Sectors vs. Countries: Does Globalization Mean You Should Change Your Views on Diversification?


Flip through the pages of the Sunday paper's business section and you'll usually be able to pick out the latest fad in portfolio risk reduction - just by reading the ads. Lately, the steady stream of advertisements for utility funds, health-care funds, and tech funds - or "sector products" - has become a downpour. Sectors, it seems, are the new hot investments.

Now think back a few years, to a world with a different set of ads - those selling Japan funds, Asian investments, and anything labeled "emerging markets" as a way to diversify and thus minimize risk. Why the about-face?

Well, risk reduction is big business, and the changing ads mean the battle's raging again. In the latest salvo, Standard & Poor's (S&P) joined the chorus and issued a report asserting that traditional, country-based diversification makes little sense in our global economy. Rather, S&P says, what matters is industry. Splitting your investments among energy, banking, and tech stocks makes a lot more sense than thinking France, China, and Italy, the report asserts.

In fact, Ryan Carrier, the S&P report's author, put together a slew of charts and other data detailing rising correlations between countries over the past thirteen years. The sharpest changes have come since 1995.

"Markets are so linked and so global now," Carrier explains, echoing the familiar theme of a global economy. But here's the interesting twist - while correlations between countries got stronger, the correlations between industries got weaker. So European energy and European financials are actually less linked statistically than say, France and Germany as countries.

These numbers prompted four recent academic studies, which Carrier built on when compiling his data-filled report. Headlines beget more headlines, and recently, The New York Times quoted several analysts recommending a zero to five percent international exposure rate - a far cry from the standard twenty-percent suggested level, and low enough to shock lots of investing professionals.

Correlations make good, easy-to-explain copy for reporters, which is why you'll probably read more about the issue. Here's how The Economist summed it up, in a piece with the headline "Dancing in step" : "The correlation between changes in American and European share prices has risen from 0.4 in the mid-1990s to 0.8 last year. Crudely, that means that movements on Wall Street can explain 80% of price movements in Europe."

That's a rather scary statistic for people who've put their retirement dollars in foreign markets. For his part, Carrier says classifying investments by sector rather than countries leaves pension-plan managers "petrified," because of the massive cost of restructuring research and analysis procedures. He thinks we'll see small changes first, like fewer country-specific funds. And already, analysts like Richard Gussow, the Israel country analyst for Lehman Brothers, say they're seeing declines in single-country specialists. "I'm the last of a dying breed," Gussow says.

But what about individual investors, making personal decisions? Should they chuck countries and switch to sectors?

For starters, that might not be as fresh an idea as it sounds. Diversifying by sector may be old news to dedicated index-fund investors, who look at entire markets anyway. In fact, Carrier says, the new sector view may not be for everyone.

"If mom and pop are sitting at home and have just indexed the world, it doesn't matter whether it's sectors of countries," Carrier explains. "But if they take it a step further, they'd subdivide into countries and sectors."

"I don't think this is a tool for the unsophisticated masses to be using," he says, since the first audience for this stuff is pension-fund managers. "That being said, the unsophisticated investor is much more in tune with sectors that outperform than countries."

Part of that, of course, is the media's love of a good story.

"It's a media phenomenon from the standpoint of tech and telecom dramatically outpacing the market. Those sectors shot up like a mountain range, and anytime something shoots up past the norm, people take notice."

"Then it crumbles," Carrier continues. "And the natural reaction is to ask why."

Despite Carrier's enthusiasm for the global sector approach, and his statistics that support the view, he's quick to point that traditional country-based international diversification is not without merit - and it's definitely still around as a strategy.

"I don't think it's dead," he says. "The best correlations we see are still 0.6 or 0.7, which means there's still room for diversification."

There's still plenty of room, argues Steven Schoenfeld, managing director and head of international equity management at Barclays Global Investors.

"Tech, telecom, and energy are truly global industries," Schoenfeld says. "But to believe that real estate and utilities are global - that's just not understanding the fundamentals."

"Let's look at utilities in California, for example," he says. "You need to be aware of not only the country, but the state!"

The strength of the sector argument is just part of the tech boom, he says. "You see the total dominance of tech - it's globally correlated on the way up, and it will be on the way down. It's based on the tech bubble."

Schoenfeld says a truly long-term, historic view is essential. "Most recent studies are just that - recent, and therefore not relevant."

"While recent data and analysis have indicated that there has been an increasing dominance of sector diversification and an increasing dominance of the U.S. in the risk-return spectrum," Schoenfeld says, "a broader historic perspective has shown that sector performance and correlations experience cycles, as well as different short-term and long-term impact."

And there's something else to think about, which an investor can realize by reading ads, glancing at headlines, or reviewing a high-school history textbook.

"No single country dominates indefinitely," Schoenfeld says. "Is the 'prudent' investor really going to be protected from the next tulip-bulb mania as in Holland or
the Japanese bubble in the 80s?"

"Certainly, it doesn't appear that they were well protected from the Nasdaq and general tech boom-and-bust in the late 1990s and early 2000s. In addition, as long as markets are imperfectly correlated -- which means any correlation of less than 1.0 -- there are always benefits to be gained through risk reduction by diversifying across markets," Schoenfeld says.

And that, Schoenfeld says, won't happen anytime soon. "Markets will continue to be imperfectly correlated as long as there exist differences in fiscal and monetary policy, macro conditions, political uncertainty, varying levels of labor productivity, geographical advantages, and natural resource endowments."

Don't hold your breath for that, Schoenfeld says. "I do not see many of these
factors converging, except monetary policy with the Euro-zone."

Still, Schoenfeld concedes that you've got to take a look at sectors - but using it as a sole basis for analysis is not something he'd suggest.

"The point I would make is that you don't ignore sectors, but to say that the sector approach will totally replace countries is rearview mirror analysis at its finest," Schoenfeld says. "That mirror is smudged, and it's three cars away."