Projecting Calendar Period IBNR and Known Loss Using Reserve Study Results

Periodically, our reinsurance company does a time consuming, in-depth reserves study of each of its underwriting areas. These studies generate detailed information on exposure, market factors, report delay patterns, ultimate expected locus ratios, et cetera, f or each homogeneous group of contracts in the underwriting area. While these studies enable the company to periodically check the adequacy of it’s reserve levels in each underwriting area, they, by themselves, do not yield 1) interim IBNR for each future calendar month until the next study, 2) projected future calendar period IBNR and known loss for company planning based on our current book of business and future writings, 3) a comparison of actual "future" calendar period known loss experience with projected “future” calendar period known loss experience, 4) the comparison in (3) by homogeneous, group of contracts and by accident period. The purpose of this paper is to show how the detailed grow information from our reinsurance study is used by our company to address (1) – (4) above. It is hoped that in presenting our methodology the reader will be able to abstract general principles that will allow him to develop a similar system based on his reserve study and its output. The methodology to be discussed here is currently in use and programmed on an IBM PC. A small teaching example is included.