Revisiting Entrepreneurial Orientation and its Contributions to Business Performance: An Industry Type Comparison employing Computer-Aided Text Analysis under Consideration of Configurational, Contingency, Environmental, and Temporal Aspects

Abstract

A firm’s entrepreneurial orientation (EO) refers to a firm-level strategic orientation that reflects its strategic choices, managerial styles, and organisational behaviours that are entrepreneurial in their basis. The majority of previous studies on a firm’s EO investigate its three most common characteristics – innovativeness, risk-taking, and proactiveness – attempting to measure and analyse their effects on business performance on a unidimensional basis while claiming a generally and overall positive impact. However, this approach is different from Lumpkin and Dess’ (1996) superior development of the conceptualisation of EO as being driven by five (not three) dimensions (they added autonomy and competitive aggressiveness). These five dimensions were conceived to vary on an independent basis, each potentially relating differently to various firm performance measures (such as sales growth, gross-profit-margin, market share, and return on assets), while being determined by both internal and external factors. Consequently, even though Lumpkin and Dess’ (1996) EO theory has rarely been previously considered empirically in the literature on the subject, it has presented a more plausible development of the conceptualisation of EO, making it highly relevant to the current entrepreneurial research. Therefore, this thesis employs the five-dimensional approach with the aim to investigate four research questions: (1) whether and how a firm can achieve an ideal profile of EO dimensions and the manner in which this fit may vary across industrial contexts, (2) whether and which dimensions may be more beneficial towards the contingency of firm performance as opposed to their counterparts when considering factors such as different industry types (high-tech versus less-tech intensive firms) as well as (3) environmental conditions (industry turbulence and munificence), and, ultimately, (4) whether the effects of EO may last longer than their initial investment period.

In brief, the proposed hypotheses were tested across a sample of US companies drawn from the Standard & Poor 500 that were selected to provide a relatively equal representation of high-technology and less-technology intensive companies, as determined by their industry types. This study pioneers a new research approach by examining the levels of the five EO dimensions through computer-aided text analysis along with a set of keywords advanced from Short et al.’s (2009) paper to extract values from the letters to shareholders and 10-K filings in the firms’ annual reports. Performance indicators and information related to the moderator and control variables were sourced from COMPUSTAT.

In describing an EO’s contextuality regarding configurational, contingency, environmental, and temporal aspects, this thesis contributes to the current knowledge of EO in the following ways.

Firstly, relating to research question 1, this study found that EO is associated with high performance in the set of ideal profile firms whereas deviance is associated with mediocre outcomes in the remaining group. Inconsistencies in the EO-performance linkage, therefore, are perceived to be driven by a poorer configuration of the EO multi-dimensions. Furthermore, it was examined to what extent the configuration associated with optimal performance remains the same across both the industry types. Herein, it was discovered that the ideal profiles do not differ across the two industry types of high-tech and less-tech.

Secondly, relating to research question 2, within the context of this study, it was discovered that EO is, in fact, to be conceived as a multi-dimensional construct comprising of five dimensions as each has either a positive or a negative impact on individual performance measures (here under consideration of the contingency approach). However, such a linkage generally does not differ with respect to the industry types of high-tech and less-tech (except for two dimensions related to the market share measure).

Thirdly, pertaining to research question 3, it was discovered that industry turbulence regarding employee stability positively moderates the EO-performance linkage for the performance indicator of market share. In contrast, for industry munificence, characterised by employee growth, a negatively moderating effect on the EO-performance relationship was observed for the same performance indicator. Thus, both employee variables are considered as central environmental influencers towards the EO-firm performance linkage regarding market share. Even so, with respect to the remaining studied performance indicators, no such effect was observed.

Lastly, relating to research question 4, innovativeness was the sole dimension that positively affected the performance indicator of gross-profit-margin over a period of two years. Moreover, an adverse effect for risk-taking on return on assets was also found over the same time-span. As a consequence, EO, when considering the nuanced research within this thesis (cross-sectional of firms and/or industry types and conditions), was neither linked with generally positive nor superior firm performance as has been assumed across earlier studies but was instead associated with varying levels of the EO-performance linkage over time.

Implications for scholarship, firms and top-level managers, limitations of this study, as well as recommendations and directions for future EO-based research close the work.