A Time Series Analysis of the Relationship between the Capital Stock and Federal Debt: Comment

Abstract

In the current controversy over the economic effects of federal debt, a conventional and widely held view is an increase in federal debt adversely affects the capital stock. The alternative and more recent view is increasingly attracting attention is increases in federal debt do not generate adverse capital stock effects.' The resolution of this important controversy depends heavily upon the empirical evidence pertaining to the realtionship between these two variables. Unfortunately, the evidence accumulated thus far is quite meager. The recent and important study of this by de Leeuw and Holloway (1985) is therefore quite timely and merits careful consideration. The reduced-forrn of de Leeuw and Holloway's model suggests that the debt/income has an unambiguous negative to the capital/output ratio (p 337) 2 de Leeuw and Holloway's analysis employs annual observations of a new measure of the cyclically adjusted U.S. federal debt and presents regression results (which) confirrn a strong negative relationship (p. 240) between this federal debt measure and the aggregate capital stock for the period 1955-1983. Our analysis extends the work of de Leeuw and Holloway in two important dimensions. First, we attempt to replicate de Leeuw and Holloway's results both for the aggregate capital stock and for the major disaggregated capital stock components using their original statistical methodology. While our results generally reconfirm de