Quantitative Analysis Of Competitive Effects For Antitrust

of 42

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
PDF
42 pages
0 downs
0 views
Share
Description
Quantitative Analysis Of Competitive Effects For Antitrust. Day 1. Luke Froeb Owen Graduate School of Management Vanderbilt University April, 2003. Case Studies Showing How Modeling is Used in Antitrust. WorldCom-Sprint Consumer Branded Product Carnival-Princess.
Transcript
Quantitative Analysis Of Competitive EffectsFor AntitrustDay 1Luke FroebOwen Graduate School of ManagementVanderbilt University April, 2003Case Studies Showing How Modeling is Used in AntitrustWorldCom-SprintConsumer Branded ProductCarnival-PrincessWorldCom-Sprint Merger:Background
  • Merger scrutinized by
  • U.S. Department of Justice
  • Federal Communications Commission
  • Interested third parties like Bell Atlantic
  • Overlap in residential long distance service
  • Regulatory restrictions keeping local phone companies out of market were soon to fall
  • WorldCom-Sprint Merger:Methodology
  • Estimate consumer choice model (demand)
  • Estimate/Calibrate Firm Model (supply)
  • Simulate “ownership” effect of merger
  • WorldCom-Sprint Merger:Simulated Merger Effects
  • Demand estimation from bill “harvesting”
  • Inelastic demand
  • Predicted price increases for merging firms
  • WorldCom: 5.4% Sprint: 8.9%
  • Calibration from WorldCom’s margin
  • Small margins imply more elastic demand
  • Predicted price increases for merging firms
  • WorldCom: 2.2% Sprint: 5.1%WorldCom-Sprint Merger:Mergers in a Post-Entry World
  • Does entry occur in response to merger?
  • Entry by incumbent local exchange carriers (“Baby Bells”)
  • State-level “experiments” show 25% share of long distance
  • What is merger effect in post-entry world?
  • Baby Bell entry cuts the industry average price effect in half
  • Branded ConsumerGoods Merger
  • This is a real case that must be kept confidential, so numbers are disguised
  • Entrant gained 25% share in two years
  • Proposed to purchase rival brand
  • Only 3 brands in “high end” segment
  • Aggregate segment elasticity 1.5
  • Branded Consumer Goods Merger:Brand Elasticity, Prices, SharesBranded Consumer Goods Merger: Findings
  • Estimated demand implies brands are fairly good substitutes for one another
  • Predicted industry price increase of 4%
  • Merging brand price increases of 5% and 8%
  • 12% and 18% marginal cost reductions required to offset price increases
  • Is Merger Prediction Consistent with Entry Experience?
  • Incumbent brands did not reduce price in response to entry with a 25% share
  • Implies entrant is bad substitute
  • Yet, we get a significant price increase following merger
  • Implies entrant is good substitute
  • Is post-merger price increase consistent with no incumbent entry response?
  • How to Answer Question
  • Calibrate model to observed data
  • “Undo” entry by raising price of recent entrant until zero quantity
  • Compare price changes of remaining (incumbent) brands
  • Entry effects are reverse of “undoing” entry effects
  • Consumer Goods Merger:Entry Model
  • Prices of incumbent brands barely change
  • Quantity drops significantly
  • Entrants steal quantity, but do not affect price
  • In general, other firms do not much affect price
  • Consumer Goods Merger:How Can We Test this Prediction?
  • Where did entrant’s quantity come from?
  • 65% from “outside” option
  • Includes lower priced brands
  • 35% from incumbent brands
  • Is this consistent with entry data?
  • If not, may want to modify model
  • 2002 Cruise Line Merger:Introduction
  • Carnival (largest) and Royal Caribbean (second-largest) each bid for Princess (third-largest)
  • Capacity constraints and “perishable” service
  • big fixed costs, small marginal costs
  • Key strategy is “revenue management”
  • Price to match uncertain demand to available capacity, i.e. to “fill the ships”
  • Merger of Parking Lot Operators
  • Central Parking acquired Allright
  • Two largest parking lot operators in US
  • Pricing: “Is lot full by 9am?”
  • If “no,” then reduce price
  • If “yes,” then raise price
  • This profit calculus unchanged by merger
  • No merger effect if lots are full
  • But Justice Department opposed merger
  • Asked for 74 lot divestitures in 18 cities
  • Cruise Lines vs. Parking Lots
  • Similar strategies: filling ships vs. filling lots
  • There is no uncertainty about parking demand, but does that make a difference?
  • Theories considered by FTC
  • Fill-the-ship pricing is unaffected by merger
  • No quantity effect, but low-elasticity consumers pay more
  • Were theories correct? Magnitude of effects?
  • Conclusions Based on Formal Model of Revenue Management
  • Two merger effects
  • Ownership effect raises price
  • Information-sharing effect raises or lowers price
  • But always increases quantity
  • Both effects small and disappear as uncertainty decreases
  • Confirms basic intuition from parking lot merger, i.e. firms price to fill the ships, and this profit calculus is unchanged by merger
  • Spatial CompetitionGrocery Store MergerParking Lot MergerGeographic Differentiation Retail Sector is Consolidating
  • In US, Wal-Mart, K-Mart, Target, Costco, and Sears account for 60 percent of general-merchandise sales
  • General-merchandise is 15 % of all retail sales
  • Productivity advantage over smaller retailers
  • Economies of scale
  • Economies of purchasing
  • Economies of distribution
  • Productivity Gains Associated with Industry ConsolidationRetail Consolidation Also in Europe
  • In EU, top 10 grocery stores forecast to increase share to 50-60%
  • In 2002, top 10 had 38%
  • Wal-Mart entering Europe
  • Policy Reaction to Retail Consolidation
  • FTC challenged some retail mergers
  • Blocked Kroger + Winn-Dixie
  • Blocked Staples + Office Depot
  • Competitive analysis based on increase in local (within-city) horizontal market power
  • “standard” horizontal analysis
  • Quantitative Horizontal Analysis: Benefit-Cost Analysis of Merger
  • Goal: quantitative estimate of merger effect
  • Necessary to weigh efficiencies against loss of competition
  • Two methodologies
  • “Natural” experiments, e.g. Staples-Office Depot
  • Model-based simulations
  • Natural Experiments: Staples-Office Depot
  • Prices in two-office-superstore cities found to be 7% lower than in one-office-superstore city
  • Is this a good metaphor for merger?
  • Pass-through rate (from cost to prices) was estimated to be 15%
  • This implies that a 85% reduction in costs necessary to offset merger effect
  • FTC successfully challenged merger
  • Model-Based Simulation
  • Model current competition
  • Estimate model parameters
  • Simulate loss of competition using estimated parameters
  • Unilateral competitive effect computed as difference between pre- and post-merger Nash equilibria
  • Estimate “gravity” choice modelSurvey density in Charlotte, NCDots represent grocery storesChoice depends on price, distance, and “noise”Model-Based Simulation:Kroger + Winn DixiePre-Merger Equilibrium:Share of Kroger + Winn-DixiePost-Merger Equilibrium:Share of Kroger+Winn-DixieParking Lot Merger
  • “Gravity” choice model
  • Lots derive market power from location, capacity
  • Higher prices if
  • Few nearby lots
  • Many nearby consumers
  • Small lot capacity
  • Parking Lot Merger Model:Conclusions
  • Constraints on merging lots attenuate merger effects by more than constraints on non-merging ones amplify them
  • Merger effects poorly approximated by shares in geographic market areas
  • No bright lines between “in” and “out”
  • Shares poor proxies for localized competition
  • Justice Department erred when it asked for divestitures in 5-square block areas where the merging firms account for more than 35%
  • Gravity Choice Models And Merger Effects
  • Merger effects “depend” on
  • Location of consumers
  • Location of merging stores
  • Location of non-merging firms
  • Cost of travel
  • Other factors affecting demand
  • Weak generalization: small price effects if location only source of market power
  • Differentiated ProductsWhat if we can’t estimate demand?Modelling choices and trade-offsAccounting for efficienciesModel-Based Methodology
  • Specify and estimate a model
  • Consumer model (demand)
  • Firm model (supply)
  • Use model to simulate counterfactual scenario
  • Mergers, collusion, damages
  • Example and Questions To “Test” Approach
  • Logit demand curve
  • What about other forms?
  • Price-setting competition
  • What about product, promotion, placement?
  • What about auctions, quantity-setting, vertical arrangements?
  • Constant marginal cost
  • What about scale economies or capacity constraints?
  • Static game
  • What about dynamic strategies?
  • Unilateral merger counterfactual
  • What about coordinated effects?
  • Critique of Market Share ScreensWith Differentiated Products
  • Competition does not stop at market boundary
  • Shares may be poor proxies for competitive effects
  • No role for efficiencies
  • How do you trade off a 10% marginal cost reduction against a 400 point change in HHI?Common Problem: Cannot Get Reliable Demand Estimate
  • Relatively flexible functional forms often lead to nonsensical estimates
  • Goods are complements (when we know they are substitutes)
  • Inelastic demand (inconsistent with optimization)
  • Data not up to task of estimating so many parameters
  • If one firm increases price, its rivals gain quantity in proportion to existing sharesImplies all goods are substitutesImplies margins proportional to sharesImplies cross elasticities proportional to sharesThese restrictive forms require less data Aggregate elasticityOne brand level elasticity or marginSolution: Ask Less of Data byMaking Intuitive AssumptionsReplacing Market Share Screens
  • Swedish beer merger
  • Aggregate elasticity ≈ 1
  • Pripps margin ≈ 30%
  • Significant industry average price increase
  • Logit Model as a Screen
  • WorldCom-Sprint Merger
  • Aggregate elasticity ≈ -1
  • WorldCom Margin ≈ 30%
  • Relatively small price effects
  • Using Logit Model as Screen III
  • Rebuttable presumption starts dialogue
  • Show products are further apart than modeled
  • Show competition is more intense than modeled
  • Requires different safe harbors
  • “Give” parties a 5% MC reduction on each of their merging products
  • Would have allowed MCI-Sprint, but not beerHow Do you Incorporate Merger-Specific Efficiencies?
  • How much do cost savings affect post-merger equilibrium?
  • One product assumes the mc of the other; or
  • Parties “prove” merger-specific cost reductions
  • Compute cost reductions sufficient to offset predicted price increase
  • In general, 2X to 3X times computed product price rise is enough.
  • Because pass-through related to post-merger price increase
  • Which Welfare Standard?Total vs. Consumer Welfare
  • Do fixed costs count?
  • Superior Propane in Canada fixed cost savings but prices went up.
  • Gain in profits was higher than consumer welfare loss
  • Canada is only country with a total welfare standard.
  • Short-run price increases vs. long-run efficiencies
  • Related Search
    Advertisements
    Similar documents
    View more...
    Advertisements
    We Need Your Support
    Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

    Thanks to everyone for your continued support.

    No, Thanks