Inflation as an Outcome of Optimal Fiscal and Monetary Policy (Job Market Paper)
I develop a neoclassical model to study how a welfare-maximizing government should conduct fiscal and monetary policy to finance exogenous government purchases and transfer payments. The government raises revenue through distortionary taxes, and finances its expenditures by issuing nominal one-period bonds and consols. Optimal policy necessitates lump-sum wealth transfers from government bondholders to the government through unexpected inflation and changes in nominal interest rates. I construct a sequence of sharply focused flexible price economies to show that the government can support the complete markets real allocations of Lucas and Stokey (1983) with moderate inflation volatility. Policy that combines state-contingent expected inflation targets with consol issuance is most effective at reducing inflation associated with adverse government purchase shocks and adverse productivity shocks. Responses to shocks in the model are broadly consistent with responses of the US economy to analogous shocks beginning in 1971. Furthermore, movements in inflation and the debt-to-gdp ratio from 2019-2023 are consistent with the directions implied by welfare-maximizing policy given the behavior of transfer payments and the policy interest rate. Motivated by the model's central redistribution channel, I use microdata from the PSID to show that households' consumption responds to unexpected changes in the real value of their fixed-income assets. This is consistent with the idea that wealth is transferred between holders of all fixed-income assets and liabilities as an externality of optimal policy.
Financial Institutions in General Equilibrium
I study an economy in which households and firms may only enter multi-period contracts with each other through banks in the financial sector. Firms use two factors of production: bank loans and labor. The use of bank loans as an additional input to production creates a time-varying labor wedge, which is a function of loan and deposit interest rates. The labor wedge is the channel through which financial shocks can influence real variables, even under flexible prices. However, data on interest rates reveal the labor wedge to be small, so that equilibrium in the presence of a financial sector is nearly identical to that in an economy without an explicitly modeled financial sector.
Banks are subject to idiosyncratic deposit withdrawal shocks at the end of each period, and banks hold excess reserves to avoid overdraft penalties imposed by the central bank. Empirically, aggregate bank holdings of excess reserves per unit of deposits are trendless from 1959 to 2007. Further, when the Federal Funds rate is equal to the rate on excess reserves, the model implies that banks will hold one dollar of reserves in excess of their required reserves for each dollar of deposits. This is consistent with the large increase in bank reserves and the size of the Federal Reserve's balance sheet beginning in 2008. In order to reduce the size of its balance sheet, the Federal Reserve can set the rate on excess reserves below the Federal Funds rate.
Estimating Inflation Using Inflation Expectations
Inflation is difficult to measure directly. Even a chain-weighted ideal price index may not accurately measure inflation because new goods enter, old goods leave, and the quality of goods may change over time. First, I show that a time series for inflation can be recovered from a time series of inflation expectations by imposing an assumption that the true model for inflation is ARX(p). I use the Kalman Filter to extract the time series for inflation implied by the Michigan Survey of Consumers question on Inflation Expectations. The PCE Deflator implies that real consumption per capita grew 80% from 1978 to 2007, and doubled between 1978 and the present. Michigan Survey Expectations imply that real consumption per capita grew 60% from 1978 to 2007, and subsequently plateaued so that real consumption per capita today is roughly 60% higher than it was in 1978.