Diversification Sweet Spot

[Rich_by_the_Bay]

I'm unsure about how to find the "sweet spot" where diversification is broad enough to reflect the major noncorrelated sectors, yet simple enough to manage with ease -- the point of diminishing returns, so to speak. So, I read lots of articles on the web, most of which deal with how many individual stocks (20 or so give you almost all the diversification you need).

I am more of a mutual fund index investor. Just not sure where slicing and dicing becomes a waste of effort or worse.

To explain what I mean, here are two portfolios (let's forget bonds for now). The percentages are just as an example, could easily be modified.

The KISS Portfolio
---------------------------
Total Stock Market Index 50%
Total International Stock Market Index 25%
REIT Fund 25%

The RONCO Portfolio (for the Ron Popiel fans out there)
-----------------------------
Large Growth 10%
Large Value 10%
Small Growth 10%
Small value 10%
Midcap 10%
REITs 20%
Commod 5%
Total International 20%
Emerging International 5%

I'm interested in whether the total returns over time would really be demonstrably better in one or the other give the same long horizon, annual rebalance, "let the dogs run" approach. I have moderate volatility tolerance for this bucket part of my portfolio.

So, how basic would you let yourself go and still feel comfortable that you were well-diversified and could rebalance effectively? Would anyone feel that the KISS apporach is perfectly acceptable? Other suggested packages?

 

[LOL!]

I think the "Sweet Spot" is rather large and the two portfolios you described will probably both do about the same.

So if you are a set it and forget it, you pick the KISS portfolio.

If you like to post messages on message boards and wonder about sweet spots and anxiously look forward to your result in 10 years, you pick the RONCO portfolio.

If you still aren't sure, then you do 50:50 of each ... or some other combination of each (60:40? 80:20?)

For many of us with 401(k)s and 403(b)s we have to do something different because our plans do not have reasonable funds to match what you have listed.

 

[yakers]

Good question!

I see folks on the diehards and other boards with 10 asset classes, serious slice & dice and the numbers seem to say it pays for the time & effort. I have a strong preference for keeping things simple so the core of my holding is in a target retirement type fund. This is well diversified, its adjusts the AA as I get older and it keeps me from wanting to market time. And it does well over time, especially adjusted for risk. Its also a fund I have available in my 401k type account.

But those slice & dice folks do have some good numbers. If you have the discipline to rebalance and not to market time, you have enough resources and enough time to manage the accounts then it is probably going for the AA with more asset classes.

 

[Red-y]

Chuck Jaffe did a column on this that appeared in the Seattle Times yesterday. He says that when asked how many funds a typical investor "needs", financial planners say between 6 and 15.

I was curious and counted my funds...I've got exactly 15 split into these categories:

Emerging Markets 10%
Foreign 25%
US Large Cap 5%
Real Estate 7%
Energy/Commodities 10%
Gold 3%
Bonds and Cash 40%

I'm expecting this mix to grow at about 8% a year. I'm retired with no SS yet (I'm 49) and this portfolio is doing pretty well at the moment. For the most part, I'm concentrated in 3 fund families (Vanguard, Fidelity, Third Avenue) plus an ETF for the Gold allocation.

This is simple enough for me to manage but the variety of asset classes seems to be doing its job in damping down volatility of the overall portfolio, over the last two years, anyway. Time will tell...

 

[wabmester]

So, how basic would you let yourself go and still feel comfortable that you were well-diversified and could rebalance effectively?

If you simply want diversification, the total market approach is fine. It's not only simple, but it also mirrors the way the market participants allocate their capital. I'm not sure what "effective" rebalancing means, but a side benefit of the total market approach is that you don't need to rebalance among various slices.

You can get just as much diversification with slice and dice, but the logic behind slice and dice is to overweight certain "factors" and underweight others. Typically, the factors you want to overweight are "small" and "value."

Since market-cap weighted "total market" funds are predominantly large cap stocks, your proposed slice+dice portfolio should perform better over the long term if you believe in the 3-factor model.

 

[audreyh1]

The loner time goes on, the simpler I make it.

I don't feel compelled to own every asset class. Commodities doesn't have the best historical record so I decided to ignore them. Emerging markets can be terribly volatile. So are international small caps.

This is what I have ended up being comfortable with:

US Large, Medium, and Small caps - with tilt towards value
International Large, International medium cap - also tilt towards value.
REITs
Bonds - short to intermediate term, higher quality tilt

So I guess that's somewhere in between KISS and RONCO

I don't hold nearly that large an amount in REITs! More like 5% of total portfolio.

Audrey

 

[Dawg52]

I copied this from the FundAdvice website. Simple with plenty of diversification. I still like the Vanguard Target funds though.

ETF BALANCED BUY-AND-HOLD
7.5% S&P Depositary Receipts (SPDR) S&P 500 (SPY)
7.5% Vanguard Value VIPERs (VTV)
7.5% I-Shares Russell Microcap Index (IWC)
7.5% Vanguard Small Cap Value VIPERs (VBR)
6% I-Shares MSCI EAFE (EFA)
6% I-Shares MSCI EAFE Value Index (EFV)
12% WisdomTree Int'l Small Cap Div Fund (DLS)
6% Vanguard Emerging Market VIPERs (VWO)
20% I-Shares Lehman 1-3 yr (SHY)
20% I-Shares Lehman Aggregate Bond (AGG)

 

[saluki9]

To explain what I mean, here are two portfolios (let's forget bonds for now). The percentages are just as an example, could easily be modified.

The KISS Portfolio
---------------------------
Total Stock Market Index 50%
Total International Stock Market Index 25%
REIT Fund 25%

The RONCO Portfolio (for the Ron Popiel fans out there)
-----------------------------
Large Growth 10%
Large Value 10%
Small Growth 10%
Small value 10%
Midcap 10%
REITs 20%
Commod 5%
Total International 20%
Emerging International 5%

I'm interested in whether the total returns over time would really be demonstrably better in one or the other give the same long horizon, annual rebalance, "let the dogs run" approach. I have moderate volatility tolerance for this bucket part of my portfolio.

So, how basic would you let yourself go and still feel comfortable that you were well-diversified and could rebalance effectively? Would anyone feel that the KISS apporach is perfectly acceptable? Other suggested packages?

Well, to make things interesting, let's put some numbers to these portfolios


KISS Portfolio 3yr 5yr 10yr


Beta 1.05 .83 .77
Sharpe 1.62 .97 .53
Mean Ret% 18.84 13.65 10.12


RONCO 1.14 .82 .74
1.57 1.09 .55
19.72 15.54 10.61

 

[Rich_by_the_Bay]

Well, to make things interesting, let's put some numbers to these portfolios...

Yep, I suppose that'd be a good thing to do .

What's Sharpe?
How would rebalancing annually affect this?
What are you concluding from the numbers?

 

[ESRBob]

I was curious and counted my funds...I've got exactly 15 split into these categories:

Emerging Markets 10%
Foreign 25%
US Large Cap 5%
Real Estate 7%
Energy/Commodities 10%
Gold 3%
Bonds and Cash 40%

I'm expecting this mix to grow at about 8% a year.

Red-y,
I happen to like this one, too -- it is 40% stock, 40% bond/Fixed Income and 20% "Other". I worked with an advisor to run all sorts of scenarios and came up with this being a good balance between income, return and volatility, with a return hight enough to cover inflation, fees and a SWR around 4%, with a little left over to grow.

Rich, the question you originally pose is the sort of thing an advisor will run scenarios with you to figure out. I know we all generally shy away from advisors here, but a few hours with a fee-only person (who has all the fancy databases and software) could shed some light on this stuff for you. Phone interview them first to make sure they aren't going to try to sell you loaded funds, though! DFA's site (www.dfaus.com) has a list of advisors who are vetted to be those who 'get' long term buy-and-hold index-style investing if that looks like a fit for you. Should be able to find someone in the (Tampa) Bay Area.

 

[yakers]

Besides the diversification and the 10 way slice & dice, it looks like simple balanced/target retirement/lifecycle funds do pretty well:

http://tinyurl.com/ybquyb

Now, to keep the male hormones working you can add asset classes, but it looks like simple is not only easy, it really is good.

 

[Rich_by_the_Bay]

I happen to like this one, too -- it is 40% stock, 40% bond/Fixed Income and 20% "Other". I worked with an advisor to run all sorts of scenarios and came up with this being a good balance between income, return and volatility, with a return hight enough to cover inflation, fees and a SWR around 4%, with a little left over to grow.

Interesting. No US small growth, midcap, or small value. And 13% effectively on commodities? Guess there's lots of ways to skin a cat.

 

[ats5g]

Yep, I suppose that'd be a good thing to do .

What's Sharpe?
How would rebalancing annually affect this?
What are you concluding from the numbers?

Rich,

The definition of the Sharpe Ratio is:

S(x) = ( rx - Rf ) / StdDev(x)

where

x is some investment
rx is the average annual rate of return of x
Rf is the best available rate of return of a "risk-free" security (i.e. cash)
StdDev(x) is the standard deviation of rx

So, even though the RONCO portfolio had a higher mean return %, both your KISS and RONCO portfolios had about the same reward-to-risk ratio. Hence, neither was more efficient [i.e. more return with less risk] than the other.

- Alec

 

[Rich_by_the_Bay]

Besides the diversification and the 10 way slice & dice, it looks like simple balanced/target retirement/lifecycle funds do pretty well:

http://tinyurl.com/ybquyb

Now, to keep the male hormones working you can add asset classes, but it looks like simple is not only easy, it really is good.

Wish there was a fund that would do this for stocks only, and let me carve out my bonds to handle separately.

Maybe DFA has one, though the fees have kept me away. Current Alliance-Bernstein is handling much of my value and REIT investments; they are doing well but fees take them down to average

The strategy and lifestyle funds have a lot of appeal to me, too, but they bundle the bond piece with the fund.

 

[Rich_by_the_Bay]

So, even though the RONCO portfolio had a higher mean return %, both your KISS and RONCO portfolios had about the same reward-to-risk ratio. Hence, neither was more efficient [i.e. more return with less risk] than the other.

Thanks, Alec. I think I get it.

Since I don't mind some risk in this scenario, I might tilt toward the RONCO approach. I've yet to figure out how rebalancing affects these comparisons because it could theoretically tilt in favor of RONCO, too, there being more opportunity to play Robin Hood between sectors.

 

[saluki9]

Rich,

DFA does have them. They are called the core equity funds. They have 2 domestic, an international, and an EM fund.

 

[wildcat]

I think the Kiss is short, sweet and effective. A potentially nice option once you retire and only have to worry about rebalancing a few funds. Diversification doesn't necessarily have to include exotic asset classes. The only change I would make would be to capture the small cap value premium and add it to the portfolio mix. I think it is worth having around.

 

[audreyh1]

Ditto Wildcat!

In the interest of simplicity (simplicity in management is good, less things to rebalance means less hassle) you really can skip the "exotic" asset classes.

But don't miss out on how small caps and a value tilt can really help your portfolio in the long run.

Audrey

 

[wildcat]

One more thing Rich and I am sure you are well aware of this but make sure you include taxes in the plan. For example on the KISS, REIT & style index funds (like small cap value wink) - tax def & total mkt & total intl' index - tax.

 

[Rich_by_the_Bay]

One more thing Rich and I am sure you are well aware of this but make sure you include taxes in the plan. For example on the KISS, REIT & style index funds (like small cap value wink) - tax def & total mkt & total intl' index - tax.

Yep. It's virtually all sheltered at this point for better or worse.

Funny thing happens during discussions of asset allocation: the simple plans seem to develop "diversification creep." Everyone likes simplicity but before long each of us wants to add just one more sector to round it out. Next thing you know, there are 7-8 funds and it begins to feel like slice-and-dice.

 

[Cut-Throat]

Yep. It's virtually all sheltered at this point for better or worse.

Funny thing happens during discussions of asset allocation: the simple plans seem to develop "diversification creep." Everyone likes simplicity but before long each of us wants to add just one more sector to round it out. Next thing you know, there are 7-8 funds and it begins to feel like slice-and-dice.

Berstein points out in four pillars, that no matter what you decide your portfoilo should be, it is important to stick with it. If you keep tweaking it, you will lose out in the benefits of holding and rebalancing. (i.e. - human nature adding sectors when they are hot, or getting rid iof them when they are cold)

He points this out many, many times when discussing portfoilo allocation - "It's not as important what your portfoilo mix is over the long term, but whether you stick to it"

 

[audreyh1]

Funny thing happens during discussions of asset allocation: the simple plans seem to develop "diversification creep." Everyone likes simplicity but before long each of us wants to add just one more sector to round it out. Next thing you know, there are 7-8 funds and it begins to feel like slice-and-dice.

Yep! I really notice that. Every 3 years there seems to be a new "must have" asset class and I see people shifting things around to make room for a wee bit of this or that. But after a while - 5% here, 3% there, you start to wonder how it can be worth it. The more slices, the more work.

Maybe people get caught tweaking because they think somehow they can "optimize" their portfolio. I don't think the point should be to optimize (I don't think you really can), I think the point is to figure out the simplest thing that is "good enough", stick to that, and go on with the rest of your life.

Audrey

 

[wildcat]

Everyone likes simplicity but before long each of us wants to add just one more sector to round it out

That's because most of us watch too much CNBC.

 

[Dawg52]

Funny thing happens during discussions of asset allocation: the simple plans seem to develop "diversification creep." Everyone likes simplicity but before long each of us wants to add just one more sector to round it out. Next thing you know, there are 7-8 funds and it begins to feel like slice-and-dice.

Yeah, that is why I'm liking the Vandguard Target funds more and more. It gives you a slice and dice approach with just one fund, but automatically rebalances to a more conservative style as you grow older. The only problem I have is 75% of my portfolio is in my taxable account. In order to convert over, I'm gonna have to sell what I have and pay capital gain taxes.

 

[yakers]

Berstein points out in four pillars, that no matter what you decide your portfoilo should be, it is important to stick with it. If you keep tweaking it, you will lose out in the benefits of holding and rebalancing. (i.e. - human nature adding sectors when they are hot, or getting rid iof them when they are cold)

He points this out many, many times when discussing portfoilo allocation - "It's not as important what your portfoilo mix is over the long term, but whether you stick to it"

Maintaining Vs Tweaking the AA

Seriously good issue. I am frequently tempted to tweak the AA as new asset classes or funds become available. My basic fund in a tax deferred target retirement fund of funds made up of S&P500, small cap, total bond, and total intl indexes and a stable value fund. That covers everything. But hay! what about REITs? Then I discover emerging markets. Then unhedged foreign bonds. What about Canadian energy trusts? Oh, yeah, precious metals, will that be hold gold itself or producer stock? Oil, energy, commodities?

How do I measure the cost impact of all this tweaking when it is done to creat a new improved AA. Is there really a cost for these changes or does it result in improved returns?