This an excerpt from
the book Hedge
Fund Market Wizards. The bold font is Jack Schwager’s remarks and the non-bold
font is Jamie Mai’s comments. This isn’t exactly how it appeared in the book,
as certain paragraph breaks and italics were omitted as I highlighted things in
batches on my Kindle. I thought it was an interesting trade that is probably
still relevant today if one wanted some cheap upside exposure to the market
without having to risk too much capital, though you probably need a significant
amount of capital under management in order to have access to this kind of trade.
…
We have a trade on now that I really like. I don’t know if
you read Jeremy Grantham of GMO. He is a widely respected value investor who
looks across all asset classes and writes commentaries and editorials about
what he is seeing. For some time now, he has been arguing that high-quality,
consumer oriented franchises, particularly those that have great international
brands, are cheap relative to the rest of the S&P based on both dividend
yield and enterprise value to cash flow. In my view, he has laid out a fairly
compelling argument that places relative valuations in the context of a cycle,
wherein the low-quality names tend to outperform early in the cycle, and the
high-quality names tend to outperform toward the end of the cycle. There is an
index called the XLP, which is an index of U.S. consumer staple companies such
as Procter & Gamble, Coca-Cola, and Johnson & Johnson. If Grantham is
right, at some point we should see a revaluation of the stocks in this index.
I assume that in the
current cycle since the 2009 low, the XLP has gone up less than the S&P?
It has gone up a lot less. Initially, we considered buying
options on the XLP, which were relatively inexpensive. But Ben came up with a
much better way to structure the same trade idea based on the XLP’s low beta of
0.5 versus the S&P500.
One observation that we found particularly striking was that
despite the XLP’s low beta, since the start of the index at the end of 1998,
the net percentage changes in the XLP and the S&P over the entire period
were almost identical. The XLP was up less in the bull markets and down less in
the bear markets, but for the period as a whole, the net change was about the
same. Seeing that both indexes had approximately the same net change over a
long period—a period that included both the Internet boom and bust and the
credit boom and bust—makes the notion that the XLP has a beta of 0.5 versus the
S&P seem counterintuitive if applied to longer periods. In addition, we
thought that cash flow and dividend valuations implied the potential for a 25
percent revaluation of the XLP versus the S&P. We went to an exotic option
dealer and asked them to price an outperformance option that would be based on
the performance of the XLP versus the S&P. What is the single measure that
the dealer is going to use to price the odds that the XLP will outperform the
S&P?
The beta.
Right. So with the beta equal to only 0.5, the model price
for an outperformance option was very cheap. Translated into English, those
inputs are saying that the XLP and S&P are likely to move in the same
direction; however, the XLP will move only half as much as the S&P.
But if we had a down
market, then the lower beta would imply a higher probability of
outperforming—namely, it would imply that the XLP would go down less than the
S&P.
That’s a great point, and it is the reason why, to get the
option cheaply, we had to strike the option at the current spot price. So there
was a dual condition for the option to pay off: The XLP had to outperform the
S&P and the S&P had to be unchanged to higher. This was essentially a
conditional long beta position. It was conditional on the XLP outperforming the
S&P, and it was long beta because it could only pay off in an up market.
What made you think
the timing for the trade was right?
We didn’t have any conviction that the market was going
higher. We almost always want to have some long beta exposure, however, and by
making the option conditional on the XLP outperforming the S&P, we were
able to get beta exposure to the market extremely cheaply. When you own options,
you’re always fighting against the time decay. Figuring out how to make the
option premium cheaper is one way of mitigating that decay.
So the basic premise
is that beta is measured based on daily relative price changes, which can be a
very poor indicator of long-term relative price changes.
Right, a fact that is obvious if you look at a long-term
chart comparison of the XLP versus the S&P. Volatility is a terrible proxy
for measuring potential price change over longer intervals of time. For
example, if an asset price changes by a constant percentage each day, its
volatility will be zero. One of our strategies is called cheap sigma and is
predicated on the idea that markets sometimes trend and that volatility will
dramatically understate the potential price move of markets that trend.