Cape Cod method

Bornhuetter and Ferguson published a method in 1972, in the Proceedings of the Casualty Actuarial Society, that is worth exactly what its a priori loss ratio is worth. That a priori often comes from pricing, that is, from the very source reserving is supposed to check. The Cape Cod method, set out by Stanard in the same Proceedings in 1985, removes the dependency by estimating a single loss ratio from the triangle itself. You divide the sum of reported cumulative amounts across all accident years by the sum of premiums weighted by each year's percent developed. A year developed to 15% contributes 15% of its premium to the denominator, which stops recent immature years from dragging the loss ratio down. The Bornhuetter-Ferguson mechanics are then applied with that loss ratio. The method assumes every accident year shares the same expected loss ratio, which stops being true once the book spans an underwriting cycle. In a market that has moved from soft to hard, Cape Cod blends years priced at 95% and at 62% and returns an average that describes neither. Gluck's generalized variant (Proceedings, 1997), which weights each year by a decay factor according to its distance from the year being estimated, softens the flaw.

• Spencer M. Gluck, "Balancing Development and Trend in Loss Reserve Analysis", Proceedings of the Casualty Actuarial Society, vol. LXXXIV, 1997, pp. 482-532