Congress created the National Railroad Retirement Investment Trust (NRRIT) in 2001 to invest assets from the federal Railroad Retirement program in equities, expecting to improve returns and help fund expanded benefits. In designing the NRRIT, Congress tried to address concerns raised by policymakers and theorists about potential political influence on investment decisions that could create conflicts, lower the program's performance, and interfere with private markets. Proposals to use centralized investment to improve Social Security's financing have recently raised similar concerns. This article reviews management and governance aspects of the NRRIT as they relate to its political independence by focusing on the Trust's legal status, mandate, governing board characteristics, investment policy, and oversight. If Social Security were to adopt such an investment policy, examining the NRRIT's design and experience in these areas could provide useful guidance.
Acknowledgments: The author thanks Dale Kintzel, Kathleen Romig Krepps, Siona Robin Listokin, Patrick Purcell, David Rajnes, Mark Sarney, Dave Shoffner, and David Weaver for their helpful comments and suggestions.
CPP | Canada Pension Plan |
NRRIT | National Railroad Retirement Investment Trust |
OIG | Office of the Inspector General |
RRB | Railroad Retirement Board |
RRSIA | Railroad Retirement and Survivors' Improvement Act |
With the Social Security system facing a projected funding shortfall within the next several decades, policymakers have offered a variety of proposals to improve the program's long-term financial outlook. One option is to increase the yield on trust fund assets through centrally managed investments in equities that offer potentially higher returns, but pose greater risks, than federal government bonds.1
Insulating investment decisions from political influences would be a core consideration of such a policy. Critics argue that regardless of any potential financial benefits, trust fund investment would be problematic because of the probability of political interference (White 1996; Ostaszewski 1997; Ferrara 1980; Greenspan 1999).2 Public opinion polls reflect a similar skepticism about the feasibility of apolitical trust fund investment (NPR 1999).3 However, other researchers have suggested that with proper design and management features, a centralized Social Security investment component could maintain its independence (Munnell and Weaver 2001; Angelis 1998; Aaron and Reischauer 1998; Templin 2007; Munnell and Sundén 1999). A common thread among these analyses is the suggestion that principles of governance found in comparable public pension plans and other models can offer guidance on avoiding political interference.
One agency relevant to Social Security in this context is the Railroad Retirement Board (RRB), which administers a separate federal program providing retirement, disability, and survivor benefits for railroad employees. The Railroad Retirement system provides two tiers of benefits: the first is designed to provide the same benefit the worker would have received if he or she were covered under Social Security, and the second is intended to replicate a private-sector defined-benefit pension. The Railroad Retirement system's funding structure currently includes money invested in equities by the National Railroad Retirement Investment Trust (NRRIT) (RRB 2006). Funds controlled by the NRRIT are used for tier II benefits, supplemental RRB annuities, and in some cases, components of tier I benefits exceeding the benefits Social Security would provide (NRRIT 2010a).
This article describes the management and governance aspects of the NRRIT as they apply to each of five common characteristics of public pension programs. These governance aspects shape program investment operations, define their level of independence, and determine the manner in which they engage with political actors.
The five pension program characteristics are:
The article begins with a summary of the NRRIT's formation. A description of the NRRIT governance aspects of each of the five program characteristics follows, along with commentary on its relevance to political independence.4 This article does not argue for or against centralized investment; rather, it provides information about some of the governance issues that are unique to this policy option.
The NRRIT was intended to fund expanded benefits. In 2001, Congress passed the Railroad Retirement and Survivors' Improvement Act (RRSIA). The RRSIA expanded benefits under the Railroad Retirement system in a variety of ways and lowered tier II payroll tax rates. To help finance these programmatic changes, the legislation also established the NRRIT,5 which was authorized to receive assets from the Railroad Retirement account and invest them in a diversified market portfolio rather than in Treasury bonds (NRRIT 2010a).
The proposal's designers expected that nominal annual equity returns would be 2 percentage points higher than the existing bond investments—that is, 8 percent versus 6 percent (House Transportation and Infrastructure Committee 2001; Romig 2008).6 The NRRIT became operational on February 1, 2002, and began making investments in September of that year.
Prior to the passage of this legislation, critics raised concerns about the potential dangers of centralized equity investment. Writing about the proposed creation of the NRRIT, David John (2000) of the Heritage Foundation wrote:
Though the board managing this investment would be nominally independent, the assets in the trust would be under the control of political appointees and government bureaucrats. Giving bureaucrats the power to invest huge amounts of money in the stock market would create a fundamental conflict of interest between the long-term needs of future retirees and short-term political goals.
As the bill was under consideration, the executive branch also voiced its dissatisfaction. Lawrence Lindsey, director of the president's National Economic Council, wrote a letter to Congress stating the administration's strong opposition to having a federal retirement program invest in the stock market. In Congress, Senator Phil Gramm (R-TX) expressed similar concerns, arguing "I do not per se object to investing the money. I think there have to be protections for the railroad worker to be sure the Government doesn't direct the investments to benefit some interests other than the worker. There needs to be some firewall between the investment committee and the Government" (RPC 2001).
In establishing the NRRIT, legislators created a variety of organizational mechanisms designed to mitigate the types of problems that critics expected. As Romig (2008) notes in her analysis of the NRRIT's investment practices, Congress explicitly structured the Trust to try to ensure that the program's operations and investments would be free of political interference. The rest of this article explores these management and governance components.
The NRRIT is legally independent from the RRB and the federal government. The RRSIA created the NRRIT as an organization entirely separate from the RRB, with no role in the administration of benefits or any other operational aspects of the Railroad Retirement program. The act states that the NRRIT "is not a department, agency, or instrumentality of the Government of the United States" and establishes the organization as subject to the rules governing trusts that apply in the District of Columbia, not those at the federal level (NRRIT 2007b).7
When given the opportunity to bring the NRRIT within the purview of government entities, Congress has explicitly demonstrated its desire to maintain the existing separation. For example, in 2007, Congress rejected a proposal from the RRB Office of the Inspector General (OIG) to combine the auditing of the NRRIT with the RRB8 by stating that "the Railroad Retirement and Survivors' Improvement Act of 2001 mandates that the Trust functions independently from the Railroad Retirement Board" and citing the requirement for a nongovernmental audit of the program (House Appropriations Committee 2007).9, 10 The NRRIT's existing audit procedures and the OIG's concerns are discussed in more detail in the section on oversight.
Because Congress established the NRRIT to fund expanded benefits, legislators charged the program with maximizing returns to achieve this goal (House Transportation and Infrastructure Committee 2001). The law provides the NRRIT and its Board of Trustees with a clear mandate, holding that their actions should be "solely in the interest of the Railroad Retirement Board and through it, the participants and beneficiaries of the programs funded under this Act" (NRRIT 2007b).
The trustees are directed to carry out this mandate by investing with "care, skill, prudence, and diligence" according to requirements like those of the Employee Retirement Income Security Act of 1974 (ERISA) (NRRIT 2007b; DOL 2008). Importantly, the fiduciary responsibilities of ERISA have been interpreted as categorizing social investment objectives—one of the core reflections of political influence—as "collateral benefits," to be considered only when a proposed investment's risk-and-return profile is at least equivalent to the best available alternative (Doyle 1998).
The performance of other public pension funds demonstrates the role mandates can play in setting real-world investment practices. For example, the California Public Employees Retirement System (CalPERS) mandate allows consideration of noneconomic factors in its investment decisions, which influences its investments in environmental initiatives and emerging markets (CalPERS 2007a, 2007b). By contrast, the Canada Pension Plan (CPP), like the NRRIT, is directed to invest only to maximize returns and the CPP has used this mandate as an explicit source of pushback when the program has been pressed to consider social goals.11
The NRRIT is led by a professional board representing both labor and management interests. Each of the seven members of the Board of Trustees serves a 3-year term.12 Three of the members are selected by labor organizations to represent employee interests, three are selected by railroad carriers to represent management interests, and one is an "independent member" chosen by a majority of the other six members. Beyond the representation requirements, the legislation also establishes professional requirements for the trustees, mandating that they have "experience and expertise in the management of financial investments and pension plans" (NRRIT 2007b).13
Congressional records demonstrate that the diverse structure of the NRRIT's Board of Trustees was considered an important protection against political interference. As the bill was being debated in the House of Representatives, its sponsor, Don Young (R-AK), singled out the Board of Trustees' diversified membership structure as critical to preventing political interference. This point was echoed by Representative Jerry Weller (R-IL), who argued that the seven-member board would make certain that "any possible implication of a government role in investing is eliminated" (US Congress 2001).
The NRRIT's Board of Trustees, in consultation with investment experts, establishes the guidelines used to direct the fund's investment decisions. The NRRIT's first set of investment guidelines echoed their mandate to focus exclusively on returns, outlining their foremost objectives as being "(i) to ensure the timely and certain payments of benefits of eligible railroad retirement plan participants and beneficiaries, and (ii) to achieve a long term rate-of-return on assets sufficient to enhance the financial strength of the Railroad Retirement System" (NRRIT 2002).14
Upon the creation of the NRRIT, the Trust's cash assets were initially transferred into equities through index funds, using the Wilshire 5000 for US equities and the MSCI World Ex-US Index for global equities (NRRIT 2007a). Fixed-income assets were later added through the Lehman Brothers Aggregate index. The Trustees suggested that passive management was required during the early stages of the program because of the lower oversight requirement and lower administrative costs (NRRIT 2002). As the program has matured, the NRRIT has introduced specific, separately managed, nonindexed equity holdings. At the close of fiscal year 2010, 74 percent of the US equities held by the NRRIT were actively managed (NRRIT 2010b).
Outside investment firms handle the management of the NRRIT's assets in accordance with the Board of Trustees' responsibility set forth in the RRSIA to "retain independent investment managers to invest the assets of the Trust in a manner consistent with such investment guidelines" (NRRIT 2007b). The asset management firm Northern Trust now handles the primary administration of the NRRIT's investments.
The NRRIT's investment targets, which cover a broad array of asset classes, are:
US equities | 26 percent |
---|---|
Non-US equities | 22 percent |
US fixed income | 17 percent |
Non-US fixed income | 10 percent |
Private equities | 10 percent |
Real estate | 10 percent |
Commodities | 5 percent |
US equities constitute the plurality of the asset class targets, with non-US equities and US fixed-income investments representing the second and third largest allocations, respectively (NRRIT 2010c).
The Trust's investment policies also address the issue of corporate control to help mitigate concerns about the influence the NRRIT could have as a single large investor. The RRSIA compels the Board of Trustees, under normal circumstances, to diversify their holdings to limit the extent to which the NRRIT can influence the corporate operations of a single company, as well as to minimize losses (NRRIT 2007b). Adding to this general guidance, the NRRIT's investment guidelines mandate that none of the private investment managers employed by the Trust can control more than 10 percent of the program's assets (Romig 2008).15
Although independent, the NRRIT is overseen by various government entities. It is subject to oversight primarily through its Board of Trustees' obligation to submit an annual management report to Congress, which it also provides to the president, the RRB, and the Director of the Office of Management and Budget (OMB). The RRSIA requires that the report contain statements of financial position, operations, and cash flows; a statement on internal accounting and administrative control systems; an independent auditor's report on the Trust's financial statements; and any other comments and information necessary to inform Congress about the operations and financial condition of the Trust (NRRIT 2007b). The NRRIT submits the management report to the aforementioned entities and posts the report online for public viewing.16
The NRRIT is also party to a memorandum of understanding with the RRB, OMB, and the Treasury Department that requires the NRRIT to provide a monthly report of its basic financial operations, specifically "receipt and disbursement of funds, purchases and sales of assets, earnings and losses on investments, value of investments held, and administrative expenses incurred" (NRRIT 2002). These methods of oversight provide another role for government entities in monitoring the NRRIT, without offering any direct mode of control.
Beyond reviewing financial and management reports, the act gives the RRB additional oversight authority by allowing it to bring lawsuits against the NRRIT for two reasons:
(i) to enjoin any act or practice by the Trust, its Board of Trustees, or its employees or agents that violates any provision of this Act;
or
(ii) to obtain other appropriate relief to redress such violations, or to enforce any provisions of this Act (NRRIT 2007b).
To date, the RRB has not invoked this authority; however, the RRB's OIG suggests that the current audit process may be inadequate for identifying the full range of potential problems that would necessitate legal action (RRB 2008). The OIG argues that in addition to the existing financial auditing, the NRRIT should be subject to performance audits, which Szymendera (2010) notes would be used to analyze "program effectiveness, economy and efficiency, internal control, and compliance with the law."17 In its published Statement of Concern on the issue, the OIG contrasts the oversight of assets administered by the NRRIT with those of the Federal Retirement Thrift Investment Board, which is subject to performance audits from the Department of Labor under the rules of the Federal Employees' Retirement System Act of 1986 (RRB 2008).
Thus far, the NRRIT appears to have achieved the political independence Congress desired. However, the role played by any single governance component in keeping the NRRIT independent and apolitical is difficult to quantify based on existing records. Although program performance can be seen as showing that the management constructs established by Congress to guard against political interference have succeeded, it may also be that their ability to withstand such challenges has not yet been sufficiently tested. As the program matures, continued study and analysis will be necessary.
The NRRIT is an appealing case study for any policymaker or theorist examining trust fund investment because of the programmatic similarities between the Railroad Retirement system and Social Security. Foremost is that both are federal-level defined-benefit retirement programs (which alleviates concerns about the limited applicability of assessing political risk by analyzing a defined-contribution pension model such as the Thrift Savings Plan). However, one should be careful not to overstate the likeness between the programs in this regard. Although the Railroad Retirement system is primarily a defined-benefit program, the portion of its assets placed in diversified investments through the NRRIT largely funds tier II benefits. Because tier II benefits are designed to replicate a private pension rather than Social Security, there may be greater willingness—and a more compelling precedent—to invest them in nongovernment equities.
Another difference that could limit the NRRIT's applicability to Social Security is the relative size of the investments. Critics of investing the Social Security trust funds in equities often suggest that the magnitude of the Old-Age, Survivors, and Disability Insurance (OASDI) Trust Funds makes political interference more likely.18 Were Social Security to invest in equities, there is little chance Congress would authorize investing all trust fund assets, but even a modest percentage of these holdings would surpass the NRRIT's assets.19
No program will ever serve as a perfect analog for Social Security; but as policymakers continue to analyze potential solutions for Social Security's projected funding shortfall, program comparisons can provide useful information. Examining the five governance components of the NRRIT (legal status, mandate, governing board characteristics, investment policy, and oversight) provides valuable insight into the management structures that would need to be considered if Social Security were to adopt centralized investment in the private market.