profit testing and the profit signature

Before launching a product, an insurer wants to know not just whether it breaks even, but whether — and when — it actually earns a profit for the shareholders who put up the capital. Profit testing is a cash-flow projection that plays the product forward year by year on a set of best-estimate assumptions, tallying every premium in, every claim, expense, and reserve change out, plus interest earned, to see what is left over for the company each year.

Precisely, you project, for a model policy (or a model point representing many), the expected cash flow in each policy year: premiums received, less expenses, less claims paid, less the increase in reserve, plus investment income, and less the cost of holding required capital. The result is the expected profit emerging at the end of each year. The sequence of these yearly profits — one number per policy year — is the profit signature (when scaled by the probability the policy is still in force) or the profit vector (before that scaling). Discounting the signature at the shareholders' required return and summing gives metrics like the net present value of profit and the internal rate of return.

Profit testing is how products are actually priced and approved in modern practice — far more flexible than a single equivalence-principle equation, because it can layer in expenses, lapses, taxes, capital costs, and realistic timing. The profit signature almost always starts NEGATIVE: heavy first-year acquisition costs and the need to set up reserves mean the company sinks money into a new policy before it ever profits ('new business strain'), and only in later years does cash emerge. A caveat: the answer is only as good as the assumptions, so profit tests are run under many scenarios and sensitivities, not just one base case.