Thank you for writing this. I am big fan of your work on work requirements. Perhaps a larger distortion than their treatment of work requirements is their treatment of lost Medicaid benefits in poverty measurement. They acknowledge that they do not attempt to estimate the increased out-of-pocket spending on premiums or care (cost sharing) for those who lose Medicaid as a result of work requirements. Those OOP expenses are subtractions from resources in SPM poverty measurment and hence would be poverty-increasing. Beyond that, Medicaid (and other health insurance benefits) are valuable resources to families that help them meet the basic need for health care, which can be just as important a need as food, clothing and shelter. Two NAS studies have recommended the use of a Health Inclusive Poverty Measure to account for the financial protection and access to care that heatlh insurance benefits provide low-income households e.g., (https://www.nationalacademies.org/read/26825) and the Census Bureau produces a HIPM as a research series and makes data files avaialable (https://www.census.gov/library/working-papers/2025/demo/sehsd-wp2025-14.html). Estimates using the HIPM show major impacts of Medicaid on poverty, e.g., (https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2017.0331). Bottom line: it is easy to show Medicaid work requirements reduce poverty substantially if you assume 80% to 100% of affected people reach the average earnings of lower-income persons, and if you ignore the financial protection and direct resource value of Medicaid for low-income households who lose insurance coverage as a result.
I completely agree. I had a discussion of that issue, but for length and to avoid getting too in the weeds, we decided to leave out. Thank you for adding this important point to the discussion!
Thank you for writing this. I am big fan of your work on work requirements. Perhaps a larger distortion than their treatment of work requirements is their treatment of lost Medicaid benefits in poverty measurement. They acknowledge that they do not attempt to estimate the increased out-of-pocket spending on premiums or care (cost sharing) for those who lose Medicaid as a result of work requirements. Those OOP expenses are subtractions from resources in SPM poverty measurment and hence would be poverty-increasing. Beyond that, Medicaid (and other health insurance benefits) are valuable resources to families that help them meet the basic need for health care, which can be just as important a need as food, clothing and shelter. Two NAS studies have recommended the use of a Health Inclusive Poverty Measure to account for the financial protection and access to care that heatlh insurance benefits provide low-income households e.g., (https://www.nationalacademies.org/read/26825) and the Census Bureau produces a HIPM as a research series and makes data files avaialable (https://www.census.gov/library/working-papers/2025/demo/sehsd-wp2025-14.html). Estimates using the HIPM show major impacts of Medicaid on poverty, e.g., (https://www.healthaffairs.org/doi/full/10.1377/hlthaff.2017.0331). Bottom line: it is easy to show Medicaid work requirements reduce poverty substantially if you assume 80% to 100% of affected people reach the average earnings of lower-income persons, and if you ignore the financial protection and direct resource value of Medicaid for low-income households who lose insurance coverage as a result.
I completely agree. I had a discussion of that issue, but for length and to avoid getting too in the weeds, we decided to leave out. Thank you for adding this important point to the discussion!
thanks for sharing this - its a target rich environment!