Efficient Price Regulations of Networks that have Sunk Costs: Should Caps be Based on Historical or Replacement Costs

Abstract

Incentive regulation allows decentralised decision-making under regulatory settings that are based upon industry characteristics. This study considers the design of regulatory profit caps and the choice of historical or replacement cost for incentive regulation when there is uncertainty sunk costs and flexibility in the timing of investment. It demonstrates that which of historical or replacement cost regulation is desirable depends upon the sources and extent of supply and demand uncertainties and trends and thereby characteristics of the industry. The welfare optimising level of the cap differs between historical and replacement cost regulation the caps are generally higher than the weighted average cost of capital and welfare is degraded much more if the cap is set below as opposed to above the optimal cap.In the presence of uncertainty and sunk costs investment thresholds that exceed the standard WACC are required to enable investment. The WACC just reflects systematic risk. It does not reflect the probability of bankruptcy or idiosyncratic risks that firms prudently consider when making investment decisions. From society's point of view the WACC is frequently too low to act as an investment hurdle rate. In practice hurdle rates are often distinctly higher than equity holders' average rates of return and much higher than return on debt.We have considered a situation where the firm has no competition. The presence of competition will generally mean that firms' investments are desirably timed from society's point of view although investment will not occur immediately. The imposition of a cap to improve welfare may if it is too tight reduce welfare substantially even relative to the situation of no regulation and no competition. It arises because investment is delayed. In such circumstances the regulator may respond by removing scope for decentralised investment by forcing investment or reaching some regulatory pact with the firm. It is easy to show that such regulation does not remove the issues of specific risks and timing considered in this paper: they are intrinsic to the industry. Unless the regulator agrees to pick up the costs of risks - eg the costs of stranded assets - a reasonable rate of return under regulatory investment requirements should cover the real options that these risks imply.We have focussed on encouraging the optimal timing of sunk investments. While the approach has been cast as provision of the entire network the same approach applies to maintenance of an existing network that is also sunk. Unless maintenance expenditure is allowable in line with the optimal caps considered in this paper networks may deteriorate or require forced maintenance by regulation. The issues are the same.