How much tilting to SCV (rule of thumb)?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
international001
Posts: 1498
Joined: Thu Feb 15, 2018 7:31 pm

How much tilting to SCV (rule of thumb)?

Post by international001 » Mon Jun 22, 2020 1:59 pm

If you believe the SCV (or just S or just V) premium still exists, then it would make sense to tilt X% to SCV?
But what is the rule of thumb for X? Probably, the more time you have left for retirement, the more you should tilt to it.

Paul Merriman suggests 1.5 x years-to-retirement

https://paulmerriman.com/two-funds-for- ... etirement/

But even for a 15 year period it would see wise to tilt heavier than that (his 4 fund portfolio 25% LCB, 25% LCV, 25% SCB, 25% SCV outperforms always LCB https://paulmerriman.com/wp-content/upl ... 8-2019.pdf). So I don't know how he gets his rule of thumb.

Any standard guidance?

KlangFool
Posts: 16639
Joined: Sat Oct 11, 2008 12:35 pm

Re: How much tilting to SCV (rule of thumb)?

Post by KlangFool » Mon Jun 22, 2020 2:06 pm

OP,

What is your goal?

A) A tilt to SCV?

B) As part of the Larry portfolio?

(A) or (B)?

KlangFool

CrossOverGuy
Posts: 311
Joined: Fri Apr 27, 2007 2:19 pm

Re: How much tilting to SCV (rule of thumb)?

Post by CrossOverGuy » Mon Jun 22, 2020 2:06 pm

I just went to the Vanguard website and checked out comparison of Small Cap Index (including both growth and value), Small-Cap Value Index and Small-Cap Growth Index. Small-Cap Growth Index and Small-Cap Index both did much better than Small-Cap Value over YTD, 1-year, 3-year, 5-year and 10 years basically. Check that against whoever has been pushing the Small-Cap Value tilt thesis. The numbers there don't really support it. It seems like some years Value does better, others Growth does better and that the Small-Cap Index captures the average (still quite good) of both. In index funds, generally, average beats big swings trying the beat the market and just captures the market.
Last edited by CrossOverGuy on Mon Jun 22, 2020 2:12 pm, edited 1 time in total.

MotoTrojan
Posts: 9952
Joined: Wed Feb 01, 2017 8:39 pm

Re: How much tilting to SCV (rule of thumb)?

Post by MotoTrojan » Mon Jun 22, 2020 2:09 pm

There really is no right answer. If it were certain to outperform, then the right answer is 100%. So you just have to do research and determine how much trust you want to put in the past, and more importantly, how well you think you can handle a 10-15 year period of underperformance (we are currently going on 14 years of drawdown).

I also would recommend you take any back-tests that do not use actual funds with a grain of salt, including the Merriman one. Here you can play around with a US and ex-US small-cap value fund pair from DFA: https://www.portfoliovisualizer.com/bac ... tion3_3=50

Even when using PV you need to be careful as their asset allocation backtest tool uses Fama French data, which doesn't include transaction costs and is from a period of time before there were funds.

Like I said though, the key is your own temperament. Many people are panicking over the long and slow drawdown which has accelerated over the last two years. Many other people are salivating at the opportunity to buy even cheaper shares. Where you land on that spectrum will determine how well suited you are for a bolder tilt.

For a data point, I am a believer and outside my 35% S&P500 investment, the remaining 65% is all in value & small leaning US & ex-US funds, with some of it in very deep value funds (<50 holdings per fund). I think 20-30% of equity is a good balance though between reducing regret, and still making a meaningful enough difference to be worth the complexity.

35% S&P500
30% FNDC (ex-US small/value-ish)
20% AVUV (US small value)
10% QVAL (US deep value)
5% IVAL (ex-US deep value)
Last edited by MotoTrojan on Mon Jun 22, 2020 2:12 pm, edited 2 times in total.

MotoTrojan
Posts: 9952
Joined: Wed Feb 01, 2017 8:39 pm

Re: How much tilting to SCV (rule of thumb)?

Post by MotoTrojan » Mon Jun 22, 2020 2:11 pm

CrossOverGuy wrote:
Mon Jun 22, 2020 2:06 pm
The numbers there don't really support it. It seems like some years Value does better, others Growth does better and that the Small-Cap Index captures the average (still quite good) of both. In index funds, generally, average beats big swings trying the beat the market and just captures the market.
If it was as simple as buying SCV and outperforming the market decade after decade, do you really think the premium would persist? Value isn't meant to beat the market more often than not, but historically over the long-run it has beat it handsomely.

Here are my two favorite recent papers on why the premium isn't dead, OP may find it interesting.

https://www.aqr.com/Insights/Perspectiv ... sting-Dead

https://www.researchaffiliates.com/en_u ... rated.html

CrossOverGuy
Posts: 311
Joined: Fri Apr 27, 2007 2:19 pm

Re: How much tilting to SCV (rule of thumb)?

Post by CrossOverGuy » Mon Jun 22, 2020 2:14 pm

If you buy some of Small-Cap Index, you'd have both value and growth stocks. But if you're willing to wait to make a killing for more than x years (15?) in small-cap value, of course, that's your prerogative.

valuables
Posts: 32
Joined: Fri Feb 07, 2020 7:55 pm

Re: How much tilting to SCV (rule of thumb)?

Post by valuables » Mon Jun 22, 2020 2:19 pm

There is no rule of thumb! Many very reasonable people here have looked at the data and suggested that 0% is what you should tilt to SCV (other than what's included in VTI). Other very reasonable people have concluded that they want to take the risks associated with tilting have tilted an amount that seemed reasonable to them. My tilt is approx. 18-20%. There are people who claim that this is a reckless strategy and there are others who tilt more. I'm committed to that tilt for life, be sure that you are too if/when you commit.

MotoTrojan
Posts: 9952
Joined: Wed Feb 01, 2017 8:39 pm

Re: How much tilting to SCV (rule of thumb)?

Post by MotoTrojan » Mon Jun 22, 2020 2:20 pm

CrossOverGuy wrote:
Mon Jun 22, 2020 2:14 pm
If you buy some of Small-Cap Index, you'd have both value and growth stocks. But if you're willing to wait to make a killing for more than x years (15?) in small-cap value, of course, that's your prerogative.
Frankly I would find it odd that someone would believe in the size premium but not the value premium. Either way, both small and value are in historic drawdowns, so you might not have to wait 15 years, you could be buying at the bottom today :twisted:.

rkhusky
Posts: 9498
Joined: Thu Aug 18, 2011 8:09 pm

Re: How much tilting to SCV (rule of thumb)?

Post by rkhusky » Mon Jun 22, 2020 2:22 pm

If you believe factor models, there is not much reason not to go 100% SCV. The models show that SCV funds achieve nearly 100% of the market return in addition to the small and value factor returns. The small and value returns are more risky though, so if you follow this route, you might want to decrease your overall equity percentage, perhaps by 10-20% absolute.
Last edited by rkhusky on Mon Jun 22, 2020 5:11 pm, edited 1 time in total.

CrossOverGuy
Posts: 311
Joined: Fri Apr 27, 2007 2:19 pm

Re: How much tilting to SCV (rule of thumb)?

Post by CrossOverGuy » Mon Jun 22, 2020 2:27 pm

Some small companies are going to eventually be medium sized ones, and some are going to become large-sized ones. Others will falter. Does anyone know in advance which? I don't believe so, so index funds have them as part of the pack. I've done reading that says at times small and medium sized companies have done better than larger ones (as in the S&P) at certain points. So one can tilt (or more to the point, invest a bit more in them) without moving towards growth or value part of the spectrum. But again, the 3-fund portfolio advocated by many here also works, too! :happy

KyleAAA
Posts: 8333
Joined: Wed Jul 01, 2009 5:35 pm
Contact:

Re: How much tilting to SCV (rule of thumb)?

Post by KyleAAA » Mon Jun 22, 2020 2:28 pm

I think 50% is as good a rule of thumb as any. I do not understand the logic behind reducing tilt as you near retirement.

Topic Author
international001
Posts: 1498
Joined: Thu Feb 15, 2018 7:31 pm

Re: How much tilting to SCV (rule of thumb)?

Post by international001 » Mon Jun 22, 2020 5:07 pm

KlangFool wrote:
Mon Jun 22, 2020 2:06 pm
OP,

What is your goal?

A) A tilt to SCV?

B) As part of the Larry portfolio?

(A) or (B)?

KlangFool
Just tilt to SCV, so I can capture higher returns over long enough time

Topic Author
international001
Posts: 1498
Joined: Thu Feb 15, 2018 7:31 pm

Re: How much tilting to SCV (rule of thumb)?

Post by international001 » Mon Jun 22, 2020 5:16 pm

I understand SCV has more risk (understood as variation of results). Thus, it would be better not to use them unless you have long enough horizon

Same concept than stocks vs bonds. If you only have 10 years horizon, 100% stocks (SP500) may under-perform in some periods. That's why most target funds start decreasing stock allocation around 20 years horizon. There may be different models and intuitions in different target funds, but they are similar because more or less they would have worked in the past to maximize returns while diminishing risk.

I am wondering if there are any studies looking at the same concept for SCV. I guess Vanguard doesn't believe in tilting to SCV. But what about Dimensional funds? Do they have any target funds with diminishing exposure to factors?

Blue456
Posts: 726
Joined: Tue Jun 04, 2019 5:46 am

Re: How much tilting to SCV (rule of thumb)?

Post by Blue456 » Mon Jun 22, 2020 7:33 pm

international001 wrote:
Mon Jun 22, 2020 1:59 pm
If you believe the SCV (or just S or just V) premium still exists, then it would make sense to tilt X% to SCV?
But what is the rule of thumb for X? Probably, the more time you have left for retirement, the more you should tilt to it.

Paul Merriman suggests 1.5 x years-to-retirement

https://paulmerriman.com/two-funds-for- ... etirement/

But even for a 15 year period it would see wise to tilt heavier than that (his 4 fund portfolio 25% LCB, 25% LCV, 25% SCB, 25% SCV outperforms always LCB https://paulmerriman.com/wp-content/upl ... 8-2019.pdf). So I don't know how he gets his rule of thumb.

Any standard guidance?
I would say you want to use % tilt that is going to move portfolio. I probably wouldn't do anything less than 15% and most likely minimum 20%.

User avatar
vineviz
Posts: 6725
Joined: Tue May 15, 2018 1:55 pm

Re: How much tilting to SCV (rule of thumb)?

Post by vineviz » Mon Jun 22, 2020 8:17 pm

international001 wrote:
Mon Jun 22, 2020 5:16 pm
I understand SCV has more risk (understood as variation of results). Thus, it would be better not to use them unless you have long enough horizon
This is a thing that people say that doesn't really make any sense. Generally speaking, the shorter your time horizon the more diversified your portfolio should be.

To the extent that you think small cap value stocks have higher expected returns than the total stock market and/or that SCV stocks are imperfectly correlated with the total stock market, a shorter time horizon would indicate a larger SCV tilt not a smaller one.

By way of historical data, the worst real annualized return for large cap US stocks over a 10-year period was -4.66%. The worst 10-year return for SCV was +0.22%. For rolling 15-year periods, the worst period for large cap US stocks was -1.14%/year. The worst annualized return for SCV over a 15-year period was +3.70%.

Basically, the shorter your time horizon the bigger the benefits from enhanced diversification. Those diversification benefits are optimized at somewhere between 20% and 50% of your US equity allocation, depending on which funds you are using.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

andrew99999
Posts: 688
Joined: Fri Jul 13, 2018 8:14 pm

Re: How much tilting to SCV (rule of thumb)?

Post by andrew99999 » Mon Jun 22, 2020 11:49 pm

vineviz wrote:
Mon Jun 22, 2020 8:17 pm
international001 wrote:
Mon Jun 22, 2020 5:16 pm
I understand SCV has more risk (understood as variation of results). Thus, it would be better not to use them unless you have long enough horizon
This is a thing that people say that doesn't really make any sense. Generally speaking, the shorter your time horizon the more diversified your portfolio should be.

To the extent that you think small cap value stocks have higher expected returns than the total stock market and/or that SCV stocks are imperfectly correlated with the total stock market, a shorter time horizon would indicate a larger SCV tilt not a smaller one.

By way of historical data, the worst real annualized return for large cap US stocks over a 10-year period was -4.66%. The worst 10-year return for SCV was +0.22%. For rolling 15-year periods, the worst period for large cap US stocks was -1.14%/year. The worst annualized return for SCV over a 15-year period was +3.70%.

Basically, the shorter your time horizon the bigger the benefits from enhanced diversification. Those diversification benefits are optimized at somewhere between 20% and 50% of your US equity allocation, depending on which funds you are using.
Correct me if I'm wrong, but my understanding is that while diversification narrows down the standard deviation, it doesn't necessarily remove downside risk due to the fact that correlations often tend to 1 in a global equity crisis, and this would be particularly apparent with small and value stocks that tend to get unloaded first and fall further in a global equity crisis.

Of course, the exception is the case of a boom popping such as 1990 for Japan and tech in 2000, where pretty much anything other than what is crashing back to earth helped by simply not being what was crashing, which is why everything outside of tech helped in 2000 (small, value, REITs, infrastructure, everything really).

But besides that type of crash, I would have thought that the increased downside risk of small and value would be a potential problem for those with short time horizons, not only due to the sharper draw downs, but also the fact that they can be below the market return for extremely long periods of time.

User avatar
Forester
Posts: 1289
Joined: Sat Jan 19, 2019 2:50 pm
Location: UK

Re: How much tilting to SCV (rule of thumb)?

Post by Forester » Tue Jun 23, 2020 2:14 am

international001 wrote:
Mon Jun 22, 2020 1:59 pm
If you believe the SCV (or just S or just V) premium still exists, then it would make sense to tilt X% to SCV?
But what is the rule of thumb for X? Probably, the more time you have left for retirement, the more you should tilt to it.

Paul Merriman suggests 1.5 x years-to-retirement

https://paulmerriman.com/two-funds-for- ... etirement/

But even for a 15 year period it would see wise to tilt heavier than that (his 4 fund portfolio 25% LCB, 25% LCV, 25% SCB, 25% SCV outperforms always LCB https://paulmerriman.com/wp-content/upl ... 8-2019.pdf). So I don't know how he gets his rule of thumb.

Any standard guidance?
25% to 50% of your equities in value or SCV. It's not a hard science. I think a tilt toward SCV and/or commodity producers is worth it even if only to serve as inflation protection.

User avatar
vineviz
Posts: 6725
Joined: Tue May 15, 2018 1:55 pm

Re: How much tilting to SCV (rule of thumb)?

Post by vineviz » Tue Jun 23, 2020 7:23 am

andrew99999 wrote:
Mon Jun 22, 2020 11:49 pm
Correct me if I'm wrong, but my understanding is that while diversification narrows down the standard deviation, it doesn't necessarily remove downside risk due to the fact that correlations often tend to 1 in a global equity crisis, and this would be particularly apparent with small and value stocks that tend to get unloaded first and fall further in a global equity crisis.
The "correlations go to 1 in a crash" is mostly a thing that traders say to each other to make themselves feel better after doing something stupid. It's not actually true, isn't very relevant for an investor of any sort, and wouldn't be very actionable even if it were true.

andrew99999 wrote:
Mon Jun 22, 2020 11:49 pm
But besides that type of crash, I would have thought that the increased downside risk of small and value would be a potential problem for those with short time horizons, not only due to the sharper draw downs, but also the fact that they can be below the market return for extremely long periods of time.
Obviously it's important that the overall AMOUNT of risk you take in a portfolio should be appropriate for your situation. If your investment horizon is very short then your overall ability to handle portfolio volatility is probably relatively low. An investor with a one-year, or even five-year, investment horizon probably shouldn't be holding a very high allocation to stocks or long-term bonds to begin with.

But at whatever level of overall portfolio volatility you deem appropriate, it must be true that a more diversified portfolio will be superior to a less diversified portfolio.

And while a SCV portfolio can underperform a total market portfolio for long periods of time (and vice versa), it is still true that the expected return will be higher for SCV for any period of time. Although it is true that SCV can underperform, that doesn't mean that you should expect SCV to underperform. In fact, we expect the opposite.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
Uncorrelated
Posts: 734
Joined: Sun Oct 13, 2019 3:16 pm

Re: How much tilting to SCV (rule of thumb)?

Post by Uncorrelated » Tue Jun 23, 2020 8:17 am

Unless an investor is leverage constrained, the optimal tilt does not vary throughout time. If you believe that it is optimal to tilt 1/3rd of your portfolio to SCV, then you can expect that to be very close to optimal for portfolio's ranging from 100% equities to 10% equities. I don't know why Paul Merriman recommends a tilt based on age, but it sounds like a crude and simple attempt to approximate lifecycle investing.

Technically, the optimal tilt is approximately equal parts of the factors MKT, HmL and SmB (among others) because all factors have approximately equal sharpe ratio's. However, obtaining this tilt requires funds that do not exist and would be to expensive, if they existed. The next best theoretical option is to go 100% SVC, but that also has disadvantages because extreme tilts increase idiosyncratic (uncompensated) risk in the market factor in ways that are very hard to measure. To my knowledge no accurate models exist that allow investors to calculate which tilt is optimal.

I suggest tilting to a maximum of 1/3 market, 1/3 SVC and 1/3 value. As you can tell by the round numbers, this suggestion is not based on mathematical rigor, but is a best-effort estimate that sounds reasonable. If you are really leverage constrained, you might want an even higher tilt. If you're not so convinced, you probably want a lower tilt.

Blue456
Posts: 726
Joined: Tue Jun 04, 2019 5:46 am

Re: How much tilting to SCV (rule of thumb)?

Post by Blue456 » Tue Jun 23, 2020 8:18 am

Uncorrelated wrote:
Tue Jun 23, 2020 8:17 am
I don't know why Paul Merriman recommends a tilt based on age, but it sounds like a crude and simple attempt to approximate lifecycle investing.
I suspect this is because small cap is more risky than large cap.

User avatar
vineviz
Posts: 6725
Joined: Tue May 15, 2018 1:55 pm

Re: How much tilting to SCV (rule of thumb)?

Post by vineviz » Tue Jun 23, 2020 8:46 am

Blue456 wrote:
Tue Jun 23, 2020 8:18 am
Uncorrelated wrote:
Tue Jun 23, 2020 8:17 am
I don't know why Paul Merriman recommends a tilt based on age, but it sounds like a crude and simple attempt to approximate lifecycle investing.
I suspect this is because small cap is more risky than large cap.
It could also be that Merriman is just winging it. A lot of his model portfolios seem to be completely decoupled from any rigorous approach to asset allocation.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

hdas
Posts: 1317
Joined: Thu Jun 11, 2015 8:24 am

-----

Post by hdas » Tue Jun 23, 2020 9:09 am

-----
Last edited by hdas on Thu Jul 09, 2020 4:58 pm, edited 1 time in total.
....

User avatar
nisiprius
Advisory Board
Posts: 41077
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: How much tilting to SCV (rule of thumb)?

Post by nisiprius » Tue Jun 23, 2020 9:31 am

One of the weaknesses of the "small-cap value tilt" idea is that I don't remember seeing any quantitative analyses of how much, although there may have been some in the forum a decade ago.

The other one is the gap between what Fama seems to have been saying about it, and what enthusiasts like Paul Merriman say. My understanding is that Fama says it is all a matter of personal taste--taste for small-cap value's "dimensions of risk." The non-idiosyncratic behavior of stocks can't be captured solely by beta, so you have two (or six or two hundred) other factors.

Your stock allocation cannot be determined objectively, but depends on your risk tolerance and your personal taste for beta, your SCV tilt. I would infer that your SCV allocation should similarly be governed by your personal tolerance and preference for the characteristically-different behavior of small-cap value from the stock market as a whole. That would include negative skew. Personally, I detest negative skew, and I won't apologize for that.

In any case, just for laughs, we can try an experiment, although the results will depend on whatever time period is available in PortfolioVisualizer--hopefully since inception of DFA US Small Cap Value Portfolio, DFSVX. And I'm picking that because it is both one of the oldest and one of the best-liked by factor mavens. I'm going to go into PortfolioVisualizer with a portfolio of 100% Vanguard Balanced Index (60/40 US stocks and bonds) and 0% DFSVX and let it tell me what the optimized allocation is, choosing "Portfolio Type: Tickers" and otherwise accepting all defaults. I am stating this upfront and posting this now, and then will post the results whatever they turn out to be. Here goes.
Last edited by nisiprius on Tue Jun 23, 2020 9:59 am, 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.

User avatar
nisiprius
Advisory Board
Posts: 41077
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: How much tilting to SCV (rule of thumb)?

Post by nisiprius » Tue Jun 23, 2020 9:42 am

And here are the results. "The time period was constrained by the available data for DFA US Small Cap Value I (DFSVX) [Mar 1993 - May 2020]." That means, incidentally that it includes the time period from 2000-2003 when small-cap value performed magnificently, and 26 of the 28 years, i.e. 93% of the time since Fama and French published their factor papers.

Based on whatever methodology is used by PortfolioVisualizer, over the time period March 1993 - May 2020,

the optimum allocation of DFSVX to add to a "portfolio" of VBINX would have been: zero.

Source

Image
Image

By the way, it doesn't depend on the starting allocation you use. If you give it a starting portfolio of 100% DFSVX, you get the same result. (I wish I had said I would do it that way, as it would have been funnier, but I needed to keep my promise and do exactly what I said up front I would do).

If you start with a "provided portfolio" = 0% VBINX, 100% DFSVX, if gives you the same optimized final portfolio of 100% VBINX. From the numbers you can see what has happened. Small-cap value had higher return, but higher risk, but the PV optimization said the extra return wasn't enough to justify the risk--and that the highest risk-adjusted return was with pure VBINX, not anything in between.

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.

RocketShipTech
Posts: 247
Joined: Sat Jun 13, 2020 10:08 pm

Re: How much tilting to SCV (rule of thumb)?

Post by RocketShipTech » Tue Jun 23, 2020 9:50 am

nisiprius wrote:
Tue Jun 23, 2020 9:42 am
And here are the results. "The time period was constrained by the available data for DFA US Small Cap Value I (DFSVX) [Mar 1993 - May 2020]." That means, incidentally that it includes the time period from 2000-2003 when small-cap value performed magnificently, and 26 of the 28 years, i.e. 93% of the time since Fama and French published their factor papers.

Based on whatever methodology is used by PortfolioVisualizer, over the time period March 1993 - May 2020,

the optimum allocation of DFSVX to add to a "portfolio" of VBINX would have been: zero.

Source

Image
Image

By the way, it doesn't depend on the starting allocation you use. If you give it a starting portfolio of 100% DFSVX, you get the same result. (I wish I had said I would do it that way, as it would have been funnier, but I needed to keep my promise and do exactly what I said up front I would do).

If you start with a "provided portfolio" = 0% VBINX, 100% DFSVX, if gives you the same optimized final portfolio of 100% VBINX. From the numbers you can see what has happened. Small-cap value had higher return, but higher risk, but the PV optimization said the extra return wasn't enough to justify the risk--and that the highest risk-adjusted return was with pure VBINX, not anything in between.

Image
Not sure why you used the balanced fund which locks the stock / bond AA

Try this on for size:

https://www.portfoliovisualizer.com/eff ... tion3_1=40

61% Long term Treasuries
25% S&P 500
14% Small cap value

User avatar
vineviz
Posts: 6725
Joined: Tue May 15, 2018 1:55 pm

Re: How much tilting to SCV (rule of thumb)?

Post by vineviz » Tue Jun 23, 2020 10:12 am

hdas wrote:
Tue Jun 23, 2020 9:09 am
Right. But the question is also not well formulated, and more generally, the idea of tilting is non sensical. You either want a exposure or not, and then you optimize the weights of the portfolio as a whole, and apply leverage. I posit that the optimal way to do this is to combine SCV with Large Cap Growth (VUG, QQQ, not a fan of MTUM). The case for Total Stock Market + SCV doesn't really make sense to me, does it to you?. Could you chime in?. H
I don't agree that "tilting" is nonsensical, although I suppose I could see an argument that it might reflect a lack of conviction.

However, the optimal allocation inherently depends on what (specifically) you're trying to optimize. If the goal is merely to diversify among the three classic factors (market, size, value) then a mix of Total Stock Market + SCV will do that better than a mix of LCG + SCV or MTUM/VUG/QQQ + SCV.

On the other hand, if the investor wants to include momentum in their diversification schema (as you might reasonably want to do) then you might very well find a combination of funds that does that better than just Total Stock Market + SCV.

Furthermore, I think it's important to acknowledge that investors are (usually) human: we are not purely rational or emotionless creatures. I have a pretty powerful streak of independence and even contrarianism when it comes to investing, but I know that not all investors have the same outlook or temperament. Long periods of significantly underperforming the S&P 500 (for instance) don't bother me much at all, so a portfolio that behaves differently isn't a problem for me. Risk tolerance has a lot of dimensions, and "aversion to being different from the crowd/default/market" is one of those dimensions. Rather than view the market vs tilt question as a binary one, I tend to view it as simply one of the continuums along which each investor needs to find their place.
"Far more money has been lost by investors preparing for corrections than has been lost in corrections themselves." ~~ Peter Lynch

User avatar
Uncorrelated
Posts: 734
Joined: Sun Oct 13, 2019 3:16 pm

Re: How much tilting to SCV (rule of thumb)?

Post by Uncorrelated » Tue Jun 23, 2020 10:15 am

nisiprius wrote:
Tue Jun 23, 2020 9:42 am
And here are the results. "The time period was constrained by the available data for DFA US Small Cap Value I (DFSVX) [Mar 1993 - May 2020]." That means, incidentally that it includes the time period from 2000-2003 when small-cap value performed magnificently, and 26 of the 28 years, i.e. 93% of the time since Fama and French published their factor papers.

Based on whatever methodology is used by PortfolioVisualizer, over the time period March 1993 - May 2020,

the optimum allocation of DFSVX to add to a "portfolio" of VBINX would have been: zero.
The tangency portfolio isn't the optimal portfolio. It's the optimal portfolio under the specific condition that leverage is free. If that condition is not met, it's one of the many possible optimal portfolio's.

If you look at the entire frontier, it's pretty clear that all portfolio's between 0% and 100% DFSVX hit some optimum. Which is the right one? That depends on the risk tolerance of the individual.

Elysium
Posts: 2940
Joined: Mon Apr 02, 2007 6:22 pm

Re: How much tilting to SCV (rule of thumb)?

Post by Elysium » Tue Jun 23, 2020 10:26 am

Finding the optimal allocation (efficient frontier) has no value for future allocations, as the entire model is dependent on past data and the patterns repeating themselves into future, which is highly unlikely. All it can tell you is what worked best in the past. Given changes in rates, valuations, business models, inflation expectations, currency valuations, the number of possibilities are infinite. It's pointless to even try.

User avatar
nisiprius
Advisory Board
Posts: 41077
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: How much tilting to SCV (rule of thumb)?

Post by nisiprius » Tue Jun 23, 2020 10:34 am

RocketShipTech wrote:
Tue Jun 23, 2020 9:50 am
Not sure why you used the balanced fund which locks the stock / bond AA

Try this on for size:
I don't want to go too far down the rabbithole of PortfolioVisualizer fun 'n' games, and there are so things one could try. But, yeah, in my personal case it probably would have been more appropriate to go in with VTSMX 60%, VBMFX 40%, DFSVX 0% and lock the VBMFX allocation at 40% (by setting minimum 40%, maximum 40.01%), on the theory that I just want to experiment with what's in the stock allocation, not with the amount allocated to stocks. That still gives you 0% DFSVX as the "optimum," as it happens, but I won't even bother to post the link because "try this, try that" more or less kills any validity of backtesting at all.

What I really would like to see in a backtest, is if you choose your desired amount of risk--such as "the risk of a 50/50 stock/bond portfolio"--add DFSVX to it in varying amount, adjusting the rest of the portfolio to hold risk constant while finding the highest return... what amount gives the largest improvement in return at constant risk, and how large is that improvement?
Last edited by nisiprius on Tue Jun 23, 2020 10:43 am, 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.

User avatar
nisiprius
Advisory Board
Posts: 41077
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: How much tilting to SCV (rule of thumb)?

Post by nisiprius » Tue Jun 23, 2020 10:37 am

To riff a bit on what Vineviz said, yes: you need to know your personal "aversion to being different from the crowd/default/market."

In my opinion backtesting over long periods of time, if it makes sense at all, only makes sense if you have a strong conviction that you are likely to maintain the backtested strategy for a long period of time into the future. A long-term strategy that is reviewed and changed every six months is not a long-term strategy, and long-term backtests become irrelevant.*

Therefore, one of the portfolio characteristics that is important to evaluate is a) what is your planned holding period, and b) what is the probability that you personally will be willing to hold it for that long?

At the lowest level, it governs your stock allocation. With SCV, as Vineviz says, it governs how far you should be willing to depart from "the crowd/default/market."



*That is why I find it so intensely annoying when e.g. Tony Robbins presents a 55%-bonds "all-seasons portfolio" and says it was dictated to him by Ray Dalio personally and that "he began to unfold the exact sequence for what his experience shows will give you and me the increased probability of the highest return in any market environment, as long as we live, with the least amount of risk." And then just six years later Ray Dalio says "you'd be pretty crazy to hold bonds' right now" and neither Robbins or Dalio takes responsibility for explaining to people, who believed the portfolio would be good in "any" market environment, ought to be doing "right now."
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
nedsaid
Posts: 13463
Joined: Fri Nov 23, 2012 12:33 pm

Re: How much tilting to SCV (rule of thumb)?

Post by nedsaid » Tue Jun 23, 2020 10:42 am

How large of a tilt to Small Value? Don't really know, my sense is that for most investors you wouldn't want more than 20% of an Equity portfolio to be in Small Value. Folks forget that this is a very volatile asset class, Vanguard Small Value Index ETF, a "sort of" Small Value product has a three year standard deviation of 23.31 compared to 17.71 for Total Market Index. iShares S&P 600 Small Value Index has a three year standard deviation of 23.37. I have seen standard deviation numbers higher than that for Small Value. Also telling that the beta of iShares S&P 600 Small Value Index is 1.27. I think of it as an additive to put more octane into the gas, I want a boost in performance, I don't want to blow up the gas tank. Not an advocate of extreme tilts.
A fool and his money are good for business.

User avatar
nisiprius
Advisory Board
Posts: 41077
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: How much tilting to SCV (rule of thumb)?

Post by nisiprius » Tue Jun 23, 2020 10:52 am

Of course another approach is "consensus among those advocating it."

A 60/40 model portfolio published by Larry Swedroe in 1998 included 10% large-cap blend, 10% US large-cap value, 5% small-cap blend, 15% US large-cap value. And 5% REIT, 15% international.

The 60/40 Bill Schultheis Coffeehouse Portfolio, originally published at about the same time, includes in US 10% large-cap blend, 10% large-cap value, 10% small-cap blend, 10% small-cap value. And 10% REIT, 10% international.

The Merriman Ultimate Buy-and-hold, at least as shown in PortfolioVisualizer, allocates 48% to stocks with 6% each in large-cap, large-cap value, small-cap, small-cap value. And 6% each to US REITs, international developed, emerging markets, international value.

I have no idea how to translate any of those into "how much SCV" but I find it odd interesting that all three of them dilute SCV so much with small-cap blend and large-cap value. I once asked about this and the factor mavens didn't really have an answer, other than "not being too extreme."

I deduce that a "rule of thumb" is not to have all your small and your value exposure in the form of "small-cap value." But I don't know why not.
Last edited by nisiprius on Tue Jun 23, 2020 10:55 am, 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.

User avatar
Steve Reading
Posts: 1878
Joined: Fri Nov 16, 2018 10:20 pm

Re: How much tilting to SCV (rule of thumb)?

Post by Steve Reading » Tue Jun 23, 2020 10:53 am

nisiprius wrote:
Tue Jun 23, 2020 9:42 am
And here are the results. "The time period was constrained by the available data for DFA US Small Cap Value I (DFSVX) [Mar 1993 - May 2020]." That means, incidentally that it includes the time period from 2000-2003 when small-cap value performed magnificently, and 26 of the 28 years, i.e. 93% of the time since Fama and French published their factor papers.

Based on whatever methodology is used by PortfolioVisualizer, over the time period March 1993 - May 2020,

the optimum allocation of DFSVX to add to a "portfolio" of VBINX would have been: zero.

Source

Image
Image

By the way, it doesn't depend on the starting allocation you use. If you give it a starting portfolio of 100% DFSVX, you get the same result. (I wish I had said I would do it that way, as it would have been funnier, but I needed to keep my promise and do exactly what I said up front I would do).

If you start with a "provided portfolio" = 0% VBINX, 100% DFSVX, if gives you the same optimized final portfolio of 100% VBINX. From the numbers you can see what has happened. Small-cap value had higher return, but higher risk, but the PV optimization said the extra return wasn't enough to justify the risk--and that the highest risk-adjusted return was with pure VBINX, not anything in between.

Image
Now all we need is to find a way to borrow at the risk-free rate and we should be all set :)

JamesDean44
Posts: 155
Joined: Sat Jan 13, 2018 7:36 pm

Re: How much tilting to SCV (rule of thumb)?

Post by JamesDean44 » Tue Jun 23, 2020 1:56 pm

Not sure I'm adding anything that hasn't been said above, but I'll throw in my 2 cents:

It doesn't make any sense to me why one would reduce the SCV allocation as a percentage of equities as one draws nearer to retirement. If you believe that exposure to certain factors diversifies a portfolio, then you'd want that diversification benefit at all times. If you believe that exposure to certain factors increases risk-adjusted returns, then you'd want that benefit at all times. (I guess if you don't think that SCV provides a diversification benefit and you don't think that SCV improves risk-adjusted returns, then sure, I can see why you'd want to reduce as you get closer to retirement. But then why are you investing in SCV in the first place?)

In terms of how much to tilt, it depends upon a number of considerations, including the composition of the rest of the portfolio, the particular factor exposures desired, the makeup of the SCV fund (e.g., compare VBR and BOSV), and the investor's ability to handle a portfolio that behaves differently than the S&P 500 or whatever benchmark. In general terms, I think holding anywhere from 25% to 50% SCV as a percentage of US equities is reasonable. Some hold US SCV as their sole US equity holding. There isn't really a "rule of thumb." And obviously some folks have less faith in the size factor as compared to the value factor, so they might overweight value more than size.

I am not sure what the point of Nisiprius's chart is. One can optimize for all sorts of different things using different time periods. That one chart isn't very illustrative, unless I am missing the point.

theorist
Posts: 631
Joined: Sat Sep 28, 2019 11:39 am

Re: How much tilting to SCV (rule of thumb)?

Post by theorist » Tue Jun 23, 2020 2:10 pm

I have an ignorant question that probably fits in this thread as well as any:

I’ve been a bit confused about why “smallness” and “valueness” (or, say, market cap and P/B for definiteness) are good ways to slice the market. Clearly, at least at any given time, the sectors are differently represented in e.g. small caps vs large caps. And one could imagine over or under weighting certain sectors instead of certain cap sizes and relative P/B values. Because sector representation has correlated with cap size and P/B (maybe in a way that has varied in a complicated manner over time), it would be work to deconvolve what is driving the bulk of return differences. But how do we know it isn’t the sectors composing small value that do better statistically (and lumpily) over time, and we wouldn’t be better off just over weighting certain sectors instead of (small) size and value?

TL/DR: do we know you shouldn’t focus on certain sectors instead of size and value factors to optimize returns?

(Actually the same question would apply to US vs international investing too.)

rkhusky
Posts: 9498
Joined: Thu Aug 18, 2011 8:09 pm

Re: How much tilting to SCV (rule of thumb)?

Post by rkhusky » Tue Jun 23, 2020 3:46 pm

theorist wrote:
Tue Jun 23, 2020 2:10 pm
I’ve been a bit confused about why “smallness” and “valueness” (or, say, market cap and P/B for definiteness) are good ways to slice the market.
Because of the success of the Fama-French model in explaining investment returns. They hit upon a combination of variables (RiskFree, Market-RiskFree, HmL, SmB) that was able to explain past returns with only a few free parameters. Perhaps someone has tried to do the same with sectors, but it hasn't worked out as well. It wouldn't be that hard to try.

User avatar
nisiprius
Advisory Board
Posts: 41077
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: How much tilting to SCV (rule of thumb)?

Post by nisiprius » Tue Jun 23, 2020 4:47 pm

theorist wrote:
Tue Jun 23, 2020 2:10 pm
I have an ignorant question that probably fits in this thread as well as any:

I’ve been a bit confused about why “smallness” and “valueness” (or, say, market cap and P/B for definiteness) are good ways to slice the market. Clearly, at least at any given time, the sectors are differently represented in e.g. small caps vs large caps. And one could imagine over or under weighting certain sectors instead of certain cap sizes and relative P/B values. Because sector representation has correlated with cap size and P/B (maybe in a way that has varied in a complicated manner over time), it would be work to deconvolve what is driving the bulk of return differences. But how do we know it isn’t the sectors composing small value that do better statistically (and lumpily) over time, and we wouldn’t be better off just over weighting certain sectors instead of (small) size and value?

TL/DR: do we know you shouldn’t focus on certain sectors instead of size and value factors to optimize returns?

(Actually the same question would apply to US vs international investing too.)
The theory is that for each sector, all that you need to know about it is how much it loads on the factors that are already known. A sector, such as "Financials," for example, doesn't have anything unique to it which isn't already captured in the existing factors.

For example, if you are trying to plot a course, all you need to know about New York is its two numbers, latitude and longitude. You don't get anything special by trying to take account of New Yorkiness or Philadelphianess, latitude and longitude do it all. The size and value factors are the equivalent of latitude and longitude. When you know them, you know all there is to know.

Of course that was before they "discovered" a dozen other factors, such as investment, momentum, charm, dysphoria, and phragmites.
Last edited by nisiprius on Tue Jun 23, 2020 5:36 pm, 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.

reln
Posts: 296
Joined: Fri Apr 19, 2019 4:01 pm

Re: How much tilting to SCV (rule of thumb)?

Post by reln » Tue Jun 23, 2020 4:52 pm

international001 wrote:
Mon Jun 22, 2020 1:59 pm
If you believe the SCV (or just S or just V) premium still exists, then it would make sense to tilt X% to SCV?
But what is the rule of thumb for X? Probably, the more time you have left for retirement, the more you should tilt to it.

Paul Merriman suggests 1.5 x years-to-retirement

https://paulmerriman.com/two-funds-for- ... etirement/

But even for a 15 year period it would see wise to tilt heavier than that (his 4 fund portfolio 25% LCB, 25% LCV, 25% SCB, 25% SCV outperforms always LCB https://paulmerriman.com/wp-content/upl ... 8-2019.pdf). So I don't know how he gets his rule of thumb.

Any standard guidance?
No standard guidance. It's simply the more you overweight any factor (not just small or value) the more exposure you'll have to said factor and the more that factor will explain your portfolio return if (or not) that extra risk paid off.

reln
Posts: 296
Joined: Fri Apr 19, 2019 4:01 pm

Re: How much tilting to SCV (rule of thumb)?

Post by reln » Tue Jun 23, 2020 5:02 pm

theorist wrote:
Tue Jun 23, 2020 2:10 pm
I have an ignorant question that probably fits in this thread as well as any:

I’ve been a bit confused about why “smallness” and “valueness” (or, say, market cap and P/B for definiteness) are good ways to slice the market. Clearly, at least at any given time, the sectors are differently represented in e.g. small caps vs large caps. And one could imagine over or under weighting certain sectors instead of certain cap sizes and relative P/B values. Because sector representation has correlated with cap size and P/B (maybe in a way that has varied in a complicated manner over time), it would be work to deconvolve what is driving the bulk of return differences. But how do we know it isn’t the sectors composing small value that do better statistically (and lumpily) over time, and we wouldn’t be better off just over weighting certain sectors instead of (small) size and value?

TL/DR: do we know you shouldn’t focus on certain sectors instead of size and value factors to optimize returns?

(Actually the same question would apply to US vs international investing too.)
The factor models are statistical models. They tested many variables for impact and found that the drivers of return are risk (market, small, value, profitablility, conservative investment, momentum, duration, credit). Sectors do not have explanatory value in statistical models of stock returns.

MotoTrojan
Posts: 9952
Joined: Wed Feb 01, 2017 8:39 pm

Re: How much tilting to SCV (rule of thumb)?

Post by MotoTrojan » Tue Jun 23, 2020 5:08 pm

theorist wrote:
Tue Jun 23, 2020 2:10 pm
I have an ignorant question that probably fits in this thread as well as any:

I’ve been a bit confused about why “smallness” and “valueness” (or, say, market cap and P/B for definiteness) are good ways to slice the market. Clearly, at least at any given time, the sectors are differently represented in e.g. small caps vs large caps. And one could imagine over or under weighting certain sectors instead of certain cap sizes and relative P/B values. Because sector representation has correlated with cap size and P/B (maybe in a way that has varied in a complicated manner over time), it would be work to deconvolve what is driving the bulk of return differences. But how do we know it isn’t the sectors composing small value that do better statistically (and lumpily) over time, and we wouldn’t be better off just over weighting certain sectors instead of (small) size and value?

TL/DR: do we know you shouldn’t focus on certain sectors instead of size and value factors to optimize returns?

(Actually the same question would apply to US vs international investing too.)
The sectors which constitute the majority of a value portfolio today aren't the same ones that did in the past. There have been long (and short) term rotations in what stocks fill the depths of the value deciles.

You are correct though that a value strategy can introduce sector-risk. I believe AQR has some strategies that remain sector-neutral and tilt to value within each sector, although this will of course increase costs and reduce the overall factor exposure obtained. If you want to keep your sector bets more reasonable you may prefer something like a RAFI Fundamental index (right now the US large cap is around 20% technology for example, splitting the difference between a value fund and the S&P500).

As to your US vs. International statement, some of the return differences certainly can be explained by sector differences. Technology is why the US is outperforming, and tech makes up far less abroad.

User avatar
Taylor Larimore
Advisory Board
Posts: 29542
Joined: Tue Feb 27, 2007 8:09 pm
Location: Miami FL

Re: How much tilting (gambling) to SCV?

Post by Taylor Larimore » Tue Jun 23, 2020 7:06 pm

international001 wrote:
Mon Jun 22, 2020 1:59 pm
If you believe the SCV (or just S or just V) premium still exists, then it would make sense to tilt X% to SCV?
But what is the rule of thumb for X? Probably, the more time you have left for retirement, the more you should tilt to it.

Paul Merriman suggests 1.5 x years-to-retirement

https://paulmerriman.com/two-funds-for- ... etirement/

But even for a 15 year period it would see wise to tilt heavier than that (his 4 fund portfolio 25% LCB, 25% LCV, 25% SCB, 25% SCV outperforms always LCB https://paulmerriman.com/wp-content/upl ... 8-2019.pdf). So I don't know how he gets his rule of thumb.

Any standard guidance?
international001:

Few things are "standard" in investing -- especially industry-promoted small-cap value stocks which are now the worst performing Morningstar asset-class.

In my view, it is a big mistake for investors to GAMBLE on which asset-class will outperform in the future. Simply buy the market (which ALREADY holds small-cap value market's weight) and forget-about-it.

Regarding Paul Merriman, his value tilted Ultimate Buy & Hold Portfolio has the worst performance of all eight MarketWatch Lazy Portfolios.

What Five Nobel Laureates Say

Best wishes
Taylor
Jack Bogle's Words of Wisdom: "Selecting funds that will significantly exceed market returns, a search in which hope springs eternal and in which past performance has proven of virtually no predictive value, is a loser’s game.” "The beauty of owning the market is that you eliminate individual stock risk, you eliminate market sector risk, and you eliminate manager risk."
"Simplicity is the master key to financial success." -- Jack Bogle

User avatar
whodidntante
Posts: 8344
Joined: Thu Jan 21, 2016 11:11 pm
Location: outside the echo chamber

Re: How much tilting to SCV (rule of thumb)?

Post by whodidntante » Tue Jun 23, 2020 7:32 pm

international001 wrote:
Mon Jun 22, 2020 5:07 pm
Just tilt to SCV, so I can capture higher returns over long enough time
OK, but that's not what a SCV tilt does. It exposes you to different risk factors, and provides higher expected returns. Especially since it's been taken to the woodshed. But you know what? Those higher expected returns might not show up for you over any time frame you're capable of investing in. It might not be sufficient to wait 10 years.

I invest in factors, but I don't go door to door trying to get others to do it. That's because I don't think most people should. I'm not trying to keep you out of the club by any stretch. Just make sure you want to join and know what it means to join. Value has been robust, and has been more reliable than market. But you can't eat the past.

User avatar
tetractys
Posts: 4830
Joined: Sat Mar 17, 2007 3:30 pm
Location: Along the Salish Sea

Re: How much tilting to SCV (rule of thumb)?

Post by tetractys » Tue Jun 23, 2020 8:12 pm

international001 wrote:
Mon Jun 22, 2020 1:59 pm
If you believe the SCV (or just S or just V) premium still exists, then it would make sense to tilt X% to SCV?
But what is the rule of thumb for X? Probably, the more time you have left for retirement, the more you should tilt to it.
Generally 50%. For example to tilt away from Total US go 50/50 Total US/US Small Value.

Keep the loadings about equal so that rebalancing pays off.

User avatar
grabiner
Advisory Board
Posts: 27157
Joined: Tue Feb 20, 2007 11:58 pm
Location: Columbia, MD

Re: How much tilting to SCV (rule of thumb)?

Post by grabiner » Tue Jun 23, 2020 10:25 pm

I would suggest weighting as a percentage of your stock portfolio, rather than as a percentage of your portfolio. (This is what I do; for years, my US stock has been 1/8 LG, 1/8 SG, 3/8 LV, 3/8 SV, but my US stock allocation decreases every year as I get closer to retirement.)
Wiki David Grabiner

YRT70
Posts: 530
Joined: Sat Apr 27, 2019 8:51 am

Re: How much tilting to SCV (rule of thumb)?

Post by YRT70 » Wed Jun 24, 2020 1:58 am

Ben Felix will be releasing his new model portfolios soon which should be interesting regarding the question how much to tilt.

hdas
Posts: 1317
Joined: Thu Jun 11, 2015 8:24 am

-----

Post by hdas » Wed Jun 24, 2020 9:01 am

-----
Last edited by hdas on Thu Jul 09, 2020 4:59 pm, edited 1 time in total.
....

Topic Author
international001
Posts: 1498
Joined: Thu Feb 15, 2018 7:31 pm

Re: How much tilting to SCV (rule of thumb)?

Post by international001 » Thu Jun 25, 2020 1:37 pm

vineviz wrote:
Mon Jun 22, 2020 8:17 pm
international001 wrote:
Mon Jun 22, 2020 5:16 pm
I understand SCV has more risk (understood as variation of results). Thus, it would be better not to use them unless you have long enough horizon
This is a thing that people say that doesn't really make any sense. Generally speaking, the shorter your time horizon the more diversified your portfolio should be.

To the extent that you think small cap value stocks have higher expected returns than the total stock market and/or that SCV stocks are imperfectly correlated with the total stock market, a shorter time horizon would indicate a larger SCV tilt not a smaller one.

By way of historical data, the worst real annualized return for large cap US stocks over a 10-year period was -4.66%. The worst 10-year return for SCV was +0.22%. For rolling 15-year periods, the worst period for large cap US stocks was -1.14%/year. The worst annualized return for SCV over a 15-year period was +3.70%.

Basically, the shorter your time horizon the bigger the benefits from enhanced diversification. Those diversification benefits are optimized at somewhere between 20% and 50% of your US equity allocation, depending on which funds you are using.
I am not sure I understand your meaning for diversification. I'll hold the same number of stocks, but I may weight more on SCV. Is this more diversified or less?

I don't want to compare time periods separately. I want to compare what is the chance that in a time period of X years SCV underperforms LCB. My understanding is that the greater X, that chance is lower

Topic Author
international001
Posts: 1498
Joined: Thu Feb 15, 2018 7:31 pm

Re: How much tilting to SCV (rule of thumb)?

Post by international001 » Thu Jun 25, 2020 1:43 pm

Uncorrelated wrote:
Tue Jun 23, 2020 8:17 am
Technically, the optimal tilt is approximately equal parts of the factors MKT, HmL and SmB (among others) because all factors have approximately equal sharpe ratio's. However, obtaining this tilt requires funds that do not exist and would be to expensive, if they existed. The next best theoretical option is to go 100% SVC, but that also has disadvantages because extreme tilts increase idiosyncratic (uncompensated) risk in the market factor in ways that are very hard to measure. To my knowledge no accurate models exist that allow investors to calculate which tilt is optimal.
I'm not familiar with sharpe ratios for factors. You consider betas separately, but you use the same stdev? That certainly wouldn't give you similar sharpes. Any link appreciated

I suggest tilting to a maximum of 1/3 market, 1/3 SVC and 1/3 value. As you can tell by the round numbers, this suggestion is not based on mathematical rigor, but is a best-effort estimate that sounds reasonable. If you are really leverage constrained, you might want an even higher tilt. If you're not so convinced, you probably want a lower tilt.
This is more exposure to value than to size, no?

Topic Author
international001
Posts: 1498
Joined: Thu Feb 15, 2018 7:31 pm

Re: How much tilting to SCV (rule of thumb)?

Post by international001 » Thu Jun 25, 2020 1:47 pm

RocketShipTech wrote:
Tue Jun 23, 2020 9:50 am
[

Not sure why you used the balanced fund which locks the stock / bond AA

Try this on for size:

https://www.portfoliovisualizer.com/eff ... tion3_1=40

61% Long term Treasuries
25% S&P 500
14% Small cap value

If you have to believe the last 25 years, and use LT as bonds, then the model is clear. The less LT bonds you have, the more tilting towards SCV you need. For 100% stocks, that you have usually when your horizon investment is 100% stocks, then you can go 100% SCV. On retirment, if you have 40% in LT, then you want 50% on LB and 50% on SCV

Topic Author
international001
Posts: 1498
Joined: Thu Feb 15, 2018 7:31 pm

Re: How much tilting to SCV (rule of thumb)?

Post by international001 » Thu Jun 25, 2020 2:00 pm

Uncorrelated wrote:
Tue Jun 23, 2020 8:17 am
Unless an investor is leverage constrained, the optimal tilt does not vary throughout time. If you believe that it is optimal to tilt 1/3rd of your portfolio to SCV, then you can expect that to be very close to optimal for portfolio's ranging from 100% equities to 10% equities. I don't know why Paul Merriman recommends a tilt based on age, but it sounds like a crude and simple attempt to approximate lifecycle investing.
Well, SCV has more return and more stdev, so it would seem normal to have more SCV if you are further from retirement. I just don't trust his exacts numbers.

Post Reply