摘要：This paper presents the architecture of the Swedish pension system, the philosophy behind it and illustrates how it works in practice. At the foundation of the Swedish pensions landscape is the public pension commitment. It is based on the philosophy that all countries must begin with: The commitment to provide a sufficient minimum standard of living in old age for all. This is the role of the necessary pillar zero. Pillars one and two consist of mandatory universal public nonfinancial defined contribution (NDC) and a financial defined contribution (FDC) schemes. These cover the earnings of all employees and self-employed – up to a ceiling.
关键词：the architecture of the Swedish pension system; nonfinancial defined contribution (NDC); financial defined contribution (FDC)
Edward Palmer (Professor (em.) and Senior Fellow at the Uppsala Center for Labor Studies, Uppsala University)
This paper presents the architecture of the Swedish pension system, the philosophy behind it and illustrates how it works in practice. At the foundation of the Swedish pensions landscape is the public pension commitment. It is based on the philosophy that all countries must begin with: The commitment to provide a sufficient minimum standard of living in old age for all. This is the role of the necessary pillar zero. Pillars one and two consist of mandatory universal public nonfinancial defined contribution (NDC) and a financial defined contribution (FDC) schemes. These cover the earnings of all employees and self-employed – up to a ceiling. In principle, the technical construction of the NDC and the guarantee benefits – explained in greater detail in the paper – can be viewed together as a package that fulfills the guaranteed minimum standard of living in old age, while making full use of the DC accounts to provide an incentive for individuals to contribute to their own retirement income. Chart 1 provides an overview.
Chart 1. Sweden’s Pension Landscape.
Sweden’s overall pension landscape also has a third pillar consisting of four quasi-mandatory occupational pension schemes based on private-sector collective labor market agreements. Together these four schemes cover about 90 percent of all employees in Sweden. The contribution rate for the entire package is 23%, excluding the cost of the guarantee benefit package, which is financed with general tax revenues (i.e., the national state budget). In addition, individual retirement savings plans can be contracted on the private insurance market.
Note that the general overall architecture of Chart 1 could represent an emerging economy, as well as a fully developed market economy such as Sweden’s. In the context of the emerging economy, NDC is the formal economy component, where the need for a guarantee declines over time with the development of the formal economy. Formality is defined in this context by paying contributions to the universal public pension system.
NDC is superior to FDC in the context of the emerging economy because it provides a context for formalizing transactions that lead to formal pension rights, while making available money that can if only in part pay for the pensions of current pensioners. In contrast, FDC requires 40+ plus years from its introduction to full maturity – when new retirees can obtain a an FDC-based benefit based on a full life working career. NDC rights are acquired in the same manner as in FDC schemes, but the system’s revenues can be made available through a rule-based connection to a bottom pillar that covers the older generation – the parents of the workers in the formal workforce.1
The bottom line is that the combination of NDC with FDC “on top,” together with an externally (budget) financed guarantee can be viewed as the generic universal pension system. Seen in this perspective, NDC is the essential “safe asset” while FDC may (but not necessarily will – especially not in an emerging economy –) obtain a higher return. The higher risk associated with FDC schemes is inherent to the nature of financial market. Increasingly higher returns are coupled to higher levels of portfolio risk consequently uncertainty, where the latter increases with the degree of risk of the specific portfolio. The goal of this paper is to provide an understanding of the underlying logic and architecture of an “optimal” structure of a country’s overall pension system with the Swedish model, combining NDC and FDC as a point of departure.
The architecture and basic design of the present Swedish pension scheme was the product of the government’s Working Group on Pensions 1992-1994. The decision to replace the then existing DB schemes (which at the time characterized the entire Swedish pension system) with NDC and FDC schemes, with a new guarantee minimum benefit at the foundation, was passed in Parliament in June of 1994.
The reasons behind the realignment from DB to DC were many: The first, was to ensure affordability and financially sustainability. The second was to create intra and inter-generational fairness. And, the third goal was that the scheme should be economically efficient. This was accomplished through the introduction of transparent personal accounts, which define NDC. NDC and FDC pension schemes are efficient at the microeconomic level because they provide an incentive for individual labor market participation with, which has the macroeconomic effect of increasing the working-age population.2
With the realignment from DB to DC, the current working population’s acquired rights in the DB scheme that was replaced were converted into NDC rights, with the latter being based on what the policymaker considers to be a fair conversion rule. There was a gradual phase in covering 15 birth cohorts and all previously granted pension benefits were honored without change. Conversion models, including Sweden’s, are provided in Palmer (2006).
The zero-pillar component of the Swedish pension system fulfills the goal of providing a sufficient minimum income in old age even for those who for one reason or another otherwise would be in poverty. What has received too little attention in the literature and discussion around pension reform – especially in considering the possibilities of NDC in the context of an emerging economy. It is how the anti-poverty guarantee benefit that interacts with NDC at the foundation of the overall system. The guarantee tapers incrementally off and the DC framework takes successively over with the increase in the income provided by the full-fledged contributory NDC and FDC pensions. In the case of the emerging economy, with time, depending on the rate of economic development, the role of the guarantee will decrease and the full coverage of NDC will increase.
Another set of topics for which there is a separate section, has to do with the roles of a) demographic reserves and b) a solvency-ratio approach to balancing the development of liabilities with assets. Also, the paper discusses how Sweden has created a unique Clearinghouse approach designed so as to provide financial market portfolios that fulfill the interests of participants – high but secure returns provided at a low cost.
The paper is organized as follows. Section 2 starts with the underlying philosophy of the Swedish pension architecture. Section 3 describes the system design in greater detail. Section 4 addresses the question of how to set the contribution rate in the context of the individual’s number of years of work and life expectancy at retirement. The section concludes by addressing the question of what constitutes an adequate replacement rate for a country’s overall UPPS commitment. Section 5 presents and discusses how NDC indexation works, the role of a reserve fund, and how a solvency-based balancing mechanism aids in maintaining a more immediate gyro towards tighter financial sustainability in NDC. Chapter 5 concludes with reflections on important lessons from the Swedish experience.
1 The process is illustrated with an example in Larsson, Leyaro, and Palmer (2020), who discuss NDC in the context of the emerging African economy of Tanzania, where the development of the formal Zero-NDC-Pillar nexus follows the formalization of the economy.
2 These features of both NDC and FDC account-based schemes and how they can be used in the context of introducing social policy into the pension system are discussed in Holzmann and Palmer (2020), the introductory chapter in Holzmann, Palmer, Palacios and Sacchi (2020).