Three essays on endogenous time preference, monetary non-superneutrality and the Mundell-Tobin effect

by Kam, Avrum Eric

The dominant opinion in neoclassical macroeconomics is that monetary growth raises the oppominity cost of holding real balances, decreases the steady state demand for money (a substitution effect) but maintains no effects on real sector variables (absence of a wealth effect). This is referred to as monetary supemeutrality, or an absence of a Mundell-Tobin effect, and is accepted as an inevitable result of monetary growth in optimization models, which assume the existence of infinitely lived, representative agents. This argument is strongly rejected since exogenous tirne preference implies monetary supemeutrality by holding constant both the real interest rate and the marginal product of capital, which removes the link between the real and monetary sectors. Paper one provides a restatement of the Mundell-Tobin effect and non-superneutrality using optimizing underpinnings of the infinitely lived, representative agent. Real balances are assurned to be a shiR parameter in the utility function and the rate of time preference, which is modeled as an increasing Function of real wealth. defined as real balances plus capital. Monetary growth decreases real wealth and the rate of time preference, which increases savings, consurnption and the capital stock. This is an optirnizing equivalent to the older ad-hoc models, which assume that savings are a decreasing function of wealth. Paper two relaxes the assumption that real balances yield utility. This is demonstrated to have no consequence on the existence of a Mundell-Tobin effect. Again, wealth effects are modeled through endogenous time preference and monetary growth raises capital and consumption. In a complementary model, real balances are removed as a shift parameter kom the rate of time preference and the utility function. The result is that the determinant of the coefficient matrix is zero, which implies that if real balances do not enter into the representative agent and their optimizing decisions, the model is intractable. However, this reinforces endogenous time preference as a necessary condition for the existence of a Mundeil-Tobin effect. Paper three examines the role of two cash-in-advance constraints. This constraint can apply to the purchase of consumption and capital goods or stnctly on consumption goods. In the presence of either type, the real effects of monetary growth are demonstrated to be independent of the decision between exogenous and endogenous tirne preference. If the constraint applies stnctly to consumption goods, monetary gowth is superneutral. But, if the constraint applies to both capital and consurnption goods, monetary growth decreases both commodities. These results are proven to be independent of time preference. Cash-in advance constraints dominate wealth effects implied by endogenous time preference and replicate the assurnption of exogenous time preference. The fint two papers prove that modeling time preference as an increasing function of real wealth is a necessary condition to generate a Mundell-Tobin effect. Supemeutrality has been broken using tools such as overlapping generations models. However, using the optimizing model of an infinitely lived, representative agent. endogenous time preference yields a less restrictive method to break monetary supemeutrality and ensure that the real sector is not immune from monetary growth. However, paper three demonstrates that the cash-in-advance or medium of exchange motive outweighs wealth effects associated with endogenous time preference and rules out the possibility of a suflciency argument.