It is not entirely clear what was being assessed in the impact assessment of the Government's skills strategy except the level of wishful thinking in 1 Victoria Street.
Reports from colleges of a vast imbalance in the proportion of men and women among the 250,000 who will lose the right to free courses under benefit-rule changes (page 29) make the impact assessment's case that equality will not be affected untenable.
Of course, it's hard for the Government to respond to the claims of surveys like this without a chance to scrutinise the methodology. There may be distortions from focusing on two of the nine English regions, for example.
But it is startling that the Department for Business, Innovation and Skills had no data of its own to fight back with. Despite all the information colleges provide and all the resources of Government, they had no assessment of the numbers affected by cutting free courses for students on a host of benefits.
Instead, the Government's case rested on an absurd assumption: that when you remove funding for some of the poorest students and bring in a loan system two years late, no one will simply take the hint and give up.
It is of course true that ministers faced tough choices given the demand to cut spending. Still, it is worth making it absolutely explicit as to what choices they made.
While help for people on benefits is slashed, businesses are grabbing millions more in cash for apprenticeships with no regard for whether they would or could have paid for the training themselves, a practice excoriated by FE minister John Hayes in opposition.
McDonald's, to pick one example, can draw upon pound;7 million for bespoke training for its apprentices, which may become irrelevant if they change jobs. But a single parent hoping to create a career once her children are in school will not be able to get funding to start her qualifications until she is already in need of work.
The Government's assumption is that our single parent will be able to go into an apprenticeship. But much of the growth achieved is among older workers; it seems likely that few new jobs are being created.
In their doomed attempt to be re-elected, a few Labour MPs last year resorted to a well-worn bit of rhetoric with a lineage back to Neil Kinnock in 1983, accusing Conservatives of not caring for the vulnerable: "don't be poor, don't be old, don't be sick". (Kinnock: "I warn you not to be ordinary. I warn you not to be young. I warn you not to fall ill. I warn you not to get old.")
As the 1983 sell-by date shows, it has never been that successful with the voters. But it also cannot be really fair to the Coalition: even on the most cynical measure, the Tories really do care about the vote of the elderly. Less cynical observers will note that people of all parties can champion the poor.
So it is distressing to see that by looking the other way when this benefits policy was cobbled together, the Government is living up to its opponents' worst accusations.
Ministers have been stalling about publishing an equality impact assessment of the effect of these same benefit changes on English language classes for immigrants, suggesting the news may not be too good. This survey shows they should also think again about the effect on hundreds of thousands of women who want a second chance in education.