South Dakota is in the process of applying for a Section 1115 waiver that will allow it to impose a work requirement on very poor parents in two counties, Minnehaha and Pennington, for five years, with the option to expand to other counties.  The state Medicaid agency posted the application for public comment on May 21; the comment period closes on June 19.

There are lots of problems with this application, but here’s the biggest:  it does not give the public the single most critical piece of information it needs to comment.  How many parents in these two counties does the Medicaid agency estimate will lose coverage as a result of the work requirements over the next five years?  Without this estimate, it is impossible for beneficiaries, providers, Tribes, and other stakeholders to understand what the agency thinks will happen to the parents targeted by the waiver.

There’s no doubt that work requirements will result in the loss of coverage, especially in states like South Dakota, which have not expanded Medicaid to cover all adults with incomes below 138% of the Federal Poverty Level.  The only question is how many parents will lose coverage, and for how long. The state seems to imply that itself.

As we have shown in other non-expansion states— Alabama, Mississippi, and South Carolina—work requirements will harm the most vulnerable children and families, those with incomes at 18%, 27%, 67%, or, in South Dakota’s case, 50% of the Federal Poverty Level.  In particular, very poor parents will lose coverage as their eligibility for Medicaid becomes contingent on whether or not they meet the new work requirements or simply get caught up in a paperwork trap.  That will put their children’s Medicaid coverage at risk as well.

So how is it possible that South Dakota’s application does not estimate the coverage loss that would result from the waiver?  It simply asserts that it is impossible to predict what the impact on enrollment will be and therefore doesn’t even try.

Here is how the state puts it:

At this time it is not clear how many individuals will increase their income above the parent and other caretaker relatives’ income limit or choose to not participate in the Career Connector program; as such the budget neutrality document does not reflect a potential change in member month due to the demonstration. Any decreases in annual enrollment would likely also result in decreases in annual expenditures. (p. 9)

As a result, when you examine the budget neutrality assumptions in the waiver application, enrollment (expressed as number of eligible member months) would be exactly the same in each of the five years of the waiver as it would be without the waiver.  For example, in FFY 23, the last year of the waiver, the application projects 144,822 member months for the Low-income Families (LIF) population without the waiver and 144,822 with the waiver.  Projected expenditures for the LIF population that year will be exactly the same—to the last dollar—under the waiver and without the waiver ($120,788,755).

These data are found in the Budget Neutrality Worksheet in Appendix 2 of the South Dakota application.  These worksheets are required in all Section 1115 waiver applications so that the federal government, in deciding whether to approve the application, can assure itself that the waiver will be budget neutral—i.e., that the state will not spend more federal Medicaid dollars under the waiver than it would without it.

These data make no sense.  Alabama projects a coverage loss of 17,300 under its waiver in the fifth year alone.  Mississippi estimates that 20,000 will lose Medicaid coverage under its waiver over five years.  In both cases, these estimates can be derived from the state’s Budget Neutrality Worksheet.  (South Carolina’s has not yet put its application out for public comment and the concept paper that we reviewed didn’t yet have budget neutrality assumptions).   These estimates from other non-expansion states may understate the coverage losses, but at least they represent a starting point for stakeholders and the public to understand the implications of the waiver application.  South Dakota’s application makes any understanding of the waiver impossible, effectively gutting the 30-day public comment period.

Federal regulations are unambiguous on this point.  A state must give the public notice of its application, and the notice must contain “a sufficient level of detail to ensure meaningful input from the public, including…An estimate of the expected increase or decrease in annual enrollment…” 42 CFR 431.408(a)(1)(i)(C).  Similarly, the application itself must include “An estimate of the expected increase or decrease in annual enrollment…” as well as “enrollment projections expected over the term of the demonstration for each category of beneficiary whose health care coverage is impacted by the demonstration.”  42 CFR 431.412(a)(1)(iii), (iv).

The state Medicaid agency needs to go back to the drawing board.  It should withdraw the current waiver application, develop a plausible Budget Neutrality Worksheet that complies with the federal regulations, and post the waiver application with the corrected Worksheet for another 30 days of public comment.  Only then can the citizens of South Dakota have an informed conversation with their policymakers about the potential impact of work requirements on Medicaid coverage for very poor parents and their children.