Plantin & Rocher (2016) document how insurers often engage in risk-shifting years before the materialization of a failure. This paper empirically examines this claim by testing the mechanisms of insurance insolvency, using a first-of-its-kind international database assembled by the authors which merges data on balance sheet and income statements together with information on impairments over the last 30 years. Employing different fixed effects logistic specifications and parametric survival models, the paper presents evidence, on top of the role of profitability as a leading indicator of failures, of the intrinsic asymmetries between the life and non-life insurance sectors. In the life sector, asset mix is highly significant in predicting an impairment, while operating efficiency plays no role. In the non-life sector, the opposite proves true.