In response to my last post, Max has clarified his remarks in a subsequent post. No need to go through it in its entirety. On his main point, he now accurately characterizes my views as to what gets worse over time:
What he means if I understand him correctly is that from the standpoint of future generations -- retirees and taxpayers -- the problem of the unfunded ten trillion dollar liability (as of 2004) gets worse because preceding cohorts of retirees are let off the hook ("held harmless" in his words) by having the good fortune of receiving their SS benefits and meeting their Maker before the bill comes due.
I'll add two additional remarks to further clarify the discussion:
First, Max makes the following statement:
Concerning Andrew's 3.5 percent of GDP, I assume this reflects another one of those calculations out to perpetuity, since the Trustees report shows the gap to be only slighly over two percent of GDP at the end of the 75 year planning period (in 2080). The increase in the cash gap relative to GDP has to take some time to rise much beyond two percent of GDP.
Yes and no. The "3.5 percent" to which I often refer is the 3.5 percent of taxable payroll that would be required immediately and in all future years to close the projected $10.4 trillion financial shortfall. It is shown here in the 2004 Trustees Report and is equivalent to about 1.2 percent of GDP. I cite this figure to put the present value calculation in the context of the traditional revenue base for the program. It is an infinite horizon measure, but it is scaled by taxable payroll not GDP.
Second, Max concludes his post with a discussion of whether "pre-funding" this unfunded obligation is good policy. This is an interesting question. As I have noted, my first choice would be not to pre-fund the unfunded obligation but instead to reduce the growth of future benefits (via increases in the retirement ages, not the replacement rate at the age of normal retirement) so that the unfunded part of the obligations is reduced to zero. Reformers to the left of me ideologically--like Diamond and Orszag--have also designed plans that do not entail substantial pre-funding. They do it with tax increases and no reduction in projected aggregate benefits for about five decades. This is a sensible distinction between right and left--smaller versus larger public spheres. I'd like to see it become the template for Congressional debate.
Without pre-funding, there is no need to work through the thorny issues of how those funds are to be accumulated and invested--centrally (via a Trust Fund with a hands off management and a "lockbox") or in a decentralized system of personal accounts. I favor the latter and prefer it to a Diamond-Orszag approach. Earlier this week, Brad DeLong commented in two posts on why he would prefer the former. Some of the difference may be the scale of the investments that we are each envisioning. Clearly, there is some size of the system's aggregate investment in equities below which even I would concede that it would be fine for the government to control the investments. I don't have a specific threshold in mind, but I'm pretty sure that it's sufficiently small (e.g., less than a hundred billion dollars) that it would do little in the way of restoring solvency. Presumably, there is some size of the system's aggregate investment that is sufficiently large that Brad would say is appropriate for a decentralized system of personal accounts. But maybe not, and I'll be the last one to put words in someone else's blog.