Since the problem of climate change has drawn attention both in political and academic agenda, the question of how much people are willing to pay for the mitigation of global warming has been a subject of keen enquiry. Several recent studies have addressed the issue of the Willingness-to-Pay (WTP) for climate protection and results have given an almost similar plurality of amounts as there have been studies. This diversity can be explained, among other reasons, by the differences in the design of the corresponding studies. While most authors employed a stated preferences approach - where subjects indicate their WTP hypothetically and without any real payment - leading to a relatively high value of WTP, some studies utilised a revealed preference framework - where real payments have to be done. Comparison of the two different approaches show that the revealed preferences studies result in much lower values for WTP.

Existing revealed preference studies, however, face the following difficulty. The values reported from such studies are obtained in a decision framework wherein individuals face the opportunity to free ride on emissions reductions by others. Subjects may withhold their contribution to the public good climate protection while others contribute. While the WTP obtained in such a framework is certainly of interest, it is conceivable that the corresponding WTP is higher in a decision framework which at least partly excludes the free riding opportunity by introducing some form of collective decision.

In this paper, we study the demand for climate protection when the purely individual perspective of previous revealed preference studies is relaxed. We investigate whether the WTP for climate protection depends (i) on the information of real behaviour of subjects in a similar decision making situation and (ii) on the introduction of collective action.

Participants in our framed field experiment were remunerated with 40€ cash for their time in partaking in this investigation and were then given the opportunity to contribute to CO2 emissions reduction by buying European Union Allowances (EUAs). Each participant was confronted with two different prices for permits in 100kg CO2. Subjects were then asked to indicate the quantity of permits they would be willing to buy at each of the prices. Finally, one of the two prices was randomly selected by the administrator and the transaction was carried out at the corresponding price. Any allowances purchased were withdrawn from the European Emissions Trading Scheme, therefore it can be stated that all contributions led to real reduction in emissions. In one treatment of the study, we introduced a collective action whereby all members of a group were forced to purchase the median amount of EUAs demanded by the group. In another treatment, subjects received information about the real demand revealed by other subjects in a similar decision making situation. The information on previous contributions may signal the level of contribution by others.

The main finding from our study is that the probability of purchasing EUAs is higher in the treatment with collective action compared to the treatment where no collective decision was implemented. The provision of information about other subjects' behaviour, however, has no treatment effect on the demand for climate protection. The mean (median) WTP amounts to 11€/tCO2 (5€/tCO2) in total. Furthermore, we observe a strong correlation between subjects' demand and their expectations about other participants' behaviour. When collective action is not available, subjects' expectations are consistent with free rider behaviour.