An inquiry concerning long-term U.S. interest rates using monthly data

ABSTRACT

This paper undertakes an empirical inquiry concerning the determinants of the long-term interest rate on U.S. Treasury securities. It applies the bounds testing procedure to cointegration and error correction models within the autoregressive distributive lag (ARDL) framework, using monthly data and estimating a wide range of Keynesian models of long-term interest rates. While previous studies have mainly relied on quarterly data, the use of monthly data substantially expands the number of observations. This in turn enables the calibration of a wide range of models to test various hypotheses. The short-term interest rate is the key determinant of the long-term interest rate, while the rate of core inflation and the pace of economic activity also influence the long-term interest rate. A rise in the ratio of the federal fiscal balance (government net lending/borrowing as a share of nominal GDP) lowers the long-term interest rate on Treasury securities. The short- and long-run effects of short-term interest rates, the rate of inflation, the pace of economic activity, and the fiscal balance ratio on the long-term interest rate are estimated. The findings reinforce Keynes’s prescient insights on the determinants of government bond yields.

KEYWORDS:

• Government bond yields
• long-term interest rates
• short-term interest rates
• monetary policy
• central bank
• John Maynard Keynes

JEL CLASSIFICATION:

• E43
• E50
• E58
• E60
• G10

Acknowledgments

The authors thank the participants of various seminars for their invaluable comments and suggestions. The authors also thank Ms. Elizabeth Dunn for her copyediting.

Disclaimer

The authors’ institutional affiliations are provided solely for identification purposes. Views expressed are solely those of the authors and are not necessarily those of Thrivent, Thrivent Asset Management, or any affiliates. This is for information purposes only and should not be construed as an offer to buy or sell any investment product or service.

Disclosure statement

No potential conflict of interest was reported by the authors.

The data set is available for replication

The data set used in the empirical part of this paper is available upon request to bona fide researchers for the replication and verification of the results.

Working paper version

Akram, Tanweer and Li, Huiqing. 2017. ‘An Inquiry Concerning Long-term US Interest Rates Using Monthly Data.’ Levy Economics Institute Working Paper No. 894 (August). http://www.levyinstitute.org/pubs/wp_894.pdf

Highlights

• The short-term interest rate is the key driver of the long-term interest rate.

• Core inflation and growth also influence the long-term interest rate.

• A rise in the fiscal balance ratio slightly lowers the long-term interest rate.

• The bounds testing procedure within the ARDL framework is applied.

• The empirical findings reinforce Keynes’s insights on interest rates.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

¹ The appendix of the working paper (Akram and Li 2017b) contains additional tables that reinforce and extend the findings presented in the paper.

² (The data in Figures 1-19 are from Macrobond [various years]).

³ The interested reader can see the mathematical models in Akram and Li (2017a).

⁴ The results of the unit root tests on the nominal yields of Treasury bills of a 6-month tenor are consistent with the nominal yields of Treasury bills of a 3-month tenor. The results of the unit root tests on the yields of Treasury securities of 2-, 5-, 7-, and 30-year tenors are consistent with the yields of Treasury securities of a 10-year tenor. The results of the unit root tests on the CPI inflation rate are consistent with the PCE inflation rate. Those results are provided in appendix tables A1 and A2 of the working paper and available upon request.

⁵ Introduced by Pesaran and Shin (1999) and Pesaran, Shin, and Smith (1999, 2001).

⁶ The rate of core CPI is used for the rate of inflation in the tables presented. Results are similar if the rate of core PCE inflation is used instead of the rate of core CPI.

⁷ See the tables in subsection 4.2 (‘Unit Roots Tests’) for further information.

⁸ Since the coefficients of the one-period lagged error correction term are not restricted, the ARDL is a UECM, also called an unconditional error correction model.

⁹ The 16 breakpoints for all models: 1967m9; 1968m8; 1971m1; 1975m11; 1976m2; 1980m12; 1981m4; 1982m10; 1985m5; 1985m8; 1995m6; 1995m7; 1998m8; 2004m2; 2005m2; and 2005m8.

¹⁰ Proposed by Shehata (2011).

¹¹ The recession data comes from NBER’s webpage for U.S. Business Cycle Expansions and Contractions, http://www.nber.org/cycles.html (accessed 20 March 2018). All six dummy variables are considered as exogenous variables during estimations.

¹² Five long-term interest rates are used in this paper: the yields of Treasury securities of 2-, 5-, 7-, 10-, and 30-year tenors. The key results using the yields of 10-year Treasury securities are discussed at length in the paper, while additional results are available in the appendix of the working paper (Akram and Li 2017b). These additional results are quite similar to those provided here.

¹³ Here the sample size is 662.

¹⁴ $\nu$is the unknown (k + 1)-vectors of intercepts of the data-generating process for $\{Z_t{\}}_{t=1}^{\mathrm{\infty }}$.

¹⁵ According to Pesaran, Shin, and Smith (1999), if the F-statistic falls above an upper critical value, the null hypothesis is rejected. The opposite is the case if the F-statistic falls below a lower critical value. If the F-statistic falls between the lower and upper critical values, the result is inconclusive.

¹⁶ Since the lagged error correction term is derived from the long-run cointegration relationship(s), the significance of the lagged error correction term(s) will indicate the long-run causal relationship.

¹⁷ In Table 6, the yields of Treasury securities of a 10-year tenor are used as the long-term interest rates. In appendix tables A3–A6 of the working paper (Akram and Li 2017b), the yields of Treasury securities of 2-, 5-, 7-, and 30-year tenors are used as the long-term interest rates.

¹⁸ The p-values for the heteroscedasticity test are less than 0.05 for models 1, 4, and 5. According to Shrestha and Chowdhury (2005, 25), ‘[s]ince the time series constituting the ARDL equation are potentially of mixed order of integration, i.e., I(0) and I(1), it is natural to detect heteroscedasticity.’

Additional information

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.