According to the latest figures from the US Census Bureau, retirement funds for public sector employees are distributed across some 2,670 retirement systems in the United States.2 Of these systems 2,451 are administered by local governments with more than 70 percent being super- vised by municipalities. In terms of membership approximately 90 percent of all participants are covered by state systems which paid benefits to more than 6 million beneficiaries in Fiscal Year 2001–2002. By comparison, more than 1 million beneficiaries received periodic benefit payments from local government pensions during this same period.
Unlike the US Social Security system, Public Employee Retirement Systems (PERS) are not pay-as-you go schemes and are jointly funded by contributions from employers and employees as well as returns on assets that are invested in financial markets. Most systems house Defined Benefit plans (DB), meaning that (a) capital along with any accumulation is later distributed according to a predetermined formula and (b) benefits, in the event of unfunded liabilities resulting from contribution and/or investment shortfalls, are essentially guaranteed by the ability of governments to raise contribution rates or to tax and borrow as needed.3 The tremendous size of these institutional investors (currently estimated to be in excess of
$2 trillion), in combination with the fact that most are overseen by a board of trustees, in a public setting, have resulted in a stream of administrative and policy related econometric research that examines PERS governance practices, investment decisions, and financial performance.
17.2.1 Governance Practices
To some extent public pension researchers face inherent ambiguity when trying to provide an explicit definition of governance. This ‘‘fact’’
within the broader realm of social science inquiries confronts other scholars in public policy and administration as well (Lynn et al., 2000, 1–2). How ever, when the term is used for retirement system research at the system level, ‘‘governance’’ typically refers to the structure of a pension plan board in combination with the complex of rules and practices that guide the
oversight of fund assets (Useem and Mitchell, 2000, 490). Given the nature of the field this rather broad definition appears to be necessary if not suffi- cient for two interrelated reasons.
First, while trustees typically have a fiduciary responsibility to represent the interests of plan participants there is no federal law requiring them to do so. Instead PERS fiduciary responsibility is usually left to state or local laws that vary across jurisdictions (Coronado et al., 2003, 580). And, as noted by Useem and Hess (2001), the trustees of public retirement systems, as well as the legislative bodies overseeing them, have interpreted their duties in varying fashions (133–134). By comparison, private sector pension representatives are obliged to conduct themselves in accordance with the Employee Retirement Security Act of 1974 (ERISA) which requires trustees to act with the care, skill, prudence and diligence of a prudent person acting in the interests of participants and beneficiaries (Miller, 1987, 8).4 This variation in fiduciary interpretations (within the public realm) and responsibility (across the sectors) has prompted researchers to examine questions concerning this particular aspect of governance and whether or not there is any deterministic impact on investment decisions and/or financial performance. Historical conclusions regarding superiority along these dimensions appear to favor the private sector. However, recent evidence suggests that this disparity may be diminishing.5
The second reason for proclaiming the definition as appropriate is that the characterization intrinsically acknowledges that those pursuing such inquiries also confront the theoretical and/or practical economics of pub- lic organizations.6 For example, researchers focusing on theoretical issues surrounding publicly managed retirement schemes have often sought to determine whether agency costs are associated with how PERS are gover- ned. Agency costs are economic inefficiencies that exist when the decision rights over an organization’s assets and cash flows are improperly aligned with the organization’s residual claimants (Nofsinger, 1998, 89). Since the majority of public pension systems house DB plans the residual claimants are generally deemed to be taxpayers who ultimately benefit/suffer from good/bad pension fund performance.7 The potential for an agency prob- lem arises because PERS boards of trustees are often comprised of mem- bers who are appointed or elected by plan participants or officials with little or perhaps no fiduciary responsibility towards the larger public.
On the other hand practically oriented research has tended to focus on the competence of trustees and the potential for administrative costs via governance practices. To illustrate, research on both the private and public sectors indicates that the size of boards, operating in any capacity, can matter in terms of operating efficiency and performance. In terms of pension funds, Useem and Hess (2001) note that having too few members denies
An Econometric Assessment of State and Local Government g 467
a board experience, expertise, and wisdom, while too many members can undermine communication, consensus, etc. (p. 137).8
Beyond this aspect, PERS representatives are usually not professional money managers and may be unfamiliar with the intimate technicalities surrounding financial markets and investment decisions (Useem and Mitchell, 2000, 47).9 Again, in terms of comparison, an important point to note is that these types of board structures do not exist in the private sector, which are usually overseen by financial professionals or officers of the sponsoring corporations. Hypotheses concerning the competence of PERS trustees tend to posit that the lack of expertise impacts both investment decisions and financial performance in the public sector. The latter of course is arguably an administrative cost from an economic perspective.
17.2.2 Investment Decisions
The examination of investment decisions by PERS can be classified as one of two types. The first is best described as investment policy decisions.
These types of decisions often involve choices concerning general system approaches, overall goals, or even authoritative controls. To illustrate, in addition to constitutional provisions or statutory limitations, many systems have internal policies placing ceilings on the percentage of assets that can be invested in stocks, the bonds of any single issuer, or foreign investments.
Some funds may even prohibit investing in specific types of companies, such as those selling tobacco or doing business with Northern Ireland.10 Others might require that some portion of a system’s assets be invested in a home state for economic development purposes. Investments of this variety are commonly referred to as Economically Targeted Investments (ETIs).
The second type of decision is more specific in that the particular use or allocation of a pension’s funds are involved with the specific intent of generating a rate of return on existing assets. Therefore, compared to investment policy decisions, these types of choices are typically referred to as investment strategy decisions.11 Some important examples include determining (a) the distribution of available funds among asset classes such as stocks, bonds, cash, etc., (b) whether or not to invest in foreign investments, (c) whether or not to invest ‘‘tactically’’ based on prevailing economic conditions or ‘‘passively’’ in broader indexes that track particular markets, and (d) the fraction of funds that should be managed internally by PERS representatives or externally by professional money managers.
The analysis of investment decisions is important because at this juncture researchers studying PERS not only face the economics of public organi- zations but the economics of financial markets. Within this realm time and uncertainty are dominating features and the empirical financial literature
indicates that investment choices can have a very substantive impact on financial performance. This has shown to be most true for asset allocation decisions which are often cited as the most important determinants of financial performance. For example, in two widely cited studies Brinson et al. (1991) and Brinson et al. (1986) demonstrate that 90 percent of an investing fund’s financial performance over time can be explained by asset allocation decisions. In an extension of these investigations Ibbotson and Kaplan (2000) go on to show that asset allocation choices explain 40 percent of the variation in financial performance among investing funds.12
In addition, the analysis of investment decisions is valuable because choices reflect prevailing PERS views regarding overall financial and social goals and the efficiency of financial markets to allocate financial resources.
PERS concerned with social issues may very well have policies prohibiting or limiting the placement of funds in specific investments, while attempting to enhance local economies with the infusion of financial resources.13 Other funds may not believe that markets are truly efficient in terms of asset pricing and seek to employ tactical investment strategies.14Concomi- tantly, trusts that do believe in relatively efficient markets might pursue longer term investment choices. Conservative pensions could opt for the safety of principle over the extent of income (Donner, 1939, p. 10). Finally, decisions to place a fund’s assets under external management might reflect the sentiments of PERS trustees or representatives about their own investment abilities, net of cost. The real question, of course, is how any one of these decisions affects the most inherently valued outcome of financial performance.