Person:

Baldwin, Carliss

The modern corporation has long been the central focus of the field of organization design. Such firms can be likened to nation-states: they have boundaries that circumscribe citizen-employees, and they engage in production and trade. But individual corporations are no longer adequate to serve as the primary unit of analysis. Over the years, systems of distributed innovation – so-called business ecosystems – have become increasingly prevalent in many industries (Adner & Kapoor, 2010; Iansiti & Levien, 2004; von Hippel, 1988). Ecosystems generally encompass numerous corporations, individuals, and communities that might be individually autonomous but related through their connection with an underlying, evolving technical system.

In the future, I believe the key problem for organization design will be the management of distributed innovation in such dynamic ecosystems. Specifically, how should diverse entities be integrated into a coherent network that generates goods in the present and new designs for the future? To answer that question, organization designers must think about how to distribute property rights, people, and activities across numerous self-governing enterprises in ways that are advantageous for the group (ecosystem) as well as for the designer’s own firm or community.

Many products are manufactured in networks of firms linked by transactions, but comparatively little is known about how or why such transaction networks differ. This article investigates the transaction networks of two large sectors in Japan at a single point in time. In characterizing these networks, our primary measure is “hierarchy,” defined as the degree to which transactions flow in one direction, from “upstream” to “downstream.” Our empirical results show that the electronics sector exhibits a much lower degree of hierarchy than the automotive sector because of the presence of numerous inter-firm transaction cycles. These cycles, in turn, reveal that a significant group of firms have two-way “vertically permeable boundaries”: (i) they participate in multiple stages of an industry’s value chain, hence are vertically integrated, but also (ii) they allow both downstream units to purchase intermediate inputs from and upstream units to sell intermediate goods to other sector firms. We demonstrate that the 10 largest electronics firms had two-way vertically permeable boundaries while almost no firms in the automotive sector had adopted that practice.

In this paper, we describe an operational methodology for characterizing the architecture of complex technical systems and demonstrate its application to a large sample of software releases. Our methodology is based upon directed network graphs, which allows us to identify all of the direct and indirect linkages between the components in a system. We use this approach to define three fundamental architectural patterns, which we label Core—periphery, multi-core, and hierarchical. Applying our methodology to a sample of 1,286 software releases from 17 applications, we find that the majority of releases possess a "core-periphery" structure. This architecture is characterized by a single dominant cyclic group of components (the "Core") that is large relative to the system as a whole as well as to other cyclic groups in the system. We show that the size of the Core varies widely, even for systems that perform the same function. These differences appear to be associated with different models of development—open, distributed organizations develop systems with smaller Cores, while closed, co-located organizations develop systems with larger Cores. Our findings establish some "stylized facts" about the fine-grained structure of large, real-world technical systems, serving as a point of departure for future empirical work.

Our paper tests a key prediction of property rights theory: that agents respond to marginal incentives embedded in property rights, when making non-contractible, revenue-enhancing investments. (Grossman and Hart, 1986; Hart and Moore, 1990). Using rich project-level data from the US film industry, we exploit variation in property right allocations, investment choices, and film revenues to test the distinctive aspects of property rights theory. Empirical tests of these key theoretical predictions have been relatively sparse due to the lack of appropriate data. The US film industry displays two distinct allocations of property rights, which differentially affect marginal returns on a particular class of investments. Studio-financed films are produced and distributed by studios which take in the lion’s share of revenue. In contrast, independent films are distributed by studios under revenue sharing agreements, which give studios 30-40% of the revenue stream. Under either regime, the allocation of scarce marketing resources is determined by and paid for by the studio. After accounting for the endogenous nature of property-right allocations, we find that studio-financed films receive superior marketing investments compared to independent films and that these investments fully mediate the positive effect of vertical integration on film revenues. As a result, this study contributes to the empirical literature on property rights by showing that both predicted linkages (from marginal returns to investment and from investment to revenue) exist in a single empirical setting.

Distributed value creation can boost the overall value created, but may create serious challenges for capturing value. In order to draw in external contributors, an innovator often waives legal exclusion rights or reveals formerly exclusive knowledge. But as a result, contributors may appropriate a large share of the jointly created value. In turn, integrating external contributions entails the risk of becoming dependent on outside owners of IP. To address this tension we propose the concept of “IP modularity.” We argue that, by managing a system’s modular structure in conjunction with its IP, firms can reconcile distributed innovation with value capture.

How do firms create and capture value in large technical systems? In this paper, I argue that the points of both value creation and value capture are the system’s bottlenecks. Bottlenecks arise first as important technical problems to be solved. Once the problem is solved, the solution in combination with organizational boundaries and property rights can be used to capture a stream of rents. The tools a firm can use to manage bottlenecks are, first, an understanding of the technical architecture of the system; and, second, an understanding of the industry architecture in which the technical system is embedded. Although these tools involve disparate bodies of knowledge, they must be used in tandem to achieve maximum effect. Dynamic architectural capabilities provide managers with the ability to see a complex technical system in an abstract way and change the system’s structure to manage bottlenecks and modules in conjunction with the firm’s organizational boundaries and property rights.

We test a method that was designed and used previously to reveal the hidden internal architectural structure of software systems. The focus of this paper is to test if it can also uncover new facts about the components and their relationships in an enterprise architecture, i.e., if the method can reveal the hidden external structure between architectural components. Our test uses data from a biopharmaceutical company. In total, we analyzed 407 components and 1,157 dependencies. Results show that the enterprise structure can be classified as a core-periphery architecture with a propagation cost of 23%, core size of 32%, and architecture flow through of 67%. We also found that business components can be classified as control elements, infrastructure components as shared, and software applications as belonging to the core. These findings suggest that the method could be effective in uncovering the hidden structure of an enterprise architecture.

Modularity is a means of partitioning technical knowledge about a product or process. When state-sanctioned intellectual property (IP) rights are ineffective or costly to enforce, modularity can be used to hide information and thus protect IP. We investigate the impact of modularity on IP protection by formally modeling the threat of expropriation by agents. The principal has three options to address this threat: trust, licensing, and paying agents to stay loyal. We show how the principal can influence the value of these options by modularizing the system and by hiring clans of agents, thus exploiting relationships among them. Extensions address screening and signaling in hiring, the effects of an imperfect legal system, and social norms of fairness. We illustrate our arguments with examples from practice.

Employing software metrics, such as size and complexity, for predicting defects has been given a lot of attention over the years and proven very useful. However, the few studies looking at software architecture and vulnerabilities are limited in scope and findings. We explore the relationship between software vulnerabilities and component metrics (like code churn and cyclomatic complexity), as well as architecture coupling metrics (direct, indirect, and cyclic coupling). Our case is based on the Google Chromium project, an open source project that has not been studied for this topic yet. Our findings show a strong relationship between vulnerabilities and both component level metrics and architecture coupling metrics. Unfortunately, the effects of different types of coupling are somewhat hard to distinguish.

This paper seeks to explain the technological forces that led to the rise of vertically integrated corporations in the late 19th Century and the opposing forces that led to a vertical-to-horizontal transition in the computer industry one hundred years later. I first model the technology of step processes with bottlenecks and show how this technology rewards vertical integration, a hierarchical organization, and the use of direct authority. These properties in turn became the organizational hallmarks of so-called “modern” corporations. I then model platform systems, showing that, in contrast to step processes, this technology rewards the multiplication of options, increasing risk, and modularity. Moreover, given a modular architecture, a platform system can be open, with different components supplied by separate firms with no loss of interoperability or efficiency. Openness multiplies options and expands diversity, thus increasing the platform system’s value. The last two decades of the 20th Century saw the rise of three distinct types of open platforms in the computer industry: (1) “forward open” platforms with downstream complementors; (2) “backward open” modular supply networks; and (3) “open exchange” platforms designed to facilitate transactions and other forms of social interaction. Whereas in 1980, vertically integrated firms dominated the industry, by 2000, the “verticals” had essentially disappeared. The largest firms in the industry in 2000 were sponsors and participants in open platform systems. I argue that the vertical-to-horizontal transition in the computer industry was an organizational response to a fundamental change in economic rewards to the technologies of rationalized step processes vs. open platform systems.