LEFT FOR DEAD: ANTI-COMPETITIVE BEHAVIOR IN ORBITAL SPACE NODIR ADILOV, BRENDAN M. CUNNINGHAM, PETER J. ALEXANDER, JERRY DUVALL and DANIEL R. SHIMAN ∗ In a dynamic investment framework with depreciation, we show incumbent satellite operators have incentives to “warehouse” a fraction of their assigned spectrum and orbital slots, keeping nonoperational assets in place, which reduces output, increases prices, and diminishes social welfare. Exploring three distinct market structures, we model frms’ incentives to warehouse, and show conditions under which frms choose to warehouse rather than replace nonfunctioning satellites. We fnd a dominant frm with a competitive fringe produces more and longer duration warehousing relative to perfect competition or monopoly. Regulators could remediate warehousing by increasing a frm’s marginal costs, or by increasing the probability of reallocating orbital slots that do not have a fully functioning satellite. (JEL L9, L5) I. INTRODUCTION We model the investment choices of a large incumbent satellite frm and perfectly competi- tive entrants in placing satellites in geostationary orbits, and show that incumbent frms have incentives to under-utilize spectrum and orbital slots by maintaining a nonoperational satellite at the assigned orbital location or not utilizing ∗ B.M.C. gratefully acknowledges funding in support of this project from a NASA Connecticut Space Grant (PTE Federal Award No.: NNX15AI12H). N.A., P.J.A., J.D., and D.R.S. did not receive NASA funding under this grant, and used no government resources during their participation on this project. We thank the editors and two anonymous referees for their many helpful suggestions in shaping this work. We also thank participants at the 14th Annual International Indus- trial Organization Conference who provided many thoughtful comments. We are indebted to David Sappington, who offered invaluable guidance and support. We thank Dimitrios Pachis for his useful suggestions. Any errors are our own. The views presented here are those of the authors, and do not necessarily refect the views of the Federal Communications Commission or any of its Commissioners or staff, NASA, or the United States Government. Adilov: Professor of Economics, Economics and Finance, Purdue University Fort Wayne, Fort Wayne, IN 46814. Phone 260-481-6497, Fax 260-481-6879, E-mail adilovn@pfw.edu Cunningham: Associate Professor, Department of Eco- nomics, Eastern Connecticut State University, Willimantic, CT 06226. Phone 860-465-0660, Fax 860-465-4469, E-mail cunninghambr@easternct.edu Alexander: Senior Economist/Researcher, Federal Commu- nications Commission, Washington, DC 20554. Phone 202-418-2183, E-mail peter.alexander@fcc.gov Duvall: Industry Economist, Federal Communications Com- mission, Washington, DC 20554. Phone 202-418-2616, E-mail jerry.duvall@fcc.gov Shiman: Industry Economist, Federal Communications Commission, Washington, DC 20554. Phone 202-418- 7153, E-mail daniel.shiman@fcc.gov the spectrum or orbital slots for extended periods of time. In short, incumbents beneft by holding under-utilized spectrum and orbital resources strategically to maintain a competitive advan- tage over potential competitors. We refer to this investment tactic as “warehousing” and suggest that warehousing spectrum and orbital slots is a form of market foreclosure. 1 In our model, frms can choose to invest by placing a fully functioning satellite in each assigned slot, or maintain a nonfunctioning satel- lite (a “zombie”) in some (or all) slots. While a nonfunctioning satellite does not produce rev- enue, we suggest a nonfunctioning satellite still provides value for the incumbent operator via foreclosure. That is, if the incumbent holds a non- functioning satellite, the orbital slot and spectrum are not available to a new entrant to launch a fully functional satellite, which may relax pricing pres- sure on the incumbent. In essence, the incumbent may raise endogenous barriers to entry. To our knowledge, this paper is the frst dynamic economic model of warehousing orbital slots and spectrum in the geostationary satel- lite industry. Somewhat related efforts explore airline slotting, grocery slotting, and electricity capacity withholding. In addition to a dynamic model with depreciation, our work is further distinguished from all these efforts in that we model investment choice within the context of a 1. We cannot hope to cover the general literature on market foreclosure, but direct the interested reader to Rey and Tirole (2007). There is also a literature on patent preemption; see, for example, Gilbert and Newbery (1982). 1497 Economic Inquiry (ISSN 0095-2583) Vol. 57, No. 3, JULY 2019, 1497–1509 doi:10.1111/ecin.12790 Online Early publication April 10, 2019 © 2019 Western Economic Association International