Title

Costs of Financing US Federal Debt: 1791-1933

Author(s)

George Hall George Hall (Brandeis University)

Jonathan Payne Jonathan Payne (Princeton University)

Thomas J. Sargent Thomas Sargent (New York University)

Bálint Szőke Bálint Szőke (Federal Reserve Board)

Abstract

We use computational Bayesian methods to estimate parameters of a statistical model of gold, greenback, and real yield curves for US federal debt from 1791 to 1933. Posterior probability coverage intervals indicate more uncertainty about yields during periods in which data are especially sparse. We detect substantial discrepancies between our approximate yield curves and standard historical series on yields on US federal debt, especially during War of 1812 and Civil War surges in government expenditures that were accompanied by units of account ambiguities. We use our approximate yield curves to describe how long it took to achieve Alexander Hamilton’s goal of reducing default risk premia in US yields by building a reputation for servicing debts as promised. We infer that during the Civil War suspension of convertibility of greenback dollars into gold dollars, US creditors anticipated a rapid post war return to convertibility at par, but that after the war they anticipated a slower return.

Creation Date

2021-09

Section URL ID

Paper Number

2021-25

URL

https://drive.google.com/file/d/1C99tAmPYu9FM8wB-j9kxArahYU8uVsAO/view

File Function

Jel

E31, E43, G12, N21, N41

Keyword(s)

Big data, default premia, yield curve, units of account, gold standard, government debt, Hamiltonian Monte Carlo, Julia, DynamicHMC.jl, pricing errors, specification analysis

Suppress

false

Series

13