Nato burden sharing revisited

Abstract

This paper revisits the NATO burden sharing debate with particular emphasis on the use of non-parametric tests that measure the equality of relative benefits and burdens. While the joint product model makes a compelling case for the equality of benefits and burdens in NATO after 1967 (as a consequence of a change in doctrine) the empirical validation of the model is not robust when using non-parametric methods. The lack of knowledge of each ally's utility function, as well as the fact that an ally may have different benefits other than the protection of its industrial base, exposed border and population to consider render the test less robust, as shown in this study. Sensitivity analysis such as the one employed in this paper should be a standard procedure when devising new proxy measures and tests for burden sharing in NATO.Burden sharing, Joint product model, NATO, Public goods,

Nato Benefits, Burdens and Borders: Reply

“…Analysis of the variation of the weights assigned to the benefit shares (bottom six rows of Table I) show that an increase in the weight of the exposed border proxy supports the original Sandler et al study, while reduction of the weights supports the Solomon (2004) findings. In particular, the reduction of the exposed border proxy from its original weight of about 30% to 20% resulted in the inclusion of the years 1965 and 1999 in the burden-benefit concordance rejection column while a 10% weight on the exposed border proxy led to the rejection of the null hypothesis (burden and benefits are equal) for all selected sample years except 1980.…”

“…The Sandler et al findings critically depend on the exposed border proxy as implied in Solomon (2004). (Khanna and Sandler, 1996;Sandler and Murdoch, 2000), Without Exposed Border (Solomon, 2004), Adjusted Exposed Border (same as original except North America western coastline only, Norway's northern border only). Weight (40/60 implies 60% to exposed border, etc) ** Significant at the 5% level.…”

“…In critiquing Solomon (2004), Sandler (this issue) suggested a fairer sensitivity analysis including varying the weights of the benefits measures, re-defining the exposed borders with an emphasis on Canadian western and Norway's northern borders and excluding outliers. The purpose of this reply is primarily to implement the suggested sensitivity measures and assess the impact on Solomon's (2004) findings.…”

“…The purpose of this reply is primarily to implement the suggested sensitivity measures and assess the impact on Solomon's (2004) findings. This reply also briefly touches on other less technical aspects of Sandler's comments.…”

The growth in military expenditure in Germany 1951–2011: did parties matter?

“…Holding off forces in military conflicts notwithstanding, Germany did not free-ride on NATO burdens (see, for example, Olson and Zeckhauser 1966;Sandler and Forbes 1980;Solomon 2004;Sandler and Shimizu 2014).…”