Random wanderings of my intellectual rather difficult in recent months, I came across an article by Messer et al 2007 on the importance of context effects in the behavior of contribution to public good. As you know some players, the economists refer to as public good property that has two characteristics, the first being the indivisibility of use (I can use the property without such use does not interfere or limit the use by another individual) and the second being the impossibility of excluding potential users (I can not stop my neighbor to enjoy the lighting even if there contributes nothing because tax fraud for example) .. .
[This is a picture, my neighbors are very nice people above all suspicion. I hope that no zealous collector does not read these lines]
The problem with public goods is the possibility that individuals behave as free riders, that is to say they do not contribute personally the public good, the contribution is generally expensive and, due to its characteristics of non-excludability, soiten tempted to wait another without an effort to produce it themselves. This possibility of free riding was raised by Knut Wicksell, an Austrian economist, from the late 19th century.
Experimental studies on this question were extremely numerous and all tend to show that the behavior of a stowaway is not as common as that, and only a small proportion of individuals are actually stowaways. The problem is that this relatively small minority is "toxic" because individuals who are not intrinsically stowaways will tend, if the contribution to the public good is repeated over time, to punish free riders by reducing their even their effort or contribution to public good (see Falk, Fehr and Fischbacher 2005, Econometrica For more details on this result.)
Well, this summary very fast (this has been discussed repeatedly in this blog) is just to stage the topic I want to address, namely the magnitude of the effect of context or presented in the phenomena of contribution to public good.
The effect of context, well known in experimental economics, has been highlighted by Kahneman and Tversky long ago (for a teacher, I'll drive back here) a few words, he said that the choice economic are sensitive to how choices are presented. (In economists' jargon, we say that there is an axiom (rarely asked explicitly for that matter) invariance of the description). A considerable number of studies showed this effect in the case of individual choices, but few have done in the case of choices involving what economists call a social dilemma, ie a situation in which the cooperation between individuals would be desirable (all contribute to public good) but for the community in which rationality will push individuals towards a poorer outcome of equilibrium in terms of general interest.
Messer's study deals specifically on this issue. The example given is that the donation of organs. Organ donation is typically a public good. Indeed, two main types of regulations exist at the international level. In the first type, you're supposed to donors by default (if you do not say anything, you can use your body for transplants or other) and the second type, you're supposed non-donor by default (it should say explicitly that you wish to donate your organs so they can be used for transplants).
A 2003 study by Johnston and Goldstein published in the journal Science ("do defaults save lives? ) shows that countries in which the status quo for the donation of organs is the consent (by default you are the donor and you must tell your loved ones explicitly if you refuse to give your organs) have a donation rate of 85.9% to 99.9% while for those countries where the status quo is a non-consent (France until recently), the rate of donation ranged between 4.3% and 27.5%.
A 2003 study by Johnston and Goldstein published in the journal Science ("do defaults save lives? ) shows that countries in which the status quo for the donation of organs is the consent (by default you are the donor and you must tell your loved ones explicitly if you refuse to give your organs) have a donation rate of 85.9% to 99.9% while for those countries where the status quo is a non-consent (France until recently), the rate of donation ranged between 4.3% and 27.5%.
Other examples are possible as regards for example the quality of our environment, the ecological sense, is it better to implement incentives to create the public good, through positive actions of remediation, or incentives to prevent the public from bad behavior polluters?
Under the policies against the greenhouse effect, the question arises. For example, in a report of recent Center for Strategic Analysis (available here ), about the political fight against the greenhouse effect, it says that the contributions of the Copenhagen agreement, describes generally as a failure, was the establishment of mechanisms (REDD +) designed to both preserve the quality of the environment but also to implement actions to in improve quality. In particular, it states:
"The Agreement explicitly included the establishment of a mechanism called" REDD + "to fight against deforestation and degradation but also to promote the planting, management forestry and conservation of carbon stocks "
Innuendo, it is well to worry about preserving the environment, but without doubt we must make more voluntary actions are implemented.
Economic theory is a priori neutral point of view: according to microeconomic theory, context does not matter (see what has been said previously) and the effectiveness of incentives in one case (creation of public goods) must be the same as in the other case (limit the development of public bad).
is precisely the subject of the article written in collaboration with Douadia Bougherara , INRA, and David Masclet soon published in an international journal, in which we seek to highlight a possible effect of context in the case of public goods. The idea is very simple: look for a creative context public good in a context of preservation of the public good, in the context of a fairly extensive experience.
For the context of preservation, assume that all resources of individuals are already placed in a common pot (the public good) and that the maintenance of these resources in the pot has an opportunity cost of private for each agent. Removing a unit of the pool, he gave up what he was withdrawing from the unit's good, but can use this resource for private consumption. Of course, by then, since this is a public good, created a negative externality because it also destroys the benefit derived by the other members of his group over to this unit was placed in the public good.
For the context of creation is simpler. Each individual has a certain amount of resources, and each of them should simply decide the amount it invests in the common pot, the amount invested with private opportunity cost (the private consumption that the agent can be made), but generates a utility after the consumption of public good and a positive externality, since the other group members also benefit from resources put into the pot. This is the famous game of VCM (Voluntary Contribution Mechanism) first used by Isaac and Walker in 1988.
To be honest, there was already an article by James Andreoni wrote on the subject in 1995 implementing almost exactly the experimental protocol that I have described, comparing the creation context and the context of preservation. It came to a surprising result, however: the level of cooperation was higher on average in the context of creating the public good in the context of preservation. The amount invested in the context in which each subject had to invest in the common pot nonexistent at the start was more important than the amount removed from the existing pot initially. As noted by Andreoni, and otherwise, subjects preferred to do good than to avoid evil (he wrote exactly "i t Must Be That people enjoy doing a good deed More Than THEY enjoy not doing a bad deed ").
In the article we wrote, we tested the robustness of this result. It must be said that it seemed quite incredible and we had some doubts about its strength: how individuals could they be more willing to destroy the public good than to create it? It seemed more natural to think that once the public good creates a form of status quo bias or bias staffing would be implemented, and that participants would find it easier to maintain than the public good create. Indeed, this is precisely what is suggested by the status quo bias highlighted by Thaler in 1980. If you give any good individuals, they are ready to make more effort to keep only if they are asked to make efforts to get the same good.
To complicate the problem a bit, and give a character greater generality in our study, we considered a public good game a bit special, with so-called threshold. In the context of creating a public good, this means that if the level of contributions is below a certain value, the public is not well established. In our experiment, we set different threshold levels, and if the threshold was reached, the public was well established in the amount of the sum of contributions by affected individuals. If, against the threshold was not reached, the contributions were lost to the subjects. This means that if, for example, the threshold to create the public good is 60 units, and that of the four participants in the group, one has invested all its resource endowment (20 units) and the other three have all invested less than forty units, then the first gets a payoff equal zero.
In the context of preservation, this means that at the beginning of the game, all the endowments of the players is placed in the pot (or 80 units, since each of the four players have 20 units right withdrawal), and they can exercise a right to withdraw a maximum of twenty units each. Each unit is removed by a player gives a point. Each unit remaining in the pot, provided that the threshold of preservation is not exceeded, giving 0.4 points to each participant, or 1.6 point for the entire group of 4 persons. For example, if the threshold for which the public is well preserved is 60 units, this means that if, for example, two participants in the group of 4 decided to remove all 20 units and the other two do not derive anything, the public good is completely destroyed. People who earn 20 points and removed those who have nothing removed earn zero points.
interest to introduce thresholds is twofold. First, it is more the reality of most public goods that surround us, insofar as the amount of public good created by the efforts of individuals within a community is probably not such linéaire.Si community does not a minimum amount of effort, the public good is not realized. Moreover, it changes the nature of equilibria theoretical contribution to the public good. In the game of public good without contributing to the threshold, the dominant strategy equilibrium is a balance of free riding, in which participants should contribute zero to the property (or withdraw all of their rights). In games Contribution threshold with the public good, we are in the presence of coordination games, in which multiple equilibria are: the balance of free riding still exists, but there are many Nash equilibria in which the sum of the contributions of players is exactly equal to the threshold of creating defined (or, in the context of preservation, the amount of withdrawals from the Nash equilibrium is equal to the threshold of preservation). What balance will ultimately selected by the players?
balance of free riding or to balance the public good to exist? The interest in experimental economics is precisely to establish results which enable an idea of the situation that is ultimately selected specifically by the group of individuals.
In the experiment we conducted, and has resorted to just under 400 participants, we implemented the two contexts (creation vs. preservation of the public good) and for four levels threshold public good (zero, 28, 60 and 80). The results are numerous, but for the most part, and to our surprise it must be said, it appears that the result of Andreoni is robust: cooperation levels are higher in the case of a creation of a public good that in the case of preserving it. The chart below briefly summarizes the main result. The higher the threshold level for the public good, the higher the level of cooperation within the group (the height of the bars which form the total contributions in each group on average) is high, and this is true in both contexts (creation and preservation in blue in red), although the maximum threshold level (80), levels of cooperation are lower than the threshold level immediately below. However, quite clearly, the gap relative levels of cooperation between the two contexts is even stronger than the threshold embodying the creation or maintenance of the public good is high.
source: Bougherara, Denant-Boemont & Masclet, 2010.
Beyond this result which underscores the strength of the experimental results concerning the behavior of contribution to public good, an interesting result and I leave to your sagacity, reader, is as follows. It seems that, contrary to what is observed for private goods, there is no status quo bias to public goods. That already existed before the public good does not guarantee that individuals will have a greater tendency to maintain that they have the tendency to create other public goods.
Amazing, no?
0 comments:
Post a Comment