Financing Experimentation

Tong Liu, Finance, The Wharton School

Abstract: This paper studies the strategy of entrepreneurs to finance their experimentation given the presence of adverse selection in capital markets. It quantifies the effect of information frictions on firm value by structurally estimating a dynamic model that features volatile market valuation, strategic experimentation, and dynamic adverse selection. Entrepreneurs make decisions to access public financing by weighing market-timing incentives against the costly delay to signal by conducting additional experimentation. Leveraging the unique setting of biotech startups with drug development, I employ data variations in leading drug stages at IPOs, durations of staying private, and IPO valuations to estimate the model. I find that adverse selection is prevalent between early-stage startups and investors. My baseline estimates suggest that information frictions cause about 24% loss of ex- ante firm value, which is due to the direct effect of market belief distortions and the indirect effect of firms remaining private longer and hence burdened with higher financing costs by approximately 15%. I also document substantial variations in magnitude of the effect across VC-backed startups and firms with more effective “patent fences.” These results show that information frictions have a large impact on the financing behavior of startups and firm values.