Pension Reform

AuthorAxel Schimmelpfennig
Pages5-6

Page 5

Under PAYG, each working generation pays for the retired generation's pensions. Typically, PAYG systems pay defined-benefit pensions with the payroll tax set at a level that finances current pensions. In a funded system, each working generation accumulates savings to provide for its own retirement. Funded pensions are typically paid on a defined-contribution basis, that is, they are determined by the rate of return.

As the dependency ratio rises, working generations will face dramatic increases in payroll taxes under existing PAYG systems, unless governments run large deficits. For example, even if the payroll tax in Japan was increased from 17 1/2 percent to 30 percent over the next 50 years, government transfers to the social security system would still double by 2050. 1 Raising payroll taxes by this amount will be politically difficult. Moreover, raising payroll taxes when the demographic shock hits may negatively affect growth by reducing investment, labor force participation, and productivity growth. 2

Other parameters besides the payroll tax can be adjusted to strengthen PAYG systems. Pension outlays can be lowered by reducing replacement rates, that is, the ratio of pensions to wages, or by moving toward less generous pension indexation formulas that give less weight to wages and more weight to inflation. Raising the retirement age in line with life expectancy would mitigate the effects of the demographic shock. In addition, the payroll tax could be raised by a small amount immediately (prefunding) to accumulate savings that would be used once the demographic shock hits. A combination of these measures, if undertaken early enough, could be sufficient to make existing systems sustainable. 3

An alternative reform strategy is to replace the PAYG systems with funded systems. The latter are often thought to be superior to PAYG systems mainly because they are more conducive to growth. First, funding leads to increased capital accumulation because it raises national saving. Second, funding increases labor supply because the payroll tax is directly linked to benefits and is thus less distortionary than a PAYG payroll tax, which is usually only weakly linked to benefits. However, replacing PAYG systems with funded systems comes at significant transitional costs. The reform generation needs to provide for current pensioners through the PAYG system while also saving for its own retirement. Whether such a reform can help solve the demographic problem is, therefore, an empirical question. For the United States, Kotlikoff, Smetters, and Walliser (1999) argue that the introduction of mandatory funding would improve the position of future generations substantially, while leaving the reform generation only slightly worse off. 4 In related work, these authors show that transition costs are lower if privatization is voluntary, and, focusing on intragenerational redistribution, privatization can be designed to pose a higher share of the transition costs on those with high lifetime incomes. 5 Cubeddu (2000) finds that the U.S. social security system redistributes in favor of females and noncollege graduates; privatizing social security would reduce this redistributive effect. 6

The claimed advantages of funding are carefully assessed in a number of studies. Brooks (2000) shows that funding also suffers from the adverse effects of a demographic shock because asset prices fall as their relative supply-which is a function of the dependency ratio-increases. 7 Barr (2000) casts doubts on the link between funding and growth because transition costs of moving from PAYG to funding would largely offset any increase in saving. Moreover, disincentives in ex-Page 6isting PAYG systems could be removed through design changes. 8 Finally, Valdivia (1999) shows that the welfare gains from longevity insurance provided by PAYG-benefits are paid until death, independent of contributions-outweigh the advantages of funding without such insurance. 9 However, longevity insurance can also be provided in funded systems through annuities. 10

A critical question is how national saving would react to the introduction of a funded system. Dayal-Gulati and Thimann (1997) find that social security expenditure...

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT