Billionaire Warren Buffett summed up his investment philosophy with the mantra “Be greedy when others are fearful and fearful when others are greedy.”

And more than two centuries ago, Baron Rothschild, one of history’s most successful financiers, offered similar advice: “Buy when there’s blood in the streets, even if the blood is your own.”

Having the self-discipline to take a dispassionate view of the market is critical to putting these aphorisms into practice.

Investors tend to become overconfident near market tops and despondent near major lows. If you ignore the herd’s animal spirits and lean against the prevailing sentiment at key turning points, you will outperform the broader market over the long haul.

Of course, determining when fear has reached levels that make for a good buying opportunity or when greed has pushed the market to unsustainable highs is easier said than done. Investors tend to remember extremes such as the late 1990s tech bubble and the 2007-09 financial crisis; however, gauging the prevailing sentiment is much more difficult in a normal market environment.

There’s no magic indicator to gauge the level of greed and fear in the market. That being said, investors’ emotions tend to rise and fall with the business cycle.

The Purchasing Managers Index (PMI) for the US manufacturing sector, one of our favorite economic indicators, offers insight into where we are in the current economic cycle.

Interpreting this diffusion index is fairly simple: Readings above 50 imply an increase in economic activity, while values below 50 indicate a contraction. Historically, PMI readings below 47 have corresponded with US recessions.

We recently examined the monthly PMI data for the US manufacturing sector and the returns generated by the S&P 500, a total of about 650 months.

Over this period, the lowest recorded PMI reading (29.4) occurred in May 1982, near the bottom of the double-dip recession that occurred in the early 1980s. The index hit a record high of 72.1 in January 1975, when the US started to emerge from a vicious recession lasting from November 1973 to March 1975.

We primarily use PMI for a quick-and-dirty check on the US economy’s health. How else can PMI help investors?

Investors looking to improve their market timing—always a fraught enterprise—might assume that buying stocks when PMI reaches elevated levels and selling when the index slips below 50 would make the most sense.

After all, PMI usually surges above 55 when the US economy starts to emerge from recession and tends to sink below 47 when the risk of an economic downturn increases.

Digging into the data suggests that PMI provides the the best buy signals when used as contrarian indicatorprimarily at extreme values.

From December 1960 to February 2015, the S&P 500 posted an average rolling six-month return of 5.5 percent. Our graph shows the average six-month total returns earned by an investor who purchased the S&P 500 when the manufacturing PMI came in between various ranges.

For example, in the 25 months when PMI ranged from 35 to 40, the S&P 500 generated a mean six-month return of 16.1 percent—almost three times the index’s overall average over this almost 55-year period.

In contrast, over the 21 months when PMI came in between 63 and 69—and indication of a robust economic expansion—the S&P 500’s averaged a 5.3 percent loss over six-month holding periods.

PMI ranged between 47 and 59 for about 70 percent of the months in our sample; the S&P 500’s average six-month return in these instances came in near the average of 5.5 percent.

Extend the average holding period to 12 months and a similar pattern emerges.

Investors who bought the S&P 500 when PMI fell below 45 and held this position for a year would have generated average total returns that exceeded the 11.5 percent mean for the entire sample set. If you purchased the S&P 500 when PMI exceeded 57, your returns would have lagged.

This pattern also applies to a 24-month holding period.

(Click graph to enlarge.)

Bottom Line: Buying stocks when the PMI for the US manufacturing sector is low and selling when the index hits elevated levels generates superior returns.

Why does this contrarian approach work? The economy reverts to the mean over time—that is, prolonged economic contractions tend to be followed by periods of above-trend growth.

For example, the booms associated with the new economy of the late 1990s and the housing bubble of the mid-2000s were followed by huge busts in those overheated segments and downturns in the overall economy.

If you buy stocks when PMI values remain low, you likely placed your bets near the trough of an economic cycle, positioning yourself to harvest the fruits of recovery. And if you sell stocks when PMI reaches elevated levels, you pare your exposure when the economy is at its strongest and investors are very bullish.

These trends also interrelate with the Federal Reserve’s monetary policy. The central bank and the government usually seek to stimulate economic growth when PMI is low, and these efforts often give stocks a boost. However, PMI readings in the low 60s and higher signal that the economy could be overheating, increasing the risk of inflation and the Fed tightening monetary policy.  

You probably wonder where we stand in the economic cycle today. In August 2014, PMI for the manufacturing sector peaked at 58.1, a level that’s well below the buy signals from previous eras.

But swings in the index have moderated since the early 1980s. From 1960 to 1982, PMI for manufacturing exhibited a standard deviation of 8.14; since 1983, this measure of dispersion has decreased to 5.14.

This trend reflects a number of developments, including the economic volatility that prevailed during the stagflation years of the 1970s and a more positive market for stocks since the early 1980s.

Regardless of the reason, PMI levels that identify peaks and troughs in the business cycle have also narrowed.

The S&P 500 has generated an average six-month rolling return of about 6.5 percent since 1982. Over the 27 months in which PMI has ranged between 58 and 60, the S&P 500’s returns over the subsequent six months averaged about 3 percent.

The effect disappears after 12 months. Following a PMI reading in the 58 to 60 range, the S&P 500 has returned more 16 percent over the ensuing 12 months, a little above the average rolling one-year return since 1982.

In other words, when PMI climbs above 58, such a move often signals that the S&P 500 could be due for at least a short-term correction. Last year, PMI spiked just before the stock market suffered a 10 percent correction in September and October.

However, relatively high PMI readings have different implications depending on where we are in the economic cycle.

When PMI spikes into the high 50s later in an economic expansion, the sell signal becomes more compelling.

For example, PMI traded below 55 for most of the early 1990s before spiking to over 57 in spring 1994. The US did not suffer another recession in the mid-90s, but this acceleration in growth prompted the Federal Reserve to tap the breaks and hike the target interest rate to 6 percent from 3 percent between early 1994 and 1995. Meanwhile, the S&P 500 exhibited a great deal of volatility in 1994 and eked out a total return of 1.3 percent—a mid-cycle slowdown par excellence.

PMI also spiked above 57 in late 1999, before retreating to the upper 40s by summer 2000. Although the economy didn’t technically enter recession until March 2001, the bursting of the tech bubble sent the stock market sharply lower between 2000 and 2002. This economic strength also persuaded the Fed to raise interest rates from 4.75 percent in mid-1999 to 6.5 percent in June 2000.

In contrast, PMI readings in the upper 50s or low 60s early in an economic expansion often signal that there’s more upside to come. For example, following the end of the vicious double-dip recession of the early 1980s, PMI registered above 57 for 13 consecutive months and the S&P 500 continued to climb.

Similar patterns occurred in the early part of the stock market’s 2002-07 rally.

As we explained in How to Spot a Bear Market. several technical indicators that we monitor suggest that we’re late in this cycle.

To worse matters, the Federal Reserve appears to be contemplating at least a modest tightening in monetary policy this year. In anticipation of this announcement, the yield on the two-year US Treasury note has spiked to about 0.71 percent from 0.50 percent in January 2015.

Although the signal isn’t as clear cut as before 1982, last summer’s spike in PMI bears a resemblance to what transpired prior to previous down-cycles, suggesting that the S&P 500 could be overdue for a pullback.

I recently made A Free Video Report highlighting other warning signs on our radar screen and three simple steps you can take to protect your portfolio in the coming bear market.

Click here to see Elliott Gues 3 Warning Signs

Ready to discover your investing potential?