國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
8
CHAPTER 2 STRATEGIC SIMILARITY AND FIRM PERFORMANCE: REPLICATIONS OF DEEPHOUSE (1999)
Executives have frequently asked: ‘Should I behave differently from others or should I behave in the same way as peers?’ Over the past few decades, researchers in strategic management and organizational theory have attempted to answer this question by examining the relationship between strategic similarity and firm performance (e.g., Baum and Mezias, 1992; Baum and Singh, 1994; Norman, Artz, and Martinez, 2007).Institutional theory argues that being similar to obtain legitimacy (DiMaggio and Powell, 1983), whereas strategic view suggests to be different to develop competitive advantage (Baum and Mezias, 1992; Baum and Sign, 1994;
Carroll, 1985). Deephouse (1999) integrated these two conflicting perspectives and developed the strategic balance perspective. He proposed and empirically confirmed that firms can actively balance the two types of pressure (i.e., social norms and competition) and generate high performance by maintaining a moderate degree of strategic similarity.
Although the strategic balance perspective has been broadly accepted by researchers for over ten years, a lack of follow-up empirical evidence and the inconsistent results found in several studies have left the knowledge base in this stream of research on shaky ground (Hubbard, Vetter, and Little, 1998). Some follow-up studies supported Deephouse’s perspective whereas others did not. On the one hand, studies supporting the strategic balance perspective found that there is an inversely U-shaped relationship between the deviation of firms’ innovation strategies and their performance (Oerlemans and Meeus, 2005) and that firms with a moderate, rather than markedly higher or lower, degree of product differentiation have higher profitability (Robinson and McDougall, 2001). On the other hand, some studies
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
9
argued that balancing the two pressures does not necessarily create better performance.
One study found that a differentiation strategy has a positive effect and a confirmatory strategy has no effect on firm performance (Kale and Arditi, 2003); another study found a U-shaped relationship between the distinctive positioning of firms and their performance (Cennamo and Santalo, 2013). These few and mixed empirical evidences not only blur the validity of strategic balance perspective, but also evoke the necessity of systematic replications.
Replication studies play several important roles in scientific research and contribute to strategic balance perspective. First, replications serve as a guard against the uncritical assimilation of erroneous empirical results into the literature (Hubbard and Vetter, 1996). Because journal editors and reviewers often favor the publications of statistically significant (p<0.05) results, this, in turn, encourages researchers to produce such results. Taken to the extreme, journals may be filled with the 5 percent of studies that are Type I errors (erroneous rejections of the null hypothesis), while the remaining 95 percent with insignificant results gather dust in the file drawers of researchers which Rosenthal (1979) named as the ‘file drawer problem’. Second, replications contribute to verification and falsification of the theory (Singh, et al., 2003). When a replication supports the findings of the original study, it provides quantum leaps in credibility. If numerous replications provide repeatedly confirmatory evidences, it is highly likely that the theory has hit upon some real structure in the social world (Tsang and Kwan, 1999). Third, replications in different contexts such as country, time, or industry help to assess the generalizability and determining the scope and limits of original findings (Hubbard and Vetter, 1996;
1998). Fourth, replication studies advance a cumulative body of empirically supported research (Betties, Helfat, and Shaver, 2016). Because replication study builds on the
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
10
original study, they are not only repeatable but also cumulative. When each replication study provides additional evidence and knowledge, the accumulation of knowledge will facilitate theory development. Tsang and Kwan (1999) take vertical integration in transaction cost economics as an example and demonstrate the knowledge advance progress from offering evidence for the transactional determinants of vertical integration, examining the effects of asset specificity on make-or-buy decision, to confirming that specific human, rather than physical, assets play a more influential role in vertical integration decisions. Fifth, cumulative information obtained by multiple replications provides implications for future research such as exploring phenomenon-based empirical regularity (Bettis et al., 2016).
Regarding strategic balance perspective literature, some follow-up studies are not consistent, even contrary, to Deephouse’s findings and have raised concerns of its internal validity and generalizability (i.e. external validity). However, the scarcity of replications disallows us to reach a accumulative body of strategic balance perspective, which, in turn, constrains theory verification or falsification and determination of the scope and limits, Given that accumulated empirical studies are needed to consolidate and develop knowledge, examining the replicability of Deephouse’s study in multiple contexts is not only essential to strengthen the strategic balance perspective, but also is beneficial to explore the boundary conditions and contextual factors potentially affecting the validation of strategic balance perspective.
Furthermore, scholars in strategy and management are attaching importance to the reproducibility of empirical results and replication studies. Articles emphasizing the necessity of replication study are published in top journals such as Academy of Management Review (Tsang and Kwan, 1999), Strategic Management Journal
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
11
(Hubbard et al., 1998), and Journal of Management (Singh, et al., 2003). Recently, Strategic Management Journal has published a special issue for replications in 2016.
The Academy of Management had a specific session for replication studies in the annual meeting in 2015. Since the replicability is substantially important and the empirical results of strategic balance perspective is not solid, I replicate the seminal study of Deephouse (1999) using panel data for 57,644 firms from 2004 to 2007 in China and test the replicability of his findings across 155 industries.