William J. Polley

The surprise there is the idea that SocSec is the primary source of income for 2/3 of retirees. After all the talk of how "everyone" had defined-benefit pensions, the idea that those benefits don't even equal $11.5K/year ($957.60*12 per this table) suggests to me that the authors of the piece are including dependents in that 2/3s.

If you read between the lines, Barro's 7/3/2000 oped and Becker's 2/22/2005 make two assumptions: (a) households have retirement wealth besides the Soc. Sec. benefits; and (b) households are rational (or have a smart advisor). Conclusion of their model (and mine): privatization will not change asset allocation. No increase in expected returns or risk. QED!

PGL,

What about young people with low but growing incomes and who have little savings outside of Social Security? Putting such a high percentage of their income into something with such a low return at their age is surely not optimal.

I think I made this point in a previous post some time ago but can't find it now. I'm with you on the fact that there won't be much change in asset allocation. But there will be some change, mostly among young people--as it should be. I think we need to increase savings among young people somehow, even if it's not through Social Security privatization. I think you would agree with that, no?

As I have said before, even if we could have partial privatization, I would put strict age limits on it. The bills in Congress allow up to age 50 or 55 I think. I'd set 40 as an absolute upper bound (and that with caveats). In a perfect world, I'd set it at 18--only brand new entrants to the labor force. Let the current generations of workers and retired folks play out the game according to the rules they had when they started. In that perfect world, I think you would see some more saving (and an endogenous increase in financial literacy) among the younger generation.

As the old joke goes when a kid gets his first paycheck: Who's this FICA guy and why is he getting my money? I think if an 18 year old knew that part (not all--part) of FICA was going into a savings account that he or she owns, it would be a good thing.

If that's too idealistic, tell me.

Ken,

I looked around a little and all I could come up with to support the 2/3 number was the SSA website itself.

http://www.ssa.gov/policy/docs/chartbooks/fast_facts/2003/ff2003.html

Not sure where they get it though.

Surprise over that number aside, I take it you agree that it will probably fall over time?

Bill - I agree that these folks MIGHT be borrowing constrained. But read Andy Abel's 1981 AER paper. If these folks are holding bonds, they are not so constrained. Maybe low income folks have higher risk aversion than others. Besides, you've just made the case for Clinton's proposal that the Cato crowd said no to.

PGL,

Do you mean Abel's 2001 AER paper? I have seen it, and it does bring something to bear on this discussion.

However, I'm not just talking about low income households in general, but about YOUNG households whose income is low relative to what it will be later. I'm talking about people whose lifetime income profile is telling them to borrow when they are young and may be borrowing to their limit, but because of compound growth they could benefit immensely from a (tax deferred) retirement savings account with a higher return (even if that means higher risk). Young households would have a higher tolerance for variability because they have more time. That's the whole reason for having a cutoff age for initial participation in any privatization plan.

I don't think the people I'm talking about hold many bonds, but maybe I'll ask some of my graduating seniors.

As for Clinton's plan, I believe you are talking about his "USA Accounts." The intent was to keep people paying all of their FICA tax towards Social Security and subsidize additional saving for low income workers. At least that's what I found in the dusty old archives of the internet.

That's not exactly what I want.

I'm more interested in reform of the system along generational lines than along income lines. If that succeeded (and I think it would), I might actually be inclined to look at something like Clinton's plan. But I would want to get ALL 18 year olds into a new system first and let everyone 19 and over play under the existing rules. If you've followed my posts this far, you know that's pretty much all I've ever wanted.

But I don't think there any chance of that happening.

The privatization plans in Congress have cutoff ages that are just too high for my blood. That's my main problem with them. I have all along been implicitly acknowledging the opposition's point that we're not in crisis yet. Right. We're not in crisis, so let's do something more gradual that has a better opportunity for long run benefit.

But let's face it, our political system does a lousy job of handling generational policies. Perhaps we have common ground there?

PGL,

In fact, I just looked back in my archives. The first post in which I talked about how a gradual transition to privatization would be necessary for it to work, beginning with very young workers:

http://www.williampolley.com/blog/archives/2005/02/privatization_n.html

Your comment was favorable. Flattering, in fact.

I haven't moved an iota off of that post.

Bill - Yes, as suspected the line is "Social Security provided at least half the income for 65% =of the aged=" [emphasis mine].

That means that spouses who never had paid employment, or only were paid sporadically (and, therefore, don't have a pension worth discussing), are included in the data.

As you see more two-earner families, you will (almost by definition) see a lower percentage of people who derives more than half their retirement income from SocSec (both because of IRA/401(k) contributions being allowed even if one spouse is not working and because dual earners have greater discretionary income, some of which is likely to be saved).

IIRC, Greenspan and O'Neill in THE PRICE OF LOYALTY project a breakeven age for SocSec at 37. But they also probably projected based on the traditional earnings lifecycle model.

I did mean 2001. Bush's general fund policies ane say free trade-do protection has me 20 years younger. But fair point on the 25 year old who expects higher future income etc. But this is the example I often use when discussing the Ando Modigliani life cycle model. And its implication is that the 25-year old might actually borrow on expected future income. OK, giving him his payroll contributions back might change the timinig of his after-tax income, but in the A-M model, timing does not affect consumption v. savings. So why would it impact allocation of savings? Perhaps the borrowing constraint argument, but then these 25-year olds have figured out how to get around borrowing constraints from what I can tell.

Maybe someone should survey 25 year olds about what assets they would buy under privatization. If anyone has, I have not heard of it.

If your last statement (about 25 year olds getting around borrowing constraints) is referring to how so many of them have gone into debt to finance current consumption, that's a point well taken. The life cycle model is powerful. You need to have proper incentives to get 25 year olds to take advantage of savings opportuities when they are young so that they get the most out of compound growth. If the incentives aren't there... well, you get what we've got.

Bill - my comment went beyond that. Suppose this borrowing is rational for a second. My point was that if one can borrow to finance the new BMW while still in graduate school, why not borrow against those Soc. Sec. benefit checks to hold stocks. I guess there is some explanation that markets aren't so perfect, but then there are lots of folks who wish to challenge Modigliani and Miller's celebrated 1958 paper.

One word: Collateral.

Or as the old joke goes...

Person to banker: "You mean in order for you to loan me money, I have to prove that I don't need it?"

Yes - collateral requirements would be a restriction on borrowing that might bite my high flying MBA student. But then some of these wizards of finance have figured out margin requirements allow them to hold twice as many stocks as they put down in cash - and perhaps they know a lot of other ways of leveraging to the hilt.

Sure, the high flying MBA student might be able do it by borrowing at margin rates.

But what about the rest of the folks for whom borrowing on the margin might not be the greatest advice?

All I'm saying is that if you give them a dollar and tell them to put it into the financial markets, some of them might buy stocks rather than bonds. Since the trust fund only holds bonds, not stocks, it would follow that there would be some effect on returns and risk. It may not be such a big deal in aggregate, but it might be a big deal for the young families of more modest means.

Again, all I'm objecting to is your statement about there being no change in expected returns or risk (and the implication that private accounts will not provide a net benefit). I have said repeatedly that the total effect on the average portfolio will be small, but if it were up to me I would devise a privatization plan that would have the greatest impact on the relatively small number of people for whom it would make a difference (who neither have MBAs nor would be advised to borrow stocks on the margin).