The Significance of Variation in Local States: The Case of Twentieth Century China

Abstract

When I was conducting fieldwork on post-Mao taxation and public finance in China during the early 1990s, I was surprised to discover that county officials in Tianjin in the north were still using Maoist mass campaign tactics to deter tax evasion. I was even more surprised to discover that county officials in Guangdong in the south said they had never heard of such a thing, were appalled by it, and relied instead only on persuasion and nonconfrontational methods. In Tianjin county tax officials dragged convicted tax evaders onto a stage and detailed their crimes to an audience of a thousand fellow entrepreneurs and anyone else who cared to watch the criticism session on local television. Meanwhile, in Guangdong entrepreneurs attended small study sessions where they read about a few egregious cases of tax evasion. The two localities' methods of countering tax evasion illustrated that officials were dealing with social groups and with resistance in different ways. How could two such different approaches to the same problem of tax evasion exist side by side in the same governmental system, especially one that is usually assumed to have a powerful, homogenizing central government? In fact, such differences in local state practice are common across many policy areas, not simply taxation, in China. What are the sources of this variation in local state structure and practice, and what are its overall consequences for the implementation of central policy, for state-society relations, for state capacity, and even for the survival of the central state? Variation in local state structure and practice-local state variation, for short-is a neglected topic in political science. Local state structure is defined as the organization of local government, and local state practice is defined as the way local governments implement central policy and make and implement their own policy in the absence of higher-level guidance. In political science, local state structure and practice have until recently been discussed only as the objects of central government standardization efforts. For example, the classic studies of the formation of central states examine why and how patterns of intrastate difference were suppressed or erased rather than how they developed.1 The standard narrative of state formation emphasizes processes of bureaucratization, centralization, and homogenization that eliminated local differences of many kinds.2 This approach loses sight of the ways