1. Define Efficiency/Pareto optimality. One issue that economists have is that in a sense people's preferences can change overtime. Some of this may involve neurological\biological changes which economics hasn't dealt much with. But lots of the change involves learning new things over time. How would you define Pareto optimality in an environment where people can learn about the quality of products over time? Can you think of two definitions taken at different moments relative to the learning about the product? If you are a policy maker what do you do if these conflict? Can you give an example?
2. One issue that economists have is that in a sense people's preferences can change over time. But think about how preferences change as your human capital as future earning potential changes. How much the definition of Pareto optimality accomodate that? How do the policy issues compare here to the above? Here is a deep extension. Let's say in the future you anticipate your preferences may change in a way that your current self dislikes. How should Pareto optimality/efficiency and policy accomodate that? Can you give an example?
3. Hard question with some outside reading required: In modem philosophy, fairness is typically bookended by two different academic models: John Rawls, believed that fairness should be measured based on a "veil of ignorance" when you didn't know your position in society and from an extremely risk averse point of view that you were heavily concerned you might end up in a very low status position. Robert Nozidc, believed fairness was based on consent and that fairness was based on voluntary trades from a hypothetical world in which you possessed certain property and each individual policy change from that state could not be supported by coercion from whatever the previous status quo was, though he allowed that in the initial status quo you may have your property stolen by stronger neighbors. Are both definitions of fairness, a subset of Pareto optimal (efficient) allocations? How would you adjust the definition of Pareto optimality to account for both of these? In the Nozidc case, how do you account for the idea that policy ideas that are consensual may emerge from a coalition of individuals short of the entire population?