What Causes A Bear Market?

Street sign image from Chris Potter and stockmonkeys.com.

Bear markets typically have an economic or monetary policy catalyst. David Rosenberg, Chief Economist & Strategist for Gluskin Sheff, summed up the things to look for during a transition from a bull market to a bear market as follows, via Business Insider:

“For stocks, it always comes down to the Fed and the economy. The reality is that bear markets do not just pop out of the air. They are caused by tight money, recessions, or both.”

2015: Recession And Rate Hike Odds

Since keeping an eye on the Fed and the odds of a recession can assist us from a risk-management perspective, we will examine both in this article.

Recessions: Taking A Look At History

The line on the graph below shows the probability of a U.S. recession (1994-2003). The shaded area shows the actual recession. Notice the probability of a recession, based on the St. Louis Fed’s model, was quite low between 1994 and 1999, a period when U.S. stocks performed very well. The probability of a recession began to rise noticeably in 2000.

The Recession Model 2000-2002 Bear Market

We prefer to use the Fed’s recession probability model as a “monitor and adjust” tool, rather than an “anticipate and hope” tool. If you were in the stock market in 2000-2003, you probably remember that most of the large and sustained losses in stocks did not begin until late September 2000. A useful technical warning came on October 6, 2000 when the S&P 500’s 200-day moving average rolled over (shown in blue below). The S&P 500 closed at 1,408 on October 6, 2000. Another warning came in the form of the Fed’s recession probability model via a reading of 22.68% in December 2000. On December 30, 2000 the S&P 500 closed at 1,320. Even if we had a data lag of 90 days with the Fed model, the S&P 500 was still trading at 1,160 on March 30, 2001.

How Helpful Was The Bearish Data?

The table below shows the three “we should be more careful with stocks” events described above. After the S&P 500’s 200-day moving average turned down, stocks lost an additional 45.41%. After the Fed’s recession probability model rose over 20%, the S&P 500 lost an additional 41.77%. Even if we assume a 90-day lag with the Fed’s model (data is released over time), the S&P 500 dropped an additional 33.74% between March 30, 2001 and the bear market intraday low made on October 10, 2002.

The Market’s Assessment Of Rate Hike Odds

If one of the common bear market triggers is “tight money”, it is prudent to examine the possible path of interest rates. This week’s stock market video covers numerous topics, including the current probabilities of a Fed rate hike before year-end, based on interest rate contracts (hard data).

Hard Data In The Financial Crisis Bear Market

A similar “was the observable evidence helpful” analysis can be performed using hard data from the 2007-2009 financial crisis period. With the exception of a move to 15.82% in 2005, the Fed’s recession probability model remained fairly tame between 2004 and 2006, a favorable period for equity investors.

Bearish Evidence 2007-2009

The Fed’s recession probability model jumped above 20% in January 2008, posting a reading of 28.56%. The S&P 500 closed at 1,378 on January 31, 2008. If we assume a 90-day data lag for the Fed’s model, the S&P 500 closed at 1,385 on April 30, 2008. From a technical perspective, the S&P 500’s 200-day moving average rolled over in a bearish manner on January 8, 2008 with the S&P 500 trading at 1,390.

How Helpful Was The Bearish Data?

The table below shows the three “we should be more careful with stocks” events described above in relation to the bear market low of 666 made on March 9, 2009. After the S&P 500’s 200-day moving average turned down, stocks lost an additional 52.09%. After the Fed’s recession probability model rose over 20%, the S&P 500 lost an additional 51.67%. Even if we assume a 90-day lag with the Fed’s model (data is released over time), the S&P 500 dropped an additional 51.91% between April 30, 2008 and the bear market intraday low made on March 9, 2009.

How Does The Hard Data Look In 2015?

The latest reading of the Fed’s recession probability model came in at 1.72%, based on hard data as of April 30, 2015. Therefore, given what we know today, the recession probability model is not screaming “the world is about to end” economically. The latest reading says there is a 98.28% chance the U.S. economy will continue to grow and avoid a recession.

How about the slope of the S&P 500’s 200-day moving average…is it shooting up bear market warning flares? No, as shown in the chart of the S&P 500 below, the slope of the 200-day moving average is still up, telling us the long-term trend in the stock market remains bullish, given what we know today.

How Can We Use All This?

Are these two inputs a foolproof way to manage stock market risk? Absolutely not…but they are helpful inputs to the weight of the evidence equation. If we use these inputs as “monitor and adjust” tools, they are both saying “try to remain patient with equities and the U.S. economy, given what we know today”.

How About Data Released August 3?

Since the Fed’s model works with a data lag, we can use the ISM Manufacturing Index figure released on Monday, August 3, 2015 as a present-day guidepost. As shown in the chart below, the ISM Manufacturing Index hovered near or below the manufacturing contraction demarcation line of 50 in late 2007 and during 2008.

The ISM manufacturing number released on Monday, August 3, 2015 came in at 52.6, a figure in line with expansion in both the manufacturing sector and broader U.S. economy. The current ISM figure also falls into the “try to be patient with stocks and the U.S. economy” category.

Forecasting Recessions

Regular readers know we are not big fans of forecasting, since it is a difficult exercise at best. However, we are fans of assessing short-term probabilities based on hard data or “knowns”, which is what our market model does. The keys are: (a) use facts in the model, (b) make short-to-intermediate term probabilistic forecasts, and (c) be flexible enough to adjust the probabilities/forecast when the facts (inputs) change. The approach is very similar to weather forecasting. It is also the approach used by the St. Louis branch of the Federal Reserve to assess U.S. recession probabilities.

A Recession Probability Model

Another thing we like in the St. Louis Fed’s description of their recession probability tool is the clear admission that predicting a recession is not an easy task:

“Predicting a recession in real time is difficult…FRED offers one of many such forecasts: a recession probability index computed by Marcelle Chauvet and Jeremy Piger. This forecast is backed up by research the authors have published in the peer-reviewed journals International Economic Review and the Journal of Business and Economic Statistics, with an early St. Louis Fed working paper added here for good measure…their forecasting method’s past performance is impressive; it predicted recession dates align well with the official NBER recession dates.”