Wednesday, June 13, 2007

This Needs to Be Hoisted from the Comments

by Ken Houghton

Robert Waldmann, commenting to Tom's post below and quoting his comment here at his blog, notes:
I personally claim that, for any policy you propose, I can write a model such that it is optimal. If I am right, there really isn't anything for liberals to disagree with. I haven't been stumped yet (OK no one has tried but come on it would be fun no ?).

That is a challenge that needs to be taken! Hints on my pending reply are in the Labels. YMMV.

Labels: , , ,

Comments:
Does he mean a policy regarding anything? What about the social security system, in terms of the schedule of payments in by workers and payments out by the government? I thought I once read Milton Friedman say there was no way that was optimal in an economic sense, so it would be interesting having this commenter prove him wrong.
 
Jeremy: Over at Robert's blog, I called all policy problems of the "intervene to fix market failure" and "don't intervene despite market failure" for Robert's Wager. (Remember, he says "a model," and I'm implicitly saying the models are of the 'second best' variety.)

In this particular case, I wouldn't even think it would require an unrealistic model (by prevailing standards). Insofar as being left to one's own devices makes people poorer overall in old age, I'd argue that Friedman's side would have to engage in more contortions to explain why it's optimal to be free to starve.
 
...and Robert weighs in:

"In this particular case, I wouldn't even think it would require an unrealistic model (by prevailing standards). Insofar as being left to one's own devices makes people poorer overall in old age, I'd argue that Friedman's side would have to engage in more contortions to explain why it's optimal to be free to starve."

Indeed. All that is needed for forced saving to be optimal are dynamically inconsistent preferences (people have problems with self control). Since people are wildly enthusiastic about social security, it is impossible to believe that they are rational enough to do without it (if they are so rational why are they wrong about social security ?).

The commenter might be talking about the bend points etc.
The optimal tax and transfer schedule is, in standard models, the solution to a very hard problem in which any result can be optimal for the right assumptions about labor supply.

 
I meant the particulars of the system, like how the payroll deduction stops at zero above an earnings level, and pairing that with the schedule of payments that people receive given that it is a pay-as-you-go system. I thought Robert's challenge involved being able to write a model that was optimal down to specifics.
 
That's at the end of Robert's comment ("bend points, etc.") -- as he says, the general result is that you can tune standard model assumptions to make any tax-and-transfer schedule optimal.
 
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?