UC Santa Cruz

This dissertation studies monetary policy design under different economic frameworks. The investigated frameworks include household heterogeneity, limitations to the policy response, and misalignments between the central bank and society's preferences concerning economic outcomes.

In the first chapter of this dissertation, a tractable two-agent, sticky-price, sticky-wage model is built to investigate the gains of wage flexibility. The model extends the traditional two-agent new Keynesian model by allowing for consumption heterogeneity among those households with restricted access to mechanisms for sharing idiosyncratic income risk. It is shown how consumption and wage inequality across the two household types, as well as the dispersion of consumption within those with limited ability to share risk, contribute new terms to the standard quadratic approximation to welfare, as well as increase the relative importance of stabilizing wage inflation. The effects of wage flexibility under a Taylor rule, as well as under optimal discretion and commitment policies, are studied using a calibrated version of the model with empirically plausible shock processes. Sources of welfare loss are affected by the type of shock and, more importantly, by how monetary policy is conducted. Under the Taylor rule, welfare loss is minimized if wages are very rigid while loss is minimized under the Ramsey policy if wages are more flexible. However, under the optimal discretionary policy, loss is lowest with either very flexible or very rigid wages. These findings challenge the common view that wage flexibility is particularly desirable, as the welfare implications of wage flexibility are strongly tied to the policy regime.

As a consequence of the Great Recession, the Federal Reserve Bank interest rate management was limited by the zero lower bound of the interest rate for at least seven years. This has been the longest duration of a zero lower bound regime ever registered in US history. Given the trend of having low levels of equilibrium interest rates in developed economies, it is believed that regimes of zero lower bound--or effective lower bounds if the bound is different from zero--will be more frequent than they were previously. In this vein, the second chapter of this dissertation evaluates the welfare implications of wage flexibility in a two-agent new Keynesian model in which monetary policy is occasionally limited by the effective lower bound on nominal interest rates. The model incorporates sticky wages and prices, yet it is tractable enough that finding a global solution to the non-linear equilibrium conditions is feasible. Thus, accurate calculations of the model dynamics and welfare, as well as different measures associated with the effective lower bound's frequency, are provided. Wage flexibility amplifies welfare cost when monetary policy responses are restricted by the effective lower bound. In fact, gains from higher wage flexibility, such as output gap stability, are far outweighed by the welfare cost induced by the rise in the volatility of prices and wages. Moreover, welfare loss does not disappear when the effective lower bound regime ends. Instead, it stays and creates long-run inefficiencies. The latter finding has been ignored in the literature, yielding systematic understatements of the welfare cost if the effective lower bound is a policy restriction.

The third chapter of this dissertation examines whether macroeconomic performance can be improved in goal- or rule-based systems, focusing on alternative goals and rules discussed in the literature as effects of the Great Recession. It evaluates whether these alternatives would have yielded welfare improvements in systems like the post-war pre-crisis US economy. Therefore, unlike most investigations, an empirically relevant model is used to assess the misalignments between the central bank and society's preferences. Goals outperform rules; however, the performance of rules that include the price level does not differ much from the performance of goals. Inflation targeting outperforms other goals regarding welfare; however, its optimal implementation could be tedious. Relative to inflation targeting, average inflation and price level targeting, as well as a persistent rate rule with price level, deliver comparable welfare gains and align better the central bank and society's preferences.