Earlier this year, the Bipartisan Policy Center, a think tank that, as you might imagine, promotes bipartisan policy solutions, released a report detailing what it calls “The New Middle on Immigration,” drawing on a survey conducted by Luntz Global. There is a lot to discuss in the report, but for now I’ll touch on two small things.

First, there is a broad consensus that immigrants should know at least some English before settling in the U.S. Though only 11 percent of respondents believe immigrants should be completely fluent, 45 percent say they should know enough to hold a basic conversation (enough to shop on their own, or to communicate with neighbors and teachers) while another 33 percent say they should know enough to easily work at their place of employment. The remaining 14 percent felt that immigrants needed no more than a few words of English on arriving in the country, presumably in the expectation that they would master the language over time.

Given that there are 1.5 billion people around the world who are studying English as a foreign language, it doesn’t strike me as unreasonable to expect immigrants to speak at least some English or to prioritize those who speak it fluently, provided we make an exception for refugees and asylees, who are, of course, in a very different boat. As it turns out, however, even when we exclude refugees and asylees, 39 percent of recent green-card recipients don’t speak English well or at all. This is despite the fact that there are over 4 million on the waiting list for family-sponsored and employment-based preference visas, many of whom either speak English fluently or would be more than willing to improve their English proficiency if doing so would move them up the queue. It is not hard to imagine a more coherent and transparent system for allocating permanent visas that encourages petitioners to acquire skills that would help them successfully navigate American life, such as the ability to communicate in English. But judging by the Gang of Eight, liberals and moderates are resistant to any reform along these lines that doesn’t also sharply increase future low-skill immigration.

Second, the BPC survey found that Americans are more concerned about immigrant reliance on social benefits — an expansive category that could include safety-net benefits, refundable tax credits, and other measures designed to meet the needs of households with low market incomes — than they were about the role of immigration in intensifying competition for jobs or lowering wages.

This makes sense to me. In Melting Pot or Civil War?, I made the case that benefits are a bigger deal than jobs or wages. Provided a country has a lightly regulated labor market, including a low statutory minimum wage, it can always incorporate low-skill newcomers by, for example, adopting more labor-intensive business models. A “shortage” of low-skill workers, in contrast, will prompt firms to adopt more capital-intensive business models and to invest in comprehensive vocational education to boost worker productivity. In either scenario, native-born workers can do just fine, whether by complementing low-skill immigrants or increasingly sophisticated machines. One difference, however, is that while machines won’t be in need of SNAP, Medicaid, and wage subsidies, the same can’t necessarily be said of immigrants earning low market wages and the family members who depend on them.

And as it turns out, among recent green-card recipients who did not enter the country as refugees or asylees, 33 percent live in households with incomes below 125 percent of the federal poverty level, which would all but guarantee that they’d be in need of benefits, while 39 percent live in households with incomes at or above 250 percent of the federal poverty level. Of course, one expect the market incomes to increase over time, but the wages of low-skill workers, whether native- or foreign-born, have been under pressure decades. Indeed, that is why there’s been so much support for boosting transfers. Insofar as voters believe (non-refugee) immigrants should be in a position to provide for themselves and their families, they ought to support moving to a more selective, skills-based system.