Saddlepath learning occurs when agents learn adaptively using a perceived law of motion that has the same form as the saddlepath relationship in rational expectations equilibrium. Under saddlepath learning, we obtain a completely general relationship between determinacy and e-stability, and...