 Hi folks, so welcome back and now we'll be talking a bit about mechanism design in general. So the idea here is that in a lot of what we've looked at so far in the course we've been looking at specific institutions, specific games, and trying to understand how they affect behavior and in particular how the structure of the game can be used to analyze and predict how people are going to behave. And now what we want to do is flip things upside down. So this will be game theory on its head and the ideas instead of taking the game is given and trying to analyze it we'll look at situations where we have some idea of what we'd like to see be the outcome and then the question is can we design a game that will yield that outcome? So can we design institutions and you know one big puzzle that we have in terms of the way people behave is when you think about the inefficiency that goes on in terms of bargaining there's often bargaining breakdowns and people spend a lot of time trying to figure out how do we mediate bargaining? What's an appropriate technique to bring two sides together so that they can reach a bargain? And in terms of complete information game theory if people can really see what the payoffs are and you looked at a particular bargaining game generally you would have situations where people would tend to reach agreements. They can look down the tree they can figure out what's going to happen and unravel things and reach an agreement and so here we can ask a question of well let's suppose now that we're in a Bayesian setting so we have situation where we have workers thinking about trying to get higher wages and we have management and shareholders on the other side trying to keep wages low and they each want different things coming out of a new agreement, a labor agreement. Can we design a way in which they're going to come to a table and make offers and counter offers in such a way that they'd actually reach an agreement and avoid a strike and avoid wasting resources and time? We see strikes all the time so the answer evidently seems to be no we can't necessarily design such a procedure. So the question becomes which procedures can we design? What are the best that we can do? How do we begin to think about that? How do we begin to model designing an institution and then seeing what the outcomes are going to be and trying to design the best institution in terms of those outcomes? So we'll be looking at things like auction design. We'll talk about why we might have breakdown and inefficiency in bargaining settings when people have private information about what they'd like to see be the outcome. We're going to look at, we've already had a look at voting rules but we'll look at a general set of decision mechanisms that public enterprises can use to improve efficiency. So we're going to look at in some specific detail at how to design things when we want to achieve efficiency. We want to get the best possible outcomes and we realize that people have private information and they're going to try and take advantage of that and have incentives to try and get the best that they can get out, which doesn't necessarily always coincide with social efficiency. So when is it that we can get socially efficient outcomes? That's going to be the central question in mechanism design. So let's get started.