Beginning in 2007, this journal will publish a new series of supplements for special issues. This new supplements series is intended to remove some special issues from the regular-issues space budget and thereby help to avoid the crowding-out problem that had previously resulted from having special issues replace regular issues. Because of the rapid growth of excellent special issues proposals, we expect the number of supplements to be published per year to grow rapidly. Initially our objective is to publish one or two such supplement issues each year. The supplement will be published “when ready.” The initial space budget for the annual supplement is 144 printed pages, plus or minus 20% to accommodate the variance in special issue sizes. Hence all special issues must be of length within the following range of printed page: 115–173.

This paper compares the forecasting performance of some leading models of inflation for G-7 countries. We show that bivariate and trivariate models suggested by economic theory or statistical analysis are not much better than univariate ones. Phillips curve specifications fit well into this class. Improvements in both the MSE of the forecasts and turning point prediction are obtained with time-varying coefficients models, which exploit international interdependencies. The performance of the latter class of models is stable throughout the 1990s.

This paper develops methods to solve for optimal discretionary policies and optimal commitment policies in rational expectations models. These algorithms, which allow the optimization constraints to be conveniently expressed in second-order structural form, are more general than existing methods and are simple to apply. We use several New Keynesian business cycle models to illustrate their application. Simulations show that the procedures developed in this paper can quickly solve small-scale models and that they can be usefully and effectively applied to medium- and large-scale models.

This paper studies stylized business-cycle properties of household production in four industrialized countries (Canada, the United States, Germany, and Japan). We employ a dynamic small open-economy business-cycle model that incorporates a household production sector. We use the model to generate data on home output, hours worked in the home sector, and hours spent on leisure. We find that in each country, home output is more volatile than market output, whereas home sector hours are about as volatile as those in the market sector. In each country, leisure is the least volatile series. Leisure hours and home hours are countercyclical in all countries, and home output is not highly correlated with market output. Home sector variables are generally less persistent than market variables and cross-country correlations related to home production tend to be lower than those related to market production. These findings demonstrate that despite some well-known structural differences in labor markets, the cyclical features of home sector variables are similar across the countries we consider.

We consider a dynamic stochastic model of currency attacks, characterized by imperfect information about a fundamental. Agents not only decide whether to attack the peg but also formulate expectations concerning the probability of future devaluation. The subjective devaluation probabilities influence the inflation expectations, which, in turn, affect the next-period wage level and unemployment. Hence, expectations affect the following-period fundamental and the policymaker's ability to resist an attack. Agents decide upon next-period wage having observed whether the current-period currency attack has been successful or not: this publicly available information is sufficient to allow for a coordination effect among agents, leading to multiple equilibria. Our results imply that a ‘transparent policy’ during a currency crisis is in the policymaker's interest.

Specification tests reject a linear inflation forecasting model over the period 1959–2002. Based on this finding, we evaluate the out-of-sample inflation forecasts of a fully nonparametric model for 1994–2002. Our two main results are that: (i) nonlinear models produce much better forecasts than linear models, and (ii) including money growth in the nonparametric model yields marginal improvements, but including velocity reduces the mean squared forecast error by as much as 40%. A threshold model fits the data well over the full sample, offering an interpretation of our findings. We conclude that it is important to account for both nonlinearity and the behavior of monetary aggregates when forecasting inflation.

This paper sets up a two-sector growth model with sectoral hysteresis because of intersectoral factor reallocation costs. The main results are: (i) The economy under study exhibits nonergodic growth implying path-dependency (history matters) and permanent consequences of temporary shocks. (ii) Flexible economies are more likely to take advantage of technological improvements. The analysis points to a new mechanism in the flexibility-growth nexus, which complements the findings of Bertola (1994). (iii) Periods of negative growth can be explained as optimal responses of an economy to favorable technology shocks. This result sheds light on the fact that economic development is associated with recurring downturns.