Morningstar's 'Vastly Superior' ETF Research?

A couple weeks ago we received an e-mail from Morningstar announcing a new type of ETF research. Here's a key passage from Morningstar's Jeffrey Ptak:

Our new Analyst Reports mark a novel approach to evaluating ETFs. We're harnessing the research that our 100-plus equity analysts conduct in order to estimate the fair value of stock ETFs.

By comparing our fair value estimate for an ETF's portfolio with the fund's market price, we can better gauge whether an ETF is cheap, dear, or reasonably priced. In so doing, we can help you profit from inviting opportunities as they arise.

Our ETF fair value estimates are based on our stock analysts' rigorous, bottom-up research. Our analysts are digging into company filings, questioning management, conferring with suppliers and competitors, and creating sophisticated financial models--all to estimate each firm's worth and risk. And they are conducting that research in accordance with a disciplined, consistently applied investment philosophy.

We think this is a vastly superior approach to the breathless macro-economic calls, technical analysis, or short-cut methods that often pass for fundamental analysis.

Vastly superior? That claim strikes us as a bit inflated. We make no case for breathlessness, of course, but we do think there's a...er...fundamental problem with this kind of "fair value" analysis. As ever, equity prices are functions of (more or less*) objective facts (earnings, revenue, margins, &c.) and investors' entirely subjective willingness to pay for such things. So not only are "future facts" difficult to project with precision, but who can say what traders and investors will pay for them?

Given this two-dimensional uncertainty, we think estimates of fair value--over what time frame? in what sort of broad market conditions? assuming current industry trends persist? or that they change?--are surrounded by such enormous error terms that they might more profitably be ignored than followed.

But if not estimates of fair value, if one doesn't have a "price target" for a given security or, in turn, a pool of securities such as an ETF, what's left? We'd say known fundamentals (i.e., recent "facts") and current price and volume action are the only serviceable sources of advantage in our relatively efficient, hyper-competitive, often random financial markets.

Now, this is a little too easy, but we can't resist. In a piece accompanying the item from which we pulled the above exceprt, Ptak concludes with this:

For example, as of this writing, we'd expect Financial Select SPDR (XLF) to generate a 22.4% annualized return based on a comparison of the fund's price to our estimate of its intrinsic worth. That ETF's hurdle rate is roughly 14% (sum of the fund's 10.8% cost of equity, 0.24% annual expense ratio, and an approximate 3% incremental risk premium). Given that the fund's 8.4% risk-adjusted excess return--22.4% expected return less its 14% hurdle rate--is greater than 3%, we're currently recommending it.

On February 27th, XLF closed at $27.56. At last check today it was changing hands at $24.23, a decline of 12.1% in two weeks. Which means that it'll have to gain 39.2% from here to reach Morningstar's "fair value" estimate of $33.73 in the next 50 weeks. We'd say that's possible. And exceedingly unlikely.

~~~~~~~~~~~~~~~~

* Yes, these "facts" can be fudged. And sometimes are. But let's just assume for the sake of this discussion that earnings are earnings.

Comments

[milo Wolff:]

Seeking Alpha makes a logical error in its rebuttal to Morningstar. SA chooses only ONE example to make its case. The value of the Morningstar portfolio of ETFs is in the AVERAGE improvement of the whole set. Of course, some will go bad but hopefully most will be good.
Cherry-picking ONE bad one is illogical.

2008 Mar 13 03:28 PM

[Lisa:]

Morningstar is struggling because the focus of investors is shifting strongly to ETFs from mutual funds. But Morningstar has little to say about index ETFs, so it has to come up with something. And this is what they came up with. A reasonable effort, but ultimately unconvincing.

2008 Mar 13 04:11 PM

[Roger Nusbaum:]

Morningstar not getting it? Amen!

2008 Mar 13 06:39 PM

[E.D. Hart:]

I subscribe to Morningstar, but it comes with a bag of salt. And limitations. They make have consistently undervalued the energy and commodities sectors because they have, in the past, incredibly low expectations for oil, gold, natural gas, and silver prices. I would argue that this is another 'dimension of uncertainty'. If you followed their advice on gold and oil stocks you would have missed the rally of the last 7 years. In this regard they are followers not leaders.

With respect to commodities they overestimate the risk--and have very low star ratings for very high value stocks because their price expectations of commodities are entirely too low. They see a cycle when there is in fact a trend.

In the financial stocks the situation is the reverse--they have very high star ratings (For Citigroup for Example) and consistently have underestimated risk, and overestimated fair value. How can we know the fair value of Citigroup, if the extent of their balance sheet deterioration is unknown, and essentially currently unknowable?

It reminds me of Bill Miller--the past must be the future--right? Keep pouring money in Amazon. Or maybe, we are in a new inflationary era where the p/es will contract (on average) margins will compress, earnings will decelerate, and only companies that have protection against rising prices will thrive?

2008 Mar 13 07:25 PM

[Kevin S. Price:]

Milo--In noting that it was "too easy," I tried to acknowledge that the single instance of XLF should not be viewed as indicative of Morningstar's entire effort. But it is reflective of the hubris of the entire undertaking, one in which consumers of this information have to hope the whole (the ETF fair value estimate) is worth more than than the sum of its parts (the individual stock fair value estimates). I don't see it working that way--and I don't think this is a particularly useful of thinking about ETFs anyway. For the vast majority of investors, they should be part of long-term asset allocation strategies, not the sort of gun-slinging implicit in Morningstar's construct.

In any case, the argument of this post isn't contingent on what happens to XLF in the near term. I just threw it in to be a little extra snarky. That may sit better with some readers than others, but it doesn't invalidate the overarching logic of the post.

2008 Mar 13 09:19 PM

[Panskeptic:]

The whole notion of fair value is predicated not on solid numbers but their manipulation to fit a set of expectations.

These predictions may be more or less informed, but the whole process remains subjective, and shows that Morningstar still hasn't a clue how to survive into the next generation.

Cite one example or many, they are going down.

2008 Mar 13 10:50 PM

[Jan Rogozinski:]

For years, I was fooled by Morningstar. I didn't buy oil at $50 a barrel because Mstar "analysts" said it was grossly overpriced and would soon fall back to a more normal level of $15 a barrel.

over all, , there is something very wrong with their entire method of viewing risk. One example. There's a fund called Permanent Portfolio, which has never had a down or negative year since about 1980. And which over time has had (unlike savings accounts or CDs) positive returns much higher than than the inflation rate. . Nevertheless, they rate it as "very risky." That's just nuts. The reality is that the fund is safer than a CD or a US treasury bond. So how could it be risky.

I finally have decided that they confuse short term volatility with long term risk. So if a stock or fund fluctuates in value as it's return is rising, they think it is risky-- even when the overall trend always is higher, year after year. Meanwhile, they think a stock or fund is without risk if it goes along rising .00000000010% a year over time but without it's price fluctauating. in teh short term..

My advice. Use their abundance of data. But totally ignore their so-called analyses.

2008 Mar 13 10:53 PM

[tjschoenlein:]

Good comments.. for me, I " shop around " I cross reference with several sources, my brokerage house, morningstar, etc. to see how their data, analysis matches up and then add \ substract info... make a decision.
It works for me. *Please note, no one should pay any attention to me. I own Citi and most of the bank stocks.. + XHB.

2008 Mar 14 10:22 AM

[Indexor:]

I certainly hope those still using Morningstar are individual investors and not advisors. Morningstar's only value is that of a historical data provider on mutual funds. Nothing more and nothing less. Anyone relying on Morningstar for "fundamental analysis" is delusional. Anyone relying on Morningstar for their "novel approach to evaluating ETFs" is probably also doing their due diligence on a real estate or oil & gas limited partnership with a 7-1 write-off. Please refer back to the second sentence.

2008 Mar 14 11:40 AM

[300mph:]

I only need one question answered: How well have Morningstar's crack equity analysts performed in the past determining the fair value of equities?

2008 Mar 14 12:24 PM

[Brice Timmons:]

Doesn't it defeat a good deal of the purpose in investing in index ETF's to pay for fundamental analysis of the makeup of the index? I thought (and I'm no expert by any means) that one of the great benefits of buying into index funds was that because the fund sought to mimic the composition and movements of an index or composite of indexes that it saved immensely on management costs and allocated risk and exposure across the width and breadth of the market. Why then would I pay for the very services that passively managed funds save money by cutting?

2008 Mar 14 02:42 PM

[JohnB:]

Morningstar's self-analysis of their star rating system performance for stocks is at
news.morningstar.com/a...
Pretty candid. In summary, their stock ratings did poorly in 2007, but beat the S&P 500 in four of the past six years.

2008 Mar 14 03:54 PM

[Jeffrey Ptak:]

Hi Kevin.

A few corrections and clarifications:

- You indicated upthread that our approach assumes the whole (ETF fair value estimate) is worth more than the sum of its parts (individual stock fair value estimates). That's factually incorrect. An ETF's fair value estimate, under our approach, is worth no more than the sum of its parts—it’s simply a weighted-average based on the weighting of the holdings and their fair value estimates. The whole can potentially exceed the sum of the parts only from a risk standpoint (i.e., benefits of diversification make ETF less volatile than its component stocks).
- In your piece you suggested that the expected return that I cited for XLF covered a 52-week time horizon. Not so. Our stock, and ETF, ratings assume a three-year holding period. In other words, one would expect the convergence of price to fair value to take place over that three-year horizon. Meaning that your math is incorrect.
- You suggest upthread that the valuation-centric ETF research that we're conducting is less useful since it fails to help investors who are using ETFs as part of a long-term strategic allocation. Never mind that we're continuing to provide research that addresses an ETF's merits as a potential portfolio building block (i.e., are fees low? is it tax-efficient? does the portfolio construction make sense? has it tracked the benchmark? etc.).
- In the piece, you indicate that our approach doesn't adequately account for 'uncertainty'. Yet, our approach explicitly allows for the uncertainty inherent in our forecasts. It's called a 'margin of safety' and it's the discount to our fair value estimate that we demand before recommending a security. The more uncertain the forecasts underpinning a fair value estimate, the larger the margin of safety we demand before recommending a security, and vice versa. Our assessment of a security's uncertainty, in turn, is a function of its quality (intractable competitive advantages, like P&G, vs., say, some flaky biotech firm) and non-financial 'event risk' (i.e., big tobacco litigation, Vioxx settlements, etc.). It’s a very similar approach to the one popularized by Warren Buffett.

I hope this is useful.

Regards,

Jeff Ptak
Morningstar, Inc.

2008 Mar 17 01:19 PM

[Kell:]

Jan, I think the permanent portfolio is considered risky because they compare it to their fund category, most of which is invested in super safe stuff like treasuries. In contrast, the Permanent Portfolio tends to hold big positions in stuff like Gold and Silver, etc. These things tend to be very volatile.

You may make the argument that gold and silver are "not risky in the long run" in that they are good inflation hedges. In the last several years, you might be right... However, over the very long run, they have done a mediocre job at best. If you had bought gold in 1980 for $900, you would probably be kicking yourself now.

2008 Mar 23 12:58 PM