Salop and David Scheffman (1983) show that firms profitably can gain market power by conduct that raises their competitors' costs. Raising rivals' costs is a more credible route to market power than is predatory pricing because it is not necessary to cause the rivals to exit, no deep pocket is required, and the additional profits are gained immediately. That paper argues that vertical restraints and contracts with input suppliers can be fertile ground for raising competitors' costs. By contracting with one or more suppliers to exclude rivals, either by dealing with them on discriminatory terms or refusing to deal with them altogether, a firm sometimes can increase its rivals' costs. As a result, it can sometimes gain the power to raise price in the market in which it sells output. This type of contract often can be characterized as the purchase of an exclusionary right from the input suppliers. That is, in addition to the purchase of inputs, the predator also purchases the right to exclude (some of) its rivals from access to the suppliers' inputs. Exclusionary rights contracts can exist in a variety of forms. At one extreme are agreements that involve only exclusionary rights; no inputs are exchanged at all. For example, it was reported in Alcoa that at one time Alcoa purchased exclusionary covenants from power companies from which it did not purchase electricity. The contracts involved only the utilities' promises not to sell electricity to other aluminum companies. Such naked exclusionary rights contracts are unusual, of course. Most exclusionary rights are bundled with the sale of inputs. For example, Stroh Beer has alleged that the two major brewers purchase not only advertising time on network sports programs, but also the right to exclude remaining brewers from advertising on those same programs, even though the networks have other advertising time available. It might be argued that such exclusionary conduct would always fail for two reasons: the excluded rivals would have available effective counterstrategies to prevent their own exclusion; and input suppliers would have no incentives to reduce their sales by excluding some customers. These criticisms imply that raising rivals' costs by contracting with suppliers would not be credible. It surely is not true that exclusionary strategies to restrict rivals' input purchases will always succeed in raising their costs. For example, where rivals easily can substitute to other equally cost-effective inputs, or where entry into the production of inputs is so easy that the excluded rivals can efficiently produce the input themselves, then cost-raising strategies will fail. Moreover, where competition in the output market would be sufficient to maintain low prices despite the exit or increased costs of the excluded competitors, then no profit-maximizing firm would spend any resources trying to exclude those rivals. These conditions are discussed in detail in our earlier paper (1985). tDiscussants: Ronald H. Coase, University of Chicago; Gregory K. Down, Yale University; David Sappington, Bell Communications Research and University of Pennsylvania.

It is very nearly a truismi amaong neoclassical economists that the public is wont to confuse changes in the general price level with relative price changes. Correspondingl y, one might speak of the independence between the two as one of the central postulates of the neoclassical tradition. William Stanley Jevons was aml-ong the first to identify this basic dichotomin economic life as well as to attempt to explain the ordinary citizen's seenming incapacity to understand it:

Recently, economists have begun to develop a theory of the perfectly competitive firm and industry in long-run equilibrium.' In contrast to the traditional model2 in which all prices are parametric, this new theory takes explicit recognition of the fact that output price must adjust to exogenous changes in input prices before the industry can be said to be in long-run competitive equilibrium. The focus of this analysis has been to derive implications of competitive theory which can be tested using data generated by observing individual firms in

In his recent article in this Review, Bruce Hamilton attempted to extend Peter Mieszkowski's (1969, 1972) analysis of property tax incidence. He correctly notes that Mieszkowski's model does not describe a market equilibrium. Hamilton's model accounts for the tax and benefit incidence of an intrajurisdictional property tax, and it delineates the competitive market adjustments due to capitalization effects. He then draws several conclusions regarding fundamental urban economic problems. However, Hamilton's model fails to describe a market equilibrium for the identical reason Mieszkowski's fails; potential supply adjustments generated by capitalization are neglected. The purpose of this comment is to correct the Hamilton model by including sufficient conditions to obtain a market equilibrium. Our reformulation of the model indicates that some of Hamilton's urban policy conclusions are utterly incorrect while others are valid only under very restrictive assumptions. Consider a Hamiltonian metropolitan area comprised of three jurisdictions: 1) a homogeneous high-income housing (HIH) community, 2) a homogeneous l'w-income housing (LIH) community, and 3) a mixed community consisting of a percent of units of LIH and (1 a) units of HIH. In the initial (pretax) equilibrium property values reflect only resource costs. A public service benefit is then provided equally to each house in every community. A proportional property tax is levied in each community based on the value of the average property in the jurisdiction. There are no net capitalization effects in the two homogeneous communities because the benefit per household equals the tax. However, the mixed community is characterized by short-run intrajurisdictional net benefit differentials (INBD). The LIH receive a fiscal surplus and the HIII incur fiscal burden due to average cost pricing of the public services. Hamilton concludes that the net benefit differences are capitalized into property values. Land is assumed to bear the capitalization effects because capital is mobile (p. 748, fn. 9). Hamilton defines this posttax capitalization state as (in supply of housing and public service) because land value differentials exactly reflect the present value of INBD (see pp. 748, 750, 752). For Hamilton's conclusion to hold we must assume the relative speeds of adjustment (i.e., mobility) of capital, renters, and land differ in that order, with capital exhibiting the greatest degree of mobility. This assumption is not inconsistent with Hamilton's model and is no doubt generally accepted.' It is essential to note that capitalization requires at least partial immobility of one or more factors (assuming nonzero elasticities of factor substitution).2 Furthermore, the Hamilton efficient state implies land is completely immobile and thus bears the full burden of INBD. The immobility of land eliminates substitution possibilities so an intrajurisdictional tax imposes only income effects. An

Can advertising by dominant firms pose a predatory threat? Theoretically, the answer clearly is yes. However, empirical evidence indicating the importance of this theoretical finding is largely missing. Economists' empirical studies of advertising, which typically focus on the question of whether high levels of advertising are indicative of the presence of a barrier to entry or signal that advertising information can substitute for consumption experience, usually employ average advertising levels that hide strategically focused price cuts or changes in short-term advertising rates.

A cost-benefit analysis of sulfur dioxide regulations concludes that the Clean Air Act should be amended to lower the cost of pollution control. Net benefits will be higher, the study shows, if marginal costs and benefits are equated instead of continuing the current method of mandating a level of emissions. A better control strategy will be more effective than stringent control even when there is great uncertainty and error. 8 references, 2 tables. (DCK)

The Home Affordable Modification Program’s (HAMP’s) Principal Reduction Alternative (PRA) is a government-sponsored program to reduce the principal balances and monthly mortgage payments of troubled borrowers. We examine the effect of principal forgiveness on borrowers’ subsequent mortgage default. The program’s rules imply a kink in the relationship between principal forgiveness and a borrower’s initial equity level. Our identification strategy exploits the quasi-experimental variation in principal forgiveness generated by this kink using a regression kink design (RKD), which compares the relationship between initial equity and default on either side of the kink. We estimate that HAMP PRA reduced the quarterly default hazard from <f>3.8\%</f> to <f>3.1\%</f>. Received February 4, 2015; accepted August 2, 2015 by Editor Philip Strahan.