[such_a_casual]

The point is that the cost of the college grad is not a basis for economic decision making. Once the cost has been accrued, it's sunk. If we're making a decision based on whether or not we want to sacrifice the baby or the college grad, we can't take into account sunk cost of the college grad (if we're using economics as the basis of our argument).

[davorak]

> The point is that the cost of the college grad is not a basis for economic decision making.

The common assumption that ajmurmann and vasiliys are likely working form is that value has been retained during the ~20 years of investment and the cost is proportional to the value and there cost is a reasonable proxy for value.

It is an assumption that does not always hold, but is common enough to do as a first order approximation.