> > why don't we take this $500B of canceled debt and direct that into supplying mosquito nets to malaria-prone regions of the world instead? > Again with the

> > why don't we take this $500B of canceled debt and direct that into supplying mosquito nets to malaria-prone regions of the world instead? > Again with the false dichotomies. i don't know how you dismiss this as a false dichotomy so quickly. i went on to present how i experience this dichotomy in my own life. we do not have limitless resources. therefore whenever we direct resources toward one thing, we are implicitly directing resources away from other pursuits. if student debt cancellation literally did consume only trivial resources, then yes, this wouldn't be a dichotomy. however on the surface, it's $500B; when you factor in debt that was already in default like 3 years ago, it decreases a bit but still in the 11 figures. if you're an economist with a goal, you can swing the data to present changes in the overall economy in whichever direction you want; the (often unstated) error bounds on those grand predictions are often too massive to be useful. > In fact many student loan debtors were in default or income-based forbearance before the 2020 pandemic suspension of payments. what about this related talk around bankruptcy reform? this is far less expenditure, visibly so with just a glance: if somebody is literally unable to pay their debt, then the lost revenue due to canceling it is trivial. whatever forecasting you use to back this up touches far fewer variables/uncertainties than the broad economic forecasting people do around this debt forgiveness program. you just sum the probability that a person in default could recover in the timespan of their loan (for which there are hundreds of thousands of reference points) multiplied by the value of that loan. i could vet their model in an hour. bankruptcy reform would help the people who have it the absolute worst, and it helps not just this year but every year for as long as such a policy stays in place. > You say you support keeping tuition low, but that involves massive, permanent taxpayer subsidy, of the kind you don't seem to like. you're right that i prefer directing subsidies into investments rather than servicing costs. i.e. subsidize developments which will lower the cost of tuition indefinitely, v.s. continuous payments which will never get smaller. examples of this are online/open courseware. the stuff out there today is good enough to replace the instruction portion of the first 2 years of my EE studies: a sizable chunk of overall tuition costs, if we made use of it. it used to be that access to lab equipment was reason enough to pay high tuition as a EE. but if i look at every piece of gear i used during my studies, i could recreate a suitable lab for < $2000 today, all COTS components. in theory, the reduced lab expenditure due to technological improvements should have reduced the actual cost of a EE degree during this time. i'm not against non-investment subsidies altogether. for example, primary school where it's as much about socializing kids as it is academic, is naturally labor intensive: it's hard to imagine any way out of paying one instructor per every 20 (or 15, or whatever) kids, every year. some amount of this applies to higher-ed, but the floor is lower: there's more room us to decrease the recurring subsidies by shaping them such that they cause an equivalent education to be less expensive each year. for the university case, tuition has risen in the face of the two deflationary technologies i showed above (which on their own, should have decreased the cost of an equivalent education). and hence it's not obvious that increasing tuition subsidies is a pragmatic approach to addressing the complaints i hear about our educational system. how much of these subsidies serve to encourage universities to scale their academic programs to support more students, v.s. how much serves to encourage universities to build cheaper, but non-beneficial programs (i.e. degrees which are cheap to provide, but don't lead to occupational paths) or even straight to the endowments? i don't think blindly increasing tuition subsidies is a great way to better our situation. at least, the mood over the last decade is that this hasn't helped. i'm hopeful of income-share agreements. the public perception of higher education at this point is that people expect it to lead to good careers, and students are upset when this hasn't been the case. conditioning the university's revenue upon the quality of the career they enable for a student is one way to encourage this outcome. the federal loan ISA option doesn't actually do this: universities get the same revenue regardless of how better off the student is. make the ISAs an agreement between the student and the university. if it's a 10% ISA, the federal government could offer to cover half of that for the student, effectively lower the burden on the student to 5% of their income. if we're only recovering half of the loan payments today, this is effectively the same cost to the federal government, but better aligned so that the university actually produces good outcomes for the student (these figures are made-up: adjust the numbers as appropriate before doing this in practice). >> their endowment is large enough to cover -- by very rough estimate -- about half of tuition in perpetuity; but not enough to refund their graduates. > I'm not sure what "refund their graduates" is supposed to mean — full refund of all tuition ever paid by every attendee? — but it sounds like much more than necessary to ameliorate the student debt problem for current debtors. the point here is that the endowments aren't large enough to both make higher-ed free to future students and compensate previous students. there's necessarily a trade-off between the two.