Recorded as enacted, because it is one of the few clean fixes in this database and the site should say so as loudly as it says the rest.

Two questions, which look similar and are not:

• Can this organisation deliver and assess training to the national standard? A training-quality question, asked by ASQA against the Standards for Registered Training Organisations.
• Should this organisation be permitted to create a debt on a student's tax file, repayable for the rest of their working life? A fiscal and consumer protection question, about who controls the organisation, how it recruits, and what happens to its revenue if the student never attends.

Under VET FEE-HELP the second question had no gate of its own worth the name. An organisation that got through the first gate could reach the loan scheme, and the scheme was uncapped: revenue arrived per enrolment, from the Commonwealth, whether or not anybody learned anything.

The trajectory is the argument.

Year	VET FEE-HELP lending
2009	$25.6 million
2012	$325 million
2014	$1.75 billion paid to providers
2015	$2.9 billion, almost 200,000 full-time-equivalent students

The number of loans doubled between 2013 and 2014. No assessment gate anywhere in the chain was capable of responding to that, because none of them was looking at volume — they were looking at whether individual organisations met training standards, one at a time, on application.

The final cost: $2.2 billion lost to the taxpayer, including an estimated $1.2 billion of loans issued inappropriately in 2014 and 2015 alone that will not be recovered. The Australian National Audit Office examined the administration of the scheme; the Senate inquiry examined the conduct.

That is not a training-quality failure and a training regulator was never going to catch it. A provider signing up students through brokers paid per head, who would never complete, was running a perfectly functional business. The Senate inquiry found marketing that misled and specifically targeted vulnerable and disadvantaged consumers.

What the VET Student Loans Act 2016 did, from 1 January 2017 — with existing students grandfathered on the old scheme until 31 December 2018:

1. Made RTO registration a precondition, not a qualification. A separate application, to the department, for approval as a VET student loans provider.
2. Asked the fitness and financial viability question directly, of the organisation and of the people controlling it — and asked about student completion rates and links to industry and employment, neither of which is a training-standards question.
3. Restricted loans to courses on a determined list — diploma, advanced diploma, graduate certificate and graduate diploma — with a smaller set approved only for specified listed and not-for-profit providers, rather than anything at diploma level and above from anyone.
4. Attached continuous conditions rather than trusting the gate: loan caps per course, payment in instalments against census days so a provider is not paid in full for a student who left in week two, and evidence the student is actually engaged.

Point 4 is the transferable lesson, and it is the same one this site drew from universal aged care registration: the durable control is usually a continuous, dull, cheap check rather than a bigger entry assessment. Entry gates are assessed once, by people who cannot see the future, on evidence the applicant chose.

And this entry is the counterweight to most of this site. Ninety-eight processes here argue that a step should be removed, merged, thresholded or run in parallel. This one records what happened when a check was missing, and it cost billions of dollars and a great many people a debt they did not understand. A method that could only ever produce arguments for less process would be a bad method, and this is the clearest case in the database of the other direction being right.