The main problem with China s pension system, the inability of paying the pensions for more and more old people, had been supposed to be dealt with by China s pension reform. But can the reform actually reach its goal? The existing literatures had already simulated the proposed reform arrangements and policy options under certainty.
But the effect of the pension reform in China by its nature is uncertain. The studies on its uncertainty will be more helpful on policy making. Employing a simple but complete model that had been used in simulating the certain results of China s pension reform, this paper presents a method to analyze the uncertainty of the effect of pension reform, to simulate and to compare the potential policy options.
The uncertainty can come from two main sources: the demographic dynamic and economic development. Demographic dynamics depends on fertility, mortality and migration, while economic development depends on future unemployment, interest rate, GDP growth, inflation, real wage, etc. The uncertainties of all these variables lead to the uncertainty of the pension system. But all these factors are unknown for us.
This paper gives the method to model the stochastic process of these random variables in term of Markov chain and introduce these variables uncertainties into the simulation model. The simulation can be done with Monte Carlo fashion. Here the method is illustrated by analysis of the uncertainty that comes from GDP growth and inflation, which both cooperate affecting the real wage. In the simple simulation model, the real wage is the key to simulate the uncertainty of the pension system variables and of the economic development. With the stochastic process description of the economic inputs, and the relationship between the economic inputs and the pension system arrangements, the uncertainty of China s pension reform can be simulated. The same work can be done to the demographic dynamic inputs, too. The suggestions on simulating the potential policy options are also given at the end of this paper.
The method is presented by using the simple model, but it is not supposed to be used only in this model. Uncertainty exists in the nature of pension system. So it is possible and necessary to apply the uncertainty analysis to all the estimates about the future of China s pension system. We can use this method to all the variables mentioned here as well as other models about pension. There are a lot of works to be done.