My team’s and our co-authors’ work on Part 1 led to several papers exploring the imprecision in perception and coding of what one sees, what determines the information seen, and the resulting impact on markets and their design. For instance, we have shown that in many markets simple cost comparisons rather than intricate informational considerations determine what market participants see. These costs determine the performance of the market and allow us to explain two empirical puzzles: the privacy paradox and the lack of disclosure. While designers and regulators cannot control information by assigning property rights they can do so by affecting the costs of information/perception choices, e.g. by setting up default information structures.
Our work on Part 2 led to papers defining simplicity of mechanisms, operationalizing comparisons of mechanisms in terms of their simplicity, and delineating classes of simple mechanisms. As an application, we showed that Random Priority is the sole simple, efficient, and fair mechanism, resolving a long-standing central conjecture in market design without transfers. We also show that ordinal simplicity is necessary if the designer wants the mechanism to be straightforward to play.
Our work on Part 3 developed a new methodology that radically simplifies designing pay-as-bid auctions. This new methodology allows one to design the government debt auctions so as to minimize the government expenditure on servicing its debt. It also allows designing the pollution permit auctions so as to maximize the societal welfare. Our other working papers evaluate the efficiency loss (or lack thereof) from restricting attention to simple mechanism. For instance, we show that in the presence of risk aversion or wealth effects simple mechanisms might achieve ex post efficient bilateral trade among informed parties, overturning the long-standing intuition that such efficiency is impossible.
