Splitting Growth And Value Leads To A Worse Return

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Topic Author
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Splitting Growth And Value Leads To A Worse Return

Post by Rick Ferri »

Does splitting value and growth into equal components help a portfolio? I haven't been able to reconcile this strategy even thought it is quite popular with investment advisers. My hunch is that the strategy provides advisers with a marketing benefit even though there is no client benefit.

Splitting Growth And Value Leads To A Worse Return

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
staythecourse
Posts: 6993
Joined: Mon Jan 03, 2011 8:40 am

Re: Splitting Growth And Value Leads To A Worse Return

Post by staythecourse »

Mr. Ferri,

Good article. The older I get the more I question EVERYTHING that is paraded as dogma as many have no substance when one "looks under the hood".

BTW, where is your excellent graph of risk/ return of Russell 1000 blended and what happens when you add more growth vs. value. you have in "AAAA"?? The one graph speaks a thousand words!!

Good luck.
"The stock market [fluctuation], therefore, is noise. A giant distraction from the business of investing.” | -Jack Bogle
User avatar
Topic Author
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by Rick Ferri »

staythecourse wrote:BTW, where is your excellent graph of risk/ return of Russell 1000 blended and what happens when you add more growth vs. value. you have in "AAAA"?? The one graph speaks a thousand words!!
I was thinking of doing that graph, but thought it best to just stick with just a 50%/50% allocation for this article to introduce the concept.

Thanks,
Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
BornInCA
Posts: 298
Joined: Tue Oct 26, 2010 8:40 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by BornInCA »

What if a US equity portfolio was equally split this way?
25% US Stock market index
25% US Large Cap Value
25% US Small Cap Blend
25% US Small Cap Value

It's a 4-way split among the "funds" but it's a tilt towards "small" and "value", is it not?
rkhusky
Posts: 17764
Joined: Thu Aug 18, 2011 8:09 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by rkhusky »

One different between the two is that value stocks pay more dividends than growth stocks. Would there be a benefit to holding value stocks in tax advantaged and growth in taxable?
Last edited by rkhusky on Mon Jul 15, 2013 12:53 pm, edited 1 time in total.
User avatar
Garco
Posts: 1078
Joined: Wed Jan 23, 2013 1:04 am
Location: U.S.A.

Re: Splitting Growth And Value Leads To A Worse Return

Post by Garco »

Yeah, if you want to own the Russell 2000, and you start splitting and overweighting, say, value, you can get dubious results. Although I split the TSM into two pieces (VFINX and VEXMX), seeking a SC dividend by overweighting VEXMX, I am happy with the performance of the VEXMX without yet another division into value and growth and the complication and cost of rebalancing between value and growth. This has worked well for me.
User avatar
Topic Author
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by Rick Ferri »

BornInCA wrote:What if a US equity portfolio was equally split this way?
25% US Stock market index
25% US Large Cap Value
25% US Small Cap Blend
25% US Small Cap Value

It's a 4-way split among the "funds" but it's a tilt towards "small" and "value", is it not?
This is a different question because you're over-weighting small and value in the above portfolio as opposed to splitting small and large by cap weights, and value and growth equally.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
User avatar
mlewis
Posts: 211
Joined: Sun Nov 04, 2012 11:39 pm
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by mlewis »

So it seems from this, and just from your portfolio in general (http://www.bogleheads.org/forum/viewtopic.php?t=58709) that you are a fan of getting some SCV but not bothering with any value tilt in the LC arena.

Are you suggesting there isn't really a value premium on its own, or that it isn't worth trying to capture?

malcolm
User avatar
Topic Author
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by Rick Ferri »

Value is value. Whether you invest in small value, large value or total market value market is irrelevant. What maters is the overall value tilt in entire portfolio.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
User avatar
mlewis
Posts: 211
Joined: Sun Nov 04, 2012 11:39 pm
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by mlewis »

Rick Ferri wrote:Value is value. Whether you invest in small value, large value or total market value market is irrelevant. What maters is the overall value tilt in entire portfolio.

Rick Ferri
I have a hard time seeing this. Do you have more information on this idea?

You are essentially saying these two portfolios have the same primary characteristics?
TSM+LCV+SC = TSM+LC+SCV

LCV and SCV might both be value, but I just think of them as separate asset classes.

malcolm
User avatar
Topic Author
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by Rick Ferri »

The idea is a little confusing at first. The amount of value stock exposure that a portfolio holds determines is expected long-term return based on the Fama-French three factor model. It doesn't matter whether these are small value stocks or large value stocks, the it is the amount of the value exposure in aggregate that matters. The difference between large and small stock returns is a different risk factor called "size" risk. See Understanding Risk: The Fama-French Three Factor Model.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
Wagnerjb
Posts: 7213
Joined: Mon Feb 19, 2007 7:44 pm
Location: Houston, Texas

Re: Splitting Growth And Value Leads To A Worse Return

Post by Wagnerjb »

Rick: I agree with your paper that splitting value and growth - just for the sake of slicing and dicing - doesn't add value. But you should note that holding value in tax-deferred and holding growth in taxable is a value adding proposition, assuming your balance of tax-deferred and taxable requires you to hold stocks in both locations.

Best wishes.
Andy
mullery
Posts: 63
Joined: Wed Nov 28, 2012 12:39 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by mullery »

Here's the size/valuation on 100% IWB:

25 26 27
7 7 7
1 0 0

And 50% IWF, 50% IWD:

25 26 27
7 7 7
1 0 0

Exactly the same. It makes sense that the reduction in annualized return is about the same as the increase ER of the split funds.
Tamahome
Posts: 2325
Joined: Wed Feb 27, 2013 12:03 pm
Location: Atlanta, GA

Re: Splitting Growth And Value Leads To A Worse Return

Post by Tamahome »

Thank you for the article! From what I gather, I am doing things wrong in my retirement account. I have two emerging markets funds (DFA funds for small cap and value in emerging markets). I could simply tilt in a US or general international fund rather than splitting up each section into micro splice and dice. Am I understanding that correctly?
I'm not a financial professional. Post is info only & not legal advice. No attorney-client relationship exists with reader. Scrutinize my ideas as if you spoke with a guy at a bar. I may be wrong.
User avatar
Topic Author
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by Rick Ferri »

In theory that is true. I don't think there is anything wrong with what you're doing. Just be aware that there is a cost to cutting a pie into many slices. The more cuts you make, the more pie sticks to the pan.

What matters most is the amount your portfolio is tilted to value and how much you're paying for this risk exposure. In other words, how much are you paying per unit of value? Some fund companies charge a lot more than others and give you less value stock exposure. DFA isn't one of these companies. Their funds tend to provide a deep tilt to value at a reasonable cost.

My method is to buy a total US stock fund and add an element of small-cap value index fund to it. I do the same in developed international using a DFA international small cap value fund. In emerging markets, I use one fund, DFA Emerging Markets Core, which takes care of the small and value tilt within one fund.

Rick Ferri

PS. I wish to reiterate that titling to value stocks is not the same as splitting growth and value down the middle. Total market index funds are NOT growth funds. They are style neutral.
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
User avatar
Garco
Posts: 1078
Joined: Wed Jan 23, 2013 1:04 am
Location: U.S.A.

Re: Splitting Growth And Value Leads To A Worse Return

Post by Garco »

Rick, I think I understand (and believe in) the theory. As a practical matter, however, many of us have a limited range of choices in our 401k's. I couldn't invest in a SCV or MCV if I wanted to, for example, unless I were allowed to go through a brokerage window in my 401k. And that's the main reason why I invest in VEXMX (VG "extended market") -- to at least get a "small" (non-large) cap tilt, even if it's neutral on a value-growth axis. For that matter, I don't have access to a true LCV either, but I've got access to one that at least tilts in that direction (DODGX). So I was happy to read above that you confirmed that the value tilt can be in any of the size classes. That had been my understanding of it from "the literature," but quite a few posters on this forum read the term "tilt toward small cap and value" to imply that you need to find specific funds in that combined category. So thank you.
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by pascalwager »

In The Intelligent Asset Allocator, Bill Bernstein included the "Gap Portfolio" (offered by DFA) which splits large-cap, small-cap, int'l small-cap, and EM large-cap between growth and value. It intuitively seemed inefficient and also I had seen this method specifically discounted in a book by Rick Ferri. Anyway, I was surprised and still wonder if the proofreader didn't make some modifications.
VT 60% / VFSUX 20% / TIPS 20%
pascalwager
Posts: 2327
Joined: Mon Oct 31, 2011 8:36 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by pascalwager »

The idea is a little confusing at first. The amount of value stock exposure that a portfolio holds determines is expected long-term return based on the Fama-French three factor model. It doesn't matter whether these are small value stocks or large value stocks, the it is the amount of the value exposure in aggregate that matters. The difference between large and small stock returns is a different risk factor called "size" risk. See Understanding Risk: The Fama-French Three Factor Model.

Rick Ferri
Then, if there is no longer a small premium (as many believe), wouldn't the premium for large value and small value be identical?
VT 60% / VFSUX 20% / TIPS 20%
Mudpuppy
Posts: 7409
Joined: Sat Aug 27, 2011 2:26 am
Location: Sunny California

Re: Splitting Growth And Value Leads To A Worse Return

Post by Mudpuppy »

Garco wrote:Rick, I think I understand (and believe in) the theory. As a practical matter, however, many of us have a limited range of choices in our 401k's.
Indeed, dealing with the choices in one's plan can be a major roadblock in optimizing retirement savings. With the CA Savings Plus Program (401k and 457b plans for certain CA state employees), one doesn't even have the luxury of choosing between value and growth. Instead, one has the choice of an index fund or a blend fund for each size. For example, there is no large cap value fund or large cap growth fund, just the large cap fund and the large cap index fund. The large cap fund is 30% value, 30% growth and 40% index. Small cap fund and mid cap fund are also split between value, growth, and index, although the exact composition is harder to find since the state switched website providers (the new website no longer provides the names of the underlying funds, just the companies providing the funds).

There is of course the self-directed brokerage account option, but they've not made it easy to take advantage of that option. I can't even find the enrollment form on the new website. The instructions on the page introducing the brokerage account are incorrect, although if one logs out and finds the brochure for the option under the public website, it does appear to have the correct instructions (judging purely by how counter-intuitive the instructions are). Additionally, one cannot find out the investment options available in the brokerage account without creating a brokerage account, although this is a rather common practice.
User avatar
Topic Author
Rick Ferri
Posts: 9707
Joined: Mon Feb 26, 2007 10:40 am
Location: Georgetown, TX. Twitter: @Rick_Ferri
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by Rick Ferri »

pascalwager wrote:
The idea is a little confusing at first. The amount of value stock exposure that a portfolio holds determines is expected long-term return based on the Fama-French three factor model. It doesn't matter whether these are small value stocks or large value stocks, the it is the amount of the value exposure in aggregate that matters. The difference between large and small stock returns is a different risk factor called "size" risk. See Understanding Risk: The Fama-French Three Factor Model.

Rick Ferri
Then, if there is no longer a small premium (as many believe), wouldn't the premium for large value and small value be identical?
The premium for "value" is the premium for value. There are not two different value premiums for large and small cap stocks.

Rick Ferri
The Education of an Index Investor: born in darkness, finds indexing enlightenment, overcomplicates everything, embraces simplicity.
User avatar
Phineas J. Whoopee
Posts: 9675
Joined: Sun Dec 18, 2011 5:18 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by Phineas J. Whoopee »

I don't see how splitting without tilting can help anything, and could easily believe it will hurt, not only from increased expense ratios, but also from increased turnover.

A mutual fund is only a vehicle by which to own underlying assets. If the underlying assets are the same, how can the return be different, except lower due to increased friction?

If there were a growth fund with only 10% of stocks in it, and a value fund with the same percentage, then that would be different from the total market. Somebody might decide to invest in that portfolio. But the somebody wouldn't be mebody.

PJW
Gecko10x
Posts: 390
Joined: Thu Jun 06, 2013 8:10 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by Gecko10x »

I'm going to resurrect this thread because it deals directly with my question: Does splitting value and growth give you a good way to hedge your bets?

Rick's posted article says "No", but this backtest seems to say "Yes". Does anyone have any additional thoughts about this? Why does this analysis contradict Rick's?
User avatar
David Jay
Posts: 14586
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Splitting Growth And Value Leads To A Worse Return

Post by David Jay »

Sorry Gecko, but your backtest is not looking at the same thing as the original thread. The original thread was comparing Total Market with 50% Value/50% Growth.

Your 3-way backtest is comparing 100% Value with 100% Growth with 50% Value/50% Growth.
It's not an engineering problem - Hersh Shefrin | To get the "risk premium", you really do have to take the risk - nisiprius
Gecko10x
Posts: 390
Joined: Thu Jun 06, 2013 8:10 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by Gecko10x »

David, the benchmark on my comparison is US Large Cap; I was only including 100% Value and 100% Growth for reference.

Yes, the original article and thread was using the total market, vs my comparison is only with large cap, but it is the same question being asked.

From what I can see in PV (even running for a few different date ranges), the 50/50 growth/value (re-balanced) portfolio does better than the 100% portfolio, which contradicts Rick's results. I get the same thing using small caps also.
User avatar
David Jay
Posts: 14586
Joined: Mon Mar 30, 2015 5:54 am
Location: Michigan

Re: Splitting Growth And Value Leads To A Worse Return

Post by David Jay »

During the period of your back test Growth outperformed. So any portfolio that held more Growth performed better.

That is one of the limitations of back testing. Did you notice "Note 1" on PV?
It's not an engineering problem - Hersh Shefrin | To get the "risk premium", you really do have to take the risk - nisiprius
Gecko10x
Posts: 390
Joined: Thu Jun 06, 2013 8:10 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by Gecko10x »

My original thought was that 50% Value / 50% growth holds the SAME assets as 100% of the market/sector/whatever. That's where I was coming from with my question.

I suppose however, that if that isn't how value/growth is defined, then that would cause problems with the analysis. If value is defined as the bottom 1/3 instead of the bottom 1/2, then I'm not comparing apples to apples. Is this what you're asserting, or something else?

Edit: I've added a 33% blend to the analysis. BOTH the re-balanced blends outperformed the benchmark. Am I being dense? Does this not show that splitting value and growth led to better return?

Edit: Just noticed the original link is broken. The article is still accessible via Forbes.
Last edited by Gecko10x on Tue Apr 25, 2017 7:55 am, edited 1 time in total.
Gecko10x
Posts: 390
Joined: Thu Jun 06, 2013 8:10 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by Gecko10x »

David Jay wrote:Did you notice "Note 1" on PV?
Really don't think this is necessary; of course I'm well aware. The entire thread is about backtesting. :annoyed
avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by avalpert »

Gecko10x wrote:My original thought was that 50% Value / 50% growth holds the SAME assets as 100% of the market/sector/whatever. That's where I was coming from with my question.

I suppose however, that if that isn't how value/growth is defined, then that would cause problems with the analysis. If value is defined as the bottom 1/3 instead of the bottom 1/2, then I'm not comparing apples to apples. Is this what you're asserting, or something else?

Edit: I've added a 33% blend to the analysis. BOTH the re-balanced blends outperformed the benchmark. Am I being dense? Does this not show that splitting value and growth led to better return?

Edit: Just noticed the original link is broken. The article is still accessible via Forbes.
Portfolio Visualizer uses Vanguard's large cap value and growth funds for the asset class returns - the combined benchmark for that would be VLACX (Vanguard's large cap index) and not their S&P 500 fund - if you use that in the backtest I believe you will see the performance difference disappear.
Gecko10x
Posts: 390
Joined: Thu Jun 06, 2013 8:10 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by Gecko10x »

avalpert wrote:Portfolio Visualizer uses Vanguard's large cap value and growth funds for the asset class returns - the combined benchmark for that would be VLACX (Vanguard's large cap index) and not their S&P 500 fund - if you use that in the backtest I believe you will see the performance difference disappear.
Ah-ha! That certainly does make the difference disappear.

Thanks!
User avatar
nisiprius
Advisory Board
Posts: 52212
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: Splitting Growth And Value Leads To A Worse Return

Post by nisiprius »

avalpert wrote:Portfolio Visualizer uses Vanguard's large cap value and growth funds for the asset class returns - the combined benchmark for that would be VLACX (Vanguard's large cap index) and not their S&P 500 fund - if you use that in the backtest I believe you will see the performance difference disappear.
I do.
Image
If that's not "identical," it's pretty darn close.

Of course, VLACX cut the time period down to Jan 2005 - Mar 2017. We can extend it just a bit by using IWB for our "large-cap" benchmark; it tracks the Russell 1000 index and is thus, even, a slightly independent check on the definition of "large-cap."

Source
Image
With a straight face and my tongue firmly in my cheek, I will say "Well, IWB isn't quite as identical to the split portfolio as VLACX, but it's still very identical." :)
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
czeckers
Posts: 1082
Joined: Thu May 17, 2007 3:49 pm
Location: USA

Re: Splitting Growth And Value Leads To A Worse Return

Post by czeckers »

Do you include rebalancing?

I just ran a portfolio of 50/50 US large cap value/growth vs US large cap with annual rebalancing and the CAGR was 10.69 vs 10.17 for the time period 1972-Mar 2017. (portfoliovisualizer)

You would expect some rebalancing bonus if you rebalanced between two asset classes that have similar expected rates of return.

-K
The Espresso portfolio: | | 20% US TSM, 20% Small Value, 10% US REIT, 10% Dev Int'l, 10% EM, 10% Commodities, 20% Inter-term US Treas | | "A journey of a thousand miles begins with a single step."
User avatar
Ketawa
Posts: 2521
Joined: Mon Aug 22, 2011 1:11 am
Location: DC

Re: Splitting Growth And Value Leads To A Worse Return

Post by Ketawa »

czeckers wrote:Do you include rebalancing?

I just ran a portfolio of 50/50 US large cap value/growth vs US large cap with annual rebalancing and the CAGR was 10.69 vs 10.17 for the time period 1972-Mar 2017. (portfoliovisualizer)

You would expect some rebalancing bonus if you rebalanced between two asset classes that have similar expected rates of return.

-K
It makes no sense to "expect" a rebalancing bonus. Otherwise, why not chop the stock market into even smaller pieces if it's this easy to get a bonus? Why not hold separate domestic, EAFE or Europe/Pacific, and emerging markets funds? Sometimes rebalancing results in a higher CAGR by assigned fixed allocations, and sometimes it doesn't.
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by willthrill81 »

Ketawa wrote:
czeckers wrote:Do you include rebalancing?

I just ran a portfolio of 50/50 US large cap value/growth vs US large cap with annual rebalancing and the CAGR was 10.69 vs 10.17 for the time period 1972-Mar 2017. (portfoliovisualizer)

You would expect some rebalancing bonus if you rebalanced between two asset classes that have similar expected rates of return.

-K
It makes no sense to "expect" a rebalancing bonus. Otherwise, why not chop the stock market into even smaller pieces if it's this easy to get a bonus? Why not hold separate domestic, EAFE or Europe/Pacific, and emerging markets funds? Sometimes rebalancing results in a higher CAGR by assigned fixed allocations, and sometimes it doesn't.
You beat me to it!

There is a persistent 'rebalancing bonus' myth. It simply does not exist. People most often think that it applies to stock/bond portfolios, where rebalancing merely keeps the desired risk level in check but does not improve returns (in that context, it has typically reduced historical returns, often significantly).
The Sensible Steward
kolea
Posts: 1322
Joined: Fri Jul 11, 2014 5:30 pm
Location: Maui and Columbia River Gorge

Re: Splitting Growth And Value Leads To A Worse Return

Post by kolea »

willthrill81 wrote: There is a persistent 'rebalancing bonus' myth. It simply does not exist. People most often think that it applies to stock/bond portfolios, where rebalancing merely keeps the desired risk level in check but does not improve returns (in that context, it has typically reduced historical returns, often significantly).
Hmm. I am on the fence about this but here is what I thought was going on with the rebalancing benefit, which is a bit more than the risk-management story...

If you believe that there is a benefit to mixing asset classes that are uncorrelated to produce the so-called diversification "free lunch", and that there is an optimum mix of these asset classes that is defined by the efficient frontier, then you have to have some means to maintain that optimum ratio of asset classes. That means is rebalancing. The benefit to rebalancing is thus the optimization of expected return by maintaining the correct balance of asset classes. In this sense rebalancing between correlated asset classes (which is what you will get if just divide TSM into a bunch of sectors) is not going to have much effect. Because expected return is not the same as actual return, the result you get by looking backwards may well appear to not be a benefit.

That is what I believe is meant by the rebalancing bonus.
Kolea (pron. ko-lay-uh). Golden plover.
avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: Splitting Growth And Value Leads To A Worse Return

Post by avalpert »

czeckers wrote:Do you include rebalancing?

I just ran a portfolio of 50/50 US large cap value/growth vs US large cap with annual rebalancing and the CAGR was 10.69 vs 10.17 for the time period 1972-Mar 2017. (portfoliovisualizer)

You would expect some rebalancing bonus if you rebalanced between two asset classes that have similar expected rates of return.

-K
See viewtopic.php?f=10&t=119837#p3342523
User avatar
willthrill81
Posts: 32250
Joined: Thu Jan 26, 2017 2:17 pm
Location: USA
Contact:

Re: Splitting Growth And Value Leads To A Worse Return

Post by willthrill81 »

kolea wrote:
willthrill81 wrote: There is a persistent 'rebalancing bonus' myth. It simply does not exist. People most often think that it applies to stock/bond portfolios, where rebalancing merely keeps the desired risk level in check but does not improve returns (in that context, it has typically reduced historical returns, often significantly).
Hmm. I am on the fence about this but here is what I thought was going on with the rebalancing benefit, which is a bit more than the risk-management story...

If you believe that there is a benefit to mixing asset classes that are uncorrelated to produce the so-called diversification "free lunch", and that there is an optimum mix of these asset classes that is defined by the efficient frontier, then you have to have some means to maintain that optimum ratio of asset classes. That means is rebalancing. The benefit to rebalancing is thus the optimization of expected return by maintaining the correct balance of asset classes. In this sense rebalancing between correlated asset classes (which is what you will get if just divide TSM into a bunch of sectors) is not going to have much effect. Because expected return is not the same as actual return, the result you get by looking backwards may well appear to not be a benefit.

That is what I believe is meant by the rebalancing bonus.
I understand your point, but the last nearly 50 years of data that we have easy access to does not support rebalancing as a means of improving returns. It hurts as often as it helps.

If we're talking about risk-adjusted returns, that's a different story. Rebalancing tends to lower volatility and improve the Sharpe ratio.
The Sensible Steward
User avatar
czeckers
Posts: 1082
Joined: Thu May 17, 2007 3:49 pm
Location: USA

Re: Splitting Growth And Value Leads To A Worse Return

Post by czeckers »

The benefits of rebalancing (or lack thereof) depend on the expected returns of the asset classes you are rebalancing between.

Rebalancing between stocks and bonds which have very different expected returns is generally a losing proposition. For the most part, you will be selling shares of the higher performing asset class (stocks) and buying shares of the lower performing one (bonds). In this case, the point of rebalancing is to control risk. An unrebalanced stock/bond allocation will drift over time toward higher percentage of stocks thereby increasing portfolio risk.

However, if you rebalance between asset classes of similar expected returns, then you in fact can expect a bit of a rebalancing bonus as it will systematically cause you to buy low and sell high. Separating large indexes into smaller but equal components does in fact increase returns. Keeping large growth/value or a large cap/extended index, or splitting developed international and emerging markets vs total international can help.

The issue is transaction costs. In many cases, the larger index can be had for a smaller ER than the smaller components. Since the benefit of this strategy is only a small incremental increase in returns, it only makes sense if you can get the subgroup portfolio at a similar cost to the overall index.

Dr. Bernstein did a nice analysis a while back. You can read it here: http://www.efficientfrontier.com/ef/996/rebal.htm

You can test it yourself using various matched pairs with any of the available back testing tools such as Portfolio Visualizer or Simba's back test spreadsheet.
The Espresso portfolio: | | 20% US TSM, 20% Small Value, 10% US REIT, 10% Dev Int'l, 10% EM, 10% Commodities, 20% Inter-term US Treas | | "A journey of a thousand miles begins with a single step."
Post Reply