The Persistence of Mythology about Earnings Estimates

A regular theme at "A Dash" is Wall Street Truthiness -- the things that many people believe whether there is evidence or not.

One such topic is the value of the analysts who produce research reports on the stocks of major corporations.  No matter what evidence is produced, there are some who just keep repeating the tired conventional wisdom.  Since these stories are once again in the news, let's take a closer look.

The Myth about Earnings Forecasts

The conventional wisdom is that "bottoms-up" analysts, those who follow companies most closely, are too optimistic about earnings forecasts.  Today there were two strong voices repeating this theme.

1. The Wall Street Journal noted that recent economic data reflected some weakness, but those following companies had not lowered earnings estimates.  Most of the article was all-out bearish.  The quotes and selective presentation do not portray an objective conclusion.  The author notes that pre-announcments have been negative, for example, without comparing this to any historical trend.  (Pre-annoucements are always bearish.  It is not the ratio, but the overall number that is significant).  He cites data on past inaccurate forecasts without saying whether they were too high or too low.  The only quotation in support of analyst forecasts is buried in the last paragraph of the article.  The featured quotations are of the "scare" variety.  As a multi-decade WSJ subscriber, I am disappointed with the quality and objectivity of this analysis.
2. Barry Ritholtz features the WSJ article and also some stale research from McKinsey, conclusions that I refuted in this original research from last October.  Here at "A Dash" we are big-time fans of Barry's blog as well as his more recent writing opportunities.  I have frequently featured his work, so this is really only a minor disagreement.  My impression is that his work would be even stronger if he would be a bit more open-minded when there is new information on a topic where he reached a conclusion on some past occasion.

A Question

With that background in mind, let me focus on an obvious question, drawn from my article:

As background, here are two elements of Wall Street Truthiness about earnings:

1. Companies and analysts are widly optimistic about forward earnings.
2. Companies lowball earnings expectations so that they can deliver an earnings "beat."

The astute readers of "A Dash" will immediately see that these widely-held beliefs seem to be inconsistent.  Many of those in the financial punditry happily advance both arguments -- but not in the same post.

There is an obvious explanation.  At some point there is a "crossover," a point where the forward estimate is accurate.  If both statements are true, we know that estimates are accurate at some point, so when does the crossover occur?

The evidence shows that bottoms-up analysts are pretty good at the one-year forecasting horizon.  I cannot find a better forecast from any other source.  Can you?  Can anyone else?  It is easy to throw stones without offering any alternatives.

Investment Conclusion

In my articles on this topic I have been careful both about the research and the conclusions from the data.  Critics of stock analysts should have the same burden.  In the Wall Street Journal article, for example, there is the suggestion that the " top down" strategists project earnings of only about $95 per share in the S&P 500.  Putting aside the difference in estimates, why should we believe the strategists?  They have an opinion about the current economic slowdown and are applying it without looking at the companies.  Meanwhile, many companies have reported that they do not see any slowdown and that the outlook is good.  As a group, the "strategists" have no particular economic expertise and no earnings track record.  The article does nothing to support their expertise, nor to explain why we should prefer them to the "bottoms-up" approach.

Even if you accept the $95 estimate from the strategists, that is only about 13 times earnings.  If the WSJ wants to help us invest, might they not have mentioned that this was the result of the more bearish estimate?

Turning to the Ritholtz article, Barry asserts (but does not prove) that analysts are wrong at economic turning points.  Let us assume that this is correct.  If so, it is a strength of the analyst approach.  As a long-time consumer of analyst reports I do not want them all to try to be amateur economists.  They would be subject to the same biases as the rest of the blogs and the media.  It is better -- much better -- for them to analyze the actual information about the companies they cover.  By following this approach I can find companies that are less affected by economic swings.

If someone thinks that the earnings estimates are going to be too high because of economic challenges, let them make their own forecast.  Then we could compare predictions and track records.

Meanwhile, analyst forecasts are very good on the one-year horizon unless there is an impending recession.  The chances of a recession are quite low, so I expect good earnings over the remainder of the year.

Comments

[oldprof]

wsm - Here is my interpretation, which I confess seems rather obvious. The researcher and the author conclude that analysts are getting worse. As evidence they offer three recessions as data points. The bottoms up analysts (the ones I find to be useful) missed by 25% the first two times and by 40% the last time. The last recession was much more severe than the first two, and included huge write downs in financial earnings. There was even a negative quarter.

So if you think that the next recession, whenever it comes, will be worse than the last, the estimates will probably be worse and the authors will have a fourth data point.

If (like most of us) you think that the next recession will be less severe, then this piece of research will quietly disappear, never to be mentioned again.

Meanwhile, it is a favorite sport to criticize analysts. The critics never seem to offer any alternative -- one that has a good track record of earnings forecasts.

Jeff

[oldprof]

wsm -- Apologist? Moi? No way! I pretty much ignore projected growth rates and definitely ignore buy and sell ratings from analysts.

I do think that the "bottoms up" analysts do the best job on one specific task: Forecasting earnings for the next twelve months.

I will comment further shortly, but first I am curious about what you think. In particular, do you think that the very interesting chart suggests that analysts are getting worse, as the author suggests? Does it prove that earnings estimates will be even less accurate in the next recession?

Thanks for bringing this up, and I'll give you a chance to answer before I do.

Jeff

[wsm]

Jeff - I know you have been an apologist for sell-side analysts, as per several of your posts over the years.

I was wondering what your reaction would be to this article:

http://www.ritholtz.com/blog/2011/09/earnings-recession-forecast-errors/

Thanks!

[oldprof]

Leigh -- I will follow your progress with great interest. The wisdom of crowds approach works well under some circumstances. One strength is that many of those in the crowd figure out which experts are worth following and take cues.

While I do not really agree with your premise about analyst motives, I strongly support the idea of having many approaches with different methods.

Thanks for alerting us, and I hope you will send links when ready.

Jeff

[LDrogen]

Jeff, we're currently building Estimize specifically for this purpose. The old regime of sell side analyst estimates is broken, we all know that their incentive structure is not aligned to giving the most accurate estimates. By sourcing estimates from the entire market, and yes that does mean "amateurs" as well, we can get a better picture of what the entire market is expecting, and most likely through the wisdom of crowds produce more accurate estimates.

I'll be very interested in your feedback when we launch shortly, we're building this for the social finance community and want to tailor it to fit that group.

Leigh