tional complexity which tends to prevent the derivation of informative results. To avoid such
complexities many environmental economists became reluctant to explicitly and properly
regard the materials-balance principle as the correct theoretical foundation of their analyses
(Pezzey and Toman 2002, 202; Pethig 2003). Therefore Ayres and Kneese's (1969, 283) ver-
diet still applies to much of the present work that production processes are viewed "... in a
manner that is somewhat at variance with the law of conservation of mass".
To be more specific, consider the simple production function Y : ≡+ → R + with
y = Y (e, £,m), (1)
where two inputs, labor £ and material m, are employed to produce two outputs as joint
products, a wanted consumer good, y, and an unwanted production residual, e (with e for
emissions). This type of technology was already applied in the early 1970s, e. g. by Forster
(1972) and Klevorick and Cramer (1972). Varying grossly in its degree of generality, it be-
came a widespread and accepted tool of analysis within few years (e. g. in Maler 1974, Pethig
1975, Baumol and Oates 1975).
In their survey on environmental economics, Cropper and Oates (1992) refer to the production
function (1) as the standard approach in the environmental economics literature3. They ob-
serve that the treatment of waste emissions "simply as another factor of production . seems
reasonable since attempts . to cut back on waste discharge will involve the diversion of
other inputs to abatement activities - thereby reducing the availability of these other inputs for
the production of goods" (Cropper and Oates 1992, p. 678). This citation reveals these au-
thors' awareness of technology (1) as a concept that implicitly involves both the generation of
a production residual and an abatement activity. Cropper and Oates (1992, p. 678) also find it
reasonable "...to assume the usual curvature properties ..." that is, they require function Y from
(1) to exhibit the
Properties (Y): Y : ≡+ → R + is concave and satisfies Ye > 0, Y > 0 and Ym > 0.
It is not clear, however, what exactly is the link between the production of a consumer good,
residuals generation and abatement which Cropper and Oates conjecture as being hidden in
(1). To put it differently, it is not clear how an explicit and comprehensive analysis of interde-
3 As compared to our equation (1), the equation (2) in Cropper and Oates (1992, p. 678) is slightly more general
in that they allow for an arbitrarily large vector of conventional inputs (which is reduced to the two-dimensional
vector (£, m) in (1)) and allow the level of pollution to be a negative productivity-reducing externality. This
externality is omitted in the present paper to keep the analysis simple.