How Good Is The S&P 500 'Forward 4-Quarter Estimate' At Predicting The Year-End Actual Estimate?

by: Brian Gilmartin, CFA

Reader "dmarilley" over on Seeking Alpha asked a good question about a month ago, but it has taken me some time to work through the backlog of articles and client meetings and tasks for the new year to properly address the question:

Dmarilley asked (and I'm paraphrasing): What was the Forward 4Q estimate at the start of the year versus the actual estimate at year end?

Great question, actually tremendous question. And here is some of the data:

  • 2016: F4Q est. as of Jan 1 '16 = $122.41, and with 80% of the S&P 500 having reported Q4 '16 earnings, the actual EPS estimate is $118.47. (That estimate will likely change over the next 5 weeks.)
  • 2015: Fwd 4Q est. as of 1/2/2015 = $126.80, and the actual S&P 500 EPS print for 2015 was $117.46.
  • 2014: Fwd 4Q est. as of 1/3/2014 was $120.77, and the actual 2014 EPS print for the S&P 500 for 2014 was $118.78.
  • 2013: Fwd 4Q est. of 1/3/2013 was $113.77, and the actual 2013 EPS print for the S&P 500 for 2013 was $109.68.
  • 2012: Fwd 4Q est. as of Jan '12 was $106, and the final print for the S&P 500 was $103.80.
  • 2011: Fwd 4Q est. as of Jan '11 was $95, and the final print for the S&P 500 was $97.82.
  • 2010: Fwd 4Q est. as of Jan '10 was $75, and the final print was $85.28.

Analysis / conclusion: With the exception of 2010, in all cases, the actual S&P 500 EPS print for the calendar year was lower than than Forward 4Q estimate at the beginning of the calendar year. In the case of 2015, the near $8 difference in EPS was the Energy sector drag and the substantially lower EPS from the Energy sector, which, at the time of the meltdown in crude oil, was almost 15% of the S&P 500. It is really 2013 that is the interesting year, since in 2013, the S&P 500 returned 32% on the calendar year and the Bernanke Taper Tantrum, and yet, the EPS estimate missed by roughly 5%. JPMorgan (NYSE:JPM) took a substantial charge for legal fees in Q3 '13 following the London Whale fiasco, and the banking industry was still under regulatory pressure at that time.

Again, can't thank Dmarilley enough for raising this question. Jeff Miller, the great blogger over at the Dash of Insight, and of NewArc Capital, has queried me many times about the "crossover" effect of earnings estimates. In other words, forward quarter estimates typically start out high, and usually get revised to too low a level just as the quarterly earnings start to get reported. Jeff has asked me many times where the "eventual actual" meets the forward estimate, and it is probably sometime in the last month, prior to a quarter's earnings release. Put another way, when Q1 '17 earnings start to get reported around April 10, 2017, and beyond, the actual growth rate for the sector will probably be a level that we saw sometime in March '17, all other things being equal. (Yes, these conversations get a little involved and nuanced.)

What will be different about 2017 is that proposed fiscal policy (and fiscal policies) could have a dramatic effect on the final 2017 S&P 500 EPS print, as this blog mentioned here.

The F4Q est. for 2017 started the year at $132.73, and it could be anywhere from $3 to $10 higher based on comprehensive tax reform, the border adjustment, and the Affordable Care repeal and replace. But we have to wait and see.

To answer the title question, though, I'd say the Forward 4Q estimate is a reasonably good predictor of the 12-month forward estimate, and that's just one opinion.

Thanks again for questions from readers, both readers of the blog and Seeking Alpha readers. And thank you for reading.