Tables of Inter-year Labor Force Status of the U.S. Population (1998–2004) to Operate the Markov Model of Worklife Expectancy

In tort litigation, plaintiffs often claim multi-year earnings losses and forensic economists are called on to calculate those losses at their expected discounted value. To make those calculations, forensic economists must determine the probabilities that plaintiffs would be able to or would choose to work and have earnings—referred to in the forensic economic literature as the determination of worklife expectancy. In this paper, under the Markov increment-decrement demographic model, we provide the theory, estimating equations, and data necessary to compute worklife probabilities for the U.S. population of males and females for five levels of educational attainment. The particular focus of the paper extends the worklife expectancy literature to the data tables necessary to produce age-based labor force participation probability estimates under the popular Markov increment-decrement model. To date, the Markov-based literature has presented its empirical estimates of worklife expectancy as the sums of unpublished age-based labor force participation probabilities. That literature shows worklife expectancy computed beginning at each age in a table through a common terminal age where worklife is assumed to end for all persons. For example, the worklife expectancies are estimated from ages 18 to 80, ages 19 to 80, ages 20 to 80,..., ages 78 to 80, and ages 79 to 80 where at age 80 all living persons are assumed to not be in the labor force. When worklife expectancy is used in forensic economics as the sum of age-based probabilities of labor force participation, at least two serious problems are created: (1) the problem of “front-loading”1 and (2) the problem of mismatching labor force participation probability estimates with estimates of earning capability. For example, assume that an estimated worklife expectancy sum is 30 years utilizing labor force data from persons ages 25 to 80. For each age, the Markov model produces a probability of labor force participation within the range of 0 < wp < 1 and the sum of those fractional year probabilities from age 25 to age 80 equals 30 whole years. Since the user of such worklife expectancy data has no information other than labor force participation will occur for a total of 30 years sometime between age 25 and age 80, he or she usually front-loads all labor force attachment years to the first 30 years beyond