Chu C Y, Lu H
J Popul Econ. 1995;8(1):35-57. doi: 10.1007/BF00172037.
Easterlin believed that there were two features associated with the birth cycles he observed: the cycles were related to the labor market, and they might be self-generating. This paper tries to set up a model that contains both of these two features. The authors suppose that the welfare of various age-specific cohorts are determined by their respective marginal productivity, and that the underlying technology which puts together labor force of various age-specific cohorts can be characterized by a general production function. Under these weak assumptions, they show that the well-analyzed cohort and period models along the lines of Lee (1974) are restricted versions of the general setting. Given that both the cohort model and the period model were rejected by statistical tests, the authors adopt the coefficient values obtained from the estimation of the unrestricted version to perform the bifurcation analysis. They go beyond the previous study which focused upon the possible existence of limit cycles, and show that the US fertility limit cycle solution is unstable. Therefore the population trajectory will never converge to that limit cycle.
伊斯特林认为，他所观察到的生育周期有两个特点：这些周期与劳动力市场有关，而且可能是自我生成的。本文试图建立一个包含这两个特点的模型。作者假设，不同年龄组人群的福利由其各自的边际生产力决定，并且将不同年龄组人群的劳动力组合在一起的基础技术可以用一个一般生产函数来描述。在这些较弱的假设下，他们表明，按照李（1974）的思路进行充分分析的队列模型和时期模型是一般框架的受限版本。鉴于队列模型和时期模型都被统计检验所否定，作者采用从无限制版本估计中获得的系数值来进行分叉分析。他们超越了之前专注于极限环可能存在性的研究，表明美国生育率极限环解是不稳定的。因此，人口轨迹永远不会收敛到那个极限环。
