In the classical literature on efficiency measurement, the notion of efficiency is intimately linked with the one of production frontier: specifically, a productive activity is called efficient if the vector of input-output quantities that describes it belongs to the (efficient) frontier of some appropriately defined production set; and the activity is called inefficient if this vector belongs to the interior of the set. Moreover, when time-related data are involved, a sharp distinction is usually made between shifts of the frontier and gains in efficiency. Running against this tradition, the authors deal in this paper with efficiency, progress and regress without using the frontier concept. For describing the relation between inputs and outputs in production, they substitute another notion called ‘benchmark production correspondence’, with respect to which the authors suggest that progress and regress be measured without having to distinguish between progress and efficiency gains, or between regress and efficiency losses. An illustration with a monthly time series covering almost 15 years of the activities of an urban transit company is presented and contrasted with DEA and FDH non-parametric frontier approaches.
