Criticism of Drum on Social Security

Victory at The Dead Parrot Society (bonus points for a Monty Python reference in your blog title) has posted a good critique of the Kevin Drum analysis on Social Security I blogged on a couple of days ago. Here are a couple of his criticisms, but I suggest you read the whole thing:

Drum conveniently ignores the fact that while the Trust Fund solvency date has shifted, the overall actuarial balance has not improved substantially (aside from improvements related to benefit cuts). The reason for this is clear: there is a fundamental long-term imbalance. You can make mild improvements to the system and even push back the TF exhaustion date a bit. But as time marches forward, we are adding years to the 75-year horizon that are terribly out of balance, preventing a serious improvement in system solvency. You can see this in the year to year reconciliation that the actuaries produce: every year the deficit gets worse by about .07% simply because time has marched forward while the problems have remained unaddressed. Drum also asserts that expected GDP growth this year of 3.9% is 'average', and blindly compares that to the ultimate GDP growth rate of the actuaries. Such a comparison conveniently ignores the fact that the labor force is expanding at an unsustainably rapidly pace. When that trend reverses, a GDP growth rate of 3.9% will hardly be 'average'. The optimistic Robert Gordon himself argues that GDP will average only 3.28% over the next 20 years, well under Drum's declared 'average', making Drum's choice of reference misleading, unwise, and rather self-contradictory.

The truth is of course we don't really know. How the economy and technology that drives it will resolve in the future is a big big unknown and predictions of five years away, let alone 20 or more years are almost certain to be hugely wrong. One thing that is obvious though, at least to me, is that this fact does show a weakness in Social Security as it is presently constituted, which a private account system would address. If what you pay in is earmarked for you, rather than spent on current beneficiaries as the current system is, regardless of what the economy does you will at least get back what you put in. You don't have to assume or trust that the future payees will be willing and able to 'pay you back' for the benefit you have given to others. Now obviously this doesn't address the income redistribution aspects of Social Security that are so beloved be the left (although I remain mystified as to why in this particular instance) but it does solve the difficulty of predicting problem. Secondly, it seems to me that connecting benefit payouts to the performance of the markets (via investment) would help to cushion against any form of macro-economic adjustments. Now obviously this doesn't address are current commitment to existing retirees and people who are near to retirement. That problem is of course what got us into this mess originally. As cube has pointed out in the comments on a couple of these posts, Social Security was created in response to an existing need to support old people whose savings was wiped out by the great depression. It was sold to younger voters as a trade, you pay for them and someone else will pay for you. Fair enough I suppose. However, it is fair to remember that it was instituted at a time of pretty extreme population growth and an even greater extreme degree of worker growth (as women were entering the workforce.) These two metrics have stagnated and to some degree reversed themselves, leading to the crux of our current problem (or potential problem if you like.) I don't know that their is any easy answer. There may not be any good answer. I do have some frustration with those who are unwilling to even explore the possibility of a different method for doing things.