关键词: 主观预期寿命, 最优消费决策, 非寿险需求, HJB方程, 寇尔法

Abstract: The deepening of population ageing has led to unprecedented perceived pressure on life expectancy and the need for wealth accumulation in households. Consumption decisions that match life expectancy and demand for non-life insurance are an important basis for households to optimise their long-term goals, balancing utility enhancement and risk management. As a result, it is more urgent than ever for residents to make efficient and rational consumer and non-life insurance decisions. Prior to the rational revolution expectations, the life-cycle-durable income hypothesis model was the main theoretical framework for studying the consumption behaviour of the population. For a long time, scholars have been trying to be closer to the reality of the population, gradually introducing term uncertainty and mortality ambiguity into the consumption decision model and non-life insurance demand model, on the basis of which they analyse the optimal consumption strategy and life insurance demand strategy.
   Based on the 2020 China Health and Retirement Longitudinal Study (CHARLS) and the 2020 China Population and Employment Statistical Yearbook, this paper uses the Kohl’s method to portray and measure the subjective life expectancy of China’s residents. The subjective life expectancy of the population is introduced into the consumption and non-life insurance demand models based on the fuzzy aversion to mortality, and the analytical solutions of the robust optimal consumption and non-life insurance demand strategies are derived through the dynamic programming principle, Girsanov’s theorem and the HJB equation. Matlab numerical simulations are applied to examine the extent to which residents’ subjective life expectancy affects residents’ consumption decisions and demand for non-life insurance, as well as the level of change in optimal decisions due to changes in important economic and social parameters. The aim of this paper is to obtain consumption and non-life insurance demand strategies that are more relevant to the reality of the population and provide a more robust wealth preparation for the population to cope with the ageing of the population.
   The analysis leads to the following conclusion: our middle-aged population underestimates subjective life expectancy. The more pessimistic the subjective life expectancy, the higher the level of consumption of the population, which is strongly characterised by “just-in-time” and “consumption distortion”. However, subjective life expectancy does not have a significant impact on residents’ non-life insurance decisions, which is related to the fact that non-life insurance is a short-term insurance product.
   In terms of the optimal consumption of the population, an increase in the risk-free interest rate, an increase in the objective mortality rate of the life table and a deepening of risk aversion significantly increase the level of consumption of the population. In terms of demand for non-life insurance, an increase in the maximum loss ratio ν, the additional premium factor θ, and the average number of insurances taken out by the population κ increase the population’s demand for non-life insurance.
   Important non-life insurance parameters have a “crowding out” effect on residents’ optimal consumption. Increases in the maximum loss ratio ν, the additional premium factor θ, and the average number of insurances taken out by the population κ dampen the population’s demand for non-life insurance. There is a substitution effect as the ratio of consumption and non-life insurance expenditures are inversely proportional to each other.

Key words: subjective life expectancy, optimal consumption decisions, non-life insurance demand, HJB equation, Kohl’s method