As 2016 begins, industry analysts and market strategists have made a number of projections for the S&P 500, including target prices and EPS estimates for individual companies in the index and for the index as a whole.
For 2016, the bottom-up EPS estimate (which reflects an aggregation of the EPS estimates for all of the companies in the index) on December 31 was $126.94. If $126.94 is the final number for the year, it will mark a new record-high EPS number for the index. However, what is the likelihood that $126.94 will be the final EPS value for the S&P 500 in 2016? How accurate is the bottom-up EPS estimate one year in advance?
Over the past 15 years, the average difference between the bottom-up EPS estimate at the beginning of the year and the final EPS number for that same year has been +7.5%. In other words, industry analysts on average have overestimated the final EPS number by about 7.5% one year in advance. Analysts overestimated the final value (i.e. the final value finished below the estimate) in ten of the fifteen years and underestimated the final value (i.e. the final value finished above the estimate) in the other five years. For the purposes of this analysis, the final EPS number for a year is the EPS number recorded two months after the end of each calendar year (February 28) to capture the actual annual EPS results reported by most companies during the fourth quarter earnings season.
However, this 7.5% average includes three years in which there were substantial differences between the bottom-up EPS estimate at the start of the year and the final EPS number: 2001 (+36.9%), 2008 (+37.8%), and 2009 (+29.0%). If these three years are excluded, the average difference between the bottom-up EPS estimate one year prior to the end of that year and the final EPS number for that year has only been +0.7% (over the past 15 years).
If one applies the overall average overestimation of 7.5% to the 2016 EPS estimate, the final value for 2016 would be $117.42. If one applies the average overestimation of 0.7% excluding the years 2001, 2008, and 2009, the final value for 2016 would be $126.00.
While the bottom-up EPS estimate for 2016 was $126.94 on December 31, the top-down mean EPS estimate (which reflects the average of the index-level EPS estimates submitted to FactSet by seven market strategists) for 2016 on that date was $126.88. Thus, bottom-up EPS estimate was just 0.05% higher than the top-down mean EPS estimate at the start of 2016. By how much do the industry analysts and the market strategists normally disagree at the start of a new year in terms of EPS projections? Is 0.05% an unusually narrow spread?
Over the past 10 years, the bottom-up EPS estimate has been 1.2% higher than the top down mean EPS estimate on average at the start of a year. Thus, the 0.05% spread for 2016 is well below the average of the past 10 years. In fact, it is the lowest spread during this period (2006-2015).
Which group has been more accurate in predicting the final (bottom-up) EPS number for each year? FactSet has historical estimates for both groups going back to 2006. This window includes the years 2008 and 2009, which were years in which both the industry analysts and the market strategists vastly overestimated the final EPS for these years, which skews the averages for both groups.
From 2006-2014, the industry analysts overestimated the final bottom-up EPS number in six years and underestimated the final bottom-up EPS number in three years. The average difference between the bottom-up EPS estimate at the start of the year and the final bottom-up EPS number was +8.2%.
Over this same period, the market strategists also overestimated the final bottom-up EPS number in six years and underestimated the final bottom-up EPS number in three years. The average difference between the top-down mean EPS estimate at the start of the year and final bottom-up EPS number was +7.7%.