Recently I've been looking for a reason to reduce my exposure to securities and increase my cash reserves. I've been searching for a signal of a future stock market correction to confirm my suspicion that the U.S. stock market is overbought and frothy. It's commonplace for analysts (and other humans for that matter) to look for and find data that confirms their suspicions, this is known as confirmation bias. I try to remind myself of the potential of such a bias when I start to stretch an indication or signal that will confirm my forecast. The market has been consolidating in the past few months, with a breakout to all time highs in the last few weeks. The breakout proved that my gut instinct was wrong and further reinforced my understanding that it's difficult to time the stock market.
However, today I noticed a signal that could be used to support the hypothesis that the market will correct in the near future. This signal was the bearish divergence between the price of the S&P 500 (SPY) and it's momentum (which is measured by the Moving Average Convergence Divergence (MACD) indicator). The bearish divergence is illustrated by the orange lines in the image below. Notice how price is increasing while momentum is decreasing over time.
It's wise to be skeptical of this bearish divergence signal due to the fact that many strong market up-trends include long periods of bearish divergences. This is due to the market's tendency to move higher even as it's momentum slows. This common occurance is illustrated in the following chart. As you can see, the MACD is signaling that momentum is slowing; a change in trend could occur in the near future. However the market continues to move significantly higher for a year.
So that begs the question, "has my suspicion been confirmed"? Nope, not even close. In the short-run, prices could continue to slide until they (1) find support on prior price resistance or (2) fall through prior support and start a new trend lower. Only then would my suspicion be confirmed.
It should be noted that any reliance on this indicator over the long-term would prove less than optimal. As the case with any single indicator, using a bearish divergence signal should be utilized in combination with other indicators to help in the confirmation of a change in trend. The value that this indicator provides is that it can warn of a change in trend in the near future. At that time other technical analysis techniques such as breaks in prior support level should be used as a signal confirmation.