684  Investor Communications rates are unpredictable, yet they can affect the profits of multinationals by 5 percent or more in a given year. Companies should therefore avoid predict- ing exchange rates and locking them into EPS targets. Rather, they should discuss their targets at constant currency rates. This would give investors a much clearer picture of expected performance. Meeting Consensus Earnings Forecasts Whether or not a company provides guidance, there will be an analyst consen- sus earnings forecast to meet or beat.15 The conventional wisdom, mistaken though it is, is that missing the consensus earnings forecast, even by a small amount, means that your share price will drop. A striking example: in early 2005, when eBay reported that it had missed the fourth-quarter 2004 consen- sus estimate by just one penny, its share price plunged 22 percent. Conversely, many executives believe that consistently beating the consensus leads to a premium share price. Thus, a common reason given for choosing to provide earnings guidance is to influence the consensus. Besides trying to influence the consensus, executives often go to some lengths to meet or beat consensus estimates—even acting in ways that could damage the longer-term health of the business. It’s not uncommon, for ex- ample, for companies to offer customers steep discounts in the final days of a reporting period in order to stoke sales numbers, in effect borrowing from the next quarter’s sales. As other researchers have shown, executives may forgo value-creating investments in favor of short-term results,16 or they might man- age earnings inappropriately to create the illusion of stability. Yet our analysis of large U.S. companies shows that these fears are un- founded.17 In the near term, falling short of consensus earnings estimates is seldom catastrophic. Even consistently beating or meeting consensus estimates over several years does not matter, once differences in companies’ growth and operating performance are considered. In fact, a company’s performance rela- tive to consensus earnings seems to matter only when the company consis- tently misses earnings estimates over several years. This doesn’t mean that companies should ignore consensus estimates, which can hint at what is on investors’ minds and why. For example, how does the industry growth outlook of investors compare with that of ex- ecutives? The consensus can also be used to assess how well analysts and 15 The section is adapted from T. Koller, R. Raj, and A. Saxena, “Avoiding the Consensus Earnings Trap,” McKinsey on Finance, no. 45 (Winter 2013). 16 J. R. Graham, C. Harvey, and S. Rajgopal, “Value Destruction and Financial Reporting Decisions,” Financial Analysts Journal 62 (2006): 27–39, which found that a majority of CFOs would “avoid initiating a positive NPV project if it meant falling short of the current quarter’s consensus earnings.” 17 This conclusion is based on analysis of the largest U.S.-based nonfinancial companies with a December 31 fiscal year-end, a sample of 266 companies.