In this dissertation I analyze the quality choices of a group of producers. In the first essay I use mechanism design to study the interaction of asymmetric information and the democratic process in the quality choices of a group of heterogeneous producers facing an opportunity to gain from establishing a reputation for their quality products. I find an asymmetry in the possible equilibria between the high and the low quality majorities. The quality level provided by the group with a low quality majority is lower than the first best, and the minority producers get rents. With high quality majority, if demand and group conditions are favourable, the quality level provided by the group is higher than the first best and the minority's type left with rents. Otherwise, the quality level provided by the group is first best and no rents are left to the low-quality producers in the minority.

The second essay proposes a methodology to measure the characteristics of intermediate products when quality is multidimensional. It uses a general representation of the multioutput technology via directional distance functions and constructs quality indicators based on differences. The quality indicators may be used to evaluate firms' output taking into account the whole set of quality attributes. I explore the relationships among the different quality attributes and the yields by a systematic investigation of the disposability properties of the technology. In addition, I show how aggregate quality may vary with the production level.

The third essay designs an optimal payment system for a group of producers implementing it empirically. In the essay I show how to implement the first best through higher prices for better quality commodities, deriving the optimal pricing schedule. I take into account producers' heterogeneity by modelling inefficiency and illustrating how technical efficiency interacts with producers' ability to produce output for a given level of inputs and hence affects revenues. The technology and the technical efficiency of producers are then estimated with a stochastic production function model. The estimation results are then used to simulate the pricing scheme.