Home |  Elder Rights |  Health |  Pension Watch |  Rural Aging |  Armed Conflict |  Aging Watch at the UN  

  SEARCH SUBSCRIBE  
 

Mission  |  Contact Us  |  Internships  |    

 



back

Can Public Social Insurance Systems Survive?

Dr. Lawrence H. Thompson
Senior Fellow, Urban Institute 

I want to frame the issue of the world-wide debate over pension reform because there is a debate going on throughout the world. It’s a debate over the structure of social security, and over what we can afford about social security. I think we need to understand what is underlying that debate before we take knee-jerk positions on the different sides of the debate. That’s my purpose today to give you a background on what’s going on, what the fight’s about, and what’s at stake.

I start with what is the role of a public pen-sion system? I think they started off a hundred years ago, basically as a system of ensuring adequate retirement income, first and fore-most through the pooling of risks, the risks of uncertain life spans, the risk of an uncertain economy, not knowing what is going to happen to employment or to price levels and so forth, pooling those risks, and, in effect, forcing some sort of savings so that workers could reach retirement age and be assured of a steady stream of monthly benefits for the rest of their life. That was the model.

Also, there is an element of promoting social cohesion that gets involved in this. It has been argued that the US Social Security System played an important role in creating a direct relationship between the federal government and the people of the United States in getting them to think of themselves as one country. That's a scenario that has gone on in other places around the world.

More recently, people have begun to notice that the pension programs can also be a source of capital. If run correctly, if the first generation through a pension program saves in advance of its benefits, it can produce a large stock of capital, which can be useful in increasing the economic activity in the society.

Pension programs are unique institutions (social institutions), because they make long range promises. They promise a worker who is starting out in his 20's that they will be there for him or her 75 years later, perhaps, when that worker is in her 90's and drawing on the pension. That's about as long range a promise as you will ever find in a social or economic institution. In order to make good on those promises, we need institutions which are stable. They have to stay in existence for 90 or 100 years. What is it that we can say about the characteristics that may promote the stability of social and economic institutions?

First, all pension programs have built into them an inherent tendency for somebody to overpromise. It's easy to promise future benefits and not put aside enough money to pay for them, to not worry about what the costs will be, let tomorrow take care of that. One of the threats to stability of any pension program is overly generous promises. These are promises that can't be fulfilled; and there-fore, we have to cut back on budgets and create social instability.

Second, inequitable distribution of costs and benefits. One of the problems that occurs in social security systems around the world is that people lose confidence that costs are being distributed fairly, or that the benefits are being distributed fairly. Americans find it hard to believe that there are other countries in the world where there is not one social security system; but instead, there are 20 or 30 different systems. The system for school teachers, and miners, and railroad workers, allows them to retire at age 50, whereas the day laborers have to work until they are 65; but yet, they are all paying the same rate; or, perhaps, the more advantaged people are paying a lower rate. Second, there are many parts of the world in which there are stated benefit packages, stated contribution rates; and, half the labor force cheats on the package. They don't pay the contribution rates. They figure out a way around it; but yet, they figure out a way to get the benefits. That sort of thing undercuts the backing for the pension program, and creates an unstable situation.

You need a broad consensus about scope and structure. A political party is not going to last 90 years in power; and so, a system that is going to get created and is going to be reliable and continue on has got to be a system which reflects a consensus of the population. Measures have to be taken to make sure that it does reflect a consensus of the population.

Finally, one thing we know about the future is that it will change. The economics will change. The demography will change. We can't predict what the future's going to hold in terms of demographic trends; but, we do know that we can't predict it. Our systems are going to have to adapt to change; and, if they are systems that get locked into place, and which can't change, then eventually they will be crushed because you have to adapt in order to survive.

What are the challenges facing social security programs around the world? The first challenge is cost increases. The population is aging. As the population ages, the cost of any pension system goes up. It doesn't even have to be a social security system--it's true of every pension system. Just think about it, if you are going to live longer after you retire, you have to set aside more money in your savings account in order to support yourself. Your employer has to set aside more money. Your society has to pay a higher contribution rate.

Second, many of these systems are maturing now. You know, when they start off a defined benefit plan, the first 20 or 30 years very few people are drawing benefits. It's easy to be lulled into believing that it's going to be cheap and that, you can get long benefits at high rates for cheap contribution rates. As these things mature, you have to face up to the fact that you have made promises that need to be financed. That's the cost challenge. There are serious institutional weaknesses around the world, and especially in the underdeveloped countries of the world, not all of them, but in many of them, institutional weaknesses that undercut their programs. You go in, and you try to file for benefits, and you discover that it is a one year marathon in which you get the sense that it's you against them, and only a few of you are going to win and get your benefits.

Finally, there are political rigidities. As I said, systems have to adapt. If they can't adapt, the political system is such that it degenerates into warring factions. Each faction is able to stop the other one from instituting any changes. The political discourse will get more and more bifurcated; and, probably, the system will be unable to adapt, and eventually it will have to undergo a wrenching change rather than an evolutionary change.

In any society, the cost of supporting the aged can be roughly decomposed into three ratios: One is the ratio of the "total consumption of the population to the total GNP"; second is the "retiree dependency ratio"-the ratio of the retirees to the total population; third is the "living standards ratio"-the living standards of the retired population as a fraction of the living standards of the total population.

The cost of taking care of the aged is dominated by the demographics of the aging of the population; and, it's also dominated by the closely related phenomenon of the social and economic decisions pertaining to when people will retire, and what will be their relative living standard. These are linear relationships. If the ratio of the elderly to the non-elderly rises by 20 percent, the cost of supporting the aged rises by 20 percent. If we have a retirement age that declines a bit, that forces the cost up. If we raise the relative standard of the aged by 10 percent relative to the rest of the population, the cost rises by 10 percent. You can see the challenge that comes as the demographics cause the populations to age and drives up the costs. You can't change this, this is immutable. The only way to change it is to change one of those ratios--to have them work longer and reduce the amount of time spent in retirement, or to adjust down the living standards of the aged--that's it.

Lastly, institutions matter, but only if they affect one of these ratios. There's a great debate over whether we should have a defined benefits pay-as-you-go system, or whether we should have individual saving systems. Keep in mind that the form of the retirement instrument, the way the pension system is organized, affects costs only if it affects one of those ratios, only if it affects the living standards or the demographic ratio.

The traditional model that came out of Bismarck in Germany, and spread throughout Europe and North America and into Latin America, is the "defined benefit" model. The model (benefit defined in advance) probably calculates your pension in relationship to pre-retirement earnings. The promise is that you will get that benefit. The person who makes that promise has to figure out how to finance it.

The reason this model grew up is that it has several inherent strengths. One strength is that it provides that predictable income stream. That is the first purpose of having a pension, providing a predictable income stream for old age. Second is that it facilitates a measure of social solidarity-a way of taking care of the lower wage people by giving them proportionally more. This is a feature in most social security systems, some more than others; but, it's found in most traditional social security systems, in a way that preserves individual dignities, so that we can have solidarity with low income people, and with people who are old when the program gets started.

There are some inherent weaknesses in this model - that's the reason that reform is becoming such a hot topic all around the world. One weakness is that the model has proven to be fairly ineffective as a way of generating capital. Publicly-operated defined benefit plans have a very poor record in being able to accumulate reserve capital. It's been tried many times. Something goes wrong, almost inevitably; and so, it is simply not a reliable way of accumulating capital, if that is what you want to do.

Second, it is clear as you look around the world that the necessary political discipline and preconditions for making these systems work does not exist in some parts of the world. One of my themes this morning is that a lot of the reforms, especially those which we see in Latin America, are reactions to failures of this particular model.

There are three alternative models that exist around the world. There's the "defined benefits" model, a "collective savings" model (defined contribution), and the one that the debate seems to be focusing on, an alternative to defined benefits, the "individual savings" model, in which each worker is required to save a certain amount, put it in an account, and the amount of benefit that worker receives in retirement is a function of how much money is in that account at the time that the worker retires. We do not define the benefit, we define the contribution; and, we determine the benefit by seeing how much is in the account.

This model is attractive where the defined benefit has failed. It constrains excessive political promises. Some of the Latin American countries have replacement rates of 80 percent, 90 percent and 100 percent that they've promised workers. Some of them have promised retirement ages of 50 and 55. Ladies and gentlemen, you cannot afford a system which pays 80 percent replacement rates at age 55. A society cannot tolerate the kind of tax rates that takes. An 80 percent replacement rate would probably mean that the worker has a higher standard of living after he retires than he did when he was working. This individual savings model stops that practice. It prevents the politicians from making those promises. It prevents perverse distribution.

The United States runs a system in which all of the money that's collected and used to pay benefits comes from payroll taxes; and, its proportionate tax, and we all understand it. In many countries of the world, there is government budget money going in to help finance social security, money collected from all taxpayers. Many of those countries have social security systems which only cover 10 or 15 or 20 percent of the population. It's the 10 or 15 or 20 percent richest part of the population; and, their pensions are being sub-sidized by the general budget, paid for by the poor as well as the rich. This prevents that sort of redistribution. You are only going to get back what you put in; so, it doesn't allow for favorable redistribution, but it prevents perverse redistribution.

Finally, it insulates the financing of these systems from compliance problems. The reason you get 40 and 50 percent payroll tax rates in many of these countries is because only half the people are paying. In this kind of an individual savings system, if you don't pay, you don't get back your benefit. The rest of us do not bear the burden of your not paying-that's the good news. The bad news is that we may end up with a whole lot of old people in poverty, because, if they cheated when they are working, and they didn't pay their social security contributions, they won't have a benefit when they retire. It sounds good, and the people who are advocates of this will stop there.

There are some weaknesses. There is a price you pay for this type of reform. The first weakness is that the workers assume all the risks. Life is full of risks. You do not know what is going to happen to financial markets. You do not know what is going to happen to the price level. You do not know what is going to happen to demography. In the individual savings model, all of those risks are born by the worker. That's the price we are going to pay for getting the politicians out of the business of setting future benefits. In the OECD countries which have struggled with and have been able to gradually adjust their social security promises - Germany, the United States, Sweden, Italy - when they have had to face the consequences of demographic change, inevitably they have faced it by cutting back on the benefits a little bit and raising the taxes a little bit. This spreads the risks throughout society. In this individual savings model, you don't do that, you give all the risks to the worker.

Secondly, the model is one which in Latin America has proven to have very high administrative costs. It costs 10 times as much to operate the Chilean system as it does to operate the US system.

Thirdly, the model is one which only works if you have very effective government regulation; otherwise, you create a situation in which you invite a savings and loan debacle in each country of the world. People compete for money by offering high returns and then investing unwisely in order to earn the money to make good on those promises.

Looking at the debate that is going on, between the three models, I urge people to not get hung up on the models, but to look first at what it is that we are trying to do in these institutions. How can we build institutions which will achieve the objectives of income adequacy, risk sharing, and social cohesion. How can we do that? How do we build institutions which are stable? Are we dealing in a world, or in a particular situation? Are we dealing in a society which cannot produce a stable defined benefit plan? If that is the case, let's not try. And how can we tailor the institutions that we create to each society's traditions and circumstances? How can we generate a consensus around how to build such an institution, a consensus which will build an institution that will last 75 to a hundred years, which will produce the kind of retirement income that we want from our pension programs?