mk's comment above was the cause of much private discussion behind his back. I was unable to square what appeared to be purely faith-based advocacy with the objective neuroscientist we know and love. Given an opportunity to see him in person, I printed our exchange along with a helpful summary:

“We agree that this program would have costs and benefits. We agree that other programs exist that have worse cost/benefit ratios. I don’t know if the benefits of this program would outweigh the costs. Why do you believe the benefits would outweigh the costs?”

“I believe the benefits will outweigh the costs.”

Discussion ensued, with Meriadoc contributing.

What I find worrisome is that mk concludes that the proposal will be a net benefit without presenting a shred of evidence. Well, surely there is evidence that improving access to education will bring benefits. I don't mean to dispute that. It's just that we have no information on the costs, the negative effects. I find it particularly alarming that mk does not dispute my prediction that the very poorest will likely be harmed by this policy, and the wealthy will benefit (a pattern lamented elsewhere).

Meriadoc made an interesting assertion that any politician who is enlightened enough to propose such a program would surely arrange to have it funded in a responsible way. I would like to see evidence for this belief. I don't know if legislators even have to specify where funding will come from, or if approved programs are simply paid out of a general budget. I was not impressed by California's model, in which a family leave program is completely paid for from a mandatory tax on all workers' salaries, while most low-income workers do not apply for the benefit, because they do not even know that it exists.

More than once I heard the refrain "We already spend money on far worse programs." That is true, and it should stop. It is not a good reason to support a program that may still be bad on balance, only less bad. If you believe that the new program will be covered by "repurposed" funds, so that it is an improvement even if it does not provide net benefit, you should have evidence for that belief.

mk and I don't think exactly alike on political matters. In my personal correspondence there was some exegesis of his words "your take is largely going to be dependent upon your own politics and value system." I might believe that forcing people to pay for community college under threat of locking them in cages is an act of coercion, and he might have a different perception. Nevertheless, his argument should be "replicable." I should to be able to come to the same conclusion as mk by thinking like mk.

Say B is the total benefit that we expect will accrue due to the implementation of the policy: more educated people, better jobs and higher salaries, reduced poverty, more prosperous community college founders and staff, etc.

Say C is the total expected cost of the policy: things not purchased because resources were diverted to community college, income and work experience not gained while going to community college, distortions to incentives for community colleges to provide good value, dilution of the perceived value of a community college degree, fraud and waste at rent-seeking institutions that pretend to be community colleges, overhead to run the program, etc. We will ignore any costs that I recognize but mk denies.

So if I am mk and don't worry that Sam has a problem, I think I would support the policy if I have reasonable expectation that B > C.

But what I actually heard mk arguing was this:

B is high enough to justify the risk that B < C.

This seems like broken logic. By this rule, we might not support programs that provide small benefits, even if we are fairly sure that they are beneficial overall. We will simply look for the program that promises the most attractive outcome and support it, irrespective of the costs.

It seems akin to arguing:

  The health budget is X.
  Health is important!
  Therefore, the health budget should be 2X.

I am not arguing that taxpayer-funded community college is good or bad. I am arguing that mk and I do not know (i.e. do not have evidence showing) that implementing a policy of funding community college through taxes would do more good than harm, and we should have reason to believe this before supporting the policy.