Corporate earnings look pretty good, don't they?

Those of us who are paid to pay attention to what’s going on in financial markets have had a number of distractions recently—the ongoing “crisis,” if that’s the word, over Greece and other countries on the periphery of the European Union, and whether or not any of them will default (probably no); the question of whether many of the millions of jobs lost to the current recession in America and Europe are likely to come back (probably no); whether Peyton Manning is the greatest quarterback of all time (probably no, and who cares?). But our main distraction these days is corporate earnings. That’s because we’re supposed to know whether companies are making or losing money, but even more importantly, whether or not they’re doing better or worse than Wall Street analysts say they’re going to do. That’s because what markets respond to is often not actual performance, but performance versus expectations. And an awful lot of the happy talk in the media about how we’re pulling out of the recession is based on the “fact” that corporate earnings are looking pretty good.

So what we—and the markets, particularly equity markets—really pay attention to is “beats” versus “misses.” Once you know the lingo, you can vary these with adjectives—a “solid beat,” a “whopping miss,” that sort of thing. And the financial press is full of this sort of language four times a year, since earnings reporting occurs quarterly in the US (and increasingly so in the UK and Europe, although we still have a number of companies that still report only twice a year). So every couple of months we all gird ourselves for the next round of reality versus illusion.

For example, we’re now generally in the middle of earnings reports for the fourth quarter (which usually ends on 31 December for most companies). And the headlines have generally been pretty positive. That is to say, there have been lots of headlines about how earnings have been generally better than expected. Now, we only get so many companies each day, but there’s usually a cumulative impact—on market psychology, on economic expectations, and on analyst bonuses. And we’re only about halfway through reporting in the US, but we’re mostly done here in Europe. Now, it turns out that Bloomberg has a handy little function (EA) that tracks surprises in terms of EPS, net income, and sales. So if we focus on EPS, what the market generally focuses on, here’s what it looks like in Europe:

So what this is telling us (in the right hand column—there’s lots of other interesting information here, too) is that positive EPS surprises (beats) exceed negative EPS surprises (misses) by about 1.1 times—probably within the standard of error (although I haven’t checked this) for this sort of series. This seems reasonable—we’ve had 533 public companies report, 171 of them beat, 155 of them missed, and the balance matched analyst predictions. Actually, let’s think about that for a second—taking out the beats and misses, only 207 of 533 companies matched analyst predictions—or about 39%. So 61% either beat or missed. Hmmm.

How about in the US? Well, actually, things look a bit different:

Because here we see that positive surprises exceed negative surprises by a factor of 2.3X. Whatever the standard of error, this probably exceeds it. This is a much larger set, of course, since the US has many more public companies than Europe does. Just as interesting, perhaps, is the fact that of the 2,181 companies that have reported, only 659 matched analyst expectations, or about 31%. This doesn’t seem like a particularly good track record, frankly. European analysts only get it right 40% of the time, not particularly impressive. But US analysts are even worse—they’ve gotten it right 30% of the time so far this season. In fact, of the 2,181 companies that have reported, 1,058 (about 49%) generated a positive earnings surprise. Which means that US analysts under-predicted earnings by a factor of about a half.

I’ve actually been following this for awhile; if we look at third quarter 2009 earnings, we see a similar pattern—US companies generated twice as many positive surprises to negative ones, or about 2X; and US analysts only accurately predicted EPS earnings less than half the time (which means that predictions for the current period are well below even this unimpressive number). For Q3 earnings in Europe, this ratio is about 1.3X, and European analysts accurately predicted EPS about 60% of the time. And if we go back to the previous quarter, we find a similar pattern. So, to the extent we can generalize, we can say the following: analysts following European companies do a better job of (1) predicting earnings in general, and (2) avoiding understating earnings in particular, than do their US counterparts. If anything, US analysts have a negative bias that their European counterparts do not have.

So maybe we have this the wrong way around. Because there are two ways to look at this phenomenon. The first is how I’ve just described it—companies do better or worse than analyst predictions. But we could just as easily flip this around and ask—how are analysts doing? Especially since the market seems to put a pretty high premium on what they’re thinking. And these people generally get paid a lot, a whole lot. Maybe the question should be did the analyst get his or her call right, rather than did the company beat or miss the analysts’ predictions? Because if we look at earnings reporting in this light, it’s pretty clear that US analysts have been consistently underestimating how US companies actually perform. And in this regard, we have to ask whether or not US analysts even deserve that paycheck—because their miss rate seems pretty high. And if these guys show that they can’t get it right, why should they be paid a lot of money? And why should we (specifically, the financial press) listen to them?

The answer to the first is easy—because they generate trades for their trading desks, which is where the money is. That seems straightforward. The answer to the second question is a bit trickier, and it’s that I have no idea. Mister Market continues to put an inordinate emphasis on surprises. But since the analysts themselves seem to be surprised most of the time, why they continue to get lots of press and to show up chatting with Maria Bartiromo on CNBC remains something of a mystery.

Now, I haven’t done a more detailed analysis of this, having neither the time nor the point to prove, but that’s what needs to be done here. It seems to me that some bright empiric in business school somewhere, one who presumably is looking for a nice academic post rather than an investment banking job, might pursue such an analysis. At least then we might have a better sense of what’s going on. Because right now we don’t. For US analysts in particular, we seem to have to choose among a range of unpalatable choices—(1) analysts aren’t very good at what they do, or are just plain stupid; (2) analysts are good at what they do (since most of them seem to keep their jobs), but we don’t really understand what it is they do, and they’ve fooled us up to now: (3) since companies know that a positive EPS surprise will boost their share price, companies are doing a very good job of going out of their way to fool analysts, and since analysts are generally dim, analysts keep falling for it; or (4) it’s all one grand capitalist conspiracy. Take your pick.

Categories: Business/Finance

30 replies »

  1. Well, I can tell you it’s not #3. The market hates surprises, so companies try not to surprise anyone. As for the analysts, I can tell you from experience that they get their most important information from the companies, themselves. I’ve been in on those phone calls. Execs aren’t allowed to make actual predictions, but if an analyst throws out a number, an exec can say that it seems too high, too low, or in the ballpark. So, if the analysts are missing the number, it means the companies are missing the number, too. Why? I don’t know. No idea.

  2. Ah, but is it really true that the market hates surprises? It hates negative surprises. But it sure seems to like positive ones. Which is why I’m a bit suspicious when half the US companies that have reported so far this quarter have surprised on the upside.That just seems like a pretty large percentage. I’d like to see some sort of longitudinal study on this. I assume there is one. If I weren’t so lazy, I’d probably try to find it.

    • If you look at a long-run average, then half of all surprises should be positive surprises while half are negative. If a long-run average didn’t show that, I’d be suspicious. Short-run averages may be a different beast, however.

      It’s like the Dilbert cartoon about sick days – the pointy-haired boss complains that 40% of all sick days are on Mondays and Fridays. Divide 100% by 5 work days (20% per day) and add up the total for Friday and Monday and it’s supposed to be 40%. If it was 60%, THEN the PHB would have a reason to complain.

  3. Yes, the market hates positive surprises. The whole issue is to try to predict future earnings and the value of stocks based on those earnings, and make buying and selling decisions accordingly. If a company outperforms, you’re going to have people pissed because they sold the stock at an undervalued price, or because they shorted the stock, or used a hedging strategy based on predicted earnings that was undermined.

    I headed up the investor relations function for a Fortune 500 company. I’ve been the subject of tongue lashings from analysts that should have been reserved for our CFO because we underestimated quarterly earnings substantially. I was just the messenger boy, of course.

    The market doesn’t like surprises unless they just happened to get lucky with a surprise, but the market would really rather not depend on luck, thank you.

    • So it’s fair to say that they prefer accuracy (since that makes planning easier), but if there have to be surprises, good ones are better than bad ones? 🙂

  4. Yes, yes, there are times when investors don’t like the fact that analysts under-predicted earnings. Like most of the time, in fact. But those times are generally not after the worst economic dislocations in decades, and one of the more significant drops in market valuations in decades as well. Not not mention the 10%-20% drop in revenues in some sectors, and the comparable drops in profits, year over year. So in this environment, people seem to like positive surprises. Hence market performance last fall and this year up to the past week. I just find it awfully…convenient.

  5. Slammy and wufnik: Look, guys, I think you have it in your heads that the only investment strategy is to buy a stock, wait for it to go up, and then sell it before it goes down. If that were the case, then — sure — higher-than-predicted earnings would be a positive thing. But that’s not the way the market works. You can make money on stocks going down in price if you guess right.

    Now, if you’re an analyst, you are making recommendations on stock buys (in most cases). If you put a sell recommendation on a stock that then outperforms your estimates, you look bad. The people who sold the stock based on your incorrect guess about future earnings are pissed, because they know they should have held on. Large mutual funds may have sold some or all of a stock to take a position in another stock, and that upsets them if they should have held.

    The market does not, does not, does not like surprises. Sure, it’s possible to have a “good” surprise if you shorted a stock and it dropped more than predicted, or hedged with options in a stock that outperformed expectations. But, generally, the market likes to function the way Vegas functions. You win a few, you lose a few, but the odds are in your favor if you set it up that way. Surprises screw up the odds. You don’t like it when the odds become unpredictable (unless you know you can predict better than the next guy, but that’s rare).

    I don’t know why stocks have outperformed earnings estimates. Wufnik, thanks for pointing this phenomenon out. I was merely pointing out that no competent investor relations department likes to miss earnings projections, either up or down, because there’s always negative fallout from it.

    • I know about put options, I think they’re called, and understand that you can make money when the market goes down. But with respect to the argument wuf is making, it would seem like that’s a pretty small crowd. The company itself doesn’t make money that way, I wouldn’t imagine, nor do its shareholders, right? So that’s what I was thinking about.

  6. sorry, posted before I meant to. But I was going to say that I’ve been dealing with IR and CFOs and Treasurers for 25 years now, so I do know how this works. And during a normal environment, it may be true that the market doesn’t like positive surprises either (although I have to say that’s not my experience, but I’m not going to pursue it further). But the point is that even assuming this is true, the past three years have not been a normative environment, and the data, such as it is, suggests a very different story.

  7. So in the grand capitalist conspiracy scenario, what would the collective motivation be to underpredict corporate performance? Are you thinking that raising the general level of optimism is the goal? After all, the majority of viewers who watch those CNBC shows and hear the “surprise” soundbites are the inexpert and clueless, like me. Am I being not-so-subtly encouraged to pull out the AmEx and start spending like it’s 1986?

  8. Wuf, I didn’t know you were an analyst, so I apologize for treating you like a tyro.

    I can only speak from my experience. In the spring of 1983 or so (I think), my CFO told me that we were probably looking at $.53 per share or so for the quarter. Usually, that would be adjusted from $.50 to $.55 or so, depending on fluctuation in traffic for the rest of the quarter. Analysts at Alex Brown, Kidder Peabody, ML, and others were aware of this normal fluctuation.

    As usual, I got calls periodically from analysts at these firms (John Pinkavage at KP, Andy Kim at Alex Brown, etc.). I knew about how far I could go in giving them ballpark estimates and I stuck to that.

    Well, quarterly profits actually came in at $.68 per share, driven by unexpected volume in areas where we had excess capacity, so that revenue growth outstripped cost growth, giving us higher margins on the traffic from those areas than we ever could have suspected. And those guys were PISSED. I got my ear blistered pretty much all day, starting from about 15 minutes after I put the quarterly earning report on PR Newswire.

    So, that was my experience with surprising analysts. It wasn’t pleasant. And that’s all I have to go on, really, other than the further dressing down we all got at our Wall Street Club meeting with analysts later that year.


  9. Interesting discussion. I have noticed that they have been low balling earnings estimates myself. Frankly, I don’t care which way the market moves as long as it moves and I get a crack at either side. Frankly, the short side is the quicker way to make money(if you’re right) as stocks fall faster than they rise. Still, even being short is tough as the stock market is a rising tide and has been for the past 180 years. In fact, the stock market is the only market to show a distinct upward trend. However, it is still essential for the general public to buy at the top and sell at the bottom.. That plus the bid/ask and vig allows for those temples of high finance on Wall, Broad, Church, and Murray street to exist and prosper.. As for the options discussed earlier, 95% of the out of the money options expire worthless, and selling those options is a big part of my strategy. The best indicator for looking at a stock is to see how it reacts on news. If good news comes out and the stock yawns, it might be time to short the stock or buy a couple of puts while selling calls if you think it’s going down. Still, it’s impossible to predict the market farther than the next few ticks. Anyone who says they can is just a charlatan.. I, myself just trade and get rid of my losers right away and let my winners run.

  10. Ann–you know, I really don’t know what the grand conspiracy might be, other than to boost market psychology. I wouldn’t go so far as to say it’s coordinated, but I guess I wouldn’t be surprised if strategists and analysts all individually decided that the last thing they wanted to be last year was too optimistic–for much of the year, it looked like there was a whole lot that could still go wrong. (And it still does,for that matter.) I personally don’t remember a whole lot of bullishness on the equity side last year (although a number of credit strategists, particularly Morgan Stanley, were very bullish). After everyone got burned so badly in 2007 and 2008, maybe everyone collectively was being ultra cautious. That seems the most plausible explanation.

    One problem here is I really don’t know how novel this data is, since my Bloomberg will only let me analyze recent data (the past two quarters), although since I’ve been keeping track since Q1 of last year, I do know that analysts under-predicted pretty much all year. What I’d like to know is whether or not this is recent (ie, last year) or has it been going on for longer and we just haven’t noticed. On Monday I’m going to see if Bloomberg can get me data for earlier years. We’ll see,

    JS–my sympathies! Wow, Kidder and Alex Brown–you’re dating yourself!

  11. Wufink, there is no grand conspiracy in hese markets, as the market is the sum total of everyone’s opinion all distilled into a price. It is no wonder that the market has reboinded like it has, as markets have exhibited the exact same characteristics since the panic of 1812. I wasn’t aware that Bloomberg had a service that only allowed you to go back only a year….how cab you look at long term trends with only a year of data. FWIW, if you don’t wish to spring for the extra data costs, there are vendors out there that will sell you data at a fraction of the cost, after all it’s history. There’s also free open-close data on the internet, that I’ve seen on dailyspeculations where I write. One rule of thumb that has held up since 1812 is that you get bullish in equities after 40% declines and buy good stocks and hold them. I’m up 368% since March on my stocks that i bought during that debacle,. One way to make money is to ignore the analysts and what the crowd is doing and take the other side of the trade. The way I see it is that if brokers and analysts were any good, they’d be trading their own account and not wasting time working for a brokerage firm.

  12. Thanks for answering.

    Just to be clear, I know you were using the “conspiracy theory” option for rhetorical effect. We all know a behavior can be widespread and consistent without being coordinated, and your idea makes sense to me – after an upheaval, a reaction of increased caution and the intense desire not to be that wrong again.

    I guess in layperson terms, it would be the “hunker down” instinct.

  13. Jeff–thanks, but I think Bloomberg probably has the data, but I just need to request it. On the EA function I can go back to the previous quarter, and that’s it. But since I’ve been able to do that all year, I suspect if I just ask them, they’ll provide it. I don’t know how far it will go back, but we’ll find out.

    Ann, dead on. But we’ll see what the longer term data shows.

  14. I may be way out of my depth here, but it looks like a sports book to me. Basically everyone’s waiting to see if companies cover the spread.

    But i’m the wrong guy to declare on these sorts of things because i have problems with the foundation of rational-utilitarianism that the system’s built on in the first place. Markets can’t be rational if people aren’t, in the majority and the majority of the time, rational. Rational-utilitarianism is too narrow a philosophy to describe human behavior and it rests on several assumptions that are beyond questionable: a preponderance of evidence proves the initial assumptions false.

    And i agree that there doesn’t need to be a secret cabal pulling the strings for the “conspiracy” to operate effectively. The collective effort of human greed working individually is more than enough to produce the outcome that looks like a conspiracy.

  15. Lex, You’re 100% correct that the markets are a sports book. It’s finding the elusive overlays that are tough. Jeff