An Empirical Analysis of the Effects of Budget Deficits (Total and Primary) and Personal Income Tax Rates on the Ex Post Real Interest Rate Yield on Long-Term U.S. Treasury Bonds

Abstract

This empirical study adopts an open-economy loanable funds model to investigate the impact of post-Bretton Woods U.S. federal government budget deficits and personal income tax rates on the ex post real interest rate yield on thirty-year Treasury bonds. In this study, the budget deficit is measured in two different ways, the total (“unified”) budget deficit and the primary deficit (the total/unified deficit minus net interest payments). Two different estimation techniques, autoregressive two stage least squares estimation and the ARCH (Autoregressive Conditional Heteroscedasticity) Method, for the 1973-2016 study period provide evidence that the ex post real interest rate yield on thirty-year Treasury bonds has been an increasing function of both federal budget deficit measures (expressed as a percent of GDP) and the maximum marginal federal personal income tax rate. The estimations all imply that elevating either the total/unified or primary federal budget deficit appears to raise the cost of borrowing in the U.S., whereas reducing the maximum marginal personal income tax rate appears to reduce the cost of borrowing. Given the potential effects of longer-term real interest rates on investment in new plant and equipment and overall economic growth, policy-makers should not overlook these findings.