January 21, 2026
By Timothy J. Bartik, with the assistance of Kyle Huisman
Using individual data from the American Community Survey, we calculated average earnings per person for Michigan residents for four groups (two gender groups by two education groups) for each individual year of age from 18 to 79. The two education groups were 1) persons with an associate degree but no higher degree, and 2) persons with a high school diploma or GED, and possibly some college, but no associate or higher degree.
These data were obtained from the American Community Survey’s 2019–2023 microdata. The data were all adjusted to 2023 prices. Given that the information on costs for Futures for Frontliners (F4F) was roughly in 2023 dollars, we made no further price adjustment to another year’s prices.
We also obtained data on the gender mix of F4F (72 percent female, 28 percent male) and the age mix as of time of application in 2020. We assumed that any earnings differential would not begin until 2025, which would allow persons four years to attain an associate degree.
For the earnings differentials by year of age, we calculated for each age cohort (as of 2020) what earnings gain they would get starting in 2025. For 2025, it was assumed that the earnings differential for that age in that year was equal in 2023 dollars to the earnings differential calculated from the ACS. For years after 2025, we followed guidelines from the Trustees of the Social Security Administration, whose latest projections are that real earnings and productivity in the U.S. economy will grow at 1.13 percent per year (The Long-Range Economic Assumptions for the 2025 Trustees Report, U.S. Social Security Administration, 2025).
For each age cohort, we then used data from the U.S. Centers for Disease Control and Prevention for average survivorship by age and gender (United States Life Tables 2023, U.S. Centers for Disease Control and Prevention, 2025). For example, the percentage of males who were age 30 in 2025 who would still be expected to be alive in 2040 would be calculated as the ratio of the percentage of males surviving to age 45 to the percentage of males surviving to age 30. These survivorship proportions were then used to adjust downward the expected earnings differentials that would be received by a given gender/age cohort.
For the calculations of average individual gains to Michigan residents who earned an associate degree, for each gender/age cohort we then calculated the present value of these expected earnings gains, from the starting age to age 79, discounted back to the year 2023. The year 2023 was chosen as this is approximately when much of the cost of the program was incurred, on average. We used a real discount rate of 3 percent, which is common in benefit-cost analysis.
These gains for each gender/age cohort were then weighted by the estimated percentage of F4F frontliners in each group. The median age of applicants was about 31. This means that the real earnings gains from associate-degree attainment would begin at around age 36, which reduces the lifespan over which lifetime earnings differentials may accrue. These calculations yield estimated direct benefits in earnings gains of $218,702, in present value as of 2023 and in 2023 dollars. The present value of earnings gains was one-third higher on average for males ($266,741) versus females ($200,020).
For the indirect gains to other Michigan residents, and for tax calculations for state and local governments in Michigan, we then adjusted the data for those remaining in Michigan. For these calculations, we examined, by gender and age, the percentage of persons who were born in Michigan who still live in Michigan. This was calculated using 2019–2023 microdata from the American Community Survey. We then assumed, for each age/gender cohort, that the percentage of persons still living in Michigan as of some later age would be equal to the percentage of that age born in Michigan who still live in the state, divided by the percentage born in Michigan who still live in the state as of their age in the starting year, in 2025.
Thus, for those aged 32 in 2025, here is how we would determine how many would still be living in Michigan at age 55, in the year 2048: we take the estimated percentage of those born in Michigan who still live in Michigan at age 55, based on the 2019–2023 ACS, and divide it by the percentage of those born in Michigan who still live in Michigan at age 32, again based on the 2019–2023 ACS. Some of these raw ratios slightly exceeded one, so the ratios were adjusted so that the value of one was a maximum.
Such a procedure has previously been used, and it roughly matches the percentage staying in the state if one could implement the ideal practice of following the same age cohort over time. (See Bartik, Timothy J., 2009, What Proportion of Children Stay in the Same State as Adults, and How Does This Vary across Location and Groups? Upjohn Institute Working Paper No. 09-145. ) We then adjusted the present value obtained for all associate-degree attainers downward to count only the earnings gains for those who stayed in the state. This ends up being a figure of $202,731—only modestly less than the direct earnings benefits of $218,702. This modest differential reflects that for the age groups participating in F4F, such persons are likely to stay in the state for most of their prime earnings years.
To measure indirect gains, we assume, based on prior research, that such spillover benefits are equal to 86 percent of the direct earnings gains. (See Moretti, Enrico, 2004, “Estimating the Social Return to Higher Education: Evidence from Longitudinal and Repeated Cross-Sectional Data,” Journal of Econometrics 121(1–2): 175–212. ) The direct effect on Michigan associate-degree attainers who stay in Michigan, multiplied by 86 percent, yields a spillover benefit for other Michigan residents of $174,349.
To measure the likely state and local tax take for Michigan from extra associate degrees, we first add together the direct gain for degree attainers who stay in Michigan of $202,731 and the indirect gain of $174,349 to get $377,080. We then multiply this by an assumed marginal tax rate in Michigan. From the Institute for Taxation and Economic Policy, we used data on average state and local tax rates by income quintile in Michigan. (See Who Pays? A Distributional Analysis of the Tax System in All 50 States, 7th Edition, Institute for Taxation and Economic Policy, 2024.) However, these average tax rates are not quite right, because this associate-degree attainment produces a marginal change in income and moves people up within a particular income quintile or to a higher income quintile. From the ITEP data, we calculated the implied marginal tax rates within and across the various income quintiles. For our Michigan calculations, we used the implied marginal tax rate of extra income for those in the second-to-lowest income quintile. This marginal income tax rate is 10.0175 percent. The resulting calculation gives a present-value gain in Michigan total state and local taxes of $37,774.
Obviously, one could make many other assumptions about future earnings trends, survival rates, discount rates, percentage of program participants remaining in Michigan, spillover benefits, and marginal tax rates. However, it seems likely that most plausible estimates would have large dollar values per associate-degree recipient, and non-negligible tax-revenue effects.
In addition, as explained in the op-ed, these rough calculations do not necessarily reflect the benefit-cost ratio that should be calculated in a more rigorous analysis for Futures for Frontliners. One should adjust upward the F4F costs to reflect that not all associate degrees paid for by the program were actually causally induced by the program. On the other hand, one should adjust upward the benefits of F4F to reflect the earnings gains from certificates or extra courses completed due to the program, which should have some earnings effects. Finally, it could be that the earnings gains for F4F participants from associate-degree attainment may not exactly match the earnings gains for the average associate-degree holder. For example, if F4F participants are more disadvantaged than the average associate-degree holder, this would adjust downward the annual earnings they would get as an associate-degree holder. It might also mean that their annual earnings without such a degree would also be lower. Thus, the effect on the differential due to an associate degree is uncertain.