Trends | Theory | Facts | Food | Environment | Aging | Elderly | Immigration | Urbanization | Family | Women

Has Privatization of Social Security Worked in Latin America?

Jeffrey Olson

 

Introduction

Old age social security schemes have been around since the beginning of the century. Mandatory old-age insurance introduced in Germany and this model known as the "Bismarkian model" became very popular in Latin America. Argentina, Brazil, Chile, Cuba, and Uruguay were the first ones to introduce the public pension schemes in the 1920s and 1930s. (Muller, 2000, 508) Latin American countries, due to the perceived success of the partially privatized social security system introduced in Chile in 1981 have introduced compulsory individually fully-funded (IFF) plans that compete with, substitute, or complement the public pay-as-you-go (PAYG) plans. The privatization of old-age security is a radical change in retirement plans, going against the welfare state. Privatization of social security is a transition from collective to individual provision and a transition from the state to the market as the main suppliers of retirement pensions. So far, it is unclear whether privatization of social security has been successful across the board in Latin America.


This paper is divided into five sections. The first section will examine the problems with the old PAYG social security scheme. The second section will discuss Chile's transition to a new IFF private pension scheme. The third section will discuss the second generation Latin American social security reforms. The fourth section will explore the results so far of the different social security reforms. The final section will summarize the paper and take a look into the future of social security privatization.

Previous Public Social Security Schemes in Latin America

Public pension schemes in Latin America had many flaws. There was a weak contribution-benefit link with large entitlement conditions and replacement rates. Because of this, contribution rates and state subsidies were elevated while evasion and underreporting of income increased. (Muller, 2000, 508) In some countries, the existing old-age security were fragmented and consisted of multiple funds with varying legislation and management, benefits and contribution rates. There was inequality between the different groups of insured and administration. The informal sector workers were very vulnerable because they were basically excluded from the existing social insurance schemes. (Muller, 2000, 508) In the 1980s with the economic crisis and the stabilization and adjustment plans that followed there was a decrease in the public resources for social policy when they were greatly needed. (Muller, 2000, 508) In many countries, the real value of the pension benefits was destroyed by inflation. In some, the public social security institutions defaulted on their pension liabilities and failed to pay retirement benefits. (Muller, 2000, 508)


Chile was the first Latin American country to privatize social security. Before the privatization in 1981 under the Pinochet regime, Chile had more than 30 separate pay-as-you-go systems based on occupation. Each of these systems had different contribution rates, requirements for retirement, and benefit levels. The number of contributors was very low because of the high unemployment, the large quantity of informal employment, and the large amount of evasion of contributions. About 93 percent of pensioners were receiving the minimum benefit because the system was about to go bankrupt and could not pay the benefits to those that were eligible. In 1980, the PAYGO system was about 2.7 percent of the GDP of Chile. (Kritzer, 2000, 7)

The Case of Chile - IFF pension funds

Chile was the first Latin American country to privatize its pension system. In 1981, with neoliberal ideologies and the Pinochet dictatorship, the PAYG (pay-as-you-go) system was replaced by a multi-pillar system of private, IFF (Individually fully-funded) pension funds. In the Chilean system, the first tier is publicly run and pays out tax-financed social assistance pensions to a limited number of elderly people with less than 20 years of contribution (Muller, 2000, 509). The second tier consists of funded private pension funds (AFPs) that have mandatory contributions paid to them by Chilean employees and investing these funds. The third tier promotes voluntary savings for old age, where workers may contribute an additional 10 percent of their wages each month as a form of voluntary savings that are tax deductible.
This new private IFF system is required for new entrants to the labor market and the PAYG scheme was to be phased out. People who were insured by the old system could enter the new system or stay in the old, public one. Under the new IFF system, all workers automatically have 10 percent of their wages deposited by their employers each month in their own Pension Savings Account (PSA). This percentage only applies to the first $22,000 of annual income (Piñera, 1996, 157). Workers chose one of the private Pension Fund Administration companies to manage his Pension Savings Account. Each AFP operates the equivalent of a mutual fund with low-risk stocks and bonds. Workers are free to change from one AFP to another. (Piñera, 1996, 157)


Employees that switched to the new privatized system received government-mandated gross wage increases of 18 percent, which were about 11 percent net wage increases (Kritzer, 2000, 7). The government issued "recognition bonds" to cover the pension entitlements of the people switching from the old to the new system (Muller, 2000, 509). These recognition bonds were indexed for inflation and funded by general revenues (Kritzer, 2000, 7).
When retiring, a worker may chose between two general payout options. Retiring in this system has a different meaning; a covered worker can retire whenever as long as the worker is over the minimum amount and does not have to retire from the workforce to retire from the pension account (Piñera, 1996, 159). A retiree may use the capital in his PSA to purchase an annuity from any private insurance company, which would provide a constant monthly income for life that is indexed to inflation. The second option that a retiree has is leaving the funds in the PSA and making programmed withdrawals with limits based upon life expectancy of the retiree and dependents in the household (Piñera, 1996, 160). Pensions in Chile are protected from inflation because they are denominated in monetary units that are adjusted to inflation monthly by the consumer price index (Kritzer, 2000, 8). The amount of the pension is based on the individual's contribution plus interest minus the administration fees. Pensions are usually paid by the AFPs at the retirement age of 60 for women and 65 for men. In the pension plan, people that meet a minimum of 20 years, but do not receive the minimum pension level are entitled to the minimum pension from the state budget (Muller, 2000, 509). The PSA system also includes insurance against premature death and disability. Each AFP provides this service to its clients by taking out group life and disability coverage from private life insurance companies, which is an additional contribution of 2.9 percent of worker salary (Piñera, 1996, 159).


The PSA system's goal is to eliminate the unfunded pension liabilities that many countries face with the pay-as-you-go system. The PSA for Chile should allow for individual preferences to become individual decisions that will produce desired outcomes; whereas, the PAYG scheme did not allow for personal preferences (Piñera, 1996, 158). Supporters of the Chilean model say that a switch to a private, funded plan increases long-term saving, investment and economic growth. These supporters claim that restricting public involvement in old-age security reduces private spending (Muller, 2000, 509). Therefore, they are implying that private savings will increase. The Chilean model has spurred pension reform all over the world.

Second Generation Reforms

Private mandatory pension provision is now established in many Latin American countries, and although there are no identical replications, variations of the Chilean model have been implemented in Argentina, Bolivia, Colombia, El Salvador, Mexico, Peru and Uruguay (Muller, 2000, 509). They include an introduction of a mandatory private pension fund that competes with, substitutes or complements the PAYG plan.


Structural pension reforms in Latin America are divided into three main groups. The first group is the substitutive model, which is where the former public system is closed down and is replaced by the private IFF system. This reform was made in Chile, Bolivia, El Salvador and Mexico. The second type is the parallel model, which is where the private IFF system is an alternative to the public system and the two systems compete. This reform was made in Peru and Colombia. The third type is the mixed model, which is where the private IFF part complements a reformed public system. Examples are Argentina and Uruguay. The reforms in all countries, except Chile, were made in the midd-1990s. (Muller, 2000, 511)


Costa Rica introduced voluntary individual accounts in 1995 as supplements to the PAYG system, and in 1999 Cost Rica passed a law requiring individual accounts that would be funded with a percentage of the employer's severance pay contribution (Kritzer, 2000, 3).

 

Results of Latin American Social Security Privatization

There are several positive aspects of privatizing social security, especially for Chile. The full or partial shifts to funding (private IFF) have led to a quick increase in retirement capital in countries with coverage. Among younger age groups, acceptance of the private schemes is high (Muller, 2000, 511). In addition, the new systems also had high real rates of return (Muller, 2000, 512). However, the two decades of pension privatization in Latin America present some discouraging empirical lessons as well.


The high transition costs result in the transitional generations paying double because they have to pay for their own retirement plans and the current pension plan that goes to the elderly. In Chile, Colombia, Peru, and El Salvador, the government issued "recognition bonds" to cover acquired pension entitlements of those switching to the funded partially privatized system, which created strain on the public budget (Muller, 2000, 512). Argentina and Bolivia used compensatory pension arrangements put less cash flow pressure on the government budget (Muller, 2000, 512).


The Chilean model change did not result in reduced public spending or increased national savings in the short or medium run (Muller, 2000, 512). In the first ten years of the new Chilean system, there was a pension deficit due to the reform of 4.0 percent of GNP per year. This pension deficit was beyond the extra personal savings stemming from the pension reform, which was 2.4 percent of GNP per year. According to economic forecasts, the pension deficit due to reform will disappear in 2020, which is 40 years after the Chilean pension reform started (Muller, 2000, 512). Therefore, the transition between the old and new system, which is a very long process, creates economic losses due to such high transition costs. Chile was in a fiscal surplus when it introduced the pension reform, whereas countries that followed introduced reform in unfavorable circumstances. The Argentine case shows that governments use their benefits of a partial shift to private funding in order to cover their transition costs (Muller, 2000, 512).


The new pension fund had an oligopolistic structure. In Chile, Uruguay and Peru over two-thirds of the insured are affiliated with the three major funds. The administrative efficiency of the private pension fund schemes has been troublesome for every country that has made privatization reforms. The administrative costs of the Chilean funds are 15 to 30 percent of annual contributions due to excessive marketing costs (Muller, 2000, 512). These high costs have persisted in Chile almost twenty years after pension privatization was initiated.


The assumption that transition to a private IFF scheme would decrease contribution evasion has not been confirmed. There is a large gap between affiliates and contributors in all Latin American reform countries. This results in an increase in risk of old-age poverty. About 50 percent of account holders fail to regularly contribute to the pension funds, especially the ones with low incomes (Muller, 2000, 513). It is likely that lower-income workers will realize that they should just count on the minimum pension rather than what they will accumulate so they will only contribute enough to get the credit for the coverage. Because administrative fees are charged only on contributions, the workers that are making the contributions are paying to maintain the inactive accounts of those who are not making regular contributions. This will significantly lower their PSAs (Kritzer, 2000, 13).


Insufficient coverage is a problem that existed before the new private schemes and was not solved with the new Latin American privatization schemes. In Chile, at the most, 75 percent of the labor force is actually in the pension system (Myers, 1996, 212). In the less developed countries, such as Bolivia, Colombia, El Salvador and Peru, over three-fourths of the population were not covered by pre-reform social security, especially the poor receiving salaries informally and the people from rural areas. This low level of coverage is not going to increase with pension privatization because the labor market and affordability problems will remain unchanged (Muller, 2000, 513). The situation of the poor might actually become worse because the high transition costs involved in pension privatization could result in using public funds to consolidate the old-age scheme of a minority when these funds should be going to public assistance, primary health care, and other programs (Muller, 2000, 513).


Women usually receive smaller pension than men for several reasons. Women receive benefits based upon their own earnings, which are generally lower than the earnings of men. In addition, using gender-specific mortality rates to calculate life expectancy produces a lower annuity because women generally live longer than men. Therefore, in order for a woman to receive the same pension as a man with the same salary, she would have to retire later than the man (Kritzer, 2000, 13). As of March of 1999, about 50 percent of all pensioners under the privatized system had retired early, 86 percent of them being men (Kritzer, 2000, 3).
The new pension scheme was supposed to make future old-age benefits more secure, but the Latin American experience shows that the PAYG risks are traded for other risks in the transition from public to the funded scheme. In an IFF scheme, the amount of future old-age benefits is dependent on rates of return of the pension assets. The investment risk therefore is the burden of the insured. Inflation may decrease the real amount of accumulated pension capital. This risk cannot be insured in a private pension fund scheme. CPI-indexed securities could solve this problem, but they are not being used in any of the second-generation countries (Muller, 2000, 513). Chile is the only country that accounted for inflation through their adjusted currency.
Another risk with the new pension scheme is the lack of investment opportunity. Pension funds are supposed to invest accumulated pension capital on the local financial market in order to close the "savings gap" because people are believed to reduce their private spending and save more. Therefore, their investments are needed in the local financial market. In order to achieve this, restrictive portfolio regulation was implemented. Crises in South East Asia, Rusia and Brazil show that local capital markets, especially for developing countries, are sensitive to major financial market crashes around the world. The Mexican Tequila crisis for instance created losses in the Chilean pension funds in 1995. (Muller, 2000, 513)

Conclusion

In Chile, the system's planners assume that the 10 percent contributions will provide adequate pensions of between 50 and 70 percent of the final salary of the contributors. This theory is based entirely on the assumption that there will be a 7 percent real rate of return. There have been high rates of return up until now, but over the long run it might not be realistic to expect over 7 percent real rates of return (Myers, 1996, 213). Therefore, it is too early to say that the Chilean system is a great success. Those who have retired by 1999, rely mainly on recognition bonds as a large portion of their pension.


Except for Chile, the new private pension schemes in Latin America have been set up very recently. It is too early for a good evaluation of their performance. Questions about social security reform will remain unanswered until the first generation of retirees is able to use their IFF pensions. Early experience with the mandatory private schemes shows that private pension funds do not necessarily guarantee efficiency and competition. Even with privatization, there are problems of coverage, evasion, and payment delays. Pension privatization is an expensive strategy, both politically and fiscally, due to the transition costs that result from the shift to private funding. Because of the high transition costs, there is also no guarantee for a net growth in national savings. Therefore, privatization is not necessarily a success for old-age social security.

 

References

 

Kritzer, Barbara E. (2000), "Social Security Privatization in Latin America", Social
Security Bulletin
v63, n2 (2000): 17-37.

Edwards, Sebastian; Edwards, Alejandra Cox (2002), "Social Security Privatization and
Labor Markets: The Case of Chile", Economic Development and Cultural Change
v50, n3 (April 2002): 465-489.

Muller, Katharina (2000) "Pension Privatization in Latin America", Journal of
International Development v12, n4 (May 2000): 507-518.

Diamond, Peter and Robert J. Myers, (1996), "Social Security Reform in Chile: Two
Views" in Social Security What Role for the Future?, edited by Peter A.
Diamond, David C. Lindeman and Howard Young, National Academy of Social Insurance, Washington, D.C., 1996, Chapter 6, pp.209-224.

Jose' Pinera, "Empowering Workers: The Privatization of Social Security in Chile", Cato
Journal
, Vol. 15, Nos. 2-3, Fall-Winter 1995/1996, pp. 155-166.