Skip to Content
Rekenthaler Report

There's More to Expected Return Than Risk

Rethinking investment convention.

Embracing the Ugly
Early on, MBA investments textbooks have a chapter on risk and return. The authors quickly explain that higher expected returns are related to higher risk, for the common-sense reason that investors demand a benefit in exchange for accepting a drawback. The authors then spend the bulk of the chapter detailing "risk" and "risk/return" calculations that are based upon a security's price fluctuations.

That treatment is useful, but it is incomplete. What, after all, is this thing called risk? It certainly is not identical with volatility, despite the textbook's assumptions. For example, low-grade municipal bonds are less volatile than top-quality credits, because low-grade munis trade infrequently. Sticky prices might reduce standard-deviation estimates, but they don't make assets safer. They only make them mispriced.

So far, so familiar. Many previously have claimed that risk is more than just volatility. This column will take the argument further, however, by arguing that risk is more than just risk. The term is unnecessarily narrow; securities offer returns for reasons other than risk, as the word conventionally is used.

I didn't invent this idea either--well not alone.* In the upcoming Journal of Portfolio Management, Roger Ibbotson and Morningstar's Tom Idzorek issue a similar call to arms. They urge that the conventional notion of a low-risk/high-risk spectrum be replaced by a popularity spectrum. Popular securities are those that people tend to like and therefore are willing to hold even if the returns are relatively modest. Unpopular securities, on the other hand, must pay extra to become dance partners.

Ibbotson and Idzorek write, "Of course, risk is unpopular--investors do not like risk and want to be compensated for it." But … "[risk] is only one dimension of popularity. Popularity can include all sorts of other characteristics that do not fit well into the risk and return paradigm."

* (The timing stretches credulity. I sketched out this column on Thursday, was sent an advance copy of the Ibbotson/Idzorek article on Friday morning, and spent the weekend in bewilderment. As those two authors can attest, our thoughts on replacing the risk/return spectrum with a different scheme were arrived at fully independently. And released at almost identical times. Peculiar that.)

I hesitate to correct my betters, but popularity/unpopularity isn't quite it, either. Roger and Tom have the idea down perfectly, but to my mind not the terminology.

The reason is that a security might be "popular" in the Ibbotson/Idzorek sense of offering low expected future returns but unpopular as investment fashion. For example, during a stock bull market, when investors become enthusiastic about the prospects of equities and are less interested than usual in Treasuries, a five-year Treasury might offer a relatively high yield. That note would therefore be currently unpopular, but it would be popular according to the Ibbotson/Idzorek scheme because of the modest size of its expected return.

Relabeling the scheme as attractive/unattractive solves that problem. In each of four areas--more on that in a moment--a security will fall somewhere on the attractiveness scale, from beautiful to pretty to average to flawed to hideous. Improving investment returns, therefore, becomes a matter of identifying the acceptable forms of ugliness, and how much of each form to accept.

If this seems merely an exercise in semantics, it is not.

As an example, a recent paper on so-called "smart beta" (Morningstar uses the term "strategic beta," which is less loaded) claims that the excess return garnered from such strategies comes almost entirely from rebalancing, not from the inherent superiority of the tactic itself. The act of selling winners to buy losers, per this article, is the act of selling overpriced and overly popular securities and replacing them with underpriced and relatively unpopular securities.

That argument cannot be described by the conventional description of risk. If the authors' contention is true--a subject for an upcoming column--then the MBA framework fails. However, the contention can comfortably be addressed through the taxonomy of attractiveness. Selling winners to buy losers means swapping securities that are currently generating good stories for securities that are no fun to be around. It means, on occasion, loading up on Eastman Kodak, as that company makes its way to bankruptcy. It means shedding Apple throughout the aughts, time and time and time again.

That is embracing the ugly. It doesn't feel good. It leads to hard questions from annoyed shareholders. Owning securities that conjure up bad feelings is a legitimate reason for an investor to expect higher future returns. It doesn't appear as a risk in the MBA treatment, but it without question is unattractive.

The four sources of unattractiveness, as I see the matter, are:

1) Economic--The common notion of investment risk. The danger that a security might not be able to make required current payments, that these payments might not prove as valuable as expected (as caused, for example, by inflation), or that projected future payments might be less than anticipated. As Ibbotson and Idzorek write, people naturally do not seek these attributes and wish to be compensated for owning them.

2) Structure--How the security is packaged. Those that are cheap to own, that are taxed favorably, that may be readily traded, and that treat shareholders well (legally and otherwise) will be prettier than those that are not. Roger Ibbotson has written extensively of the return premium afforded to illiquid stocks. Accepting lower liquidity can be thought of as taking on more risk; but it's better yet to think of it as accepting an unattractive structure.

3) Restrictions--Barriers to ownership. These might be institutional, such as the claim that low-volatility stocks outperform high-volatility stocks because institutions generally cannot leverage. As a result, investors flock toward higher-volatility stocks (assuming more economic unattractiveness, in my scheme), thereby reducing those stocks' expected returns. Or the difficulty of trading across national borders, tax problems, legal and fiduciary concerns, and so forth. There are many ways that restrictions can affect security prices. They are not as the term commonly is used, investment risks.

4) Behavioral--The psychological aspects of ownership. The pain of holding unpopular securities--and the possible extra return associated with accepting that pain--has already been mentioned. Any investment tactic that seeks to profit from other investors' behavioral patterns, as caused by common psychological responses (that is, not caused by structure or restrictions, which would fit into sources number two and number three, respectively). These, too, cannot be called investment risks. 

Upcoming columns will trace through these four sources of return, outlining which investment tactics feed into them and which investor profiles are suited for each source. Think of the exercise as an alternative version of a risk questionnaire. Rather than determine how much extra return one can pursue by measuring tolerance for volatility, the attractiveness exercise prods the investor into accepting types and amounts of ugliness. 

John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.