Information Spillovers from Protests Against Corporations: A Tale of Walmart and Target Lori Yue, HayagreevaRao, Paul Ingram
Motivation • From 1998 to 2008, Walmart made 2,050 proposals, encountered 805 protests, and managed to open 1,234 stores. • Target floated 1,110 proposals, attracted only 74 protests, and was able to open 1,046 stores. • What explained the striking difference in the protest rates experienced by Walmart and Target?
Popular Media’s Explanation • Walmart is a poster child of extreme capitalism • Target has a benevolent corporate image • Activists argue that Target is not really that different from Wal-Mart • Serres, 2005; Bustillo and Zimmerman, 2010 • A tough anti-union attitude • Wages and benefits mirror Walmart’s • Super Target has similar impacts as Walmart Supercenter • Walmart is a regular contributor to community causes • “Why is Walmart the bad guy?”
A New Explanation: Information Spillover from Social Movement Activism • A framework with three parties: a first-mover (Walmart), a second-mover (Target), activists • Information asymmetries. Activism against the first-mover signals a community’s preference • Second-mover receives the signals and better positions itself
Walmart and Target Store Location, Before 1998 2322 Walmart stores 764 Target stores
Walmart and Target Store Opening: 1998-2008 1311 Walmart in triangles 1004 Target in dots
Hypotheses • H1: Protest events against Walmart have a curvilinear (+,-) relationship with Target’s entry. • H2: Walmart’swithdrawals due to protests have a curvilinear (+,-) relationship with Target’s entry. • H3: Walmart’s successes in overcoming activists have a positive relationship with Target’s entry.
Data and Methods • Data: All the 25,375 places in the U.S. from 1998 to 2008. Due to the creation of time-lagged variables, the sample used in the final estimation include all the years from 2001 to 2008. • Store Proposal and Protest: A media search from 1998 to 2008 using the Lexis-Nexis, the America’s News, and the Sprawl-Busters databases. Matched with a list of Walmart/Target store openings from 1998 to 2010.
Variables • Dependent Variable: A dummy variable to indicate whether Target proposed to open a new store in a place in a year • Independent Variables: • Anti-Walmart protest • Walmart’s withdraw due to protest • Walmart’s success despite protest • Control Variables: Community characteristics (population, income per capita, unemployment, urban, race, and region), political ideology (Pro Democrat, hazard of institutional escalation ), organization of mobilization (anti-Target protests, union, church, Main Street Program) ; retail economy (distance to Target d.c., Wal-Mart and Target stores within 5 miles); the effect of media (media’s attention, unfavorable editorials)
Estimation • Problem: Unobserved heterogeneities that simultaneously affect Walmart’s proposal (Y1), anti-Walmartprotests (Y2), Walmart’s withdrawals (Y3), and Target’s entry (Y4) • Solution: Conditional Mixed Process (CMP) Model (Roodman, 2011)
Estimated Probability of Target’s Entry with 95% Confidence Interval Walmart’sWithdrawl Due to Protest
Estimated Probability of Target’s Entry with 95% Confidence Interval Walmart’sSuccess Despite Protest
Discussion & Conclusion • Theoretical Contribution • The literature on private politics has emphasized the direct effect of protests on their intended opponent, and overlooked information spillovers from protests. • We distinguished between two types of spillovers, one type produced by the incidence of protest and a second type from a targeted firm’s response. • We contribute to the strategy literature on market entry by showing second-movers benefit from a first-mover’s exposure to protests. • For organizational learning, we show that social movement activism constitutes another form of information spillover between organizations. • Extension & Revision-mechanism & market size