Measuring efficiency when competitive prices aggregate differences in product quality and risk

In perfectly competitive markets with homogenous goods, prices aggregate inputs and outputs into a money metric that allows production plans and, hence, firms to be ranked by their profitability. Standard techniques of efficiency measurement use this metric to estimate cost and profit frontiers that identify “best-practice” production, conditioned on these exogenous prices. However, when prices vary due to differences among firms in the quality of outputs and inputs and in how asymmetric informational problems are resolved, both quality and the production of information can be decision variables of the firm, and prices will have endogenous components linked to production decisions. For example, in banking, prices of financial inputs and outputs are linked to credit quality and, hence, to risk and, thus, aggregate both inputs and outputs and their risk characteristics as well as reflect how informational asymmetries in credit markets are ameliorated. This paper argues that these cases pose two serious problems for the standard techniques of efficiency measurement: (1) they underestimate inefficiency because, in conditioning on prices, they fail to account for the effects of suboptimal pricing strategies on profitability; and (2) in ignoring how production plans and pricing strategies affect market-priced risk, the standard techniques neglect the effects of different pricing strategies on the discount rate, on expected profit and, hence, on market value. Two alternative techniques that do not condition their frontiers on prices and that account for risk are described to show how they measure the efficiency of different pricing strategies as well as production plans. These two alternative models for measuring efficiency are employed to study how differences in pricing strategies affect the efficiency and market value of highest-level bank holding companies in the United States in 1994.