Ah, true, but let's distinguish between two types of political uses of the Trust Fund/ lockbox concept. First -- and this is the case from the start of the system, not just as it was "reformed." The SocSec system is treated, politically, as if it's self-contained, and for lots of good reasons. You don't want to have the funding and benefits subject to renegotiation every budget year. So a rate of funding and benefits calculations are set that won't require constant adjustments and, procedurally, would present major challenges to anyone who wanted to "fine tune" the system all the time. That's necessary both for the effectiveness of the program -- people need to be able to plan from year-to-year on the main features of the program -- and for its political survival. Once it was reformed, the system had the added benefit of reducing the amount of deficit financing that the federal government had to raise each year while we were producing such big large annual net negative cash flows in the non-SocSec part of the fiscal system. The lockbox was the most recent politically expedient iteration of the self-containment concept. In this case, it was extended beyond SocSec itself for lots of admirable reasons as explained above. But to make that argument, the Dems had to invest a huge amount of energy and "communication capital" into the "lockbox" concept, which built on and reinforced the Trust Fund concept. Now they're dealing with the consequences of a reasonably successful sales job on the lockbox. First a bunch of them (both politicians and wonks/pundits) may really believe it, or at least really believe it's the soundest way to conceptualize Social Security. But also, they're on record about the reality and inviolability of the Trust Fund. So when Bush launched the whole privatization debate based on "crisis," they were in an awkward position. Let's assume they asked themselves "what if we base our argument on a cash-flow analysis." The answer was very unattractive: the Repubs will have a field day throwing sand in everybody's eyes about how we're contradicting ourselves. And that will let Bush off the hook, because they can spend their time discrediting us rather than having to defend their own programs. Furthermore, focusing on the Trust Fund pushed the crisis date out into the "can't get your brain around it" future, where not only the timing but the projected size of the so-called crisis was also easy to question. So it certainly was good tactics to "defend" the Trust Fund while demolishing the crisis argument. That was, in fact, the thrust of (at least part of ) Krauthammer's article -- Bush had chosen a losing tactic politically. I'd argue, however, that anti-privatizers are now starting to reach the point of diminishing returns on that set of tactics. If the Republicans now want to talk cash flow, it may indeed be time to start talking cash flow -- but put it in the overall fiscal situation, not just Social Security, where you'll simply get lost in the bowels of Capitol Hill. BTW -- another little dirty secret that always seems to get lost in the trees, and why Social Security is really better framed within a much broader macroeconomic discussion. In the macro-sense, payments to retirees, whether from private investments or from government payments, will either be easy to afford or hard to afford based on the overall structure of the US economy in future decades. It's really immaterial in the aggregate sense what form retirement payments take (though clearly there's quite a difference for individuals as well as the overall distribution of payments to retirees). Let's always remember that financial assets are merely claims on future revenue streams. So it doesn't really matter in an aggregate sense whether retirees are relying on a bunch of investment portfolios or on the government's taxing and benefits payments mechanisms, nor on whether a retirement system is "fully funded" or "pay as you go". [Note -- that's true in the aggregate, obviously not true re specific pension plans in which employees have direct property rights or which depend on the solvency of the employer at some future date. In that case, "fully funded" and "funded with what" (i.e. not employer stock) is rather crucial.] Although it's overly simplistic, it's helpful to recognize that, as an economy, retirees will receive funds either out of what US firms and individuals are producing at that time, or from claims we have on the production of foreigners. [Let's not get into the option of "selling down capital stock" rather than "current return on investment" or the effects of leverage etc that add lots of extra complications in timing and pricing of assets. Those assumptions may be quite relevant to designing the particulars of a given system, including especially tax incentives. But don't really change the basic argument.] Right now, the US is financing its accumulation of current account deficits , by increasing the claims of foreigners against our future production and reducing our own claims against future production of foreigners. Barnett, for example, points out that instead we ought to be sending investment capital to the developing countries, with their burgeoning youthful populations, because they're going to be producing for our retirements. Instead, it's operating in reverse. Rather perverse, eh? [It's for me the biggest gaping analytical weakness of Barnett -- he's willing to have our current account deficits financed indefinitely as part of financing the Leviathan function and somehow believes the "balance of terror" between China's production and US consumption is sustainable. At the same time his big picture Core-Gap analysis says the capital flows should be working in quite a different direction.] The one comfort we have is that our aging problem ain't nothin compared to what the Chinese are facing!