Jeff Miller, one of my friends in the business and the excellent ‘Weighing the Week Ahead” and “Dash of Insight” blogger here in Chicago (Jeff’s blog was voted one of the Top 100 Financial blogs in America a few years back,) once asked me at one of our occasional lunches in Naperville, Illinois, where Jeff’s offices are located, “Have you ever looked at how the S&P 500’s “forward 4-quarter estimate” compares to actual S&P 500 earnings ?” and I responded I hadn’t but will do so, and then promptly forgot about the conversation.
With just a week of trading left in the 3rd quarter and basically all of the revisions done in for the coming 3rd quarter, 2017 S&P 500 earnings, we are kind of in a dead zone for data, or revisions, so I thought I’d take a stab and see how forward estimates “predict” the actual S&P 500 earnings:
- 2017: $128.57 was the forward 4-qtr estimate as of the last Friday of 2016, and the current full-year 2017 EPS estimate is $131.38
- 2016: $122.15 was the forward 4-qtr estimate as of the last Friday of 2015, and the actual 2016 EPS was $118.10
- 2015: $122.85 was the forward 4-qtr estimate as of the last Friday of 2014, and actual 2015 EPS was $117.46
- 2014: $116.94 was the forward 4-qtr estimate as of the last Friday of 2013, and actual 2014 S&P 500 EPS was $118.78
- 2013: $108.83 was the forward 4-qtr estimate on the last Friday of 2012, and actual 2013 S&P 500 EPS was $109.68
- 2012: $103.33 was the forward 4-qtr estimate on the last Friday of 2011, and actual 2012 S&P 500 EPS was $103.80
- 2011: $92.22 was the forward 4-qtr estimate on the last Friday of 2010, and actual 2011 S&P 500 EPS was $97.82
Source: the source of the majority of all our earnings data and tracking is Thomson Reuters I/B/E/S data.
The forward 4-quarter estimate is published every Friday night in Thomson’s “This Week in Earnings” and it has been tracked on a spreadsheet by me since the early 2000’s. Since an earnings update isnt typically published every day by Thomson, that last Friday of the calendar year becomes the defacto forward estimate used for tracking purposes.
Analysis: The forward estimate does a pretty good job of estimating S&P 500 in what might be considered “normal” years. For instance, in late 2007, the forward estimate for 2008 was looking for roughly $100 in EPS, and the actual 2008 EPS was closer to $65. Another example might be 2015: in late 2014, the S&P 500 forward estimate was looking for $122 in S&P 500 EPS and the final was $118.78 thanks to the collapse in crude oil, the strength of the dollar and the Chinese yuan devaluation that occurred in late summer, 2015.
Still those events only cost the S&P 500 approximately $5 in EPS in 2015, which tells you what a tremendous shock the Financial Crisis was to S&P 500 to cause a decline of roughly 35% from “estimated-to-actual” S&P 500 EPS.
What isn’t included in this discussion, and I’ll follow up with shortly is the importance of “market-cap” on the benchmarks and “P/E expansion / contraction”.
Before updating readers with the weekly S&P 500 data, here is a quick question for readers that I’ll provide the answer to early this coming week: The NASDAQ Composite declined 75% – 80% from its peak of March, 2000 at 5,312 to its low of 1,192.42 in late July, 2002. What was the actual change in S&P 500 earnings EPS for calendar, 2001 and 2002 ?
Thomson Reuters data (by the numbers, from TWIE dated 9/22/17):
- Fwd 4-Q estimate: $137.54
- P.E ratio: 18(x)
- PEG ratio: 1.8(x)
- S&P 500 earnings yield: 5.50%
- Year-over-year growth of forward estimate: +9.92% just a smidge shy of the late August ’17 high of 9.95%.
The good news this week was the jump in the year-over-year growth of the forward estimate after getting knocked down on Irma and Harvey in prior weeks. 9.92% expected growth is not bad.
Lots of food for discussion in the data this week.