Revealing Incentives for Compatibility Provision in Vertically Differentiated Network Industries

• Author: Filomena Garcia,Cecilia Vergari
• Published date: 01 September 2016
• DOI: http://doi.org/10.1111/jems.12146
• Date: 01 September 2016

Revealing Incentives for Compatibility Provision in

Vertically Differentiated Network Industries

FILOMENA GARCIA

Indiana University and Research Unit on Complexity and Economics (UECE)

ISEG (School of Economics and Management)

Universidade de Lisboa

filosgarcia@gmail.com

CECILIA VERGARI

Department of Economics

University of Bologna

Bologna, Italy

cecilia.vergari@unibo.it

We determine the incentives for compatibility provision of firms that produce network goods

with different intrinsic qualities when firms do not have veto power over compatibility. When

network effects are strong, there are multiple equilibria in pricing and consumer decisions. We

show that in some equilibria, it is the high-quality firm that invests in compatibility, whereas

in others, the low-quality firm triggers compatibility. The socially optimal compatibility degree

is zero, except under very strong network effects, where one of the equilibria has all consumers

buying the low-quality good. In this case, a partial degree of compatibility is optimal.

1. Introduction

Since Rohlfs (1974) first described his theory of the interdependent demand for com-

munication services, economists have studied different aspects of the so-called network

industries. These are characterized by the existence of positive externalities in consump-

tion, that is, the utility of agents is increasing in the number of users of the same good or

compatible goods. The consequence of this is that consumers choosing one such good

always forego the benefits of interacting with the agents that choose other (incompatible)

goods. A possible solution for firms to increase the willingness to pay of the consumers

is to establish some degree of compatibility. Indeed, firms decide whether to make their

goods compatible with those of their rivals, thus competing in the market, or to render

them incompatible thus competing for the market.1

We thank two anonymous referees,the editor, and the co-editor for detailed and constructive comments. We

are grateful to O. Amerighi, R. Amir, R.A. Becker, P. Belleflamme, P. Colla, J. Correia da Silva, G. De Feo,

V. Denicol`

o, M. Denni, L. Filippini, J. Gabszewicz, L. Lambertini, L.D. Opromolla, M. Peitz, M. Tirelli, G.

Torrens,V. Vannetelbosch, X. Wauthy,and several seminar participants. Financial support by FCT (Fundac¸˜

ao

paraaCi

ˆ

encia e Tecnologia, Portugal) is gratefully acknowledged. This paper is part of project PTDC/EGE-

ECO/122507/2010.

1. As Besen and Farrell (1994) put forward, “there is no general answer to the question of whether firms

will prefer competition for the potentially enormous prizes under inter-technology competition, or the more

conventional competition that occurs when there are common standards.” An example is Microsoft and Sony

Team on Digital Entertainment Content Management System: although rivals in the gaming-console market,

both companies found mutual gains in working closely to integrate the Sony VAIOXL1 Digital Living System

with Microsoft Windows XP Media Center Edition 2005.

C2015 Wiley Periodicals, Inc.

Journal of Economics & Management Strategy, Volume25, Number 3, Autumn 2016, 720–749

Incentives for Compatibility Provision 721

The objective of this paper is to determine the incentives for compatibility provision

of vertically differentiated firms operating in a network market. Our paper is the first

to concentrate on the provision of partial compatibility. We also extend the literature by

determining the compatibility incentives under strong network effects, that is when the

importance of the network for consumers is so high that it induces multiple equilibria

in consumption.

We can think of several examples for which our analysis could be used. These

examples have in common the existence of direct or indirect network effects in con-

sumption and differentiated quality of the products offered by distinct firms.2As a first

example of a good with direct network effects we can think of Basic LEGO Bricks. This

is a network good as the more compatible bricks you and your friends have, the larger

your building possibilities. Although LEGO remains the most famous bricks-maker and

is perceived by consumers as a high-quality good (at least for its long history and brand

recognition), there are competitors, like Mega Blocks that might offer or not a compati-

ble product.3Other examples, mostly characterized by indirect network effects, are: the

game console industry, the operating systems industry and the high-definition DVD

industry.T hese industries can be classified as two-sided markets in the sense that users of

one group (game console’s, OS’s, and high-definition DVD’s users) are better off as the

number of users of the other group (game, software, and DVD developers) increases.4

We do not model explicitly the two-sided market and focus on the role of (indirect)

network effects. We can say that a consumer’s utility for a particular product increases

(indirectly) with the number of consumers of the same or compatible products when

the availability of components or support services increases with consumption. In these

industries, it is frequent to find that one firm is perceived by the consumers as being the

high-quality firm. For instance, in the case of MAC vs. Microsoft Windows operating

systems, it is often argued that Apple Macintosh is a higher quality product, for example,

in terms of high-resolution graphics, virus detection, etc., whereas, historically and also

due to large price differences, Microsoft has captured the largest market share. Another

example is provided by the high-definition DVD industry. HD-DVD was produced by

Toshiba and had storage potential around 40% lower than the Blu-Ray DVD produced

by Sony. Also, we can think about the car industry in which consumers are better off if

others buy the same type of vehicle as the supply of customer services will be wider.

Compatibility can be achieved either by standardization, or by the introduction of

aconverter, a device which allows consumers of one product to enjoy (partially or fully)

the network of the other product.5Often, the converter device represents a compromise

of quality, in the sense that compatibility may be imperfect. Likewise, standardization

is a costly requirement because it limits product variety. In our model, we assume that

2. It is common to distinguish between direct network effects, stemming from direct interaction of the

consumers, and indirect network effects, which stem from other externalities related to the dimension of the

consumer base, such as increased quality,services and components, etc.

3. We borrowthis example from Belleflamme and Peitz (2010, p. 579).

4. For an analysis of competition in two-sided markets, see Gabszewicz and Wauthy(2014) and Armstrong

(2006).

5. For instance, in the HD-DVD/Blu-Ray case, producers of DVD discs at some point have come out with

a disc which had the HD-DVD version of the movie on one side and the Blu-Ray version on the other.Also, car

producers share the service facilities, rendering their products compatible. For instance, a buyer of a Nissan

sedan can enjoy the indirect network effects of the Renault service facilities. We borrow this example from

Alexandrov (2015).