An Econometric Contribution to the U.S. Defense–Growth Nexus: Evidence from Error Correction Model

Abstract

Past and contemporary research has examined the relationship between military spending and economic growth and has reported mixed results. These conflicting findings are perhaps due to the methodologies employed. Consequently, this study employs a simple, parsimonious New Growth model to investigate the link between military spending and economic growth in the United States over the period 1/1959–1/2001 by adopting a more robust estimation technique. It follows the Johansen co-integration and error correction methodology coupled with vector autoregression (VAR) and innovation accounting techniques. Findings are robustly substantiated and reveal that military spending and growth have neither a statistical nor an economic impact on each other. This suggests that current U.S. political debates opposing or favoring military spending on the grounds of its economic merit are irrelevant.

Keywords:

• co-integration
• error correction model
• innovation accounting
• military spending
• economic growth

Without any implication, I would like to thank Dr. Glenn Palmer, Dr. Hannarong Shamsub, and four anonymous referees for their valuable comments. The usual disclaimer applies.

Notes

¹ In this paper, “defense spending” is used interchangeably with “military spending” and is represented by military spending as a percentage of total government spending. Similarly, “economic growth” is used interchangeably with “growth” and is measured by the growth rate in real per capita gross domestic product (GDP).

² The Johansen co-integration and error correction methodology could provide robust results once it is followed by a VAR analysis with innovative accounting techniques (i.e., impulse–response functions, variance decompositions, and Granger-causality tests). For further discussion of the Johansen methodology, see Ender (1995), Applied Econometric Time Series.

³ Barro and Sala-i-Martin (1995) examined the link between defense spending and growth in a cross-country regression. They argued that, theoretically, defense spending is expected to correlate with long-term growth rates. For further details on the role of defense spending within the new growth framework, see Barro and Sala-i-Martin (1995).

⁴ This study chooses to analyze the period 1959–2001 based on the data available to the author, as well as the standardized methodology used in measuring military spending beginning in the early 1950s.

⁵ For further discussion of the Johansen co-integration and error correction methodology, see Enders (1995).

∗ Indicates rejection of the null hypothesis of nonstationarity at less than 5% level of significance.

∗ Indicates 1% level of significance.