Research Database

A cash-in-advance model with sequential markets is constructed, where unanticipated monetary injections are nonneutral and can potentially produce large liquidity effects. However, if the monetary authority adheres to an optimal money rule, money should not respond to unanticipated shocks, so that a Friedman rule is suboptimal and the monetary authority does not exploit the liquidity effect. Quantitatively, the model can generate variability in money and nominal interest rates close to what is observed, and can produce data with no obvious evidence of the existence of liquidity effects.

I show that a simple sticky price model based on Rotemberg (1982) is consistent with a variety of facts concerning the correlation of prices, hours and output. In particular, I show that it is consistent with a negative correlation between the detrended levels of output and prices when the Beveridge-Nelson method is used to detrend both the price and output data. Such a correlation, i.e.,a negative correlation between the predictable movements in output and the predictable movements in prices is present (and very strong) in U.S. data. Consistent with the model, this correlation is stronger than correlations between prices and hours of work. I also study the size of the predictable price movements that are associated with predictable output movements as well as the degree to which there are predictable movements in monetary aggregates associated with predictable movements in output. These facts are used to shed light on the degree to which the Federal Reserve has pursued a policy designed to stabilize expected inflation.

We examine price formation in a simple static model with asymmetric information, a countable number of risk neutral traders and without noise traders. Prices can exhibit excess volatility (the variance of prices exceeds the variance of dividends), even in such a simple model. More generally, we show that for an open set of parameter values no equilibrium has prices which turn out to equal the value of dividends state by state, while for another open set of parameter values there exist equilibria such that equilibrium prices equal the value of dividends state by state. When information collection is endogenous and costly, expected prices exhibit a "V-shape" as a function of the cost of information: They are maximized when information is either costless so that everyone acquires it, or else is so costly that no one chooses to acquire it. Prices are depressed if information is cheap enough so that some agents become informed, while others do not. If the model is altered so that information is useful in making productive decisions, then the V-shape is altered, reducing the attractiveness of prohibitively high costs.

This paper develops a dynamic model of general imperfect competition by embedding the Shapley-Shubik model of market games into an overlapping generations framework. Existence of an open market equilibrium where there is trading at each post is demonstrated when there are an arbitrary (finite) number of commodities in each period and an arbitrary (finite) number of consumers in each generation. The open market equilibria are fully characterized when there is a single consumption good in each period and it is shown that stationary open market equilibria exist if endowments are not Pareto optimal. Two examples are also given. The first calculates the stationary equilibrium in an economy, and the second shows that the on replicating the economy the stationary equilibria converge to the unique non-autarky stationary equilibrium in the corresponding Walrasian overlapping generations economy. Preliminary on-going work indicates the possibility of cycles and other fluctuations even in the log-linear economy.

A cash-in-advance model with sequential markets is constructed, where unanticipated monetary injections are nonneutral and can potentially produce large liquidity effects. However, if the monetary authority adheres to an optimal money rule, money should not respond to unanticipated shocks, so that a Friedman rule is suboptimal and the monetary authority does not exploit the liquidity effect. Quantitatively, the model can generate variability in money and nominal interest rates close to what is observed, and can produce data with no obvious evidence of the existence of liquidity effects.

We formulate a representative consumer model of intertemporal resource reallocation in which fluctuations in equity prices contribute to the smoothing of consumption flows. Features of the model include (a) an incompletely observable stochastic process of productivity shocks leading to fluctuating confidence of beliefs and (b) technologies involving commitments of a resource good. These features are exploited to show that (1) equities are not a representative form of total wealth and (2) the valuation of currently active firms is not representative of the valuation of all firms. We examine the implications of (1) and (2) to argue that empirical findings for the volatility and 'value shortfall' of equity prices may be consistent with a frictionless representative consumer model having a low degree of risk-aversion. Simulation of a calibrated version of the model for a risk-neutral consumer shows that when the 'data' is analyzed according to current econometric procedures, it is found to exhibit volatility of the same order of magnitude as that found in the actual data, although the model contains no excess volatility.

Over the past decade, a substantial literature on the estimation of discrete choice dynamic programming (DC-DP) models of behavior has developed. However, this literature now faces major computational barriers. Specifically, in order to solve the dynamic programming (DP) problems that generate agents' decision rules in DC-DP models, high dimensional integrations must be performed at each point in the state space of the DP problem. In this paper we explore the performance of approximate solutions to DP problems. Our approximation method consists of: 1) using Monte Carlo integration to simulate the required multiple integrals at a subset of the state points, and 2) interpolating the non-simulated values using a regression function. The overall performance of this approximation method appears to be excellent, both in terms of the degree to which it mimics the exact solution, and in terms of the parameter estimates it generates when embedded in an estimation algorithm.

Standard models of public education provision predict an implicit transfer of resources from higher income individuals toward lower income individuals. Many studies have documented that public higher education involves a transfer in the reverse direction. We show that this pattern of redistribution is an equilibrium outcome in a model in which education is only partially publicly provided and individuals vote over the extent to which it is subsidized. We show that increased inequality in the income distribution makes this outcome more likely and that the efficiency implications of this exclusion depend on the wealth of the economy.

A standard explanation for why sovereign governments repay their debts is that they must maintain a good reputation to easily borrow more. We show that the ability of reputation to support debt depends critically on the assumptions made about institutions. At one extreme, we assume that bankers can default on payments they owe to governments. At the other, we assume that bankers are committed to honoring contracts made with governments. We show that if bankers can default, then a government gets enduring benefits from maintaining a good relationship with bankers and its reputation can support a large amount of borrowing. If, however, bankers must honor their contracts, then a government gets only transient benefits from maintaining a good relationship and its reputation can support zero borrowing.