The president retained most of these voters in 2012 despite the poor economy. Journalist Ron Brownstein calculated that Obama carried whites without college degrees again in Iowa and ran even with Romney among them in New Hampshire. He carried 45 percent of them in Wisconsin and 42 percent in Ohio, high enough in each state to allow his urban coalition of blacks and liberal, well-educated whites to carry him to victory.

Data compiled by political scientist Larry Bartels show that the president’s support among non-college-educated whites came largely from the working and lower-middle classes, exactly the people most affected by the Great Recession and the wage pressures of the preceding decade. Bartels’s analysis shows that non-Southern whites making less than about $45,000 a year voted for the president in 2012, and those making between $45,000 and $62,500 gave him more than 45 percent of their votes. Conservative analysts may decry the dependency safety-net programs engender, but clearly those voters who have struggled for decades were not so concerned.

These facts have not gone unnoticed by these states’ Republican governors. Every GOP governor in these states except Wisconsin’s Scott Walker has either expanded Medicaid or endorsed expansion, and even Walker reformed his state’s generous Medicaid program to expand insurance coverage by nearly one hundred thousand people without taking new federal money. Ohio governor John Kasich was so intent on expanding Medicaid prior to his 2014 reelection vote that he used his appointees to an obscure state board to get around opposition from Republicans in his state legislature to expand the program. The nation’s two Hispanic Republican governors, Nevada’s Brian Sandoval and New Mexico’s Susana Martinez, also took leading roles in establishing their state’s exchanges pursuant to Obamacare. Republican governors in Michigan and Pennsylvania, two other potential swing states that lean slightly Democratic, also endorsed Medicaid expansion.

INDIANA REPUBLICAN governor Mike Pence’s 2014 announcement that he, too, would seek to expand Medicaid further shows how strong the political pressure is. Pence is so conservative that as a congressman he was one of just twenty-one GOP stalwarts not to vote for President Bush’s bill creating prescription-drug coverage for seniors. Yet Indiana is another state politically dominated by whites without college degrees. This group comprised 56 percent of the 2008 electorate, the highest percentage of any midwestern state. While this group also has a significantly larger evangelical Christian portion than elsewhere in the Midwest, Pence has surely recognized that the same economic pressures that make Medicaid expansion tempting for low-income whites elsewhere apply to Hoosiers as well.

Pence’s proposal is slightly different, and as such offers a lens through which one can view the options conservatives have in crafting an original response to these pressures. The largely hostile reception his idea has met among conservative intellectuals, however, shows that the “just say no” impetus remains strong on the right.

Pence’s idea builds on a program serving a part of the Medicaid population started by his predecessor, Mitch Daniels. The Daniels plan, called the Healthy Indiana Program (HIP), created Health Savings Accounts (HSAs) for forty thousand recipients. The state funded part of the recipient’s HSA, so long as the recipient also contributed. The state used its own funds for the HSA financing, although it continued to rely on a mix of state and federal funds for the underlying basic health-insurance coverage.

Pence’s idea would take this concept and expand it to over 330,000 able-bodied adults earning between 100 and 138 percent of the federal poverty limit. This is also the limit for expanded Medicaid eligibility under Obamacare, and Pence’s plan would use the money available for Obamacare expansion to pay for the HIP expansion. It would change the HIP plan significantly, lowering the amount of money recipients would be asked to pay into their HSAs and dropping any requirement they pay completely (those who would not contribute would get a less comprehensive plan). It would, however, add a premium-support component so that Hoosiers eligible for the new program who also could take employer coverage could use HSA dollars to pay their share of the premium in the employer-offered program.

Conservative opposition to Pence’s plan covers a host of issues, but essentially boils down to two: cost and dependency. The Heritage Foundation’s Nina Owcharenko focused on the first complaint in her attack, arguing that even if the state could, as it says, fully fund the expansion through 2020, “original estimates are likely not the true full cost.” Furthermore, since federal promises to pay a high portion of the expansion could be changed in the future, Indiana taxpayers could be legally stuck with paying a higher share of a much more costly program.

Owcharenko’s cost argument has been essentially echoed by many other critics, such as the Mercatus Center’s Veronique de Rugy. Unlike Owcharenko, however, de Rugy raises the issue of increasing dependency. Reducing dependence on government is increasingly cited by many conservatives as justification for opposing Obamacare, increased food-stamp usage, and a host of other safety-net or entitlement programs. This term is rarely clearly defined (the Heritage Foundation’s annual Index of Dependence on Government, which it published for over a decade before discontinuing it in 2014, included Social Security and Medicare as dependency-inducing programs, a claim virtually no conservative politician will seriously echo). It does, however, clearly include the receipt of government checks or in-kind services such as health insurance by able-bodied people of working age. The most vociferous conservative opposition to Pence’s proposal and, indeed, to Medicaid expansion generally, comes from this idea.

This is best elucidated by a blistering critique penned by the Foundation for Government Accountability’s Josh Archambault and his colleagues. They argued that Pence’s plan “creates a new entitlement largely for able-bodied, non-working adults.” They cite data showing that half of Indiana adults earning less than 138 percent of the federal poverty limit do not work at all, and only 14 percent work full-time and year-round. They contend that Medicaid expansion discourages work, citing studies showing that Medicaid expansion previously has reduced full-time employment and the likelihood of working at all by up to 11 percent. “This is particularly troubling given the fact that full-time employment moves people off of government dependence and into self-sufficiency [emphasis added].”

THIS ASSERTION brings us back to the original point, which is that the economic changes of the past fifteen years increasingly mean that full-time employment for the low-skilled does not create self-sufficiency, if by that term we mean relatively secure access to basic housing, medical care, transportation and food without government assistance. The federal poverty limit for a family of four is currently $23,850. A single earner working forty hours a week, fifty-two weeks a year would have to make $11.50 per hour to barely exceed that amount. Millions of people today can work full-time and still not lift themselves out of poverty.

The state of Ohio maintains detailed data breaking down hourly wages by occupation. Of twenty-two typical low-skilled occupations in the data set,[1] only two pay enough that a person making the occupation’s median wage and working full-time, all year could support a family of four above the poverty line. Most fall much short of that, earning less than $10 an hour.

These jobs are not a trivial part of the economy. In Ohio, over one million people have such jobs, nearly 20 percent of the state’s total employment. This means that over half a million people in Ohio work in jobs that don’t allow them to provide for themselves. And this is before benefits are mentioned; surely many of these jobs do not offer health insurance or retirement plans of the sort most of the readers of this article are accustomed to.

OF COURSE, in many families there are two adults and both could work to bring their household above the poverty line. Given the reality of children, however, it is highly unlikely that both adults would work full-time, or at least full-time at the same time of day. Most families in this situation can’t afford child care without government subsidy, so it makes economic sense for one member of the family to work part-time so that person can take care of the children before and after school. Shockingly, it is precisely this arrangement that Archambault and his colleagues also attack.

They note that Indiana data show that nearly half of all people in households earning less than 138 percent of the poverty limit do not currently work. They further note that only 14 percent work full-time all year round (whether that is by choice or because of unemployment they don’t say). Another 11 percent work full-time for part of the year, while another 26 percent work part-time for some or all of the year. Since they have previously argued that Medicaid reduces work, they use this data to contend that Pence’s proposal would make a bad situation worse.

Really? A single person without dependents who chooses to work less so he can play more is clearly morally distinguishable from a harried mom who chooses to work less to be at home when her kids get there. Not to distinguish between these different cases is very problematic, yet it is all too common for some in the “just say no” faction to fall prey to this temptation.