摘要
由于缺乏科学的预识别机制,已有研究并未甄别出参保人中谁会是延迟退休者。本文遵从参保人退休行为的异质性,将"退休"设为分类型潜变量,并使用2个显变量作为其测量指标,进而使用潜分类模型LCM对被访者进行识别。估计结果显示被访者的退休取向可分为非延迟退休、延迟退休2种不同类型,并分别解释了"退休"方差的58. 5%和41. 5%。观测值个体的贝叶斯后验概率结果表明,非延迟退休者占62. 04%,而愿意延迟退休的占37. 95%。这说明有多于1/3的职工愿意推迟退休。最后建议延迟退休应主要依靠发挥基本养老保险制度的经济激励作用。
Due to the absence of a scientific pre-identification mechanism,current studies have failed to identify who is likely to be a late retiree. According to the heterogeneity of the retirement behavior of the insured,this paper set "retirement"as a classified latent variable,and used two explicit variables as its measurement indicators,and then used the latent classification model LCM to identify the respondents. The results showed that the respondents’ retirement orientation could be divided into two types: non-delayed retirement and delayed retirement,and explained the variance of "retirement"of 58. 5% and 41. 5% respectively. The Bayesian posterior probability of the observed individuals showed that 62. 04% of them were non-delayed retirement,while 37. 95% were willing to delay retirement. This showed that more than one-third of employees were willing to postpone retirement,which was not "universally condemned"as the Internet survey said. Finally,it is suggested that postponing retirement should mainly rely on the economic incentive effect of the basic pension insurance system.
作者
阳义南
肖建华
YANG Yi-nan;XIAO Jian-hua
出处
《保险研究》
CSSCI
北大核心
2018年第11期107-116,共10页
Insurance Studies
基金
广东省自然科学基金项目(2016A030313209)
华南理工大学基本科研业务费重大项目培育(ZDPY05)
