Unobserved Capacity Constraints and Entry Deterrence

More documents

Recommendations

Info

Unobserved Capacity Constraints andEntry Deterrence *Felix Munoz-Garcia † Gulnara Zaynutdinova ‡School of Economic SciencesWashington State UniversityDepartment of Financeand Management ScienceWashington State UniversityMay 2011AbstractThis paper examines entry deterrence and signaling when an incumbent firm experiences a capacityconstraint, arising from either her productive efficiency or the high market demand she faces. In bothcases, we demonstrate that separating and pooling equilibria can be sustained. Our results show that ifthe costs that constrained and unconstrained incumbents face when expanding their facilities aresubstantially different, the separating equilibrium can be supported under large parameter values. Inthis case, information is perfectly transmitted to the entrant. If, in contrast, both types of incumbentface similar expansion costs, a policy reducing expansion costs can help move the industry from apooling equilibrium to the separating equilibrium with associated efficient entry. Nonetheless, ourresults show that if this policy is overemphasized entry patterns remain unaffected, suggesting apotential disadvantage of policies that significantly reduce firms’ expansion costs.KEYWORDS: Business expansion; Signaling; Entry deterrence.JEL CLASSIFICATION: L12, D82.* We would like to especially thank Ana Espinola-Arredondo, Jia Yan, Jill McCluskey and LevanElbakidze for their insightful comments and suggestions.† 103G Hulbert Hall, School of Economic Sciences, Washington State University, Pullman, WA, 99164-6210. Tel. 509-335-8402. Fax 509-335-1173. E-mail: fmunoz@wsu.edu.‡ 481 Todd Hall, Department of Finance and Management Science, Washington State University, Pullman,WA, 99164-4750. E-mail: gzaynutdinova@wsu.edu.
1. IntroductionCapacity constraints constitute a limiting factor for industries experiencing a sudden increase in demand.Indeed, some firms may only be able to satisfy additional demand by producing at higher marginal costs,while others may find it extremely costly to produce above their current capacity, thus making themunable to satisfy unexpected increases in demand. An expansion of the current facility might be anattractive option, since it alleviates such capacity constraint. In the absence of entry threats, firms expandif their direct benefits from expanding are positive, i.e., if the increase in future profits associated to theexpansion offset expansion costs. Under entry threats, however, firms must consider not only this directbenefit but also the indirect effects that such expansion might entail. In particular, expansion might signala high demand and attract potential entrants to the industry, thus suggesting that firms suffering a capacityconstraint might face a tradeoff when considering whether or not to expand their facility. Such a tradeoffis specifically relevant in periods of economic recovery, where several firms start experiencing largercustomer traffic and sales, making them more likely to experience capacity constraints. 1In this paper we examine this tradeoff by studying entry deterrence in a context where theincumbent is privately informed about her capacity constraint. Specifically, the incumbent is constrainedif she cannot produce her profit-maximizing output because she faces a limited plant capacity. Thisoccurs, for instance, when her technological efficiency or the market demand she faces are relatively high.In contrast, an unconstrained incumbent can produce her profit-maximizing output. Our model considersthat, first, the incumbent chooses whether to expand her facility where the fixed costs from suchexpansion may differ between the constrained and unconstrained type of incumbent. The potential entrantdoes not observe whether the incumbent suffers a capacity constraint, and must therefore base her entrydecision on the information he infers from the incumbent’s expansion.We first show that both separating and pooling equilibria can be sustained, where information iseither perfectly conveyed to the potential entrant or concealed from him, respectively. 2 In the separatingequilibrium, such information allows the entrant to base his entry decision on more accurate informationabout his post-entry competition. In contrast, in the pooling equilibrium the entrant cannot accuratelyassess the profitability of the market, and thus may enter a market that is actually unprofitable. Hence, theseparating equilibrium supports entry in similar contexts as under complete information. Instead, the1 Industries that have recently reported relatively severe capacity constraints include, among others, the oil industry(as documented by the U.S. Energy Information Administration), the biopharmaceutical manufacturing industry (asreported in a survey conducted among European and U.S. firms by BioPlan Associates, Inc.), and the freighttransportationindustry (according to the University of Denver’s Intermodal Transportation Institute).2 In addition, both the separating and pooling equilibria survive standard equilibrium refinements in signalinggames, i.e., the Cho and Kreps’ (1987) Intuitive Criterion, under relatively general parameter conditions.1
Page 1: School of Economic SciencesWorking
Page 5 and 6: Our equilibrium results hold under
Page 8 and 9: increase for all units surpassing c
Page 10 and 11: BCC PLE BCC , whereas the high-cos
Page 12 and 13: the high-cost incumbent are relativ
Page 14: where both types of incumbent expan
Page 18 and 19: Proposition 4. In the expansion sig
Page 20 and 21: significantly reduced at C that bot
Page 22: choose among a continuum of investm
Page 25 and 26: M , NEthe high-cost incumbent does
Page 27 and 28: NEEgiven beliefs ( H | NoExp) p and
Page 29 and 30: KHKHr rM, E D,E Einc, Hinc,HPLEHw
Page 31 and 32: Similarly, note that probability r
Page 33 and 34: DE ,beliefs become ( H | Exp) 0 af
Page 35 and 36: 1E 5costs are BCCH , PLEH ,16 36for
Page 37 and 38: M, NED,EDE , M,NEincumbent does not
Page 39 and 40: Edo not expand their facility can b
Page 41 and 42: ENEIf p p and p then the entrant
Page 43 and 44: DNE ,F ent,L NENESolving for γ, w

Unobserved Capacity Constraints and Entry Deterrence

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?