David M. Mandy

Killing the Goose That May Have Laid the Golden Egg: Only the Data Know Whether Sabotage Pays

A lively debate has developed concerning the incentive for a price-regulated input monopolist to engage in non-price discrimination if it vertically integrates into downstream markets. This paper identifies the key industry characteristics that determine whether such discrimination is likely, and studies both theoretically and numerically the tradeoffs among three important characteristics in the simplest extant model. The incentive to discriminate is theoretically ambiguous, but data for the US telecommunications industry indicate that discrimination is likely in the absence of policy-induced countervailing incentives. Countervailing policies include direct penalties for discriminatory behavior, forcing vertical autonomy, and fostering upstream competition.

Dynamic Pricing and Investment from Static Proxy Models

This paper evaluates the use of static cost proxy models in setting forward-looking prices such as the prices set according to the FCCs TELRIC methodology. First, it compares the time paths of prices and depreciation under traditional regulatory accounting with the prices and depreciation implied by various versions of TELRIC. When TELRIC prices are recomputed at intervals shorter than asset lives, the firm will generally not earn the target rate of return. In these cases, a correction factor must be applied to the TELRIC price path in order for revenues to exactly recover investment cost, including the target rate of return. Next, the paper considers a firms cost minimizing investment decisions under two different assumptions about asset obsolescence. In both scenarios, cost minimizing investment paths and implied utilization rates for the firms assets are derived under a variety of assumptions about the relevant input parameters. Some implications for TELRIC pricing are then derived.

Nonuniform Bertrand Competition

The feasibility of Bertrand undercutting with nonuniform prices is established and properties are derived for Bertrand equilibrium in nonuniform price strategies. With free entry, equilibrium entails zero-profit minimum average cost production. If there is more than one producing firm, all prices collapse to a minimum average cost uniform price. An existence condition is compared to conditions from uniform price theory. Without free entry equilibrium, prices may not collapse to a uniform price. Positive profit may occur but all firms earn equal profit and incur equal marginal cost, while consumers pay average outlay no greater than marginal cost. Copyright 1992 by The Econometric Society.

Seemingly unrelated regressions under additive heteroscedasticity Theory and share equation applications

Abstract We derive consistent, asymptotically efficient, and asymptotically normal estimators for SUR systems that have additive heteroscedastic contemporaneous correlation. Both our estimator for the location vector and the parameters of the covariance matrix possess these properties. The procedure is superior to other methods because we use GLS to estimate the parameters of the covariance matrix. Our method also permits the use of cross-equation parameter restrictions. We discuss how this type of heteroscedasticity arises naturally in share equation systems and random coefficient models, and how these models can be uniquely estimated with our two-step estimation technique.

TELRIC Pricing with Vintage Capital

We study the dynamic effects of technical progress on competitive prices. The model facilitates comparison between competitive prices and the TELRIC prices the FCC adopted for determining universal service subsidies. The prices differ due to differences in (1) discount factors, (2) the stream of operating costs, and (3) the method used to discount revenues. A calibrated comparison reveals the last difference as most important, with TELRIC prices understating competitive prices by billions of present value dollars nationwide. Competitive prices can be derived without any depreciation assumptions, contrary to regulatory practice, but cannot be calculated in advance of costs once capital utilization is endogenized.

The sampling performance of pre-test estimators of the scale parameter under squared error loss

Abstract This paper evaluates the sampling properties of the conventional, inequality and minimum mean square pre-test estimators of the scale parameter under a squared error loss.

A Unified Approach to Asymptotic Equivalence of Aitken and Feasible Aitken Instrumental Variables Estimators

Asymptotic equivalence of Aitken and feasible Aitken estimators in linear models with nonscalar identity error covariance matrices is usually established in a tedious case-by-case manner. Some general sufficient conditions for this equivalence exist but there are problems with the extant conditions. These problems are discussed and new widely applicable sufficient conditions are presented and applied to a variety of error structures. Copyright 1994 by Economics Department of the University of Pennsylvania and the Osaka University Institute of Social and Economic Research Association.

The moments of a pre-test estimator under possible heteroscedasticity

Abstract This paper investigates the sampling properties of a pre-test estimator for the location vector of the two-sample heteroscedastic linear model.

Competitive Two-Part Tariffs as a Response to Differential Rates of Time Preference

The major focus of the nonlinear pricing literature has been to demonstrate how welfare in a monopoly can be improved through the use of a nonlinear payment schedule. However, nonlinear prices arise in many different markets. This paper explores nonlinear price equilibria in which both multiple firms and price schedules exist. These outcomes are driven by the use of a sustainability criterion for equilibrium and differential rates of discount as the motivating force for the formation of nonlinear prices. Copyright 1991 by The London School of Economics and Political Science.

Forecasting unemployment insurance trust funds: The case of Tennessee

Abstract Many state unemployment insurance (UI) trust funds have reached levels inadequate to fund UI programs, resulting in renewed efforts to forecast UI variables. This paper describes efforts in Tennessee to formulate a UI forecasting model that overcomes the weaknesses of existing models. The intent is to help forecasters in other states cope with the problems of UI forecasting and to provide a framework to build upon. The two major goals of the model are to provide accurate forecasts and to facilitate strategic planning. Forecasting errors illuminate some important inaccuracies in the model. However, the model captures the variables most frequently subject to legislative change.