Underwriting Apophenia and Cryptids: Are Cycles Statistical Figments of our Imagination?

This paper re-examines the evidence in favour of the existence of underwriting cycles in property and casualty insurance and their economical significance. Using a meta-analysis of published papers in the area of insurance economics, we show that the evidence supporting the existence of underwriting cycles is misleading. There is, in fact, little evidence in favour of insurance cycles with a linear autoregressive character. This means that any cyclicality in firm profitability in the property and casualty insurance industry is not predictable in a classical econometric framework. It follows that pricing in the property and casualty insurance industry is not incompatible with that of a competitive market.