execute than a complex one. As Albert Einstein said: “Everything should be made as simple as possible. But not simpler”. The challenge is to identify that “simpler” threshold.
The law of diminishing returns applies here. Harry Markowitz explored this investment dimension in his early 1950s studies. His Mean-Variance Modern Portfolio Theory model set the stage for the more recent work. Adding complexity (numbers) to a portfolio quickly falls victim to the diminishing returns phenomena. Markowitz demonstrated that effect for individual stock holdings.
The various Lazy Portfolios serve the same purpose for a mutual fund investor. Using the PortfolioVisualizer portfolio performance tool, the diminishing returns principle can be verified by any user of that fine tool. Outcomes are sensitive to the study timeframe.
I have done a few portfolio diversification calculations on the referenced website. It works in theory. It also works in practice. Anecdotally, the broad diversification that I implement in our family portfolio worked to our advantage in the 2000-2002 period market meltdown. Our portfolio delivered positive returns in that depressing period. That same diversification plan failed to completely protect us during the 2008 horrid year.
Nothing in the investment universe works universally each and every time.
Best Wishes.
What are the notable (actionable) benefits of diversification the last decade and more. In other words, reasons not to simplify, with age, into US LC indexlike holdings.
Sorry that I did not answer your diversification benefit question to your satisfaction. Sometimes it’s difficult to gauge exactly what an MFOer knows and what he needs to fill the gaps in his understanding. My second attempt will be hopefully more directly to the point. I hope my lunch wine will not compromise my response too much.
The primary benefit of diversification comes from the historical observation that various asset classes do not respond and behave identically to changing market conditions. The numerical representation of that observation is that correlation coefficients differ from a perfect value of One. And correlation coefficients are one factor in determining a portfolio’s standard deviation.
That fact is nicely captured in capsule form by the Periodic Table of Investment annual returns. That graphical representation demonstrates the random mobility within the ordering of annual returns; what is at the top of the totem pole one year is not the same asset the following year. Change happens.
Your referenced Periodic Table nicely demonstrates that mobility. But wait, that same Table also demonstrates that a combination of holdings, like a balanced set of funds significantly attenuates the annual fluctuations. That is a reduction in the portfolio’s overall standard deviation.
When assembling a portfolio, the primary benefit of adding all those various asset class holdings is to reduce portfolio volatility (standard deviation) without seriously compromising projected annual return. This comes directly from Harry Markowitz’s work. The tradeoff between risk, as measured by standard deviation, and portfolio returns always exists.
Mathematically, a reduction in standard deviation translates into a higher cumulative annual compound return since standard deviation operates to subtract from average annual return according to the approximate mathematical equation that annual compound return equals annual average return minus one-half the square of the portfolio’s standard deviation. Standard deviation always subtracts from average annual return taken over time. Compound annual return determines end wealth.
Emotionally, a reduction in standard deviation often translates into an investor “staying the course”. Wild annual returns are reduced and an investor is typically more comfortable with such a portfolio. He is less inclined to panic at the wrong time. In representative cases, a portfolio’s standard deviation can be shortened by a factor of 50% while only sacrificing a 10% projected annual return. Many investors favor such a tradeoff.
I hope this explanation satisfies your discomfort with the concept and benefits of portfolio diversification. I had assumed a level of understanding that was apparently not present. My error.
I also hope the wine has not distorted my reply too much.
Best Wishes.
My investing penchants reduce extreme volatility perforce (divs etc.). The bromide about stddev reduction = higher returns would of course indicate low-vol investing for all, yet one regularly reads such as this from a WSJ article last fall:
In an analysis last year, Prof. Novy-Marx found that in a 45-year span ended in 2013, moderately volatile stocks performed fairly similarly to stocks with the lowest volatility—and significantly better than the most volatile stocks. (A key reason is that the most volatile stocks often are those of unprofitable companies classified as “small growth” stocks.) In large cap value, the study added, aggressive stocks actually outpaced more-defensive issues.
So I was interested in that. Why ever bother with AOR or GLRBX or FPACX instead of betting the house on NOBL or FCNTX or PRBLX or, with a longer term, FLPSX? Why diversify further, given the recent past? Maybe these are unanswerable ponderings.
Aha! Thanks to your more detailed description of your original question, I believe I have finally had a Eureka Moment. I suspect that I can now provide a better reply to the thrust of your original post.
In a sense, you are asking if portfolio concentration in an anticipated outperforming equity class is superior in terms of expected returns over a more diversified equity positioning. The debate about concentration vs. diversification has raged for decades with both sides capable of referencing studies that favor their judgment.
Much depends on timeframe. A universal answer likely does not exist, although most investment gurus support some diversification. The law of diminishing returns kicks in rather early. One danger of diversification can be characterized as over-diversification which operates to negate the initial benefits from some diversification.
Regardless, my answer to your question is “Yes”, a portfolio for most investors would benefit from the more common diversification equity classes. Note I restricted my answer to most investors; smart investors like Warren Buffett are excluded from my generic reply.
My “Yes” response is based on a few simulations I completed using PortfolioVisualizer tools, specifically to address your question. To simplify the task, I postulated 4 all equity portfolios and used historical equity class (Index like) returns.
A small fraction of active fund managers will deliver better results, but to project which ones and by how much is hazardous duty. The random-like checkerboard pattern that the Periodic Returns Tables typically display and the DALBAR studies support my contention.
The 4 portfolios that I explored were: (1) the S&P 500 Index, (2) a 100% Large Cap Blend, (3) 80% Large Cap Blend, 10% Large Cap value, 10% Small Cap Value, and (4) 40% Large Cap Blend, 10% Large Cap Value, 10% Small Cap Value, 10% Mid-Cap Blend, 10% REIT, 10% Int’l Stocks, 10% Emerging Markets.
No attempt was made to optimize the Number (4) portfolio. The primary aim was to diversify the class holdings. The portfolio leanings towards Value products reflect the famous Fama-French findings.
I ran 4 timeframes, all ending in 2015. The 4 periods were 10, 15, 20, and 25 years long. Standard deviations differed for each period and for each portfolio.
My “Yes” answer is based on the superior returns produced by diversified portfolio Number (4). For all 4 timeframes, that diversified portfolio generated Compound Annual Growth Rates (CAGR) in excess of those recorded for the other portfolios.
Will that always be the case? Probably not. Unlike the laws that govern physics, investing has no universal laws. Exceptions to investment rules can always be discovered.
I hope this adequately addresses your question.
Best Wishes.
I think the answer is probably yes, always the case.
Interesting that 4, the diverse one, prevailed for 10y, not just the longer periods.
What approximately were the excess percentages or overages?
Merriman should have some MJG portfolios, I say.
Longer timeframes, as you say; otherwise:
http://www.marketwatch.com/lazyportfolio
Put VWINX or its Admiral mate (if you can) under your pillow and sleep well.
http://performance.morningstar.com/fund/performance-return.action?t=VWINX®ion=usa&culture=en-US
When you're done viewing this for long term performance, scroll up again, select the "risk" tab and when that loads, scroll down to view the upside/downside.
The category rank isn't displaying at the time of this post, but it is "1's and 2's" back 15 years.
This Fidelity link is for the most recent reported composition for VWINX:
https://fundresearch.fidelity.com/mutual-funds/composition/921938106?type=sq-NavBar
One has to "love" the ticker, eh? vWINx
I would need to see documented proof that very many average, retail investors have obtained a better than 7% annualized return from 10 years ago; which obviously contains the "market melt". Only the very smart or very patient, and/or the combo of.....
Take care,
Catch
It does show "notable (actionable) benefits of diversification the last decade", unless you were lucky, and there was no way to choose luck's path: one could just as well have been in the highly recommended DODGX or WVALX or SSHFX (which it outperforms) as in the equally recommended FCNTX or TWEIX or JENSX (which outperform it).
Thanks- OJ
I am split b/w ML and Fido; each has advantages and offerings. ML offers lots of completely free ETF and stock purchases w enough holdings, plus some fund deals Fido does not. E.g., if PRWCX were not closed, I could buy it NTF/NL.
Not sure about the "Fido does not" you noted:
If the 70 etfs (these are the I-shares list, there are more) in this list won't do it for most folks, they're hav'in more fun than I.
https://www.fidelity.com/etfs/ishares
'Course, if one's account is set with the brokerage feature (all accounts that qualify, should be); one may travel just about any direction seeking the "money grail".
Catch, no period b/w 'deals' and 'Fido'; simply meant ML offers some vehicles for free which Fidelity does not.
Vanguard generally does not participate in revenue sharing with brokerages so they are not ntf most places, with exceptions in some of their etfs.
Allow me one final thought on this topic.
Diversification is good under most circumstances, over-diversification can ruin that goodness, but on the other hand, concentration might offer benefits depending on the proclivities of an investor.
Some investors accumulate mutual funds like a housewife accumulates recipes. That’s especially true if the investor favors active fund management. That might not be a good idea. And that’s because of the 80/20 rule which loosely applies to active fund managers.
The accumulated data strongly supports the observation that only about 20% of fund managers deliver returns in excess of appropriate benchmarks. The remaining 80% fail that meaningful target. The managers filling the elite 20% pool constantly change and are difficult to identify in advance of their superior performance.
Monte Carlo methods are ideal tools to use in modeling the inherent uncertainties that are characteristic of this issue . This has been done several times. One earlier effort was completed by financial advisor Allan Roth. More recently Richard Ferri added his version of a Monte Carlo study.
Allan Roth studied active fund returns and measured their success against relevant benchmarks. He explored two dimensions of the problem: holding time and number of funds in a portfolio. His Monte Carlo simulation examined timeframes from 1 to 25 years and number of portfolio funds from 1 to 5 to 10.
Roth concluded that those funds that outdistanced their benchmarks eroded as a function of time and as a function of the number of active funds in the portfolio. After 25 years, only 1% of a 10 active fund portfolio outperformed its benchmark equivalent. At shorter holding periods of 10 years and 5 years, the odds of outperforming benchmarks improved to 6% and 9%, respectively.
These are dire results indeed. The odds definitely tilt against active fund management. As an investor, do you believe you can be in the prescient 1% cohort? That’s not me.
Here is a Link to a summary of the Ferri study:
http://www.rickferri.com/WhitePaper.pdf
The key summary findings of the Ferri work are that (1) “Index funds have a higher probability of outperforming actively managed funds”. (2) “The probability of index fund portfolio outperformance increased when the time period was extended from 5 years to 15 years”, and (3) “The probability of index fund portfolio outperformance increased when two or more actively managed funds were held in each asset class.”
These conclusions reinforce the Roth results. If you want to integrate active fund management into your portfolios, these Monte Carlo simulations make a strong case to severely limit the number of those active fund managers. In this instance, the less, the better is good policy. So diversification in actively managed funds hurt end wealth.
Thank you for your patience. I have referenced these Monte Carlo studies in earlier postings. They warrant repeating.
MFOer Expatsp asked if I only included stocks in my analysis because I believed bonds had no place in a portfolio? No, not so! I merely limited my brief study to equities to simplify the analysis. My portfolio includes major fixed income components.
Best Wishes
1. Transaction Fee:
A transaction fee is similar to a brokerage fee or commission which you pay when you buy or sell a stock. For some funds available through Fidelity you are required to pay a transaction fee. However, you will not pay a sales load on Transaction Fee (TF) funds. You will only be charged a transaction fee when you buy a FundsNetwork TF fund, not when you sell one. All other fees and expenses described in a fund's prospectus still apply. You can choose to buy or sell shares directly from the fund itself or its principal underwriter or distributor without paying a transaction fee to Fidelity.
Online Transaction Fees: $49.95 for most funds. Certain funds will have a transaction fee of $75. To identify any applicable transaction fees associated with the purchase of a given fund, please refer to the "Fees and Distributions" tab.
Fidelity Automated Service Telephone (FAST): 25% off representative-assisted rates, Maximum: $187.50, Minimum:$75
Representative-Assisted: 0.75% of principal, Maximum: $250, Minimum: $100
Automatic Investment: $5 per transaction, after the initial investment.
Slick is correct in this case, the fee will be $75. I checked because all of my previous buys of non-NTF funds have been $49.95.
What you could do Joe is open the account at Vanguard and then have the VWINX fund transferred to Fidelity. Additional "automatic" purchases would then only cost you $5 per purchase. If you intend to add to your fund with random purchases those would cost you $75 each however. Something about no free lunch.
OJ
In my last post I mentioned the yeoman work that Allan Roth performed about a decade ago. I failed to provide a specific reference to his work or his interpretation of it. I just located a very brief commentary that Roth made to the AARP community last year. He is still singing the same song, and solidly defends his earlier pathfinder study. Here is a Link to the article:
http://blog.aarp.org/2015/11/09/play-the-odds-when-investing/
When making any financial decision, it’s always a good policy to “Play the odds when investing”. The Roth study definitely made a few assumptions and simplifications that not everyone would agree too. But that’s true for any modeling study. His work does properly capture the significant tradeoffs.
When comparing various Monte Carlo simulations, specific outcome numbers change depending on modeling details and timeframe dates, but the fundamental outcome trends and conclusions are consistent.
I recognize that some of you guys distrust Monte Carlo analysis feeling that it is too sophisticated mathematically. In reality it is none of that. It is conceptually simple. However, it can only be executed for the necessary thousands of random portfolio choices because of the speed of modern computers.
Many MFOers did spreadsheet portfolio analyses when making retirement decisions and didn’t fully realize they were doing a very limited (actually a too limited) Monte Carlo type analysis using estimates for yearly portfolio returns. Monte Carlo codes yield a more reliable base for that crucial retirement decision simply because it completes that same analysis thousands of times.
Best Wishes.