Style Strategy: Why Sectors Resonate

By Jim Wiandt

What Matt Hougan says is true about my asset allocation but that doesn't mean it's set in stone.

I got a lot of email in response to my blogs about sectors and how we currently look at asset allocation. That tells me that this subject and my thinking resonates. And let me tell you, people can work themselves into a frenzy defending this perspective or that, even when it's peripheral to the sorts of issues I am looking at.

Here's my favorite quote from the emails (I won't use names unless you email me and would like me to add them in). My favorite quote was the beginning of an email that began:

Jim:

Let me help you, as "conventional wisdom" is indeed generally anathema to our individual and collective prosperity.

"When did sector investing become all about tactical investing, while style holds the asset allocation mantle?"

In spirit it was the same time it was discovered that 1 + 1 = 2.

You've got to love that. Though I'm not quite sure how the first statement works with the second part as style IS conventional wisdom, I would say.

Investing sometimes feels like religion. People get so worked up about their beliefs that they will defend them to the point of demagoguery even when common sense—and the evidence—may point in another direction. Still, you've got to love these guys (and I count myself as one of their number) around indexing getting all righteous and fired up. And I guarantee you that they're more around the truth, and are doing a better job of investing than 95% of everyone else.

So take a deep breath and come to your senses: it's all about sectors, baby!

Seriously, thanks for all of the emails and the insights. There's some other great material in emails you send me that I will try to bring into this discussion as well ... feel free to keep those coming. All of this is spurring a call to action ... in how I'm thinking about my own portfolio and the markets. Let me start here with the most substantive response by the original emailer who responded to my initial blog. He wrote this following Matt's last response:

Matt / Jim,

I appreciate the comments from both of you and the promotion of discussion.

I haven't yet reviewed content at the link you provided, Matt, but will do so shortly.

I did seriously consider incorporating global sector alternatives into the discussion. And the question of why I asked about U.S. sectors rather than global is a very, very reasonable one—and, I think, the discussion does need to go there. There are a couple of other significant wrinkles that one has to address when expanding it to the global level. Here are a few of them ...

iShares S&P global ETFs is the only complete set of global ETFs currently available (or am I wrong on this?—I know that one could cobble parts of an allocation from different sponsors and based on different sets of indexes). From that single, complete set ...

U.S. vs. non-U.S. split: A portfolio holding 10% of each sector piece would currently have an allocation which is 45% U.S., 55% non-U.S.; about 95% developed market, 5% emerging market;

Market capitalization: 91 to 100% large cap and 0 to 9% mid-cap (depending on how you look at the classification); and, not surprisingly, significant variance in country location of concentration across sectors.

Going down the global path, with only one fully global set of ETFs (iShares S&P), one has to also swallow the U.S./non-U.S. split, country-specific allocations and hugely large-cap orientation. Even then, many of the holdings in any sector ETF, whether U.S., global or otherwise, will have operations that extend beyond a single country's borders—making any country or regional constraint far less than pure.

Still, I would love to see S&P ex-US sector ETFs (and based upon other indexes) so that one could at least exert some level of allocation control over US vs non-US. For those who suggest that we have little need for more ETFs and that all the various asset buckets are sufficiently covered—I'd have to strongly disagree.

Regarding the use of market cap as determinant of country weighting within the global sector ETFs and, by extension, overall U.S./non-U.S. allocation—does nearly all large-cap with 45% US make sense for US investors? If that approach is valid for a US investor, then isn't it also valid for a resident of Spain to have the same portfolio and weightings? Or a resident of Singapore? I don't have an answer to that question but I do know that the global sector ETF removes any US vs non-US allocation flexibility—in my view likely leaving a U.S. investor a bit light on US holdings (heavy on international developed mkt), heavy on large cap (light on mid and small cap), light on emerging markets. I haven't yet dug into the individual global sector ETFs to have an opinion about country splits on the individual ETFs.

So, to frame the discussion on U.S. sector ETFs or global sector ETFs? Both need to be done but I think the global side of it can be a natural and more manageable extension of starting with the U.S.

All 3 of the tables below speak to your question, Matt, about global sector plays (and here he attached a bunch of nifty charts).

I'll end this blog and send you into the 4^th of July with just this one chart ... it's an amazing one representing the sector weights of the S&P 500 at different points in time. I know Matt thinks, with some justification that global gets you even closer on diversification by sectors, but I think the point is illustrated well here even with just the U.S. weightings:

	2005	2000	1995	1990	1985	1980	1975	1970
Consumer Discretionary	10.7	10.3	13.0	10.1	12.5	7.3	12.9	16.5
Consumer Staples	9.5	8.1	12.8	16.4	12.5	8.7	11.2	10.3
Energy	9.3	6.6	9.1	13.1	11.6	28.2	16.6	15.7
Financials	21.3	17.3	13.1	7.2	7.0	5.0	0.6	0.8
Health Care	13.4	14.4	10.8	10.3	6.9	8.0	6.9	4.8
Industrials	11.3	10.6	12.6	11.9	14.4	15.0	15.2	16.6
Information Technology	15.1	21.2	9.4	8.8	14.8	8.7	10.7	11.5
Materials	3.0	2.3	6.1	7.1	7.1	9.7	13.0	10.0
Telecommunications	3.0	5.5	8.5	2.0	1.8	3.8	5.7	5.9
Utilities	3.4	3.8	4.5	13.1	11.4	5.6	7.2	7.9

And here's the "now":

	12/31/2007	6/30/2008
Consumer Discretionary	8.23	7.88
Consumer Staples	10.26	10.81
Energy	12.9	16.25
Financials	17.63	14.28
Health Care	12.01	11.94
Industrials	11.55	11.15
Information Technology	16.81	16.46
Materials	3.34	3.91
Telecommunications Services	3.64	3.34
Utilities	3.63	3.99

Pretty cool, huh?

The question is what we make of all of this.

It's a big issue—and one that's been sort of nagging at me for years. Even as I sort my own core portfolio by growth and value (yes, it's true Matt—though I do NOT view my portfolio construction the way some people view the Bible or the Constitution, as inalienable truth), I am fully aware of the fact that what's going on is often actually determined by what is happening in the underlying sector. The question is whether the issues are cyclical or systemic, and I think it's a combination of both. But just intuitively, I've got to feel that sector is a more fundamental driver of returns than company valuation.

Anyway, I'd like to see us do research in that area ... and to encourage academia to look at it a bit more as well. Matt does link to one study that touches on the topic looking at attribution in his blog that follows mine. I think we could have some positive influence. Taking a look at that sector data above is incredibly interesting ... the question is whether that works FOR or AGAINST sectors as a core way to asset allocate. Are you looking for diversification of risk (I'd say so) or persistence of outperformance (the case that is sometimes made for small and value).

I think it's extremely interesting and a big, big topic that we could nudge forward. Somehow sectors got pigeonholed in the tactical zone, when perhaps they could be an equal or even a better way to look at asset allocation. But then maybe I should remember that 1+1 does equal 2.