![HoustonKemp.com
Economics is only relevant if it can assist in the
assessment of conduct with reference to the law
• Section 50 of the Competition and Consumer Act
(CCA) prohibits mergers that…
‘would have the effect, or be likely to have the effect,
of substantially lessening competition [SLC] in any
market’
5](https://image.slidesharecdn.com/c8cb8dec-f059-4a18-8fe8-07e5f9090452-161007020234/85/Lecture-at-Sydney-University-Mergers-and-Vertical-Restraints-5-320.jpg)
Lecture at Sydney University - Mergers and Vertical Restraints
- 1. HoustonKemp.com Mergers and Vertical Restraints The application of economics to examine the competitive effect of mergers and vertical restraints Dr Luke Wainscoat Senior Economist, HoustonKemp University of Sydney 6 October 2016 © 2016
- 2. HoustonKemp.com Mergers How economics can help assess the effect of a merger on competition
- 3. HoustonKemp.com Outline • How economists can help assess the competitive effect of mergers • How does the market function? • Coordinated effects • Barriers to entry • Unilateral effects (firms set quantity) • Unilateral effects (firms set prices) 3
- 4. HoustonKemp.com Mergers can be very valuable for the economy • ACCC assessed 322 mergers in 2014-15, conducting public reviews of 42 • Cost synergies estimated to be 3-4% of transaction value, on average • Value of mergers estimated to be US$1.3 trillion in Asia-Pacific region in 2015 • BUT, some mergers can reduce competition, driving up prices 4
- 5. HoustonKemp.com Economics is only relevant if it can assist in the assessment of conduct with reference to the law • Section 50 of the Competition and Consumer Act (CCA) prohibits mergers that… ‘would have the effect, or be likely to have the effect, of substantially lessening competition [SLC] in any market’ 5
- 6. HoustonKemp.com Economics can help determine whether there has been a SLC 6 How does the market function? What are the credible theories of harm? Does the evidence support a credible theory of harm? • What information is needed? • Market definition • Use qualitative and quantitative evidence • What is the likely effect of the merger? • Theories of harm need to be consistent with how market operates • Under what conditions would theory of harm be true? SLC • Relate back to competition law
- 7. HoustonKemp.com How does the market function? • Strategic environment › Price setting › Quantity/capacity setting › Auction › Bargaining • Type of products › Homogenous or differentiated? › Substitutes 7 • Entry/expansion › Capacity constraints • Market structure › Concentration (over time) › Vertical relationships • Type of customers › Small › Large • Others › Two-sided market › Network effects › Innovation
- 8. HoustonKemp.com What are the credible theories of harm? • Vertical mergers › Similar arguments for vertical restraints • Horizontal mergers › Coordinated effects Collusion becomes more likely/stable › Unilateral effects Firms act independently 8 Firm A Firm B Firm C Firm D
- 9. HoustonKemp.com Coordinated effects • Three conditions necessary for collusion to take place › Firms reach agreement as to conduct › Collusive agreement is internally stable › Collusive agreement is externally stable • Merger may affect first two conditions › Easier to reach agreement with fewer firms (in particular if merge with maverick) › Reduced incentive to cheat in more concentrated market 9
- 10. HoustonKemp.com Three main sources of constraint to unilateral conduct 10 Firms in the market Potential entrants Customers
- 11. HoustonKemp.com Contestable markets and barriers to entry 11 • Perfectly contestable market › One supplier › Technology/knowhow for production available to all › Fixed, but no sunk costs • Outcome › Price=average costs › Mergers have no effect • But › Fixed costs often sunk – no ‘hit and run’ entry › New entrants may have higher variable costs
- 12. HoustonKemp.com Barriers to entry allow incumbents to set prices above the competitive level without entry occurring • Some disagreement on exact definition • Something that allows the incumbents to earn above-normal profits • A cost of producing that must be borne by an entrant but not incumbents • Additional profit earned as a sole consequence of being established in the industry • Barriers to entry allow firms to have market power • Extent to which the threat of entry restricts market power depends upon › Likelihood of entry › Timeliness of entry › Impact of entry on the incumbents 12
- 13. HoustonKemp.com Unilateral effects when firms set quantities • Cournot model › Firms independently and simultaneously decide their own quantity/capacities › Products are homogeneous › Prices determined by total amount supplied by all firms • Examples › Airlines › Farms › Mining 13
- 14. HoustonKemp.com Herfindahl-Hirschman Index (HHI) • HHI is the sum of the squared market shares › 𝐻𝐻𝐼 = 𝑠𝑖 2 • Eg › Market shares, 20%, 30% and 50% › HHI = 202+302+502=3,800 • Take into account › HHI post merger › change in HHI as a result of the merger • ‘The ACCC will generally be less likely to identify horizontal competition concerns when the post-merger HHI is: › less than 2000, or › greater than 2000 with a delta less than 100.’ 14
- 15. HoustonKemp.com Greater Herfindahl-Hirschman Index (HHI) leads to higher prices 15 𝜋𝑖 = 𝑥𝑖 𝑃 𝑋 − 𝐶 𝑥𝑖 𝑃 + 𝑥𝑖 𝑃 𝑋 − 𝐶 𝑥𝑖 = 0Profit for firm i FOC 𝜀 = 1 𝑃 𝑋 𝑃 𝑋 𝑚𝑖 ≡ 𝑃 − 𝐶 𝑥𝑖 𝑃 Price cost margin Elasticity of demand 𝑃 − 𝐶 𝑥𝑖 𝑃 = −𝑥𝑖 𝑃 𝑋 𝑃 𝑚𝑖 = −𝑥𝑖 𝑃 𝑋 𝑃 Greater HHI More elastic demand Lower prices Higher prices 𝑠𝑖 𝑚𝑖 = −𝑠𝑖 2 𝜀 𝑠𝑖 𝑚𝑖 = − 1 𝜀 𝑠𝑖 2 𝑚 = − 𝐻𝐻𝐼 𝜀 Profit max = −𝑥𝑖 𝑋 𝑃 𝑋 𝑋 𝑃 = −𝑠𝑖 𝜀 𝑠𝑖 = 𝑥𝑖 𝑋 Market share Share weighted margin
- 16. HoustonKemp.com Limitations of Herfindahl-Hirschman Index • Post merger market shares will not be sum of pre merger market shares if the merger is anticompetitive • Industry average marginal cost will change if there are efficiencies and/or if market shares change • Need to consider possibility of entry • Requires market shares of all firms • Only applies to Cournot competition 16
- 17. HoustonKemp.com Unilateral effects when firms set prices and sell differentiated products • Bertrand competition › Firms set prices › Products are differentiated – they are imperfect substitutes › Eg, differentiation may be Brand Quality Location of stores • Examples › Restaurants/ coffee shops › Retail 17 Store A Store B Store C
- 18. HoustonKemp.com Merger simulation for differentiated products • Use model calibrated to observed features of market to predict the effects of a merger › Estimate cost and demand parameters Assume form of demand › Calibrate demand system to give prices and market shares actually observed › Specify model of competition – Bertrand in this case › Re-estimate model with new market structure, and potentially, efficiencies 18
- 19. HoustonKemp.com Drawbacks of merger simulation • Assume form of competition, demand system and functional form of marginal cost unchanged by merger • Assumes no supply side response • Data availability • May not provide precise result • Time intensive • Results can appear to be opaque 19
- 20. HoustonKemp.com Simplified merger simulation – pricing pressure tests • Number of assumptions have been suggested › Hold prices of third party firms constant Only need price elasticities and cross-price elasticities of merging firms › Use Lerner index to determine own-price elasticities from mark-ups › Use diversion rations instead of cross price elasticities › Demand curvature Assume linear or constant elasticity of demand Assess effects at current prices • Leads to a variety of different tests 20
- 21. HoustonKemp.com 21 Upwards Pricing Pressure (UPP) test Merger Merger Merger Merger Merger
- 22. HoustonKemp.com The economic intuition behind UPP 22 Price Output Revenue RevenuePrice Output
- 23. HoustonKemp.com The economic intuition behind UPP 23 Price rises Quantity falls Quantity Increases
- 24. HoustonKemp.com The economic intuition behind UPP 24 Price rises Quantity falls Quantity Increases Significance of this incentive depends upon: • Quantity of sales that switches (closeness of competition) • Profit margin for silver (benefit from each switched sale)
- 25. HoustonKemp.com The economic intuition behind UPP 25 Output $ Demand Marginal Revenue Marginal Cost Q P
- 26. HoustonKemp.com The economic intuition behind UPP 26 Output $ Demand Marginal Revenue Marginal Cost Q P P Q
- 27. HoustonKemp.com Calculating UPP • Calculating profit maximising prices involves determining the FOC and setting this equal to 0 • Upward pricing pressure when FOC>0 • So UPP only measures direction of effect 27
- 28. HoustonKemp.com Calculating UPP (firm 1) 28 Pressure for prices to rise Pressure for prices to fall Marginal cost reduction (firm 1) as fraction of the price e1 = ∆𝑐1 𝑝1 Profit margin (firm 2) p2−c2 p2 Diversion ratio (from 1 to 2) D12 = − ε21 ε1 × Q2 Q1 𝑝2 𝑝1 p2−c2 p2 × 𝐷12 − 𝑒1
- 29. HoustonKemp.com Sales Volume 10 15 30 Price 120 115 80 Marginal Cost 90 85 75 Margin 0.25 0.2609 0.0625 Market Share 23% 32% 45% UPP vs concentration measures 29 3,578Pre-merger HHI
- 30. HoustonKemp.com Using a market share approach 30 55% 68% Post merger HHI: 5,050 Change in HHI: 1,472 Post merger HHI: 5,648 Change in HHI: 2,070
- 31. HoustonKemp.com Using UPP 31 𝑝2 𝑝1 p2 − c2 p2 ∗ 𝐷12 − 𝑒1 DGS = 0.5 UPPG = 0.025 UPP = 115 120 ∗ 0.261 ∗ 0.5 − 0.1 = 0.025
- 32. HoustonKemp.com Using UPP 32 𝑝2 𝑝1 p2 − c2 p2 ∗ 𝐷12 − 𝑒1 DGS = 0.5 UPPG = 0.025 UPP = 120 115 ∗ 0.258 ∗ 0.4 − 0.1 = 0.0043 D 𝑆𝐺 = 0.4 UPPS = 0.0043
- 33. HoustonKemp.com Using UPP 33 DGS = 0.5 UPPG = 0.025 UPP = 80 120 ∗ 0.0625 ∗ 0.25 − 0.1 = −0.0896 D 𝑆𝐺 = 0.4 UPPS = 0.0043 DG𝑃 = 0.25 UPPG = −0.0896 𝑝2 𝑝1 p2 − c2 p2 ∗ 𝐷12 − 𝑒1
- 34. HoustonKemp.com Using UPP 34 DGS = 0.5 UPPG = 0.025 UPP = 120 80 ∗ 0.25 ∗ 0.15 − 0.1 = −0.0438 D 𝑆𝐺 = 0.4 UPPS = 0.0043 DG𝑃 = 0.25 UPPG = −0.0896 UPP𝑃 = −0.0438 D 𝑃𝐺 = 0.15 𝑝2 𝑝1 p2 − c2 p2 ∗ 𝐷12 − 𝑒1
- 35. HoustonKemp.com Market share and UPP approaches give different results 35 68% DG𝑃 = 0.25 UPPG = −0.0896 UPP𝑃 = −0.0438 D 𝑃𝐺 = 0.15 Post merger HHI: 5,648 Change in HHI: 2,070 Market share UPP
- 36. HoustonKemp.com Benefits and limitations of using UPP • Benefits › UPP is relatively easy to calculate › Does not rely on market definition • Limitations › Assumptions about functional form for demand › Does not account for supply-side response (entry and repositioning) multi-product nature of firms (eg effect on complementary products) response to price changes by other firms › Often relies on surveys or switching analysis for diversion ratios 36
- 37. HoustonKemp.com Vertical restraints How economics can help assess the effect of vertical restraints on competition
- 38. HoustonKemp.com Outline • What are vertical restraints? • How economists can help to assess the competitive effects of vertical restraints • Pro-competitive effects of vertical restraints • Anti-competitive effects of vertical restraints • Overall assessment 38
- 39. HoustonKemp.com What are vertical restraints? 39 Coca-Cola Coles Downstream Upstream Supply agreement
- 40. HoustonKemp.com Common vertical restraints • Selective distribution • Exclusive distribution • Exclusive supply • Single branding • Quantity forcing • Restraints to prices › Resale price maintenance › Most favoured nation contracts 40
- 41. HoustonKemp.com Vertical restraints are important in the economy • Vertical restraints are common › Includes lots of types of conduct • Can be pro and anti competitive • Topical issue 41
- 42. HoustonKemp.com Vertical restraints can be prohibited under the Competition and Consumer Act • Anti-competitive agreements (s 45) › SLC test • Misuse of market power (s 46) › Substantial market power › Take advantage of market power › Damage an actual or potential competitor • Exclusive dealing (s 47) › Parts per se, parts subject to SLC test • Resale price maintenance (s 48) › Per se offence 42
- 43. HoustonKemp.com Economics can help determine whether there has been a substantial lessening of competition 43 How does the market function? What are the credible theories of harm and benefit? Does the evidence support a credible theory of harm or benefit? • What additional information is needed? • Market definition • Use qualitative and quantitative evidence • What is the likely net effect of the conduct? • Theories of harm need to be consistent with how market operates • Under what conditions would theory of harm be true? SLC • Relate back to competition law
- 44. HoustonKemp.com Vertical restraints prevent double markup 44 Coca-Cola Coles Consumers = (Manufacturing cost *Markup)*Markup Wholesale price = Manufacturing cost *Markup *MarkupWholesale priceRetail Price =
- 45. HoustonKemp.com Vertical restraints can prevent free-riding 45 Manufacturer High quality retailer Discount Store Customer visits store Customer purchases product
- 46. HoustonKemp.com Vertical restraints can prevent hold-up 46 Manufacturer Retailer 1. Investment in crates 2. Refuses to pay for crates (hold up)
- 47. HoustonKemp.com Vertical foreclosure can restrict competition 47 Manufacturer A Retailer A Retailer B Manufacturer B
- 48. HoustonKemp.com Vertical restraints can facilitate downstream collusion 48 Manufacturer A Retailer A Retailer B Agree retail price Agree retail price Agree to collude using vertical restraints
- 49. HoustonKemp.com Vertical restraints can facilitate collusion upstream 49 Manufacturer A Retailer A Agree retail price Agree retail price Manufacturer B Collusion over wholesale prices
- 50. HoustonKemp.com Vertical restraints can solve a monopolist’s commitment problem 50 Franchise A Franchisor New Town Franchise B Franchise C
- 51. HoustonKemp.com Vertical restraints can soften competition 51 Manufacturer A Manufacturer B Retailer A Retailer B
- 52. HoustonKemp.com Risk of vertical restraints being anti-competitive increases with market power 52 Risk of anti-competitive effect increasing No market power Downstream market power only Upstream and downstream market power Upstream market power only
- 53. HoustonKemp.com Assessing vertical conduct Do any of the firms involved have significant market power? Is there an anti- competitive theory of harm that is consistent with the facts? Is there a pro- competitive theory that is consistent with the facts? Weigh up pro and anti- competitive effects 53 No No SLC Yes, significant benefit, and little harm No SLC Yes Small, or no significant benefit No No SLC Yes
- 54. HoustonKemp.com Sydney Level 40 161 Castlereagh Street Sydney NSW 2000 Phone: +61 2 8880 4800 Singapore 12 Marina View #21-08 Asia Square Tower 2 Singapore 018961 Phone: +65 6653 3420 Contact Us