This paper examines the relationships between the reported accidents of workers in construction firms and the probability of those firms' survival. Method: Between 2004 and 2010, a sample of 344 Spanish construction firms from Majorca were selected. The study built panel data with the reported official accidents from the Labor Authority records and the firm survival or mortality from the Bureau van Dijḱs Iberian Balance Sheet Analysis System database. The hypothesis is that a higher number of accidents directly affects the probability of the company surviving in the sector. By using a probit regression model with panel data, the relationship between these two variables were explored to test the hypothesis. Results: The study found evidence that an increment in accidents decreases the probability of the company continuing to operate, or worse, going bankrupt. The results can be useful to highlight the importance of defining policies to control those accidents effectively, since this may be a key factor in the sustainability, competitiveness, and growth of the construction sector for the economy of a region.