Markets

Nir Kaissar is a Bloomberg Gadfly columnist covering the markets. He is the founder of Unison Advisors, an asset management firm. He has worked as a lawyer at Sullivan & Cromwell and a consultant at Ernst & Young.

Investors are in a defensive mood.

As my Gadfly colleague Mike Regan recently pointed out, the U.S. stock market has eked out modest gains since the S&P 500 notched a record high in May 2015, but defensive sectors have experienced gangbuster growth since then.

The S&P 500 returned 5.7 percent from June 2015 to July 2016, including dividends, while the S&P 500 Consumer Staples Index and the S&P 500 Utilities Index -- two typical defensive sectors -- returned 15.7 percent and 22.7 percent, respectively.

Defensive sectors have historically given investors a smoother ride than the broader market. Consumer staples and utilities, for example, had the lowest standard deviation of the 10 S&P 500 sectors from October 1989 to July 2016, the longest period for which data is available. At the same time, financials and technology -- two classic boom-and-bust sectors -- had the highest standard deviation. (Standard deviation reflects the performance volatility of an investment; a lower standard deviation indicates a less bumpy ride.)

Defensive Stance
Traditional defensive sectors have historically enjoyed lower volatility than cyclical ones
Source: Bloomberg
Note: Volatility is annualized standard deviation for each of the S&P 500 sectors from October 1989 to July 2016.

That lower volatility isn’t just theoretical blather;  it actually gave investors some protection during the two bear markets since 1990. The correlation between the sectors’ historical volatility and their declines was a negative 0.59 during the dot-com bust and a negative 0.65 during the housing bust. In other words, the sectors with the highest volatility tended to suffer the deepest declines, and vice versa.

volitility-NIR-082916

You can’t blame investors for wanting to batten down the hatches. It has been more than seven years since the last bear market in the U.S., and stocks have had an epic run since then. The last two times investors saw this movie -- in the run-up to the dot-com bubble and again a few years later when housing prices ballooned -- it didn’t end well.   

But can defensive sectors still be counted on for protection, or is their newfound popularity setting them up for a fall?      

Investors’ enthusiasm for defensive sectors has stretched their valuation relative to cyclical ones. For example, the average normalized price-to-earnings ratio (using 10-year trailing average earnings) for the five least volatile sectors -- consumer staples, utilities, health care, industrials and consumer discretionary -- is 25. The average P/E ratio for the five most volatile sectors -- energy, telecom, materials, financials and technology -- is 21.

valuation-NIR-82917

It was just the opposite in the lead-up to the two previous bear markets. The five most volatile sectors had an average P/E ratio of 60 in early 2000, whereas the five least volatile sectors had an average P/E ratio of 38. The higher volatility sectors were also modestly more expensive in early 2007, with an average P/E ratio of 26 compared with an average P/E ratio of 25 for the lower volatility sectors.

So we’re in uncharted territory, but that doesn’t mean that defensive sectors will necessarily disappoint.

There was no reliable correlation between sector valuations just before the previous two bear markets and their subsequent performance. The correlation between the sectors’ P/E ratios in the beginning of 2000 and their declines from September 2000 to September 2002 was a negative 0.71. That would imply that high valuations lead to gut-wrenching declines. But if you remove technology -- whose absurdly high P/E ratio of 156 was destined to end very, very badly -- that correlation drops to a negative 0.21. In other words, hardly any correlation at all.

The result was similar the next time around. The correlation between the sectors’ P/E ratios in the beginning of 2007 and their declines from November 2007 to February 2009 was a positive 0.25. Again, little correlation to speak of.  

Those weak correlations meant that investors who were guided by valuations for safe passage during the last two bear markets were in for some surprises. Materials, for example, traded at a rich P/E ratio of 31 in early 2000 but gave up just 2.1 percent in the ensuing bear market on an annualized total-return basis. The financial sector, on the other hand, was the cheapest sector in early 2007, with a P/E ratio of just 16, but it gave up a heart-stopping 66.2 percent in the ensuing collapse.    

So sidestepping the worst of the next bear market may not be as easy as leaning on the cheapest or least volatile sectors.

This column does not necessarily reflect the opinion of Bloomberg LP and its owners.

To contact the author of this story:
Nir Kaissar in Washington at nkaissar1@bloomberg.net

To contact the editor responsible for this story:
Daniel Niemi at dniemi1@bloomberg.net