This paper studies whether newly created firms have higher injury rates than established firms. We use data on a large sample of single-establishment firms in Pennsylvania from 2001–2005 to examine the relationship between firm age and the risk of lost workday injuries. Using the full set of firms, there appears to be little overall correlation between firm age and risk. If anything, newer firms appear less likely to have lost workday injuries. When we condition on having at least one injury reported in 2000, however, we find that in later years the injury risk of firms declines with age. This pattern is consistent with systematic underreporting of injuries by new firms.