Research Database

From introduction: This paper is intended to introduce economists to a simple SIR model of the progression of COVID-19 to aid understanding of how such a model might be incorporated into more standard macroeconomic models. An SIR model is a Markov model of the spread of an epidemic in which the total population is divided into categories of being susceptible to the disease (S); actively infected with the disease (I); and resistant (R), meaning those that have recovered, died from the disease, or have been vaccinated. The initial distribution of the population across these states and the transition rates at which agents move between these three states determine how an epidemic plays out over time. These transition rates are determined by characteristics of the underlying disease and by the extent of mitigation and social distancing measures. This model allows for quantitative statements regarding the tradeoff between the severity and timing of suppression of the disease through social distancing and the progression of the disease in the population.

This paper shows that some key stylized facts of exchange-rate-based stabilization plans can be explained by the uncertain duration of the plans themselves. Uncertain duration is modeled to reflect evidence showing that devaluation probabilities are higher when the plans are introduced and abandoned than in the period in between. If contingent-claims markets are incomplete, this uncertain duration distortion introduces temporary fiscal cuts with large wealth effects. Investment and employment are also distorted, and the resulting supply-side effects play a critical role. Stabilizations of uncertain duration entail large welfare costs, but they are preferred to persistent high inflation. México’s experience is examined in the light of these predictions.

We determine the second best rule for the inflation tax in monetary general equilibrium models where money is dominated in rate of return. The results in the literature are ambiguous and inconsistent across different monetary environments. We compare the derived optimal inflation tax solutions across the different environments and find that Friedman’s policy recommendation of a zero nominal interest rate is the right one.

Two types of unemployment insurance systems are studied. In one, unemployed workers receive benefits while those on reduced hours do not, as in North America (at least until recently). In the other, short-time compensation is paid to workers on reduced hours, as in Europe. With incomplete experience-rating of unemployment insurance taxes, the first system leads to inefficient temporary layoffs. The latter system does not lead to layoffs but does lead to inefficient hours per worker. Some cross-country evidence is presented regarding these effects. The implication of the analysis is that policy reform on the tax, not the benefit, side of the system is the best way to reduce the inefficiencies implied by both types of unemployment insurance.

This study describes a model built on the long-held view that the use of money as a medium of exchange is the result of an absence of double coincidence of wants. The model can account for two of the most challenging observations facing monetary theory: The disparate short-run and long-run effects of changes in the quantity of money and the coexistence of money and assets with higher rates of return. For both observations, the model's ability to provide a rich analysis depends on little more than the ingredients implicit in the absence-of-double-coincidence view.

We develop a Q model of investment with multiple capital goods that delivers a one-to-one relation between the growth rate of the capital aggregate and the stock market-based Q. We estimate the growth-Q relation using a panel of over six hundred Japanese manufacturing firms taking into account the endogeneity of Q. Identification is achieved by combining the theoretical structure of the Q model and an assumed serial correlation structure of the technology shock that comprises the error term in the growth-Q relation. The Q variable is significantly related to firm growth. Much, but not all, of the apparent explanatory power of cash flow disappears if its endogeneity is corrected for. The estimated Q coefficient is not implausibly small if the growth rate of the capital aggregate contains measurement error.

This article uses data from the 1998 Survey of Consumer Finances and from recent waves of the Panel Study of Income Dynamics to update a study of economic inequality in the United States based on 1992 and earlier data. The article reports data on the U.S. distributions of earnings, income, and wealth and on related features of inequality, such as age, employment status, educational attainment, and marital status. It also reports data on the economic inequality among U.S. households in financial trouble and on the economic mobility of U.S. households. The article finds that earnings, income, and wealth were very unequally distributed among U.S. households late in the 1990s, just as they had been at the beginning of the decade. It concludes that the basic facts about economic inequality in the United States did not change much during the 1990s.