Although many have studied various incarnations of Internet–based reputation systems, past analyses have mostly been piecemeal in method and have focused only on market applications. I propose a general “reputation–based governance” framework that has interesting properties and implications. The concept also provides useful insights on the issues of openness (as in “open source” software) and of participative forms of design and production.Contents
Introduction: Reputation in governance
What is reputation–based governance?
An application: The management of public works
New methods of production of statistical information and new venues for quantitative analyses
Fighting corruption
“Openness”, “participation” and “reputation”: A convergence?
Pending issues
A roadmap to reputation–based governance
Introduction: Reputation in governance
Most readers are familiar with examples of Internet–based reputation [1] systems. On eBay, buyers and sellers may rate one another at the end of a transaction. On many online communities, such as Slashdot, readers may assess each other’s comments. Around the world, many restaurants’ guides and hotel booking services allow past clients to rate their experience.
In all these cases, past feedbacks posted on the Internet allow to compute a “reputation index” that provides strong incentives to behave in a socially advantageous fashion. On eBay, persons most times refrain from cheating also because they fear to receive retribution in the form of negative “feedbacks”. On Slashdot, readers may decide to filter out writers who typically receive poor ratings. Persons who care for their writings to be read then have an incentive to craft them carefully. The clients of hotels and of restaurants clients increasingly use reputation information that they access through the Internet as a base for their decisions on where to stay or dine. Hotel keepers and restaurant managers typically take notice.
The importance of reputation is hardly a novelty. Reputational considerations have always played an important role in human affairs, and contribute to the predictability of behaviours, a prerequisite for most social intercourses. In many contexts, they provide an alternative to the contractualization of behaviors and even to the presence of a well–working judicial system [2].
In this respect, the Internet innovates in three important ways, by allowing what Dellarocas (2003) defines the word–of–mouth.”
First, it allows to spread voice–of–mouth to an unprecedented level. This, in turn, permits the existence of reputation–based interactions — be them of the market type, or other — at a global level and among persons many degrees of separation apart.
Secondly, the presence of a digital information infrastructure allows for a careful engineering of many details that contribute to the overall outcome of the system, such as: The condition under which the assessments are made, the metric according to which they aggregate to form a reputation index, the rules for participating and the possibility of changing one’s identity, etc.
Third, the Internet democratizes reputation systems, because it allows for their design so that all relevant parties may play the game under similar conditions. In conventional contexts information on reputation mostly spreads informally and via social networks: people who are better placed within them are at an advantage because they obtain better information. This, in turn, creates an incentive to spend time and resources to place oneself within such advantaged networks, a socially wasteful activity that economists would define as “rent seeking”.
What is reputation–based governance?
In principle, reputation–based governance applies both to private and to public activities; however, I’ll here discuss public governance and, more to the point, the management of policies. Reputation–based governance is a way of executing (and as we will see, possibly, of choosing) policies that hinges on the reputation of the actors involved, by a systematic use of an appropriate digital information system that computes reputation measures and makes them accessible to all.
Policies can be either projects, i.e., initiatives having a goal well specified in advance and lasting a definite amount of time (such as, the building of a new bridge), or programs, i.e., services that are delivered over a period of time (for example, an educational degree offered at a public university). Note however that paying attention to public policies does not imply that private actors are outside our visual range, because governments increasingly collaborate with non–governmental actors and with firms to pursue their goals. We’ll consider a public procurement example below, where a public administration explicitly asks private actors to provide goods or services.
Reputation affects incentives. The underlying behavioural assumption is rooted in what social scientists call “rational choice theory”. In a nutshell, I posit that social actors act to advance their self–interest. Firms maximize profits, and having a good reputation helps them in this respect. Within public administrations, I adopt the so called “economics of career concerns approach” (see Holmström, 1982; Dewatripont, et al., 1999), according to which bureaucrats are motivated by their desire to step up the career ladder, in a situation where a good past performance advances one’s chances of obtaining a promotion.
Note that such view on human behaviour does not rule out the presence of non–monetary incentives, including some of those that have been advanced in advocates’ accounts of open source software production — such as “egoboo”, or the personal satisfaction that derives from public recognition of one’s (voluntary) work (see Raymond, 2000). Personal satisfaction also is a positive function of reputation, at least for all who are not snobbish to the point of enjoying not being appreciated by the masses, so incentives of this type would actually reinforce the role of reputation.
Reputational effects have three main positive effects on governance.
First, at a given moment in time they help discriminating between providers of different quality — at least when a choice is possible. If the task is choosing a restaurant, knowing about their reputation allows to pick a good one.
Secondly, they allow selection forces to weed out the least fit. Bad restaurants eventually go out of business, and mediocre bureaucrats, when they do not get fired, at least do not obtain promotions so that they progressively become less relevant.
Thirdly, they provide incentives to invest in quality. The owner of a restaurant, or the bureaucrat, operating in a context where performance matters, has a strong reason to try to improve his skills.
These effect, in varying measures, are present wherever reputational considerations play at least some role in governance. Within reputation–based governance, the link between past performances and reputation is particularly strong, following the presence of two of the three motives mentioned above: The unprecedented publicity of reputational information afforded by the Internet, and the possibility of designing the reputation system and metric according to needs.
An application: The management of public works
To help fix ideas I discuss the application of reputation–based governance to the management of public works [3]. Public works are an important instance of public procurement; their reputation–based management, moreover, is the object of a software demonstrator (named “Rebag–Ware”) that the reader is encouraged to access [4], by picking one of the user identifications and passwords that are provided for the purpose of experimentation [5].
Public procurement is subject to a set of constraints that private enterprises do not face when they buy goods or services. In particular, private negotiations prevail within the private sector, while the public sector is typically obliged to use public bids for its procurement needs (Bajari, et al., 2003).
Public bids and auctions may have optimal characteristics when the goods or services required can be specified in all their details. This however only happens rarely. In most cases it would be beneficial to consider the past reputation of potential contractors, so as to constrain their actions by the thought that if they under–perform, they will find it difficult to secure future contracts.
Reputation–based governance of public works is based on a taxonomy of possible projects. For example, new public works can be “roads”, “bridges”, “buildings”, etc. Within each type of good, there are further sub–classes: Buildings may be schools, police headquarters, and so forth. All projects are documented in a database accessible by all.
Upon approval, the public administration inserts each project into the database. Preliminary information about a project include its general description according to the codified taxonomy (type of good, location, expected cost, etc.), technical drawings, and any documentation that may accompany the early stages of the project, such as an environmental impact and a cost–benefit analysis. As the project evolves, more information is provided, eventually to include details on costs and information on the final outcome, comprising, for instance, pictures of the completed works. Each project is associated with the administration and with the contracting firms that are responsible for its execution. In particular, the information system keeps track of, and makes visible, all the projects carried out by each administration and firm.
Once completed, a project is assessed by accessing the information system. The public administrators involved express their opinion on the quality of the work carried out by the contracting firms. Firms in turn assess the performance of the public administrations with which they interacted. Most importantly, the public judges public works that affect them. For example, a school is assessed by the local community where it is built, by the personnel working there and, possibly, by its students. A local road is judged by the people who live in its proximity, while a freeway is of interest to residents of a vaster area. Such assessments are highly structured and refer to a small set of well–defined characteristics. For example, the public could judge “aesthetic qualities”, “usefulness” and “accessibility” of a public building.
These digitalized information allows for the provision of a series of automatically generated summary statistics. When all the relevant information on public works are contained in a digital repository, computing summary statistics of interest does not require any ad hoc data gathering and processing, but only the production of a “view” of the available digitalized data. Such method of producing statistics in itself represents a novelty with respect to current practices, a topic on which I will return.
The statistics produced inform on the general characteristics of the projects, the unit costs of works of comparable types (for example, of a mile of road of a given category) and their completion time. They moreover express the overall assessment by the public. Also, the availability of information on many projects allows for the computation of rankings of individual projects on several dimensions. Summary statistics could be available also for various subsets of the data, providing views by geographic unit, by type of administration, and by contracting firm.
The assessments of the quality of the completed public works then reverberates to administrations, administrators and firms: The reputation of all these actors is a function of their past performances, as perceived by others. Political appointees and parties also could come into play. The assessments of projects could also reverberate to the political appointees who are at the head of public administrations executing them, and to their political parties, providing electors with useful policy–derived “hard data”.
Reputation–based governance of public works would establish a number of incentives and of disincentives. The visibility of various rankings of completed public works, of the administrations who carried them out, of the administrators involved, and of the contracting firms, by itself puts a premium on honest and efficient behaviour, inasmuch as such information influences electoral choices.
Also, public administrators would be pressured not to be looked down by their peers. The visibility of their reputation rankings would encourage meritocratic career decision by their superiors. As for firms, to the extent that reputation matters in determining the outcome of public procurement, they would want to receive good ratings. Moreover, if a firm is active in both the public and in the private sector, a good reputation in the former would be expendable in the latter.
Most importantly, reputation considerations could be employed within source selection procedures, that is, contractors could be chosen not only according to the price and technical evaluation of their bids, but also depending on their reputation [6].
New methods of production of statistical information and new venues for quantitative analyses
Reputation–based governance would bring about important changes in the availability of governance–related statistics. In a context where all business is carried out online, governance–related information can be automatically retrieved and analysed. Consider for example the opinion of Dalle, et al. (2004) regarding the study of the open source mode of production:
[There is a] “growing abundance and accessibility of quantitative material concerning the internal workings of ‘open epistemic communities’. The kinds of data that are available for extraction and analysis by automated techniques from repositories of open source code and newsgroup archives, and from the email subscriber lists generated by F/LOSS project members themselves, also offer a rapidly widening window for research on the generic features of processes of collective discovery and invention.”So, while the issue is almost obvious and has been raised before, it is useful to consider it from a more general point of view. Official statistics are produced by dedicated institutions, the most important being the various “National Statistical Offices” (NSOs), such as the Census Bureau in the United States. In democracies, these institutions developed during the twentieth century with in view the need of guaranteeing the necessary accuracy and correctness of official statistics against possible political manipulations.
The data that NSOs publish are typically the result of surveys of various types. Their production process includes a design phase, then the collection of data, their processing and, finally, their publication. Several months or even years may elapse from the actual collection of the data to the publication of the resulting statistical information.
On the other hand, if the prevailing governance model were reputation–based, an important part of the information on which official statistics are based also would be available digitally and in real time. Such a situation would allow for an important shift away from the present situation where official statistics are “gathered” and then “processed”, to one where statistics would simply be a view of the data contained within a suitable information system.
From being the result of ad hoc activities, resulting in a high set–up and unit cost (or “marginal”) cost of production, official statistics would become integrated into the routine working of the overall information system supporting governance, displaying negligible marginal costs of production once the necessary procedures have been put in place.
The consequences of this transformation are not limited to the fact that new statistics would be available that measure phenomena today escaping quantification, and that the existing statistics would be more accurate and timely. Quite likely, the new governance landscape would have important implications also on the overall institutional and social environment that today leads to the production of official data on human activities, and on the societal role played by statistical information.
For example, the availability of a wealth of digitalized policy–related information would allow for new uses of data, and in particular to new and more powerful methods for the quantitative evaluation of policies and of their effects. Such evaluations have traditionally proven to be difficult due to conceptual issues and to a paucity of data. The availability of the individual assessments on policy outcomes would constitute an almost obvious route towards a systematic assessment of governance. Also, the presence of abundant objective digitalized information would allow for the implementation of model-driven evaluations techniques [7].
Fighting corruption
Corruption is a serious problem in many countries around the world. It can take many forms, from the “petty” variety — such as when a police officer asks for some money not to impose a traffic fine — to the “grand” activities of highly placed bureaucrats and politicians that divert public funds to their pockets or to the coffers of their political campaigns. There is a broad consensus that corruption has very detrimental consequences on economic growth and social cohesion [8].
Measuring corruption is difficult, because it is an illegal activity that is kept as covert as possible. The best known measures of corruption are “perception–based”, deriving from surveys carried out among people in the know. An example of this approach is provided by Transparency International’s “Corruption Perception Index” [9].
There have also been attempts to develop “objective” measures of corruption. One example is provided in Golden and Picci (2005), who consider public infrastructure in the Italian regions. They compare two sets of measures: One is based on how much was spent over time to build infrastructure (such as roads, public buildings, etc.), and the other is a physical inventory of what was actually built. They find that, on average, in the south of Italy much was spent, and relatively little was built. This contrast permits to compute an “index of corruption” [10].
Many cures have been proposed for corruption, but the fact that the problem is still so serious indicates that the issue is all but settled. At least two of the factors connected with corruption would be directly affected by the adoption of reputation–based governance. First, transparency of government acts puts a check on corrupt behaviours. If the citizens, civic society organizations and the free press (where it exists) have easy access to information that pertains to the resources that are publicly spent, then bureaucrats find it harder to be dishonest. As we discussed in the previous section, the information system on which reputation–based governance rests would allow for the computation of a wealth of governance–related measures, and among them of objective indexes of corruption similar to the one proposed in Golden and Picci (2005). These measures could be computed routinely at a negligible cost, and not, as it happens today, only after a painstaking effort of data collection. The government itself could use such publicly available information to introduce forms of “probabilistic auditing”: activities whose cost is above a standardized average, or whose assessments are poor, would have a higher probability of being thoroughly controlled.
Secondly, strong reputational effects would provide an incentive to bureaucrats not to squander resources, and to weight the personal benefits of corruption not only against the risk of punishment by the law, but also against the foregone career opportunities that a poor performance, even when undetected as an act of corruption, would imply.
“Openness”, “participation” and “reputation”: A convergence?
To the extent that reputation–based governance gives voice to the people who are affected by the policies chosen by elected officials, and that it provides a framework to increase the accountability of the actors of governance, it can be seen as a set of procedures that, broadly speaking, make governance more democratic. Here, the effect on the “quantity and quality” of democracy is through the presence of ex post incentives: administrators and politicians would anticipate the increased level of accountability of their actions, and ex ante would presumably listen more to the needs of the people.
Reputation–based governance also lends itself to an increase of participation by allowing forms of mixed direct–representative democracy, such as the ones that have been adopted, albeit without a central role of the Internet, in the Brazilian city of Porto Alegre and in many other places around the worlds, to create a “participative budget” (Baiocchi, 2005). In fact, the availability of an information system as the one here envisioned would lend itself to innovative participative practices, where appropriate procedures allow the people also to express their opinion on prospective policies and to contribute to forms of collaborative design of policy options.
This extension, in turn, would allow for the introduction of reputational information also on the citizens. Such reputation could be formed retroactively: Citizens who contributed to the design of policies that are later successfully evaluated, would have their reputation improved. Or, it could more simply derive from the assessment that anyone’s contribution receives from her fellow citizens.
We witness here an interesting possible convergence between three issues that, in different circles, are today widely debated: “openness” (as in, open source software), “participation” (as in, participatory design) and “reputation”. The debate on open source software has already benefited from an enlargement of perspective, thanks to the consideration of historical antecedents, such as the rise of open science in the sixteenth and seventeenth centuries (David, 2004; Willinsky, 2005), and to the study of forms of mixed design/production of goods that overcome the traditional distinction of inventor/designer, producer, and consumer [11].
In the debate on “openness”, reputational considerations play an important role. In the production of open source software, accounts written by insightful advocates (as an example, see Raymond, 2000) and academic enquires (Dalle, et al., 2004) agree on the pivotal role of reputation in the governance of open source projects. Having a good reputation, besides advancing one’s position within the open source community, also signals good programmer’s qualities to potential employers (Lerner and Tirole, 2002).
From a broader point of view, the whole of “Internet governance” — indicating with this loose term the way that the Internet has been ruled — is to a great extent “open” and “reputation–based”: based on open standards, much open source software, and a series of institutions (such as the Internet Engineering Task Force or the World Wide Web Consortium) that are rather horizontal in the way the build consensus on technical standards, and where individual reputations play a prominent role in determining who has a bigger say on any problem.
It is important to realize that this emerging governance model is not determined by the Internet. Technologies may enable organizational forms, but they never determine them. Depending on the context, a better communication technology may support both more intense horizontal or more vertical relations within an organizational landscape.
The Roman Empire of the first century had many characteristics that resembled what today we would call a “network organization”. This situation derived in part from the difficulties of communication technologies of the time that, among other things, impeded navigation across the Mediterranean sea for most of the year. A century or so later, the same communication technology was supporting a much more centralized structure [12]. For a contemporary example, the emergence of the Internet today is enabling not only horizontal hackerdom, but also quite vertical Wal–Mart.
Pending issues
So, all is well with reputation–based governance? Not quite, and as James Wilson somberly pointed out some time ago, “public management is not an arena in which to find Big Answers; it is a world of settled institutions designed to allow imperfect people to use flawed procedures to cope with insoluble problems.” [13]
Of all the open issues of reputation–based governance I here provide a nonexhaustive list, with some indications of possible solutions.
Some of the problems are shared with the Internet–based reputation systems of today such as eBay. For example, ratings of policies is a public good, so that we would expect it to be underprovided. However, we do witness a high degree of participation in many such systems, suggesting that sometimes people actually enjoy providing services that benefit others.
Other problems that affect existing systems would not be relevant within reputation–based governance. For example, the real identities of all actors would be known, so that opportunistic changes of identity to provide a fresh start to cheaters, or multiple identities, to provide fake consensus (see David and Pinch, 2006, for an example), would simply not be an issue.
Some issues are new with the proposed model of governance. One follows from the possibility that corrupt bureaucrats team up with opportunistic citizens to secure an unfair share of public resources. A bad and wasteful school in one’s neighbourhood may still be preferable to a good school located a long drive away. While this is a potentially serious issue, note that the “integrated statistics” that reputation–based governance would make available would include timely information both on the geographical distribution of resources, and on the standardized, or average, cost of a given output. This would make it more difficult for a coalition of bureaucrats, politicians and citizens to cover–up corrupt behaviours in exchange for an unfair share of public resources to a given constituency — or “pork barrel”, as it is sometimes called. Note also that these “distributive” practices are permitted by the relative secrecy of the legislative process. In the United States, for example, expenditure laws are often wrapped in “omnibus legislation”, discussed within committees and attracting support from both sides of the isle. Within reputation–based governance, at least in principle, it would be very easy to monitor who is getting what [14].
A further problematic issue involves the management of the information system and of privacy rights. Within today’s Internet–based reputation systems the problem is not very severe: one’s opinion, say, about a restaurant or a hotel that one has visited is hardly contentious. Within reputation–based governance, on the other hand, personal opinions are on sensitive political matters, and every person accesses the system with a real identity, as a citizen. Not guaranteeing the necessary privacy would allow abuses, making it possible, for example, for corrupt and dishonest politicians or bureaucrats to “buy” positive ratings from cynical, or simply poor, citizens. The management of the information system would need an appropriate institutional set–up. Participative practices always need adequate supporting institutions, and reputation–based governance is no exception.
A roadmap to reputation–based governance
Important organizational and institutional changes are always difficult to obtain, particularly when they depend on the presence of a complex technological infrastructure. Seen as something entirely new to be established, reputation–based governance could appear to be an overly ambitious goal.
However, there is a more useful view point on the issue. There is an ongoing process where information technologies are becoming ubiquitous in the management of governance processes. To an increasing degree, policies are managed with the help of information systems where they are codified and accompanied by a vast amount of information. It is a fairly safe bet to affirm that the process of “digitalization of policies” sill spread and deepen, in a sense, regardless of what we think and do.
The issue, then, is on what particular shape these changes will take. Will they empower citizens, or will they not? Will they support an horizontal, network–shaped, organizational landscape or, to the contrary, will they tighten the bolts of traditional vertical organizations? Will they model the reputation of the actors of governance, or will they only describe the aspects of governance that are useful for societal control, with reputational information still exchanged within traditional social networks?
Reputation–based governance, then, can be seen as a proposal for the direction that many concurrent changes that are already happening should take. Implementing reputation–based governance is about steering an ongoing process, providing a coherent shape to it, and making sure that, among the many organizational forms that the Internet may support, the ones that will emerge will allow be characterized by an unprecedented degree of transparency and by an allocation of resources and of power that is strictly linked to past performances and to the reputation that derives from them. Then, reputation–based governance would also provide an ideal platform on which to implant innovative procedures, reputation–based themselves, that permit the participation of citizens in the design of policies.
About the author
Lucio Picci holds a Ph.D in Economics from the University of California, San Diego. Currently on leave from the University of Bologna, he is a senior researcher of the Institute for Prospective Technological Studies (IPTS) of the Joint Research Centre (JRC) of the European Commission. Based in Seville, Spain, the Institute is one of seven scientific institutes of the JRC.
His personal page, from which most of his publications may be accessed, is http://www.spbo.unibo.it/picci. He is currently writing a book entitled “Reputation–based Governance”. The site of the research project summarized in this paper is http://www.rebag.it.
The views expressed in this paper are the sole responsibility of the author and do not necessarily reflect the views of the European Commission. As research into the technological aspects of Information Society Technologies (IST) is well developed in Europe, the ICT unit of IPTS addresses the policy aspects of the European Information Society in general and the socio–economic impacts of IST in Europe in particular.
Notes
1. In what follows, only an everyday language understanding of the concept of reputation is required. Reputation (and trust) can be analyzed using two alternative sets of game theoretic tools. Reputation (or, better, trust) emerges according to the Folk Theorem: in an infinitely repeated “prisoner’s dilemma” game, players may prefer the long–run benefit of not cheating, to the short–run advantage of cheating. Also, reputation can be considered within a bayesian game context, where the quality of a player is not immediately evident to others. See Cabral (2005).
2. See Greif (1993) and Greif, et al. (1994) on reputation and trade in the Middle Ages.
3. For a fuller account see Picci (2007) on which this section draws.
4. Available at http://andromeda.ei.unibo.it/rebagware/, accessed 5 September 2007.
5. See also Confalonieri, et al. (2007).
6. During the 1990’s a reform of Federal procurement in the United States effectively introduced reputational considerations into source selection procedures. See Picci (2007) and, for the account of one of the protagonists of the reform, Kelman (2002).
7. See, as an example, Picci (2006).
8. An up–to–date account of research on the determinants and consequences of corruption is provided in Rose–Ackerman (2006).
9. http://www.transparency.org/policy_research/surveys_indices/cpi, accessed 5 September 2007.
10. Obviously, the difference between the two measures could be explained by factors other than corruption, such as differences in efficiency or in building costs. This issue is considered in Golden and Picci (2005). What is relevant here is the derivation of a measure of governance from hard data that originate from the governance process itself.
11. As in the development of windsurfing, skateboarding and snowboarding; see Shah (2005).
12. Luttwack, 1974; Picci, 1999, p. 172.
13. Wilson, 1989, p. 375.
14. The U.S. “Federal Funding Accountability and Transparency Act of 2006” (Public Law of 26 September 2006 — S.2590) broadly addresses the same problem, by directing [...] “the Office of Management and Budget (OMB), by January 1, 2008, to ensure the existence and operation of a single searchable website [sic] accessible by the public at no cost that includes for each federal award of federal financial assistance and expenditures [...]” (Source: http://thomas.loc.gov/, accessed 5 September 2007).
References
P.L. Bajari, R.S. McMillan and S. Tadelis, 2003. “Auctions versus negotiations in procurement: An empirical analysis,” NBER Working Paper, number 9757, and at http://www-econ.stanford.edu/faculty/workp/swp02007.pdf, accessed 5 September 2007.
G. Baiocchi, 2005. Militants and citizens: The politics of participatory democracy in Porto Alegre. Stanford, Calif.: Stanford University Press.
L. Cabral, 2005. “The economics of reputation and trust,” mimeo, and at http://pages.stern.nyu.edu/~lcabral/reputation/Reputation_June05.pdf, accessed 5 September 2007.
R. Confalonieri, C. Leoni and L. Picci, 2007. “Rebag–Ware: Reputation–based governance of public works,” paper presented at IST–Africa Conference 2007 (Maputo, Mozambique, 8–10 May 2007), and at http://didattica.spbo.unibo.it/picci/Rebag-Ware-ISTAfrica2007. pdf, accessed 5 September 2007.
J. Dalle, P. David, R.A.. Ghosh and W.E. Steinmuller, 2004. “Advancing economic research on the free and open source software mode of production,” SIEPR Discussion Paper, number 04–03, Stanford Institute for Economic Policy Research, at http://siepr.stanford.edu/papers/pdf/04-03.html, accessed 5 September 2007.
P. David, 2004. “From keeping ‘nature’s secrets’ to the institutionalization of ‘open science,’” In: R.A. Ghosh (editor). CODE: Collaborative ownership and the digital economy. Cambridge, Mass.: MIT Press.
S. David and T. Pinch, 2006. “Six degrees of reputation: The use and abuse of online review and recommendation systems,” First Monday, volume 11, number 3 (March), at http://www.firstmonday.org/issues/issue11_3/david/, accessed 5 September 2007.
C. Dellarocas, 2003. “The digitalization of word of mouth: Promise and challenge of online feedback mechanisms,” Management Science, volume 49, number 10, pp. 1407–1424. http://dx.doi.org/10.1287/mnsc.49.10.1407.17308
M. Dewatripont, I. Jewitt and J. Tirole, 1999. “The economics of career concerns, Part II: Application to missions and accountability of government agencies,” Review of Economic Studies, volume 66, number 1, pp. 199–217. http://dx.doi.org/10.1111/1467-937X.00085
M. Golden and L. Picci, 2005. “Proposal for a new measure of corruption illustrated using Italian data,” Economics and Politics, volume 17, number 1, pp. 37–75, and at http://didattica.spbo.unibo.it/picci/article.pdf, accessed 5 September 2007.
A. Greif, 1993. “Contract enforceability and economic institutions in early trade: The Maghribi Traders’ Coalition,” American Economic Review, volume 83, number 3, pp. 525–548.
A. Greif, P. Milgrom and B.R. Weingast, 1994. “Coordination, commitment, and enforcement: The case of the Merchant Guild,” Journal of Political Economy, volume 102, number 4, pp. 745–776. http://dx.doi.org/10.1086/261953
B. Holmström, 1982. “Managerial incentive problems: A dynamic perspective,” In: Vetenskap och företagsledning: Studier i ekonomi och ledarskap tillägnade Lars Wahlbeck: Essays in economics and management in honor of Lars Wahlbeck. Helsingfors: Svenska handelshögskolan; see also B. Holmström, 1999. “Managerial incentive problems: A dynamic perspective,” volume 66, number 1, pp. 169–182.
S. Kelman, 2002. “Remaking federal procurement,” Public Contracts Law Journal, volume 31, pp. 581–622; see also http://www.ksg.harvard.edu/visions/publication/kelman.pdf, accessed 5 September 2007.
J. Lerner and J. Tirole, 2002. “Some simple economics of open source,” Journal of Industrial Economics, volume 50, number 2, pp. 197–234. http://dx.doi.org/10.1111/1467-6451.00174
E. Luttwak, 1979. The grand strategy of the Roman Empire: From the first century A.D. to the third. Baltimore: Johns Hopkins University Press.
L. Picci, 2007. “Reputation–based governance of public works,” In: G. Piga and K.V. Thai (editors). Economics of public procurement. London: Palgrave; a draft of the chapter is available at http://didattica.spbo.unibo.it/picci/rebag-publicworks.pdf, accessed 5 September 2007.
L. Picci, 2006. “The quantitative evaluation of the economic impact of e–government: A structural modelling approach,” Information Economics & Policy, volume 18, pp. 107–123; also at http://didattica.spbo.unibo.it/picci/egoveval.pdf, accessed 5 September 2007.
L. Picci, 1999. La sfera telematica: [come le reti trasformano la società]. Bologna: Baskerville.
E.S. Raymond, 2000. “The cathedral and the bazaar,” at http://www.catb.org/~esr/writings/cathedral-bazaar/cathedral-bazaar, accessed 5 September 2007.
S. Rose–Ackerman (editor), 2006. International handbook of economic corruption. Cheltenham: Edward Elgar.
S.K. Shah, 2005. “Open beyond software,” In: C. DiBona, D. Cooper and M. Stone (editors). Open sources 2.0: The continuing evolution. Sebastopol, Calif.: O’Reilly Media.
J. Willinsky, 2005. “The unacknowledged convergence of open source, open access, and open science,” First Monday, volume 10, number 8 (August), at http://firstmonday.org/issues/issue10_8/willinsky/, accessed 5 September 2007.
J.Q. Wilson, 1989. Bureaucracy: What government agencies do and why they do it. New York: Basic Books.
Editorial history
Paper received 9 June 2007; accepted 12 August 2007.
Reputation-based Governance is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 2.5 Italy LicenseReputation–based governance by Lucio Picci
First Monday, volume 12, number 9 (September 2007),
URL: http://firstmonday.org/issues/issue12_9/picci/index.html
