Statement of Sylvester J. Schieber*, Vice President, Research and Information,Watson Wyatt Worldwide
Testimony Before the House Committee on Ways and Means
Hearing on Retirement Security and Defined Contribution Plans
February 26, 2002
Mr. Chairman and members of the Committee on Ways and Means, I am Sylvester J. Schieber, Vice President of Research and Information at Watson Wyatt Worldwide. I am testifying today on issues regarding Retirement Security and Defined Contribution Plans. My comments are my own and do not reflect those of Watson Wyatt Worldwide, or any of its other associates.
I have spent more than 30 years studying the retirement systems in the United States and elsewhere around the world. I understand why there are concerns today about the retirement security system in this country and specifically about the operation of employer-sponsored defined contribution plans given recent developments. But I would like to begin by cautioning the members of this Committee and other members of Congress against doing anything that jeopardizes an extremely robust and resilient element of our retirement system.
I firmly believe in the importance of public policy in regulation of employer sponsored retirement plans. I believe that the Employee Retirement Income Security Act (ERISA) has done a great deal to improve the retirement prospects of millions of workers in this country. But I also believe there is strong evidence that the over regulation of pensions during the 1980s and early 1990s led to the reduction in the availability of pensions and to drastic changes in the sorts of plans that have been offered to workers. In other words, I believe regulation is important but that over regulation is potentially counterproductive.
Today, many people are concerned about the risks associated with defined contribution plans and would saddle these plans with new sets of requirements, restrictions, and expenses in order to ameliorate such risks. A significant problem, however, is that the reduction of current risks in these plans has the potential to create another set of risks for them and their participants. In highlighting the recent concerns about defined contribution plans, much has been said about these plans and others that is very misleading and has the potential to result in the development of bad public policies. I am concerned that such policies might lead to the curtailment of plans in the short term with a long-term result that few would consider appropriate or desirable.
Prior Evidence on the Importance of Regulation
Nearly 10 years ago, Professor John Shoven of Stanford University and I presented a paper at a policy conference here in Washington, DC, that analyzed the implications of pension funding restrictions that had been imposed on private sector employers during the 1980s and 1990s.[1] Our analysis concluded that these policies had significantly delayed the funding of pension obligations for the baby boom generation of workers and would ultimately result in much higher costs to employers than if prior rules had been left in place. We suggested the implications of these policies were likely to be adverse to the pension prospects of baby boomers. At the conference when we first presented our paper, a policy analyst from the Department of Labor suggested the implication of our analysis was simply that employers would have to contribute more to their pension plans late in the baby boomers’ careers than under prior funding regulations. We observed that there was an alternative prospect that employers might simply curtail their defined benefit plans as the delayed liabilities came due. I believe that there is strong evidence over the past decade that our concerns about the potential curtailment of private defined benefit plans were well founded. I am convinced that public policy has played a major role in what has transpired.
In a subsequent policy paper that Professor Robert Clark of North Carolina State University, a colleague of mine at Watson Wyatt Worldwide, Janemarie Mulvey, and I wrote, we analyzed the effects of pension nondiscrimination rules on private sector pension participation.[2] In an effort to prevent plan sponsors from targeting tax benefits accorded pensions to high-wage employees, Congress established nondiscrimination standards that require employers to include a wide range of workers in pension plans in order for these plans to achieve tax-qualified status. In addition, regulations have been introduced to limit maximum benefits to high-income workers and to restrict the integration of pension benefits with Social Security. The objective of these nondiscrimination rules has been to increase the participation rate of low-wage workers while limiting the loss in tax revenues associated with benefits to highly paid employees.
In our analysis, we examined changes in pension coverage rates between 1979 and 1998 to determine if the absolute and relative participation of low-wage workers increased following the implementation of new, more restrictive nondiscrimination standards adopted during the 1980s. In our analyses, we found no support for the hypothesis that more restrictive discrimination rules forced or enticed employers to provide pensions to low-paid workers. Participation rates for low earners simply did not rise in absolute terms or relative to the participation rates of high-wage workers following the implementation of new standards.
These new nondiscrimination standards along with other pension regulations have increased the cost of providing pensions. We showed in our analysis that in many cases, the administrative costs associated with government regulation of employer-sponsored plans can exceed the tax advantage of pension saving for workers at lower pay levels especially in smaller plans. As a result, it is not surprising that many small employers terminated defined benefit plans over the past two decades. This indirect effect of these regulations is one of the reasons that participation rates of low-income workers have remained relatively low.
Administrative costs are a disincentive for employers to provide pension coverage to low-income workers. Yet, most of the legislative efforts aimed at increasing participation have actually increased the regulatory burden to employers and thus their overall administrative costs. In reality, these regulations have done little to increase participation among low-wage workers over the past twenty years. Workers at low and middle earnings levels actually experienced small declines in pension participation following the adoption of these regulations. If Congress wants to expand participation for low-income workers it should look for ways to reduce, rather than increase, the regulatory burdens on employers.
Recent Developments and the Need for New Regulation
A renewed awareness of the fragility of our retirement system has arisen from a number of public accounts of Enron employees losing their retirement savings as their employer plunged into bankruptcy late last year. Remarkably less has been said about what happened to their defined benefit savings. A widely published problem in this case was that many Enron employees had invested most, if not all, of their 401(k) assets in Enron stock. To further complicate a bad situation, it appears that the most senior managers in Enron encouraged workers to buy Enron stock even after they became aware of the likelihood that the company was in financial peril. This combination of problems was further exacerbated by the fact that the participants in the Enron 401(k) plan had gone through a blackout period when they could not sell the Enron stock in their plan during a period when the value of the stock was plunging. This latter situation arose because of a shift from one plan administrator to another. And finally, to add insult to injury, the senior managers in Enron are reported to have been selling substantial blocks of Enron stock at exactly the same time the rank-and-file employees were trapped in the blackout on selling the Enron stock in their 401(k) accounts.
Out of this situation several proposals have evolved that would limit the exposure that employers could impose on workers to employer stock in their 401(k) plans. Other proposals would require certain communication with workers. There have even been proposals that we adopt some sort of benefit insurance covering defined contribution plans that would be similar to the insurance protection provided to participants in defined benefit plans through the Pension Benefit Guaranty Corporation (PBGC). Before turning to an assessment of the policy proposals, it is important to put some facts on the table regarding the demise of the Enron 401(k) plan and the general situation of 401(k) plans.
One of the concerns arising out of Enron is that employers are forcing their employees to hold their stock in their 401(k) accounts and thus putting them at excessive risk in terms of their retirement security. The risk to retirement security comes partially from over concentration in a single stock but is exacerbated by the correlation with employment risks associated with employers that go bankrupt. In other words, the employees at Enron faced double jeopardy as the company went bankrupt—they not only lost much of their retirement security they also lost the security of their existing jobs. There are two issues here that are important. The first of these is the extent to which workers are forced to hold company stock. The second is the extent to which their retirement security is at risk because their retirement portfolio is not sufficiently diversified in the assets securing it.
While there may be a misperception about the case, the fact is that most of the company stock that Enron employees held in their 401(k) plan was there at the employees’ discretion. Ignoring for the moment, the blackout period, Enron workers in the plan were not precluded from selling most of their employer’s stock and buying some other financial security. There may be three potential explanations for why the workers in this case held so much Enron stock in their 401(k) portfolios. One is that they had been misled about the potential performance of the stock in the future relative to alternative investment options. Second is that they did not appreciate the risks associated with investing in a single company’s stock. Third is that they knew there were downside risks but perceived the upside potential outweighed the cost of taking the risk of investing heavily in Enron.
To the extent that workers were duped into buying Enron stock because senior management in the company was misleading them about the prospects of the company’s performance, there are already Securities and Exchange Commission rules on what senior managers of publicly traded corporations can tell any potential investors in their stock. If these rules were violated, the senior managers who violated them should be prosecuted to the maximum extent possible. If a thief on the street who broke into Enron employees’ or executives’ homes is subject to prosecution, mandatory sentences including three-strike rules, and lengthy jail time, any thieves stealing from retirement plans should be just as subject to the same potential punishment. While the SEC might need to beef up accounting and disclosure rules, the best deterrent to protect 401(k) plan participants from corporate managers who mislead them about the prospects of their companies might be vigorous enforcement of existing laws.
There is some likelihood that Enron employees and many other employees around the country do not appreciate the risks they take on in investing heavily in their employer’s stock, especially in doing so in their retirement plans. This problem can either be addressed by providing more education for workers or by imposing limits on them in terms of the extent to which they can buy their employers’ stock through their 401(k) plans. While the latter approach might be the more effective one from the perspective of an enlightened regulator, I would caution policymakers from rushing headlong into this approach. What seems enlightened from the perspective of Washington, sometimes seems less so outside the beltway.
This leaves us with a question of whether we should restrict employees who are not misled about their employer’s financial prospects and who understand the risks associated with investing in a single company’s stock from investing most or all of their 401(k) assets in that stock. I believe that one of the strongest aspects of the 401(k) system in the United States is the sense of ownership that workers have in the programs. Workers are adamant that the assets in their retirement accounts are theirs. Next to the basic freedoms we enjoy in this country, property rights are something that we guard with tremendous fervor.
For every Enron where employees have lost most of their funds from investing in their employer’s stock, there are many other examples of employees in other companies who have done very well over extended periods of time by voluntarily investing in their employers’ stock. Prohibiting workers from investing their retirement money in the assets they wish to invest in has the potential to create an adverse public outcry that policymakers ought to seriously consider before they adopt restrictive regulations in this area. You might recall that during the debates over tax reform during the mid-1980s that both the Reagan Administration and the Chairman of the Ways and Means Committee entertained proposals to restrict 401(k) plans that were quickly abandoned when workers voiced their displeasure en masse.
One of the problems that we face in devising limits that protect 401(k) participants is the highly variable set of circumstances under which these plans are offered. In some cases, employers offer their 401(k) plan as a supplement to a relatively generous defined benefit plan. In others, it is the only retirement saving vehicle the company offers. If an employer has a defined benefit plan that in combination with Social Security provides career workers with pension annuities that allow them to maintain preretirement standards of living, what risks to their retirement security do workers pose when they invest their 401(k) assets in employer stock? Even in cases where workers are predominantly dependent on their 401(k) savings for retirement, there are tremendous differences in the risks associated with investing in company stock at ages 25, 35, 45, or 55. How do you control for those in setting rules limiting where workers can invest their retirement assets?
Insuring Against Risk in Defined Contribution Plans
Going beyond simply limiting where employees can invest their 401(k) retirement funds, some policy analysts are now advocating that we actually insure the investment performance in these plans. The argument here is that the insurance guarantee provided to defined benefit participants is the equivalent of a minimum investment return guarantee. If the government is going to be the insurer of one sort of plan, then why not the other. Indeed, cash balance plans are insured under the PBGC and basically insures the implied rates of return on these plans.[3] There are several problems with this logic and the proposals that flow out of it.
First of all, the argument that insuring investment performance in a defined contribution plan with participant-directed investment and insuring benefits in a defined benefit plan are equivalent is far fetched. The PBGC insures benefits only in cases of bankruptcy resulting in the inability of a pension sponsor to pay promised benefits under the plan. In cases where the insurance comes into play, the PBGC has a claim against any residual assets in the sponsoring company. This insurance is provided in conjunction with a stringent set of funding requirements and variable premiums that seek to entice if not force plan sponsors to keep asset levels in the plan at roughly the level of liabilities that exist within them. Adverse experience in the investment of the assets in these plans does not trigger an insurance payment by the PBGC, it triggers added contributions on the part of plan sponsors. Even in cash balance plans, the plan sponsor’s failure to realize rates of return on plan assets that are as high as the credited rate of return on the notional accounts has to be made up with added sponsor contributions.
In a defined contribution world, the provision of similar insurance to that provided in the defined benefit world would conceivably put the employer in the position of being the insurer of first resort. Most of the employers who were motivated to shift from offering defined benefit plans to offering defined contribution plans because of their unwillingness to accept investment risks in retirement plan sponsorship would likely quit offering plans. Those that continued to offer them would likely move back toward a highly restricted set of investment options in their plans. In the early days of 401(k) plans much of the investment was in guaranteed investment contracts (GICs) or similar instuments that paid relatively low fixed rates of return over the long term. In part, the move to self-directed investment in these plans was the result of workers wanting the higher returns from more aggressive investment that plan sponsors were not willing to pursue directly with their employees’ vested account balances.
The problem here cannot be diversified away. Figure 1 shows the variability in annual nominal returns payable to investors in broad stock or bond indexes in the United States between 1942 and 2000. Over the period shown, the average return on the S&P 500 index fund was 14.6 percent per year compared to 5.8 percent per year for the bond fund. But the volatility in the stock fund, as measured by the standard deviation of the historical returns, was also higher at 16.5 percent compared to 9 percent for the bond fund. Workers want the higher returns over time they seemingly get from investing in stocks, but employers are unwilling to take on the added risks associated with investing in stock to provide these higher returns.
Figure 1: Annual Returns from the Standard and Poors
500 Stock IndexI
ncluding Dividends and from an Index of U.S. Ten-Year Treasury Bonds
Source: Derived by Olivia Mitchell and Marie-Eve Lachance, Wharton School, University of Pennsylvania.
The advocates of providing some sort of return guarantee in defined contribution plans argue that by setting up cash balance plans, employers have demonstrated they are willing to provide such guarantees. But these advocates ignore that employers have imposed a relatively heavy price on participants when they provide return guarantees in these plans. In data we have gathered on approximately 120 cash balance plans, two-thirds of them provided interest credits at the equivalent to either the consumer price index rate or some federal bond rate. A number of others had fixed credit rates that were even lower than federal bond rates. It is highly unlikely that the majority of 401(k) participants would be willing to accept a guaranteed rate of return at such a steep price.
If the federal government is going to provide this insurance instead of attempting to force employers to do it, it would almost certainly mean the creation of some sort of pooled account with centralized administration. Even if we were willing to create such an entity, it is not clear that policymakers would be willing to impose the price of return guarantees on participants. In fact, President Bush’s recent Social Security Commission considered some sort of return guarantees for the individual accounts created in the Social Security reform options they devised. But the Commission did not include a guarantee in any of its reform options. In large part, the Commission members thought the cost would be too high to guarantee returns in this sort of program.
If we can figure out the mechanism for providing investment insurance, it would still mean a radical reorientation of the investment of assets in these plans. If we allowed the current method of investment to persist along with an investment return insurance program, we would create a tremendous moral hazard situation. If I know that I have a large up-side potential from pursuing a risky investment strategy but realize that I have little downside exposure because of the insurance program, then why would I do anything but pursue the risky strategy? I would accrue all the benefits of such an approach and the insurer would sustain all the risks.
Making Defined Contribution Benefits More Secure
In his State of the Union Address this year, President Bush noted the public concern about 401(k) plans that has arisen out of the Enron bankruptcy situation. He has formed a task force including the Secretaries of Treasury, Labor, and Commerce to develop new safeguards for these plans. The President has recommended that workers be given greater freedom to diversify and manage their retirement funds; that corporate managers be restricted in their ability to trade company stock during 401(k) trading blackout periods; that workers be given quarterly information on their asset balances; and that they be given more access to investment advice. While the Bush Administration has not put forward specific legislation, a bill that has been introduced by Representatives Rob Portman (R-OH) and Benjamin Cardin (D-MD) would substantially cover the principles that have been laid out by the President.
In some regards, it is regrettable that any new restrictions have to be put on these plans as the track record they have achieved is remarkable. Where plans are offered, 70 to 80 percent of eligible workers participate in them. Total contributions going into these plans equal 8 to 9 percent of pay.[4] Jim Poterba, Steven Venti, and David Wise estimate that by 2030 the 401(k) system in the United States will be generating retirement benefits that are larger than Social Security.[5] In other words, this totally voluntary system has the potential to completely outstrip Social Security in terms of aggregate benefit delivery by 2030, a only half century after the first plan was put in place. On a totally voluntary basis it will outstrip the government program that requires more in tax revenue to support it than any other government program. The 401(k) system is so admired or envied by policymakers elsewhere in the world that other countries are moving to implement similar programs. Germany and Japan recently adopted systems that seek to mimic ours to a considerable extent. We should be very careful about doing anything that jeopardizes this system.
As a matter of public policy, I believe that absolute restrictions on the amount of employer stock a worker can hold in his or her retirement savings account will cause a strong adverse reaction on the part of plan sponsors and participants and is not warranted. Employers use their benefit programs for a variety of purposes and they use them in combination to attract, retain, and motivate workers. Providing matching contributions in the form of employer stock is one tool that employers have in achieving their goals. Employees in successful companies, often seek to participate in some of the benefits of that success beyond simply taking home a paycheck. Our research suggests that companies with higher levels of employee ownership of stock generally out perform those where employees do not have such a financial interest.[6] The success of our economy, the labor markets, and the growth of retirement saving over the period since 401(k) plans have come into operation highlight the reason we should be wary of adopting any massive overhaul of the 401(k) system.
While I oppose restrictions that would preclude workers from freely investing in their employers’ stocks, I am sympathetic to the argument that a workers’ vested benefits in their retirement plan are an economic asset intended to secure their retirement needs. As such, the ability for anyone to dictate that such assets be invested in a particular way should be limited. Some employers may be unhappy that such restrictions might limit their ability to give workers a vested interest in the success of their organizations. If the new restrictions do not include absolute limits, however, good companies will still be desirable places for workers to invest. Like many other aspects of the organization of our economy, this requirement will place an added premium on good management, but it is good management of our private sector businesses that has made our economy such a dominant force in the world.
Given the growing dependence of American workers on the accumulating balances in their retirement savings plans, any effort to provide them with more information about the appropriate investment behavior should be favorably considered. As with many things in life, however, retirement savings plans are often offered by small employers or in highly competitive environments where lavish budgets to provide extensive communication and investment advice are limited. We do not want to relearn the lessons of the 1980s that too much regulation leads to fewer plans rather than more security in the ones that already exist.
Finally, any provisions that seek to provide guaranteed returns in these plans should be viewed with an extremely wary eye. I cannot think of any single policy change that would have the potential to so radically alter the landscape of our retirement system in an adverse way. If this guarantee is going to be foisted on employers, policymakers should expect to see a significant exodus of sponsors from offering plans. If the federal government is going to establish and run such a program, policymakers should have a full understanding of the costs involved in it and who is going to be assessed those costs.
*The opinions and conclusions stated here are the author’s and should not be construed to be those of Watson Wyatt Worldwide or any of its other associates.
[1] This paper was first presented at a conference during September 1993 and was subsequently published in Sylvester J. Schieber and John B. Shoven, “The Consequences of Population Aging on Private Pension Fund Saving and Asset Markets,” in Sylvester J. Schieber and John B. Shoven. eds., Public Policy Toward Pensions (Cambridge, MA: The MIT Press, 1997), pp. 279-246.
[2] Robert L. Clark, Janemarie Mulvey, and Sylvester J. Schieber, “The Effects of Pension Nondiscrimination Rules on Private Sector Pension Participation,” in William Gale, John Shoven, and Mark Warshawshky, eds., Public Policies and Private Pensions (Washington, DC: The Brookings Institution, 2002 forthcoming). Until released in its published form this paper may be found at: http://www.watsonwyatt.com/progress_files/whitepapers/wp-05.pdf.
[3] Regina T. Jefferson, “Rethinking the Risk of Defined Contribution Plans,” Florida Tax Review (2000), vol. 4, no. 9.
[4] Robert L. Clark, Gordon P. Goodfellow, Sylvester J. Schieber, and Drew Warwick, "Making the Most of 401(k) Plans: Who's Choosing What and Why," in Olivia S. Mitchell, P. Brett Hammond, and Anna M. Rappaport, eds., Forecasting Retirement Needs and Retirement Wealth (Philadelphia: University of Pennsylvania Press, 2000, p. 104.
[5] James M. Poterba, Steven F. Venti, and David A. Wise, “401(k) Plans and Future Retirement Patterns,” American Economic Review (May, 1998), p. 183.
[6] Watson Wyatt Worldwide, Human Capital Index (Washington, DC, 2001), p. 5.