Open source methods for creating software rely on developers who voluntarily reveal code in the expectation that other developers will reciprocate. Open source incentives are distinct from earlier uses of intellectual property, leading to different types of inefficiencies and different biasesin R&D
investment. Open source style of software development remedies a defect of intellectual property protection, namely, that it does not generally require or encourage disclosure of source code. We review a considerable body of survey evidence and theory that seeks to explain why developers participateinopensource collaborationsinsteadofkeepingtheir codeproprietary, andevaluatesthe extent to which open source may improve welfare compared to proprietary development.

We began this research with two overlapping objectives. The first, and of most general
academic interest, is to gain insight about the following puzzle: why has there been
essentially nil implementation of any of the institutional ideas in the economics literature
for improving the efficiency of public goods decisions? These ideas have been proposed
and refined in literally hundreds of academic articles over the past 30 years, many of
them have undergone extensive laboratory testing, and we have an extensive network of
public policy practitioners and academics that might be expected to help bring them into
practice. The second objective is more specific and of immediate policy relevance: to
understand if there are opportunities to increase the effectiveness of the federal Legal
Services Corporation (LSC) by improving its decisions about its own internal public
goods, largely the provision of information to attorneys that directly service the eligible
low-income population. Providing these public goods requires locating, customizing,
synthesizing, and creating documents and templates, doing research, leading training, and
answering questions. We hope that our joint consideration of these two objectives might
be beneficial to each: identifying practical implications of the public goods literature may
benefit the LSC, and a case-study of LSC may identify general challenges that public
goods mechanism literature should address.

Though standard accounting practice requires that all assets of a firm be manifest in accounts, an exception to this principle allows museums to omit their entire collections from their balance sheets. As the collection of any top-rank museum has more value than the rest of a typical museum's assets put together, this practice greatly obstructs good institutional decisionmaking. The practice is justified by an assertion common in the museum community that the collection is not a financial asset, neither available for sale nor obligation as a loan security, and by a variety of other assertions regarding the cost of assessing it. These assertions are shown to be choices with no fundamental support beyond museum management comfort, or to be erroneous. 

The allocation of artistic patrimony across museums generates a return on investment below 1%, indicating that these assets are not being used efficiently to create the benefits museums (whether non-profit or public) are supposed to provide: no other organization would be allowed to control assets with such a low rate of return. Capitalizing collections is practical and would lead to a variety of beneficial changes in museum practice, on the criterion of inducing "more, better, engagement with more art by more people."

This paper provides an extension of general equilibrium theory that incorporates the actions of individuals both as demanders and suppliers of goods
and as members of ¯rms, schools, social groups, and contractual relationships. The central notion of the paper is a group: a collection of individuals associated with one another for some purpose. The model takes as primitive an exogenous set of group types, interpretable as (potential) ¯rms, schools, social groups, contracts etc. The types of schools and ¯rms that materialize in equilibrium, as well as the way that agents acquire skills, are determined endogenously in a competitive market, as are the contracts they enter into, and the production and consumption of private commodities. The model is
well-founded (equilibrium exists) and passes a basic test of perfect competition (coincidence of the core with the set of equilibrium states). Examples
and Applications illustrate the °exibility and power of the framework.