Michigan law waiving safety net work requirement would have disparate impact on urban (blacks), favor rural (whites)

I'm in the process of writing a book chapter about how work requirements for SNAP and Medicaid--the legislative and administrative "flavor of the month" in Trump's America--are a mismatch for rural America, where labor markets are thin and the few jobs available are often mis-matches for local labor forces.  Another problem is that poor public transportation infrastructure makes it difficult for folks to get to jobs or the substitute training requirements imposed by some states.  This book chapter will follow up on a 2007 article I published about how welfare reform's work requirements were a poor fit in rural America.     

That's all background for why a colleague called my attention to this NY Times op-ed this week.  It it titled "Michigan's Discriminatory Work Requirement," and the authors are a pair of University of Michigan law professors, Nicholas Bagley and Eli Savit.  The core argument is reflected in this excerpt:

Last month, the [Michigan] State Senate passed a bill that would require Medicaid beneficiaries to find work or else lose their coverage. The bill, now under consideration in the Michigan House, has come under fire for harming the poor and disabled, as well as for burdening struggling families with needless paperwork. More than 100,000 people may lose health instance if it passes. 

There’s another flaw in the bill, however, one that exposes it to serious legal challenge: It’s racially discriminatory. 

Many of the legislators supporting Michigan’s work requirements come from rural districts with high unemployment. Many of those districts are predominantly white. To protect their constituents, these legislators have included a safety valve in the bill: If you live in a county with a high unemployment rate (over 8.5 percent), you’re exempt from the work requirements. The rationale? When there are no jobs to be had, it doesn’t make sense to punish you for not working.

Yet that safety valve does not apply equally. Specifically, it does little for Michigan’s black residents, who are concentrated in cities like Detroit, Muskegon and Flint. Those cities suffer from chronically high unemployment rates, but they’re all in counties with low rates.

The Washington Post followed up a few days later with this analysis from Jeff Stein and Andrew Van Dam. The headline blares: "Michigan's GOP has a Plan to Shield some People from Medicaid Work Requirements.  They're Overwhelmingly White." 

Medicaid enrollment data provided to The Post by the Michigan Department of Health and Human Services shows that this exemption would overwhelmingly benefit white people while leaving the work requirements in place for all but a sliver of the affected African American population.

Without the exemption, the work requirements are projected to apply primarily to approximately 700,000 Michigan residents enrolled in Medicaid under broader eligibility rules passed under Gov. Rick Snyder (R). 

African Americans make up about 23 percent of that population, but they would make up only 1.2 percent of the people eligible for the unemployment exemption. White people make up 57 percent of the total potential affected population, but they make up 85 percent of the group eligible for the unemployment exemption, according to an analysis of the state's data.

I read both pieces to imply that the proposed law's sponsors intended to discriminate against African Americans living in the state's major cities.  That is entirely possible, but another explanation is also possible:  the legislators were simply trying to do a good turn for rural folks facing crappy labor market opportunities.  Perhaps this is just an illustration that the scale of the county is a poor one for these purposes.  That is, the scale of the county makes some sense in relatively rural areas where the employment rate doesn't vary much across the county.  It's a poor proxy for the robustness of the labor market in a county that includes both core urban and suburban (and even exurban) areas.

Bagley and Savit do at one point acknowledge that discriminatory intent may be absent here, and they also helpfully explain that it doesn't matter under the relevant federal law, where the test is discriminatory impact.  The authors close with reference to bigger picture issues that implicate the ease (or lack thereof) of mobility to places with better job markets.  They also then highlight rural-urban differences and, as I read it, effectively and unhelpfully pit rural and urban against each other:

If work requirements were a good idea, conservative Michigan legislators wouldn’t need to exempt their rural constituents. They’d just offer a tough-love message: If you want health insurance on the public dime, you should move to a place where you can find work. 

That’s not the message, though. The message, instead, is that work requirements are good for people who live in hard-bitten cities and bad for those who live in hard-bitten counties.  

Given the lack of political power I perceive to be held by rural people and places, I'm surprised the Michigan legislators thought of them at all.