As part of wider efforts at social-security reform over the past 25 years, most governments in Latin America have endeavored to reduce the financial burden of paying pensions. Chile was the pioneer, introducing mandatory individual savings accounts in the early 1980s; in later years, Argentina, Bolivia, Colombia, Ecuador, Mexico, Peru, and Uruguay followed suit. The measures have been closely watched in the developed world, where aging populations are straining publicly funded pension systems.

The switch to private funding has brought Latin America significant benefits: lower public-sector liabilities in the long term, more robust capital markets, and improved transparency for individual benefits. However, two important challenges to the system have emerged, both requiring urgent attention.

First, income-replacement rates—which determine the amount of money an individual can expect at retirement—have been much lower than desired because of high fund-management commissions, decreasing government-bond yields, and lapses in contributions. As a result, many workers who relied on this savings mechanism have been disappointed.

Regulators have sought to address this problem by injecting new competition into the sector. But unexpectedly, this move resulted in added costs, as financial institutions built up their sales forces, and new entrants emerged with only limited experience in asset management. To offset the risk associated with limited expertise, regulators imposed excessively stringent controls on the asset classes for pension investments, so returns and expected replacement rates have fallen further.

The second issue is that the new system, like the old one, excludes large numbers of people, particularly the self-employed and informal workers, who fall outside the official social-security net. In some countries, this segment represents up to 55 percent of the economically active population.

These two issues are at the forefront of pension reform in Latin America today. In Chile, where it all got started, formal workers reaching retirement age have observed replacement rates as low as 40 percent. Last year Chile’s new government, under President Michelle Bachelet, embarked on a further set of reforms in an effort to plug such gaps, make the pensions industry more competitive, and increase replacement rates.

Other countries in the region need to consider making similar adjustments to their pension systems. We believe that a more stable and rewarding pension regime can be created by linking incentives for fund managers to the performance of investments, easing the current restrictions on investment classes, and imposing risk-management best practices on the industry. In addition, the reach of pension plans can be extended through the more active promotion of voluntary savings.

Transition challenges

The Chilean government first presented an alternative pension model in 1981; it required each person to make contributions to individual savings accounts, managed by one of several private institutions, known as AFPs¹ across the region (or as Afores² in Mexico). Under the system, widely copied by other countries (but with important local variations), workers save for their own retirement, the accumulated sum is converted into an annuity at the end of their working lives, and the old government-administered pension pool account eventually disappears.

This defined-contribution system—so named because the factor known at the outset is the level of contributions rather than the benefits paid on retirement—has three main advantages: transparency, security, and the avoidance of government subsidies.

Transparency comes not only from workers knowing exactly how much they have contributed (the balance in an individual account includes the interest earned less the commissions paid in each period) but also from their freedom to choose the fund manager and, most importantly, from the security of having retirement funds outside the influence of changing government policies.

Public subsidies are required under the plans only if the final balance proves insufficient to guarantee a minimum pension. Under the old pay-as-you-go system, governments had to use the federal budget to make up the funding shortfall. When they couldn’t afford to do so, they had to cut benefits or increase the retirement age, as Costa Rica did in 1995.

While all countries embracing the new system have seen their short-term liabilities increase, defined-contribution systems will reduce the public-sector burden in the long run. In Mexico, for example, the pension deficit had been projected to grow rapidly through 2050, but with the new system the deficit is expected to peak in 2020 and then gradually decline to below its 2001 level (Exhibit 1). The initial increase occurs because, in the short term, governments do not get funds from the economically active population (those who have transferred their contributions from the government pool to their own individual savings accounts) but still have to pay for the pensions of those who have already retired. Eventually, as these retirees become a smaller share of the total, the deficit begins to decline. Chile’s experience is instructive: a pension deficit that in the late 1970s was projected to reach 20 percent of GDP in 2000 in fact fell below 4 percent in that year and continues to decrease.

A further advantage is the way the defined-contribution system can strengthen national savings and capital markets. The assets managed under Chile’s system, for example, are equivalent to 50 percent of the country’s GDP and represent more than half of its financial assets.

Despite these advantages, two major challenges remain. While the gross investment returns that funds achieved in some countries have been impressive—an annual average of 10.29 percent in Chile from 1982 to 2005—the big failing of the region’s defined-contribution system has been inadequate income-replacement rates for the retired population. In Chile, as the cost of acquiring customers and running back-office operations grew, fund-management fees and commissions eroded one-fifth of the potential capital accumulation of savings over the 1982-2005 period. In Mexico up-front commissions can amount to 20 percent of an individual’s contributions, with annual management fees two or three times those that a voluntary mutual-fund equivalent in the United States would charge.

In the early years of the new pension systems, governments tended to offer incentives to AFPs and Afores, trying to make them profitable in hopes of attracting skills and capital to the fund-management business. More recently, as income-replacement rates for retirees have become an immediate social and political priority, some regulators have capped fee increases, fostered price wars among market participants, and sought to eliminate barriers to entry. Mexico’s regulator has approved the entry of eight new Afores over the past five years and eased earlier restrictions on the right to switch between funds, especially if commissions are lower.

Governments have also tried to limit risk by controlling the asset classes in which fund managers may invest, as well as the total value of the fund at risk. The rules vary significantly by country, but in practice they require fund managers to invest heavily in government securities (80 percent in the case of Mexico). Such curbs make it more difficult for AFPs and Afores to provide higher returns to savers—ultimately the best way to improve income-replacement rates.

The other challenge—how to widen the scope of the new system—has yet to be tackled seriously. Recent reforms may encourage unaffiliated workers to start contributing voluntarily to their retirement, but the region’s policy makers continue to seek new incentives for those outside the formal sector. Colombia, for example, has promoted voluntary pensions through tax benefits and by requiring users of the public-health system to have an account in an AFP.

The future for pension plans

Governments across the region should consider action in three areas: fee structures, investment restrictions and risk management, and efforts to reach the excluded. They could give income-replacement levels a much-needed boost by realigning fee structures so that the profitability of AFPs and Afores is linked more closely to (risk-adjusted) returns than to up-front fees. However, the primary influence on pension fund replacement rates is investment returns (Exhibit 2). To improve them, governments should widen the list of asset classes in which financial groups can invest and adopt a more nuanced approach to risk. Instead of focusing on government securities, fund managers should be encouraged to explore options such as mortgage-backed securities and corporate bonds, which offer higher yields and low to medium levels of risk. Regulators should concentrate on monitoring the risk-management practices of fund managers—a change that would further shift the basis of competition in the direction of technical skill. Combined with a reward system emphasizing the rate of returns rather than up-front fees, these changes would force private institutions to improve their operational efficiency, hire better fund managers, and offer more generous yields to individual savers. In general, regulatory measures to promote competition should avoid incentives that encourage fund-management groups to invest in big sales forces and other operating costs.

Governments must promote voluntary pension contributions in more imaginative and determined ways to attract increasing numbers of workers from the informal sector into the new system. Encouragement should be given to those who might be inclined to make contributions in nontraditional settings (such as supermarkets) or by direct charge to a credit card, perhaps with the added inducement of lower fees or tax incentives. Such alternatives should attract higher-income people who normally rely on other investment vehicles for their retirement needs but could also become retirement-saving channels for low-income workers not currently covered by the social-security system. Our experience suggests that simplicity, clearly articulated benefits, and creative promotions (for example, raffles with tempting pension-related prizes) are needed to encourage nonparticipants.

Pension reform in Latin America has so far created a more transparent system with the promise—already fulfilled in Chile—of a reduced financial burden for governments in the future. However, the most urgent challenge is to increase replacement rates. We believe that there are ways to do so while strengthening the technical skills of fund managers and reducing overall frictional costs, in the process creating a more dynamic and performance-oriented investment-management sector. In addition, innovative strategies to attract unaffiliated people are needed to bring the system’s benefits to a larger share of the population.