International Capital Inflows, Federal Budget Deficits, and Interest Rates, 1971-1984

Article excerpt

INTRODUCTION

The impact of federal budget deficits on the rate of interest has been studied extensively.(1) Most of these studies focus on the short-term rate of interest, especially the three month U.S. Treasury bill rate or the four to six month commercial paper rate; a few of these studies focus on long-term rates of interest, such as the ten year U.S. Treasury note rate, the 20 year U.S. Treasury bond rate, the Moody's Aaa-rated corporate bond rate, or the Moody's Baa-rated corporate bond rate. The various studies have focused on different time periods, principally between the years 1954 and 1984.

There is no consensus regarding the interest rate impact of the budget deficit. Many of the studies that focus on the short-term rate of interest find that the deficit exercises no significant impact, although there are exceptions (cf. Barth, Iden, and Russek, 1984 and 1985; Cebula, 1987; Tanzi, 1985; and Zahid, 1988). A majority of the studies that focus on the longer-term rate of interest, however, find that the deficit does exercise a positive and significant impact, although there are a number of exceptions (cf. Evans, 1985 and 1987; Mascaro and Meltzer, 1983). Unfortunately, interpretation of results is complicated by differences in the time periods examined in the various studies. Several studies deal with the period 1955 to 1984; others deal with time periods such as 1979 to 1983 (using monthly data); still others begin with the year 1971 or 1972 or roughly the time period when the regime of fixed exchange rates (Bretton Woods) began to collapse and net international capital flows into the United States began to grow enormously. Because of the latter considerations, different time periods involve different macroeconomic circumstances. Most of the studies dealing with the time period beginning in the early to late 1970s have two traits in common:

* They generally find that federal budget deficits in the United States do not influence interest rates; and

* They generally omit international capital flows from their models.

Johnson (1992, pp. 145-146) recently has observed that it is essential for us to "... understand why the data consistently support mutually exclusive views concerning the impact of debt financing. For a question that has far-reaching policy implications, this ambiguity is not acceptable." Accordingly, to identify the impact of federal budget deficits on interest rates in the United States from 1971 to 1984, this study seeks to investigate empirically the impact of both the choice of interest rate measure and of time period studied upon the interest rate impact of the federal budget deficit.

The model examined below uses quarterly data, deals with five different interest rate measures, includes net international capital inflows, and examines two different time periods. The empirical analysis demonstrates that both the choice of interest rate studied and the time period studied affect the empirical results obtained.

THE MODEL

The analysis is couched within the familiar IS-LM framework, although a standard loanable funds model can be shown to yield nearly identical conclusions. This study empirically examines the open economy version of the standard IS-LM model. As noted in Hoelscher (1986) and Johnson (1992), this model basically differs from the closed IS-LM model by including net international capital inflows [NCAP].

Following Cebula (1987), Evans (1985), Makin (1983), and Ostrosky (1990), we assert that, according to the IS-LM paradigm, the nominal rate of interest (R) is determined principally by real government purchases of goods and services (G), the real budget deficit (D), the real exogenously determined money stock (M), and expected inflation (P):

(1) R = R(G, D, M, P)

where we expect:

(2) [R.sub.G] [greater than] 0, [R.sub.D] [greater than] 0, [R.sub.M] [less than] 0, [R.sub.p] [greater than] 0

with subscripts denoting partial differentiation. …