Askar Mukashov (Kiel Institute)
Identifying optimal policies can be very complicated under the standard CGE modeling framework. First, the set of available policy options can consist of several instruments, each with its own range of potential policy shocks. Second, the definition of best policies might vary depending on assigned parameters and reference scenarios. This paper introduces portfolio optimization methods to CGE models that allow analyzing the spectrum of policy options under uncertainty over other model parameters. As a case study country, we select Senegal and focus on agribusiness policy targeted at increasing rural welfare. As a domain of policy instruments, we consider productivity growth of 22 agribusiness sectors; as a domain of uncertain model parameters, we consider exogenous shocks. We sample 3000 policy scenarios, and for each potential policy, simulate 250 possible scenarios of exogenous shocks. The obtained sample allows us to analyze risk and return indicators of each potential policy, while the application of portfolio optimization theory allows us to determine the set of potentially optimal policy shocks. However, the definition of the (exact) optimal policy is not certain and depends on the risk/return preferences of a particular stakeholder or investor.
Virtuall via Gotomeeting