Staff Growth in International Organizations: A Principal-Agent Problem? An Empirical Analysis

Abstract

The analysis covers 27 international organizations in the years 1950-2001. From the first to the last year, staff increased at a compound average rate of 3.2 percent per annum. Since the number of member states rose by only 2.5 percent, the elasticity of staff to member states is larger than one (e 1 = 1.28). As this may be due to an expansion of tasks, we estimate time-series regressions and panel-data regressions which contain output proxies or task dummies wherever possible. The pooled analysis of 817 observations reveals that (i) the elasticity of staff to membership is much larger than unity (1.36) if, and only if, the non-stationary component of staff size is not removed, (ii) United Nations organizations have significantly more staff, (iii) international organizations in the United States and Switzerland have significantly less staff, (iv) heterogeneity in terms of per capita income limits the size of an international organization and that (v) its staff is larger if its membership comprises many industrial or communist countries. In a reduced sample, the financing share of the largest contributor in combination with the party or programmatic orientation of its government has a significantly negative effect on staff because the size of the largest financing share determines the incentive to monitor. U.S. exit from an international organization reduces its staff significantly. Most of these results depend on the condition that the nonstationary component of staff size is not taken account of by time dummies or trends.