Earnings surprises often generate rapid and large share price movements. The above example was a negative earnings surprise. Positive surprises, earnings that come in ahead of consensus expectations, can push share prices higher. Multex.com tracks earnings surprises over time and this information is presented to you in the Earnings Surprise table.

Why Do Surprises Matter?

To understand the dynamics underlying earnings surprise and share prices, it's necessary to review the development of corporate "investor relations." The phrase "investor relations" has several meanings. On the one hand, it can refer to mundane aspects of company-investor interaction for such purposes as disseminating company reports and news releases or advising shareholders about transfer agents, dividend reinvestment, earnings release dates, annual meetings etc. But we'll focus on the aspect of investor relations that involves communication between companies and the professional investment community (analysts and portfolio managers) for the purpose of helping the latter develop earnings estimates.

Strictly speaking, companies aren't required to have investor relations (i.e. company/analyst communications). Any NYSE, AMEX or NASDQ company would be well within its rights to do nothing more than file required documents such as the proxy statement, the annual 10-K report, and the quarterly 10-Q reports, and hold annual (and, where warranted by events, special) shareholder meetings. Twenty years ago, before the modern bull market got under way, many companies, including some that were large and well known, did little more than that. Companies that chose to communicate more openly varied in terms of who spoke for the company (often it was the Chief Financial Officer), how much information they shared with analysts and the extent to which they which they would comment on analyst earnings estimates. A few companies actually told analysts what they expected to earn, but most kept silent about specific numbers. Among the older generation of investors, earnings tallies that fell within 10%-15% of analyst estimates were considered to be within a reasonable range of "estimate error." Surprises above or below that approximate range did influence share prices, but to a far lesser and more gradual extent than is the case nowadays.

As the bull markets of the Eighties and Nineties progressed, institutional investors became more prominent in the equity markets and Wall Street research organizations grew larger to serve this increasingly powerful customer group. Many companies, especially bigger ones, found that CFOs could no longer devote as much time to investor relations as they had in the past. That led to the creation of separate departments dedicated solely to investor relations, and the growth of investor relations (IR) as a profession. CFOs and CEOs still enter the picture periodically to meet with analysts (one-on-one and in large group meetings), but the bulk of the day-to-day analyst/company communication is centered on the corporate IR staffs (this is less likely to be so for smaller corporations).

As increasingly sophisticated analysts spent more and more time communicating with increasingly talented IR professionals, it became harder and harder for companies to avoid discussing earnings estimates. The transition was a gradual one. In phase one, an analyst might ask IR: "Do you comment on earnings estimates?" The answer, more often than not, was "No." Eventually, analysts stopped asking and simply told IR what the estimates were, counting on IR to react in some fashion if the estimates seemed unreasonable. Over time, IR came to react to estimates by saying something like this: "You're within/above/below the range of other estimates we're seeing." That evolved to the point where IR also commented on whether or not management was comfortable with the range of analyst estimates. Today, IR has evolved to the point where companies will overtly guide analysts toward an estimate range which management believes to be reasonable and in line with internal corporate estimates. It is not uncommon nowadays for an individual analyst whose estimate lies outside the range to receive an unsolicited phone call from an IR professional who will try to "guide" the analyst toward the consensus range. The analyst need not comply. But most will, because corporate managers are presumed to possess better information about their own businesses and because the modern corporation tends to approach IR in an honest manner (in contrast to a generation ago, when companies were widely perceived to be painting as rosy a picture as possible). As companies participated more closely in the process of creating estimates, analysts became increasingly demanding about the information they desired and confident in the conclusions they drew. Estimates that were once "rounded" to the nearest nickel or dime are now set to the exact penny. Hence an analyst who, twenty years ago, would have been satisfied to estimate quarterly EPS within a 15% margin of error now expects to forecast the number to the penny.

In light of the historical development of IR, consider the changed meaning of an earnings report that falls shy of expectations. What might you think if you were the disappointed analyst who expected EPS of $0.50 only to be confronted with an actual number of $0.46?

• Twenty years ago, you might have thought: "I missed by only 8%. That's pretty good considering the lack of help I received from management."
  
  
• Nowadays, you may think: "I spent an hour with the CEO and CFO, and then two more hours with the IR Vice President. I started out looking for $0.48, but they kept talking about a $0.50-$0.53 consensus range, saying they were comfortable with it. So what went wrong? Do they have a proper handle on their markets? Are their internal controls functioning properly? If they were wrong about this quarter, what does that say about the next period... and the full year estimate... and next year's estimate?"

Is this fair, from the vantage point of corporate management? Absolutely not. No company exists in a vacuum. Every company's earnings are influenced by factors outside its control. And companies are no more able than analysts to predict many of these issues. Indeed, corporations that do extensive internal forecasting typically subscribe to the exact same econometric organizations as do the brokerage houses that employ the analysts! Companies are more able than analysts to assess internal issues, but when it comes to external factors, their forecasting ability is no greater than that of Wall Street (and in some cases, it might be argued that analysts are better able to forecast the externals because their firms may employ their own economists, something many companies do not do). But fair or unfair, the fact remains that earnings surprises today are perceived as company errors, not analyst errors. And the simultaneous receipt and reaction to the bad news by a large number of institutional investors sets off the selling waves that drive share prices sharply lower as soon as the news is disseminated.

In theory, the same set of reactions should occur when surprises are favorable. If one questions management's grasp of its own business when it reports $0.46 a share instead of the expected $0.50, couldn't one be equally leery of management's comprehension if the company reports $0.54 a share? Reasoning further, shouldn't the stock likewise fall in response to the favorable earnings surprise? But in fact, this doesn't happen. That's because the issue of company error exists along side of a long-standing tendency on the part of Wall Street to "extrapolate" the latest trend; to assume that the most recently reported trend will persist indefinitely. You see this all the time when economic data is released. At times, for example, even modest increases in factory utilization can send stocks sharply lower, as Wall Street assumes utilization will continue to increase, shortages will push prices upward, the inflationary price trend will lead to higher interest rates, etc. A drop in utilization the next time data is reported would reverse the process (as Wall Street anticipates a long trend of falling utilization, disinflationary pressures and economic sluggishness, lower interest rates, etc.) and send the stock market higher.

When corporations report negative earnings surprises, the tendency to extrapolate the weakness into the future aggravates the perception of company error and sparks a selloff in the stock. (Remember, stocks fell in response to negative earnings surprises long before IR came of age. The issue of company error merely increased the severity of a trend that had been in place many years ago.) When the surprise is favorable, euphoric reaction to the traditional perception that good news will persist indefinitely wins out over concerns about company error. This is not a logical balance. Nevertheless, it is the modern stock market culture.

How Should You React To Surprises?

The easy answer is to say you should be bullish on companies that tend to report positive surprises, and bearish on companies that that fall short of consensus estimates. And indeed, for many investors on many occasions, that is a correct answer. If you want to own stocks that are moving up right now, you will be hard pressed to find many choices among those that feature significant negative surprises over the course of several quarters. But before resting here, we need to offer two additional considerations:

• Temper the formula with common sense, since a surprise will be produced by a variation of a single penny. A gap between a $0.20 report and a $0.21 estimate would have one implication. A gap between $3.66 and $3.67 may be evaluated differently even though both are negative surprises. A penny shortfall below a consensus estimate of $0.76 is less likely to be deemed significant if the company guided to a range of, say $$0.75 to $0.80 as contrasted to the case if a company guided to, say $0.75-$0.77. The vigor with which management offers its guidance will be another factor. So, before you make an investment decision based on past earnings surprises, check price charts and make sure the surprises that occurred were actually perceived as significant by Wall Street.
  
  
• It is very hard to resist the temptation to follow the age-old Wall Street tendency to assume that latest trend will persist indefinitely. But that's not the way the world functions. Remember, too, the fundamental unfairness of quick assumptions that earnings surprises raise competence questions about management (they can, but that's a rare occurrence). So you may find that it pays to be sensitive to contrarian investment opportunities presented by earnings surprises. One example would be a stock that was pummeled to deeply depressed levels by a surprise that does not truly signal recurring trouble or management problems. Conversely a soaring stock price following a string of significantly favorable earnings surprises might signal unreasonably optimistic expectations that the company may be hard pressed to continue meeting.