Biased and overconfident, unbiased but going for it : How framing and anchoring affect the decision to start a new venture

Cognitive heuristics, biases, and overconfidence have been suggested as an explanation for entrepreneurial entry. Nevertheless, empirical research on the subject has produced mixed findings and has under-explored the cognitive mechanisms leading to overconfidence in entrepreneurial settings. In two within-subject experiments, we focus on three cognitive heuristics—reference point framing, outcome salience framing, and anchoring in conjunctive events—and examine their effects on perceived risk, confidence, required and estimated probabilities of success, and the decision to start a new venture. Our findings show that reference point framing and outcome salience framing affect the decision to enter directly and indirectly via risk perception, but do not affect confidence. In addition, the effect of anchoring is contingent on the congruence between its semantic and its numeric influences. Overconfidence only obtains when the numeric and semantic influences of anchoring are aligned and aimed at enhancing the salience of potential positive outcomes, i.e., through high probabilities of success.