17 Jan 2008Working Paper Summaries

Competition in Modular Clusters

by Carliss Y. Baldwin and C. Jason Woodard

The last 20 years have witnessed the rise of disaggregated "clusters," "networks," or "ecosystems" of firms in a number of industries, including computers, telecommunications, and pharmaceuticals. In these clusters, different firms design and produce the various components of a complex artifact (such as the processor, peripherals, and software of a computer system), and different firms specialize in the various stages of a complex production process. This paper considers the pricing behavior and profitability of these so-called modular clusters. Baldwin and Woodard isolate the offsetting price effects in a model, and show how they might operate in large as well as in small clusters. Key concepts include:

Clusters operating under open, public standards may have higher prices and profits than those operating under closed, proprietary standards.
Cluster forms of industrial organization may not be conducive to all kinds of innovation. In particular, innovations that add new layers of functionality to the system, and thus increase total demand, will not be adequately rewarded relative to the value they create.
It is important to learn how cluster configurations affect incentives to supply different forms of innovation, and how firms respond to these cross-layer dependencies in formulating their long-term strategies.

Author Abstract

The last twenty years have witnessed the rise of disaggregated "clusters," "networks," or "ecosystems" of firms. In these clusters the activities of R&D, product design, production, distribution, and system integration may be split up among hundreds or even thousands of firms. Different firms will design and produce the different components of a complex artifact (like the processor, peripherals, and software of a computer system), and different firms will specialize in different stages of a complex production process. This paper considers the pricing behavior and profitability of these so-called modular clusters. In particular, we investigate a possibility hinted at in prior work: that for composite goods, a vertical pricing externality operating across complements can offset horizontal competition between substitutes. In this paper, we isolate the offsetting price effects and show how they operate in large (as well as small) clusters. We argue that it is possible in principle for a modular cluster of firms to mimic the pricing behavior and profitability of a vertically integrated monopoly. We then use our model to compare open and closed standards regimes, to understand how commoditization affects a cluster, to determine the relative profits of platform firms and firms that depend on the platform, and to assess the impact of horizontal and vertical mergers. Our model highlights a collective action problem: what is good for an individual firm is often not good for the cluster. We speculate that this conflict may be a source of strategic tension in platform firms.