Goodwill Excel Center
Introduction
Merola, Phillips, and Turner (2025) estimate the MVPF of enrolling into a tuition-free, public charter high school for adults. Goodwill of Central and Southern Indiana created the Excel Center in 2010 to help adult learners overcome barriers to high school graduation by providing an accelerated curriculum tailored to the individual, flexible class schedules, and non-academic support services like free on-site childcare, life coaches, and transportation assistance. The program incorporates professional certificates and community college credits into the traditional high school curriculum. Brough, Phillips, and Turner (2024) use linked administrative data on earnings to estimate the return to graduating by comparing the earnings of graduates before and after application to both students who enrolled in the Excel Center but exited before graduating and people who applied but did not enroll. The paper finds graduates earn substantially more—a roughly 40%—and are able to access higher-paying sectors than non-enrolled applicants in the years following application. Merola, Phillips, and Turner (2025) extend this analysis by linking applicants to administrative court case records and find that diploma earners see a 49% decline in charges in the year after applying, with effects lasting at least five years.
MVPF = 60.6
Net Cost
In 2014, the State of Indiana paid adult charter high schools a $6,600 payment per student per year. The average student in the data attends TEC for 1.08 years, and graduates attend the Excel Center longer (1.49 years) than exiters (0.96 years). While the upfront cost of enrolling an additional student is $7,111—$9,850 for graduates and $6,321 for exiters— the net cost of providing a high school education is lower as students pay additional taxes throughout their working career, particularly during their highest earning mid-career years, and graduates are less likely to interact with the criminal justice system. This analysis uses estimates of the earnings gains (and losses) from enrolling in and graduating from The Excel Center in the first 20 quarters after enrollment, as well as estimates of the reductions in criminal charges for students who complete a degree. In modeling the returns to earnings over the life cycle, the paper follows Hendren and Sprung-Keyser (2020) in assuming that the relative return in years 6–40 (i.e., the typical applicant’s remaining working career) is equal to the year-5 return, and model the evolution of earnings for the non-enrolled group using population average incomes of high school non-completers by age in the 2015 ACS. The paper discounts future costs and benefits with an annual discount rate of 3%. In estimating the impacts of the program on taxes and transfers, the paper uses CBO estimates linked to the 2016 federal poverty threshold for a household of two ($16,151). In quantifying the fiscal savings of reduced criminal justice interactions, the paper follows Bhatt et al. (2024) in assigning costs based on crime type to each guilty disposition in the court data.
Willingness to Pay
The paper takes the willingness to pay (WTP) for the Excel Center to be the change in after-tax earnings gains of Excel Center students that occur over the 5 to 40 years after application, as well as the reductions in the social cost of crime during the 5 years after application. The average enrollee experiences a $2,559 increase in the net present value of after-tax earnings within five years of application (see figure for quarterly effects on earnings during five years following application), and the paper conservatively estimates the reduction in the social cost of crime over that same period to be $11,152.
The paper focuses on a primary scenario in which earnings effects persist through retirement—an additional 35 years—and find enrolling in the Excel Center yields a willingness to pay estimate of $45,215—where $34,063 in benefits comes from the net present value of lifetime after-tax earnings and the additional $11,152 comes from reducing the social cost of crime. The paper follows Hendren and Sprung-Keyser (2020) in assuming that the relative return in years 6–40 (i.e., the typical applicant’s remaining working career) is equal to the year-5 return, and model the lifecycle evolution of earnings for the non-enrolled group using population average incomes of high school non-completers by age in the 2015 ACS. The paper discounts future earnings with an annual discount rate of 3%. In estimating the impacts of the program on taxes and transfers, the paper uses CBO estimates linked to the 2016 federal poverty threshold for a household of two ($16,151). In quantifying the social cost savings of reduced criminal justice interactions, the paper follows Bhatt et al. (2024) in assigning costs based on crime type to each guilty disposition in the court data, using a lower-bound estimate that does not scale up costs by reporting and arrest rates.
MVPF Summary
Dividing the willingness to pay by the net cost, the paper gets an MVPF estimate of $34,063 / $747 = 60.6.
This MVPF estimate combines the earnings effects of Excel Center enrollment reported by Brough, Phillips, and Turner (2024) with the criminal justice effects of Excel Center enrollment reported by Merola, Phillips and Turner (2025).
Unsurprisingly, the magnitude of the MVPF depends on the length of time that earnings effects persist. Because the program affects educational attainment, the paper assumes earnings gains persist through retirement. The estimate is sensitive to how the paper assumes effects persist into the future. For instance, if the year-5 earnings effect persists in levels, rather than as a proportion of the comparison group average earnings, then the estimated MVPF is 11.3 (95% CI of 5.5 to 23.0). Conversely, if the paper uses a more inclusive measure of the social cost savings from crime that scale up effects by reporting and arrest rates, the estimated MVPF is 180.3 (95% CI of 31.7 to ).
References
Bhatt, Monica P., Sara B. Heller, Max Kapustin, Marianne Bertrand, and Christopher Blattman (2024). “Predicting and Preventing Gun Violence: An Experimental Evaluation of READI Chicago,” The Quarterly Journal of Economics, 139(1), 1–56. DOI: https://doi.org/10.1093/qje/qjad031
Brough, Rebecca, David C. Phillips, and Patrick S. Turner (2024). “High Schools Tailored to Adults Can Help Them Complete a Traditional Diploma and Excel in the Labor Market.” American Economic Journal: Economic Policy, 16(4): 34-67. DOI: https://doi.org/10.1257/pol.20230053
Hendren, Nathaniel and Ben Sprung-Keyser (2020). “A Unified Welfare Analysis of Government Policies.” The Quarterly Journal of Economics, 135(3): 1209–1318. DOI: https://doi.org/10.1093/qje/qjaa006
Merola, Emily, David C. Phillips, and Patrick S. Turner (2025). “Reversing the School to Prison Pipeline: The Impact of an Adult High School Program,” IZA Discussion Paper No. 18168, October 2025. https://www.iza.org/publications/dp/18168/reversing-the-school-to-prison-pipeline-the-impact-of-an-adult-high-school-program
Policy Features
- Category
- Education
- Sub-Category
- Job Training
- Beneficiary Type(s)
- Adults
- Average Age
- 25
- Average Income
- 5191
- Country of Implementation
- United States
- Year of Implementation
- 2013
- Empirical Method
- Difference in Differences
- Research Type
- Primary
- Peer Reviewed
- No