In contrast to its decentralized political economy model of the 1980s, China took a surprising turn towards recentralization in the mid-1990s. Its fiscal centralization, starting with the 1994 tax reforms, is well known, but political recentralization also has been under way to control cadres directly at township and village levels. Little-noticed measures designed to tighten administrative and fiscal regulation began to be implemented during approximately the same period in the mid-1990s. Over time these measures have succeeded in hollowing out the power of village and township cadres. The increasing reach of the central state is the direct result of explicit state policies that have taken power over economic resources that were once under the control of village and township cadres. This article examines the broad shift towards recentralization by examining the fiscal and political consequences of these policies at the village and township levels. Evidence for this shift comes from new survey data on village-level investments, administrative regulation and fiscal oversight, as well as township-level fiscal revenues, expenditures, transfers (between counties and townships) and public-goods investments.

Recruiting and retaining leaders and public servants at the grass-roots level in developing countries creates a potential tension between providing sufficient returns to attract talent and limiting the scope for excessive rent-seeking behavior. In China, researchers have frequently argued that village cadres, who are the lowest level of administrators in rural areas, exploit personal political status for economic gain. Much existing research, however, compares the earnings of cadre and non-cadre households in rural China without controlling for unobserved dimensions of ability that are also correlated with success as entrepreneurs or in non-agricultural activities. The findings of this paper suggest a measurable return to cadre status, but the magnitudes are not large and provide only a modest incentive to participate in village-level public administration. The paper does not find evidence that households of village cadres earn significant rents from having a family member who is a cadre. Given the increasing return to non-agricultural employment since China’s economic reforms began, it is not surprising that the return to working as a village cadre has also increased over time. Returns to cadre-status (such as they are) are derived both from direct compensation and subsidies for cadres and indirectly through returns earned in offfarm employment from businesses and economic activities managed by villages

In this paper we have two objectives - one empirical; one methodological. Although China's leaders are beginning to pay attention to health care in rural China, there are still concerns about access to health services. To examine this issue, we use measure of travel distances to health services to examine the nature of coverage in Shaanxi Province, our case study. The mean distance by road to the nearest health center is more than 6 km. When we use thresholds for access of 5 and 10 km we find that more than 40 (15) percent of the population lives outside of these 5 (10) kilometer service areas for health centers. The nature of the access differs by geographical region and demographic composition of the household.. The methodological contribution of our paper originates from a key feature of our analysis in which we use Geographic Information System (GIS) network analysis methods to measure traveling distance along the road network. We compare these measures to straight-line distance measures. Road distances (produced by network analysis) produce measures (using means) that are nearly twice as great as straight-line distances. Moreover, the errors in the measures (that is, the difference between road distances and straight-line distances) are not random. Therefore, traditional econometric methodsof ameliorating the effects of measurements errors, such as instrument variables regression, will not produce consistent results when used with straight-line distances.

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica; min-height: 14.0px} p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; font: 8.0px Helvetica} span.s1 {font: 12.0px Helvetica} span.s2 {font: 9.0px Helvetica} span.s3 {font: 8.5px Helvetica} span.s4 {font: 7.5px Helvetica}

p.p1 {margin: 0.0px 0.0px 0.0px 0.0px; font: 12.0px Helvetica; min-height: 14.0px} p.p2 {margin: 0.0px 0.0px 0.0px 0.0px; font: 8.0px Helvetica} span.s1 {font: 12.0px Helvetica} span.s2 {font: 8.5px Helvetica} span.s3 {font: 9.0px Helvetica}