How Can "Smart Beta" Go Horribly Wrong?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
Beliavsky
Posts: 933
Joined: Sun Jun 29, 2014 10:21 am

How Can "Smart Beta" Go Horribly Wrong?

Post by Beliavsky » Sat Feb 13, 2016 12:18 pm

How Can "Smart Beta" Go Horribly Wrong?
by Rob Arnott, Vitali Kalesnik, John West, and Noah Beck
Research Affiliates
February 2016
Because active equity management has largely failed to deliver on investors’ expectations, investors have acquired a notable appetite for any ideas that seem likely to boost returns. In this environment, impressive past results for so-called smart beta strategies, even if only on paper, are attracting enormous inflows. Investors often choose these strategies, as they previously chose their active managers, based on recent performance. If the strong performance comes from structural alpha, terrific! If the performance is due to the strategy becoming more and more expensive relative to the market, watch out!

Performance chasing, the root cause of many investors’ travails, has three inextricably linked components. Rising valuation levels of a stock, sector, asset class, or strategy inflate past performance and create an illusion of superiority. At the same time, rising valuations reduce the future return prospects of that stock, sector, asset class, or strategy, even if the new valuation levels hold. Finally, the higher valuations create an added risk of mean reversion to historical valuation norms.

Many of the most popular new factors and strategies have succeeded solely because they have become more and more expensive. Is the financial engineering community at risk of encouraging performance chasing, under the rubric of smart beta? If so, then smart beta is, well, not very smart.

Are we being alarmist? We don’t believe so. If anything, we think it’s reasonably likely a smart beta crash will be a consequence of the soaring popularity of factor-tilt strategies. This provocative statement—especially by one of the original smart beta practitioners—requires careful documentation. In this article we examine the impact of rising valuations on many popular smart beta categories.
According to Panel B of this paper, the 10-year annualized excess return of low volatility through Sep 30, 2015 (2015Q3) was 0.82%. According to S&P http://us.spindices.com/indices/strateg ... lity-index , the 10-year return of the S&P 500 Low Volatility and "regular" indices through Feb 12, 2016 was 8.91% vs. 6.18%, a much larger difference of 2.73%. I wonder if the difference is due just to the different time periods covered or if different definitions of low volatility are being used.

User avatar
Robert T
Posts: 2610
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Robert T » Sat Feb 13, 2016 1:00 pm

.
Great article. Some extracts.
"... the value effect is far from dead! In fact, it’s in its cheapest decile in history."

"Large asset flows into low beta products are now driving valuation levels far above their historical norms. Low beta’s end-point in relative valuation is at an all-time peak..."

"In the last decade, quality generated the best performance of all smart betas at 2.37% a year, helped by rising valuations. This result is a backtest, and the best way to validate a backtest is to use out-of-sample data. During the 39-year period 1967–2005, quality delivered a −0.14% annualized return. Not surprisingly, in the longer 49-year sample, quality had the worst performance of all smart betas at 0.37% a year. Interestingly, unlike for the gross profitability factor for which close to 100% of both 10-year and full-sample returns is attributable to rising valuations, for quality the 0.37% a year return did not come from rising valuations. This highlights another problem with quality investing: quality portfolios can be quite sensitive to the definition being used. A portfolio formed on gross profitability can be very different from a portfolio formed on quality, which is based on profitability, leverage, and earnings volatility."
.

Hank Moody
Posts: 271
Joined: Fri Oct 17, 2014 9:41 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Hank Moody » Sun Feb 14, 2016 2:50 pm

I'm curious what brighter minds think of Arnott's conclusions about Profitability and Momentum. I've been exploring AQR's research on these factors, and I'm wondering how to put these two seemingly disparate thoughts together.

On Gross Profitability, he writes:

Over the last 10 years, of the six factors gross profitability (the quality definition most popular in current academic research) had the best performance, while value had the worst. We also find that essentially all of the outperformance for profitability is due to rising valuations. When we subtract the returns associated with the rising popularity, and therefore rising relative valuation, of high-profit companies versus low-profit companies, the gross profitability factor loses more than 90% of its historical efficacy, delivering 10-year performance net of valuation change of just 0.39%. The more conservative regression-based performance estimate trims excess return by two-thirds, from 4.54% to a much less spectacular 1.57%.
On Momentum, he writes:
Rising valuations also provided a substantial tailwind for momentum and illiquidity over the last decade. This means gross profitability, momentum, and illiquidity are all considerably more expensive today than they were ten years ago. Over this same period, we see value has floundered because it was out of favor and becoming ever cheaper!
Anyone?
-HM

lack_ey
Posts: 6701
Joined: Wed Nov 19, 2014 11:55 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by lack_ey » Sun Feb 14, 2016 3:10 pm

The linkage between a non-market factor return or spread and subsequent performance for the factor is real but may not be as strong as you might expect. That's because the groups of stocks change over time. It's possible, for example, for say "quality" stocks to be expensive now and become less expensive relative to "junk" while simultaneously outperforming junk, in part from the constituents in the portfolios changing.

But sure, if part of the performance differential over a period is a result of a relative change in valuations, that is probably not a great sign for the future.

You should always expect the effects of most factors to be overstated for a variety of reasons.

On momentum I'm pretty sure AQR would point to very long datasets for momentum and evidence across multiple markets and asset classes. About profitability, I'm not sure.

Disclaimer: I haven't done more than barely skim RA's work here (planning to read later), and if you're looking for brighter minds on the subject, you'll certainly have to keep looking past me. :wink:

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Sun Feb 14, 2016 3:15 pm

Hank
FWIW
To me there is no good risk story for profitability/quality, although I can create one or two (more profitable companies tend to be growth companies and more of their cash flow is thus in the longer term future and thus subject to more risk-higher discount rate-- or they attract more competition seeking that high profitability). Now we know the very low profitability companies can be are those lottery tickets, leading to some of the premium. So I'm not as much a fan of this factor. And simply by using p/e and/or p/cf when creating value portfolios you gain access to profitability/quality indirectly while IMO improving value factor returns (diversification benefits).

Re momentum, no good risk story at all, just crash risk in reversals but that doesn't justify a premium IMO, but there are so many studies showing MOM exists everywhere and across all asset classes that IMO it's more prudent to take it into account. Note that there is good support in the data for both cross-sectional MOM (relative MOM) and time series MOM (trend following or absolute MOM). As to whether the size of the MOM premium will continue seems logical that it has shrunk given the amount of assets chasing it. But no evidence IMO that it's gone.

Larry

Hank Moody
Posts: 271
Joined: Fri Oct 17, 2014 9:41 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Hank Moody » Sun Feb 14, 2016 3:21 pm

Can you discuss BAM's use of AQR Large US Multi-Style as a substitute for DFA Large Value or one of the DFA Core funds? It seems the one attribute of AQR over DFA in general is its strongly belief in the MOM factor.
larryswedroe wrote:Hank
FWIW
To me there is no good risk story for profitability/quality, although I can create one or two (more profitable companies tend to be growth companies and more of their cash flow is thus in the longer term future and thus subject to more risk-higher discount rate-- or they attract more competition seeking that high profitability). Now we know the very low profitability companies can be are those lottery tickets, leading to some of the premium. So I'm not as much a fan of this factor. And simply by using p/e and/or p/cf when creating value portfolios you gain access to profitability/quality indirectly while IMO improving value factor returns (diversification benefits).

Re momentum, no good risk story at all, just crash risk in reversals but that doesn't justify a premium IMO, but there are so many studies showing MOM exists everywhere and across all asset classes that IMO it's more prudent to take it into account. Note that there is good support in the data for both cross-sectional MOM (relative MOM) and time series MOM (trend following or absolute MOM). As to whether the size of the MOM premium will continue seems logical that it has shrunk given the amount of assets chasing it. But no evidence IMO that it's gone.

Larry
-HM

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Sun Feb 14, 2016 3:27 pm

Hank
couple more things.
First, as the paper notes the momentum factor has very high turnover and thus the stocks within it are often changing in type and thus changing valuations are very different than for factors such as value or size.

Second, we chose to use AQR multi-style for US large value (and looking at it for other asset classes) as we want to diversify further the SOURCES of returns away from just beta, size and value and AQR targets THREE factors for positive loadings DIRECTCLY (value, MOM, and profitability with the first two at 40% and MOM at 20%). So that moves the momentum loading from say 0 (for DFA value funds, used to be highly negative but they now screen out negative momentum stocks so that gets them to about zero) to a positive figure, diversifying sources of returns, and also reduces tracking error, making the portfolio more "market like" as MOM tends to be growthier. So that helps with investors subject to that dread disease (and provides diversification benefit).

Hope that helps
Larry

Hank Moody
Posts: 271
Joined: Fri Oct 17, 2014 9:41 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Hank Moody » Sun Feb 14, 2016 3:34 pm

It definitely helps, especially the comment of having more sources of return.

Why are you subbing in AQR M/S for DFA Value instead of Core? When Momentum is introduced, doesn't it counteract some of the risk of Value already?

Also, how do you feel about using AQR M/S in taxable accounts given the higher turnover compared to DFA?
larryswedroe wrote:Hank
couple more things.
First, as the paper notes the momentum factor has very high turnover and thus the stocks within it are often changing in type and thus changing valuations are very different than for factors such as value or size.

Second, we chose to use AQR multi-style for US large value (and looking at it for other asset classes) as we want to diversify further the SOURCES of returns away from just beta, size and value and AQR targets THREE factors for positive loadings DIRECTCLY (value, MOM, and profitability with the first two at 40% and MOM at 20%). So that moves the momentum loading from say 0 (for DFA value funds, used to be highly negative but they now screen out negative momentum stocks so that gets them to about zero) to a positive figure, diversifying sources of returns, and also reduces tracking error, making the portfolio more "market like" as MOM tends to be growthier. So that helps with investors subject to that dread disease (and provides diversification benefit).

Hope that helps
Larry
-HM

User avatar
JoMoney
Posts: 8211
Joined: Tue Jul 23, 2013 5:31 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by JoMoney » Sun Feb 14, 2016 3:49 pm

Hank Moody wrote:... Anyone?
These guys shy away from making analysis of individual businesses, but then make sweeping claims with regard to what the future "value" of trends in active trading styles should be?
Seems like just more rhetoric from someone who has an opinion and a product to sell.

Changes in the demand for sock garters or butter over margarine may be bucking the long term historical trend, but it doesn't mean i can forecast mean-reversion within any particular timeline... but I could be wrong, he is a professional in the field of marketing active mutual funds and I'm not. Maybe he knows something, but I'm sure there's professionals with an opposing view as well.
Benjamin Graham wrote:...The stock market resembles a huge laundry in which institutions take in large blocks of each other's washing...without true rhyme or reason.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Sun Feb 14, 2016 4:36 pm

Hank, we are CURRENTLY using on in tax advantaged accounts but considering for taxable as well.
And we are working with the firm to develop a core type fund with the tilt we prefer for most clients. So currently we use DFA core and then add small and large value to tailor to individuals who want/desire/need more tilt.
Larry

User avatar
Robert T
Posts: 2610
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Robert T » Mon Feb 15, 2016 6:03 am

.
On of the most interesting findings in the paper is the following

1967-2015: Annualized factor returns (%), net of valuation changes

3.12 = Value
3.83 = Momentum
1.86 = Small cap
1.89 = Illiquidity
0.00 = Low beta
0.05 = Gross profitability

0.00% and 0.05% for low beta and gross profitability respectively.
Net of rising valuation, the value added by low beta disappears entirely or is reduced by a third, considering the more conservative regression-adjusted estimate; the same holds true for gross profitability. Should investors really expect to be rewarded for profitability or quality? Shouldn’t we accept a lower return for safer assets? Some of the strategies, for example, low beta, may still be an attractive investment, but for their risk-reducing characteristics not for the alpha they have historically provided, net of their rising popularity and relative valuation. These data suggest that common sense prevails: lower risk, higher quality, and safety have all earned a strong premium only as a consequence of becoming more expensive! Will this upward adjustment in relative valuation prove permanent? Figure 2 suggests otherwise. But even if it is, we cannot rely on rising valuations to continue to create a continuing illusion of alpha.
I think we need to distinguish between changing valuations of stocks within a strategy, and changing valuations of the aggregate strategy.

(i) on changing valuations of stocks within a strategy: In one sense, all the value premium is from changes in stock valuations. For example the entire FF US small cap value return in excess of the market between 1926 and 2005 was from just over 20 percent of SV stocks who’s valuations increased and they migrated to small growth (neutral), or large caps - http://www.jasonhsu.org/uploads/1/0/0/7 ... ration.pdf .

(ii) on changes in aggregate valuations of a strategy. While the pattern is clearer for say value. E.g. over the last 10 years value has become cheaper relative to growth, as growth a recently outperformed. You can replicate a similar pattern from Ken French’s database for LG relative to LV and SG relative to SV. For momentum the message is less clear – figure 2 seems to suggest that relative performance is only very partially impacted by relative valuation (unlike growth and value, where they are defined by valuations, as with small caps).

In addition, as the article looked at factors independently, its more challenging to draw conclusions on multi-factor strategies such as the AQR multi-style funds. M* indicates that QCELX (the US version of AQR multi-style fund) has p/b of 1.6, compared to the S&P500 of 2.4, and not too far off the Russell 1000 Value p/b of 1.6.

Anyway, we may see a response to the article coming from Asness (as he did with Fama)…

Robert
.

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Mon Feb 15, 2016 8:50 am

Robert
I know AQR is working on papers related to the subject and already has one out on defensive style.
Larry

Hank Moody
Posts: 271
Joined: Fri Oct 17, 2014 9:41 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Hank Moody » Mon Feb 15, 2016 11:21 am

Larry, why does an allocation to AQR reduce DFA Large Value instead of DFA Core?
larryswedroe wrote:Hank, we are CURRENTLY using on in tax advantaged accounts but considering for taxable as well.
And we are working with the firm to develop a core type fund with the tilt we prefer for most clients. So currently we use DFA core and then add small and large value to tailor to individuals who want/desire/need more tilt.
Larry
-HM

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Mon Feb 15, 2016 12:00 pm

Hank
Because currently only using their large fund and they don't have a "core" mutli-style premium fund, just large and small. We want to use cores for the main portion and then add bit of other funds to get the desired tilts.
larry

stlutz
Posts: 5514
Joined: Fri Jan 02, 2009 1:08 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by stlutz » Mon Feb 15, 2016 12:31 pm

Thanks for sharing the article. It's a very good one. One nit I would pick:
... the value effect is far from dead! In fact, it’s in its cheapest decile in history.
It's probably worth noting that this is true if you are looking at price-to-book but not true if you are looking at price-to-earnings.

The price-to-forward earnings of the Russell 3000 (all cap) value index is 15.24 while it's 18.26 for the R-3000 Growth--not that big of a disparity at all.

Using P/E, one might conclude the value is just as "overpriced" as all of the "factors" currently.

With the exception of momentum and (arguably) low volatility, all of these factors have differences in how one defines them. In most cases those differences even out over decades of time, but they do create challenges if one is trying to market time factors.

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Mon Feb 15, 2016 1:31 pm

stlutz
The problem with comparing p/es today with p/es prior to 2001 when GAAP changed rules on requirement immediate write down of impairment of intangible assets such as goodwill (and of course then book values as well) is that while the new rules are clearly superior they overstate the current valuations relative to past ones. In case of P/Es if you used the pre 2001 rules today's p/es would be about 4 lower. Don't know impact on p/b

Larry

Hank Moody
Posts: 271
Joined: Fri Oct 17, 2014 9:41 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Hank Moody » Mon Feb 15, 2016 3:20 pm

In other words, you don't want to give up the broader, market-wide diversification that DFA Core offers?
larryswedroe wrote:Hank
Because currently only using their large fund and they don't have a "core" mutli-style premium fund, just large and small. We want to use cores for the main portion and then add bit of other funds to get the desired tilts.
larry
-HM

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Mon Feb 15, 2016 4:29 pm

Hank
Not exactly

Here's simple example. Vanguard copied DFA's action recently and combined EM and developed into a total market or core fund. It makes much more sense to do that then to own components in the same percentages. The reasons are lower transactions costs as stocks/countries migrate from one asset class to another and that also for taxable investors improves tax efficiency. And for taxable investors the biggest benefit may be that you get to rebalance with other people's money, not yours as the fund rebalances with cash flows.

So let's say we want to tilt to size and value more than does the DFA core 2 portfolio. So you don't go and build the components individually. Instead you own the core 2 for MOST of the portfolio and then add on enough of the other asset classes (factors) to get the desired tilt.

Similarly it makes more sense to own one mutli-style fund that gives you exposure to all the factors in one fund than to own say three funds with each one giving you exposure to one of the three factors.

I hope that is helpful
Larry

User avatar
Robert T
Posts: 2610
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Robert T » Sat Feb 20, 2016 7:40 am

.
http://www.researchaffiliates.com/Pages ... videoID=34

Interesting to note final sentence by Arnott.

User avatar
nisiprius
Advisory Board
Posts: 39790
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 Can "Smart Beta" Go Horribly Wrong?

Post by nisiprius » Sat Feb 20, 2016 10:51 am

So I skimmed the article.

Question #1: One of the phrases that does not appear in the article is "fundamental index" or "RAFI fundamental index." So what about it? Are these, or are they not "smart beta" strategies? Are they exempt from the article's observation or not, and, if not, why not?

Question #2: If we accept everything in the article, is the word "horribly" justified? As Taylor says there are "many roads to Dublin." Just how "horrible" is horrible? How much is that in dollars?

To me, examples of "gone horribly wrong" would include:

a) People who bought into Bill Miller's Legg Mason Value Trust during the years when his 15-year streak were becoming widely known, and failed to get the hunch that told them to sell before 2007.

b) People who bought the innovative new OIL or USO ETFs during 2009-2011, because they were sure the price of oil would go up, which it did, but their ETFs didn't.

c) People who bought into the 130/30 funds back when they were confidently expected to supplant traditional long-only funds and become a $2 trillion category.

What's so "horrible" about smart beta? If it's just another way to charge 0.50%-1.00% expense ratios for something that investors expect to beat the index, and instead it sometimes does noticeably worse--but still builds your retirement portfolio--shrug. All kinds of minor cheats and scams will always be with us.
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
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by in_reality » Sat Feb 20, 2016 11:48 am

nisiprius wrote:So I skimmed the article.

Question #1: One of the phrases that does not appear in the article is "fundamental index" or "RAFI fundamental index." So what about it? Are these, or are they not "smart beta" strategies? Are they exempt from the article's observation or not, and, if not, why not?
Yes, they are but no I don't consider them to be "smart beta" strategies in the same way that equal weighting, low volatility or momentum are. Neither does Arnott who both penned the article and created fundamental indexes.

If stocks in fundamental indexes get high valuations, you will be rotated out of them and into something with more favorable valuations (as they measure them). This is because they are choosing on low valuations (relative to a 5 year measure of economic footprint). This worked fine in the '08/'09 financial crisis when banks got really cheap and then recovered. Let's see if the same goes for oil now

I really believe they can out perform because well let's face it, investing in banks in 08/09 and oil now is, how do we say, "taking on more risk" isn't it? There's more room for upside but no guarantee of things going up and prices have fallen for some reason.

So no by definition, I don't see the fundamental indexes having the problem of a crowded trade. They have the problems of loading up on what appears to be cheap based on the past 5 years but actually deserve their pricing because the company is going bad, higher trading costs, and higher ER.

I use them as I think they counter cap weighting. Are they really worth the price or is just the cheapest cap weighting the best? I can't really say.
Last edited by in_reality on Sat Feb 20, 2016 11:55 am, edited 1 time in total.

User avatar
nisiprius
Advisory Board
Posts: 39790
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 Can "Smart Beta" Go Horribly Wrong?

Post by nisiprius » Sat Feb 20, 2016 11:49 am

Well, then, what is the definition of "smart beta?"
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
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by in_reality » Sat Feb 20, 2016 11:54 am

nisiprius wrote:Well, then, what is the definition of "smart beta?"
Smart beta defines a set of investment strategies that emphasize the use of alternative index construction rules to traditional market capitalization based indices.

Read more: Smart Beta Definition | Investopedia http://www.investopedia.com/terms/s/sma ... z40jCGwH45

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Sat Feb 20, 2016 1:39 pm

in reality
I disagree with that definition as it is nothing more than different betas, with nothing NECESSARILY smart about it.
What's smart beta are things like patient trading vs. forced trading to replicate an index, screening out stocks that have historically had poor returns like penny stocks, stocks in bankruptcy, small growth with low profitability and high investment. That's "smart" beta. The rest is just beta, beta (loading or exposure) on factors other than market beta

User avatar
nisiprius
Advisory Board
Posts: 39790
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 Can "Smart Beta" Go Horribly Wrong?

Post by nisiprius » Sat Feb 20, 2016 4:35 pm

in_reality wrote:
nisiprius wrote:Well, then, what is the definition of "smart beta?"
Smart beta defines a set of investment strategies that emphasize the use of alternative index construction rules to traditional market capitalization based indices.--Investopedia
With the proviso that a) Larry Swedroe disagrees with that definition, and b) it's IMHO an intentionally meaningless term that's intended to obfuscate and persuade, not inform... if we accept Investopedia's definition, or the Financial Times' definition
...an umbrella term for rules based investment strategies that do not use the conventional market capitalisation weights that have been criticised for delivering sub-optimal returns by overweighting overvalued stocks and, conversely, underweighting undervalued ones.

Smart beta strategies attempt to deliver a better risk and return trade-off than conventional market cap weighted indices by using alternative weighting schemes based on measures such as volatility or dividends.

Smart beta refers to an investment style where the manager passively follows an index designed to take advantage of perceived systematic biases or inefficiencies in the market....
isn't that an exact description of the RAFI and other fundamental indexes?
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
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by in_reality » Sat Feb 20, 2016 7:23 pm

larryswedroe wrote:in reality
I disagree with that definition as it is nothing more than different betas, with nothing NECESSARILY smart about it.
What's smart beta are things like patient trading vs. forced trading to replicate an index, screening out stocks that have historically had poor returns like penny stocks, stocks in bankruptcy, small growth with low profitability and high investment. That's "smart" beta. The rest is just beta, beta (loading or exposure) on factors other than market beta
Well obviously the term "smart" beta is stupid because it implies something successful (no guarantees in investing though). I'd prefer "strategic beta" because strategies are attempts which can fail or succeed which deviating from market cap weighting is. It's more descriptive.
...screening out stocks...like small growth with low profitability and high investment.
That's your opinion (and likely correct) of what a good "smart beta" strategy would be. However, a fund like FYX (a) which tilts small growth is considered to be under the smart beta umbrella or "alternative indices".

So in the end, there are good smart beta strategies and poor smart beta strategies and the label doesn't tell you which one, and there is no way to know which is really which without understanding the strategy. Alls you know with smart beta is that the costs will be higher and there is a strategy that needs evaluating. It's dangerous I think to equate smart beta with "good" strategies. That was the point of the article actually.

But you win Larry by knowing more about investing that I do. I concede even if I am right because I don't want you to pull out a 250 page AQR research paper on me.... :sharebeer

(a) http://www.etftrends.com/2015/01/smart- ... utperform/

User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by in_reality » Sat Feb 20, 2016 7:26 pm

nisiprius wrote: it's IMHO an intentionally meaningless term that's intended to obfuscate and persuade, not inform...
I completely agree. It tells you the fund is not market cap weighted and that is it. It could be a great strategy, a horrible strategy or anything in-between. I don't think it's even able to tell if a fund in attempting to do factor investing or not. The term is really useless beyond identifying the fund is not market cap weighted. That's good info to have though.
nisiprius wrote:
Smart beta refers to an investment style where the manager passively follows an index designed to take advantage of perceived systematic biases or inefficiencies in the market....
isn't that an exact description of the RAFI and other fundamental indexes?
Yes, Fundamental Indexes are "Smart Beta" i.e. non-purely-cap-weighted strategy-following indexes.

That doesn't mean though that Fundamental Indexes have the same flaws as all other smart beta strategies. Different strategies have different flaws. Buying overpriced stocks in a crowded trade isn't something the Fundamental Index is going to do [unless you consider all US stocks to be overpriced and so even a Fundamental Index would be purchasing too high -- but then again at least it isn't purchasing the most of the highest priced stocks which market capitalization (for good or bad) would have you do].

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Sun Feb 21, 2016 9:50 am

in reality
Yes the problem is that there are smart beta strategies and dumb beta strategies (like the R2K was a really dumb beta strategy for loading on small cap factor). So the term as used is dumb, meaningless, because it covers any non market cap weighted strategies. But yes IMO there are good ways to implement beta strategies (so there are smart beta strategies) and bad ones.

Leesbro63
Posts: 6094
Joined: Mon Nov 08, 2010 4:36 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Leesbro63 » Sun Feb 21, 2016 11:46 am

There is mention of P/Es today versus in yesteryear before an accounting change. Can someone please expand on that?

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Sun Feb 21, 2016 12:05 pm

Leesbro
Prior to 2001 companies wrote off intangible assets (such as goodwill from acquisitions) over 40 years. FAS 142 and 144 changed that to require annual impairment tests and immediate writedown. Obviously superior accounting, but that makes comparing earnings from one regime to another very much apples to dates, not apples to oranges. One study found that if we were still operating under the old rules the PEs would be about 4 lower. So instead of CAPE 10 of about 24 would be about 20.

Larry

User avatar
nisiprius
Advisory Board
Posts: 39790
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 Can "Smart Beta" Go Horribly Wrong?

Post by nisiprius » Sun Feb 21, 2016 12:20 pm

in_reality wrote:...I completely agree. It tells you the fund is not market cap weighted and that is it. It could be a great strategy, a horrible strategy or anything in-between. I don't think it's even able to tell if a fund in attempting to do factor investing or not. The term is really useless beyond identifying the fund is not market cap weighted. That's good info to have though...
Another detail that is very hard to pin down, but... at least by the time the ideas get filtered through a couple of people... I have the impression that some people are really, truly saying that cap-weighting is actually the worst of all possible weightings and that all other weightings are better (thus "smart.")

They're not quite that categorical, they usually say that "just about any" weighting is better, leaving it unclear how you know which are "just about any" weightings and which are... not just any weightings. But the impression is certainly that any old competent, workmanlike, sensible departure from cap weighting will be better, and therefore that, as a class, funds that are self-described as "smart beta" ought to be better than cap-weighted funds.

That's testable, of course.
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
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by in_reality » Sun Feb 21, 2016 6:13 pm

nisiprius wrote:Another detail that is very hard to pin down, but... at least by the time the ideas get filtered through a couple of people... I have the impression that some people are really, truly saying that cap-weighting is actually the worst of all possible weightings and that all other weightings are better (thus "smart.")

They're not quite that categorical, they usually say that "just about any" weighting is better, leaving it unclear how you know which are "just about any" weightings and which are... not just any weightings. But the impression is certainly that any old competent, workmanlike, sensible departure from cap weighting will be better, and therefore that, as a class, funds that are self-described as "smart beta" ought to be better than cap-weighted funds.

That's testable, of course.
Well that is what Arnott is addressing.

"just about any" weighting will tilt towards small and value (relative to market cap weighting at least). (Unless you deliberately select the largest most overpriced securities). So it should outperform.

However, as Arnott is saying, if a strategy (i.e. weighting) makes a crowded trade and pushes up valuations, then you are no longer tilting value. Sure in backtests it was value, and it has outperformed recently as people rushed into the trade, but the valuations are now high and therefor have more muted expectations going forward.

noonakasmis
Posts: 6
Joined: Wed Feb 20, 2008 8:30 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by noonakasmis » Mon Feb 22, 2016 9:51 am

Larry,
larryswedroe wrote:Hank, we are CURRENTLY using on in tax advantaged accounts but considering for taxable as well.
And we are working with the firm to develop a core type fund with the tilt we prefer for most clients. So currently we use DFA core and then add small and large value to tailor to individuals who want/desire/need more tilt.
Larry
If you don't mind me asking. If you add a fund like QMNIX (AQR Market Neutral) or even QSPIX to a DFA core + DFA SV + AQR M/S Large portfolio, does it make a sense to replace some of the DFA/AQR M/S allocation with a cheap beta funds like TSM?

noon

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Mon Feb 22, 2016 12:44 pm

Noon
I don't think so since for various reasons (including rebalancing costs) the cheapest way to own the tilted portfolios is the core DFA funds for various reasons
larry

User avatar
Robert T
Posts: 2610
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Robert T » Fri Apr 29, 2016 7:00 pm

.
Webinar by Rob Arnott How Can “Smart Beta” Go Horribly Wrong?

Slide 18 - expected 5-year premiums (US markets)

Value = +7.9%
Size = +4.4%
Momentum = +2.9%
Low beta = -0.9%
Profitability = -4.6%
.

User avatar
grap0013
Posts: 1884
Joined: Thu Mar 18, 2010 1:24 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by grap0013 » Tue May 03, 2016 8:02 am

Very interesting presentation by Arnott. I think DFA, AQR, and RA all basically agree on the core elements of this stuff. The debates about nomenclature, smart beta, marketing etc... are immaterial. I'd wager there will be times when each of the companies appear to have superior strategies but I bet they will all have similar returns over long periods eg 20 years. AQR's EM fund was beating RA's fund for a while and now the tides have turned. Best bet is to own some of each.

I also think this presentation illustrated the benefit to diversifying your value strategy. Price to book and price to dividend are classic value plays whereas price to earnings and price to cash flow give you some profitability loadings as well. Hence, it is good to spread it around if for example profitability looks expensive etc...

Finally, like I have been saying, EM value does look pretty sweet. DFA's core EM fund has beaten their EM value fund for the past 6 consecutive years. So far YTD the tide is turning.
There are no guarantees, only probabilities.

Random Walker
Posts: 4233
Joined: Fri Feb 23, 2007 8:21 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Random Walker » Tue May 03, 2016 9:44 am

Here's another reason for the core fund approach. Core funds typically cheaper than the individual asset class funds, so overall portfolio ER probably lower using core as mainstay and only adding asset class funds as desired rather than doing it all with individual asset class funds.

Dave

Random Walker
Posts: 4233
Joined: Fri Feb 23, 2007 8:21 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Random Walker » Tue May 03, 2016 11:08 am

Regarding "How Can Smart Beta Go Horribly Wrong",

The only way I can see a portfolio diversified across sources of return going wrong is by increased costs. (I realize that is huge!). But I thing when one is diversified across risk factors / sources of return, if shortfall occurs it won't be by that much.

Dave

Theoretical
Posts: 1485
Joined: Tue Aug 19, 2014 10:09 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Theoretical » Fri Sep 02, 2016 2:00 pm

in_reality wrote:
nisiprius wrote:So I skimmed the article.

Question #1: One of the phrases that does not appear in the article is "fundamental index" or "RAFI fundamental index." So what about it? Are these, or are they not "smart beta" strategies? Are they exempt from the article's observation or not, and, if not, why not?
Yes, they are but no I don't consider them to be "smart beta" strategies in the same way that equal weighting, low volatility or momentum are. Neither does Arnott who both penned the article and created fundamental indexes.

If stocks in fundamental indexes get high valuations, you will be rotated out of them and into something with more favorable valuations (as they measure them). This is because they are choosing on low valuations (relative to a 5 year measure of economic footprint). This worked fine in the '08/'09 financial crisis when banks got really cheap and then recovered. Let's see if the same goes for oil now
Based on looking at the returns of PXH and FNDE this year, I'd say the long-term patient buying of the trash heap looks to be finally paying off.

User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by in_reality » Fri Sep 02, 2016 11:34 pm

Theoretical wrote:
in_reality wrote:
nisiprius wrote:So I skimmed the article.

Question #1: One of the phrases that does not appear in the article is "fundamental index" or "RAFI fundamental index." So what about it? Are these, or are they not "smart beta" strategies? Are they exempt from the article's observation or not, and, if not, why not?
Yes, they are but no I don't consider them to be "smart beta" strategies in the same way that equal weighting, low volatility or momentum are. Neither does Arnott who both penned the article and created fundamental indexes.

If stocks in fundamental indexes get high valuations, you will be rotated out of them and into something with more favorable valuations (as they measure them). This is because they are choosing on low valuations (relative to a 5 year measure of economic footprint). This worked fine in the '08/'09 financial crisis when banks got really cheap and then recovered. Let's see if the same goes for oil now
Based on looking at the returns of PXH and FNDE this year, I'd say the long-term patient buying of the trash heap looks to be finally paying off.
"Trash heap" is a term of endearment I presume. Actually though, I don't think PXH and FNDE are geared towards buying junky value traps. You can see this in their momentum loadings which don't go as negative. Sure they are going for companies that have a lower price than you'd expect for that company's 5 year performance, but they do hold more "growth" companies than a typical value fund -- of course those growth companies still fit the "lower price than you'd expect for that's company's 5 year performance" criteria.

FNDE did have me worried for a while I'll admit ... to be expected in value investing though right.

runnerguy
Posts: 52
Joined: Tue Dec 01, 2015 9:09 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by runnerguy » Sat Sep 03, 2016 9:16 am

Random Walker wrote:Regarding "How Can Smart Beta Go Horribly Wrong",

The only way I can see a portfolio diversified across sources of return going wrong is by increased costs. (I realize that is huge!). But I thing when one is diversified across risk factors / sources of return, if shortfall occurs it won't be by that much.

Dave
Thats my position as well. A low cost smart beta fund gives a higher chance of beating the market than a mutual fund, but if it doesn't the loss isn't life changing or regrettable. Here's an interesting take on it: http://www.realsmartbeta.com/if-low-vol ... s-crashes/

Theoretical
Posts: 1485
Joined: Tue Aug 19, 2014 10:09 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Theoretical » Sat Sep 03, 2016 1:43 pm

[/quote]

"Trash heap" is a term of endearment I presume. Actually though, I don't think PXH and FNDE are geared towards buying junky value traps. You can see this in their momentum loadings which don't go as negative. Sure they are going for companies that have a lower price than you'd expect for that company's 5 year performance, but they do hold more "growth" companies than a typical value fund -- of course those growth companies still fit the "lower price than you'd expect for that's company's 5 year performance" criteria.

FNDE did have me worried for a while I'll admit ... to be expected in value investing though right.[/quote]

It is totally a term of endearment. I was more going to people's perceptions of bank stocks after 2008 or huge stakes in Russian and Brazil oil companies now. And I very much like the lower turnover and inclusion of "better relative deal" growth stocks.

pkcrafter
Posts: 13748
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by pkcrafter » Sun Sep 04, 2016 9:32 am

Robert T wrote:
Slide 18 - expected 5-year premiums (US markets)

Value = +7.9%
Size = +4.4%
Momentum = +2.9%
Low beta = -0.9%
Profitability = -4.6%
.


Robert T,

I would like to see your final summary thoughts on factor investing.

Unrelated thoughts--It used to be that the F&F 3-factor model explained 80% of returns, so are all these other factors now attempting to capture the remaining 20%, or were F&F wrong?

We've known for a long time that value and growth swap leads from time to time and what we have been seeing is growth leading. This happened in the tech bubble as investors chased tech performance, and they are doing the same thing now. The huge surge into indexing has been the driver. Does that mean there is an index (Bogleheads) factor?

Ok, a bit of a spoof, but I'm not feeling the thrill. :happy


Paul
Last edited by pkcrafter on Sun Sep 04, 2016 6:46 pm, edited 1 time in total.
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

Random Walker
Posts: 4233
Joined: Fri Feb 23, 2007 8:21 pm

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Random Walker » Sun Sep 04, 2016 11:25 am

The only way I can see diversifying across risk factors / sources of return going wrong is via costs. Of course that's huge! Otherwise, I think if a portfolio diversified across sources of return underperformed, it would be by a small amount.

Dave

User avatar
Robert T
Posts: 2610
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Robert T » Mon Sep 05, 2016 3:32 am

pkcrafter wrote:Robert T,

I would like to see your final summary thoughts on factor investing.
My views have not changed much.

(i) the Fama-French paper 23 years ago on Common Risk Factors in the Returns on Stocks and Bonds is still a stand-out.

(ii) the CAPM (beta) explains about 70% of the variability in stock returns, the Fama-French three factor equity model explains about 95% of the variability in stock returns (see Exhibit 1 pg. 12 – Multifactor investing). Subsequent models have not significantly improved this explanatory power (perhaps increasing it by only about 1%).

(iii) as per the 1993 Fama-French paper I believe the (equity, size, value, term, and default) factor premiums - over time - are rewards for higher risk. As Swensen says ““Finance theory and capital market history provide analytical and practical underpinnings for the notion of a risk premium. Without expectations of superior returns for risky assets, the financial work would be turned on its head”…”The risk premium must exist for capital markets to function effectively”. I believe it is well documented in the literature that equities, small stocks, and value stocks have relatively higher risk. Fama provides a nice summary (see 5.45 minute mark in “Why Small Caps and Value Stocks Outperform” ).

(iv) as there is a less clear risk story for other ‘factors’ such as profitability and momentum, I am less confident in their durability. As a result, I currently don’t have portfolio exposure targets for profitability, but prefer a small cap and value tilted portfolio with a zero momentum load.

(v) there can be long periods of underperformance of factor (equity, small, value, term, default) premiums. We should not be surprised by this going forward. (for example see table 2 in Volatility and Premiums in US Equity Returns).

(vi) It’s difficult to explicitly time factor premiums (equity, size and value).

(vii) It is possible to capture the size and value premiums with long only strategies (which make up almost all the size premium and 60 percent of the value premium), in a fairly cost effective way.

(viii) The ‘optimal’ factor tilt is one you can stick with over the long-term.

Happy to say that I have had the same factor tilts over the last 14 years or so (same stock:bond mix; US:EAFE:EM allocation, and target size and value loads). No plans to change these anytime soon.

Robert
.

selters
Posts: 658
Joined: Thu Feb 27, 2014 9:26 am

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by selters » Mon Sep 05, 2016 5:59 am

It seems like all factors outperform the market. Which are the ones that underperform?

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Mon Sep 05, 2016 8:03 am

selters
Factors are LONG/SHORT portfolios so the factors that underperform are the SHORT side, so large, growth, low profitability, high investment, negative momentum, and so on.
Larry

larryswedroe
Posts: 16022
Joined: Thu Feb 22, 2007 8:28 am
Location: St Louis MO

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by larryswedroe » Mon Sep 05, 2016 8:07 am

Just to add to Robert T's comments, IMO there are attempts at providing risk-based explanations for profitability and momentum, but IMO none hold up to scrutiny. Thus they should be classified as behavioral anomalies. With that said momentum for example has been well documented for decades now and it persists and there are other well known anomalies that persist because limits to arbitrage and the fear and costs of shorting, especially in small, illiquid stocks, can allow anomalies to persist, and human behavior doesn't change very easily.

Best wishes
Larry

pkcrafter
Posts: 13748
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by pkcrafter » Mon Sep 05, 2016 10:41 am

Thanks Robert T. I thought the original 3-factor model only explained 80%, but if it's 95% that is even more convincing. Do you have confidence in the 95% number, or is it possible that it's actually lower (miscalculated) or has decreased due to attempts at exploitation? Even at 80% that doesn't leave much for behavioral funds.

I also agree that behavior is a real issue and it will continue.

Robert and Larry, have you looked at the Fuller/Thaler behavioral funds? Larry, I know you did a comparison to DFA, but what's the latest on these funds. I know one F/T behavioral fund has been closed, so even though we know misbehavior will continue, it appears difficult to exploit.

Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

User avatar
Robert T
Posts: 2610
Joined: Tue Feb 27, 2007 9:40 pm
Location: 1, 0.2, 0.4, 0.5
Contact:

Re: How Can "Smart Beta" Go Horribly Wrong?

Post by Robert T » Mon Sep 05, 2016 12:10 pm

pkcrafter wrote:Robert and Larry, have you looked at the Fuller/Thaler behavioral funds? Larry, I know you did a comparison to DFA, but what's the latest on these funds. I know one F/T behavioral fund has been closed, so even though we know misbehavior will continue, it appears difficult to exploit.
When I looked at this 8 years ago there were three "Fuller and Thailer" (behavioral) funds available to individual investors - none added significant alpha beyond factor exposure as per this earlier analysis. Since then, two of the funds have closed (no longer exist), with only the Undiscovered Mgrs Behavioral Value A (UBVAX) still operating.

The updated analysis tells the same story - no significant alpha beyond factor exposure (the Vanguard Small Value Index fund was added to the analysis for comparison).

Fuller and Thaler Asset Management say "Investors make mental mistakes. Fuller and Thaler’s objective is to exploit them. Our investment approach applies insights from some of the foremost sholars in Behavioral Finance to identify these opportunities and gain a competitive edge over the market.” The above results suggest (that despite their best efforts) they have not managed to exploit investor mistakes to add significant alpha beyond factor exposure.

Robert
.

Post Reply