The article does compare them (albeit without much detail).

“Delinquency rates on student loans are much higher than those on auto loans or mortgages, due to loose student loan underwriting standards, the unsecured nature of student debt, and the inability to charge off non-performing student loans in bankruptcy,” Goldman Sachs Group Inc. analysts Marty Young and Lotfi Karoui wrote in a note Tuesday. “The substantial majority of student loan default risk is borne by the U.S. Treasury.”

If the default risk is borne by the US Treasury, then what risk do we have in general?

The risk that the increased supply of money lowers the value of that money. Not that this is the straw that's gonna break the camels back, but it's not like there aren't consequences to the taxpayer for allowing the government to allocate funds in a sub optimal manner.

This money isn't unbounded; it's confined by the budget allocated to the Department of Education and various government organizations (like Fannie Mae, et. al.).

Ultimately the Department of Education is responsible for this money. The last 30 years are remarkable in the amount of controversy surrounding student loans and the default of large education funds.

The following article describes this in greater detail: https://tcf.org/content/report/student-loan-guaranty-agencie....

I think the main risk would be a deterioration in the government's finances since the government is expecting to get paid back and is factoring in those payments to the financial outlook. That could in turn lead to an increase in the government's borrowing costs, which could affect lots of other things negatively.

Classic case of moral hazard.

Just FYI the average default rate for the 2014 Cohort was 11.5% but that varies WIDELY depending upon the school from 0.10% to 56.20%.

Official numbers can be found at: https://www2.ed.gov/offices/OSFAP/defaultmanagement/cdr.html

Default rates in good times can be very misleading (as the subprime crisis showed).

Can't you extrapolate default rates to when we don't have good times.

That's what banks do to manage their credit risk, they look at defaults rates "through the cycle", i.e. including the last downturn. But for new asset classes, like subprime in their time and like student loans now (if you look at chart in the article in term of growth), people have to make their own assumptions in term of how it will behave in a downturn. And that's where the credit markets can be really dangerous. These assumptions are what banks and rating agencies got wrong last time. And I think it is fair to assume that people under financial hardship will probably default on their student loan before they default on their mortgage (roof over their head) or their car.