In terms specifically of the writeoff, the situation around that and how that's changed over time, the committee may find it helpful to have a historical trend line on what that looks like.
The best measure of student default that we use is a default rate after three years. That's a kind of standard tracking rate we use. That rate has declined almost like clockwork by about 1% a year since 2003-04, when it was at 19%. The rate this year, as we're asking for the writeoff, is the lowest rate we've had, which is 9.0%, but it's been a relatively straight decline year after year.
That's a combination of factors over that time that includes more flexibility for students in terms of repayment. As you may be aware, students do not have to start repaying until they start earning an income of $25,000. If they run into trouble, there are flexibilities provided to them in terms of when they start the payments. It also reflects more effective collection efforts on the other end so that we can reduce the amount of the writeoff. I would say the trend line is moving in the right direction.
I did ask for a bit of an international comparator, because that's sometimes helpful. It's hard to do apples to apples. The U.S. system is the closest we came to, and their rate in comparison to our 9.0% rate is about 10.8% or 10.9%. The trend line has been good on this, but it reflects both sides.