Predicting Startup Survival Using the Normalized Burn Rate

We study the association of startup firm spending with firm survival. We propose that spending per employee (the “normalized burn rate”) captures entrepreneur's ability to avoid failure better than total spending (the popular “burn rate”). We derive an analytical model to describe how spending per employee reflects entrepreneur's knowledge about the contribution of human and non-human input flows to firm value. The model prescribes a U-shape relationship between the level of spending per employee and firm failure, and that most firms spend below the optimum. These hypotheses are borne out by a seven year representative panel dataset of U.S. businesses founded in 2004. The data also show that high levels of education and work experience (proxies for precise knowledge) lead entrepreneurs to spend closer to the optimum. These results suggest that spending per employee can be a useful metric to compare across firms and to assess managerial decisions within firms