In response to a question raised by a scholar on the "hidden incomes" of numerous urban residents that went uncounted by statistical departments, the National Bureau of Statistics (NBS) published two articles on its website on Aug 24 and Aug 25, asserting that the report was "unreasonable" and has flaws in the methods of choosing survey samples and making calculations.
Wang Xiaolu, deputy director of the National Economic Research Institute under the China Reform Foundation, said in a survey report that the amount of "hidden incomes" of urban residents that was missed in the official statistics hit 9.3 trillion yuan (China's national income would amount to 23.3 trillion yuan if the allegedly omitted stats were included), of which, 5.4 trillion yuan is "gray income." This means an even lower proportion of labor remuneration to GDP and a wider wealth gap.
Shi Faqi from the Department of National Accounts of the NBS noted that the survey samples are too small in number and non-representative. The number of households sampled was 4,900, about 8 percent of the household survey carried out by the NBS. The calculated results based on such a small number of samples are not reliable, Shi said.
Wang Youjuan, chief of the Household Office of the Department of Urban and Social Economic Survey under the NBS, said that to secure the data on residents' incomes and a series of consumption structures, the inquirers first interviewed their relatives, friends or neighbors and then asked the respondents to persuade their relatives, friends or neighbors to participate in the surveys. The survey approach, known as the snowball sampling method in sociology, is subject to the influence of subjective factors.
Shi said in the article that apart from the irrationality in the quoting data, there are evident defects in the estimation methods adopted by the survey. When the survey projected the disposable income of Chinese residents, it applied the rural and urban population data collected at the end of a year, while the statistical departments generally use the average population data of a year. The annual disposable income of Chinese residents in 2008 that was estimated based on the average population of the year is 600 billion yuan lower than that estimated based on the population at the end of the year.
Wang said in the article that the Engel's coefficient, a major analysis method applied by the survey, should not be used as a main basis to project residents' income in the current China. The changes in China's current consumption structure are greater than those seen in developed countries' relatively stable consumption structure. Moreover, the change in the Engel's coefficient is associated with many factors, so it is not just affected by the single factor of income.