Vertical Integration in Video Broadcasting the basic competitive

Vertical Integration in Video Broadcasting: the basic competitive effects Roger Ware, Queen’s University Confidential, 1

Outline 1. Overview of market structure 2. An accelerating trend of vertical integration of programming networks and distributors 3. The traditional “Industrial Organization Economics” model of vertical integration 4. The bargaining model of vertical integration 5. Why they yield dramatically different conclusions with respect to competitive effects Confidential, 2

Market Structure in Programming and Distribution Some descriptive charts Confidential, 3

Programming Market Structure Distribution Market Structure (Share by Revenues) (Share by Subscribers) CANADA Bell Media (CTV+Astral) Shaw Media (Can. West) Rogers Bell Canada Enterprises Videotron (Quebecor) CBC Remstar U. S. A. Videotron Cogeco Corus Telus Other Disney/ABC Comcast/NBC News Corp/Fox Time Warner CBS Viacom A&E Discovery Others Comcast Direc. TV Dish Network Time Warner Cox Communications Charter Communications Others Confidential, 4

Distribution Market Structure (2006) Overall Large Metro Other CANADA DTH Cable Over-the-Air U. S. A. ADS Cable Over-the-Air Market Shares in Distribution (by share of TV-Owning Households) Confidential, 5

Geographic Markets for Distribution Are Local • Note that distribution markets are local, in that subscribers can only choose between services that are actually available at their address – which generally means one cableco, 2 satellite providers, over-the-air TV, and, in some areas IPTV provided by fibre optic cable (Bell and Telus are the major operators of fibre TV). • A typical US urban market is similar. Confidential, 6

The Big Vertical Mergers 1. Canada: Quebecor-TVA (2001); Rogers-City. TV (2007); Shaw-Can. West (2010); BCE-CTV (2011). BCE-Astral Media (2012) 2. U. S. A. : Direc. TV (Hughes-News Corp. (2003), subsequently unwound) ; Comcast-NBCU (2010); ATT – Time Warner (2018). Confidential, 7

What are the Competition/Regulatory Concerns? • Increased prices for programming content that will feed through to subscription prices for consumers. • Possible foreclosures: some customers may lose access to some programming content (major concern is pricing for programming – can be called partial foreclosure). • Effect on incentives for programming production and innovation? Confidential, 8

The Regulatory Response So Far • The response has been one of imposing behavioural or regulatory remedies, rather than structural intervention (divestitures or non-approval of mergers). • An open question is whether these behavioural remedies are sufficient to police the anti-competitive effects of the mergers. Confidential, 9

Traditional Economic Model of Vertical Integration Confidential, 10

The “ 2 X 2” Model of Vertical Integration U 1 D 1 U 2 D 2 Confidential, 11

Traditional Industrial Organization Economics of Vertical Integration 1. The Double Marginalization Effect: a downstream distributor will face a marginal cost equal to the wholesale price. If this has already been marked up by the upstream supplier above their marginal cost, then there will be “double marginalization” causing prices to the final consumer to be “too high” even from the perspective of joint profit maximization of the two firms. 2. Vertical Integration is usually assumed to eliminate this problem, leading to an increase in pricing efficiency, or “the efficiency effect of vertical integration”. Confidential, 12

Strategic Effects (Raising Rivals Costs): Vertical Integration can be profitable if it leads to input cost increases for downstream rivals. In the broadcasting context this would take the form of an increase in programming costs to unaffiliated distributors. Confidential, 13

Foreclosure: some terminology • Partial Foreclosure - Vertical integration occurs, but the integrated firm continues to potentially both sell inputs into the competitor downstream market and to buy inputs from the competitor upstream market. • Complete Foreclosure – after Vertical Integration the integrated firm neither buys nor sells inputs from or to other firms. Confidential, 14

The Traditional Analysis • This problem has been studied for many decades. One of the more recent and sophisticated treatments is Church, Majumdar and Baldauf (2011). • The consensus from this traditional analysis is that efficiency effects are likely to dominate strategic effects in all but exceptional cases. • “In symmetric cases, the literature suggests that the output expansion effects from the internalization of double marginalization exceed any raising rivals’ costs effect. Confidential, 15

Strategic Structure • A key feature of traditional models is the upstream suppliers make take-it-or-leave-it offers to downstream firms. • The downstream oligopoly (of distributors) take the programming fees as given and choose their rates for subscribers in competition with their rival distributors. • The upstreaming programming networks set programming fees based on derived demand for programs (or bundles of programs) Confidential, 16

A Realistic Simplification I am going to make a further simplifying assumption, that each of the upstream programming networks are essential i. e. to get any subscribers at all, a distributor must offer these programs (think CTV (which includes TSN) prior to the Bell merger). Given this assumption, we can just consider each programming network separately, as in the following diagram. Confidential, 17

Each upstream programming network is essential Programming Network (essential) Distributors

The Traditional Model Where Both Programming Packages Are “Must Have” § With programming packages essential, in the traditional model there is no strategic effect, because before vertical integration the upstream programming company can charge any price that they want to (i. e. , the monopoly price) to each distribution platform. § As a result of VI, there will be an efficiency effect as the input price of the affiliated platform drops to marginal cost. § Therefore, the efficiency effect must dominate and VI will should be pro-competitive (i. e. , good for consumers). Confidential, 19

The Bargaining Model of Vertical Integration Confidential, 20

The Bargaining Model • William Rogerson (former FCC Chief Economist) has developed the most sophisticated application of this model, in the context of the Comcast-NBCU transaction. • Let us return to the model, and again assume that both programming packages are essential (i. e. , a distributor must offer them if they wish to retain subscribers). • In effect, that makes programming suppliers monopolists. Confidential, 21

Review of the Nash Bargaining Solution (NBS) • One Seller, One Buyer. Assume a suite of programs, already produced, costs are sunk. Seller’s reservation price = $0, Buyer’s reservation price = $200. Nash Bargaining Solution suggests they will split the surplus equally, and price = $100. • Now suppose that the Seller does have an opportunity cost, the ability to make a profit by selling the programs to their own downstream affiliate after VI. Suppose the profit from doing this would be $50. Then NBS indicates that price will increase by $25 (the surplus from trade is still split equally) to $125. Confidential, 22

Calculation of Increase in Programming Fees Formula for calculating the increase in programming fees: where π = profit per subscriber from affiliated distributor, α = fraction of leaving subscribers that switch to the affiliated distributor, and d = fraction of subscribers to unaffiliated distributors that will leave if programming is withdrawn Note: possible that both α and d = 1 (e. g. if all Cogeco subscribers would leave without CTV/TSN and all would switch to Bell) [Source: William Rogerson] Confidential, 23

Nash Bargaining Model • In this model the strategic effect is always positive (i. e. , always leads to an increase in prices for programming, because the opportunity cost of the affiliated program supplier always increases after VI). • But what about Double Marginalization, or the Efficiency effect? Confidential, 24

When is the Double Marginalization (efficiency) effect just plain wrong? • After VI what is the opportunity cost of adding another subscriber to the affiliated distributor? • The conventional double marginalization theory says that this cost falls from the wholesale price w to the true marginal cost, which we are assuming to be zero. Confidential, 25

Double Marginalization (Efficiency) Effect (cont’d) • But … if the new subscriber is switching from the other unaffiliated distributor, then the opportunity cost of adding another subscriber is actually the lost profit from selling the programming package to the unaffiliated distributor – which is approximately w. • In other words, there is no double marginalization, or efficiency, effect. Confidential, 26

Nash Bargaining Theory • Again, sticking within our simple modelling framework where each programming package is viewed as essential by subscribers. • In the Nash Bargaining model, there is no efficiency effect of vertical integration but there is always a strategic effect that leads to an increase in prices for programming. • Vertical Integration is always anti-competitive. Confidential, 27

Review • In the traditional model, there is no strategic effect, and thus the efficiency effect must dominate, so that the vertical merger is unambiguously pro-competitive. • In the bargaining model, there is no efficiency effect (to first order) and thus the strategic effect must dominate, so that the vertical merger is unambiguously anticompetitive. Confidential, 28

Extensions and Qualifications • There may be some efficiency effect of Vertical Integration for two reasons: i. If new subscribers to the affiliated platform (Bell Media) do not all come from an unaffiliated distributor, then the opportunity cost will drop accordingly (note the existence of Shaw Direct does not have this effect, as they are carrying CTV network programs also). Rogerson shows that this effect is likely to be small, for example if 5% of new subscribers come from households not subscribing elsewhere to the CTV network programs, then the opportunity cost of adding another subscriber is 95% of the foregone profits from selling to Cogeco and Shaw. ii. There may be production efficiency effects of VI, rather than effects due to “unravelling double marginalization”. Confidential, 29

Broadcasting Regulatory Policy CRTC 2011 -601 • In 2011, the CRTC imposed some restrictions on vertical integration among programming and broadcasting companies. These behavioural remedies included: § Restriction on exclusivity; § More programming choices, such as “pick and pay” for consumers § Changes to dispute resolution arrangements, including standstill and reverse onus provisions Confidential, 30

Can Regulation adequately prevent a Substantial Lessening of Competition arising from Vertical Mergers? • Regulation would have to effectively prevent the increases in programming fees that economic theory predicts • The dispute resolution mechanisms implemented by the CRTC may not achieve this goal. • It is still possible for dominant programming networks, to in effect discriminate against unaffiliated distributors by increasing fees, delaying dispute resolution, and offering de facto exclusive content while disputes are being resolved. Confidential, 31

Conclusions and Review • The “conventional wisdom” on the economics of vertical integration – that vertical integration is generally pro-competitive – is wrong when applied to large vertical integration events in video broadcasting markets. • The bargaining model, which has been accepted by both the FCC and academic Industrial Organization economists, predicts that such vertical mergers will always be anti-competitive. • Key assumptions are (i) that the integrating upstream programming network is essential for downstream distributors, whether affiliated or not; (ii) most subscribers who switch to the affiliated distributor come from a distributor also carrying the integrated firm’s programming package. Confidential, 32

The AT&T – Time Warner Merger 2017 -18 • $85 billion acquisition • Contested by US DOJ in high profile antitrust trial 2018 • Judge ruled against the US government • Arguments at trial by expert economists were exactly the ones we have discussed Confidential, 33