Competitive pressure: competitive reactions at the group-level

ABSTRACT

We offer a new perspective on competitive dynamics research by analyzing firms’ reactions to groups of rivals, both at the industry and the strategic group level. We propose that firms observe a buildup of rival actions and perceive increasing ‘competitive pressure’ to respond, which results in greater action volume. Data on the German mobile telephony supports our argument: the volume of firms’ actions is influenced by a buildup of actions by multiple rivals, and firms react more strongly to rivals in their own strategic group.

KEYWORDS:

• Competitive dynamics
• managerial cognition
• mobile telecommunications
• strategic groups

JEL:

• D83
• M10
• D22
• L11

Acknowledgments

We gratefully acknowledge comments by Peter Bryant, Christoph Dehne, JP Eggers, Johannes Koenen, Alex Oettl, Nils Stieglitz, Catherine Tucker, Joachim Winter, Walter Zucchini and audiences at the AOM Meetings 2011, DRUID Annual Conference 2011, TIME Colloquium Munich, EBS Business School, Freie Universität Berlin, Università Ca’ Foscari Venice, and the University of Southern Denmark. Any remaining errors are our own.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary Material

Supplemental data for this article can be accessed here.

Notes

¹ Note that we neither propose nor observe a snowballing or ‘Red Queen’ effect (Barnett and Sorenson 2002). This is discussed below.

² To be precise ‘managerial teams observe actions and perceive competitive pressure. As our focus is not on micro-processes within firms we use ‘firm’ as shorthand for the groups of managers perceiving competitive pressure (Barr and Huff 1997).

³ Another interesting measure might have been firms’ action magnitude, e.g. the size of a price drop. Unfortunately, the ‘price’ each customer pays for his tariff depends on his usage pattern. In the absence of micro-data on mobile phone usage for each tariff we were therefore unable to calculate the tariffs’ ‘price’ and consequently also the magnitude of price drops or rises.

⁴ Note however, that using a stock measure to capture the buildup of pressure prevents us from estimating hazard models for the lag between actions and reactions, since the main dependent variables are highly correlated with time.