Mergers with interfirm bundling: a case of pharmaceutical cocktails

• Author: Minjae Song,Claudio Lucarelli,Sean Nicholson
• DOI: http://doi.org/10.1111/1756-2171.12192
• Date: 01 August 2017
• Published date: 01 August 2017

RAND Journal of Economics

Vol.48, No. 3, Fall 2017

pp. 810–834

Mergers with interfirm bundling: a case

of pharmaceutical cocktails

Minjae Song∗

Sean Nicholson∗∗

and

Claudio Lucarelli∗∗∗

Pharmaceutical cocktails often consist of two or more drugs produced by competing firms.

The component drugs are often also sold as stand-alone products. We analyze the effects of a

merger between two pharmaceutical firms selling complements for colorectal cancer treatment.

In this setting there are two merger effects: the standard upward pricing pressure due to firms

internalizing the substitution between the stand-alone products,and an additional effect wherethe

firms internalize the impact of selling complements and reduce the price of the cocktail product.

The net impact of a merger is a modest price increase, or even a price decrease.

1. Introduction

Pharmaceutical combination therapies, or “cocktails,” are examples of interfirm bundling,

where a bundle consists of drugs produced by competing firms and the components of the bundle

are also sold as stand-alone products. In this article, we empirically analyze the welfare effectsof

a merger between two firms selling products used in such bundling, focusing on pharmaceutical

treatment of colorectal cancer patients. In the colorectal cancer treatment market, most drugs used

in cocktails are also available as a stand-alone product, and physicians choose a single regimen

(at a point in time) for their patients. In the early 2000s, for example, five patent-protected drugs

were used in 12 major regimens. Six of these were stand-alone regimens; six cocktail regimens

∗Bates White Economic Consulting; minjae.song@bateswhite.com.

∗∗Cor nell Universityand NBER; sn243@cornell.edu.

∗∗∗Universidad de los Andes; claudio.lucarelli@uandes.cl.

We have benefited from discussions with Yongmin Chen, Joshua Lustig, Jae Nahm, Michael Raith, Michael Riordan,

Michael Waldman,and comments by seminar participants at Columbia University, University of Southern California, the

Cowles Foundation,the Federal Trade Commission, the 2009 Internal Industrial Organization Conference (Boston, MA),

Korea University, SKK University (Seoul, Korea), the University of Rochester,and the University of Wisconsin-Madison.

James Elwell provided excellent research assistance. We especially thank the Editor, Aviv Nevo, and two ananymous

referees who helped us improve this article substantially. All errors are ours.

810 C2017, The RAND Corporation.

SONG, NICHOLSON AND LUCARELLI / 811

were created by combining the five drugs in various ways. In 2008, 31% of US colorectal cancer

patients receiving drug treatment were administered a combination therapy, or cocktail regimen.1

If there were only an interfirm cocktail regimen, without stand-alone regimens available in

the market, the pricing problem would be effectively the same as a situation where two firms

unilaterally set prices of complements. When one firm raises the price of its component drug,

demand is reduced both for its own drug and the rival firm’s drug. Cournot (1838) shows that

the price of complements is higher when set unilaterally by competing firms than when set by a

monopolist. In such a setting, a merger can increase both producer surplus and consumer welfare

by eliminating double marginalization.2

However, when a drug is used in multiple regimens, including as a stand-alone product and as

a component of a cocktail regimen, it is not clear whether a merger between the twofirms sharing

a cocktail regimen would increase welfare. Although the firms would want to lower the price of

the cocktail product postmerger to internalize the effect of selling complements, which we refer

to as the complementarity effect, they would wantto raise prices for their stand-alone products to

internalize the substitution between these products, which we refer to as the market power effect.

Therefore, the welfare effects of a merger between any two firms producing components of a

cocktail will be ambiguous.

This trade-off created by an interfirm bundle is present in other industries. The merger

between AT&T and DIRECTV, approved by the Federal Communications Commission (FCC)

in July 2015, provides such an example (FCC, 2015). Prior to the merger, DIRECTV offered

customers video programming services only, whereas AT&T offered video only, broadband

Internet access only, and a bundle of those two products in some markets for customers seeking

a more interactive viewing experience. Although DIRECTV did not offer broadband services

before the merger, consumers could create a synthetic bundle consisting of video and broadband

components offered by two separate firms, including AT&T with DIRECTV, and firms other than

AT&T with DIRECTV. The merging fir ms were therefore producing complements that could be

consumed together as a “cocktail” as well as stand-alone products. Furthermore, the stand-alone

products and interfirm bundle competed with one another at the product level, which is also the

situation in our context.

Weestimate a structural model that allows us to analyze hypothetical mergereffects. We begin

by estimating a demand system at the regimen level using data on regimen prices, market shares,

and attributes. Regimens, which can be single drugs or cocktails of two or more drugs, are well

defined and standardized. The National Comprehensive Cancer Network (NCCN) recommends

the amount of each drug an oncologist should use in each regimen. “Market share” is defined as the

proportion of colorectal cancer patients treated with a particular regimen. Data from randomized

clinical trials provide information on attributes such as regimen efficacy (e.g., the median number

of months patients survived in the clinical trial) and side effects (e.g., the percent of patients in

the clinical trial who experienced abdominal pain).

We use the demand estimates and profit-maximization conditions to recover the marginal

cost of each drug. We then fix the marginal costs and demand parameters and conduct a series

of counterfactual merger scenarios, where we separately measure the market power and the

complementarity effects. To capture the former effect, we remove the cocktail regimen that two

given firms have in common and compare the premerger and the postmerger equilibrium prices.

We show that merging firms in the colorectal cancer treatment market would raise drug prices

substantially,more than twice in some cases, if they merged with no cocktail regimen in common.

1Combination therapies are also common in other segments of the pharmaceutical industry. Most HIV/AIDS

patients, for example, receive a cocktail regimen, such as a combination of efavirenz, lamivudine, and zidovudine.

Harvoni, a combination of the drugs ledipasvir and sofosbuvir, has a 95% cure rate for hepatitis C and generated $13.9

billion in sales in 2015.

2Although the pricing of complements has been studied theoretically in various settings, there are few empirical

analysesin the economic literature. Examples of theoretical treatments include Economides and Salop (1992), Economides

(1998), Davis and Murphy (2000), Choi (2008), Dari-Mattiacci and Parisi (2006), and Yan and Bandyopadhyay (2011).

C

The RAND Corporation 2017.