Debt and growth: Is there a debt threshold? An empirical investigation for the United States
Date
2021
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
This dissertation takes a pure time-series approach to investigate the current debate sparked by Reinhart and Rogoff (2010) on whether there is a public debt threshold beyond which debt has a significantly detrimental effect on growth. To abstract from the cross-country heterogeneity, this dissertation focuses on one country, the United States, thereby complementing the literature of panel studies on the debt-growth nonlinear relationship. ☐ This dissertation makes several contributions to the literature. First, my dissertation presents the first econometric evidence on the threshold relationship between growth and net government debt, which is deemed as a better government debt measure than the gross government debt. Second, my dissertation applies a Threshold Vector Error Correction Model (TVECM) to the debt-growth study. This model setup allows for a more general form of the threshold variable but does not rule out the possibility that the debt over GDP ratio is the threshold. Third, My dissertation develops a two-step procedure to test the presence of cointegration and nonlinearity in the dynamic relationship between debt and GDP. ☐ After examining six different debt measures based on three data samples, my dissertation concludes that: First, the threshold level should not be constant if the debt over GDP ratio is used as the threshold variable. This finding also casts doubt on the validity of the empirical literature on the debt threshold effect on GDP. Second, this dissertation does not find support for Reinhart and Rogoff's hypothesis. In particular, there is no evidence that high debt beyond a certain threshold will slow down economic growth in the United States. One possible explanation for the lack of evidence of Reinhart and Rogoff's hypothesis is that the U.S. government debt has not yet reached its threshold.
Description
Keywords
Generalized impulse response, Public debt, Reinhart and Rogoff's hypothesis, Threshold, TVECM