Yoram C. Peles, Marshall Sarnat, Corporate Taxes and Capital Structure: Some Evidence Drawn from the British Experience, The Review of Economics and Statistics, Vol. 61, No. 1 (Feb., 1979), pp. 118-120

We are delighted that Nicholas Rau has attempted to generalize our result that: the of rate of return regulation is highly sensitive to the nature of the uncertainty. His paper has stimulated us to reflect further on generalizing and simplifying our joint results. Originally, we stated the following results: a) If uncertainty affects the maximal quasi-rents function R(K, u) in a multiplicative way, R(K,u) = R(K)(1 + u), then a sufficiently large gap must exist between the regulated rate of return (s) and the cost of capital (i) to induce the firm to select ex ante a scale of plant greater than the scale chosen by the unregulated monopolist. The closer is the regulated rate to the cost of capital, the more likely is it that the regulated firm will select ex ante a smaller scale of plant than is chosen by the unregulated firm. Were that to occur, regulation would definitely be worse than no regulation. b) If the uncertainty enters the maximal quasi-rents function in an additive way, R(K,u) = R(K) + u, then the conventional Harvey Averch and Leland Johnson (A-J) occurs (unless the regulation drives it out of business). Rau's main conclusion is that: .... if the state of nature affects both the average and marginal return on capital, then whether an A-J or anti A-J prevails depends on the amount of randomness in the environment. The more 'noise,' the more likely is an anti A-J effect (p. 190). A general and simple statement of the regulation theorems is derived below which contains points a) and b) and Rau's conclusion as special cases. 1. A General Formulation of the Problem