Chapter 6 Management of value conflict in the design of retirement policy

Warning View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Conceptual framework
The art of separation in Australia, the Netherlands and the United States
Linkage
Blurring
Concluding comment
Bibliography

Conceptual framework

The design of retirement policy requires a strategy for addressing some unavoidable value conflicts. This chapter exposits key aspects of the conflicts and describes some major similarities and differences among selected countries in the strategies adopted to manage the conflicts.

Social security systems in mature industrial economies are typically committed to at least three basic values: the economic well-being of retirees, the productivity of firms and individual self-reliance. Each of these values can be described in terms of a principle by which income should be distributed to retirees: distribution according to need (well-being), distribution according to the interests of the firm (productivity) and distribution according to the level of personal savings during the working life (self-reliance). We will discuss at length below how other values often influence social security systems.

Ideally, we want to create a national policy as a set of complementary instruments designed to uphold all three of these norms within an integrated system. However, at a practical level these desirable values come into conflict with each other.

For example, when societies have created redistributive systems to ensure economic well-being for those who are economically impoverished, then the more adequate the resources that are distributed, the more intense has been the conflict between the goals of poverty-reducing redistributive initiatives and the goal of promoting individual self-reliance. Such conflicts among the core goals of retirement policy have given rise to some of the most persistent disputes in social security policy. A key question in the design of social security systems is how best to cope with these pervasive conflicts to ensure that retirement policy adequately attends to each of the social values we have described.

One conventional answer to the problem of value conflict amounts to an application of Walzer's art of separation to social security systems, so that each of the values we just described is associated with different institutional "pillars". The simplest version of this approach holds that each of the three values just mentioned should hold sway in each of the major institutional domains of society: the value of firm productivity should govern employer-provided occupational pension systems (the firm pillar), the value of self reliance should govern individual personal savings accounts (the personal savings pillar) and concern for retiree's minimum well-being should shape state-run social security systems (the public pillar). This institutional separation of functions is widely referred to as a "three pillars" or three-tier model.

The three pillars model rests on the idea that institutions are limited in their ability to attend to multiple and conflicting values, so each institution should not attempt to balance all the competing goals of retirement policy. Instead each institutional sphere of society should contribute to retirement benefits in a distinctive way, governed by a distinctive and limited set of values appropriate to that sphere of life. This division of labor allows each pillar to focus on a clear goal-an arrangement that increases accountability and transparency and avoids the potential for paralysis and confusion that complex mandates bring (Thacher and Rein 2004).

Nevertheless, separation can lose some of its essential features in different ways. For example, one way arises from interactions among the institutions that constitute the division of labor. The different institutions have been committed towards different principles and they attempt to influence one another. Contextual changes can also have such effects. To understand the pattern of interactions, one needs to take account of the explicit and implicit rules that govern how these institutions are interconnected.

We will call the explicit cross-institutional rules "linkages". This term refers to the manner in which distributive decisions taken in an institution influence those made in the others (In the context of social security, such decisions determine who is eligible for benefits and what amount she can receive from the two institutions.) Linkage can take several forms, but we will cite particularly two types. The first form of linkage applies the principle of complementarities, according to which each institution is built upon others (for example the public system offers the same level of retirement pensions to an individual irrespective of what the other pillars offer.) In complementary systems, the admissibility to benefits is entirely independent for the two sectors. This is the model of the U.S.A.

Another type of linkage is harmonization, a principle according to which contributions undertaken by the different pillars add to an amount that is fixed, for example 75% of the final net salary. This replacement rate is interpreted as a target to be achieved by the combined efforts of all the separated institutions.1 In harmonized systems, the level of support provided by one pillar obviously influences the level of support provided by others. The Dutch model illustrates this kind of harmonized system.

Because linkage is in some sense the precise opposite of separation, in its strongest forms it eventually undermines the division of labor completely. We describe that collapse of the art of separation as "blurring" of the separated spheres. A separated system has blurred when the commitment of each institution in a system to a particular, limited set of principles or values falters. We believe that two major forces lead to blurring: (1) interference into the retirement system by institutional actors that stand outside the main domain of retirement, and (2) changes in the context within which the retirement system operates, such as demographic, labor force, and economic fluctuations, and the natural maturing of the retirement system over time. When either or both of these kinds of changes occur, each pillar of the system (e.g. the public pension system) faces a need to adapt.

In the remainder of this paper we will explore the role the art of separation plays in the design and operation of retirement systems, and try to develop a more nuanced understanding of what it involves, by examining three case studies: the United States, Australia and the Netherlands. We have selected these three countries because of the different principles of separation in the public and private spheres they have created.

The art of separation in Australia, the Netherlands and the United States

Australia

Australia constitutes a clear example of a social security system that leans on the art of separation. Before 1990, the private contractual labor agreement provided pensions to approximately one-third of the active population, most of the time to better-off persons. By contrast, the State offered a social protection to the aged population based on need. The definition of the needs was large enough to cover 83% the elderly eligible for an Age Pension or a Veterans Pension. Thus, the State's program covered all the workers except those with highest incomes.

In this case we have two separated spheres, and even the populations they served had little overlap. Each institution focused on its own distinctive concerns: The contractual agreements were broadly interpreted as delayed wages and thus served the purposes of attracting, retaining, and motivating a loyal and committed labor force. At the same time, the Age Pension reinforced social solidarity, in the sense that it provided near universal pensions for all but the richest 20% of well off workers, and it was redistributive since the system was financed from general taxation. Nevertheless, in the last twenty years, the division of the labor has begun to be blurred.

The Netherlands

The Dutch system also demonstrates the art of separation, though not to the same extent as Australia. The public system in Holland reflects the principle of solidarity: the entire aged population receives a public pension with the same level of benefits-even retirement-aged persons who had little or no income earlier in life are entitled to full flat-rate benefits2 -and the entire working population contributes the same flat-rate contributions to fund the system. All contributions to the system are the responsibility of the individual.

By contrast, the private sphere in the Netherlands generally distributes pensions according to the principle of equivalence, i.e. benefits related to earning (Most private pensions are defined benefit systems, and most commonly they cover about 70% of final wages.) The private pension system covers virtually the entire working population.

The United States

In the U.S.A., the retirement income system is comprised of obligatory public pensions and a private and voluntary pension system. These two systems rest partly on the principle of equivalence, while in the other countries this principle is of the exclusive domain of the private sphere. The obligatory public system of social security is a universal pay-as-you-go program providing earnings-related benefits that cover over 95% of the working age population.3 The system is financed by the contributions of the employed and of employers, using a flat rate.

The private pension system in the U.S.A. is a voluntary system that covers less than half of the working age population, and firms have generally had considerable freedom to fix retirement benefits in whatever way would serve the interests of productivity. Over time, however, changing economic conditions led firms to place less value on longevity, as younger workers with up-to-date technological skills became more important to the success of many firms. In this new context, it is not surprising to witness a trend toward individual accounts, 401 (K) plans, stock purchase plans, cash purchase plans and many other hybrid forms of firm pensions.4 In this respect, U.S.A. private pensions have undergone major structural changes in the last two decades, from a defined benefit system to one based on individual accounts.

Linkage

The conceptual distinctions that we have offered above create a framework for discussion concerning the practice of linkage in two of the three countries being considered here. We will begin with Holland, which has the most elaborate system of linkage. We will turn then to the U.S.A.

The Netherlands

How has the Netherlands connected the pillars of the private and public sectors? They have done so by means of three central concepts, which are rather unique to the Dutch system. These three concepts are: the Convention, the Franchise and the Covenant. These concepts arose because it was necessary to cope with an extremely polarized and fragmented system involving a division of the society along religious and secular lines. The Convention defines non-legal5 but binding norms which led to a wide but tacitly understood agreement that a worker should be able to get at the time of this retirement 70% of his final net wages. The Franchise refers to the value of each individual's public pension, which sets the starting point from which firms calculate the level of pension support that they must provide to ensure that an employee's total pension will reach 70% of final earnings.6 The Covenant is a set of implicit norms and rules that have evolved over a long period of time. In particular, the Dutch system rests on the assumption that adequate pensions can be realized only through the combination of resources from both the public and private domains.

The Dutch case exemplifies a harmonized system, since the benefit provided by one system determines the benefit level needed for the other system, given a broad convention that establishes the amount of income individuals need in retirement. To repeat, in the Netherlands a private pension fund first computes the value of the public benefit, which is the so-called franchise level, that each individual receives in exchange for their lifetime contributions to the public pension. It is then the balance that falls short of 70% of final wages that is the responsibility of the private occupational pension system to cover.

The United States

The system of the U.S.A. is generally based on the principle of supplementation, since income arising from each subsystem increases the available total income at the time of retirement. The public pension system is supplemented by occupational pensions, and personal savings are further added to the occupational pensions. Each pillar of benefits adds to the total amount, which is then subject to income taxes (however, the rate of taxation is below that for earned income.)7 The important point is that in the U.S.A. there is no national administratively agreed on adjustment techniques to tie the systems together, as exists in Holland where the franchise specifies how the two systems should be uniformly linked.

Even in the U.S.A., however, there are exceptions, which are worth reviewing briefly to illustrate how linkage in this system differs from that of a system that has tight linkage between the public and private pillars (as in Holland). These exceptions involve firms who have developed contractual agreements with some or all of their employees that promise to protect the individual in the event that the public pension decreased in value. Recently, an independent analysis reported by the Center for Retirement Research at Boston College estimates that one quarter of private plans are integrated in this manner and, surprisingly, that this proportion actually increased between 1993 and 1997 (Perun 2002:2). Regardless of their scope, what is important to emphasize about these arrangements is that they are voluntary.

Blurring

As mentioned earlier, blurring refers to the collapse of the art of separation. A system becomes blurred when there is a decline of each institution's commitment to a series of particular principles or values. When such changes happen, each pillar of the system (for example the public pension system) must adapt. The necessary adaptation does not always entail blurring. For example, in some cases, the system could revise its linkage rules so as to preserve the underlying normative principles upon which the institutional rules of distribution rely (for example the distribution according to needs). On the other hand, a more radical response may be necessary and each institution must then revise these normative principles in their totality. Illustrations of these issues will now be provided in the cases of the Netherlands and the U.S.A.

The Netherlands

In Holland, the blurring seems to be an outcome of the issue of redistribution of costs among the sectors. How can each sector modify its share to the totality of the income as well as its share of the burden relative to the gross domestic product? The State is pressing the private sector to expand its coverage to 100% of the population and to use a franchise principle that imposes on firms a higher percentage of pension costs. While the State wants to alter the roughly 50:30 share of public and private occupational pensions, some firms are responding by moving in the opposite direction, not by changing the rules of the covenant but by changing the norms governing the non legally binding covenant and lowering the 70% replacement rule of final earnings. The underlying basic issue is as follows: Will the public share of total pension costs decline as the private contribution increases or will there be a shift toward more self-reliance as personal accounts and personal assets are encouraged to expand?

The Dutch story also illustrates how developments outside of the pension system, in this case a court decision about the equal treatment of women, can influence change in the retirement system. The following example illustrates how linkage can lead to blurring when rule changes introduced outside of the pension system weaken the incentive of firms to comply with rules that they did not directly participate in creating.

The key event in this story was a ruling of the European court, which had an important but unintended impact on the contribution of firm pensions to the replacement of final earnings. Under the Dutch system, a non-working wife did not get a public pension in her own right. The European Court ruled that this was illegal, so the rule was changed so that the wife got half of the husband's pension. This ruling had the effect of reducing the value of husband's basic pension by half, and therefore the lower franchise technically meant that firms needed to increase his private pension dramatically in order to reach the 70% replacement goal. In order to avoid this higher cost to the occupational pension, many firms ignored the ruling and continued the earlier practice of defining the franchise as being based on the older public benefits given to the husband. This practice has deep cultural roots known as the male breadwinner model, and it was based on an implicit pooling assumption that family income was equally shared and the wife got half of the public benefits received by the husband. But this is now history.

In sum, the introduction of new EU rules created a blurring of the rules of linking the public and the private contributions. The blurring persisted because as the State got its way in computing the franchise, the firms began to redefine the social conventions by lowering the replacement rate of 70%.

The United States

There exists a distinct form of blurring in the system of pensions of the U.S.A. It seems that occupational pensions based on the principle of defined benefits are being replaced by a system based on personal accounts that derive from defined contributions. New personal accounts like investments in stock options are now serving the function of promoting the loyalty and productivity incentives that the older occupational pensions once served. It seems therefore that the relative sizes of the different spheres of responsibility involving occupational pensions and personal accounts are shifting without there being any major legislative pension reform inspiring the change in practice.

In this narrative, we read the basic dynamic of the story as one where the state gradually makes greater and greater demands on the private sphere. If the public system can't provide the replacement rate for wages that policymakers want to provide, do they try to pressure the private system to do the state's bidding by the use of tax incentives, regulation and mandating, according to the state's interpretation of broadly accepted public values? That would undermine the private system and contribute to changes within the private system, as it tries to pass on the costs to other institutional systems. In the process there is change in the relative balance of values that was created in the initial system of firewalls. The result is a continuous dynamic process of change in the public-private mix.

Concluding comment

What is the role of blurring in institutional change? Our general argument is that we have not seen a direct effort to change the value system that a country has historically adapted. What we find in our three cases is that the values underlying the system do change, but the change seems to occur first by the emergence of blurring when the principles governing the systems of linkage introduce inconsistencies or other visible non-intended effects. In other words, it is changes in governing the linkage systems that activates a process that eventually introduces change in the separated institutions.

Bibliography

Durkheim, Emile. 1997. The Division of Labor in Society [1893]. New York. Free Press.

Hacker, Jacob. 2004. "Call it the family risk factor." New York Times. Op Ed. January 11: 15.

Hacker, Jacob. 2004. "Privatizing risk without privalizing the welfare state: The hidden politics of social policy retrenchment in the United States." American Political Science Review. 98, 2: 244, 254.

Lindbreck, Asar and Mats Persoso. 2003. "Gains from pension reform." Journal of Economic Literature. XLI: 74 to 112.

Perun, Pamela. 2002. Social Security and the Private Pension System: The Significance of Integrated Plans. Working Paper no. 2. Center for Retirement Research at Boston College.

Ploug, Niels. 2003. "The recalibration of the Danish old-age pension system." International Social Security Review. 56, 2: 70.

Special Committee on Aging, United States Senate. April 1982. Linkage Between Private Pensions and Social Security Reform. Washington. U.S.

Thacher, David and Martin Rein. 2004. "Managing value conflict in public policy." Governance. 17: 457 to 486.

Van Riel, Bart, A. Hererijck and J. Vissar. 2003. "Is there a Dutch way to Pension Reform?" Chapter 3 in Pension Security in the 21st Century: Redrawing the Public Private Divide. Gordon Clark and Noel Whiteside (eds.). Oxford. Oxford University Press.

Wagner, Peter. 1994. "Dispute, uncertainty and institution in recent French debates." Journal of Political Philosophy. 2, 3: 270 to 289.

Walsh, Mary Williams. 2004. "U.S insurer of pensions says its deficit has soared." New York Times. January 16, p.C1.

Walzer, Michael. 1984. "Liberalism and the art of separation." Political Theory. 12, 3: 315 to 330.


Notes

  1. A similar logic of linkage also enters if the objective is to reach a minimum income level based on a means test rather than a replacment rate.
  2. The only exception concerns couples, who typically do not receive the full benefits to which they would be entitled as two single persons, since the benefits formula assumes that larger family units enjoy economies of scale.
  3. Historically, the United States has a minimum pension for selected groups of low wage employees. However this provision was dropped at the same time that a national means tested program, knowsn as Supplementary Security Income (SSI), was introduced as a separate administrative organization independent from the social security program. This new program provided financial aid to aged pensioners based on need. But it is also a public program to supplement the income needs for those in the bottom end of the earning distribution.
  4. This shift does not, of course, reflect a re-weighing of the values of self-reliance and industrial productivity; it reflects the continued dominance of firm productivity as the overriding value in the firm pillar, now applied in a more volatile economic environment where longevity is less important Indeed, the very existence of occupational pensions may be threatened by changing economic conditions.
  5. Though the main force of the convention is non-legal, some legislative provisions reinforce it. For example, personal accounts which provide an annuity insurance are tax exempt if their benefits do not exceed 70% of final earnings, thus there is little scope for the growth of these personal accounts (Van Riel et al. 2003:3).
  6. More precisely, the franchise level is close to, but not identical with the actual benefit levels that individuals and couples receive as the basic flat rate public pension. We do not review the technical issues in this computation here.
  7. Earnings from work can also supplement public social security payments. This was not true historically, when the main purpose of public pensions was to get older workers out of the labor market and to make room for the young. But in the last several years Congress changed the law and removed this restriction on earned income. Such income is, of course, subject to income tax, which does not affect the size of the public pension, but does effect net disposable income, i.e. what the individual actually receives.