For other versions of this document, see http://wikileaks.org/wiki/CRS-RL31551 ------------------------------------------------------------------------------ Order Code RL31551 CRS Report for Congress Received through the CRS Web Employer Stock in Pension Plans: Economic and Tax Issues September 4, 2002 Jane G. Gravelle Senior Specialist in Economic Policy Government and Finance Division Congressional Research Service ~ The Library of Congress Employer Stock in Pension Plans: Economic and Tax Issues Summary The loss of retirement assets held in Enron stock by Enron employees has stimulated proposals to restrict the holding of employer stock in retirement plans, and other proposals to regulate these plans. Stock in the Enron plan came from firm contributions in the form of stock that was not allowed to be sold and from voluntary investment by employees. This report focuses on rationales for providing employer retirement plans and for holding (or not holding) employer stock in these plans, both from the perspective of the private sector and of government policy. Retirement plans fall into two types: defined benefit plans, where a pension based on earnings and years of service is provided; and defined contribution plans, where individuals receive benefits based on accumulated principal and interest. Either plan can hold employer stock, but holdings are limited to 10% of assets in the case of defined benefit plans. These plans receive tax subsidies, as do employee stock purchase plans and certain types of stock options. The analysis suggests that there are economic reasons that firms and employees may engage in pension and profit sharing (or stock ownership) plans even in the absence of tax subsidies. Pension plans, primarily defined benefit plans, may be attractive for administrative and risk-reduction reasons, for dealing with inadequate investment in on-the-job training, and for smoothing the retirement of older workers. Stock options and stock ownership plans may be useful for addressing inconsistency in objectives between shareholders and managers and worker monitoring problems, although these benefits are not likely to accrue to stock ownership by the rank and file of large companies. These employee stock ownership plans may also deter hostile takeovers, which may undermine economic efficiency and stockholder interests. Stock contributions are also popular because they do not reduce cash flow, which has both benefits and costs from an economic efficiency perspective. Government subsidies to plans may be justified to increase retirement incomes and access to annuities because of a shortfall in optimal savings and certain economic problems with self selection in purchasing annuities. These objectives also underlie the justification of Social Security. Pursuing these goals may actually conflict with another worthy objective, on-the-job training. However, objectives that are addressed via employee stock ownership do not appear important in shaping the nature of large, broadly-based, retirement plans. Diversification of plan assets and prudent investment portfolios do appear consistent with rationales for government intervention. Attempts to address this issue might take several forms: restrictions on shares of employer stock in plans, prohibiting employer stock contributions or lifting restrictions on sale, denying a tax deduction on employer contributions until they could be sold, and requiring independent investment advice. The last proposal is not costless and could undermine participation for small firms. For administrative and other reasons, it may be rational to impose share restrictions on allocations of contributions, rather than on assets. There are no plans to update this report. Contents Basic Types of Plans . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Private Rationales for Pension and Profit-Sharing Plans . . . . . . . . . . . . . . . . . . . . 2 Administrative Savings, Insurance and Risk-Pooling . . . . . . . . . . . . . . . . . . 3 Human Capital Investment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 Facilitating Retirement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 The Principal-Agent and Worker-Monitoring Problems . . . . . . . . . . . . . . . . 4 Minimizing Cash Flow Effects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Rationales for Government Intervention . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Implications for Reform Proposals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Employer Stock in Pension Plans: Economic and Tax Issues The Enron bankruptcy caused many employees to lose a large portion of their retirement savings that was invested in Enron stock. Approximately 60% of Enron's 401(k) plan was in Enron stock. Some of the concentration was because Enron matching contributions were in stock, and that stock could not be sold until age 50 or upon leaving the company. However, a large fraction was due to employees choosing to purchase Enron stock. (The loss was somewhat exacerbated because of a blackout period when individuals were not allowed to trade because of a change in plan administrators.) Critics have also argued that Enron officials were touting the soundness of the stock as an investment (and deducting stock contributions at full value) even as they were aware of looming problems for the firm. While the Enron case was probably the most publicized instance of employee loss of retirement assets due to holdings of company stock, it is not an isolated case. Several legislative proposals have been advanced to address these problems including dramatically shortening (to three years) the required holding time for company donated stock, restricting the amount of employer stock held in 401(k) plans, and requiring firms to provide independent investment advice. Critics of these proposals have argued that some of these restrictions, particularly those restricting the ability of firms to contribute in the form of stock, would lead to reduced employer contributions. Moreover, proposals that might seem reasonable for large firms but that add cost and complexity (such as providing investment advisors), may reduce the participation of smaller firms in 401(k) plans. This report provides an economic perspective on the issue, focusing on the potential rationale for government intervention into employer retirement and profit sharing plans. Readers are referred to another CRS report that contains a more detailed discussion of the Enron bankruptcy, the general issues associated with the holding of employer stock by retirement plans, and legislative proposals: CRS Report RL31507, Employer Stock in Retirement Plans: Investment Risk and Retirement Security, by Pat Purcell. The next section of this report discusses briefly the kinds of retirement and profit sharing plans firms use. The following section discusses why pension and profit sharing (or stock ownership) plans might exist even without government intervention. Understanding those rationales is key to understanding the behavioral response to government regulatory changes, and determining whether government policy should further encourage or restrict such plans. The next section discusses economic rationales for government intervention and the final section discusses implications of these rationales for possible legislative revisions. CRS-2 Basic Types of Plans A broad range of plans that involve benefits to workers not in the form of cash currently exist, including a wide variety of plans that are for retirement purposes (and that employees cannot receive until they leave the firm). These plans receive tax benefits in that contributions and earnings are not taxed to individuals until received as pensions (or payments on separation). These plans fall into two basic types: defined benefit plans (where workers are guaranteed a certain benefit related to earnings and years of service) and defined contribution plans, where employees receive benefits based on the size of assets and accumulated earnings. For most types of plans, contributions to these plans are typically made by employers, and thus do not permit employees to choose between them and cash wages. For a variety of reasons, defined benefit plans, which once dominated the pension landscape, have been in a decline over the past quarter of a century and defined contribution plans are on the increase. This rise in defined contribution plans particularly reflects a popular form of defined contribution plan: a 401(k) plan, where individual accounts are maintained, both employers and employees contribute, and employee contributions are voluntary. Employees have some choice in the allocation of their own contributions. Employer stock may be one of the choices; employer contributions may also be made in the form of employer stock. While all pension plans are subject to regulations of some type, the restrictions are greatest for defined benefit plans. Defined benefit plans are also covered by pension insurance, which insures against total loss of assets. Some of the restrictions on plans, particularly defined benefit plans, were designed to insure sound investments and to make sure that the plans were not simply a tax shelter for high ranking employees and managers. Only defined benefit plans have restrictions on the amount of employer stock that can be held (10%). For defined contribution plans, employees bear the risk of loss of investment or low returns. Some types of plans with retirement features (Employee Stock Ownership Plans, or ESOPs) must hold assets primarily in the form of employer stock. Plans may be combined. Tax benefits are also provided for acquiring employer stock that is not restricted to retirement plans. These benefits include benefits for stock options and stock purchase plans: the former are generally directed to high ranking employees and managers, while the latter are generally available to all employees.1 Private Rationales for Pension and Profit-Sharing Plans Why should firms pay employee compensation in the form of pension or profit sharing plans (or other fringe benefits) rather than in cash? Does the existence of these plans hinge solely or largely on tax benefits? In this section we briefly discuss 1 See CRS Report RL31458, Employer Stock Options: Tax Treatment and Tax Issues, by Jack Taylor. CRS-3 several reasons that firms might desire to pay compensation, or that individuals might like to receive it, in the form of pensions or stock. The first three reasons are largely associated with defined benefit pension plans while the last two are associated with profit-sharing or stock ownership plans. Defined contribution plans not in the form of stock may, indeed, have largely arisen from a desire to exploit tax benefits, as it is difficult to discern another reason for using them. Administrative Savings, Insurance and Risk-Pooling One reason for the popularity of these retirement plans might be that collective plans, such as pension plans, could be considered desirable because of administrative savings and risk reduction. Pension plans are also often constructed to provide insurance elements, such as disability and survivor payments and life annuities. Insurance of this nature is often not efficiently provided in private markets because of adverse selection. (Adverse selection occurs when individuals sort themselves out due to private knowledge of risks. For example, a person with a terminal illness might desire to buy life insurance but would not purchase a life annuity.) While administrative savings could apply to any plan, risk-pooling and insurance elements would tend to apply most strongly to defined benefit pensions plans with diversified assets and would not apply to stock ownership plans which tend to increase risk. Administrative and risk reduction benefits have probably declined over time with the current availability of large mutual funds and the benefits are particularly less important for defined contribution plans that do not have unique elements of insurance and risk reduction associated with defined-benefit plans. Human Capital Investment Investment in human capital is generally under-provided in a market economy, because individuals cannot engage in involuntary servitude or commit to a certain form of employment to insure that future earnings are used to repay human capital investments. These problems apply both to general education and to certain types of on-the-job training. Returns to on-the-job training that teaches skills specific to the firm can only be exploited by remaining with the firm. However, returns to on-the- job training that teaches skills transferable to other jobs can be realized by either staying with the firm or moving to another firm. It is in the interest of the firm to retain employees when the firm has invested in their training and whose net product is low or even negative in the early period of the career. Defined benefit pension plans that are not quickly vested, and whose benefits become greatest when spending a long career with a company, can be used to allow firms and employees to mutually exploit gains from these types of human capital investment. (Government pension regulations which require early vesting and portability as ways to protect worker retirement security and guard against tax sheltering may, however, have eroded this benefit.) Facilitating Retirement Pension plans may also be a way of encouraging employees to retire as their marginal product falls, without undermining morale by firing older workers who have CRS-4 become less productive (or, these days, placing the employer in the position of potentially violating the law). Defined benefit pension plans are particularly effective because workers who have reached full retirement and remain at work forgo their pensions, and their net wage is reduced, creating a powerful substitution effect that encourages retirement while also offsetting lost income. Defined contribution plans permit retirement but do not create as powerful an incentive to retire and thus do not perform this function as well. The Principal-Agent and Worker-Monitoring Problems Another problem that large firms with many stockholders and/or employees face is the problem of a misalignment of shareholder and worker interests. In the case of managers, this problem is often referred to as the principal-agent or agency cost problem: managers who run the company as an agent for shareholders may make decisions that do not necessarily maximize stockholder profits. Moreover, large firms may find it difficult to monitor the performance of workers; in particular they cannot easily distinguish between the effects of work effort versus outside influences on productivity. One way that firms may attempt to remedy this problem is through use of stock options (which provide employees an incentive to increase the firm's value) and stock ownership plans, which provide some alignment with the shareholders' objectives. Of course, in theory, this approach would not be particularly beneficial for the rank and file employees of large firms, where additional work effort by any one employee would have a negligible effect on the value of that employee's stock. Stock options and ownership could have an effect on top management, whose actions have important consequences, and on closely held firms. Moreover, many managers believe that employer ownership boosts employee loyalty and morale.2 Employee stock ownership can also work against shareholder interests and economic efficiency. Firms where employees hold a large fraction of stock are more impervious to hostile takeovers, as employees and managers may otherwise fear loss of pay and jobs in such a circumstance. However, threats of takeovers are also a market mechanism that may keep the principal-agent problem under control and both takeover threats and actual takeovers may lead to a more efficient company. Reducing takeovers may be advantageous for managers and workers but may not be desirable socially. 2 There is an extensive literature on the effects of employee ownership. See Douglas Kruse, "Research Evidence on Prevalence and Effects of Employee Ownership," Presented in testimony before the Subcommittee on Employer-Employee Relations, Committee on Education and Workforce, U.S. House of Representatives, Feb. 13, 2002; Chris Doucouliagos, "Worker Participation and Productivity in Labor-Managed and Participatory Capitalist Firms: A Meta Analysis," Industrial and Labor Relations Review, Vol. 49, Oct. 1995 (which includes an extensive list of references); Linda A. Bell and David Neumark, "Lump-Sum Payments and Profit-sharing Plans in the Union Sector of the United States Economy," The Economic Journal , Vol. 103, May 1993. CRS-5 Minimizing Cash Flow Effects Providing compensation in the form of stock is often considered cheaper because it does not reduce current cash-flow. Such an approach may be particularly attractive to new and fast growing firms, where access to capital markets is difficult and initial profitability is low. This motive may lead to more economic efficiency if capital markets consistently overestimate expected risk. It may lead to less efficient markets if the dilution of stock makes information on the firm's profitability more difficult to assess by investors. Rationales for Government Intervention The previous section has suggested private motives even in the absence of tax benefits for defined benefit pension plans and for stock ownership plans, but there does not appear to be a strong private rationale for defined contribution plans not held in the form of employer stock. Clearly another reason for pension and profit sharing plans that may play a crucial role in encouraging these plans, particularly defined contribution plans not invested in employer stock, is the tax benefits associated with them. Amounts contributed to a pension plan and their earnings are not subject to tax until received as pensions, usually many years into the future. The combination of deferral of tax on the initial contribution and deferral of tax on earnings is the equivalent of an exemption of earnings from tax, and is thus a valuable tax benefit. Why should the government intervene with tax subsidies (and regulations) to shape the compensation package? And why should it intervene with mixed signals, including both benefits for and restrictions on ownership of employer stock in pension plans? The first set of rationales for supporting pension plans are also are among the reasons for establishing Social Security: adverse selection in annuity markets and failure of individual optimization (failure to save a desirable amount).3 Adverse selection in annuity markets occurs because individuals expecting to have a shorter life span will avoid the annuities market, thereby making the annuities unattractive for the average individual. This rationale would justify tax subsidies to pension plans that provide retirement annuities, early vesting, and mandates for broad coverage of employees. (The latter rule is probably necessary in any case to prevent the tax benefits from becoming a tax shelter for highly compensated employees.) Of course, it is not clear that individuals do not save enough, and it is not clear why using resources (in the form of reduced taxes) to encourage a private pension system that covers about half of workers should be preferred to devoting those resources to an expansion of Social Security. The entire pension system has been shifting away from plans that address these goals: many defined contribution plans have a lump-sum payoff option, and defined 3 See CRS Report RL31498, Social Security Reform: Economic Issues, by Marc Labonte and Jane G. Gravelle. CRS-6 benefit plans have been increasingly displaced by defined contribution plans. 401(k) plans, which have become very popular, allow voluntary, not mandatory, contributions. Outside of certain defined benefit plans, the current systems have simply become primarily ways to obtain tax benefits. Of course, arguments are made that these tax benefits have encouraged saving, but neither economic theory nor empirical evidence has confirmed that view.4 In any case, this rationale suggests that prudence in investment is important to encourage. In that case, many of the proposals for revision, including limits on employer stock holding in any type of pension plan, and reducing or eliminating any required holding period, might be justified. Of course, such an approach also suggests that employee stock ownership plan subsidies be discontinued or modified. Some of the rules made to insure safe and broadly available worker pensions have been in conflict with solutions to other economic problems. For example, as noted earlier, a market economy tends to under-invest in human capital. There are obviously massive government interventions through direct spending, low-interest loans, and tax benefits to provide for formal education and training. Pension rules may, however, have undermined the use of defined benefit pensions for encouraging spending on on-the-job training. For example, rules mandating early vesting are in conflict with objectives to increase on-the-job training (objectives that might be worthy of pursuit by the government as well), although they may be appropriate to increase pension coverage and employee security. Similarly, non-discrimination requirements have been used to prevent the use of the tax subsidies as a tax shelter for highly compensated employees and increase coverage, even though on-the-job training benefits may not be uniform across employees. Diversification of assets is consistent, however, with both government and private rationales for defined benefit pension plans. Economic problems arising from principal-agent costs or worker monitoring costs may be argued to justify government subsidies to stock ownership plans to increase efficiency beyond what private markets may do. However, they are unlikely to be addressed by ownership of employer stock among rank and file employees of large companies, where most tax subsidies for pensions are directed. Moreover, encouraging practices that make companies resistant to takeover may reduce rather than increase economic efficiency. Nor is it clear that stock ownership plans benefits in preserving cash flow outweigh the possible negative effects of using stock as compensation on stockholder information. Thus, the justification for government subsidies to stock ownership is not really established. 4 Considerable research on tax benefits for retirement, often focused on individual retirement accounts and 401(k) plans, has found mixed results. For a summary see CRS Report RL30255, Individual Retirement Accounts (IRAs): Issues and Proposed Expansion, by Jane G. Gravelle. CRS-7 Implications for Reform Proposals As discussed above, the discussion of justifications for government intervention based on economic problems suggests rationales for traditional pension plans, particularly those with defined benefits, but do not necessarily suggest a justification of tax subsidies for plans such as 401(k) plans, where participation by employees is voluntary. These subsidies may not even increase retirement saving. However, given that subsidies do exist, there does appear to be a case for increasing the safety of investments by limiting investment in employer stock, while the rationale for supporting employer stock ownership plans appears weak. If limits on ownership are to be imposed, how should constraints work? It would appear better for administrative and other reasons to impose the limits on the share of contributions made rather than the share of assets. A reason for limiting the share of assets in employer stock in defined benefit plans is in part because of pension insurance and because of limits on excess contributions, issues that do not exist for 401(k) plans. In the case of a defined contribution plan, a successful employer stock may simply become a larger part of the asset base because it is appreciating rapidly and forced sales may not be sensible. Other legislative changes might be requirements that employees be able to sell employer contributed stock immediately or within a few years (which is proposed in some legislation), disallowing deductions for stock contributions until employees are allowed to sell, or even prohibiting these types of contributions altogether. It is not clear whether these restrictions will be successful, however, if ESOPs remain as an option. Another legislative proposal that has been made to increase the security of individual investment plans (which might be used instead of explicit restrictions) is to require firms to provide investment advice, since firm managers may have a conflict of interest. Of course, investment advice is not costless, and it could significantly discourage the use of these plans in the case of small firms. Moreover, virtually any independent investment advisor would counsel against holding a large part of retirement assets in a single stock, but many advisors would suggest taking on general stock market risk. If explicit restrictions are imposed on ownership of a single stock, such individual advice might not be necessary. Alternatively, a general public information campaign may be considered. Moreover, the publicity associated with the Enron bankruptcy, along with other firm failures and the decline in the stock market, may be adequate to alert individuals to the need for portfolio diversification. One criticism that has been made of these proposals is that firms may respond by reducing their contributions. It is not clear that such a reduction would occur or should be a problem if it did (and many contributions are currently made in cash rather than stock). If stock contributions have any value, then they are likely to substitute for cash wages. It is not clear that policy should be concerned about employer matches in a risky asset as a substitution for cash wages, particularly when the participation is voluntary, both on the part of individuals contributing to plans and firms setting up plans. The beneficiaries in both cases are only part of the population (and the more affluent part). An argument could be made that the revenue derived from increased taxes on cash wages might be better used for other purposes. ------------------------------------------------------------------------------ For other versions of this document, see http://wikileaks.org/wiki/CRS-RL31551