earticle

영어

Replacement rate is the key parameter for social pension systems. To derive more accurate replacement rates, this paper modifies Lluch's (1973) ELES (Extended Linear Expenditure System) model through the consideration of two additional variables that have been overlooked in the previous studies. Those variables considered here are the differences of living expenditures among various income groups and the yearly changes in the price index. Instead of just one replacement rate for the entire Chinese urban workers currently used by the Chinese social pension insurer, this paper derives several replacement rates according to their income levels. The results of the study suggest the followings: For low income group, the marginal propensity to consume is relatively high and their expenditures mostly concentrate on the basic living goods before and after their retirement. Therefore, for this group, replacement rate should maintain higher level to keep their basic living standards. On the contrary, for high income group, the marginal propensity to consume is relatively low while their propensity to save is higher. The replacement rate does not need to be so high. Before retirement, high income people spend most of their income on housing, durable goods and relatively low proportion on the basic living needs. Even after their retirement, they can usually afford to non-essential expenditures through their private savings. And for the medium income group, both the propensity to consume and propensity to save are in the medium level between low income and high income, close to the social average level. For them, the replacement rate could be little higher than the lowest living standard, and close to the mean level.