What do we know about Single-Firm Conduct?

Luke M. Froeb
Vanderbilt University
ERS Group

Sept 26, Washington, DC
DOJ & FTC Hearings on single-firm
conduct

Slide 2

Acknowledgement

(does not indicate agreement)

• Tim Brennan (UMBC & Canadian Competition Bureau)

• Dan O’brien(DOJ)

• Mike Vita (FTC)

• Greg Werden(DOJ)

Slide 3

Why is Horizontal Merger Analysis Relatively Easy?

• Ignores long-run, indirect effects
  • Jargon: “distant” vs. “proximate” (short-run, direct)

• Distinct mechanisms through which mergers affect short-run welfare
  • Unilateral, Entry, Product Repositioning, Efficiencies
    • How much do we know about Coordinated?
  • Gather evidence on each mechanism

• Compute net effect (“balancing”, “trade off”)
  • Weighted sum (likelihood + magnitude)
    = expected net effect

Slide 4

Why is SingleConduct Analysis Relatively Hard?

• Long-run, indirect, strategic effects

• Opposing mechanisms appear together
  • Predation
    • Proximate: reduced price

    • Distant: fewer competitors

  • Vertical integration
    • Proximate: eliminate double marginalization

    • Distant: RRC or RRR

  • Exclusive dealing
    • Proximate: eliminate competitors

    • Distant: incentive alignment between retailer & manufacturer

• How do we balance proximate vs. distant effects?

Slide 5

Simple Taxonomy

		Distant Mechanism
		Bad	Good
Proximate Mechanism	Bad	(bad, bad)	(bad, good)
	Good	(good, bad)	(good, good)

• Proximate: immediate, direct, current
  • e.g., combining complements reduces price

• Distant: strategic, indirect, future
  • e.g., incentive to innovate, invest, enter, advertise

• NOTE: 2X2 box assumes only 2 mechanismsbalance proximate vs. distant effects?

Slide 6

Simple Taxonomy (cont.)

		Distant Mechanism
		Bad	Good
Proximate Mechanism	Bad	Makes no business sense, but for... (Dentsply, Conwood)	Exclusive dealing
	Good	Predation Bundling Vertical Integration Loyalty discounts	Most transactions Max RPM

Slide 7

US Monopolization vs. EU Abuse of Dominance

• Different enforcement regimes
  • US: More concerned with Type I errors

  • EU: More concerned withType II errors

		Distant Effect
		Bad	Good
Proximate Effect	Bad	Type II Errors (False Acquittals)
	Good	Type I Errors (False convinctions)

Slide 8

How does one Determine Effects?

• “Effect” question compares two states of the world (“with” vs. “without” behavior)
  • but only one is observed

• Two ways of drawing inference about unobserved state of world, “counterfactual”
  • Natural “experiments”
    • Good data may be rare

  • Theory-based inference
    • Theory of single-firm conduct often indeterminate

Slide 9

Natural Experiments

• Experimental market(with behavior)

• Control market(without behavior)

• Difference is estimate of effect.

• BIG questions
  • How well does experiment mimic effect?

  • Did you hold everything else constant?

Slide 10

Estimate of “Distant” effect Estimate of “Distant” effect

• FCC regulations (and Cable Act) required cable TV to carry local over-the-air channels

• Natural experiment: In 1980’s, Appeals Court overturned “must carry” on 1^st Amendment grounds

• Question: Would cable monopolist drop competitors, when allowed?

Slide 11

Results: Probability a Channel was Dropped or “Excluded”

	NOT Advertising competitor	Advertising competitor
Broadcast competitor	1.5%	1.1%
NOT Broadcast competitor	8.4%	6.2%

• Channels with low ratings dropped

• Competitors LESS likely to be dropped

• Refutes anticompetivehypothesis

Slide 12

Representative Studies

• Gasoline: prices 2.7¢/gallon higher in states with vertical divorcement laws
  • Vita (2000)

• Beer: Indiana’s 1979 ban on exclusive territories reduced per-capita beer consumption by 6%
  • Sass and Sauman(1996)

• Beer: UK divorcement of “tied” pubs raised price
  • Slade (1998)

• Fast food: prices at company-owned stores 2.8% lower
  • Graddy(1997)

Slide 13

LaFontaine-Slade Taxonomy: Voluntary vs. Govt.-imposed imposed

• when manufacturers … impose … restraints, not only do they make themselves better off, but they also typically allow consumers to benefit from higher quality products and better service provision.

• In contrast, when contract limitations are imposed on manufacturers via government intervention, the effect is typically to reduce consumer welfare as prices increase and service levels fall.

• ..., the interests of manufacturer and consumer welfare are apt to be aligned, while interference in the market is accomplished at the expense of consumers (and of course manufacturers).

Slide 14

Do we know what we don’t know?

• How to generalize to other cases?

• How to test for “distant” effects?
  • e.g., effects on innovation?

• How to test effects of antitrust intervention?

Slide 15

References

• Cooper, James, Luke Froeb, Daniel O'Brien, and Michael Vita, Vertical Antitrust Policy as a Problem of Inference, International Journal of Industrial Organization, 23 (2005) 639–664

• LaFontaine, Francine, and Margaret Slade, Exclusive Contracts and Vertical Restraints: Empirical Evidence and Public Policy, Handbook of Antitrust Economics, Paolo Buccirossi(ed.), Cambridge: MIT Press, forthcoming.