ABSTRACT

Hypothesis 5a: In societies characterized by a high degree of performance orientation, entrepreneurial self-efficacy will be a more potent facilitator of entrepreneurial entry.

Performance orientation might also come at a cost if performance can be directed at a variety of activities. While self-efficacy may influence the performance of both managers and entrepreneurs (Chen, Greene, and Crick 1998), a societal emphasis on high performance may lead individuals towards activities with the highest predicted economic outcome rather than towards activities that are more uncertain. It has been widely documented that on average, individuals with equal skills and experience have higher economic returns as paid workers than as entrepreneurs – a career from which some reap economic returns but most earn little (Hamilton 2000). Hence, in cultural settings in which professionalism, performance and perseverance are seen as virtues, potential loss from engaging in more uncertain economic activities (if they do not pay off) may lead individuals to shun away from entrepreneurship if they are fearful of failing (Vaillant and Lafuente 2007). Thus, a society’s performance orientation may enhance the effect of both self-efficacy and fear of failure on entry: Hypothesis 5b: In societies characterized by a high degree of performance orientation,

fear of failure will be a more potent inhibitor of entrepreneurial entry.

3. Methodology 3.1. Data We test our predictions using 8 years of survey data from the GEM data-set (Reynolds, Bosma, and Autio 2005). We combined this with data on national cultural attributes collected by the GLOBE study (House et al. 2004). Together, 42 countries and 324,566 (unweighted) individual-level interviews from 2001 to 2008 were available after combining both data sources. We added exogenous controls of national-level attributes – country’s population and gross domestic product (GDP) per capita and two additional cultural measures – obtained from the International Monetary Fund, EuroStat and GLOBE data-sets, respectively.