Signaling corporate strategy in IPO communication: a study of biotechnology IPOs on the NASDAQ.

Signal clarity. Extant research finds that an effective signal should be observable and clear (Certo et al., 2001) so that market participants can easily capture the signal (Eliashberg & Robertson, 1988; Spence, 1973, 1974). This article proposes that IPO prospectuses allow IPO firms to present a corporate strategy type by stating their intention to favor one strategy type over another. If the IPO prospectus clearly states the firm's corporate strategy, the prospectus reader will quickly grasp and understand the intended corporate strategy (Heil & Robertson, 1991). Clear understanding of the firm's strategy will then encourage the investors or analysts to assess other strategic information, such as the structure of the firm and its adaptive processes (Baruch, 1992).

Signal intensity. Scholars suggest that frequently communicated signals are likely to capture the audience's attention (Riley, 1975; Spence, 1974, 2002). The audience will grasp the signals and subsequently use them to infer the firm's intended strategy type.

Signal consistency. Signal effectiveness also depends on signal consistency (Riley, 1975), which means that firms need to consistently communicate multiple signals from different dimensions and show little contradiction across dimensions. However, past IPO communication research has not investigated this aspect of signal quality; signals under investigation are typically simple, unidimensional, and focused (Certo et al., 2001; Megginson & Weiss, 1991; Neck et al., 2000). A clear and intense articulation of the firm's strategy type alone will not enhance strategy credibility; a firm also needs to show consistent evidence of a fit between strategy dimensions in order to establish strategy credibility (Miles & Snow, 1978). Therefore, this article further employs signal consistency across the dimensions of corporate strategy to examine the overall quality of corporate strategy signals-something not done in other IPO studies.

Having identified the characteristics of a credible market signal, how can a firm determine whether it has been successful in signaling corporate strategy from an IPO event? To date, a small number of studies apply signaling perspective to understand an IPO event and its performance (e.g., Carter et al., 1998; Certo et al., 2001; Megginson & Weiss, 1991; Neck et al., 2000; Rosenstein & Wyatt, 1997). Most of these studies examine the effect of IPO signals on the 1st day of initial returns (i.e., underpricing) as an indicator of whether or not market signaling has reduced information asymmetry between informed and uninformed investors (Carter et al., 1998; Certo et al., 2001; Megginson & Weiss, 1991). These studies justify the 1st-day returns based upon market efficiency theory, which suggests that the market responds immediately to information.

In this study, we work with both the 1st-day and 30-day market returns. We propose that the credible communication of corporate strategy will result in a very small gap between the offer and 1st-day price and, therefore, very little underpricing. We also propose that the three components we have identified as indicating a credible market strategy signal (clarity, intensity, and consistency) will have a negative impact on underpricing because they work to reduce information asymmetry between informed and uninformed investors. Accordingly, this study suggests the following research hypotheses:

Hypothesis 1a: The clarity of the firm's communication of its corporate strategy relates negatively to IPO underpricing on the 1st day of the IPO.

Hypothesis 1b: The intensity of the firm's communication of its corporate strategy relates negatively to IPO underpricing on the 1st day of the IPO.

Hypothesis 1c: The consistency of the firm's communication of its corporate strategy relates negatively to IPO underpricing on the 1st day of the IPO.

As previously mentioned, this study also includes the signaling effects of corporate strategy communication on returns 30 days after an IPO event for two reasons. First, opportunistic investors, who are more attracted by the potential for underpricing, might not be affected by the corporate strategy communicated in the IPO prospectus. This is especially the case for biotechnology IPOs where faddish purchasing behavior, due to overoptimism of the likely "abnormal return" from IPOs, is frequently present (Finkle & Lamb, 2002; Ritter, 1991, 1998). Second, analysts and investors may need time to read, understand, and evaluate the corporate strategy communicated in IPO prospectuses (Baruch, 1992; Bukh et al., 2002), so they will have a better understanding of the IPO value after a period of time. Therefore, the effective communication of corporate strategy might reduce 30-day initial returns by taking a slightly longer term view of underpricing. This article also investigates 30-day returns after the IPO event. This backdrop informs the following hypotheses:

Hypothesis 2a: The clarity of the firm's communication of its corporate strategy relates negatively to 30-day IPO returns.

Hypothesis 2b: The intensity of the firm's communication of its corporate strategy relates negatively to 30-day IPO returns.

Hypothesis 2c: The consistency of the firm's communication of its corporate strategy relates negatively to 30-day IPO returns.

[FIGURE 1 OMITTED]

METHOD

This article reports the findings of a single industry study: the biotechnology industry. The biotechnology industry is characterized by emerging, diverse, fast-moving, and hard-to-value products (Ranchhod, Gurau, & Lace, 2002; Weisenfeld-Schenk, 1994). Accordingly, these characteristics are likely to drive biotechnology IPO firms to employ signals in order to reduce uncertainty for investors. This research analyzes the prospectuses of 57 biotechnology IPOs, representing the entire population of biotechnology IPOs listed on the NASDAQ between 1997 and 2002.

Bhabra and Pettway (2003) find that prospectus information is more useful for predicting survival/failure compared with subsequent information. Thus, the vast majority of IPO signal studies rely on information available from prospectuses (Cohen & Dean, 2001; Daily et al., 2003; Neck et al., 2000).

The analysis includes two stages. The first stage applies content analysis, which allows us to test our research proposition that an IPO prospectus can be used to identify corporate strategy communication signals. The study applies a three-phase approach to content analysis. Phase 1 establishes 13 index variables using the Miles and Snow (1978) typology of strategy and develops operational definitions for these variables (see Table 1). This phase also includes the construction of a coding scheme (i.e., coding book, coding category table, and coding sheet). We provided a separate column in the coding sheet that allowed recording of a fourth, ambiguous signal class besides D (defender), A (analyzer), and P (prospector) for the three signal characteristics of clarity, intensity, and consistency. This coding allows for the three levels of class variables for each of the three characteristic types to be explicitly specified and tested for significance in a regression model. The fourth class, ambiguous, is appropriately confounded with the intercept, thereby eliminating unwanted linear dependence among the variables of interest.

Phase 2 includes a pilot study with two independent coders working on a subsample of IPO prospectuses to test the validity and reliability of the coding scheme developed in Phase 1. The first author of the article and an independent coder participated in the pilot coding process. The independent coder was experienced in the method of content analysis and familiar with the Miles and Snow strategy typology. The coders placed "+" on the coding sheet and entered this as "1" in the database when the coders identified an index variable from a case; otherwise, they used "-" and entered "0." Our research used the Perreault and Leigh (1989) index of reliability to test for intercoder reliability. In the first pretest, the reliability indices for the coding of two prospectuses were 62% and 73%, respectively. By comparing the coding notes, the disagreements were mainly attributed to ambiguous descriptions of operational definitions of index variables. Subsequently, the coding scheme was revised to improve the clarity of the definitions. A second pretest was then conducted to confirm the improvement in coder reliability. The reliability index was increased to 80%, which met recommended norms (Neuendorf, 2002; Perreault & Leigh, 1989). After establishing an acceptable reliability index, the coders worked independently (Perreault & Leigh, 1989).

Phase 3 applies the final coding scheme against the full sample of IPO firms. During the coding process, we focused on examining the communication of strategy in terms of quality of the content of strategic information. The coders gave particular attention to the standardized section of the IPO prospectus "business strategy" but also considered other parts of the prospectus related to strategic information. The coders searched for all relevant information associated with the Miles and Snow typology types throughout the prospectuses and filled out the coding sheet, referring to the coding book for guidance.

The second stage applies multiple regression analysis to examine the relationship between the quality of signals--clarity (CLA), intensity (IN), and consistency (CON)--and initial returns (IR) of the 1st day ([IR.sub.1]) and 30 days ([IR.sub.30]) after the IPO event. The initial regression model and description of measures follows: