Disparities in the Level of Poverty in China: Evidence from China Family Panel Studies 2010

It should be noted that according to Gao (2012), health insurance in rural areas was worse than in the urban areas in 2000, but had improved by 2009.

Moreover, the coverage of the social insurance is sensitive to policy changes (e.g. the coverage of the health insurance increased sharply after the New Cooperative Medical Care System established).

Data constraint is the reason for not incorporating more dimensions.

See Jing and Mukhopadhaya (forthcoming) for an elaborate discussion. Here we discuss the importance and relevance of the selected dimensions/indicators in China context.

Among the total 13,339 sample households, 43.39 % households have at least one child. Among these households, 71.02 % households have at least one child in schooling age (6–15). If there is no child in the household (56.61 % cases) the household is not considered as poor. It should be noted that in the MPI, the education dimension includes two indicators, years of schooling and school attendance. Also, in MPI “all household members are considered deprived if any of their school-age children are not attending grades 1 to 8 of school and the households with no school-aged children are considered non-deprived. Hence incidence of deprivation in this indicator will reflect the demographic structure of the household and country as well as the educational attainments. Empirical studies suggest that this indicator provides different and complementary information to mean years of schooling (Santos et al. 2010). Furthermore, this indicator will be immediately sensitive to policy changes, whereas mean years of schooling will change more slowly.”(Alkire and Santos 2010, pp. 14). Our selection is similar to MPI. However, the adult literacy is better than the years of schooling to reflect the quality of education.

According to Wang and Alkire (2009), clean drinking water is considered to be tap water or well water that is from a depth of more than 5 metres underground. However, as with the definition in the MDG, safe water includes piped water, public tap, borehole or pump, protected well, protected spring or rainwater.

Food, clothing, housing, medical care and burial expenses are the five guaranteed forms of support.

This system was implemented to provide some assistance temporarily to the households or persons who get into the trap of basic living debt due to emergencies, accidents, serious illnesses or other special reasons but are not covered by the other social relief systems. In 2014 the State Council released the “Notice about establishment of the comprehensive temporary assistance system”. In the Notice the local governments (at the level of county and above) are responsible for the temporary assistance system. No further information on this is available.

H is obtained by dividing the number of poor households by total households, while A is calculated by adding the proportion of total weighted deprivation scores that each household suffers from and then dividing by the number of poor households. M ₀ is the product of H and A. Note that if the household is identified as poor by two cut-offs (the poverty line and number of predetermined indicators) then the deprivation score assigned is 1 otherwise 0. To compute A these scores are multiplied by respective weights of the indicators.

It is worth noting that the result in Jiangxi may be due to the variation in the sample size: while there are 176 rural households in the sample, the urban sample size is only 59; about 1/3 of the rural samples (see Table 1). Therefore, this result may be due to the sample size of urban households being too small.

Note that the numerical contribution of indicator/dimension may change with different cut-offs. However, the variation of cut-off will not change identified indicators that contribute most in multidimensional poverty. We are aware of the arbitrariness of the selection of dimension as a shortcomings of the multidimensional poverty measurement (See Ravallion 2011, 2012). That is why we have systematically chosen the dimensions and indicators using existing theories and conduct the robustness tests using various weights.

No previous study considers social security in the way we use it here. Some studies, however, use health insurance as social security (Yu 2013) and report that the deprivation in health insurance has declined since 2004.

Reform of the urban housing system in China started in 1998. After that, the work unit stopped allocating houses to workers. Instead, urban households had to buy houses from the market. Since the house prices have been rising for more than 10 years in urban China, many urban households could not afford to buy a house but had to rent. On the other hand, rural households can still build their houses on the homestead. From this perspective, urban households are more deprived.

There is no empirical study on the multidimensional poverty in China available that shows robustness by changing the weights. Only Guo and Wu (2012) checked the impact of the changing weights. However, they used one weighting system (normative weight). They first use the same indicators as MPI and then drop two indicators, school attendance and child mortality. Thus, the weight of the years of schooling and the nutrition changed from 1/6 to 1/3. They found that the results changed much.