Total Stock vs. Small Cap Value/Emerging Split

[hrux]

In reading various posts it seems to be divided by several on what can produce the best risk adjusted return.  I would like to see Paul and Craig discuss this on a podcast.

Clive and Paul Boyer seem to favor the small cap value/emerging split.

Craif, Medium Tex and others seem to endorse not tinkering with thes tock portion and to keep in total stock market.

Which strategy will offer the maximum risk adjusted return?

[foglifter]

In reading various posts it seems to be divided by several on what can produce the best risk adjusted return.  I would like to see Paul and Craig discuss this on a podcast.

Clive and Paul Boyer seem to favor the small cap value/emerging split.

Craigr, Medium Tex and others seem to endorse not tinkering with the stock portion and to keep in total stock market.

Which strategy will offer the maximum risk adjusted return?

Short answer: nobody knows

Long answer: I'd rephrase the question as "Which strategy may offer the maximum risk adjusted return within the next XX years". This questions has been debated in a broader context in many threads on BH and you can count a small army of proponents facing another army of opponents. I do tilt into small value and EM small, but I understand the risks. I may one day say "craigr, you were right" after I see small stocks under-performing TSM. I think it maybe a safer approach to consider small cap and EM as part of one's variable portfolio.

[BobS]

Perhaps it would help to read - http://www.fundadvice.com/articles/buy- ... ategy.html

It discusses the things that can be done to decrease the risk in a portfolio.  I don't see why the equity side of that wouldn't
apply to the equity part of the PP.

Also, I found that - http://econ.duke.edu/Papers/PDF/Vanguar ... y_2007.pdf was useful
if Vanguard funds were to used for the equity portion.  A core set of funds that could implement the equity side
of the Ultimate Buy and Hold Strategy was described in the paper as -

  VISVX - 42%
  NAESX - 28%
  VTRIX - 14%
  VIVAX - 6%
  VPACX - 10%

But the above is just the core.  While it could be implemented with etfs, it would still be necessary to add other
funds as discussed in Ultimate Strategy.

[craigr]

I think simple is better. Less moving parts. Less chance to tinker and time components. Less chance to get a bug in your ear from some advisor about trading positions. Less taxes. Less expense ratio. Less regrets with respect to market tracking error.

I think even some index advisors advocating DIY slice and dice are coming to grips with the notion that most investors just can't do it well. The emotions (fear, greed, etc.) get in the way of making rational decisions vs. a simpler solution. There is much temptation to tinker, time and otherwise mess around with the portfolio and this hurts performance.

I think it is easier to track a big stock index and when it gets too big to sell it down and buy bonds, gold, cash vs. tracking a bunch of different stock sectors and deciding whether to let one ride, to rebalance between the stocks, etc. Not just this, but the simpler stock strategy is far cheaper to implement both in taxes and expense ratios and that's free money.

I say keep it simple. I think it's a good idea to recognize how well a simple solution works and not try to outsmart yourself.

[KevinW]

I am convinced that the Fama-French model is correct, and there are small-cap and value premia.  However I am not convinced that those premia necessarily manifest themselves on timeframes that are compatible with individual investors' time horizons.  In other words, I am confident that on time scales of multiple centuries, small-cap value will exhibit a better risk-adjusted reward than the total market.  However it is entirely possible that small-cap value can lag the market for decades, and that those decades could happen to coincide with an individual's entire investing career.

So I can't condone using small cap value as the sole domestic equity position in an individual's portfolio.  However it might be appropriate in some kind of Rip Van Winkle scenario where the assets will be held longer than a single lifetime, such as an endowment or intergenerational wealth.

As discussed elsewhere I think foreign stocks can augment other kinds of portfolios, but have no place in a permanent portfolio.  That goes for emerging market stocks.

Another issue is that the permanent portfolio depends on the stock allocation to carry the portfolio in times of prosperity.  It is well established that the total market rallies under prosperity, but there is no guarantee that small cap value or emerging market stocks will rally during domestic prosperity.  It is easy to imagine a boom that favors established multinational firms with good credit, coinciding with a strong domestic currency that suppresses foreign stocks.  The "Swedroe concentrated" portfolio doesn't really care when equity rallies happen as long as they do, but a PP without a strong prosperity-stock correlation may seriously underperform during prosperity.

[Roy]

I am convinced that the Fama-French model is correct, and there are small-cap and value premia.  However I am not convinced that those premia necessarily manifest themselves on timeframes that are compatible with individual investors' time horizons.  In other words, I am confident that on time scales of multiple centuries, small-cap value will exhibit a better risk-adjusted reward than the total market.  However it is entirely possible that small-cap value can lag the market for decades, and that those decades could happen to coincide with an individual's entire investing career.

So I can't condone using small cap value as the sole domestic equity position in an individual's portfolio.  However it might be appropriate in some kind of Rip Van Winkle scenario where the assets will be held longer than a single lifetime, such as an endowment or intergenerational wealth.

As discussed elsewhere I think foreign stocks can augment other kinds of portfolios, but have no place in a permanent portfolio.  That goes for emerging market stocks.

Another issue is that the permanent portfolio depends on the stock allocation to carry the portfolio in times of prosperity.  It is well established that the total market rallies under prosperity, but there is no guarantee that small cap value or emerging market stocks will rally during domestic prosperity.  It is easy to imagine a boom that favors established multinational firms with good credit, coinciding with a strong domestic currency that suppresses foreign stocks.  The "Swedroe concentrated" portfolio doesn't really care when equity rallies happen as long as they do, but a PP without a strong prosperity-stock correlation may seriously underperform during prosperity.

I understand not wanting to use SV exclusively.  Though, I believe SV has done better than large growth/blend over most rolling ten year periods, and that isn't a long time.  That is, it is a risk premium that gets rewarded fairly persistently.  But the threat is that if one gets a prolonged period of underperformance (of any sort), the investor jumps ship due to tracking error regret.

Also, Swedroe doesn't care if SV underperforms in positive returns years because he is looking to minimize losses, mainly.  In years that the market gets clobbered, most equities go down uniformly.  His only concern is if SV goes down while the overall market goes up--and that has not happened, I think;  that is, there were no 2 consecutive years where SV went down and TSM up.  Of course, when markets decline, the Swedroe portfolio is very likely to outperform conventional models due to its lower Beta exposure.  Still, it is tracking error regret of any sort that is likely the biggest threat when focusing on particular asset classes vice broad markets. 

In FAIL SAFE INVESTING, on my page 108, Browne says, when discussing the Stocks component, "you want funds that invest in volatile stocks in the hope they will move farther than the general stock market when times are good."  It is not unreasonable to interpret this as small cap/value, which has some component of being tied strongly to the economic cycle.  But in his recommended list, you get only Large Caps, though one Large Value fund.

[MediumTex]

Back in the "Nifty 50" days didn't you see large caps outperforming small caps for quite a while?

I would have wanted some of that Nifty 50 action if I had been wanting to benefit from stock market gains.

[craigr]

Back in the "Nifty 50" days didn't you see large caps outperforming small caps for quite a while?

I would have wanted some of that Nifty 50 action if I had been wanting to benefit from stock market gains.

They outperformed in the 1990s as well. It's a crap shoot. IMO. Any complexity you add to a portfolio adds a new decision point for an investor to second guess themselves. More stock funds means more chances to not rebalance due to various reasons. More chance to want to tinker. More chance to be influenced by various "experts". Etc. 

I would advise keeping things simple. You can always add complexity later to the variable portfolio if things aren't exciting enough. It's hard to take complexity out of a portfolio though once present.

[KevinW]

Though, I believe SV has done better than large growth/blend over most rolling ten year periods, and that isn't a long time.

Sure, but I don't know of any economic theory that says the SV premium must bloom within 10 years.  With investing I follow the scientific method which means I want to see a plausible hypothesis first, followed by supporting experimental evidence.  The problem with extrapolating from a backtest is that it's possible the dataset isn't exhaustive.

IMO small/value tracking is closely intertwined with government policy on regulation and contract enforcement.  Most backtest datasets cover a subset of the 20th century USA, which is a regulation/enforcement environment we're all familiar with, but I'm not confident patterns from that era necessarily generalize to other centuries or countries.

In FAIL SAFE INVESTING, on my page 108, Browne says, when discussing the Stocks component, "you want funds that invest in volatile stocks in the hope they will move farther than the general stock market when times are good."  It is not unreasonable to interpret this as small cap/value, which has some component of being tied strongly to the economic cycle.  But in his recommended list, you get only Large Caps, though one Large Value fund.

Some authorities say small/value has more upward volatility in prosperity, and some say growth does.  Both camps have a story that sounds plausible.  I believe early versions of the permanent portfolio held growth stocks, and the stock portion of PRPFX, PAGRX, is a growth fund.  TSM holds both and no one ever got fired for picking TSM.

Different strokes for different folks...

[craigr]

In FAIL SAFE INVESTING, on my page 108, Browne says, when discussing the Stocks component, "you want funds that invest in volatile stocks in the hope they will move farther than the general stock market when times are good."  It is not unreasonable to interpret this as small cap/value, which has some component of being tied strongly to the economic cycle.  But in his recommended list, you get only Large Caps, though one Large Value fund.

Here's the problem with this idea. And, I'd add, that his former partner John Chandler confirmed with me when I have spoken with him. That is you can't be sure a fund with high beta will remain high beta going forward. These are all backwards looking measures. When you own the market you do so with no regrets. You capture the market returns and all the sectors it represents. You never get caught on the downdraft of a sector going out of favor. So you avoid, again, second guessing and feeling like you need to tinker.

I think it is a salient point to say that towards the end of Browne's career he was recommending an index fund that focused on capturing maximum market returns (s&p 500) and not high beta funds.

[Roy]

Though, I believe SV has done better than large growth/blend over most rolling ten year periods, and that isn't a long time.

Sure, but I don't know of any economic theory that says the SV premium must bloom within 10 years.  With investing I follow the scientific method which means I want to see a plausible hypothesis first, followed by supporting experimental evidence.  The problem with extrapolating from a backtest is that it's possible the dataset isn't exhaustive.

First, there isn't any economic theory that says anything must happen, including any backward-inspired anticipations of how the PP will respond to future economic conditions. 

There are, of course, plausible economic reasons why risk premia—of all sorts—(including those contained in the PP) have higher expected returns than less risky securities.  The past has simply demonstrated those reasonings, and the longer the time frame the greater the demonstration power.  "Multiple centuries" have not been required to demonstrate any of this, in many instances, rolling ten-year periods have been good enough, even as it is obvious there is no requisite for this to continue—in reality or theory. But had these economic underpinnings not been demonstrated, there would be no risk premia.

[Roy]

In FAIL SAFE INVESTING, on my page 108, Browne says, when discussing the Stocks component, "you want funds that invest in volatile stocks in the hope they will move farther than the general stock market when times are good."  It is not unreasonable to interpret this as small cap/value, which has some component of being tied strongly to the economic cycle.  But in his recommended list, you get only Large Caps, though one Large Value fund.

Here's the problem with this idea. And, I'd add, that his former partner John Chandler confirmed with me when I have spoken with him. That is you can't be sure a fund with high beta will remain high beta going forward. These are all backwards looking measures. When you own the market you do so with no regrets. You capture the market returns and all the sectors it represents. You never get caught on the downdraft of a sector going out of favor. So you avoid, again, second guessing and feeling like you need to tinker.

I simply quoted a passage from HB's paperback version of FAIL-SAFE INVESTING (from 1999). That passage stated an intention to outperform the market:

"In addition, you want funds that invest in volatile stocks, in the hope they will move farther than the general stock market when times are good."

Even as his recommendations (in that edition) were all active (Large Cap) funds, and one Large Cap Value fund, I was curious what this modifying statement meant or how it could be interpreted.  That's all.  I wasn't suggesting that one must invest in Small Caps for the PP, but Small and Value have, over time, been the very securities that accomplished his additional objective in that passage. 

And what are the non/less-volatile stocks?  It is also clear, by extension, that if one could find "volatile stocks" that would outperform, one might also underperform "the market" using those same stocks.  Just wasn't sure what these were.

I have always agreed (and still believe) that TSM (or S&P 500) is the best option for the PP stock component (and though my edition is 1999, I was also unsure why an edit did not mention the Index funds).  But I have nothing really other than the fact that is has worked so well for so long (in combination with its other 3 asset classes) to support that belief.

[MediumTex]

And what are the non/less-volatile stocks?  It is also clear, by extension, that if one could find "volatile stocks" that would outperform, one might also underperform "the market" using those same stocks.  Just wasn't sure what these were.

Banks used to be considered low volatility stocks.  Now they are high volatility stocks.

Tobacco has its periods of low and high volatility, as do the drug makers.

The car companies?  Non-volatile some of the time, very volatile others.

How about old school technology like IBM?  Same story--volatility comes and goes.

Volatility visits every sector sooner or later.

[Roy]

Volatility visits every sector sooner or later.

Yes indeed.  And sooner rather than later, as magnificently demonstrated in flights. 

What were the volatile stocks you believe HB was referring to in his book (the ones that "move farther than the general stock market when times are good")?

[moda0306]

High vs. Low dividend yield maybe?

Growth Stocks?

[Roy]

High vs. Low dividend yield maybe?

Growth Stocks?

He does recommend some Growth funds.  And a Large Value fund. 

I suppose were he to edit it today, it would likely be the TSM, as Craig says. This has combined well with its asset classes over these 40 years, and the PP is dominated by non-stock classes anyway.  Though, I imagine even particular asset classes would have done well also, maybe better (I seem to recall Clive or Trev playing with equity asset classes for the stock portion over the same period).  But TSM is inexpensive, worked just fine, and to spectacular portfolio outcomes at times (like these past 3 years).

[Gumby]

The problem with using Emerging Markets in a Permanent Portfolio is that it's simply performance chasing, and nothing more than adding speculation (on foreign markets and currencies) into an otherwise neutral portfolio. That would easily break the warranty on a Permanent Portfolio.

In other words, the reason why HB never considered Emerging Markets is because Emerging Markets are not directly tied to "Prosperity" — one of the four pillars of the PP. Emerging Markets did their poorest from 1995 to 2000 — during what was perhaps the greatest 6-year stretch of prosperity, ever.



A PP that uses Emerging Markets might do very well over a 40 year period, but it may also very well be left behind during actual times of prosperity — which would be quite painful.

[moda0306]

Great point on 1995-2000 Gumby.  I didn't realize that.  Thanks for the info.

[Gumby]

The problem with using Emerging Markets in a Permanent Portfolio is that it's simply performance chasing, and nothing more than adding speculation (on foreign markets and currencies) into an otherwise neutral portfolio. That would easily break the warranty on a Permanent Portfolio.

In other words, the reason why HB never considered Emerging Markets is because Emerging Markets are not directly tied to "Prosperity" — one of the four pillars of the PP. Emerging Markets did their poorest from 1995 to 2000 — during what was perhaps the greatest 6-year stretch of prosperity, ever.
....snip

The performance of EM's over that period is entirely as you might expect.  Being foreign currency and somewhat commodity driven they're more like gold than stocks.  In times of USD crisis EM's might be expected to do relatively well for the offshore currency and commodity like qualities as can be seen from more recent EM performance figures.

Blended with perhaps SCV and you have a sort of 'neutral' holding or a mini-PP with SCV acting as the 'stocks' and EM as the 'gold'.

For a fair comparison you should really compare between two periods when both assets had achieved the same overall gain (or loss), unless that is you expect one of the assets to never re-align with the other in which case you'd just hold the better choice.

Given two assets that achieve the same gain between two points in time (such as over a period when they both achieved the same rewards), then holding equal amounts of both and periodically rebalancing back to equal weightings when deviations became apparent, has you add-low/reduce-high and will produce higher rewards overall than having held one or the other alone, and also likely do so with less risk (assuming volatility or standard deviation is considered as a measure of risk).  Having two assets that individually have relatively high volatility, but zig and zag in somewhat opposite directions has the effect of reducing the overall combined volatility, but you're rewarded on an individual 'higher risk' basis (average of the two gains).

That's all well and good, and I'm sure it would do well over the long run, but my point is that EM/SCV clearly doesn't represent Prosperity — which is likely why Harry Browne never considered it as a legitimate choice for the PP.

By replacing your Prosperity pillar of the Permanent Portfolio with a "mini-PP" you would not have done very well during the great prosperity of 1995 to 2000. That's the undeniable consequence.

I've been told that the most challenging part of holding the Permanent Portfolio is that you may be tempted to ditch the strategy during times of prosperity to chase higher returns. So, imagine holding the EM/SCV split in your PP from 1995 to 2000 and underperforming a traditional Permanent Portfolio when everyone else is making enormous amounts of money. It would be nearly impossible to stay the course. Not to mention, you would not be given the proper opportunity to re-balance large profits into key assets like Gold and Bonds in time for an oncoming recession.

[moda0306]

Since we have no decent TSM index in my 401(k), I've been using a lot of Emerging Markets thinking it's just "TSM on steroids," and it was basically just a more risky set of stocks but will make the same movement directions.

I am glad I now know that's not the case.

[Gumby]

Hi Gumby

By replacing your Prosperity pillar of the Permanent Portfolio with a "mini-PP" you would not have done very well during the great prosperity of 1995 to 2000. That's the undeniable consequence

As with the PP you have to consider the collective rather than the individuals alone

Small (Growth) and Value - combined SCV, might be considered as a combination of stocks (small) and LT's (value).  Each can do OK alone, as a pair they tend to have a degree of low correlation

EM's might be considered a form of gold play.

So a blend of 50% SCV, 25% EM and 25% 5 year treasury (as a 'cash' proxy)

...

produced some reasonable gains over the 1995 to 2000 period.

The ride was quite bumpy however, so an option might be to water that down by halving the amount of exposure in SCV and EM's to

...

Or perhaps even water down to a third and :

...

The water down to a third did relatively OK, but did have some real (after inflation) volatility during the 1970's which could have been reduced by taking some of the cash and investing in gold for a bit more high inflation/domestic currency crisis protection

...

That overall provided somewhat similar risk (few down years and down only a little in those years) and relatively consistent real gains over time.

I've been told that the most challenging part of holding the Permanent Portfolio is that you may be tempted to ditch the strategy during times of prosperity to chase higher returns. So, imagine holding the EM/SCV split in your PP from 1995 to 2000 and underperforming a traditional Permanent Portfolio when everyone else is making enormous amounts of money

Compared to the Coffee House (60/40 stock/bond type blend), having included SCV and EM's to me looks potentially less likely to have been capitulated than the PP, including over the 1995 to 2000 period.

You'll have to excuse me, but I can't follow any of that.

When you start including a handful of lengthy charts and using a dozen funds — including REITs and European stocks — in an allocation that's supposed to simply be SCV and EMs, my eyes glaze over. I was under the impression that this thread was simply about replacing the PP's stock allocation with a SCV fund and a EM fund, no? That's what I meant when I referred to your "mini-PP" analogy.

Put the charts away for a moment. If the Permanent Portfolio works because there are four assets where at least one will perform well during each of the four stages of the economy, can you explain — using language alone — how replacing the PP's stock allocation with a simple SCV/EM split will save my hide during times of Prosperity?

I don't see how you can when you look at 1995-2000.

[Gumby]

can you explain — using language alone — how replacing the PP's stock allocation with a simple SCV/EM split will save my hide during times of Prosperity?

No

I think what you've found is a nice diversified variable portfolio that has worked very well over long periods of time. I was under the impression that the SCV/EM split was an even 50/50, but if you've chosen to diminish the weighting of the EM down to 8.4%, I suppose that will improve your numbers for that period of prosperity. No one is doubting that it has done well over the long haul. And I'm sure it will do fine.

The reason I chose 1995-2000 is because it was a prosperity bubble. It was the run-up to the S&P 500 index reaching an all-time intraday high of 1,552.87 in trading on March 24, 2000. If that's not a perfect example of a period of prosperity, I don't know what is. Stocks represent prosperity within the PP, and therefore, the all-time high of the S&P 500 is easily explained as a product of that period. EMs did not do well during that period — which makes one wonder why it should be a part of a powerful prosperity allocation.

Which, again, brings me back to my point. You are so far unable to explain — in simple terms — how a PP with a SCV/EM stock allocation should perform during a period of prosperity. It just seems a bit irresponsible to recommend a rebalanced PP strategy to others if you can't easily explain how it will work in every economic environment. That, I feel, is the reason why HB never suggested it.

If your portfolio's prosperity pillar can't be explained, or doesn't exist, that's fine. To each his own. We don't need to keep going back and forth on this ad nauseum.  :)

[Lone Wolf]

In other words, the reason why HB never considered Emerging Markets is because Emerging Markets are not directly tied to "Prosperity" — one of the four pillars of the PP. Emerging Markets did their poorest from 1995 to 2000 — during what was perhaps the greatest 6-year stretch of prosperity, ever.

This is a very important point.  I had not realized what happened in 1995-2000 in EM.  If your stock allocation doesn't perform well during prosperity, the portfolio simply won't act like a "Permanent Portfolio" in the way that the 4x25 is designed.  The 4x25 has good, steady returns with low volatility.

Furthermore, there is an economic rationale for why 4x25 performs this way in all known economic conditions.  While I think that there is an excellent rationale for why SCV will perform well in times of prosperity (although I personally prefer "total stock market"), this rationale simply does not exist for EM.  EM might do "fine" in times of domestic prosperity but there's certainly no guarantee that this would be the case.  With gold probably doing terribly during those prosperous times, weak performance on the part of EM is likely to mean a much more volatile portfolio.

EM just doesn't seem like it fits in a "Permanent Portfolio", particularly for an investor that puts much emphasis on volatility.

[Pkg Man]

I initially had a small allocation of EM in the stock portion of the PP but based on feedback here decided to just use VTI, the S&P500, and a bit of a mid-cap index fund (the later two are in my 401K as there is no direct option for the total market; most of the stock portion is in VTI).

I like EM a lot and hold some individual stocks from EM countries, but will invest only only VP funds there from now on.

[Gumby]

Which, again, brings me back to my point. You are so far unable to explain — in simple terms —

In simple terms



For the explanation read Fama and French

For the record, what I actually said was...

Which, again, brings me back to my point. You are so far unable to explain — in simple terms — how a PP with a SCV/EM stock allocation should perform during a period of prosperity.

Prosperity, prosperity, prosperity, prosperity....prosperity. It would seem I can't repeat that enough.

The formula doesn't have anything to do with ensuring I'll have the most powerful stock allocation in my PP during times of "prosperity".