This paper makes use of optimal control relaxed problems to prove the absence of optimal trajectory in continuous time models with social increasing returns to scale where indeterminacy occurs. Although an efficient optimal policy does not exist, some chattering stabilization policies can mimic...