By introducing automation development into a labor search model, this paper obtains that the increasing importance of automation in production may be responsible for the reduction in job reallocation along the transitional dynamics path. In the long run, we find automation also increases the total unemployment rate and reduces overall labor force participation. In addition, decreasing any disparity between differently skilled labor is detrimental to job reallocation along the transitional dynamics path, and both the long-run total unemployment rate and overall labor market participation will fall. Nevertheless, appropriate government subsidy policies can improve business dynamics across the labor market.

Japan’s so-called Lost Decade of the 1990s presents a unique case study of an economy with a recent severe and prolonged recession, with large changes in the labor market and fiscal policy as the main policy available to the government. Japanese unemployment rate surged from 2.1% in 1991 to 5.4% in 2002. Meanwhile, the Japanese economy experienced a rise in government expenditures, while taxes remained fairly stable. This paper quantitatively evaluates the impact of these changes in fiscal policies on labor market variables, in particular the unemployment rate, during the 1990s. We build, calibrate, and simulate a dynamic general equilibrium model with search frictions in the labor market, a productive government sector, heterogenous government spendings, and different categories of taxes. Our model is able to reproduce the paths of the main labor market variables, and the counterfactual experiments show that the changes that took place in the different spending components affected the unemployment rate heterogeneously, although overall they kept unemployment lower than it could have been. We also find that had the government also implemented countercyclical tax policies, unemployment would not have risen as much as it did by 2002.

The development of artificial intelligence (AI) does influence human jobs but not necessarily in a negative way. Although labor force participation rates and firms’ job vacancies for human labor decline, the unemployment rate may be lower than that in an economy without AI. In an economy with heterogeneously skilled workers, the invention of AI usually has a negative effect on the skilled labor market but a positive effect on the unskilled labor market. The overall unemployment rate may decline as AI develops.

This paper shows how fiscal policy affects unemployment in a New Keynesian model with search and matching frictions and distortionary taxation. The model is estimated using US data that includes labor market flows and distinct fiscal instruments. Several findings stand out. First, unemployment multipliers for spending and consumption tax cuts are substantial, even though output multipliers turn out to be less than one. Second, multipliers for labor tax cuts are small. Third, fiscal rules enhance the positive effects of discretionary fiscal policy. However, these expansionary effects on the multipliers are modest compared to earlier studies.

This paper builds a standard search model with flexible prices and wages, and extensive and intensive labor adjustments. Money is introduced into the model through a cash-in-advance constraint in which only consumption is cash constrained. The model reproduces labor-market dynamics under a productivity shock and/or a monetary shock. I can replicate the Beveridge and Phillips curves that are observed in the data, and do not need to rely on the New Keynesian model or real wage rigidity. I find that the nonexistence of an extensive margin and different money mechanisms, such as cash constraints on investment and money in the utility function, do not change the above replications. Furthermore, I can still replicate the Beveridge curve even without money or with rigid prices.

We calculate that the extension of unemployment insurance benefits during downturns has significantly increased the variability of unemployment and vacancies in the United States. Taking this into account reduces the value of leisure necessary to match the wide labor market business cycles experienced in the United States using the Mortensen--Pissarides model. For this calculation, we analyze a version of the model where unemployment insurance benefits not only expire but must be earned with prior employment. With these features, we can calibrate the model to be consistent with unemployment responding strongly to productivity shocks and mildly to changes in unemployment insurance policies. Our preferred calibration predicts that the standard deviation of unemployment since 1945 would have fallen by around 37% if there had not been programs extending unemployment benefits during recessions. We also find that the enactment of the Emergency Unemployment Compensation program in 2008 increased the unemployment rate by 0.5 percentage points.

This paper attempts to determine the factors generating the persistence of unemployment over the business cycle. The observations show that the total unemployment rate is highly persistent, and that the persistence of the unemployment rate of unskilled workers is higher than that of skilled workers. To account for these observations, the paper develops a framework that features search frictions. Individuals are either high educated or low educated, and firms post two types of vacancies: the complex, which can be matched with the high educated, and the simple, which can be matched with the high and the low educated. On-the-job search for a complex occupation is undertaken by the high educated in simple occupations. A negative aggregate technological shock induces the high educated unemployed to compete with the low educated by increasing their search intensity for simple vacancies. As the high educated occupy simple vacancies, they crowd out the low educated into unemployment. This downgrading of jobs in a cyclical downturn, or the increase in the labor input of the high educated in simple occupations, and the subsequent crowding out of the low educated into unemployment, provide a possible explanation for unemployment persistence.