Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
We quantify the empirical relevance of the pecking order hypothesis using a novel empirical model and testing strategy that addresses statistical power concerns with previous tests. While the classificatory ability of the pecking order varies significantly depending on whether one interprets the hypothesis in a strict or liberal (e.g., "modified" pecking order) manner, the pecking order is never able to accurately classify more than half of the observed financing decisions. However, when we expand the model to incorporate factors typically attributed to alternative theories, the predictive accuracy of the model increases dramatically--accurately classifying over 80% of the observed debt and equity issuances. Finally, we show that what little pecking order behavior can be found in the data is driven more by incentive conflicts, as opposed to information asymmetry.