The Price is Right

Energy consumption has been a subject of great concern to policy- makers during the last decade. Has the increase in energy prices since 1973 encouraged more efficient use of energy? A change in "energy productivity', the relationship between total energy consumption and output, might provide an answer to this question. In order to assess this relationship we need a variable representing "aggregate' energy consumption, the total consumption of all fuels, including coal, oil, gas and electricity. In a recent Discussion Paper Richard Westoby and CEPR Research Fellow David Pearce conclude that many of the measures which have been widely used in empirical studies are inappropriate for economic analysis.

Measures of useful energy have generally been calculated on the basis of relative fuel efficiencies. These are obtained from engineering data and take into account the thermal equivalence of fuels, losses in transmission to the consumer and the final use of the energy by the consumer. Economists, by contrast, have constructed measures of energy consumption based on production functions, relating measured output to the observed inputs of capital, labour and different types of fuel. The estimated coefficients of the fuel inputs in the production function are identified as the "relative fuel efficiencies', which are then used as the weights in an index of aggregate energy consumption. Westoby and Pearce argue that the production functions used in this procedure are often misspecified. Moreover, investigators may be misled if they regress such an energy index on the same output data used to construct the index.

Westoby and Pearce argue instead that fuel prices are the most appropriate means of weighting the consumption of different fuels in order to obtain an aggregate energy consumption index. The argument is that prices reflect the consumer's rates of substitution between fuels at the margin, which in turn reflect the relative efficiencies of the fuels. This procedure was originally recommended by Turvey and Nobay in 1965. Westoby and Pearce extend their "Net Energy Expenditure' index to provide a series for the UK running from 1954 to 1980.

The index does not appear to have any special advantage for forecasting purposes, but it is preferable for assessing trends in energy productivity, the authors argue. The conventional measure of energy productivity, the ratio of aggregate physical energy to GNP, exhibits a secular decline since the 1950s. The ratio of the authors' Net Energy Expenditure index to GNP shows a tendency to rise until the early 1970s, when it begins to decline. The authors attribute the divergence between the two measures in the 1960s to the substitution of other fuels for coal.


Net Energy Expenditure and Energy Demand in the UK
Richard Westoby and David Pearce

Discussion Paper No. 46, February 1985 (IT)