[About F&D] [Write F&D] [Contents] [Back Issues] [Search] [Subscribe] [F&D Home]
A Reform Proposal for Costa Rica's Pension System
ASLI DEMIRGÜÇ-KUNT AND ANITA SCHWARZ
Costa Rica's public pension system, like those of many other countries, faces long-run financial problems. As its authorities consider reform, they should bear in mind that although a radical approach inevitably entails higher initial costs than a gradual approach, the benefits of the former may make it worth pursuing.
ANY developing countries are grappling with pension reform in
the face of growing pressures from aging populations. The typical
systems are defined benefit, pay-as-you-go schemes in which
contributions and benefits are weakly linked. (For background
information, see the three articles on pension reform that appeared
in the June 1995 issue of Finance & Development.)
Consequently, the schemes usually run into financial difficulties,
pose a large fiscal burden for future generations, and are typically
quite inequitable in their impact. The question facing policymakers
is how to reform the pension system and address the transitional
costs associated with such reform. Cost Rica provides a good example
of the problems of a traditional public pension scheme and the
policy options facing the government.
The present system
The Costa Rican Social Insurance Fund (CCSS, the main social security institution), was established in 1941 to provide compulsory social insurance coverage for employees on old-age, disability, and survivor pensions, as well as sickness and maternity benefits. CCSS covers 45 percent of the working population, insuring both public and private workers. The table summarizes the main features of the CCSS system.
In general, the current system is too costly to be sustainable in the long run. On the one hand, the contribution rate--a total of 7.5 percent of an employee's salary-- is quite low by international standards. Contribution rates in pension systems in Latin America and the Caribbean, for example, average 10.5 percent, and rates are even higher in the industrial countries, where they average 16.3 percent. On the other hand, targeted replacement rates (the percentage of a retired person's wage that is replaced by a pension) are very high. A worker with 40 years of service can expect a replacement rate of over 80 percent. This compares with an average replacement rate of 50 percent in industrial countries. Although inadequate inflation indexation is used to lower the actual replacement rates and ease the financial burden on the state, this practice significantly erodes retirees' living standards as they age. The retirement age, which was recently raised to 60 and 62 years for women and men, respectively, is also relatively low. Lax regulations have allowed individuals to use disability pensions as a very generous early retirement system; one-third of pension recipients claim disabilities compared with less than 10 percent in most other countries.
| ||||||||||||||||||||||||||||||
Despite the low contribution rate, the CCSS suffers from extensive evasion, since the weak link between contributions and benefits leads contributors to view the system as a tax on labor. Private sector employees, especially, evade the system either by not contributing or by underreporting their incomes during most of their career. Since pensions are based on salaries during the last five years before retirement, individuals report much higher incomes in these years to qualify for the maximum pension. Furthermore, the poor investment performance of the CCSS' reserves worsens the funds' financial problems--real returns on pension investments have been either low or negative, shrinking the fund in real terms.
In Costa Rica, as in most countries, the long-run prospects for financial viability are dominated by demographics. Owing to the longer life expectancies and lower birth rates that accompany higher incomes, the ratio of the number of people of retirement age to the number of people of working age increases over time. In 1990, there were nine individuals of working age for every individual of retirement age, suggesting that if each of the former contributed 7.5 percent of his or her average wage, the resulting contributions could support a pension worth about 66 percent of the average wage for each retiree. However, by 2070, there will be only two workers per retiree, which would mean that the same 7.5 percent contribution would be able to support a pension worth only 14 percent of each retiree's average wage.
This demographic change clearly will translate into higher contribution rates, lower replacement rates, or budget deficits for the government. The cost to the government if both the contribution rate and the benefit rate remain unchanged would rise from 2 percent of GDP in 1992 to 32 percent of GDP by the year 2070.
Options for reform
There are two major reform paths Costa Rica's pension system might take. The first is a limited reform, which would involve keeping the present system intact but making minor changes to make it both more financially sustainable in the near term and more equitable. The second is a more radical reform that would involve restructuring the system to create a multipillar old-age-security system.
Limited reform would attempt to make the system more financially sustainable by raising the normal retirement age, raising the contribution rate, decreasing benefits, or better investing the system's reserves. It should be noted, however, that all of these reforms oriented toward maintaining financial solvency are likely to be successful only in the short-to-medium term and are not likely to provide long-term security.
Radical reform (essentially the model described in the pension reform articles that appeared in the June 1995 issue of Finance & Development) would involve instituting a mandatory fully funded, defined-contribution pillar, through which individuals could save to provide income for when they are retired or unable to work. This pillar would be supplemented by a public pension system, much smaller than the current one, that would provide a social safety net. Finally, the third pillar would consist of optional but funded company-based or occupational pension schemes and voluntary savings in the form of bank deposits, marketable securities, and investments in real assets.
The costs and characteristics of the two strategies depend on the specific components chosen. While limited reform could consist of raising the retirement age, lowering benefits, or raising contribution rates, we assume that only the retirement age is raised--to age 65--in our judgment, the least politically sensitive option in Costa Rica (Chart 1). Radical reform can include more options. The proposal considered here is fairly expensive. We assume that all individuals aged 65 and above would receive a pension worth 50 percent of the minimum wage, or about 25 percent of the average wage, funded from general government revenues; and individuals and employers would contribute a combined 7.5 percent of earnings to a mandatory savings pillar.
The limited reform would save money, but would still entail relatively high costs over the long term. The universal pension to be provided under the radical reform would also entail rising costs owing to the aging of the population. However, these costs could be lowered by restricting pensions to those who had contributed to the pension system or only to those whose incomes fell below the poverty line. At first, the multipillar system pays less because retirees will have accumulated savings for only 10 years. But over the longer term, the radical reform results in higher benefits than either the current system or the limited reform plan (Chart 2).
The transition
In any pension reform, the design and management of the transition is a crucial issue, with potentially large consequences for the stability, equity, and political acceptability of the reform. In Costa Rica, an immediate change to a new pension system, particularly a pension system which is substantially different from the existing one, would result in gross unfairness to some current workers and jeopardize the political acceptability of the reform. An appropriate regulatory framework would also have to be put in place. As with the reform options, there is a wide choice of transition options with considerable variance in costs.
One possibility would be to calculate what the government already owes each current contributor and pensioner, and then have the government transfer that sum to the respective individuals. Another option, which would be more costly, is for the government to allow all individuals in the existing plan to remain in that plan with the current level of contributions and benefits, but mandate that all new contributors join the new plan. This option (see lower panel of Chart 1) is more costly since workers would continue to accumulate service credits for benefits that were not covered by their contributions. Many other options exist, including allowing the old plan to operate for those over a particular age but mandating that all younger workers transfer to the new pension system. While the cost of radical reform with the high-cost transition option built in is higher than the cost of limited reform at first, the future pension cost for new entrants to the labor force will be substantially lower under radical reform. Choosing a less costly transition option would, of course, substantially lower total costs under radical reform.
The radical-reform solution we are proposing results in lower costs and higher benefits than our limited reform proposal in the long term, thus providing both the government and covered employees with a win-win solution. In the medium term, however, transition costs may result in higher total costs under radical reform.
Conclusion
Should transition costs deter countries from choosing the best long-term strategy? While the answer may depend on each country's particular circumstances, there are additional long-term advantages to pursuing radical reform that are hard to quantify but may be sufficient to outweigh the medium-term disadvantages:
- More poverty alleviation is achieved than under the original
pension system, which provided relief only to aging workers in the
formal sector. Furthermore, tax revenues taken from the whole
population will provide benefits to the whole population rather than
to a select group. This improves equity as well as providing a more
efficient social safety net.
- Incentives for evasion and for decreased labor force
participation are reduced--those who do not participate in formal
labor markets and who do not contribute or who contribute very
little, receive low pensions.
- Pay-as-you-go schemes in countries likely to experience major
demographic changes require sharply rising payroll taxes to remain
solvent, all things being equal, which undermines employment in the
formal sector and labor competitiveness. Arresting this dynamic
through pension reform should increase both employment and the
productivity of labor.
- Capital market development is strengthened, making long-term capital available for both public and private sector investment. In addition, improved incentives for better management of pension funds can raise returns, and hence workers' benefits, significantly.
Thus, despite the initial high costs of radical reform, the
benefits that can be derived--long-term fiscal sustainability,
better pension benefits, an improved income distribution, fewer
labor market distortions, and capital market growth--make this
approach well worth exploring.
|
Asli Demirgüç-Kunt,
a Turkish national, is a Senior Economist in the Finance and Private Sector Development Division of the World Bank's Policy Research Department. |
Anita Schwarz,
a US national, is a Human Resources Economist with the Labor Markets, Social Insurance, and Public Sector Management Group of the World Bank's Poverty and Social Policy Department. |
Authors' note: After we submitted this article to Finance & Development, the Costa Rican government embarked on a reform strategy similar to the radical reform proposal the article describes. In place of a universal pension, the authorities are proposing to set up a contributory pension for the first pillar and a smaller mandatory savings pillar, which is to be instituted once an appropriate regulatory framework is in place. These two pillars would be complemented by additional contributions made by individual workers on a voluntary basis.
This article is derived from the authors' "Costa Rican Pension System: Options for Reform," World Bank Policy Research Working Paper No. 1483, June 1995 (Washington).
[About F&D] [Write F&D] [Contents] [Back Issues] [Search] [Subscribe] [F&D Home]