The future of public-pension systems: An interview with the president of the Chilean Pension Reform Commission

11 05 2007

The chairman of a commission set up to investigate Chile’s private-pension system explains that the goal now is to supplement rather than replace it.

Web exclusive, April 2007

Chile’s pioneering pension reform attracted keen interest in Latin America and beyond when first introduced, in 1981. But a generation later, the much lauded obligatory private-pension scheme is showing signs of age. Projections have demonstrated that returns from the system will likely provide disappointing incomes for many participants. A large number of others—notably the poor and the self-employed—will receive no benefits at all.

President Michelle Bachelet, elected in early 2006, made this unexpected midlife crisis a major priority of her government and quickly appointed a special commission to investigate. After receiving its report, last August, her ministers included most of the wide-ranging recommendations in a bill currently before Congress.

The commission—chaired by Mario Marcel, an economist who was the budget director in the administration of Bachelet’s predecessor, Ricardo Lagos—played a crucial role in devising solutions to Chile’s pension problems. Rafael Rofman, the World Bank’s lead social-protection specialist for Latin America and the Caribbean, says that other countries in the region will pay close attention.

McKinsey’s Tim Dickson and Marcelo Larraguibel talked with Professor Marcel in the Santiago office of his private consultancy, Politeia Public Solutions, about the pressure for renewed change, the likely impact of his proposals, and Chile’s lessons for the outside world.

The Quarterly: How would you describe the pension outlook in Latin America as a whole?

Mario Marcel: The situation in Latin America is very heterogeneous, with different types of pension regimes and different demographic profiles. Roughly half of the region has adopted some kind of individual-capitalization component, in some cases replacing pay-as-you-go systems, as with the Chilean one; in others as part of “multipillar” schemes, as in Mexico; or in a complex cohabitation with defined-benefit regimes—for example, Colombia. Chile’s reforms came first, in 1981, but even so the first full generation to contribute throughout their working lives will not retire until 2025. So we are still in the middle of the transition. Those countries that introduced their reforms later, such as Mexico and Uruguay, are at a much earlier stage of evolution.

The challenges across the region are also different. Brazil’s unfunded state system is creating major fiscal problems for that country, partly because the reforms promoted by President Lula1 in his first term have not been implemented. At the other end of the spectrum, low coverage is the issue, with just 12 to 20 percent of the population receiving any sort of pension in some places. Those countries in the middle, which introduced individual capitalization, like Chile, are grappling with high fees and limited competition between pension fund administrators. A major exception is Mexico, which changed its regulations six years ago, increasing the number of administrators, including banks, and putting more competitive pressure on fees.

The Quarterly: What were the reasons for setting up your commission and for the subsequent bill now before Chile’s Congress?

Mario Marcel: The reform of 1981, which happened in the middle of a dictatorship, was certainly very revolutionary in public-policy terms. There were several clear benefits, including, from a macroeconomic perspective, the deepening of domestic financial markets. And because of these benefits, discussion of change was forbidden territory for many years. The industry was very defensive.

The Governments of Concertación, the center-left coalition that was elected after 1990, had mixed attitudes. They did not feel completely comfortable with the new system but feared that any debate might undermine trust in Chilean economic policy. Lack of hard information was also a factor, and every time someone made a projection it tended to be criticized as speculative or based on unrealistic assumptions.

However, the administration of President Ricardo Lagos supported some very important research that both collected data from the files of the pension funds and carried out a survey of individual contributors. That enabled us to make some reasonably accurate forecasts of what would happen when pension reform matures, 20 years from now. The projections were much less optimistic than people had thought they would be, especially at the outset. In 1981 the promise was that the system would generate pensions very close to final salaries and that self-employed workers would be encouraged to contribute.

The Quarterly: So what did that research reveal? How serious was the problem?

Mario Marcel: The picture was much more mixed. We found out that people who made regular contributions over 40 to 45 years, notwithstanding some interruptions for unemployment, would get pensions close to their final salaries. But such cases were not representative of the bulk of contributors. Since the early 1980s, around one-quarter of the labor force has been self-employed, other people have come in and out of work, and some have withdrawn from the labor market for family and other reasons. Many people now have part-time or seasonal jobs.

The self-employed are not just people who have worked on their own all their working lives. Often they move between independent work, contract work, unemployment, seasonal work, and a full-time job. There are many different arrangements. On average, therefore, we found that the generation due to retire between 2020 and 2025 was likely to receive a pension of about 44 percent of their final salaries, with a wide variation—from close to 100 percent, at one extreme, to nearly zero, at the other. We also found that nearly half of the elderly in the future would be receiving pensions equal to, or lower than, the minimum pension, a guaranteed income which is only available if you have accumulated contributions over a 20-year period. Many workers are only likely to have contributed for 5, 10, or 15 years and so may end up with as little as $50 a month.

The Quarterly: So is that what prompted the president to act?

Mario Marcel: While she was a candidate in the last election, in early 2006, Michelle Bachelet announced that if she was successful, pension reform would be one of her priorities. That created quite a stir given that, as I told you, it had almost been a taboo subject. She didn’t spell out what kind of reform would take place but committed herself to creating a commission that would diagnose the problem and set out proposals for improving the system.

The Quarterly: What were the main recommendations of the commission?

Mario Marcel: The fundamental change was to propose a new way of looking at pensions, not as an extension of people’s working lives, but as a means of sustaining the elderly and preventing poverty in old age. In Chile, as in other countries, the pension system has been based on the notion of a male breadwinner working full-time throughout his life to provide something for himself and his family when he retires. But that image is just not representative of the labor force as a whole.

We established certain principles—that the system’s objective should be to provide reasonable replacement rates of, say, 70 to 80 percent, that everyone should be guaranteed an income above the poverty line in old age, and that discrimination between different sets of workers should be avoided, notably between men and women.

With these principles in mind, we made 72 separate proposals aimed at constructing a system with three regimes. One of them is a strengthened version of the individual-capitalization scheme introduced in 1981. Second, there is what we call a solidarity pillar that replaces and better integrates some of the subsidiary benefits provided by the state. And third, there is a voluntary pillar for people who want to contribute more than what is compulsory.

The Quarterly: Can you say more about the new solidarity pillar?

Mario Marcel: What we had were two government-funded benefits—one the minimum-pension guarantee that nevertheless required a 20-year contribution record under the individual-capitalization scheme, the other known as the Welfare Pension, for those below the poverty line who do not have any self-funded arrangements at all. Projections told us that only 2 percent of the elderly would be able to benefit from the minimum-pension guarantee, while to some extent these benefits undermined incentives to contribute voluntarily to the private pillar after 20 years.

What we propose is a solidarity pillar that guarantees about $150 a month for everyone in the poorest 60 percent of households. Support from the solidarity pillar is compatible with a self-financed pension from the individual-capitalization regime and should be reduced only gradually as the self-financed pension increases. The aim is to provide additional support for those unable to accumulate a substantial fund through contributions, thereby preventing poverty in old age, but to avoid weakening the incentives to contribute. We meet this aim by ensuring that the higher the savings made by individuals during their working years, the larger their total pension, as funded by both pillars.

The Quarterly: And what about plans to encourage voluntary contributions?

Mario Marcel: Our proposals create some new tax incentives underpinning agreements between workers and their employers for voluntary contributions, but always involving specialist pension funds or other qualified asset managers. We do not, for example, allow self-managed pensions as in Britain and the United States. And of course we aim to strengthen the individual-capitalization system itself by bringing in more competition to lower fees, widening the asset classes in which pension funds can invest, and expanding the possibility of investing abroad. Today there is a ceiling of 30 percent on foreign investments by pension funds.

The Quarterly: Will you strengthen the private-sector component mainly by bringing in banks as fund administrators for the first time?

Mario Marcel: We should distinguish here between supply measures and demand measures. On the demand side, we propose to separate some of the functions that currently have to be performed by pension funds, such as collecting contributions, pursuing and prosecuting people who try to avoid contributing, managing branch offices, and providing information. What has happened is that all these “platform” services had a fixed-cost component and created barriers to entry by requiring large economies of scale—at least 1,000,000 contributors to be profitable. By separating them out so that pension funds concentrate on investing and the other functions are contracted out to an administrator, it will be possible to establish a fund with, say, 200,000 contributors.

We have also proposed an auction mechanism under which the pension funds offering the lowest fees would earn the right to affiliate all new entrants over a 12-month period. The individuals would then be forced to stay with that fund for up to 18 months, after which they could move elsewhere if they wanted to. The idea here is to save on marketing, which currently represents about 25 percent of total costs.

The Quarterly: So it’s a winner-takes-all system that is repeated every year?

Mario Marcel: Exactly. It’s very similar to the mechanism that Mexico has used in the last five years. It will not only put more pressure on fees but should facilitate the entry of new competitors. Rather than going into an expensive marketing war with other funds, a new player can participate in the auction before going through all the quite costly procedures to become an AFP.2 If it wins, there is time for all the necessary due diligence and formality; if it loses, it has incurred little expense and there is nothing more to do until the next auction, a year later.

The Quarterly: And what about the measures on the supply side?

Mario Marcel: Here the proposal in the bill is not so much that of the commission but something the government introduced when submitting the bill to Congress. Banks will be allowed to create pension funds—but, importantly, they must be dedicated only to managing pensions. When the commission analyzed this possibility, we were concerned about the risk of conflicts of interest, notably the cross-selling of some other financial products that might weaken the pension product. While opting to allow banking groups to create pension fund subsidiaries, the government also considered the caveats of the commission and included some regulations, in the bill submitted to Congress, to prevent these risks.

We believe that we can bring down fees through new competition but not at the expense of profitability. That’s why there are a lot of proposals to spread the portfolio of eligible investments and to generate more room for competition.

The Quarterly: Do you think there are major implementation issues ahead? How, for example, do you think the solidarity pillar will be financed?

Mario Marcel: The financing of the solidarity pillar should not be that difficult given where Chile is right now. The 1981 reforms were certainly very expensive because the government still had to pay out pensions from the state fund for many years while losing contributions to the individual-capitalization scheme. But the pension deficit is now going down, and there will be built-in savings for the government over the next 20 years. Those savings can be redirected to finance the solidarity pillar. I would say that discussion of the reforms will take about one year in Congress, while issues like the institutional rearrangements and the selection of the beneficiaries of the solidarity pillar could take another three. The president has emphasized that she wants the government to start paying some benefits under the new system before the end of her administration.

The Quarterly: What is the lesson of Chile for the rest of the world? Do recent events suggest that the Chilean model was not as good as everyone thought?

Mario Marcel: The mistake Chile made was to put too much emphasis on the individual-capitalization regime introduced in 1981. Doing so created problems for those unable to contribute. What we are doing now is not throwing out what we had but reinventing it as one component of a wider pension system. Individual capitalization is a very powerful accumulation mechanism, and that’s very important given the demographic trends in many countries. But the lesson from Chile is that it works better when supplemented by, rather than replacing, defined-benefit regimes. Profitability requires contributions—a basis from which to accumulate—which is not always possible where there is a large informal sector or substantial numbers of self-employed. Recent proposals in the UK, Sweden, and parts of Eastern Europe suggest that other governments acknowledge this too.

The Quarterly: Looking ahead, what will be the biggest challenge of making the new system work?

Mario Marcel: Coverage will be one issue, notably the series of proposals to increase the participation of the self-employed. In Chile, unlike many other countries in Latin America, most self-employed workers are not part of the so-called informal economy. This is not a “lost” sector. They fill out tax returns, pay taxes, and are subject to regulation. They require municipal permits to do their work. For these reasons, it should be possible to make such people contribute to their pensions, provided there are more user-friendly mechanisms, there are additional benefits not provided by the social-security system today, and efforts are made to educate and inform during a transitional period, before things are made compulsory. At some point, contributions by the self-employed will have to be compulsory—only in this way will we eliminate this asymmetry between them and other workers. I am hopeful, though. The banking industry has developed special credit card and mortgage products aimed at the self-employed, so it should be able to do the same for the pensions system.

About the Authors

Tim Dickson is a member of the board of editors of The McKinsey Quarterly, and Marcelo Larraguibel is a director in McKinsey’s Santiago office.


1 Luiz Inácio Lula da Silva.

2 Pension fund administrator (Administradores de Fondos de Pensiones).





2 responses

14 12 2012

Why is this so. There are a number of international brand watches that are available in
India nowadays. The Qnet Fraud had been quite on the headlines for many days and so was Mr.

26 01 2013

This site was… how do you say it? Relevant!! Finally I’ve found something which helped me. Cheers!


Introduce tus datos o haz clic en un icono para iniciar sesión:

Logo de

Estás comentando usando tu cuenta de Cerrar sesión / Cambiar )

Imagen de Twitter

Estás comentando usando tu cuenta de Twitter. Cerrar sesión / Cambiar )

Foto de Facebook

Estás comentando usando tu cuenta de Facebook. Cerrar sesión / Cambiar )

Google+ photo

Estás comentando usando tu cuenta de Google+. Cerrar sesión / Cambiar )

Conectando a %s