Coase's Penguin, Or, Linux and The Nature of the Firm

You don't have to be an economist or into open source programming to appreciate this lucid explanation of how computers and networks and new norms of "peer production" present a novel alternative to the market or the firm (the two modes of production Ronald Coase explained decades ago, for which he won the Nobel in economics). 

These summaries are part of the cooperation project, carried out in collaboration with Institute for the Future circa 2006. At that time, with the objective of catalyzing an interdisciplinary study of cooperation, we created these summaries so that cooperation researchers in different disciplines could become aware of and get up to speed on what others were doing.

Publication Date:  August 2002
URL:  https://dash.harvard.edu/handle/1/11363034

Authors: Benkler, Yochai



Findings:

• The  emerging pervasively networked information economy has four  characteristics that have enabled the emergence of peer production.  These are (1) information is non-rival and "the social cost of using  existing information as input to new information production is zero",  (2) the "decline in the capital cost of information production", (3)  information production relies on human talent and creativity which is  highly variable, (4) the "dramatic decline in communication costs".

• 
  

• Benkler hypothesizes, in short, that peer-based cooperation scales.  More specifically that "rich information exchange among large sets of  agents free to communicate and use existing information resources  cheaply will create sufficiently substantial information gains". And  that peer production has "potential allocation gains enabled by the  differences in how peer production, firms, and markets reduce  uncertainty" of production outcome. And finally that these information  gains together with allocation gains "overcome the added information  exchange costs necessary to overcome the absence of pricing and  managerial direction, and the added coordination costs created by the  lack of property and contract as institutional bases for structuring  coordination".

• 
  

• From (2) above he concludes that "where the physical capital costs  of information production are low, and where existing information  resources are freely or cheaply available, the low cost of cost of  communication of very large sets of agents allows agents to collect  information through extensive communication and feedback instead of by  using information compression mechanisms like prices or managerial  instructions." This will result in efficient information production and  is a basis to argue in favor of common or freely available information  over the current system that favors property.

• 
  

• Benkler claims that human intellectual effort is highly variable. As  a result, "human creativity is very difficult to standardize and  specify in the contracts and necessary for either market-cleared or  hierarchically organized production."  That is, there is an information  loss. Some companies try to compensate for this with incentive  compensation and other methods however, "it is unclear how well they can  overcome the core difficulty."

• 
  

• "Peer production relies on making an unbounded set of resources  available to an unbounded set of agents." Whereas market and firm based  production rely on bounded sets of agents and resources secured through  property and contract. "The permeability of the boundaries of these sets  is limited by the cost of making decisions in a firm."  This, coupled  with the variability of human talent, leads to allocation gains for peer  production over the other modes of production.

• 
  

• Benkler notices that not all projects are suited for peer  production, including ones that don't require highly variable talent or  that cannot be adequately partitioned.

• 
  

• This is all well and good, but it doesn't explain how peer  production actually works or more importantly how to design a successful  peer production project. There are obvious questions about (1)  potential substantial duplication of effort placing a drag on  production, (2) motivation and threats to motivation, and (3) the  integration of contributions.  For (1), redundancy provides a better  product and a successful project may simply be taking production time  from unproductive activities (e.g., watching TV) and not result in an  overall loss of productivity. For (2) Benkler notes that "given a  sufficiently large number of contribution, direct monetary incentives  necessary to bring about contributions are trivial" and that an  important part of a successful project is its ability to be decomposed  into small contributions. In addition, there are two forms of  demotivational activities: failure to integrate a contribution, and  taking over a project or portion of a project, both of which a  successful project must design to avoid. And finally for (3), a method  for successful integration and quality control is critical for any  successful project.

One Sentence:

Commons  based peer production (e.g., free software) has emerged in the  pervasively networked digital information economy as a third method of  production which for some projects, has productivity gains, in the form  of information and allocation gains, over market and firm-based  production.

One Paragraph:

The traditional  framework of the organization of economic production includes two modes  of production: individuals order their productive activities either  under the direction of managers at firms, or as individuals in markets  following price signals. Free Software is one example of a broader  social-economic phenomenon that Benkler calls 'commons-based peer  production', a new, third mode of production. Because of the highly  variable nature of human expertise, and given a pervasively networked  information economy, commons-based peer production has advantages over  the two traditional forms of organization in both information and  allocation gains. Motivation and organization are different in peer  production, Benkler concludes that (1) "Given a sufficiently large  number of contributions, direct monetary incentives necessary to bring  about contributions are trivial." and (2)  "Peer production is limited  not by the total cost or complexity of a project, but by its modularity,  granularity, and the cost of integration."

One Page:

The  traditional framework of the organization of economic production  includes two modes of production: individuals order their productive  activities either under the direction of managers at firms, or as  individuals in markets following price signals. Free Software is one  example of a broader social-economic phenomenon that Benkler calls  'commons-based peer production', a new, third mode of production in  digitally networked environments. In order to explain the emergence of  this third mode of production Benkler augments the traditional  production framework. Because of the highly variable nature of human  expertise, and given a pervasively networked information economy,  commons-based peer production has advantages over the two traditional  forms of organization. The paper concludes with a discussion of the  problems of motivation, loss of motivation, and integration in peer  production enterprises. Benkler concludes that (1) "Given a sufficiently  large number of contributions, direct monetary incentives necessary to  bring about contributions are trivial." and (2)  "Peer production is  limited not by the total cost or complexity of a project, but by its  modularity, granularity, and the cost of integration."

The paper concludes with a discussion of the problems of collective  action and how they are solved in the absence of property and the  presence of high transaction costs of monetary compensation. Relevant  factors include the fact that the resource being produced (information)  is non-rival, that problems are divisible into a fine level of  granularity, the ability to provide integration (quality control and  handling of contributions) in a socially acceptable manner, that the  pervasively networked information economy provides access to a large  number of potential contributors, and the willingness of contributors to  accept non-monetary rewards.

Benkler posits that understanding peer production in the same  framework as the mainstream economic theory of organizations could  explain the emergence of commons-based peer production. The mainstream  economic theory of  organizations says that individuals organize into  firms whenever the cost of achieving an outcome is greater using a price  system. Peer production emerge whenever the cost of peer-based  production is lower than either market-based or firm-based production.  Property rights emerge whenever the value of a resource is such that its  utilization through a property-based appropriation offsets the cost of  implementing and enforcing the property rights regime. Commons emerge  when the cost of implementing a property regime is higher than the  opportunity cost of the property. Market and firm based production can  be divided into property based production and commons-based production.  Peer-based production fits well into the framework with plenty of  examples of both property based production (e.g., Xerox's Eureka) and  commons-based peer production (e.g., free software, academic science,  Wikipedia).

The emergence of peer production is tied to a pervasively networked  information economy. Commons-based peer production has systematic  advantages over market and firm based production when (1) the object of  production is information or culture, and (2) the physical capital  necessary for production is widely distributed. Both of the advantages  of peer production are a function of the variability of human capital.  First, commons-based peer production has an advantage of having a lowest  cost of determining who is the best person for a given task (Benkler  calls this 'information opportunity cost'). Second, it has an advantage  of allocation efficiency where large groups of potential contributors  interact with large groups of resources in the search for new tasks.  That is, the practice of firms -- and to a lesser extent markets -- of  securing access to limited sets of contributors and resources through  contracts and property entails a systematic loss of productivity.

Benkler addresses the problems of motivation, loss of motivation, and  integration in peer production enterprises. Benkler concludes the  following:

• "Given a sufficiently large number of contributions, direct monetary  incentives necessary to bring about contributions are trivial."

• "Peer production is limited not by the total cost or complexity of a  project, but by its modularity, granularity, and the cost of  integration."

Two kinds of actions represent threats to motivation (1)  (the most  important) unilateral appropriation by an individual or group of the  project and (2) some behavior affecting the intrinsic value of  participation for contributors (e.g., failure to integrate a  contribution). Free-riding is a common demotivating action in commons.  Since information is non-rival, free-riding is a non-issue so long as  the pool of contributors is sufficiently large and the act of  free-riding does not undermine production.  In this sense, 'Absence of  exclusion' is the organizing feature of commons-based peer production.  Finally, integration requires (1) a quality control or integrity  assurance mechanism, and (2) a method for combining individual  contributions into the whole.