Recession-Testing Your Portfolio

There is correlation between economic phases and sector performance. Which funds will best ride the wave?

If you subscribe to common wisdom, the late-summer market selloff must have you thinking about the potential for a true recessionary slide. After all, the stock market looked like it had topped out after a stunning six-year run. Many investors and pundits subscribe to the notion that the stock market is a leading indicator of the economy’s direction.

True believers assume that certain market sectors will price in improvements some six to nine months before their fundamentals actually perk up. They use these signals to overweight favored industries while paring exposure to those segments most likely to falter.

The trick to this rotation business is, of course, correctly identifying the market’s current state. How do you, after all, spot ascendant sectors? And which ones fare best—or worst—at different points in the
economic cycle?

The second question’s pretty easy to answer. The first not so much, but we’ll get to that in a minute.

Research has shown pretty strong correlations between sector performance and economic phases. Sam Stovall, chief equity strategist at S&P Capital IQ Equity Research, famously mapped the relationship when he authored “The S&P Guide to Sector Investing” in 1995 (see chart).

If Stovall et al are to be believed, we could confirm a market top if we found bullish signals in the sectors that outperform during contractions, i.e., consumer staples (non-cyclicals), health care, utilities, financials and consumer discretionary (cyclicals).

So let’s get back to that first question. Just how do we spot sectors poised for liftoff? Fundamental analysis won’t avail us for a timing decision like this. Here, technical indicators and trend analysis work better. But what indicator? And what trend?

Sector Performance is Time-Dependent

It’s best to look at a mix of near-term and longer-term signals rather than relying upon a single marker. After all, sector performance is time-dependent: some industries will do better in the early phase of a recession, others in the midphase. You can then weight each indicator according to your confidence in its predictive power. There’s a wide range of indicators you can employ, but a half dozen or so seems ideal and nondilutive. Try these as examples: