Fundamental Indexing

[twa2w]

o'shaughnessy provided 40+ years of back testing and he now has 10 years of actual investing results for 2 of his stratgies.) His new funds are now offering new strategies - also back tested.

Not sure if the American based fund he sold to Hennessey Funds are still following his strategies but if they are then he has about 12 or so years of results on those strategies whihc are I think slightly different than the stratgies He used on his RBC funds.

Cheers
j

[DenisD]

o'shaughnessy provided 40+ years of back testing and he now has 10 years of actual investing results for 2 of his stratgies.) His new funds are now offering new strategies - also back tested.

He made major changes to the US Value fund algorithm two or three years ago when he added buyback percent to dividend yield. So I don't think most of that 10-year record applies to the current fund. He made minor changes to the US Growth fund algorithm at the same time.

Do you know how long the back test is for the new strategies? Some of them aren't in his book.

[NormR]

Do you know how long the back test is for the new strategies? Some of them aren't in his book.

Are you sure? The most recent edition of the book contains the yield plus buyback approach. That and updates of the older strategies. More importantly, the new book has all sorts of other stats with various downside risk measures.

But you can just go over here

[DenisD]

Do you know how long the back test is for the new strategies? Some of them aren't in his book.

Are you sure?

Pretty sure. I've read the latest edition of his book (but don't own it) and I don't remember seeing the following strategies from the RBC prospectus:

The All-Canadian Value Strategy:
> starts with the top 300 securities as ranked by average monthly
trading dollar volume and selects 50 securities with the lowest
price-to-cash flow ratios which rank in the top two-thirds of the
following criteria:
> year-over-year 12-month earnings growth,
> six-month total return, and
> analyst earnings revisions.

The All-Canadian Growth Strategy:
> starts with the top 300 securities as ranked by average monthly
trading dollar volume and selects 50 securities with the highest
price increase for the previous six months which rank in the top
two-thirds of the following criteria:
– price to cash flow,
– gross profit margins, and
– earnings quality.

The Global Blend Strategy:
> selects the 50 securities with the highest multifactor ranking. The
securities must have sufficient market capitalization and trading
volume. The Global Blend Strategy screens for securities with
attractive growth and value characteristics using a factor-based
model. Each security is ranked according to the following factors:
– price momentum (above-average three-, six-, nine-, and
12-month price increase),
– value (price-to-sales ratio, price-to-cash flow ratio and
dividend yield),
– earnings quality, and
– analyst earnings revisions.

The link you gave just gives the yearly returns from the strategies in the book. And only to 2004.

[NormR]

Humm, I'm not sure how far back you can get info on 'analyst earnings revisions'. Perhaps they'll be in the next version of the book?

[DanH]

Humm, I'm not sure how far back you can get info on 'analyst earnings revisions'. Perhaps they'll be in the next version of the book?

Momentum managers that have presented back-tested numbers state that such data was available as far back as the mid-1980s. I haven't see the data, just the footnotes citing data sources and associated time frames.

[tidal]

In what seems as though it is becoming a "permanent feature", there are more articles on the "Fundamental Indexing Debate" in the Sept/Oct 2007 issue of Journal of Indexes, including:

[url=http://www.indexuniverse.com/index.php?option=com_content&view=article&Itemid=34&issue=117&id=2995]Fundamental Indexing Smackdown
Rob Arnott, Gus Sauter, Jeremy Siegel
Debating the merits of alternative weighting methodologies.[/url]

and

[url=http://www.indexuniverse.com/index.php?option=com_content&view=article&Itemid=34&issue=117&id=2998]A Fundamental Challenge
Luciano Siracusano
Is there a better way to index?[/url]

Is there an emoticon for "YAWN"?

[NormR]

Past is Not Prologue, and Hope Is Not a Strategy -- Guest Column by Rob Arnott

[DenisD]

What is Fundamental about a Fundamental Index? by Rob Arnott

Forget about fundamental indexing! What I want to buy are these companies:

The companies that are on the Russell 1000 and not the FTSE-RAFI 1000 are – by definition – large-cap small companies. In other words, they’re little companies trading at lofty enough multiples to enter the large-cap universe. The companies that are on the FTSE-RAFI 1000 and not the Russell 1000 are – again by definition – small-cap large companies. They’re big companies trading at distressed enough multiples to fall into the small-cap domain. Which list performs better? Small-cap large companies have outperformed large-cap small companies, since 1962, by nearly 1000 basis points per annum. If we merely reweight the Russell 1000, we miss these wonderful companies.

Added: On second thought, maybe the small-cap large companies aren't so great. The large-cap small companies probably underperformed the index substantially.

[DenisD]

2007—A PROOF STATEMENT!

The RAFI® strategy performed brilliantly in 2005 and 2006. So what? Value beat growth in those years, so we had a tailwind. As Gus Sauter suggested in 2005, the real test is how it does when growth beats value. By all measures, 2007 was a strong growth-dominated market. So, how did the RAFI approach perform in what should be a hostile environment for a value-oriented strategy? Actually, quite well. Indeed, startlingly better than classic value strategies and even far better than most quantitative enhanced index strategies.

[Icarus]

2007—A PROOF STATEMENT!

The RAFI® strategy performed brilliantly in 2005 and 2006. So what? Value beat growth in those years, so we had a tailwind. As Gus Sauter suggested in 2005, the real test is how it does when growth beats value. By all measures, 2007 was a strong growth-dominated market. So, how did the RAFI approach perform in what should be a hostile environment for a value-oriented strategy? Actually, quite well. Indeed, startlingly better than classic value strategies and even far better than most quantitative enhanced index strategies.

The devil is in the details. See footnote 2. RAFI beat the S&P 500 by 80 bps, but this does not include trading costs or management expenses. The Net Ratio is 70 basis points vs. 7 bps for VTI. So the approximate outperformance is 17 bps (using VTI as a proxy for S&P 500, which is not technically correct). And the RAFIs would be expected to have higher turnover and higher dividends, so there would be additional tax drag, much worse for a taxable Canadian investor.

Meanwhile, this comparison is inappropriate anyway. Value outperformed growth over those three years, so a better comparison is a portfolio with a value tilt, rebalanced annually. A rebalanced portfolio would also "ramp up its value exposure when value offers a larger premium," a claimed benefit of the RAFI indices.

What this really needs is a regression analysis, to see how much of the premium is due to the technique and how much due to naive value and size factors. I'm not smart enough to do that, but RAFI is and the fact that they chose not to is suspicious. It would be most meaningful if the results factored in management fees (and taxes for taxable investors).

Three years is too short for a meaningful comparison, of course, but RAFI chose to do it. Even without doing any fancy math, their results are really pretty unimpressive to me, and I did not start out with a bias against them. (I was actually thinking of buying them.)

[WishingWealth]

Arrived at the Interfluidity blog through some convoluted ways.
Surprising to read that kind of prose in 2008.

http://www.interfluidity.com/posts/1205101710.shtml

In Praise of Active Investing.
...So active investors, as a class, do better than they otherwise would have by bearing the high cost of active investment, even though in doing so they must endure the indignity of being outperformed on average by those who free-ride off their work! It is perverse, under these circumstances, to accuse active investors of squandering $100B, and recommend that we all move to index funds. On the contrary, it is passive investors who ought be discouraged. Passive investors pay none of the costs of generating good investment decisions, but enjoy the benefits by free-riding on the work of others. Their copycatting reduces compensation to those who have earned returns by performing or underwriting informational work. Passive investing also introduces feedback effects and noise into asset prices, rendering the work of active investors more costly and less effective. (See, e.g. information cascades — ht Mark Thoma — and this interesting model of bubbles and crashes — ht jck of Alea — for ways that copying others' investment decisions as reflected in price moves can lead to instability and error in markets.)

The world of money management is full of shysters and charlatans who'll take "active investment" fees and do nothing useful with them, sure. But part of that headline $100B "cost" funds real information work, without which markets would devolve entirely to lotteries. Advising people to buy index funds rather than select investments is akin to advising people not to vote, since the cost of voting far exceeds any individual benefit. Those who don't vote get the same government everyone else does, but at lower cost! A citizenry that takes this reasoning to its logical extreme will get the government it deserves. An investor class that flocks to index funds may soon have the stock market it deserves.

(Did not find a Worth Reading thread in the ETF section)

WW

[adrian2]

Morgan Stanley fundamentally analyzes fundamental indexing

In Conclusion
In short, the study worked out like this: during periods where value outperformed, the fundamentally weighted indexes beat market-cap benchmarks; during periods where growth outperformed, they trailed. The fundamental indexes may not be just value indexes, but they are certainly correlated with value-based outperformance.

Kind of brought back to mind by the incessant ads on BNN from Claymore: share prices don't tell the whole story; fundamental indexes weigh stocks based on sales, number of employees etc.

My point, made repeatedly over the years, is that fundamental indexing is a form of value investing, not a "better" weighting than market cap. So my timely question to all supporters of fundamental vs. market cap indexing is now that value-style does not do so well, how do you feel about giving Citi (e.g.) a much higher weighting in the U.S. index than its market cap? After all, it has a lot of employees etc. Without going into the merits or lack of them for Citi specifically, how does it feel with a "better implementation" of indexing?

[Taggart]

Morgan Stanley fundamentally analyzes fundamental indexing

In Conclusion
In short, the study worked out like this: during periods where value outperformed, the fundamentally weighted indexes beat market-cap benchmarks; during periods where growth outperformed, they trailed. The fundamental indexes may not be just value indexes, but they are certainly correlated with value-based outperformance.

Kind of brought back to mind by the incessant ads on BNN from Claymore: share prices don't tell the whole story; fundamental indexes weigh stocks based on sales, number of employees etc.

My point, made repeatedly over the years, is that fundamental indexing is a form of value investing, not a "better" weighting than market cap. So my timely question to all supporters of fundamental vs. market cap indexing is now that value-style does not do so well, how do you feel about giving Citi (e.g.) a much higher weighting in the U.S. index than its market cap? After all, it has a lot of employees etc. Without going into the merits or lack of them for Citi specifically, how does it feel with a "better implementation" of indexing?

Adrian,

I've always known that fundamental indexing is a form of value investing. Nothing wrong with that. The danger lies in too high a percentage allocated to a few sectors, but then the problem with market cap indexing is that the bubble stocks get too high a percentage in the portfolio like Nortel did in 1999.

[adrian2]

I've always known that fundamental indexing is a form of value investing. Nothing wrong with that.

Agreed. My beef is with the claim that it better represents the market as a whole.

The danger lies in too high a percentage allocated to a few sectors, but then the problem with market cap indexing is that the bubble stocks get too high a percentage in the portfolio like Nortel did in 1999.

The answer to that problem is to use a capped index, not a fundamental one. With the fundamental version, you always have turnover, which may or may not help (Bylo + Selhi), but with the capped index, the only turnover is when the cap is breached (plus when stocks drop out of the index, but in that case there is usually no capital gain to worry about).

[Taggart]

I've always known that fundamental indexing is a form of value investing. Nothing wrong with that.

Agreed. My beef is with the claim that it better represents the market as a whole.

Perhaps that claim came from someone, but certainly not from me. I've always felt that a capped index represents the market as a whole, and yes I do use them to a small extent through TD e-funds. I also own a larger proportion of foreign dividend ETF's in the portfolios. I ignored the warnings about too much concentration in certain sectors.

The danger lies in too high a percentage allocated to a few sectors, but then the problem with market cap indexing is that the bubble stocks get too high a percentage in the portfolio like Nortel did in 1999.

The answer to that problem is to use a capped index, not a fundamental one. With the fundamental version, you always have turnover, which may or may not help (Bylo + Selhi), but with the capped index, the only turnover is when the cap is breached (plus when stocks drop out of the index, but in that case there is usually no capital gain to worry about).

Yes, I'm well aware of the lower turnover and costs in the capped indexes, certainly a consideration for anyone wanting to index.

[DenisD]

So my timely question to all supporters of fundamental vs. market cap indexing is now that value-style does not do so well, how do you feel about giving Citi (e.g.) a much higher weighting in the U.S. index than its market cap?

I feel the way I always feel. Ignore the short term, as much is possible that is. Concentrate on the long term.

Coincidentally, your concern is addressed in the latest issue of RAFI Fundamentals.

Discounts and Relative Performance

[Norbert Schlenker]

I'm sure there's a thread somewhere about ETFs shutting down but I can't find it offhand, so I'm putting this here because 12 of 19 are RAFI based funds in the US.

Invesco Announces "Changes" to ETF Family

[ockham]

I'm sure there's a thread somewhere about ETFs shutting down but I can't find it offhand, so I'm putting this here because 12 of 19 are RAFI based funds in the US.

Invesco Announces "Changes" to ETF Family

Meanwhile, in Canada, on April 7/09, Claymore launched a non-hedged US RAFI index ETF. So far, trade volume appears to be nil.

[DenisD]

Invesco Announces "Changes" to ETF Family

Luckily, they're not axing any of mine. The RAFI ETF's had an excellent month in April. All those financials they picked up in the rebalance.

Fundamental Indexing Is Working (Recently)

Just for balance, here's a con opinion.

Can You Out-Index The S&P

[DenisD]

The RAFI ETF's had an excellent month in April. All those financials they picked up in the rebalance.

And now Rob is gloating [s]too[/s].

THE GREAT CONTRA-TRADE

Many observers credit the value tilt of RAFI strategies as the source of its long-term historical success. The reality is more subtle. The main source of value-added is not the average value tilt of the RAFI portfolios, but its dynamic contra-trading against the most extreme market bets. Value stocks got cheaper and cheaper and—as a direct consequence—our value tilt got larger and larger. These dynamic style tilts are primarily the result of contra-trading against the market's constantly shifting expectations, fads, bubbles, and crashes.

...

April was the best month ever for FTSE RAFI US 1000 relative to both the S&P 500 Index (9.33% excess return) and the Russell 1000 Value (8.18% excess return.)

[Taggart]

The Wall Street Journal

MAY 31, 2009, 7:00 P.M. ET

Indexing Debate

Which is better for investors: an S&P-500 index fund— or an ETF with a twist?

By JOHN SPENCE

[DenisD]

Indexing Patent to Test Fund Firms

Research Affiliates LLC, based in Newport Beach, Calif., has received a patent for an indexing methodology that selects and weights securities using fundamental measures of company size, such as dividends and sales. Many traditional benchmarks weight stocks by their market capitalization.
.
.
.
The patent could raise questions for exchange-traded-fund provider WisdomTree, which offers a number of funds tracking its own dividend-weighted and earnings-weighted indexes. The Research Affiliates patent covers indexes built on all sorts of fundamental measures, such as sales, earnings and dividends, but not market-cap weighting, price weighting or equal weighting.

WisdomTree, which had $5.5 billion in assets under management as of Sept. 30, didn't return calls for comment.
.
.
.
But those seeking to follow in Research Affiliates' footsteps may have a long wait. The firm's original patent application was filed in 2002, Mr. Arnott says. "It's been a long slog."

[Taggart]

Monday, 15th February 2010

Investors lose faith in active managers

One paragraph in particular in this piece really caught my eye.

"Elroy Dimson, Paul Marsh and Mike Staunton of the London Business School warned index opportunities could be lost as a result of a weight of money pushing markets one way or another: “Once a premium has been identified and the research disseminated, the rewards to factor exposure might change,” they said."

[DenisD]

FTSE/RAFI indexes rebalance after the close on the third Friday of March. This was an excellent time to rebalance last year. I decided to see how things changed this year. So, I recorded the company and sector weightings for CRQ for March 18-23.

First, we have the company weights for the four days sorted from high to low on March 19. It looks like the rebalance did happen on Monday. The last column shows the weighting change between Friday and Monday. Big changes in many companies.

Code: Select all

                                          CRQ
Name                                   18-Mar  19-Mar  22-Mar  23-Mar  Change
TECK COMINCO LTD CL B                   9.17%   8.97%   1.96%   2.00%    -78%
ROYAL BANK OF CANADA                    8.27%   8.32%   7.49%   7.53%    -10%
TORONTO-DOMINION BANK                   6.61%   6.71%   6.45%   6.45%     -4%
BANK OF MONTREAL (I/L)                  5.36%   5.41%   4.25%   4.28%    -21%
BANK OF NOVA SCOTIA CAD                 5.29%   5.33%   5.36%   5.39%      1%
MANULIFE FINANCIAL CORP                 5.16%   5.13%   4.94%   4.89%     -4%
SUNCOR ENERGY INC                       4.79%   4.70%   2.19%   2.19%    -53%
CANADIAN IMPERIAL BANK OF COMMERCE      3.30%   3.35%   3.37%   3.38%      1%
SUN LIFE FINANCIAL                      2.90%   2.91%   3.07%   3.07%      5%
CANADIAN NATURAL RESOURCES LTD          2.71%   2.71%   2.53%   2.52%     -7%
POWER CORP OF CANADA SERIES D           2.66%   2.64%   2.67%   2.68%      1%
THOMSON CORP                            2.64%   2.63%   1.40%   1.42%    -47%
MAGNA INTERNATIONAL INC                 2.39%   2.43%   1.96%   1.95%    -19%
CANADIAN NATIONAL RAILWAY CO (I/L)      1.81%   1.85%   2.23%   2.20%     21%
TRANSCANADA CORPORATION                 1.76%   1.78%   2.59%   2.58%     46%
ENCANA CORP                             1.63%   1.60%   4.06%   4.09%    154%
TALISMAN ENERGY INC (I/L)               1.59%   1.55%   1.60%   1.58%      3%
BROOKFIELD ASSET MANAGEMENT INC CL A    1.41%   1.39%   1.71%   1.70%     23%
BOMBARDIER INC B SUB VTG SHARES         1.35%   1.38%   0.89%   0.89%    -36%
CELESTICA INC. (I/L)                    1.36%   1.35%   0.59%   0.61%    -56%
CENOVUS ENERGY INC                      1.31%   1.30%   0.78%   0.78%    -40%
POWER FINANCIAL CORP                    1.29%   1.28%   1.32%   1.33%      3%
BARRICK GOLD CORP COM (I/L)             1.27%   1.27%   1.92%   1.88%     51%
NATIONAL BANK OF CANADA                 1.15%   1.16%   1.21%   1.20%      4%
ENBRIDGE INC                            1.08%   1.08%   1.75%   1.76%     62%
CANADIAN PACIFIC RAILWAY LTD.           1.02%   1.02%   1.19%   1.17%     17%
BCE INC (CAD)                           1.02%   1.01%   1.47%   1.47%     46%
ONEX CORP SUB VTG                       0.99%   0.98%   1.04%   1.04%      6%
GREAT WEST LIFECO INC                   0.93%   0.93%   0.99%   1.00%      6%
NEXEN INC                               0.89%   0.91%   1.17%   1.16%     29%
QUEBECOR INC CL B SVS                   0.82%   0.82%   0.60%   0.60%    -27%
BROOKFIELD PROPERTIES CORP              0.80%   0.82%   0.60%   0.60%    -27%
ROGERS COMMS INC CL B (I/L) NV          0.80%   0.80%   1.60%   1.59%    100%
GOLDCORP INC                            0.78%   0.78%   1.24%   1.25%     59%
POTASH CORP OF SASKATCHEWAN             0.76%   0.76%   1.35%   1.34%     78%
FAIRFAX FINANCIAL HOLDINGS LTD          0.72%   0.72%   0.93%   0.93%     29%
RESEARCH IN MOTION LTD(CAD$-SHS)        0.72%   0.71%   1.10%   1.12%     55%
CANADIAN TIRE CORP LTD CLASS A NON V    0.58%   0.59%   0.74%   0.73%     25%
GEORGE WESTON LTD                       0.56%   0.58%   0.78%   0.78%     34%
LOBLAW COS LTD                          0.58%   0.58%   0.75%   0.74%     29%
EMPIRE CO LTD N/V CL A                  0.55%   0.56%   0.79%   0.80%     41%
AGRIUM INC                              0.52%   0.53%   0.79%   0.80%     49%
IMPERIAL OIL LTD (I/L)                  0.51%   0.52%   0.99%   0.99%     90%
HUSKY ENERGY INC                        0.51%   0.52%   1.12%   1.11%    115%
ALIMENTATION COUCHE-TARD CLB            0.50%   0.50%   0.59%   0.59%     18%
INTACT FINANCIAL CORP.                  0.50%   0.50%   0.55%   0.55%     10%
CGI GRP INC                             0.51%   0.50%   0.50%   0.49%      0%
IGM FINANCIAL INC                       0.47%   0.48%   0.53%   0.53%     10%
SHOPPERS DRUG MART INC.                 0.46%   0.46%   0.85%   0.85%     85%
CANADIAN UTILITIES LTD N/V CL A         0.45%   0.45%   0.71%   0.71%     58%
RIOCAN REAL ESTATE INVESTMENT TRUST     0.44%   0.44%   0.44%   0.44%      0%
METRO INC                               0.43%   0.43%   0.67%   0.66%     56%
FINNING INTERNATIONAL LTD               0.43%   0.42%   0.44%   0.44%      5%
SHAW COMMUNICATION INC CLASS B          0.42%   0.42%   0.76%   0.76%     81%
JEAN COUTU GROUP (PJC) INC              0.41%   0.41%   0.41%   0.41%      0%
ATCO LTD CL I N/V                       0.40%   0.39%   0.55%   0.55%     41%
TRANSALTA CORPORATION -COM              0.34%   0.34%   0.75%   0.75%    121%
KINROSS GOLD CORP                       0.33%   0.33%   0.42%   0.42%     27%
BIOVAIL CORPORATION                     0.31%   0.32%
YAMANA GOLD INC COM                     0.27%   0.27%   0.45%   0.45%     67%
GERDAU AMERISTEEL                       0.25%   0.25%   0.27%   0.27%      8%
SEARS CANADA INC                        0.22%   0.22%   0.22%   0.22%      0%
TELUS CORPORATION                       0.21%   0.22%   0.40%   0.40%     82%
TELUS CORPORATION NON VOTING            0.17%   0.17%   0.31%   0.31%     82%
MANITOBA TELECOM SERVICES INC           0.12%   0.13%   0.28%   0.27%    115%
BOMBARDIER INC A CV MULTIPLE VTG        0.01%   0.01%   0.04%   0.04%    300%
FORTIS INC                                              0.57%   0.57%
CI FINANCIAL CORP                                       0.49%   0.49%
CAMECO CORP (I/L)                                       0.48%   0.48%
VITERRA INC.                                            0.48%   0.47%
SAPUTO GROUP INC (I/L)                                  0.31%   0.31%


Next, we have the sector weights. The financials didn't go down as much as I'd hoped. I've included the XIU weights for comparison.

Code: Select all

                               CRQ                                     XIU
Sector                      18-Mar  19-Mar  22-Mar  23-Mar  Change  24-Mar
Financials                  48.28%  48.50%  47.41%  47.49%     -2%  34.35%
Energy                      16.78%  16.67%  19.26%  19.23%     16%  25.61%
Materials                   13.36%  13.15%   8.41%   8.39%    -36%  17.89%
Consumer Discretionary       7.06%   7.11%   5.69%   5.69%    -20%   4.27%
Industrials                  4.62%   4.67%   4.79%   4.74%      3%   5.20%
Consumer Staples             3.49%   3.52%   5.62%   5.61%     60%   2.46%
Information Technology       2.59%   2.56%   2.18%   2.22%    -15%   3.61%
Telecommunication Services   2.33%   2.33%   4.05%   4.05%     74%   5.08%
Utilities                    1.19%   1.18%   2.58%   2.59%    119%   0.97%
Health Care                  0.31%   0.32%                           0.26%