Despite decades of research, the consistent estimation of structural forward-looking macroeconomic equations remains a formidable empirical challenge because of pervasive endogeneity issues. Prominent cases—the estimation of Phillips curves, Euler equations, or monetary policy rules—have typically relied on using predetermined variables as instruments, with mixed success. In this work, we propose a new approach that consists in using sequences of independently identified structural shocks as instrumental variables. Our approach is robust to weak instruments and is valid regardless of the shocks’ variance contribution. We estimate a Phillips curve using monetary shocks as instruments and find that conventional methods substantially underestimate the slope of the Phillips curve.

The unemployment rate is one of the most important business cycle indicators, but its interpretation can be difficult because slow changes in the demographic composition of the labor force affect the level of unemployment and make comparisons across business cycles difficult. To purge the unemployment rate from demographic factors, labor force shares are routinely used to control for compositional changes. This paper shows that this approach is ill defined, because the labor force share of a demographic group is mechanically linked to that group's unemployment rate, as both variables are driven by the same underlying worker flows. We propose a new demographic-adjustment procedure that uses a dynamic factor model for the worker flows to separate aggregate labor market forces and demographic-specific trends. Using the U.S. labor market as an illustration, our demographic-adjusted unemployment rate indicates that the 2008–2009 recession was much more severe and generated substantially more slack than the early 1980s recession.

The evaluation of macroeconomic policy decisions has traditionally relied on the formulation of a specific economic model. In this work, we show that two statistics are sufficient to detect, often even correct, nonoptimal policies, i.e., policies that do not minimize the loss function. The two sufficient statistics are (i) forecasts for the policy objectives conditional on the policy choice and (ii) the impulse responses of the policy objectives to policy shocks. Both statistics can be estimated without relying on a specific structural economic model. We illustrate the method by studying US monetary policy decisions. (JEL E24, E31, E32, E43, E52, E58)