DOES MILITARY EXPENDITURE MATTER FOR INFLATION AND ECONOMIC GROWTH?

Abstract

This paper sets up a monetary endogenous growth model, and uses it to explain the ambiguous linkage between the military burden and the inflation rate observed in existing empirical studies. It is found that an expansion in the military burden has an ambiguous effect on the inflation rate depending upon the relative extent of two conflicting forces. More specifically, if the increase in the marginal benefit from holding money exceeds (falls short of) the increase in the marginal product of private capital, the inflation rate will rise (fall) in response. Moreover, it is found that an increase in the military burden will stimulate the balanced growth rate, confirming Benoit’s famous empirical findings.

Keywords:

• Endogenous growth
• Military expenditure
• Inflation
• Benoit hypothesis

Jel Codes:

• O42
• H56

ACKNOWLEDGEMENTS

The authors are grateful to Chi‐ting Chin, Chih‐hsing Liao, and two anonymous referees who provided us with many insightful comments and suggestions on an earlier version of this article. We alone are responsible for all the views and any remaining shortcomings herein.

Notes

¹ Shieh et al. (2002b) set up an endogenous growth model in which military expenditure is financed by a lump‐sum tax. Their results indicate that military expenditure and balanced growth are positively correlated. Under a similar framework in which defense spending is financed by income tax, Shieh et al. (2002a) find evidence of a non‐linear relationship between defense and economic growth.

² Benoit (1973, 1978), Brumm (1997) and Murdoch et al. (1997) find a positive linkage between defense spending and economic growth. Huang and Mintz (1990, 1991) reveal that there is no significant effect of defense spending on economic growth. Deger and Smith (1983) and Faini et al. (1984) point out the existence of a negative relationship between defense expenditure and growth.

³ For a more complete review of the Benoit hypothesis, please see Sandler and Hartley (1995) and Ram (1995).

⁴ In fact, Deger and Sen (1983, 1984), Chang et al. (1996), and Shieh et al. (2002a, b) have adopted defense capital as a proxy for national security.

⁵ This effect indicates that the defense sector will provide a production externality to the private sector through infrastructure, technology, training, education, and other human capital‐enhancing activities. See Sandler and Hartley (1995: 201–202) for a more detailed explanation.

⁶ Dunne et al. (2005: 456) argue that there seem to be strong theoretical and econometric reasons not to use the Feder–Ram model. This might explain why the Feder–Ram model is now rarely used outside the defense economics literature.

⁷ In order to have an ongoing growth rate, homogeneity of degree one (or constant returns to scale) in the growth variables (e.g. private and defense capital in this paper) of the production function is necessary.

⁸ See footnote 13 for a detailed discussion.

⁹ Equation (5a) states that the marginal benefit from consumption (1/C) equals the marginal cost of consumption in terms of utility (λ+ψ). Accordingly, (λ+ψ)/λ denotes the marginal cost (marginal benefit) from consumption in terms of money. Moreover, in order to hold a unit of money, the household will give up the nominal reward from lending the money to other people, namely, one plus the nominal interest rate (1 + π + (1 − α)A(D / K) ^α , where the marginal product of private capital (1 − α)A(D / K) ^α is equivalent to the real interest rate). According to the CIA constraint reported in equation (4), since consumption must be equal to money, it follows that the marginal benefit from consumption equals the marginal benefit from holding money. As a result, the condition that the marginal benefit from holding money equals the marginal cost of holding money can be expressed as:

By subtracting one from both sides of the equal sign in the above expression, it is easy to infer that the above expression is equivalent to the no‐arbitrage condition reported in equation (5b).

¹⁰ For simplicity, we assume that all government expenditures are devoted to weapon procurement purchases, and hence the government’s military expenditure leads to an accumulation of the defense capital stock. Our conclusions are valid if a specific fraction of the government expenditures is devoted to weapon procurement purchases.

¹¹ Although D is interpreted as defense capital in this study, it can be also interpreted as the stock of infrastructure. However, some may argue that there exists a major difference between the defense stock and the infrastructure stock: The defense stock is related to both the household’s utility and the firm’s production, while the infrastructure stock is related to the firm’s production only. In fact, some infrastructures such as a reservoir can be used for both tourism (entering the utility function) and for generating power (entering the production function). Under such a situation, the framework of this paper can be used to examine the effect of a rise in the infrastructure burden on inflation and economic growth.

¹² The ambiguous relationship between military expenditure and the inflation rate still holds if defense spending is financed by income tax. A mathematical proof is available from the authors upon request.

¹³ In their recent study, Aizenman and Glick (2006) specify that defense spending enhances private output in the production function, and find a negative impact of higher military expenditure on economic growth. Moreover, Shieh et al. (2002a) find an ambiguous relationship between the military burden and the balanced growth rate, even if a positive linkage between the defense burden and private output is established. The results of Aizenman and Glick (2006) and Shieh et al. (2002a) reveal the fact that the positive relationship between the defense burden and the balanced growth rate is not sensitive, so that defense capital has a positive impact on output in the production function.