The Case for Indexing - More Evidence

In his article Measuring the True Cost of Active Management, Ross Miller showed how expensive the “active” portion of traditional long-only mutual funds really is.

Earlier this year, he followed up with another article on the same subject. Stansky’s Monster: A Critical Examination of Fidelity Magellan’s Frankenfund” makes even more painfully clear how expensive actively managed mutual funds can really be to investors.

Miller’s work provides very strong evidence that, instead of investing in traditional “long only” actively managed mutual funds, most investors would be much better off diversifying their portfolios across different types of beta risk (i.e., broadly defined asset class index funds) and separately deciding how much to allocate to pure active management products (in the form of market neutral funds), where the expenses and returns of active management are made clear.

To cite one example of how this might be done, in our model portfolios, our reported results for market neutral products represent an equally weighted allocation to four equity oriented products. They are: James Market Neutral, JAMNX; Hussman Strategic Growth, HSGFX; Analytic Investors Global Long-Short, ANGLX; J.P. Morgan Market Neutral, OGNAX; and the Deutsche Bank G10 Currency Harvest ETF (DBV) that pursues a market neutral foreign exchange based strategy.

Comments

[Brad Zigler]

Very few investors will become cognizant of the true costs of active management unless market observers and pundits start using metrics like Miller's.

Most, however, shun the mention of such yardsticks for fear of confusing investors.

Take the instance described in the Journal of Indexes back in 2006:


Journal of Indexes - January/February 2006

www.indexuniverse.com/...

The Curmudgeon
By Brad Zigler

"Hooray for 'new math'...
It's so simple,
So very simple
That only a child can do it!"

-Apologies to Tom Lehrer

Now, I'm not one to complain, but this column didn't get named Curmudgeon for nothing.

Readers of the November 2005 issue of Money magazine were treated to an article by Jason Zweig on the innovative work of Ross Miller, a risk consultant and finance professor at the State University of New York. Specifically, Zweig took aim at Miller's recently published paper (1) offering a novel, and mathematically simple, way to measure the true cost of active management by mutual funds.

Miller contends that a fund's variability-and its costs-can be parsed into active and passive components to better gauge a manager's value.

So, what's the problem?

Well, while Zweig acknowledged the sensibility of Miller's approach, he ends up pooh-poohing Miller's work because "the underlying math is too unwieldy."

Now what's so unwieldy about:

(See Web article at URL above for math formula)

As Miller puts it: "If you can use the memory function of your hand-held calculator, you can derive a portfolio's active weight." And with that in hand, he says, a fund's active expense ratio-the "true cost" of alpha seeking-and its active alpha can be readily derived. So basically, all you need to start peering under a fund's hood is its r-squared correlation cheerfully supplied by Morningstar or some other data vendor and your battery powered slipstick. How unwieldy is that?

By Miller's calculus, for example, about 26 percent (2) of TIAA-CREF's actively managed Mid-Cap Value Fund (TRVRX) has mimicked its Russell Midcap Value benchmark this year.

Knowing that, our 'new math' allows us to concentrate the fund's excess cost-that is, the premium over the cheapest available index fund's expenses-on the actively managed portion of the portfolio to come up with the ex-beta cost of 115 basis points. That's more than twice the fund's then-advertised ratio of 48 bps.

There's sense in knowing this. Why would anyone want to pay full boat for market tracking, after all, if there are cheaper options available?

Zweig says, "Miller's measure isn't ready to replace the cost standard we all use."

We all use?

Well, we all used horses as our primary means of transport once.

More properly, I think, pundits like Zweig aren't ready to give up the buggy whip of fund-manufactured data. Miller's math offers fine detail about the operation of funds that many portfolio runners would like to keep hush-hush.

Oddly enough, index fund managers may have the most to hide. Let's go back to TIAA-CREF for an example. In addition to the actively managed TRVRX portfolio, TIAA-CREF also offers a passive portfolio-TRVUX-which tracks the Russell Midcap Value Index. Over the past few months, TRVUX's r-squared coefficient has been .9958, which seemingly makes the fund a doppelganger for the Russell benchmark (TRVRX's contemporaneous correlation was .8946).

Miller's math, however, tells us that a sliver-six percent-of the portfolio is gamed. This intuitively makes sense when you consider that some dodging and weaving must be done by the fund's managers to avoid being picked off when index reconstitutions force them into the market to buy and sell holdings.

Our calculus now allows us to readily quantify the price of index portfolio gamesmanship. If an active weight of six percent is "normal," we should be able to easily tell when an index manager develops wanderlust.

More importantly, we can see how much that ticket costs. TRVUX's active expense ratio is 338 basis points, nearly eight times its published cost. Active management "around the edges," it seems, can be very costly.

Now where'd I put my calculator?

Endnotes:

(1) Measuring the True Cost of Active Management by Mutual Funds at papers.ssrn.com/abstra...= 746926.
(2) Based on weekly data, February 22 -November 18, 2005.

[Brad Zigler]

Very few investors will become cognizant of the true costs of active management unless market observers and pundits start using metrics like Miller's.

Most, however, shun the mention of such yardsticks for fear of confusing investors.

Take the instance described in the Journal of Indexes back in 2006:


Journal of Indexes - January/February 2006

www.indexuniverse.com/...

The Curmudgeon
By Brad Zigler

"Hooray for 'new math'...
It's so simple,
So very simple
That only a child can do it!"

-Apologies to Tom Lehrer

Now, I'm not one to complain, but this column didn't get named Curmudgeon for nothing.

Readers of the November 2005 issue of Money magazine were treated to an article by Jason Zweig on the innovative work of Ross Miller, a risk consultant and finance professor at the State University of New York. Specifically, Zweig took aim at Miller's recently published paper (1) offering a novel, and mathematically simple, way to measure the true cost of active management by mutual funds.

Miller contends that a fund's variability-and its costs-can be parsed into active and passive components to better gauge a manager's value.

So, what's the problem?

Well, while Zweig acknowledged the sensibility of Miller's approach, he ends up pooh-poohing Miller's work because "the underlying math is too unwieldy."

Now what's so unwieldy about:

(See Web article at URL above for math formula)

As Miller puts it: "If you can use the memory function of your hand-held calculator, you can derive a portfolio's active weight." And with that in hand, he says, a fund's active expense ratio-the "true cost" of alpha seeking-and its active alpha can be readily derived. So basically, all you need to start peering under a fund's hood is its r-squared correlation cheerfully supplied by Morningstar or some other data vendor and your battery powered slipstick. How unwieldy is that?

By Miller's calculus, for example, about 26 percent (2) of TIAA-CREF's actively managed Mid-Cap Value Fund (TRVRX) has mimicked its Russell Midcap Value benchmark this year.

Knowing that, our 'new math' allows us to concentrate the fund's excess cost-that is, the premium over the cheapest available index fund's expenses-on the actively managed portion of the portfolio to come up with the ex-beta cost of 115 basis points. That's more than twice the fund's then-advertised ratio of 48 bps.

There's sense in knowing this. Why would anyone want to pay full boat for market tracking, after all, if there are cheaper options available?

Zweig says, "Miller's measure isn't ready to replace the cost standard we all use."

We all use?

Well, we all used horses as our primary means of transport once.

More properly, I think, pundits like Zweig aren't ready to give up the buggy whip of fund-manufactured data. Miller's math offers fine detail about the operation of funds that many portfolio runners would like to keep hush-hush.

Oddly enough, index fund managers may have the most to hide. Let's go back to TIAA-CREF for an example. In addition to the actively managed TRVRX portfolio, TIAA-CREF also offers a passive portfolio-TRVUX-which tracks the Russell Midcap Value Index. Over the past few months, TRVUX's r-squared coefficient has been .9958, which seemingly makes the fund a doppelganger for the Russell benchmark (TRVRX's contemporaneous correlation was .8946).

Miller's math, however, tells us that a sliver-six percent-of the portfolio is gamed. This intuitively makes sense when you consider that some dodging and weaving must be done by the fund's managers to avoid being picked off when index reconstitutions force them into the market to buy and sell holdings.

Our calculus now allows us to readily quantify the price of index portfolio gamesmanship. If an active weight of six percent is "normal," we should be able to easily tell when an index manager develops wanderlust.

More importantly, we can see how much that ticket costs. TRVUX's active expense ratio is 338 basis points, nearly eight times its published cost. Active management "around the edges," it seems, can be very costly.

Now where'd I put my calculator?

Endnotes:

(1) Measuring the True Cost of Active Management by Mutual Funds at papers.ssrn.com/abstra...= 746926.
(2) Based on weekly data, February 22 -November 18, 2005.

[ETF Wanderer]

Hi, Can you make an index out of the four funds: James Market Neutral, JAMNX; Hussman Strategic Growth, HSGFX; Analytic Investors Global Long-Short, ANGLX; J.P. Morgan Market Neutral, OGNAX and show how the index has performed. Alternatively, it will be great if you can provide this "Listed Market Neutral Fund Index" on Bloomberg & Reuters - it will be a great way for people to track these funds.

Thanks.