Research Database

The consequences of a straightforward monetary targeting scheme are examined for a simple dynamic macro model. The notion of “targeting” used is the strategic one introduced by Rogoff (1985). Numerical calculations are used to demonstrate that for the model under consideration, monetary targeting is likely to lead to a deterioration of policy performance. These examples cast doubt upon the general efficacy of simple targeting schemes in dynamic rational expectations models.

Deaton (1986) has noted that if income is a first-order autoregressive process in first differences, then a simple version of Friedman’s permanent income hypothesis (SPIH) implies that measured U.S. consumption is insufficiently sensitive to innovations in income. This paper argues that this implication of the SPIH is a consequence of the fact that it ignores the role of the substitution effect in the consumption decision. Using a parametric version of the standard model of economic growth, the paper shows that very small movements in interest rates are sufficient to induce an empirically plausible amount of consumption smoothing. Since an overall evaluation of the model’s explanation for the observed smoothness of consumption requires examining its implications for other aspects of the data, the paper also explores some of these.

This paper reexamines U.S. postwar data to investigate if the observed comovements between money, interest rates, inflation, and output are compatible with the money to real interest to output links suggested by existing monetary theories of the business cycle, which include both Keynesian and equilibrium models. We find these theories are incompatible with the data, and in light of these results, we propose an alternative structural model which can account for the major dynamic interactions among the variables. This model has two central features: (i) output is unaffected by the money supply; and (ii) the money supply process is influenced by policies designed to achieve short-run price stability.

In this paper we propose and test a new explanation of bank behavior during the Free Banking Era, 1837–63. Arguing against the view that free bank failures were due to fraud, we claim that they were caused by exposure to term structure risk. Testing this new explanation with a new and extensive body of data, we find strong support for it: periods of falling bond prices correspond to the periods with most of the free bank failures. The new data do not support the view that fraud caused the failures.

We analyze a general equilibrium model with search frictions and differentiated commodities. Because of the many differentiated commodities, barter is difficult because it requires a double coincidence of wants, and this provides a medium of exchange role for fiat money. We prove the existence of equilibrium with valued fiat money and show it is robust to certain changes in the environment, including imposing transactions costs, storage costs, and taxes on the use of money. Rate of return dominance, liquidity, and the potential welfare improving role of fiat money are discussed.

During the period 1870–1913, Canada had a well-diversified branch banking system while banks in the U.S. unit banking system were less diversified. Canadian banks could issue large-denomination notes with no restrictions on their backing, while all U.S. currency was essentially an obligation of the U.S. government. Also, experience in the two countries with regard to bank failures and banking panics was quite different. A general equilibrium business cycle model with endogenous financial intermediation is constructed that captures these historical Canadian and American monetary and banking arrangements as special cases. The predictions of the model contradict conventional wisdom with regard to the cyclical effects of banking panics. Support for these predictions is found in aggregate annual time series data for Canada and the United States.

Some Revenue Sharing programs, including the Federal government’s General Revenue Sharing program, reward higher tax effort with larger aid payments. A natural, game-theoretic generalization of the standard consumer demand based theory of grants-in-aid is used to examine the impacts such tax effort provisions have on the recipient government’s tax effort, spending levels, and welfare. Nonlinear simulation is used to provide rough quantitative estimates of the impacts General Revenue Sharing had in 1972.

Restrictions that general equilibrium theory place upon average returns are found to be strongly violated by the U.S. data in the 1889–1978 period. This result is robust to model specification and measurement problems. We conclude that equilibrium models which are not Arrow-Debreu economies are needed to rationalize the large average equity premium that prevailed during the last 90 years.

The qualitative dynamics of a discrete time version of a deterministic, continuous time, nonlinear macro model formulated by Haavelmo are fully characterized. Recently developed methods of symbolic dynamics and ergodic theory are shown to provide a simple, effective means of analyzing the behavior of the resulting one-parameter family of first-order, deterministic, nonlinear difference equations. A complex periodic and random "aperiodic" orbit structure dependent on a key structural parameter is present, which contrasts with the total absence of such complexity in Haavelmo's continuous time version. Several implications for dynamic economic modelling are discussed.

This paper describes a Bayesian specification procedure used to generate a vector autoregressive model for forecasting macroeconomic variables. The specification search is over parameters of a prior. This quasi-Bayesian approach is viewed as a flexible tool for constructing a filter which optimally extracts information about the future from a set of macroeconomic data. The procedure is applied to a set of data and a consistent improvement in forecasting performance is documented.

The statistical significance of variance decompositions and impulse response functions for unrestricted vector autoregressions is questionable. Most previous studies are suspect because they have not provided confidence intervals for variance decompositions and impulse response functions. Here two methods of computing such intervals are developed, one using a normal approximation, the other using bootstrapped resampling. An example from Sims’ work illustrates the importance of computing these confidence intervals. In the example, the 95 percent confidence intervals for variance decompositions span up to 66 percentage points at that usual forecasting horizon.

On our interpretation, real bills advocates favor unfettered intermediation, while their critics, who we call quantity theorists, favor legal restrictions on intermediation geared to separate “money” from “credit.” We display examples of economies in which quantity-theory assertions about “money-supply” and price-level behavior under the real bills regime are valid. In particular, both the price level and an asset total that quantity theorists would identify as money fluctuate more under a real bills regime than under a regime with restrictions like those favored by quantity theorists. Despite this, the Pareto criterion does not support the quantity-theory position.

Silberberg [6] and Pauwels [2] have produced and clarified seminal results in the comparative statics of single-agent classical optimization problems. This paper extends Pauwels’ method to derive analogous results for stable Nath equilibria in a subclass of the widely used class of concave orthogonal games defined by Rosen [3]. Application of these results to cost curve shifts in the asymmetric Cournot oligopoly immediately uncovers apparently new comparative statics results.

This paper derives a variance bounds test for a broad class of linear rational expectations models. According to this test, if observed data accord with the model, then a weighted sum of auto-covariances of the covariance-stationary components of the endogenous state variables should be nonnegative. The new test reinterprets West’s (1986) variance bounds test and extends its applicability by not requiring observable exogenous state variables, covariance-stationary exogenous or endogenous state variables, or a zero initial value for the endogenous state variable. The paper also discusses the possibility of the new test’s application to nonlinear models.

This paper proposes a method for estimating the parameters of continuous time, stochastic rational expectations models from discrete time observations. The method is important since various heuristic procedures for deducing the implications for discrete time data of continuous time models, such as replacing derivatives with first differences, can sometimes give rise to very misleading conclusions about parameters. Our proposal is to express the restrictions imposed by the rational expectations model on the continuous time process generating the observable variables. Then the likelihood function of a discrete time sample of observations from this process is obtained. Parameter estimates are computed by maximizing the likelihood function with respect to the free parameters of the continuous time model.

We examine the limiting behavior of cooperative and noncooperative fiscal policies as countries’ market power goes to zero. We show that these policies converge if countries raise revenues through lump-sum taxation. However, if there are unremovable domestic distortions, such as distorting taxes, there can be gains to coordination even when a single country’s policy cannot affect world prices. These results differ from the received wisdom in the optimal tariff literature. The key distinction is that, unlike in the tariff literature, the spending decisions of governments are explicitly modeled.

This paper explores some of the implications for econometric practice of the principle that people’s observed behavior will change when their constraints change. In dynamic contexts, a proper definition of people’s constraints includes among them laws of motion that describe the evolution of the taxes they must pay and the prices of the goods that they buy and sell. Changes in agents’ perceptions of these laws of motion (or constraints) will in general produce changes in the schedules that describe the choices they make as a function of the information that they possess. Until very recently, received dynamic econometric practice ignored this principle. The practice of dynamic econometrics should be changed so that it is consistent with the principle that people’s rules of choice are influenced by their constraints. This is a substantial undertaking, and involves major adjustments in the ways that we formulate, estimate, and simulate econometric models.

Methods are presented for solving a certain class of rational expectations models, principally those that arise from dynamic games. The methods allow for numerical solution using spectral factorization algorithms and for estimation of these models using maximum likelihood techniques.

Using a simple model, we show why previous empirical studies of budget policy effects are flawed. Due to an identification problem, those studies’ findings can be shown to be consistent with either policies mattering or not.

This paper describes methods for estimating the parameters of continuous time linear stochastic rational expectations models from discrete time observations. The economic models that we study are continuous time, multiple variable, stochastic, linear-quadratic rational expectations models. The paper shows how such continuous time models can properly be used to place restrictions on discrete time data. Various heuristic procedures for deducing the implications for discrete time data of these models, such as replacing derivatives with first differences, can sometimes give rise to very misleading conclusions about parameters. The idea is to express the restrictions imposed by the rational expectations model on the continuous time process of the observable variables. Then the likelihood function of a discrete-time sample of observations from this process is obtained. Estimators are obtained by maximizing the likelihood function with respect to the free parameters of the continuous time model.