The issue of declining value of education is one of recent history that would be distorted by looking at a thirty year time frame. The question you have to ask is where are we going? Does the last thirty years offer the better feel for it or does the last five years? How much will outsourcing not only cost jobs here but also serve to put downward pressure on salaries because of implicit threats to outsource if our spoiled tech workers ask for too much? I have read of professionals in both the IT and engineering fields who say they would not encourage their children to enter the field. Nowhere do you address these psychological issues.

I'm not sure I agree with the point: "But high credit card rates are mainly due to the high cost of servicing these accounts because they have relatively low dollar amounts per account, and high default rates." No doubt those factors are part of credit card companies' pricing analysis, but isn't the other big part, and the part that makes the cost-analysis possible, that demand for credit card loans is relatively inelastic? That is, credit card companies get away, in the marketplace, with charging very high interest rates because there are many people in the market for consumer credit for whom the high cost is insufficient disincentive to borrow.

That doesn't make credit issuers bad people, or necessarily make high interest rate borrowers foolish, though there's an argument for both propositions. Rather, it's just to say that high credit costs among consumer borrowers is a function both of the costs of extending the credit and the demand for the credit. My sense is that if the demand for credit, on as unfavorable terms as consumer lenders are often able to exact, were lower, lenders would find a way to minimize the transactional costs that figure into the higher rates by, among other things, vetting borrowers better to reduce defaults, and taking advantage of technology efficiencies to reduce processing costs.

The larger point, though, is that the demand for credit by students seeking money for educational purposes is, like consumer credit, relatively demand inelastic. (In one sense, that's cause for celebration in that it suggests Americans value higher education with fervor somewhat comparable to how much they value big screen tv's.) So, if demand for educational loans is relatively inelastic, and the transactional costs of lending money are manageable in both the consumer and education loan business, what is to stop the market from lending to students at a much higher rate than they currently do?

Once someone makes the judgment that they will take themselves out of contention to earn very much money for two or three or four years, it seems to me unlikely that credit costs of an additional two hundred or four hundred or six hundred basis points, amortized over ten or twenty years, will dissuade very many students at all.

Now that may, in free market terms, be a perfectly reasonable and even desirable result. But I also think it's acceptable, and desirable, for government, in certain arenas, to make the judgment that some diminution in the freedom of markets is justified by some overriding social goals. One of those is making education affordable and attainable for its citizens.

I'll concede that that argument counsels in favor of indexing government subsidized or guaranteed or protected loans to students' income since, for middle-class and upper-class students, as I say a few percentage points won't deter them in obtaining an education. What my own argument omits, though, are these two factors (at least): (i) most students who think of themselves as middle or upper-class aren't -- their families are; (ii) an extra $200 per month in loan payments, though not a very weighty prospect viewed three or years removed from actually having to pay those loans back, is a lot when it comes to deciding what to do with one's shiny new degree. $200 may not be much to a new associate in a silk stocking firm in a big city, but it is a lot to a newly minted legal aid attorney.

In support of government guaranteed and regulated student loans are, then, these two notions. Students ought to be considered for their eligibility for government loans and subsidies, once over the age of majority, based on their own financial status, not that of their families. On the margins this will result in a few Kennedys and Trumps getting discounted rates on student loans, but in the vast majority of instances will be vastly fairer to middle income families who will need help to pay $100,000 for an undergraduate or graduate degree. Second, government does have an interest in seeing that it's well-educated citizens can afford to choose to work in places, and ways, that are important socially. Student loan subsidies are a reasonable enough way of going some distance toward doing that.

I wonder if one consequence of greater involvement by private lenders in the market for student loans would be risk-rating on the basis of the track record of the high school the student graduated from. Suppose a high school had a good record in the sense that its graduates had a high probability of completing their university degree. If that led to its graduates having to pay significantly lower rates on their university student loans, it should tend to give high schools an incentive to do a better job of preparing their students for university. The effect of the high school on interest rates would be something which could be estimated independently of the effect of factors like family income.