Industry analysts have a long history of being more overly optimistic about earnings results the further those results are in the future.
On the other hand, they tend to lower their estimates in the weeks before upcoming earnings seasons, and more often than not are too pessimistic about the latest quarterly results.
It is the nature of industry analysts, I suppose.
They seem to be idealists about the future of their companies (why else follow them?), but are excessively hard-nosed realists about the here and now. Some may even be in cahoots with company managements, who guide estimates lower so that their actual results will be positive surprises.
That’s why Joe and I prefer to track forward earnings, which is a time-weighted average of analysts’ earnings expectations for the current year and the coming year. We do this on both a monthly basis and a weekly basis using data compiled by Thomson Reuters on analysts’ consensus expectations for S&P 500 operating earnings on a market-cap-weighted basis.
The monthly data start during January 1979 (Fig. 1 and Fig. 2). We update analysts’ consensus expectations for the current and coming year each month. The “Earnings Squiggles” are 25-month series for each year starting in February of the previous year and running through February of the following year. So the series for 2015 started in February 2014 and ended in February 2016, after the results were in for the final quarter of 2015.
Here are some observations on all of this and the implications for the outlooks for earnings and stock prices:
(1) Monthly Earnings Squiggles show long history of too much optimism. During the 36 years from 1980 through 2015, the endpoint of the Earnings Squiggles was below the starting point during 28 of those years by 18.5% on average. The only years with upside surprises were 1980, 1988, 1995, 2004-2006 and 2010-2011.
(2) Monthly forward earnings tends to be a good leading indicator of actual earnings most of the time. The Earnings Squiggles are based on data compiled by Thomson Reuters for S&P 500 operating earnings. Forward earnings data can be calculated on a monthly basis starting in 1979, while weekly data are available since April 1994.
Forward earnings tends to be a remarkably accurate predictor of actual earnings (available since 1986 as compiled by Thomson Reuters on the same basis) over the coming four quarters (Fig. 3 and Fig. 4). However, there is one important exception to this accuracy: Industry analysts don’t see recessions coming and so slash their estimates only as one unfolds.
(3) Forward earnings bullish again. Predicting recessions is my job. I don’t see one coming over the next 12 months, so forward earnings should be a good predictor of actual earnings over this period. On a weekly basis, forward earnings have stalled at record highs for the S&P 500/400/600 since mid-2014, when commodity prices started to fall sharply led by a dive in oil prices (Fig. 5).
In recent weeks, all three have moved higher into record territory (Fig. 6). Forward earnings for each of the three market-cap groups are rising to new highs along with forward revenues (Fig. 7). That’s all bullish stuff as long as there is no recession in sight.
(4) Market math adding up to continuation of secular bull market. As of the 7/21 week, the forward earnings per share for the S&P 500 was $127.39, which is a time-weighted average of this year’s latest consensus estimate ($118.22) and the estimate for next year ($134.67). Joe and I agree with the analysts about this year, but think they are too optimistic about next year, as usual. However, we are slightly raising our estimate from $126 to $129 (up 8.4% y/y) for next year, which means that forward earnings may not have a lot of upside over the rest of the year since at yearend it will match the consensus estimate for next year, which is likely to come down closer to our estimate.
However, once we get into next year, the 2018 estimate will start to get more weight as the year progresses. Industry analysts are currently forecasting $148.20 for 2018’s S&P 500 operating earnings. Our forecast--which theirs should drop towards in time--is $136.75. So as the market starts giving that number more weight as next year progresses, the S&P 500 could rise to 2325 by the end of 2017 assuming the multiple remains at its current level of 17.0 (Fig. 8). Last week, we raised our target range for next year from 2200-2300 to 2300-2400.
Dr. Ed Yardeni is the President of Yardeni Research, Inc., a provider of independent global investment strategy research. To read more of his blogs, CLICK HERE NOW.
© 2022 Newsmax Finance. All rights reserved.