Research says shorting ETFs in a 1X0/X0 portfolio holds unique benefits

130/30 07 Apr 2008

Hedge fund managers often contend that long-only managers lack the skills required to short-sell.  They will point to things like the fact that short positions will actually grow as they move against you (unlike long positions which shrink as they move against you).  They will also point to the fact that shorts tend to be driven more by catalysts than longs.  But one of the most legitimate concerns raised by hedge fund managers is the simple fact that good short ideas are often in short supply.

This was a concern raised last June in this MarketWatch article (see related posting, “So Much for Double Alpha”):

“Some equity hedge funds have quit short selling stocks because the strategy is riskier in a rising market and has become too crowded to be profitable. Instead, more managers are shorting exchange-traded funds. That’s a problem, according to some experts, who argue that using ETFs to hedge equity portfolios is a poor substitute for the real thing.”

“ETFs are also indexes, and so, by definition, they provide so-called beta — that is, the return generated by the market. Hedge-fund managers are in the business of creating alpha and outpacing the market benchmarks. So if they build short positions with ETFs, that part of their strategy will track whatever portion of the market they’re betting against. That could end up looking more like beta than alpha.”

While it may be true that hedge funds are in the business of creating alpha, 130/30 funds may be a little different.  Institutional investors, the early adopters of these hybrid strategies, seek alpha too.  But they do so with a watchful eye to volatility (known to them as “tracking error” vs. a benchmark index).  The key ratio for them is the “information ratio” (alpha/tracking error).

While shorting ETFs may not provide any alpha, it can match-off some of the exposures in the long book – thus reducing volatility while maintaining alpha and increasing the information ratio.

This is exactly the point made in a note to clients by Morgan Stanley’s Marty Leibowitz and Anthony Bova (available here at AllAboutAlpha.com with free registration).  The duo refers to ETF’s as “generics” and argues:

“In active 130/30 extensions, it can sometimes be difficult to find a full complement of active short positions. This paper presents a simplified model that illustrates how zero-alpha generics can provide funds needed for reinvestment, augment long alphas, and offset potential factor effects.”

What they say next has alpha/beta bifurcation written all over it:

“A generalization of the generic concept can be applied to any active portfolio by separating out an Activity component containing all positions whose primary purpose is alpha generation. The remaining positions then consist of basic market weights, fragmented overweights and underweights, as well as literal generic instruments…”

Their research note compares three variants on the short book in a 1X0/X0 fund: 1) “all generics”, 2) “all actives” and 3) 40/60 active passive (a scenario one might encounter if they simply ran out of short ideas).

Leibowitz and Bova plot the alpha and tracking error of these three scenarios on the chart below.  Note that the alpha over the tracking error is equivalent to the slope of the lines.  In other words, the steeper the line, the higher the information ratio.  As the chart illustrates, the “all generics” short book actually produced alpha because proceeds from short positions are reinvested in alpha-producing long positions.  As you might expect, the “all actives” short book produced the most alpha (the so-called “double-alpha”).  But what is striking is that the active/passive combination actually produces a steeper line than the all active short book.

As the authors say:

“…the use of offsetting generics allows for higher extension percentages — and higher alphas — to be obtained while staying within more clearly specified TE bounds….Such generic completions would prove particularly valuable in a highly TE-sensitive situation.”

While active managers (particularly quant managers) may never actually run out of active short ideas, it is quite possible that at some point their longs may have greater alpha potential than their shorts.  In fact, short extension enthusiasts may recall that this was central to the arguments made by Citibank’s Manolis Liodakis in a posting here last summer.

The moral of the story is that in world of scarce great shorting opportunities, shorting ETFs may not be that bad after all – as long as the information ratio is your key metric.

Be Sociable, Share!

2 Comments

  1. Bill aka NO DooDahs!
    April 7, 2008 at 8:01 pm

    You know, take ANY TWO equity strategies, call them “A” and “B”. Assume they each have a positive expectancy, and that they have a less than perfect correlation to each other.

    Now it’s pretty obvious that combining them would end up reducing volatility and probably increasing CAGR. This could be 50/50, 67/33, or any other split. Let’s say … 81.25/18.75! That’s “A” having 13/16ths weight.

    Now take this blended strategy, and lever it up 1.6 to 1.

    Gee, think it would outperform either one of its components (“A” or “B”)? Why would 130/30 be an exception to that logic?


Leave A Reply

← 130/30 in the 1930s New Study: No hedge fund bubble...but a potentially serious capacity constraint →