Essays on indeterminacy in dynamic general equilibrium models

This thesis consists of three essays on the issues of monetary policy, indeterminacy and expectational stability (E-stability).;The first essay discusses about indeterminacy under interest rate policy. We show that, with endogenous investment, virtually all monetary policy rules that set a nominal interest rate in response solely to expected future inflation induce real indeterminacy in models with (i) staggered prices, (ii) staggered prices and staggered wages, and (iii) staggered prices, staggered wages, and firm-specific capital. In (i), policy's response to current output can help significantly in ensuring determinacy with an infinite labor supply elasticity, but little with empirically plausible labor supply elasticity. In (ii), responding to output always helps a great deal, though under low price stickiness and without capital adjustment cost it may call for a moderate response to output in order to ensure determinacy for a wide range of response to inflation. In (iii), even a tiny response to output can always render equilibrium determinate for a wide range of response to inflation. We also find that the policy's response to lagged interest rate further enhances macroeconomic stability, though in many cases some response to output is still essential for ensuring a nonempty determinacy region, and even in the other cases responding to output often remains important in order to ensure determinacy for a wide range of response to inflation. Our results are quantitatively invariant to the presence of habit persistence in consumption.;In the second essay, we analyze the possibility of generating indeterminacy in the monetary models with nominal rigidities under increasing returns to scale technology. We find that in the monetary models with flexible wage (prices can be either flexible or staggered), the minimum degree of returns to scale necessary for local indeterminacy is inversely related to the magnitude of the labor-supply elasticity. This result is consistent with the one obtained in a real model in Benhabib and Farmer (1994), in particular, indeterminacy can only occur under very large labor-supply elasticity. Furthermore, we show that when incorporating nominal wage rigidities, indeterminacy can occur under an empirically plausible labor supply elasticity. However, the role of nominal rigidities for indeterminacy is very sensitive to the degree of capital adjustment costs, but it is robust to the assumptions of timing and the degrees of nominal rigidities.;The third essay studies the stability puzzle (Evans and McGough, 2002): why does indeterminacy almost always imply expectational unstability in RBC models? Following Meng and Yip (2008), we relax the restrictions on the magnitude of capital externalities with Cobb-Douglas technology. We find regions for joint indeterminacy and E-stability (i) when the felicity function is separable in consumption and leisure and there are negative capital externalities; or (ii) when the felicity function is non-separable and the social elasticity of production with respect to capital exceeds one. We further show that with the general utility function, a necessary condition for joint indeterminacy and E-stability is that the labor-demand curve is upward-sloping and steeper than the Frisch labor-supply curve.