“To be more specific, I wouldn’t deny that there is SOME smoothing done, but current consumption isn’t independent of current income”

No-one made that claim though – all you require is some consumption smoothing, and thereby some credit constraint, to make the result hold.

I do not disagree that the extreme position of “all people would smooth consumption perfectly” is very unlikely to be true. However, I find the comment “people don’t smooth consumption over their lifetime” equally extreme.

It sounds like we agree on this in the first place – but I would like to note that the assumption of perfect consumption smooth was never, and does not need to be, made anywhere in what I have said.

I don’t think it’s patently false, just unqualified as it stands. To be more specific, I wouldn’t deny that there is SOME smoothing done, but current consumption isn’t independent of current income. No model where consumption is a function only of lifetime income can explain observed consumption patterns.

“I have no doubt that credit constraints are part of the explanation; however, it would be rash to presume that they are the primary reason why consumption follows income over people’s lifetimes”

I agree that people consumption tilt for sure, that is part of impatience 😉 . However, I also believe that people consumption smooth.

“I’m not disagreeing with the point of your post, merely adding that there are many other considerations when thinking about lifetime welfare and consumption smoothing.”

Also agreed. However, the welfare cost I wanted to discuss stems from the specific issue of credit constraints – hence why I focused on that.

Furthermore, my comment was in response to you stating that “people don’t smooth consumption over their lifetime”, a statement that I believe is patently false.

“Note also that Thaler’s estimates were based on a behavioural consumption model, so his estimates aren’t directly comparable to standard estimates of MPC.”

Interesting. Wouldn’t mind having a look at that tbh, the methodology of behavioural models fascinates me – mainly because it feel fresh methinks.

I’m not disagreeing with the point of your post, merely adding that there are many other considerations when thinking about lifetime welfare and consumption smoothing.

Note also that Thaler’s estimates were based on a behavioural consumption model, so his estimates aren’t directly comparable to standard estimates of MPC.

Thaler …

Lots of people have done estimates, and Thaler’s estimates were by far the most extreme. Mankiw and Barro have both written a bunch of stuff on it.

Furthermore, we have to ask what causes the MPC from current income to be positive (for unchanged lifetime income). One of the major explanations is … credit constraints. As a result, if Thaler’s results are true it might indicate that the credit constraint is EXTREMELY binding – which would imply that the welfare cost is HIGHER.

Party times had by all.

Agreed. I know plenty of people that maxed out and just put it in a savings account.

“Of course, the problem could be solved entirely by adding provision for greater loans at market interest rates”

Agreed. Although I would prefer it if they waited till I’d payed back a bit more of my debt 😛

However, if I remove my self-interest I would suggest they should do it immediately.

There could be a credit constraint, but I’d warn against taking the proportion of students currently maxed out on student loans as evidence of anything. Of course, the problem could be solved entirely by adding provision for greater loans at market interest rates (along with a ball and chain to keep them from fleeing the country without paying debts, I suppose). Not that you’re taking those numbers as evidence. But that a majority of my students were on no loans prior to 0% suggests constraints are generally non-binding.