3 Ways to Predict Post-Earnings Shocks
04/19/2012 4:45 pm EST
Technical tools like Fibonacci, relative performance, and on-balance volume analysis can often fire early warning signals and keep stock investors safe from earnings-inspired meltdowns.
Corporate earnings season is generally a stressful time for investors and traders. Those holding company stocks before the earnings are released have to decide whether to take any action or simply stay with their long positions. Others try to determine whether they can earn a profit by buying a particular stock ahead of the company’s report.
Many have told me how they almost sold before a report, but ultimately didn’t, only to have their stock open sharply lower. Some also get convinced by a good fundamental story that the stock is turning around and they buy just before the report. If they fail to have a stop in place, they are then faced with another tough decision. Too many are either afraid to sell, or even worse, they lower their stop and ultimately end up with a larger loss.
There are three technical tools that I have found to be useful in warning that an upcoming earnings report may disappoint investors and that the stock could come under heavy selling pressure. Generally, at least two of these tools will warn you in advance that the outlook for a stock has weakened and that it is more likely to drop in reaction to the earnings report.
One tech stock that has disappointed investors frequently in the past two years has been Cisco Systems Inc. (CSCO). The pattern of earnings misses started in August 2010 when the company reported a 79% increase in earnings, but the sales numbers were lower than expected. The stock dropped 9% after the report, and by the end of the month (point 2), it was 17.5% lower.
For the next few months, CSCO was helped by positive spin from some fundamental analysts and rallied back to a high of $24.60 (point 3). This was just below August’s pre-earnings high of $24.87. More importantly, the rally in CSCO stopped just short of the 61.8% Fibonacci retracement resistance, as calculated from the April 2010 high (point 1) and the August low (point 2).
Fibonacci retracement analysis can be a very helpful tool for analyzing a stock ahead of an earnings report. If the stock has rallied but failed below the 50% or 61.8% retracement resistance, it gives little reason to initiate new buying. For those who are already long, the failure to move through a key retracement level is a sign that the stock has lost upward momentum.
The relative performance, or RS analysis, is often quite helpful because for stocks in well-defined downtrends, it would indicate the stock is weaker than the S&P 500 and therefore more vulnerable to selling pressure while those who are already long grow more disappointed by the underperformance.
The daily relative performance for CSCO had been in a downtrend since June 2010, as it had formed lower highs, line a, and lower lows, line b. The downtrend had been reinforced by the drop in August. The RS line was able to hold above its weighted moving average (WMA) until after the report was released.
CSCO closed on November 10, 2010 near the day’s high at $24.49 and opened the following day at $20.46 in reaction to the earnings report. This was a drop of 16.5%.
The on-balance volume (OBV) is the third tool that I find very useful, as it often leads prices lower and can reveal deterioration before the stock price drops. In the case of CSCO, the on-balance volume had broken an eight-month uptrend, line c, in June and had been in a shallow uptrend since the August lows. The August spike in volume was doubled in November when 553 million shares were traded. The negative impact of this heavy selling takes some time to overcome.
CSCO finally bottomed at $19.00 (point 4) and once again edged higher. Over the next few months, some analysts were convinced that the worst of the selling was over and that the company could even be a takeover candidate. By early February, CSCO had reached a high of $22.34.
This level corresponded to the 38.2% Fibonacci retracement resistance of the decline from the May 2010 high (point 1) and the December low (point 4). The rally had fallen just short of the 61.8% Fibonacci retracement resistance as calculated from the November high, point 3, and the December low, point 4.
The RS line had just regained about 50% of the drop from the November high but had stayed in its overall downtrend. The OBV also had an anemic rally, and it is interesting to note that (as was the case in November), volume picked up the day before the report.
In both instances, CSCO closed the day higher. This may have given investors a false sense of optimism, and on February 10, the stock gapped 10.8% lower when earnings were again worse than expected.
The earnings history of Oracle Corporation (ORCL) has been much different, as the company has a pattern of beating—or at least meeting—earnings estimates. Therefore, the miss on December 20, 2011 was quite a surprise considering the company had beaten estimates in each of the previous five quarters. ORCL bottomed in August at $24.72 and reached a high in October at $33.81, which was just below the resistance from the July high at $34.13 (line a).
In late November, ORCL had declined far enough to violate its uptrend, line b. The ensuing rebound in early December was not impressive, as both the RS and OBV analysis were raising warning flags.
The RS line had started a new downtrend by the end of November and was below its declining weighted moving average. On December 15, three days before the report, the RS line broke its uptrend, line c, suggesting that prices would soon follow. (Similar signals seen in big bank stocks in 2011 forecasted many of their declines as well.)
The OBV broke its short-term uptrend (line d) and dropped below its weighted moving average on November 17 and declined even further by the end of the month. It formed a pattern of lower lows in December, and like the RS, it was acting quite weak just three days ahead of the report. ORCL closed at $29.17, up 2%, the day before earnings were released. The following day, ORCL opened 12% lower at $25.67.
Best Buy Co. Inc. (BBY) has been disappointing investors with its earnings over the past two years. In November 2010, BBY made marginal new highs at $45.63 before breaking short-term support, line a, in early December.
The drop was confirmed by the RS analysis, which dropped through support, line b, and its weighted moving average. A similar formation was evident in the OBV, which violated support at line c. Both were declining sharply by December 7 (line 1).
Five days later, on December 14, BBY reported a decline in sales and lowered its full-year guidance. The stock opened at $35.59, which was down 14.6% from the prior day’s close, and over 60 million shares were traded.
A year later, things were not much different, and by October 4, 2011, BBY had dropped to a low of $21.79 after trading as high as $32.85 during June of that same year. The stock reached a high of $28.36 in November before correcting sharply and breaking its uptrend, line a.
The November highs were slightly exceeded in early December when BBY reached $28.52 but failed to move above $28.73, which was the 61.8% Fibonacci retracement resistance from the June highs.
The RS analysis had been in a shallow uptrend from the October lows and formed a short-term negative divergence, line b, just before the earnings report. The RS line plunged through support, line c, when BBY opened at $25.48 on December 13, down 9.2% from the prior day’s close.
The OBV did a better job of warning investors, as it broke support, line c, and its weighted moving average on November 21. The OBV failed to move above its weighted moving average (see circle e) and was dropping more sharply ahead of earnings. This suggested that long positions were being liquidated ahead of the report.
Hewlett-Packard Co (HPQ) has also disappointed investors since early 2011, as its stock price dropped almost 20% in February 2011. Even though HPQ beat estimates, the company’s guidance spooked the market.
HPQ eventually dropped as low as $37.60 in March before rebounding ahead of the May earnings report. The report was released a day early, as rumors had spread that it was another tough quarter for the company.
The relative performance had been declining (line c) while HPQ’s stock price was moving higher and had formed lower lows. The OBV stayed in a narrow range (see box) with a downside bias. It plunged after the report was released.
Another trading range (or continuation pattern) developed over the next three months, lines g and h, which was a sign of weakness. The rally was so weak that it failed to move above the short-term 50% Fibonacci retracement resistance from the early-May highs.
HPQ broke support (line h) in early August and the RS line dropped below its weighted moving average. The OBV had formed lower highs during the summer and had made new correction lows before price support was broken. From the late-July high at $37.70, HPQ dropped to a low of $22.75 on August 19 when the market punished the stock following another negative earnings report. From the July high, this was a decline of 39.6%.
Even tech titan Apple, Inc. (AAPL) can disappoint the market with earnings, although it doesn’t happen often. On October 17, 2011, APPL made a new rally high at $426.70, which was just above the September high of $421.59, line a. The following day (line 1), APPL closed higher for the day, but some short-term divergences were evident.
The RS line failed to confirm the new highs, line d, and had turned down. The divergence in the OBV was more pronounced, as it had already violated its uptrend, line e, in early October. As I have noted in past articles about on-balance volume, such breaks of OBV support can often be fairly reliable early warning signs.
The earnings miss on October 18 caught the market by surprise and was the first time Apple had missed revenue expectations since 2008. The stock opened 5% lower on August 19 and the RS line closed below its uptrend, line c. By November 25, AAPL had dropped as low as $363.32, which was down 14% from the October 18 close.
An important part of divergence analysis is that the divergence line also represents an important level of support or resistance. Watching the negative divergence resistance (line d) in the RS line provided a way to spot the improvement in the RS. In this case, the move above the bearish divergence resistance, line d, signaled that AAPL was again outperforming the S&P 500.
This confirmed the action of the OBV, and its downtrend, line f, was broken in early December, suggesting that the worst of the selling might be over. By December 23 (line 2), the OBV had clearly begun a new uptrend, and this was the start of the rally that has taken AAPL as high as $644 per share.
So, let’s review. If the stock you are analyzing ahead of an earnings report has been rallying, calculate the Fibonacci retracement resistance levels. Sometimes, as was the case with Cisco Systems, Inc. (CSCO), you will find that the minor and major retracement levels converge. If a stock has moved sideways and failed to move above the resistance level, it is signaling that supply may have caught up with demand.
Also, look at both the RS and OBV analysis to see whether they are moving higher with prices or diverging. If they are rising gradually and still holding above support, take a longer-term look to see whether there is a pattern of lower highs and lower lows. This will help to identify the major trend.
Pay attention to whether any support levels or uptrends in either the RS or OBV have been broken, as they will often precede a break of price support as well.
Lastly, be sure you note the earnings release dates for your stock holdings, and specify whether the reports will be issued before the opening or after the close. If you make your stocks pass these three tests prior to earnings reports, I think you will reduce the chances of being caught in an earnings downdraft.