Do “stationary Bandits” Promote Economic Growth? Evidence from Warlords in Early Twentieth Century China

Abstract: Do warlords, as substitutes for the state or “stationary bandits” in the sense of Mancur Olson, promote economic growth? This paper provides evidence on the warlords’ significant positive effect on China’s economic growth in the early twentieth century. By treating the sudden death of Yuan Shikai (1859-1916), “the Father of Warlords”, as a quasi-natural experiment, I find that firms in counties occupied by warlords (i.e., with at least one garrison there) grew faster than counties without such warlords during the succeeding Warlord Era (1916-1927). To exclude potential reverse causality I use the distribution of the New Army garrisons, the national military forces in the previous empire as an instrumental variable. The results of the instrumental variable estimation are consistent with the baseline results. The evidence supports the positive effect of warlords, both directly through investment and indirectly through the provision of public goods, reduction of the tax burden, and maintenance of social stability. These results are consistent with Mancur Olson’s argument that stationary bandit has an encompass interest to provide a peaceful order and other public goods in exchange for taxation. The heterogeneous effects of warlords are also examined. The empirical results suggest that the positive effect of warlords is weakened in those counties that (1) were more easily attacked by others, causing warlords to become short-sighted and less likely to promote growth, as Olson argued, and (2) had abundant mineral resources that could be easily exploited without the need to promote broader development, consistent with the resource curse argument.