In Chile we accomplished a revolutionary reform. We knew that cosmetic changes — increasing the retirement age, increasing taxes — would not beenough. We understood that the pay‐​as‐​you‐​go system had a fundamental flaw, one rooted in a false conception of how human beings behave. That flaw was lack of a link between what people put into their pension program and what they take out. In a government system, contributions and benefits are unrelated because they are defined politically, by the power of pressure groups.

So we decided to go in the other direction, to link benefits to contributions. The money that a worker pays into the system goes into an account that is owned by the worker. We called the idea a “capitalization scheme.”

We decided that the minimum contribution should be 10 percent of wages. But workers may contribute up to 20 percent. The money contributed is deducted from the worker’s taxable income. The money is invested by a private institution, and the returns are untaxed. By the time a worker reaches retirement age — 65 for men, 60 for women — a sizable sum of capital has accumulated in the account. At retirement the worker transforms that lump sum into an annuity with an insurance company. He can shop among different insurance companies to find the plan that best suits his personal and family situation. (He pays taxes when the money is withdrawn but usually at a lower rate than he would have paid when he was working.)

As I said, a worker can contribute more than 10 percent if he wants a higher pension or if he wants to retire early. Individuals have different preferences: some want to work until they are 85; others want to go fishing at 55, or 50, or 45, if they can. The uniform pay‐​as‐​you‐​go social security system does not recognize differences in individual preferences. In my country, those differences had led to pressure on the congress to legislate different retirement ages for different groups. As a result, we had a discriminatory retirement‐​age system. Blue‐​collar workers could retire at 65; white‐​collar workers could retire more or less at 55; bank employees could retire after 25 years of work; and the most powerful group of all, those who make the laws, the congressmen, were able to retire after 15 years of work.

Under our new system, you don’t have to pressure anyone. If you want to retire at 55, you go to one of the pension‐​fund companies and sit in front of a user‐​friendly computer. It asks you at what age you want to retire. You answer 55. The computer then does some calculations and says that you must contribute 12.1 percent of your income to carry out your plan. You then go back to your employer and instruct him to deduct the appropriate amount. Workers thus translate their personal preferences into tailored pension plans. If a worker’s pension savings are not enough at the legal retirement age, the government makes up the difference from general tax revenue.

The system is managed by competitive private companies called AFPs (from the Spanish for pension fund administrators). Each AFP operates the equivalent of a mutual fund that invests in stocks, bonds, and government debt. The AFP is separate from the mutual fund; so if the AFP goes bankrupt, the assets of the mutual fund — that is, workers’ investments — are not affected. The regulatory board takes over the fund and asks the workers to change to another AFP. Not a dime of the workers’ money is touched in the process. Workers are free to change from one AFP to another. That creates competition among the companies to provide a higher return on investment and better customer service, or to charge lower commissions.

The AFP market opened on May 1, 1981, which is Labor Day in Chile and most of the world. It was supposed to open May 4, but I made a last‐​minute change to May 1. When my colleagues asked why, I explained that May 1 had always been celebrated all over the world as a day of class confrontation, when workers fight employers as if their interests were completely divergent. But in a free‐​market economy, their interests are convergent. “Let’s begin this system on May 1,” I said, “so that in the future, Labor Day can be celebrated as a day when workers freed themselves from the state and moved to a privately managed capitalization system.” That’s what we did.

Today we have 20 AFPs. In 14 years no AFP has gone bankrupt. Workers have not lost a dime. Of course, we created a regulatory body that, along with the central bank, set some investment diversification rules. Funds cannot invest more than x percent in government bonds, y percent in private companies’ debentures, or z percent in common stocks. Nor can more than a specified amount be in the stock of any given company, and all companies in which funds are invested must have credit ratings above a given level.

We set up such transitional rules with a bias for safety because our plan was to be radical (even revolutionary) in approach but conservative and prudent in execution. We trust the private sector, but we are not naive. We knew that there were companies that might invest in derivatives and lose a lot of money. We didn’t want the pension funds investing workers’ money in derivatives in Singapore. If the system had failed in the first years, we would never have been able to try it again. So we set strict rules 14 years ago, but we are relaxing those rules. For example, only three years ago we began to allow the funds to invest abroad, which they weren’t allowed to do initially, because Chilean institutions had no experience in investing abroad. The day will come when the rules will be much more flexible.

Let me say something about the transition to the new system. We began by assuring every retired worker that the state would guarantee his pension; he had absolutely nothing to fear from the change. Pension reform should not damage those who have contributed all their lives. If that takes a constitutional amendment, so be it.

Second, the workers already in the workforce, who had contributed tothe state system, were given the option of staying in the system even though we thought its future was problematic. Those who moved to the new system received what we call a “recognition bond,” which acknowledges their contributions to the old system. When those workers retire, the government will cash the bonds.

New workers have to go into the new private system because the old system is bankrupt. Thus, the old system will inevitably die on the day that the last person who entered that system passes away. On that day the government will have no pension system whatsoever. The private system is not a complementary system; it is a replacement that we believe is more efficient.

The real transition cost of the system is the money the government ceases to obtain from the workers who moved to the new system, because the government is committed to pay the pensions of the people already retired and of those who will retire in the future. That transition cost can be calculated. In Chile it was around 3 percent of gross national product. How we financed it is another story. It will be done differently in each country. Suffice it to say that even though governments have enormous pension liabilities, they also have enormous assets. In Chile we had state‐​owned enterprises. In America I understand that the federal government owns a third of the land. I don’t know why the government owns land, and I don’t know the value. Nor am I saying that you should sell the land tomorrow. What I am saying is that when you consider privatizing Social Security, you must look at assets as well as liabilities. I am sure that the U.S. government has gigantic assets. Are they more or less than the liabilities of the Social Security system? I don’t know, but the Cato project on privatizing Social Security will study that. In Chile we calculated the real balance sheet and, knowing there were enough assets, financed the transition without raising tax rates, generating inflation, or pressuring interest rates upward. In the last several years we have had a fiscal surplus of 1 to 2 percent of GNP.

The main goal and consequence of the pension reform is to improve the lot of workers during their old age. As I will explain, the reform has a lot of side effects: savings, growth, capital markets. But we should never forget that the reform was enacted to assure workers decent pensions so that they can enjoy their old age in tranquility. That goal has been met already. After 14 years and because of compound interest, the system is paying old‐​age pensions that are 40 to 50 percent higher than those paid under the old system. (In the case of disability and survivor pensions, another privatized insurance, pensions are 70 to 100 percent higher than under the old system.) We are extremely happy.

But there have been other enormous effects. A second — and, to me, extremely important — one is that the new system reduces what can be called the payroll tax on labor. The social security contribution was seen by workers and employers as basically a tax on the use of labor; and a tax on the use of labor reduces employment. But a contribution to an individual’s pension account is not seen as a tax on the use of labor. Unemployment in Chile is less than 5 percent. And that is without disguised unemployment in the federal government. We are approaching what could be called full employment in Chile. That’s very different from a country like Spain, with a socialist government for the last 12 years, that has an unemployment rate of 24 percent and a youth unemployment rate of 40 percent.

Chile’s private pension system has been the main factor in increasing the savings rate to the level of an Asian tiger. Our rate is 26 percent of GNP, compared to about 15 percent in Latin America. The Asian tigers are at 30 percent. The dramatic increase in the savings rate is the main reason that Chile is not suffering from the so‐​called tequila effect that plagues Mexico. We do not depend on short‐​run capital flows because we have an enormous pool of internal savings to finance our investment strategies. Chile will grow by about 6 percent of GNP this year, the year of the “tequila effect.” The stock exchange has gone down by only 1 or 2 percent and will be higher at the end of the year. Chile has been isolated from short‐​run capital movement because its development is basically rooted in a high savings rate.

Pension reform has contributed strongly to an increase in the rate of economic growth. Before the 1970s Chile had a real growth rate of 3.5 percent. For the last 10 years we have been growing at the rate of 7 percent, double our historic rate. That is the most powerful means of eliminating poverty because growth increases employment and wages. Several experts have attributed the doubling of the growth rate to the private pension system.

Finally, the private pension system has had a very important political and cultural consequence. Ninety percent of Chile’s workers chose to move into the new system. They moved faster than Germans going from East to West after the fall of the Berlin Wall. Those workers freely decided to abandon the state system, even though some of the trade‐​union leaders and the old political class advised against it. But workers are able to make wise decisions on matters close to their lives, such as pensions, education, and health. That’s why I believe so much in their freedom to choose.

Every Chilean worker knows that he is the owner of an individual pension account. We have calculated that the typical Chilean worker’s main asset is not his small house or his used car but the capital in his pension account. The Chilean worker is an owner, a capitalist. There is no more powerful way to stabilize a free‐​market economy and to get the support of the workers than to link them directly to the benefits of the market economy. When Chile grows at 7 percent or when the stock market doubles — as it has done in the last three years — Chilean workers benefit directly, not only through higher wages, not only through more employment, but through additional capital in their individual pension accounts.

Private pensions are undoubtedly creating cultural change. When workers feel that they own a fraction of a country, not through the party bosses, not through a politburo (like the Russians thought), but through ownership of part of the financial assets of the country, they are much more attached to the free market, a free society, and democracy.

By taking politicians out of the social security business we have done them a great favor because they can now focus on what they should do: stop crime, run a good justice system, manage foreign affairs — the real duties of a government. By removing the government from social security, we have accomplished the biggest privatization in Chilean history — someone even called it, paraphrasing Saddam Hussein, the mother of all privatizations, because it has allowed us to go on to privatize the energy and telecommunications companies.

That has been our experience. Of course, there have been some mistakes. There are some things that should be improved. There is no perfect reform. With time and experience, I know I would do some things differently. But on the whole, I can tell you that it has been a success beyond all our dreams.