This project extends the investigator's work on unemployment in four directions. The first part continues work on unemployment in continental Western Europe. Quantitative time series are constructed for various French labor market institutions and analyzed in order to understand the role of changes in the political and economic environments affecting them. The project explores the role of trust in a society in affecting the operation of labor markets. It explores differential wage and unemployment history in Ontario and Quebec, particularly the role of political and cultural conflict in the latter province. And it explores why the labor share fell in Europe in the last decades of the twentieth century.
The second returns to the evolution of unemployment in Central and Eastern Europe. The project studies why unemployment in many of the previously socialist economies has remained so high. This is in spite of the fact that institutions have become "employment friendly," i.e., that unemployment insurance is meagre and legal restraints on firing are few. According to accepted labor theory, this should reduce the unemployment rate. Since labor market institutions have been imported from the West without the predicted outcomes, other aspects must be involved. This project studies various factors that may contribute to mismatch between the characteristics demanded by employers and those supplied by prospective workers.
The third extends earlier normative work on labor market institutions, especially on the design of unemployment insurance and employment protection. In past work, Jean Tirole and the investigator have analyzed these issues in a static context; the justification was to clarify the role of a various imperfections, starting from a simple benchmark. They now want to introduce dynamics in a number of dimensions. More specifically, they will extend their work shifting funding of unemployment benefits from payroll taxes to a layoff tax, in exchange for reduction in the legal protections workers enjoyed against termination.
The fourth would continue work on monetary policy and the labor market. In recent work, Jordi Gali and the investigator have shown how the presence of imperfections in the labor market substantially modifies the conclusions of the standard New-Keynesian model with respect to optimal monetary policy. They are working further, both on the microfoundations, and on practical issues such as the optimal response to movements in the price of oil.
The work described in the proposal can make a substantive intellectual contribution, namely a better understanding of the factors behind variations in unemployment and the work can make a substantive broader contribution, namely help in the design of better structural and macro policies.
The project has pursued various open questions on the determinants of aggregate economic activity and employment in the short run and on the optimal response of policy. Let me discuss three articles that were supported by the project. "Labor Markets and Monetary Policy" by Gali and Blanchard is one of the first papers to incorporate an explicit model of workers searching for jobs in a new Keynesian model, which is the standard model used today in central banks around the world to think about the impact of different monetary policy decisions. This allows the authors to study the consequences of different monetary policies for unemployment. In particular, the authors can ask the question: in what cases the central bank can pursue both price stability and full employment with the same policy and in what cases there is a trade-off between inflation and unemployment. When a tradeoff exists, the authors can identify the parameters to be estimated in empirical work that are crucial to evaluate the costs and benefits of alternative policies. In "News, Noise and Fluctuations," Blanchard, L'Huillier and Lorenzoni explore empirically the idea of economic fluctuations as driven by consumer sentiment. While this idea is common in policy discussions, it has received relatively little attention in formal modeling and estimation. The authors build a model that captures the idea that consumers continuously receive information about the future growth potential of the economy and that this information is a prime driver of consumer spending in the short run. Consumer spending then determines aggregate activity through a standard Keynesian channel. The paper explores both issues of method and of substance. On the method front, the paper shows what econometric method can and cannot be used to estimate these class of models. Structural VAR methods seem ill suited for the task. Other approaches (like Maximum Likelihood) work, but require more trust in the structure of the model used. Looking at US post-war time series, the authors show that disturbances that look like consumer sentiment shocks seem to account for about half of consumption volatility at business cycle frequencies. In "Credit Crises, Precautionary Savings and the Liquidity Trap" Guerrieri and Lorenzoni look at the effects of a contraction in credit to consumers on consumption and aggregate activity. If the financial sector is less willing to extend credit to consumers who need to borrow, this translates into a bigger overall demand for safe and liquid assets by consumers. If the supply of safe and liquid assets is inelastic, this translates into a large drop in equilibrium interest rates. In an economy where good prices adjust slowly, the nominal interest rate may hit its lower bound at zero. In this case, a credit crunch can have bigger recessionary effects, as the drop in consumer demand cannot be compensated by a drop in interest rates.
