Privatization of Pensions

In the Fehr, et al paper, “Will China Eat Our Lunch or Take us to Dinner?,” the authors discuss various ways for the government to finance the welfare programs promised to the retiring baby boomers. One analysis looked at the privatization of pensions, modeled by “the elimination of any new public pension benefit accrual coupled with the establishment of individual accounts.” Basically, this privatization of pensions relies heavily on the effectiveness of free markets, as well as an “intergenerational redistribution.” The privatization of pensions would place the burden on the current elderly, as they would be “forced to pay for their benefits via the consumption tax.” Instead, “younger and future generations benefit enormously” because their wages are much higher without the payroll tax burden. While the Fehr papers notes that “the poor initially experience larger welfare losses,” Allan Meltzer thinks differently.

In his paper, “Privatizing Social Security,” he claims that “private accounts are a much greater benefit for the poor than for the rich.” Because most of the poor do not have private pensions, they would “gain the most from the opportunity to get higher returns, if part of social security is privatized.” Meltzer advocates for complete privatization of social security, leaving the government no room to manage any of the redistribution. Meltzer notes that “the appeal of privatization…is that it avoids some of the need for higher tax rates or lower benefits to balance future accounts.” He places full confidence that the free market economy would most efficiently deal with social security. Of course, there is no mention in either of these papers on how a damaged market economy, in the midst of a slow recovery from a considerable recession, could efficiently handle the privatization of social security. Would US citizens be willing to trust that their future benefits will be safe in the hands of the volatile financial market?

http://repository.cmu.edu/cgi/viewcontent.cgi?article=2052&context=tepper

the prof

Meltzer apparently has not bothered to look at the rather large literature on private returns, which does not find that private pensions systematically outperform social security systems. The old rule of thumb in the US was that if you could wait 7 years, stocks always proved superior. But that no longer holds, in effect economists cherry-picked data (postwar US) to bolster their case, though not necessarily intentionally because (i) US-based economists didn’t study the rest of the world and (ii) datasets were more limited.

It’s now clear that private savings face very substantial risk, even over a much longer working-life time horizon. If you can time when you were born and hence when you retire you can come out well. If you chose the wrong age, you’re screwed. Such risks can only be insured on a social basis, not on a private basis — but it turns out that when done at a societal level such insurance only appears expensive (remember it’s an opportunity cost definition!) when you have a period of sustained gains on the stock market.

Finally all of this glosses over how you handle the transition. So you are absolutely right to question Metlzer’s claim, and to turn to Fehr et al for support where they explicitly model what happens to different groups when you impose large changes on the system.

That said, it’s not clear why privatizing does well on some dimensions in the Fehr et al. model. There are basically two margins of adjustment, gains from higher savings (which in the model always generates higher investment and hence more capital per worker), and gains from reactions to shifts in tax rates that boost labor force participation and hence GDP and hence savings. I am skeptical that real-world elasticities are sufficiently strong to produce large gains, particularly in the US because (political diatribes aside) we have unusually low tax rates and we don’t see systematically worse outcomes in other economies. (Greece may be different!) Behavior changes, but not enough to provide a big windfall.

I also fail to see how privatization will lead to more savings which (in the short run) must mean lower consumption. But back to the main model issue: these sorts of long-run models have a Solow growth process at their core, which leads to odd interactions (or at least odd to me). So I need to go through that part of the model much more carefully to try to discern why they get the big gain; they don’t provide a cogent explanation.

November 13, 2012

One Comment

the prof

Meltzer apparently has not bothered to look at the rather large literature on private returns, which does not find that private pensions systematically outperform social security systems. The old rule of thumb in the US was that if you could wait 7 years, stocks always proved superior. But that no longer holds, in effect economists cherry-picked data (postwar US) to bolster their case, though not necessarily intentionally because (i) US-based economists didn’t study the rest of the world and (ii) datasets were more limited.

It’s now clear that private savings face very substantial risk, even over a much longer working-life time horizon. If you can time when you were born and hence when you retire you can come out well. If you chose the wrong age, you’re screwed. Such risks can only be insured on a social basis, not on a private basis — but it turns out that when done at a societal level such insurance only appears expensive (remember it’s an opportunity cost definition!) when you have a period of sustained gains on the stock market.

Finally all of this glosses over how you handle the transition. So you are absolutely right to question Metlzer’s claim, and to turn to Fehr et al for support where they explicitly model what happens to different groups when you impose large changes on the system.

That said, it’s not clear why privatizing does well on some dimensions in the Fehr et al. model. There are basically two margins of adjustment, gains from higher savings (which in the model always generates higher investment and hence more capital per worker), and gains from reactions to shifts in tax rates that boost labor force participation and hence GDP and hence savings. I am skeptical that real-world elasticities are sufficiently strong to produce large gains, particularly in the US because (political diatribes aside) we have unusually low tax rates and we don’t see systematically worse outcomes in other economies. (Greece may be different!) Behavior changes, but not enough to provide a big windfall.

I also fail to see how privatization will lead to more savings which (in the short run) must mean lower consumption. But back to the main model issue: these sorts of long-run models have a Solow growth process at their core, which leads to odd interactions (or at least odd to me). So I need to go through that part of the model much more carefully to try to discern why they get the big gain; they don’t provide a cogent explanation.

November 13, 2012

Comments are closed.

Privatization of Pensions

Share this:

One Comment