 In my research, I am interested in elections as a way for society or smaller groups of agents to reach a decision. Existing research has studied the effectiveness of formal elections in terms of solving collective choice problems like a conflict of interests or a conflict of opinion. What we are interested in is the effectiveness and nature of informal political processes that share certain features with elections. So examples are polls and petitions are protests. In these informal political processes, the more people participate, the more effective we would think a petition or a poll or a protest is in convincing a decision maker. There is no pre-specified rule that determines how the decision maker should react to a protest or a petition. And so we want to understand how effective they are depending on determinants like the motifs of the participants and their beliefs about the participation choices of the others. So we can think about the motifs of the participants in along three dimensions. So one is how convinced they are of the cause, whether they think it's really important to do something against climate change. The second dimension is sort of what are their personal costs and benefits. How costly it is to go to a protest or whether there are other benefits that are orthogonal to their convincing of the protest. Third, what will be important for the participants is what they think about the other participants. How many other participants are going and why the other participants are going because this will determine how the protest is perceived by the intended audience. To put these three dimensions into a framework we use game theory, which is how economists think about strategic situations. People have studied protests and such informal political processes using game theory before, but they have looked only at two of these dimensions, namely at whether or not people are convinced of the cause of the protest and what they think about what the others are doing. Existing research has found that political processes are ineffective because there are no costs of going to the protest. Everyone joins the protest and the protest at the end is ineffective. By adding the third dimension, which was the personal cost and benefit of going to the protest, we add this extra credibility dimension, which is I go to the protest despite having high personal costs. So when we think about their costs of going to the protest, thinking about how many other participants there are becomes particularly important. Why? Because if very few other people are there, the protest is very unlikely to be effective and if going to the protest is costly, I would like to save on my costs and will also not go. So a few other people are going, very few others will want to go and the protest is ineffective. On the other hand, if already many other people are going and it's costly for me to join the protest myself, then I may want to free ride in the terminology of economists in the sense of I will let the others protest and I stay home and save my personal costs. The first effect, the fact that if very few other people are going, I don't want to go, that's going to be captured by what economists are calling strategic compliments, which is the more others are doing something, the more I want to do it, the fewer others are doing it, the less I want to do it. On the other hand, the more other people are doing it, the less I want to do it, that's a strategic substitute. In our research we have two sets of findings. The first set of finding is related to the fact that we consider the personal costs of participation and what we find is that the presence of personal costs of participation often makes a political process, informal political processes effective when they haven't been before. The fact that it can be costly to go to the protest makes the protest more credible and that may explain why we actually often see protests in practice while existing research often suggested that informal political processes like protests are not effective. The second set of findings is related to the strategic nature of going to the protest so related to the fact that protests may be strategic substitutes or compliments. So the finding that protests are something like a strategic compliment where the more others are going, the more I want to go, but the fewer others are going, the less I want to go, the presence of this strategic compliment also makes protests maybe more fragile and the fragility has two implications. First, a small change in belief. So thinking that few others are joining will make people less likely to join and this can lead to an unraveling of the protests. Second, even small changes in fundamentals then can cause an unraveling of protests. So what we could imagine is a situation where a small change in fundamentals will lead suddenly to an eruption of protests or small changes in fundamentals may lead large protest movements to suddenly stop. First, the finding that costly participation may increase the effectiveness of protests has implications for the effect of say social media on protests. If social media can lower the costs of joining a protest and social media lowers the cost of coordination then what we would expect is that we see more protests or more people participating but at the same time we may expect that protests themselves become less effective. Making it easier to sign a petition or to participate in a poll may actually lower the informativeness of petitions and polls. The second set of implications is related to the strategic nature of protests. The fact that protests are often a strategic compliment means that protests are maybe inherently difficult to predict. So in future work we want to refine our methods and we want to consider the possibility that participants reasons differently. So rather than saying what happens if I join the protests rather than not we will consider a participant who says what happens if everyone like me joins the protest or what happens if people like me do not join the protest. So this is a little bit like a Kantian thinking. When I think about many people like me joining the protests or not joining the protest it becomes much easier to think through the consequences of my actions and also to think through what would be the right way to do in this situation. And we think using this as an alternative model will be closer to the actual reasoning of people who join protests.