Using SCV and EM

[mbh]

Using the Simba spreadsheet, I took a look at the impact of using SCV (VISVX) and EM (VEIEX) for the equity portion.  The tables below summarize the results of one scenario. 

Note the following:
- Both options assume 25% Gold, 25% LT Bonds and 25% TBills/Treasury Money Market
- Option A assumes 25% TSM (VTSMX) for the equity portion
- Option B assumes 5% TSM, 15% SCV and 5% EM for the equity portion


Option A: Total US MarketStart      CAGR      Std. Dev  Down SD
1972      9.10%    8.02%   2.59%
1985      7.82%    5.83%   2.39%
1990      6.84%    5.44%   2.67%
1995      7.37%    5.33%   2.39%
2000      6.60%    5.40%      2.93%



Option B: SCV + EM EmphasisStart      CAGR        Std. Dev    Down SD
1972        10.17%    8.02%        2.29%
1985        8.51%    6.13%        2.39%
1990        7.46%    5.86%        2.67%
1995        7.76%    4.99%        1.99%
2000        8.54%    5.17%        2.24%


No matter which year I pick as the starting year, the CAGR for option B is better, the overall Std. Dev. is similar or better and the Down SD is similar or better.  In other words, using SCV and EM seems to boost returns and reduce risk.

Am I missing something?  Is there a reason why SCV and EM would be a bad choice in the equity piece?

[craigr]

Am I missing something?  Is there a reason why SCV and EM would be a bad choice in the equity piece?

Theoretically the more volatile funds could mesh OK. But there are several problems for me:

1) EM has currency risk as it is non-US dollar.
2) SCV funds are so widely available today that the past advantages may disappear because the cat is out of the bag.
3) SCV has had significant tracking error from the total stock market at times lagging it for some time.
4) For taxable investors these funds can have higher costs making the extra returns disappear to Uncle Sam.
5) Some of these specialty asset class returns are reconstructed from historical data where the index didn't exist in the past. There could be some biases.

Total stock market funds are theoretically the most efficient way to capture total market returns and SCV is a sector play that may or may not work out better going forward. I stick with TSM because it is simple and efficient. Some others will probably chime in shortly to explain why I'm wrong.

Overall though I was a slice and dice guy many years ago. But after looking at the issue in depth I just found the gamble to be too unreliable vs. a TSM approach. As a taxable investor, TSM also is a guaranteed tax savings for my situation where SCV is a loser. So if SCV lags the total stock market I have the additional pain of paying more taxes for the honor.

I posted a blog entry on this a while back:

https://web.archive.org/web/20160324133 ... cktesting/

The take-away is this:

"With the Permanent Portfolio your diversification comes from your Stock, Bond, Cash and Gold split and not from owning different stock asset classes."

Basically, I'm extremely cautious about backtesting. There are a lot of ways the data could not be valid but appear to be a free lunch.

But this topic always stirs up controversy. In the end, I'm a big fan of Occam.

[Roy]

2) SCV funds are so widely available today that the past advantages may disappear because the cat is out of the bag.
3) SCV has had significant tracking error from the total stock market at times lagging it for some time.

Total stock market funds are theoretically the most efficient way to capture total market returns and SCV is a sector play that may or may not work out better going forward. I stick with TSM because it is simple and efficient. Some others will probably chime in shortly to explain why I'm wrong.

https://web.archive.org/web/20160324133 ... cktesting/

The take-away is this:

"With the Permanent Portfolio your diversification comes from your Stock, Bond, Cash and Gold split and not from owning different stock asset classes."

Basically, I'm extremely cautious about backtesting. There are a lot of ways the data could not be valid but appear to be a free lunch.

But this topic always stirs up controversy. In the end, I'm a big fan of Occam.

Like Craig, I believe in implementing the PP using just four index-like funds (OK, maybe some international).  My opinion may not produce optimal results, but from a correlative and volatility standpoint, it makes sense to me from a returns and emotional viewpoint.  That said (and this just talking on the topics...):

Simplicity: Whatever that means, may not in itself produce better outcomes.  I agree one should not be any more complex than necessary to achieve a stated goal.  But in investing, that may require one fund or ten funds.  Some complain the 4x25 is not simple enough and feel they need just one fund.  At present, the 4x25 is not in one fund and a tiny bit of math is necessary, perhaps once or twice a year, to balance what many think is the best implementation of a well-diversified strategy.  If somehow one believed in the HB PP, and if it took 10 funds to accomplish it, should one find another strategy because it took 30-minutes a year to rebalance vice 5 minutes?  Many people making these arguments spend so much more time in serious discussion on investing debates;  assuming one is a buy and hold investor, minutes spent in rebalancing time can not possibly be the issue. So how many funds does one need?  I think the right amount is best.


Tracking Error Regret:  If one uses SCV or some combo of slices other than TSM, it won't really matter much since the PP portfolio deviance is already massive.  Anyone undertaking the traditional PP must immediately accept tracking error from a broad market standpoint (which is the point of comparison), or he will go mad.  Tracking Error is a necessary part of the PP.  Just look at Friday, for a one day example (the PP went up significantly as the broad markets got crushed), or in any year that the broad market declined or rose sharply, while the PP held steady or moved contrariwise. There are good reasons to avoid using SCV in the PP.  I don't believe Tracking Error regret—within the context of the PP—is one of them.  Conversely, I do think TE is a very good reason to avoid SCV if contemplating a traditional approach.  


Small Cap Value:  In no way is "tilting" (SCV or EM, say)  a "free lunch."  Neither is using common stocks a free lunch over bonds.  These are all based on a sound economic principle of risk/reward;  the backtesting simply demonstrates how those principles have worked over time—in all cases including for common stocks over bonds.

If freebies ever existed, they would have long since been arbitraged away.  Fama, the man who created the research most cited, argues that it is compensation for added risk.  Furthermore, Fama sees the "tilting" question as being one of investor preference that may tilt in any direction.  Recent research looks at relative liquidity (risk) in addition to Fama's work.  Bottom line, risk is key.  This research has nothing to do with any one approach being better/best.

Also, Fama examined individual stock performance (Large, Small, Growth, Value).  Bogle examined Stock FUNDS—that necessarily style drift, sometimes massively—which is why his conclusions are flawed;  he measured the wrong thing—which is why Bogle's "research" is more equivocal on the asset classes.

That said, TSM is the "optimal portfolio" (in stocks).  But optimal does not mean—necessarily—that all, or any, investors should own it.


Asset Class availability:  Fama recently commented on the wider accessibility of value funds.  Fama said:  "There is no evidence that the portfolio of all US equity mutual funds has become increasingly tilted toward small and value stocks over time. The aggregate portfolio still looks a lot like the market portfolio."

I think Fama is likely correct since many investors will style drift in and out of funds to chase returns or safety.  Risk, greed and fear are always present, and added fund availability may not affect those emotional factors.  


Backtesting:  I think it is both useful in understanding expected return, correlations between classes, and should never be the sole reason for allocation decisions.  One can't say past returns don't matter at all and then point to past performance, which is often done or implied in portfolio compositions.      


Again, I like the PP Stock component— as is—because during periods of extra-volatility (when all risky assets decline together) there is always the problem that the "extra" volatility of SCV and EM will prove too much to avoid capitulation.  And that extra volatility can happen over a very short period.  It is easy to look at past data, dispassionately, and "prove" something is (was) better.  But until you LIVE through that volatility, emotionally, you don't know how you'll respond. There probably is a better arrangement of classes than the HB PP.  But I don't know what that is and, emotionally, the past success of it, across multiple market cycles and economic conditions, discourages me from improving it.

Let's face it, no matter how well-argued the PP remains from an economics standpoint, the reason it is popular in discussion, or used, is because it has worked in the past to achieve its stated objective.  Had it behaved more like a traditional portfolio, we would not be having these conversations.

Roy
 

[MediumTex]

Roy,

That's a great post.  It got me to thinking about what an investor really needs to know when evaluating the PP (or any other approach to investing).

First, in evaluating an approach to investing and its suitability to a particular investor, it is important for the investor to understand his own emotional make-up when it comes to risk tolerance and the effect various market conditions will have on his emotional state.

Second, it is vital to have some depth of understanding with respect to the strategy behind the allocation method the investor is considering.  When I say "strategy" I am talking about the historical performance, the risk of the approach when compared to other investments, the tracking error, etc.  Sort of what shows up on Morningstar, but sans the marketing materials like goofy green ribbons on the sidewalk leading to retirement.  

Finally, when one has identified his risk tolerance and matched an investment strategy to that risk tolerance, I think it is then important to look into the philosophy behind the allocation strategy one has tentatively selected. When I say "philosophy" in this context, here is what I mean: Every investment approach has a set of economic, political, social, cultural, and ecological beliefs embedded in the assumptions on which the strategy is based (and these assumptions rarely show up in the prospectus). The way these beliefs are configured in the investment strategy is what I am referring to as the philosophy of the allocation strategy, and I think it is very important for the investor to understand the philosophy to which I am referring and make sure it is consistent with his own beliefs.

Here are philosophy-related questions that I think are important to ask about a proposed investment strategy:

1. To what extent does it reflect a belief in the ability to predict the future?  (Most managed funds reflect this belief.)

2. To what extent does it rely upon the durability of one or more existing political structures (e.g., populist democracy, corporate controlled democracy, monarchy, dictatorship, etc.)?

3. To what extent does it rely upon the durability of one or more economic structures (e.g., pure capitalism, hybrid capitalist-socialist structure, or something else)?

4. To what extent does it rely upon the durability of one or more monetary structures (e.g., a floating exchange rate, a gold standard, or certain levels of price stability or inflation)?

5. To what extent does it account for the tendency of humans to overdo it when it comes to both optimism on the way up (tending toward delusion) and pessimism on the way down (tending toward despair)?

6. To what extent does it rely on recent past performance (e.g., within the past few decades) as a predictor of future results?

7. To what extent does it seem to grasp the exponential function and its role in a world of finite resources?  In other words, since we know trees don't grow to the sky, the question then is how high they do grow before they stop.  Many investment strategies don't do a very good job of reconciling the abstract world of finance with the real world of human effort being applied to convert natural resources into capital and consumption items (and the inherent limitations built into this process).

8. To what extent does it allow for the maintenance of core levels of capital in the midst of natural disasters or man-made political and economic disasters?

9. To what extent does it compensate for the herding behavior and other cognitive biases of the humans who are making the tactical allocation decisions within the particular strategy?

10. To what extent does the viability of the strategy basically depend upon a "greater fool" methodology for taking any profits off the table?

11. When the strategy suffered its largest losses, was anyone surprised?  Did the initial beliefs about the strategy suggest such losses were possible?

12. To what extent does the strategy rely upon the ability to model the behavior of markets (and the humans who make up those markets) using quantitative methods of analysis?  

My personal preference for the PP is the result of thinking through these issues for myself and deciding what approach best matches my personal risk tolerance and my beliefs about investment strategy and investment philosophy.

[MadMoneyMachine]

1) EM has currency risk as it is non-US dollar.
2) SCV funds are so widely available today that the past advantages may disappear because the cat is out of the bag.
3) SCV has had significant tracking error from the total stock market at times lagging it for some time.
4) For taxable investors these funds can have higher costs making the extra returns disappear to Uncle Sam.
5) Some of these specialty asset class returns are reconstructed from historical data where the index didn't exist in the past. There could be some biases.

I agree with most things craigr and MediumTex teach on these forums with the exception of the Stocks component. I believe that the "Prosperity" portion of the PP should be invested in riskier stock indexes, and half SCV and half EM are fantastic here. Fama said that risk is the source of returns. The reason SCV and EM outperform TSM over the long term is that SCV and EM are riskier. I also *like* the currency risk here. I can't predict the future so why would I want to be 100% in dollars?
I believe TSM is too LC weighted to obtain the risk I want for the Prosperity portion of the PP. The Simba spreadsheet backtesting results are icing on the cake.

As far as expenses, VBR and VWO both have 0.20% expense ratio. They do give off some dividends annually, which I don't actually like, but the trade-off is OK for the expected capital gain.
BTW: I am in the process of putting my money where my mouth is and buying VBR and VWO.
But darn, I wish their prices would fall!

[Roy]

Every investment approach has a set of economic, political, social, cultural, and ecological beliefs embedded in the assumptions on which the strategy is based (and these assumptions rarely show up in the prospectus).

Tex,

This is damned good.  I'm going to read that repeatedly.  The questions are especially helpful in evaluating, or re-evaluating the PP, since its allocation/strategy grinds with and against more assumptions than any other strategy I've seen.

Sometimes, oftentimes, investors do not realize they are making bets.  For example, some Total Marketeers (TSM and TBM) believe they make no bets at all, since they are using market-weighted, "optimal" Total Market funds;  and in one sense that is true.  Yet, in every Target Retirement fund there are huge bets being made, regardless of the equity/fixed split.  And even if you find a fund that accords better with one's "risk profile" a bet is invariably made on the continuance of the equity risk premium—it's all factored in but never really talked about.  Bogle's "age in bonds" advice assumes this, even as the backstory goes largely unexplicated.  It may not be a mortal sin to make broad assumptions based on the relative premia, after all, they derive from reasonably sound economic principles, but when one depends on continuance of an established premia, there can be problems.

Similarly, there are many "no bets" variants out there, genuinely believing they indeed make no bets.  Now, if such a portfolio accords with an investor's personality (as Tex suggests), so that if some asset class does poorly another may do better (and that makes one feel good), then that may make emotional sense.  But we have discussed the problems with these portfolios before—that being the shared exposure to Beta, which can be lethal in downturns:  you can get duped at the worst possible time.

I don't care one wit about CAPM (talking about conventional ports, now).  To me, it means nothing and 3-Factor explains returns better.  But Tracking Error may mean a lot to one individual and so a Total Markets approach has emotionally validity for that individual.  (I think Tracking Error means a lot to all who peek (assuming conventional ports), whether they say so or not.)  TE can be managed (read, emotionally) with education, but it is definitely something to be managed. 

Low Beta portfolios weighted heavily to SCV, for example, may also assume added premia. But in Larry's reasoning, (as with the PP), there is an a priori understanding that the primary reason for using it is wealth protection, so if SCV does not outperform as it has in past decades, no big deal.  But if someone copying Larry depends on SCV outperformance, there can be problems (not just Tracking Error but failure to meet goals).  Most conservative portfolios (Total Market or Sliced/Diced) depend on the stability of their large bond positions.  This is reasonable, perhaps, especially with US Treasuries, but massive inflation is one risk there.

The PP seems to take more "Jokers" into account in its symmetrical deck of assumptions or non-assumptions, as one may view it.  It takes a lot of understanding but that's why we study and discuss. 

Suggestions to expand the Stock component of the PP may be good ones.  But I think one must consider always its no-lose mandate, and how the extra volatility caused by riskier asset classes may affect investor emotions—perhaps impelling capitulation or strategy switching back to TSM for the stock portion, when in crisis.  That is, for me, I'd worry that the Stock portion would suffer greater than it normally would, more than I would miss the added return the riskier assets provide.  I could be wrong, though...

Roy

[MediumTex]

Every investment approach has a set of economic, political, social, cultural, and ecological beliefs embedded in the assumptions on which the strategy is based (and these assumptions rarely show up in the prospectus).

Sometimes, oftentimes, investors do not realize they are making bets.  For example, some Total Marketeers (TSM and TBM) believe they make no bets at all, since they are using market-weighted, "optimal" Total Market funds;  and in one sense that is true.   Yet, in every Target Retirement fund there are huge bets being made, regardless of the equity/fixed split.  And even if you find a fund that accords better with one's "risk profile" a bet is invariably made on the continuance of the equity risk premium—it's all factored in but never really talked about.  Bogle's "age in bonds" advice assumes this, even as the backstory goes largely unexplicated.  It may not be a mortal sin to make broad assumptions based on the relative premia, after all, they derive from reasonably sound economic principles, but when one depends on continuance of an established premia, there can be problems.

As some readers know, I am a retirement plan attorney, which means I design retirement plans, assist plan sponsors with compliance matters, and do a lot of other stuff which will put you to sleep quickly if you're not careful.  One of the things I do is design investment policy statements for retirement plans (which is more of a mechanical than strategic exercise for most retirement plans).  I also find myself sitting in on a lot of investment strategy discussions and presentations from investment managers.  I am not a professional investment advisor, and thus I do not offer my clients any professional opinion on investments they may be considering.

Based upon my experience, the level of quality at the institutional investor level is often little better than Jim Cramer (though there are some good ones out there).  In most cases, it's still all about salesmanship, force of personality and the ability to tell a good story that validates the rationale behind a large fee (normally something like "this is complicated and cutting edge, no one else is doing this, we are way out ahead of the curve with our thinking on this", etc.).

To Roy's point, I often find myself listening to institutional investment guys talk about strategies and I just marvel at their ability to convince clients (and seemingly convince themselves) of what are basically high fee speculations based upon a narrow range of possible future scenarios.  It is as if by saying something enough times it becomes true (my favorite is "interest rates have nowhere to go but up"...which I've been hearing for 10 years and it hasn't been right yet).

This scramble for target retirement funds in recent years is sad.  If they actually did what they promised, they would be great...but they don't.  They are in many ways the worst of all possible worlds--they've got high fees, too much equity exposure, manager risk, and in the retirement plan world much less scrutiny than one would think.  The only larger investment product screw job I have seen in recent years has been many state 529 college saving plans--the funds in most of these plans are truly 100% garbage (fees have started to come down in some plans after recent lawsuits, however).

[craigr]

1) EM has currency risk as it is non-US dollar.
2) SCV funds are so widely available today that the past advantages may disappear because the cat is out of the bag.
3) SCV has had significant tracking error from the total stock market at times lagging it for some time.
4) For taxable investors these funds can have higher costs making the extra returns disappear to Uncle Sam.
5) Some of these specialty asset class returns are reconstructed from historical data where the index didn't exist in the past. There could be some biases.

I agree with most things craigr and MediumTex teach on these forums with the exception of the Stocks component. I believe that the "Prosperity" portion of the PP should be invested in riskier stock indexes, and half SCV and half EM are fantastic here. Fama said that risk is the source of returns. The reason SCV and EM outperform TSM over the long term is that SCV and EM are riskier. I also *like* the currency risk here. I can't predict the future so why would I want to be 100% in dollars?

If you read Browne and Coxon's earlier books they generally wanted stock funds with a high Beta so they would be more volatile than the market as a whole. So the advice then was to seek out "aggressive" stock funds. So the idea of using these more volatile funds is theoretically sound. However the advice eventually changed to use use a broad based index fund.

I don't know the exact reasoning, but I suspect it's because it's just so hard to find a fund that is consistently more volatile on the upside to the market as a whole. The target is just moving around way too much depending on what's en vogue. This is something I spoke about with John Chandler, Browne's former busines publisher and partner, and he concurred.

As for currency risk in EM, I really don't think it's needed. The portfolio holds 25% in gold which is currency neutral and provides very powerful diversification against the dollar. Currency risk could hurt or help. It just depends. With that said, I hold a small allocation to international but it's not enough to be a big impact. Maybe I'll change and just switch it over to EM in the future if I get the chance and count it as part of my variable portfolio.

In the end I'm much less religious about this issue now. If someone is using a low cost index fund and doing SCV I don't think it is going to hurt the portfolio as long as they know and expect some more volatility depending on how overweight they are. They may also expect it to lag at times and maybe excel at times to the total market. The investor just needs to have the fortitude to stick it out.

But then again I'd just say that past performance is no guarantee of future results. The SCV advantages may not continue because there are so many products out now that try to capture the past excess returns. Many years back I had a broker trying to sell me his services and his pitch was SCV tilting. This was a really huge firm doing this for all their clients (Merrill Lynch? Don't recall right now.). So I appreciate the risk story associated with SCV, but the markets just have a way of doing away with these advantages from my cynical perspective.

[craigr]

Every investment approach has a set of economic, political, social, cultural, and ecological beliefs embedded in the assumptions on which the strategy is based (and these assumptions rarely show up in the prospectus).

Quite true! Fund managers often have political leanings that influence their judgment. I've spoken with a well known investment guru in the past and he is very left leaning. I couldn't help but leave afterwards thinking that I will be very careful about listening to his advice from now on. I couldn't shake the idea that his political dogma was probably influencing a lot of his analysis.

Browne was of course a hard core Libertarian. But he was unusual in that his investing advice was fairly neutral. I can't think of many Libertarians that would own US Govt. Long Term bonds for instance.

[Roy]

If you read Browne and Coxon's earlier books they generally wanted stock funds with a high Beta so they would be more volatile than the market as a whole. So the advice then was to seek out "aggressive" stock funds. So the idea of using these more volatile funds is theoretically sound. However the advice eventually changed to use use a broad based index fund.

As for currency risk in EM, I really don't think it's needed. The portfolio holds 25% in gold which is currency neutral and provides very powerful diversification against the dollar. Currency risk could hurt or help. It just depends. With that said, I hold a small allocation to international but it's not enough to be a big impact. Maybe I'll change and just switch it over to EM in the future if I get the chance and count it as part of my variable portfolio.

In the end I'm much less religious about this issue now. If someone is using a low cost index fund and doing SCV I don't think it is going to hurt the portfolio as long as they know and expect some more volatility depending on how overweight they are. They may also expect it to lag at times and maybe excel at times to the total market. The investor just needs to have the fortitude to stick it out.

I agree.  I think currency risk, if needed, should be taken on the equity side.  The reason is that if taken in Bonds, the return isn't greater but the risks (foreign default) are moreso and that offsets the diversification gained.  But the currency diversification concept applies to conventional ports more.  In the PP, because of the Gold weight, it really is not needed.  And if it were, the comparatively small weighting to EM suggested would not be enough.  Also, if one follows all HB's advice, it is likely better overall simply to have part of one's wealth foreign-placed.

On the "tilt" side of things it may indeed make sense to add SCV, and EM, but I would be concerned about periods of high volatility where several classes are going down and then the Stocks tanking yet further than they would normally. That degree of risk to principal (or to investor emotion) is not consistent with the PP main objective.  But quite maybe I'm wrong on that as the overall Beta in the PP would still be low.  I do think if the PP can be improved, theoretically, the Stocks end is where that is, and maybe the Cash.  Though Craigr's suggestion to use ST Treasuries, vice some Cash, already accomplishes the latter.

Roy

[doug6zj9]

Quite true! Fund managers often have political leanings that influence their judgment. I've spoken with a well known investment guru in the past and he is very left leaning. I couldn't help but leave afterwards thinking that I will be very careful about listening to his advice from now on. I couldn't shake the idea that his political dogma was probably influencing a lot of his analysis.

Boy, I'll say...  I dropped Berkshire Hathaway and all Baron funds last year after reading both Buffet and Ron Baron gushing on about Obama in their writings.  I was shocked to see such childish adoration of a political hack coming from people managing significant sums of my money.    Of course they lost about 40% of it...

[doug6zj9]

...On the "tilt" side of things it may indeed make sense to add SCV, and EM, but I would be concerned about periods of high volatility where several classes are going down and then the Stocks tanking yet further than they would normally....

And I am unwilling to concede that past over or under-performance of these "slices" is indicative of their future.  Anything can happen and nothing must happen...  That is the whole point.

[Roy]

I believe TSM is too LC weighted to obtain the risk I want for the Prosperity portion of the PP.

Guys,

I find I am thinking about the Stocks component much more than I had. 

Aggressive Stocks:  the original mandate for Prosperity
As others have done, I too wondered if there is a better approach to achieve the "aggressive" mandate suggested by HB in the Stock portion for "Prosperity".  Whatever that word "aggressive" means, it seems to imply higher returns than less aggressive securities.  I don't think it means this must come from "Growth Stocks". I think it is the behavior of the security type that is the issue.  Remember, any benefits from SV derive not from the class itself but because of the migration of a few its stocks into a Larger and/or Growthier direction.  Now, many of us do not believe higher returns possible without higher risk.  So what Stock types do provide this sort of "aggression" in "prosperity"? 

I know many of us have looked at this repeatedly.  And for different reasons, many of us (me included) have settled on TSM as being the best representation of what the Stock component ought to be, just as HB might agree.  Further, many of us agree that the greatest diversification and explanatory power of the PP comes from its quartile composition.  But since this site permits a more focused analysis of the PP, and because 25% of any portfolio is still a large amount, I've been thinking on Harry's original mandate a bit more.    Since the OP and others mention Small Cap Value, and Craigr mentioned the Aggression issue, I decided initially to look again at the SV class. 

The following yearly analysis (going back as far as I could find stocks data, around 1927 up to recently but not sure the endpoint) shows the powerful effect of Beta exposure—that which is common to all stock types.  (I did not look at Emerging Markets or International Small Value yet.)  I believe Larry discussed these findings before.  I mention this data simply to offer some framework other than pure CAGR and Risk calculations.


• TSM and SV both had 24 years where the returns were Negative.

• 18 of those 24 years both TSM and SV were Negative (same time).

• There were 6 years where TSM was Negative and SV Positive.

• There were 6 years where SV was Negative and TSM Positive.

• There were NO extended periods where SV was Negative and TSM Positive.

• The years where SV was Negative and TSM positive = 1939,1948,1953,1960,1987,1998.

• The years where TSM was Negative and SV positive = 1970,1977,1981,2000,2001,2007.

• There were no 2 consecutive years where SV went down and TSM up.

• There was 1 string of 2 years where TSM went down and SV up (2000-2001).

Over the period examined, Small Value had higher returns, much like what Fama showed.


Tracking Error  and other concerns
Speaking only about Stocks for the moment, any allocation is making a bet on its composition.  Past returns may not assure future returns, but that applies to everything.

The nature of CAPM means, necessarily, that a massive weighting will be to Large Caps (which always dominate "the market").  That is not indifference;  it matters.  CAPM may called "neutral" owing to its structure, but that structure defaults to a huge commitment to a particular market segment.  (Madmoneymachine indicated this, above.)

Those who fear Tracking Error understandably want to own "the market" in its CAP weight.  With conventional portfolios, that argument makes lots of sense.  But the Tracking Error issues would not concern me in the PP.  First, the PP—as a portfolio— already has enormous Tracking Error, and it is doubtful that the low Beta exposure of the PP stock component would alter that significantly, but over time, returns may be improved without incurring significant risk (measured partly by Standard Deviation, which clearly is not the only risk). 

My concern has not been losing out on higher expected returns that SV (in some addition) might provide.  My concern was the periods of under-perfomance (even months) where additional SV losses may test my resolve especially when the other PP asset classes are also tanking (I'm thinking short term acute periods not just whole years).  As the PP concept is conservative, the Large Cap domination made some sense.  But consider the Stock component was originally supposed to be "aggressive and growthy". 

But, if the Beta exposure within the PP is low (and it is pretty low), this may be less a concern than I imagined.  My other reason was the past success of the PP, using TSM (an indexed version of HB's original active funds).  But if looking back matters at all, we can see any number of Stock arrangements could have increased portfolio returns without sacrificing much in Standard Deviation or  drawdown.  Thus, "aggression" paid off.  And if one believed an adjustment in the Stocks component were justified, should the adjustment be absolute or relative to where an investor is in their own retirement picture?  That is, would a younger investor feel better with a "more aggressive" portion than one already in retirement (More SV in the former, less, if any, in the latter)?

(I have not looked back to 1972 to see how SV would have impacted the PP in each year.  And I plan to look at pure Large Growth Stocks too)

Now if looking back does not matter, then strike every mention of past from all obelisks, remove the yearly returns charts, etc.  But I think each part of this (how to view past correlations, risks, and returns) holds some portion of the truth without any part being absolute.


No answers here, but for the moment, I think TSM is likely the best choice.  Though, I am less sure of that than before, even as I am certain the basic PP quartile approach is the factor with the greatest impact.  And maybe at day's end the Stock type does not matter much. But the Aggressive Stocks mandate is something I'd want to discuss a bit more.


Roy

[craigr]

No answers here, but for the moment, I think TSM is likely the best choice.  Though, I am less sure of that than before, even as I am certain the basic PP quartile approach is the factor with the greatest impact.  And maybe at day's end the Stock type does not matter much. But the Aggressive Stocks mandate is something I'd want to discuss a bit more.

Great post, Roy.

I think that the above is basically where I am. I feel the diversification in the portfolio comes from the core assets chosen. We can debate about whether gold is best for inflation vs. alternatives (I think it is). We can also debate if US Treasury LT bonds are best for deflation (I think it is). We can debate if ultra-safe Treasury Money Markets are best for cash (I think it is). So it is a perfectly rational debate to look at the stocks. Is TSM the best of our choices?

Well, let's back up a bit. I am quite certain that of stock fund types, an index fund is the best choice. Active management is very spotty and not likely to beat the market over the long run. This has been decisively shown. So we know we want the stocks to be an index. This is a change from the original portfolio that Browne also made.

So what index then?

TSM is my way of basically throwing up my hands in a true "I don't know" fashion. My main reasons for using it is because it may not be the best, but I'm unlikely to have any serious regrets.

In the past I was on the SV side of things and did slice and dice my (non-Permanent) portfolio. But after running it for a few years I found that the funds were expensive and the tax costs of rebalancing a bunch of different assets was very high. Finally, the funds sometimes had higher turnover increasing internal capital gains that spilled out. My own analysis showed that for a taxable investor much of the surplus gains (if any) went to the Govt. In essence, I was taking the risk but Uncle Sam was getting the rewards from my value tilting.

Now I think there is an argument though that for those that can stomach the extra volatility that maybe an SCV fund is fine for the stock portion. Or perhaps it's best as a mix with TSM as many like to do. It's possible. And in fact I really don't have any serious objections other than the ones above about tracking error and possible higher costs. But with new SCV index funds the costs have been going down and in a tax-deferred location it may not matter.

My advice overall is if someone wants to do it that they should only use an index fund. Of the primary indices available for SCV (S&P 600 Small Cap Value and Russell 2000) I think that the S&P 600 is better. They use a stricter screening criteria for companies entering the index and have lower turnover than the R2000 index. This makes the fund tolerable for taxable investors but also boosts returns vs. the R2000 index. I do know that the R2000 index criteria had been re-swizzled the past couple years so perhaps they fixed the other problems. I have to research it more. The criteria specifically would be:

1) A passive index
2) Low cost
3) Low turnover
4) Best tax-efficiency possible
5) Offered from an established index fund provider
6) Not a specialty index
7) Good liquidity and asset base size (see #6)
8) No gimmicks (e.g. Fundamental indexing)

But I use TSM basically because it is simple, cost efficient, and theoretically the best diversified as it captures all the market segments. If someone wanted to go the SCV route and use an index fund to do it it would probably work out fine if they stick to the plan.

[Roy]

Well, let's back up a bit. I am quite certain that of stock fund types, an index fund is the best choice. Active management is very spotty and not likely to beat the market over the long run. This has been decisively shown. So we know we want the stocks to be an index. This is a change from the original portfolio that Browne also made.

Now I think there is an argument though that for those that can stomach the extra volatility that maybe an SCV fund is fine for the stock portion. Or perhaps it's best as a mix with TSM as many like to do. It's possible. And in fact I really don't have any serious objections other than the ones above about tracking error and possible higher costs. But with new SCV index funds the costs have been going down and in a tax-deferred location it may not matter.

1) A passive index
2) Low cost
3) Low turnover
4) Best tax-efficiency possible
5) Offered from an established index fund provider
6) Not a specialty index
7) Good liquidity and asset base size (see #6)
No gimmicks (e.g. Fundamental indexing)

But I use TSM basically because it is simple, cost efficient, and theoretically the best diversified as it captures all the market segments. If someone wanted to go the SCV route and use an index fund to do it it would probably work out fine if they stick to the plan.

Craigr,

Wholly agreed on the requirements for a proper fund of any sort. 

Your last point is can not be overemphasized.  It is the sine qua non Fama himself makes for those who choose to "tilt".  He says they should be prepared to "ride their decision to the beach."  (Remember too, Fama sees "tilting" as possible in any direction, not just SV!  It's just that Tracking Error Regret applies when deviating from only the broad market—TSM or S&P 500)  So Fama's admonition would apply powerfully in 1998 when the broad market was up big and SV was down. And Gold would not be helping then either.  Now, as we see, that sort of up/down disparity with SV did not happen much, but the severe lag lasted 2 years, which in today's information age may feel like eternity for those who actually track the movements of their portfolio.  In recent years, anyone following discussion boards, saw how REITS went from "must haves" to dogs;  Like SV, REITS are influenced strongly by economic cycles.

Are you ("you" meaning anyone) prepared to ride your allocation to the beach?  This is true for all portfolios, but especially when specific class sectors are involved.  I'm guessing that no matter which Equity division produces technically superior outcomes (and one can apportion that 25% in many "logical" ways), the one that may be emotionally superior is TSM.  And for me, the biggest appeal for the PP proper lies there.

Roy

[MediumTex]

Are you ("you" meaning anyone) prepared to ride your allocation to the beach?  This is true for all portfolios, but especially when specific class sectors are involved.  I'm guessing that no matter which Equity division produces technically superior outcomes (and one can apportion that 25% in many "logical" ways), the one that may be emotionally superior is TSM.  And for me, the biggest appeal for the PP proper lies there.

Roy

Just to make this point explicit, I don't like getting upset, and losing money that I have already earned upsets me.  I like the PP because I know I am unlikely to get upset as a result of choosing this allocation.  Tracking error is, to me, an abstraction that is basically meaningless and tends not to upset me in the least (though I understand that it does upset some people).

As simple and obvious as these points sound, I think there are a lot more people like me out there who also don't like getting upset, but who have portfolios that are going to upset them if they stick with them long enough.

I think that deep down most investors are conservative--many of them just haven't figured it out yet, and the time they spend being upset about their investments slows down the process of discovering their own nature.

[hrux]

After much consideration, deliberation, research, etc.  in my personal opinion one should have a 50/50 US-International equity exposure as well as an equal mix of small to large cap.  The following is my suggestion:

For taxable accounts:
25% VV (US large blend)
25% IJS (US small value)
15% EFV (Int'l large value)
10% VWO (emerging)
25% VSS (international small to mid for both developed and emerging)