Full Citation

Title: Point estimation of certain measures in organizational demography using variable-r methods

Citation Type: Journal Article

Publication Year: 2023

ISBN:

ISSN:

DOI: 10.4054/DEMRES.2023.49.33

NSFID:

PMCID:

PMID:

Abstract: Background: The distribution of employees’ eventual tenure plays an important role in economic sociology, labor economics, and social stratification. Objective: I aim to show how variable-r methods used for the point estimation of life table quantities for nonstable human populations can also be used to obtain expected employee tenures and related measures like conditional expected job tenure, the likelihood of a job ending via displacement, and the static effect of eliminating job displacements on expected employee tenure. Methods and Data: When birth rates are changing over time analysts can use variable-r methods to estimate period life tables with a count of intercensal births and two censuses. Once a single-decrement life table is obtained, researchers with data on specific decrements can construct multiple decrement life tables. I argue that the “population of employment relationships” is analogous to a human population. An employee tenure table, analogous to a life table, can be constructed using variable-r methods. Only two retrospective surveys of employee tenure lengths and a count of between-survey hires are required. Tenure-specific sources of decrement, e.g., job displacements, allow an analyst to construct multiple and associated single-decrement tenure tables isolating the effect of that decrement on the period employment tenure distribution. I illustrate and briefly evaluate the method using the 2002-2004 IPUMS Current Population Survey Job Tenure and Displaced Worker Supplements as retrospective surveys of employment tenure lengths and tenure-specific decrements, respectively. I obtain between-survey hiring counts from the 2002 – 2003 Job Openings and Labor Turnover Survey. Results: Variable-r methods generated reasonable expected employee tenure and job survival curves, replicating stylized features of the employee tenure distribution. Multiple decrement methods can be used to estimate the probability of an employment relationship ending via job displacement. Cause-deleted tenure tables can capture the static effect of eliminating a particular risk to the population of employment relationships. Conclusions: Demographers, economists, and sociologists should consider the use of variable-r methods when analyzing trends in expected employee tenure inequalities. Variable-r methods have limited data requirements and enable the estimation of interesting new measures in labor demography. Contribution: Arthur and Vaupel (1984) present an original derivation of the variable-r equations that I utilize in this article. Vaupel had an interest in formal demography throughout his life but started his academic career in business statistics. This paper combines those interests.

Url: https://www.demographic-research.org/articles/volume/49/33/

User Submitted?: No

Authors: Lachanski, Michael

Periodical (Full): Demographic Research

Issue: 33

Volume: 49

Pages: 865-904

Data Collections: IPUMS CPS

Topics: Labor Force and Occupational Structure

Countries: