Essays in Networks and Macroeconomics

Abstract

I develop an analytical framework to study monetary policy in an economy with multiple sectors and a general input-output network. I derive the Phillips curve and welfare as a function of the underlying production primitives, obtaining important new insights. (i) The presence of input-output linkages flattens the Phillips curve. (ii) There is an endogenous tradeoff between stabilizing output and stabilizing consumer prices. (iii) I construct a novel inflation index that eliminates this tradeoff, which I refer to as the “divine coincidence” index. (iv) Monetary policy faces a tradeoff between stabilizing aggregate output and relative output across sectors. While targeting the “divine coincidence” index stabilizes aggregate output, the optimal policy can be implemented by targeting an alternative inflation index which incorporates this tradeoff. I calibrate the model to the U.S. economy, providing a quantitative counterpart to (i)-(iv). Mirroring (i), the slope of the consumer price Phillips curve is one order of magnitude smaller in the muti-sector model than in the one-sector benchmark, and matches empirical estimates. The model also predicts a 30% decline over the past 70 years, arising from increased intermediate inputs in production. Validating (ii) and (iii), the divine coincidence inflation rate provides a better fit for Phillips curve regressions than conventional consumer price specifications. The baseline policy of targeting consumer inflation leads to a welfare loss of 1:12% of per-period GDP. Switching to the optimal policy in (iv) reduces this loss to 0:28%, but cannot fully eliminate it. Targeting the output gap almost replicates the optimal policy.