THE MEASUREMENT OF EFFICIENCY IN LIFE-INSURANCE - ESTIMATES OF A MIXED NORMAL-GAMMA ERROR MODEL

This paper make three contributions to the literature on efficiency in financial services. First, it presents estimates of a mixed error cost frontier in which the variances of both the normal and gamma distributions can vary with firm size. Second, it extends the literature on life insurance scale and product mix economies by incorporating and measuring X-inefficiency. Estimates of X-inefficiency range from 35 to 50%; ray scale economies exist up to $15 billion in assets. Third, comparisons of normal-gamma estimates with other methods - thick frontier, weighted least squares, half-normal distributions demonstrate some of their pitfalls.