There is an immediate positive and significant relationship between suppliers’ abnormal returns and customers’ surprise earnings announcements. Furthermore, there is a delayed response over the customers’ post-earnings drift period: The cumulative abnormal returns of suppliers continue to be positively related to the direction and magnitude of the customers’ earnings surprises for two months following the earnings announcement date.

The unexpected earnings announcements of key customers contain important information about economically linked suppliers. It appears, however, that investors pay limited attention to these important customer–supplier links, suggesting that investors underreact at the time of an earnings surprise, which leads to predictable future returns for the supplier.

In other words, key customers’ earnings information is not fully incorporated into the suppliers’ outlook at the time of the surprise even though future returns are not entirely the result of limited investor attention.

Active portfolio managers, as well as other practitioners, may consider using trading strategies to take advantage of investors’ limited attention by buying (shorting) stocks of suppliers whose key customers report positive (negative) earnings surprises.

The limited investor attention hypothesis states that investors’ limited attention to the arrival of new information causes return anomalies. The author tests the hypothesis for economically linked suppliers. Linking exists when a customer accounts for more than 10% of a supplier’s total sales.

To test the hypothesis, the author collects daily stock returns from the Center for Research in Securities Prices (CRSP) and quarterly earnings estimates from the Institutional Brokers’ Estimate System (I/B/E/S) on 1,083 unique customer–supplier relationships gathered from Compustat and covering the 1983–2011 time period. As such, many of the firms in the study are small.

Two cumulative abnormal return windows of a supplier’s returns—a 3-day window and a 60-day window—are regressed on the customer’s standardized unexpected earnings (SUE). Controls are placed for industry effects, year effects, and other variables. SUE is a ranked variable that essentially measures a customer’s actual reported earnings per share (EPS) minus the median of the analysts’ forecasted EPS. Regression results indicate a positive and significant relationship between suppliers’ abnormal returns and key customers’ earnings surprises surrounding and following the earnings announcement date.

The author performs robustness checks, including a test to demonstrate that the customer–supplier link is related to investor inattentiveness and is not simply a consequence of post–earnings announcement drift in customers’ stock returns. Although post-earnings drift contributes to the reaction, the author’s main conclusion is that investors are generally inattentive to customer–supplier links, which leads to abnormal supplier returns surrounding and following key customers’ earnings announcements. This result provides empirical support for the limited investor attention hypothesis.

This research is interesting, and the author provides very compelling evidence that suppliers’ returns are somewhat slow to react to key customers’ unexpected earnings news. It may be interesting to test this issue from another angle by examining the impact of suppliers’ earnings announcements on customers’ abnormal returns. For example, when the input being supplied is customized and not a commodity, the economic performance of a small supplier may be more closely linked to the economic performance of a much larger customer than expected.