Earnings season opens this month, and the pattern will repeat several hundred times: a company reports results that beat expectations, and the shares fall. Or it misses, and they rise. To anyone treating the reported quarter as the news, this looks irrational. It is not.
A quarterly report contains two things. One is a record of a period that has already ended. The other is management’s statement about periods that have not. Only the second is genuinely new information, and it is where almost all of the share price reaction comes from.
Why the reported quarter is largely known
By the time a company reports, the quarter ended weeks ago. In the interim, analysts have modelled it using industry data, competitor results, channel checks and the company’s own guidance. Consensus estimates are the aggregation of that work.
The market does not trade on the reported figure. It trades on the difference between the reported figure and what was expected — and even that is compressed, because expectations themselves have been adjusted in the run-up. A widely anticipated strong quarter is priced before the announcement.
There is a further wrinkle. Companies guide analysts toward achievable numbers, so the consensus a company beats is frequently one it steered. A modest beat is the normal outcome and carries little information. The absence of one is the anomaly.
What guidance actually is
Guidance is management’s forecast of future performance, and it is a peculiar document. It is issued by people who have superior information about the business, strong incentives regarding how it is received, and — in most jurisdictions — considerable discretion over whether to issue it at all.
Reading it well means reading the incentives alongside the numbers.
- The range matters more than the midpoint. A wide range signals genuine uncertainty about demand or costs. A narrowing range as the year progresses is normal; a widening one mid-year is a real signal.
- Reaffirmed guidance is not neutral. If a company reaffirms full-year guidance after a strong quarter, it is implicitly guiding the remaining quarters down. The arithmetic is simple and it is regularly missed.
- Withdrawn guidance is the strongest negative signal available. Management has concluded it cannot forecast its own business. That is worse than a bad forecast.
- Newly introduced metrics deserve suspicion. A company that begins guiding on a measure it has never emphasised before is usually doing so because the established measure has stopped flattering it.
The quality of the beat
Two companies can beat by the same amount for entirely different reasons.
A beat driven by revenue above expectations indicates demand. A beat driven by cost control on in-line revenue indicates management competence but says nothing about the market. A beat driven by a lower tax rate, a one-off gain, or a favourable currency movement indicates essentially nothing about the operating business, and will not repeat.
The market distinguishes between these quickly, which is the usual explanation for a share price falling on a headline beat. The beat was low quality, and the guidance that accompanied it was weak.
Margin direction is the most compact test. Revenue growth with expanding margins indicates pricing power. Revenue growth with contracting margins indicates growth purchased through discounting or rising input costs — considerably less valuable, and often a leading indicator of a bad quarter ahead.
The call is where the information is
The press release is drafted by the company and reviewed by lawyers. The prepared remarks on the earnings call are scripted. The question-and-answer session is not, and it is the only part of the process where management responds to questions it did not choose.
Useful things to track:
- Questions that get answered with a different question’s answer. Analysts rarely press a second time; a deflected question is often the most informative moment on the call.
- Changes in emphasis between quarters. A metric discussed at length last quarter and omitted this quarter has usually deteriorated.
- Which analysts get called on. Operators work from a queue management can influence.
Adjusted figures
Nearly every company reports adjusted earnings alongside statutory figures, excluding items it characterises as non-recurring. Some of these adjustments are entirely legitimate — a genuine one-off restructuring does obscure underlying performance.
The test is recurrence. A restructuring charge in one year is an adjustment. A restructuring charge in each of five consecutive years is an operating cost being presented as an exception. Similarly, share-based compensation excluded from adjusted earnings is a real cost — it dilutes existing shareholders — and excluding it is a choice, not a correction.
The reliable defence is to check the cash flow statement, which records what actually moved rather than what was characterised. The gap between adjusted earnings and operating cash flow, tracked over several years, is one of the more informative numbers a company publishes without meaning to — a point we set out in reading a cash flow statement.
