This paper studies the unemployment accelerator, a mechanism where workers directly affect the firms’ financial conditions, and, in turn, firms’ financial conditions feedback again to the real economy. The unemployment accelerator builds on two key assumptions: search frictions in the labor market and firms’ default risk. The former assumption implies a positive relation between the firm’s value and its number of workers; the latter assumption entails a tight connection between the value of the workers and the firm’s incentives to default. We develop and estimate a model with these two frictions together with firm-level heterogeneity; and show the model matches firm-level statistics as well as business cycle fluctuations in labor and financial markets. We provide compelling micro-evidence of the unemployment accelerator: a 10% increase in a firm’s number of workers is associated with a 4% increase in its market value and a 6% decline in its probability of default. We show that our model can account for these facts, and that the two key assumptions we make are essential for this.