Several of the global models (see the review in Dean and Hoeller 1992:26-7) employ a parameter called AEEI (autonomous energy efficiency improvement) to capture a perceived economic tendency to move towards greater energy efficiency independently of relative price changes (i.e. of substitution effects, which Manne and Richels (1991), for example, model through a separate parameter, ESUB, the elasticity of price-induced substitution).
The values of AEEI vary in the models between 0.25 per cent and 1.0 per cent. Even this range does not encompass the uncertainty associated with AEEI. Williams (1990), for example, considers that a 1 per cent value is too low, on the basis of historical decreases in energy intensity from 1973 to 1986.
Hogan and Jorgenson (1991), on the other hand, find on the basis of econometric estimates using US data that AEEI is effectively negative, in other words that, without substitutions between factors induced by relative price changes, the energy-GDP ratio would increase. This is because their results indicate a positive technical bias for energy. Their analysis further suggests that increasing the price of energy, for example through a carbon tax, could reduce the growth of total factor productivity and thereby bring about a reduction in the growth rate. Their initial calculations suggest that the effect of this on output could be of the same order of magnitude as the reduction in output due directly to energy's higher price.
The value chosen for AEEI makes a substantial difference to the baseline CO2 generated, as Dean and Hoeller report: 'A difference of 0.5 per cent in this parameter [AEEI], given compounding, can lead to an outcome in 2100 which is as much as 20 billion tons [of carbon emissions] different.. Uncertainty about the size of this parameter is likely to remain large as it depends on future technical progress' (Dean and Hoeller 1992:19; and p. 28, Table 4). Obviously differences in baseline emissions of this magnitude would greatly affect the cost of reducing these emissions to any particular level.
Whether or not investment and improvements in energy efficiency will accelerate with a carbon tax, many analysts have argued that market failures are preventing the implementation of some cost-efficient energy conservation measures now (e.g. Jackson and Jacobs 1991, Lovins and Lovins 1991; Chapter 10 in this volume). After reviewing this issue, Cline (1992:227) decides that a reasonable estimate is that the first 22 per cent of carbon emissions from base can be cut back at zero cost. The incorporation of such costless cuts in the cost-generating models would obviously reduce their negative outcomes from a carbon tax.
Was this article helpful?