Using Monte Carlo Sensitivity Analysis to Estimate Lost Earning Capacity
I recently estimated the lost earning capacity of a surgeon by using Monte Carlo sensitivity analysis to satisfy the criterion of “within a reasonable degree of economic probability.” This method was necessary because the surgeon’s earnings were based on numerous factors, all of which were variable, that is, predictable only within a range of possible values. However, the variable factors were capable of being defined by a probability distribution. For instance, along the but-for earnings path the variables included the probability of exceeding threshold net clinical revenues (which determined whether base salary would change in the subsequent period), the percent increase/decrease in base salary, the probability of earning a bonus for non-surgical services and activities, the dollar value of that bonus, the probability of promotion, and the timing of that promotion. Similarly, along the mitigating earnings path the variables included the probability of promotion, the timing of that promotion, and the percent increase/decrease in net earnings. Even the length of the forecast loss period was subject to variation. The range of the various variables was determined, in part, on the testimony of physician-surgeon experts and other data. 10,000 iterations of the Monte Carlo process were used to derive the mean (average) value of lost earning capacity as well as construct confidence intervals. Alternate specifications of the underlying probability distributions of the variables were applied to test the robustness of the initial result, each of which confirmed the reliability of the initial estimate of lost earning capacity.