The movement to provide disadvantaged students with ¡°free¡± ÖйúAƬ is enjoying a bit of a boom these days in the US: at least 16 states now have such offerings, half of them joining in the past year.

Yet as so-called Promise programmes gain momentum, they¡¯re also attracting sceptical attention ¨C both from their expected critics on the political right and, now, from their traditional allies on the left.

Three separate policy-study groups that advocate for low-income students have warned?states that some of their Promise programme structures may be hurting the students they mean to help.

The Promise concept, which dates to the 1990s, generally represents an attempt to go beyond traditional student aid models by covering tuition costs that remain for low-income students after counting all of their other available funds.

ÖйúAƬ

That ¡°last-dollar¡± approach, however, is a central concern now being raised by experts from the three study groups ¨C the Institute for ÖйúAƬ Policy, the Education Trust and the Century Foundation.

The groups are asking states to instead consider a that would give students an initial chunk of aid, leaving federal grant money and any other assistance to help the students with important non-tuition costs such as housing, transport and books.

ÖйúAƬ

In many cases, the groups said, last-dollar models that count only tuition and fees can leave college unaffordable for low-income students. Last-dollar models also make it easier for higher-income students to compete for scarce state aid, their analyses found.

The Education Trust proposed of a ¡°truly equity-focused¡± free college programme, including the coverage of all costs for all low-income students for at least four years and a ban on eligibility requirements tied to grades or credits earned.

The group studied 15 existing statewide programmes and 16 proposed new schemes, concluding that the programmes on average met about half of the eight criteria and none met all of them.

The Institute for ÖйúAƬ Policy used similar benchmarks to assess programmes in two states ¨C New York and Tennessee ¨C and . Both programmes provide thousands of extra dollars for high-income students who don¡¯t need it, while leaving low-income students thousands of dollars short of their actual costs, the institute said.

ÖйúAƬ

Officials from both states are pushing back. A Tennessee Promise founder said the programme has helped an additional 1,700 of the state¡¯s neediest students attend college each year.

A spokesman for New York¡¯s governor, Andrew Cuomo, called the institute¡¯s analysis ¡°uninformed and fundamentally flawed¡± because it did not sufficiently consider a separate scheme that provides nearly $1 billion (?766 million) in aid to the state¡¯s neediest students. Unlike other state programmes, which limit eligibility to two-year colleges, New York¡¯s has since 2014 included four-year institutions.

Some in Congress ¨C??Senator Bernie Sanders, a 2016 presidential candidate, and Senator Elizabeth Warren, a potential 2020 candidate ¨C have been free college programmes on the federal level. Another, Senator Brian Schatz of Hawaii, a measure aimed at making university? for students, rather than tuition-free.

The differences over such details on the state and federal levels underlie a key political consideration seen by analysts as affecting the future of such programmes: . While affordability advocates recognise the importance of helping those most in need, some have that excluding middle- and upper-class families may make Promise-type programmes politically difficult to sustain.

ÖйúAƬ

Data from another group, the State ÖйúAƬ Executive Officers association, may hint at that. The group calculates annual state expenditures on ÖйúAƬ, and show that the six states with the oldest Promise-type programmes ¨C Indiana, Oklahoma, Missouri, Louisiana, Delaware and Washington ¨C are all now spending below nationwide averages for ÖйúAƬ.

paul.basken@timeshighereducation.com