 |
José Piñera at the World Bank Reform Club (The World Bank Pension Reform Club invited Dr. Piñera to give an address at its launch on November 7, 2000, in the World Bank Institute Studio MCC2. In addition to a studio audience, the session was broadcast by satellite to WB field offices in Riga, Zagreb, Sophia, Vilnius, and Tashkent; the participant sheet is attached. This is a transcript of the session). On the podium, Azedine Ouerghi, Senior Human Development Specialist, World Bank Institute; Michal Rutkowski, Sector Manager, Human Development, ECA Region; and Dr. Pinera began by stating that November 4, 2000, was the 20th anniversary of the signing of the Chilean pension reform program which has given a great boost to the Chilean economy and to workers' sense of freedom and dignity. After the approval of the reform three administrations in Chile have consolidated this program and tried to improve it, and now it is solidly a national program. In these 20 years there were a lot of successes and failures (e.g. Brazil), although much more successes than failures. Now there are seven Latin American countries following the Chilean model in one way or another. Recently in ECA several countries also followed this road (Hungary, Poland, Kazakhstan) and some of them are close to it (Croatia, Romania). Rather than go into technical details on the Chilean reform program itself, or the nitty gritty of how a specific country proceeded in adopting/adapting the Chilean example, Dr. Pinera said he would focus his presentation on three large and important issues, which have emerged again and again during the course of his travels worldwide during the past ten years: 1. how the pay-as-you-go system is structurally flawed; Pay-as-you-go retirement system The pay-as-you-go state-run (PAYG) social security systems are all going bankrupt, not because of marginal problems which can be fixed, but because of a structural flaw. The PAYG program is defined as taxing current workers in order to finance the benefits of the elderly. The problem with this program lies at its root. The link between contribution and benefits is destroyed, that is the link between the contribution to the retirement system and the benefits one receives from it. The contributions are defined by a law in terms of the payroll tax, the benefits are defined by a separate law in terms of the benefits and those laws may change over the 40/45 years of each worker in the labor force. And it would be only by coincidence if a given worker were to receive in present value benefits the same amount he has contributed in present value payroll taxes. Of course, one can separate the link between contributions and benefits in some specific focused programs aimed at helping the very poor - that kind of programs are financed from general taxation. But one can not do that for the entire population and for the entire life of that population - this is Dr. Pinera's basic difference with the theoreticians of PAYG. A massive de-linking of contributions and benefits tends to generate individual and group behavior that sooner or later destroys the program. Individuals try to minimize what they contribute to the system (e.g. through under- declared income, evasion) and maximize what they get out of it (e.g. through union pressure on the government to get early retirement or other special privileges), and the end result is a growing deficit. As for the group behavior, with a PAYG system politicians can please current workers/voters with promises (e.g. early retirement to a given group of people) that will be paid by future workers through higher payroll taxes (those future workers may not even vote today). This is ultimate demagoguery. The pay-as-you-go de-link destroys the necessary link between who pays and who benefits in the same moment of time. So the so called intergenerational transfer is also an intergenerational political distortion (which has been so much the case in Latin America and Europe, where the systems are so much in deficit). Some people place a lot of emphasis only on demographics which is wrong. Demographics is not the ultimate source of the problem, it just accelerates/aggravates the problem. Raising retirement age can atenuate the impact of population aging. The ultimate source of the problem is that the PAYG system is built in such a way that in a democratic political discourse the real problem cannot be addressed without demagogy. This is a structural problem of PAYG and that is why it will go bankrupt all over the world - it's just a matter of time. Personal Retirement Accounts Dr. Pinera is the architect of this system which has been working in Chile for the past 20 years. If the problem in the PAYG system is the de-link in contributions and benefits, then the solution is obvious - to link them, allowing workers to invest the payroll tax in personal retirement accounts (PRAs). In 45 years investments in a portfolio in the market will definitely give a positive rate of return which will always be higher than the rate of return a PAYG system can give at the end of the day (because of the above mentioned distortions). For the worker this means a higher benefit and a much more secure system because it does not depend on demographics. In a defined contribution system it is at the individual level where workers can adjust their contributions to changing life expectancy so the system of PRAs is ultimately about the freedom to choose how to live one's retirement. Retirement age is not determined by politics but by the amount of money accumulated in the worker's account. It allows the worker flexibility to live and manage his own life with dignity. A worker, knowing that he is retiring on his own money, has pride in ownership of property (retirement funds). It empowers the worker. Through payroll taxes governments are expropriating the saving capacity of the majority of workers instead of allowing them to benefit from the extraordinary force of compound interest. It's astonishingly unfair. Transition Cost? This whole discussion about transition costs is overblown, exaggerated and so much confuses the problem. If we agree on that (1) PAYG is going bankrupt and the unfunded liabilities (promises given to workers and the elderly) and so the costs are already there, and (2) a system of PRAs is superior in moral, economic, social, political terms - that's it. Once you have decided to move from PAYG to PRAs, how to move is, of course, a very important and complex issue, but it should never stop you from going from a bad system to a good one. Transition financing problems can always be solved. Chile and other countries did that in one way or another. In fact, there are no transition economic costs (a cost in GNP), as there are, for example, in reducing abruptly inflation when expectations do not follow accordingly, or when there is a radical trade liberalization program. On the contrary, there is an enormous economic benefit. For example, according to Klaus Schmidt-Hebbel's research at least one percentage point of the Chilean economic growth can be clearly attributed to the pension reform, or Martin Feldstein estimate of a 10-20 trillion economic gain if social security is fully privatized in the USA. The so called transition "cost" is a fiscal, cash flow problem, which can be solved by debt, taxes, government spending,WB aid, etc. The way you finance transition fiscal costs is, to an important degree, a political decision. Questions Question 1 The rules about diversification - both inside the country and between domestic and foreign financial instruments - have to be different in each country. In general, the weaker the capital market, the higher the share of the international investments should be. As for the macroeconomics, Dr. Pinera does not believe that, with flexible exchange rates, there should be a balance of payments problem since the system funds, by definiton, grow gradually. Question 2 Virtual accounts are as sexy as virtual sex. It is a very bad idea. As virtual accounts are defined by political processes, this system again depends on political decisions which can be changed any time. If you decide to reform your pension system, you should go for real retirement accounts rather than virtual. Question 3 In the PRA system, Chile kept the minimum pension provisions as they were in the PAYG system. Dr. Pinera is in favor of having a safety net. But no one is intrinsically poor at the age of 20 and people should not be labeled as poor at 20, otherwise they may develop a culture of dependency and poverty. Society must come in only when the person is 65 to finance the gap (if any) between the pension that his accumulated balance can provide and the minimum pension for the very poor. So a valid role for society is to define a minimum pension and to decide how to finance the gap for the poor (through taxation). The system of minimum pension guarantees - as it is now in Chile - is designed not in the best possible way, e.g. there is a moral hazard problem caused by the requirement of minimum 20 years contribution. Designing a better system is an important area where WB experts could contribute. Questions from Zagreb Question 1 The main worry in designing a PRA system should focus on making the system safe and secure, rather than on minimizing administrative costs. Given that focus, of course, one should try to minimize those costs. Twenty years ago in Chile almost a full industry had to be created to manage PRAs. That was expensive at the beginning but in that way safety and security were maximized. Now - according to a CBO study - administrative costs are 1% of assets managed which does not seem to be high. Especially if compared with 11% that has been the average real rate of return in Chile during 20 years. Now there is a lot of discussions in Chile on how to further reduce the administrative costs eliminating unnecessary regulation, fostering more competition, etc. But the issue of administrative costs is not something that should prevent the reform. Question 2 Of course it matters, and I said there is no universal recipe for financing transition. In Chile it was done in a way that was most applicable in that country at that moment. Dr. Pinera does not "export" transition financing formulas, but the concept of PRAs systems. Question 3 With well functioning capital markets, the solution is simple: one could buy a portfolio that completely hedges the risk or move to funds with a very safe portfolio years before retirement. With less developed capital markets there could be another way, e.g. - as in Chile - a person does not have to buy an annuity at the moment of retirement but keep the money in the account (making monthly withdrawals) until the cycle goes up and then buy an annuity. Question from Vilnius Did Chile solve coverage problem, mainly coverage of self-employed? On this regard, Chile kept the system as it was before, covering only the employed, and not covering the self-employed. Dr. Pinera has no particularly strong views on how to address this issue. One idea would be to make the system compulsory for self-employed, but then - with a large informal sector - a "pension police" would be needed. Riga: In cases of severe economic and financial crisis people don't have enough trust in private management of retirement savings. What would Dr. Pinera recommend at the beginning of reform with respect to private vs. public management? Security in the private system is extremely important. The design must be extremely careful in order to ensure transparency, safety, and security of the system. People need to be convinced that the system is safe. The campaign line he used in Chile was "this will be a radical reform with a conservative execution". Government regulation can minimize the problem of security by ensuring portfolio diversification and by providing technical, honest, and depolitized supervision. Transition rules may be needed. For example, Chile began with most of the funds in bonds, then domestic shares (with a ceiling, total and individual(, and then gradual international investing. It has worked well for 20 years, without ever a fraud or scandal. Questions from Tashkent What will happen with the Chilean pension system if there is a severe financial crisis and currency devaluation? What is the probability of such a crisis during a 40-45 year period (an approximate duration of the accumulation phase)? International diversification ensures more security since the probability of another world depression is very low (and even if it happens, PAYG systems will have as many problems as private pension funds). Besides, the world financial markets will reward countries allowing international diversification with inflow of capital in the form of direct investments, loans, bonds as international investors will have more trust in those countries. So there will be no problem in the long run, with globalization going on, whereas in the short run the problem is relevant and that is why in the transition period there have to be prudent portfolio regulations in place. At the beginning, if you don't have a good capital market, all possible efforts should be done to strengthen it, e,g, develop clear disclosure laws, stock exchange laws, etc., but in the meantime regulations should be conservative to ensure safety. Questions from HQ Question 1 Dr. Pinera is of course in favor of unemployment accounts based on ythe PRA model. The optimum should be to introduce them as to replace the existing system of severance payments which rigidize the labor market, not in addition to it. Dr. Pinera believes that the PRA model should be the paradigm for all kinds of social protection programs, including for example introducing medical savings accounts in addition to health catastrofic insurance. Question 2 Dr. Pinera agrees with the comment regarding annuitization. The more flexibility you have the better you are. As for the longevity risk, he believes that longevity is much less of a problem in a PRA system than in PAYG. In a PAYG system a worker becomes entitled to a certain retirement age the moment he enters the system so longevity related uncertainty for a cohort lasts for around 60 years. Retirement age becomes a political contention issue and gets very difficult to change it. This means the adjustment to longevity increase is lagged. In a PRA system most of the longevity increase can be compensated by an individual decision of the worker (e.g. through putting more savings during his working career or working more years, part time or full time), and the uncertainty during the years of retirement can be transferred to the insurance company providing the life annuity. Participants at the Pension Reform Club Launch, November 7, 2000 From the Bank 1. Truman Packard, Social Protection Specialist, LAC From Sofia 1. Dimiter Ninov - Head of Division, MOF From Tashkent 1. Mr. Zabikhodjaev, department head, ministry of social maintenance; From Riga 1. Ms. Elita Krumina, deputy head, Social sphere financing division, Ministry of Finance From Vilnius 1. Alfredas Nazarovas, Vice Minister of Social Security and Labor From Zagreb 1. Sandor Sipos, Office Head, World Bank Croatia Office
|
2010 © www.josepinera.org