The social planner problem is \begin{align} \max_{\lbrace x_i,y_i,x_j,y_j \rbrace } &\theta_i U_i(x_i,y_i) +\theta_j U_i(x_j,y_j) \\ & \text{s.t.}\\ &x_i+x_j = \omega_i^x + \omega_j^x\\ &y_i+y_j = \omega_i^y + \omega_j^y\\ \end{align} where $\theta_i,\theta_j$ are weights.
As I understand it, with a proper choice of weights, one can turn the Social Planner's problem into
- A Lagrangian optimization problem that solves for a Pareto optimal allocation.
- A Lagrangian optimization problem that solves for an allocation that is a Walrasian equilibrium.
Is this accurate? If so, what constraints are there on the weights for each case?