Falling Knives: Do Stocks Really Drop 3 Times Faster Than They Rise?

| About: SPDR S&P (SPY)
This article is now exclusive for PRO subscribers.

Summary

It is a commonly accepted truth that stocks fall 3 times faster than they rise, a phenomenon usually attributed to trader psychology fearing loss of profit more over greed.

This article examines whether over the last 20 years the ETF SPY declines more rapidly than it rises, both during Black Swan Events and during typical market activity.

A buy-and-hold strategy is compared against a strategy with tight stops intended to protect profits.

Based on this data, trading strategies and protecting against market risk are evaluated and discussed.

It is a commonly accepted truth on Wall Street that stocks fall 3 times faster than they rise. The typical explanation borrows heavily from psychology and assumes fear of loss is greater than desire for gain. Investors will flee en masse when their profits are at risk, but are more timid when it comes to buying into a market. Short sellers who espouse these principles point to the great stock market crashes of the 21st century to illustrate their points.

Figure 1 below shows a chart of the SPDR S&P 500 Trust (NYSEARCA:SPY) from March 9, 2003 through March 9, 2009 when the stock market finally bottomed after the Great Recession.

Figure 1: SPY performance from March 2003 to March 2009 suggesting showing rate of decline > rate of advancement. [Source: Yahoo Finance]

From March 9, 2003 to October 16, 2007, a period of 1682 trading days, SPY rallied from $63.85 to $131.31, a gain of 106%. From October 16, 2007 to March 9, 2009, a period of just 510 trading days, SPY gave all of those gains back, and then some, falling to $59.90. Thus, it took SPY 1682 days to gain 67 points, but just 510--or less than one-third the time--to lose the same amount.

Fast-forwarding 2 years, Figure 2 below shows SPY from October 3, 2010 to October 3, 2011 during the period of the European Crisis.

Figure 2: SPY performance from October 2010 to October 2011 against showing the rate of decline to be greater than the rate of advancement. [Source: Yahoo Finance]

Again, we see that in the same period that it took to gain 22 points--277 days--it took just 88 days or 32% of the time to shed those gains.

These two charts suggest anecdotally that the market does appear to fall much more quickly than it goes up. However, Black Swan events, such as those discussed above, are rare, occurring once per 5-10 years. The stock market is typically much more orderly with rallies followed by small 5-10% pullbacks. Do stocks fall 3 times more rapidly than they rise during these periods as well? This article analyzes the S&P 500 over the last 22 years to determine whether outside of historic market crashes downward momentum outpaces upward momentum. SPY was used as the basis for the analysis presented in this article as it is the oldest index ETF and provides a broad assessment of the overall market holding a large basket of different stocks.

Figure 3 separates SPY into periods of rallies and periods of corrections over time frames ranging from a single day to 90 days from 1993 to 2015.

Figure 3: Comparison of 1-90 average of advancing periods versus declining periods of SPY from 1993 to present showing that while the rate of decline narrowly outpaces the rate of advancement for 1-50 day averages, thereafter over longer periods, SPY actually advances more rapidly than it pulls back. [Source: Yahoo Finance]

For SPY, average 1-day through 50-day losses exceeded average gains, but never by more than 10%. This divergence peaked after about three trading weeks with the 16-day average of all uptrending periods at +2.9% and the 16-day average on all downtrending periods at -3.3% for a decline:advance ratio of 1.12:1. Nonetheless, this is nowhere near the 3:1 ratio in question. As the time period broadens out, average advancements approach average decliners. The 50-day average of declining periods and advancing periods is identical at +/-5.5%. The average 90-day gains, on the other hand, are greater than the average losses at +8.2% vs -7.5%, a ratio of 0.91, suggesting that, over the longer term, SPY, on average, actually rises slightly faster than it falls.

But what about during times of increased volatility? Instead of the average change for advancing periods and declining periods, Figure 4 below shows the top 5% of advancing periods and the bottom 5% of declining periods.

Figure 4: Comparison of the top 5% of rallies in SPY against the top 5% largest pullbacks in SPY for 1993-Present showing that the rate of decline outpaces the rate of advancement, but falls well short of the commonly accepted rate of 3:1. [Source: Yahoo Finance]

During periods of increased volatility, SPY does indeed decline at a faster rate than it rallies. The top 5% of 90-day rallies was at 18.4% while the bottom 5% of 90-day declines was at -26.6%, a decline:advance ratio of 1.44. However, again, this is still nowhere near the 3:1 ratio discussed above.

Let's examine some other ETFs and individual stocks. Figure 5 below shows a table of the average 30-day gaining periods versus declining periods for a set of ETFs and equities. The 30-day period was chosen as it will catch most significant rallies and pullbacks, but is not so broad that pullbacks will be buried in an overall uptrend, or vice-versa, skewing the results.

Figure 5: Comparison of advances and pullbacks of several other stocks and ETFs which again shows that both rates are approximately the same, regardless of marketcap, sector, volatility, or trend in price performance.[Source: Yahoo Finance]

There appears to be no tendency for accelerated declines depending on sector. Technology [Apple (NASDAQ:AAPL), Microsoft (MSFT)], Energy [Chesapeake (NYSE:CHK)], Biotech [Gilead Sciences (NASDAQ:GILD)], Volatility (NYSEARCA:VXX), or Precious Metals (NYSEARCA:JNUG). Likewise, there does not seem to be a relationship between stocks that have been in a protracted downtrended versus stocks that have seen their prices steadily surge (GILD, AAPL) versus stocks that have traded flat during the periods examined. Likewise, more volatile stocks (JNUG, VXX) do not show a propensity to decline more rapidly than they rally. On the contrary, looking at the top 5% of rallies and corrections, both JNUG and VXX showed a marked tendency to rally faster than they decline. JNUG's top 5% of 30-day rallies averaged a spectacular 116% while its bottom 5% of 30-day declines averaged just -66%, a ratio of 0.57. Likewise, VXX averaged +76% versus -32% for a ratio of 0.41.

Based on this data, the SPY 500, as measured by SPY, does not fall more rapidly than it rises outside of the Black Swan events discussed in Figures 1 and 2. During normal market activity, the rate of rallies and pullbacks are approximately equal. What does this mean for the typical trader? One can argue that it is FOR Black Swan events that one needs to be cautious and protect profits.

However, a fear-driven, sell-first-and-ask-questions-later strategy can be devastating to a portfolio. Figure 6 below shows the results of a strategy that sells out of SPY after a 5% pullback from the peak and only buys back in once tranquil conditions have returned to the market and SPY is 2% above the sell price.

Figure 6: Comparison of a Buy-SPY-And-Hold strategy against a strategy that sells after the ETF falls 5% from its peak value and only rebuys after the stock normalizes and trades 2% above its sale price. The result is dramatic underperformance of the Sell-The-Fear Strategy. [Source: Yahoo Finance]

Since the ETF's inception in 1993, such a strategy would have returned only 216% versus a buy-and-hold strategy which would have returned 622%. Even during historic market corrections in the early 2000s after the Dot-Com bubble collapsed and 2008-2009 during the Financial Crisis, this strategy at best was equal to a buy-and-hold strategy. Otherwise, it underperformed dramatically. This underperformance does not even include losses associated with commissions from frequent buy-and-sells. Further, with multiyear gaps during the early 2000s and 2010s during which time the portfolio sat in cash, additional underperformance is added through loss of dividends, towards which SPY pays roughly 2% annually.

An effective trader has a healthy respect for the market and its ability to demolish a portfolio in short order, but also has the mental fortitude to not panic at the first sign of trouble. To quote the bounty hunter Greedo from Star Wars, "Jabba can't have a smuggler who dumps his cargo at the first sign of an Imperial Cruiser." An appetite for risk is an inherent part of investing and too-low an appetite can be just as costly as too great an appetite. Commission-hungry brokers may advise tight stops on positions, but this is only a recipe to sacrifice profits. If you can't take a 5% or a 10% loss on a position, you probably should not be in the stock in the first place. My advise is to use a "fundamental stop," that is, sell when fundamentals change. Yes, you may swing and miss and have a few large losses associated with "Black Swan" events, but I expect that you will more than compensate these losses by letter winners run.

In conclusion, my data suggests that during typical market activity, SPY does not fall 3 times faster than it rises. In fact, the rates of pullbacks and rallies from a 1-day average to a 90-day average are within 10% of each other. Only when looking at the peak +/-5% of largest rallies and pullbacks do corrections occur more rapidly than rallies, and even then no more than 1.5x more rapid. This trend holds true looking at index funds, high and low market cap stocks, uptrending and downtrending stocks, and highly volatile and minimally volatile stocks. Based on this data, I advise letting positions run. Keeping positions on too-short a short leash is a surefire way to sacrifice profits and will only rarely protect from serious losses. To be clear, I am not saying "don't have an exit strategy." On the contrary, the point of sale--for profit or loss--should be made before even buying the position and then modifying as necessary thereafter. Rather, I simply advise against selling a position too eagerly based on numbers and fear. You started that position for a reason--stick to it.

Disclosure: The author has no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.

The author wrote this article themselves, and it expresses their own opinions. The author is not receiving compensation for it (other than from Seeking Alpha). The author has no business relationship with any company whose stock is mentioned in this article.