Issue Chairman Lau, Vice Issue Chairman Turner and Members of the Advisory Council, thank you for inviting me to testify before you today on this important issue.
I am David C. John, a Senior Research Fellow at the Heritage Foundation and Deputy Director of the Retirement Security Project. I also serve as senior policy advisor to Retirement Made Simpler—and it is in that capacity that I appear before you today.
Retirement Made Simpler is a coalition formed by AARP, the Financial Industry Regulatory Authority (FINRA) and the Retirement Security Project (RSP). The coalition was created in November 2007 specifically to inspire and support employers who want to help their employees save more for retirement through adding automatic features to their retirement plans. By providing companies with the tools and information they need to better understand and embrace such features as automatic enrollment and automatic escalation, RMS believes that we can help more Americans achieve a safe and secure retirement.
The timing of this meeting—with its focus on 403(b) plans and best practices of 403(b) plans, including automatic features—is opportune, and I would like to thank Jack Towarnicky and others on the Council for inviting RMS to participate in this hearing.
The gyrating financial markets that we have experienced recently encourage the wrong type of employee inertia—the kind that keeps employees who are not already participating in a retirement savings plan on the savings sidelines. Those markets also encourage some employees to either stop contributing or to make sudden changes in their investments. Both responses keep workers away from the incremental and long-term investing that is essential to building a sufficient retirement nest egg. This failure to save comes despite new evidence from Fidelity that participants who remained in their 401(k) plans, kept the same level of savings and maintained some level of equity allocation had better outcomes between September 30, 2008 and June 3 of this year than those who did not.
The best way to encourage employees to maintain steady savings patterns is automatic features. Automated features—from automatic enrollment to automatic escalation—lay the groundwork for a different kind of inertia: the kind that starts employees participating in their organization’s retirement plan early, generally in a balanced or lifecycle fund. And when properly designed, automatic plan design allows employees to save enough to lay the groundwork for a comfortable retirement.
As you are aware, the 403(b) plan landscape is different than it was even just a few years ago. Legislative and regulatory actions have put employers squarely in the center of responsibility for 403(b) plans. The changes have made 403(b) and 401(k) plans virtually synonymous. In fact, 403(b) plan providers are in very much the same situation that 401(k) plan providers were in when the Pension Protection Act was passed in 2006. They are asking the same questions, actively looking for guidance, seeking out best practices but, most important, starting to adopt automatic features with increased frequency. According to the Profit Sharing/401(k) Council of America’s 2010 403(b) Study, 11.5 percent of plans have adopted automatic enrollment. That percentage rises to 22 percent for plans with 1,000 or more participants. With an automatic plan, employees are enrolled at a pre-set contribution rate into a preselected investment fund, and are in the plan unless they opt out.
My comments today can be boiled down to one major point: The success that automatic plan design has experienced with 401(k) plans is replicable with 403(b) plans also. While there are some technical differences between 401(k)s and 403(b)s—and while there are also workforce composition differences—for instance many teachers, healthcare workers and employees in the nonprofit sector have 403(b) plans—automatic plan design will yield similar results in either environment. Specifically, automatic features can be expected to boost participation rates and increase savings rates.
Furthermore, as with 401(k) plans, automatic features are likely to have the most impact on lower wage employees who tend to participate and save at a much lower rate than higher wage employees. Automatic enrollment and escalation have a special value to minority and moderate income employees. In general, these groups tend to be much less likely to participate in a retirement plan—be it 401(k) or 403(b). However, studies show that automatic enrollment can dramatically increase participation by the four groups—women, minorities, younger workers and moderate-income employees—most likely to under-save. Automatic enrollment can increase participation in these four groups from about 1 out of 5 to 4 out of 5 and above.
South Dakota Case Study
A significant 2010 study directly illustrates this point. The study was commissioned by Retirement Made Simpler and conducted by the Center for State and Local Government Excellence. It examines the background and passage of automatic enrollment legislation in South Dakota and its initial impact on participation rates in the state’s defined contribution retirement plan. Ninety-one percent of new, eligible employees are participating in the SRP in units that adopted automatic enrollment, versus only 1 percent of new hires in units without automatic enrollment.
Although still in its early stages, it is already clear that automatic enrollment is positively affecting participation rates in South Dakota’s Supplemental Retirement Plan (SRP)—this is a 457 plan that is supplemental to the state’s pension, and actually shares many similarities with 403(b) plans which themselves often supplement a defined benefit plan.
Under the South Dakota supplemental plan, state employees are allowed to make optional pre-tax contributions to augment their retirement income. The plan is considered to be an integral component of retirement benefits and employees are encouraged to participate in the SRP when planning for retirement. After considering national trends in savings rates and the need for additional retirement income for participants in the South Dakota retirement plan, the South Dakota Retirement System decided to increase the prominence of the SRP. Changes included revising the SDRS mission statement to emphasize the need for additional personal saving through the supplemental plan and adopting automatic enrollment in the SRP for newly hired employees.
Prior to 2009, newly hired workers had to make a positive election to participate in the SRP: if the employee did nothing, he or she was not enrolled in the plan. In 2008, the state legislature passed a bill authorizing automatic enrollment so that newly hired employees were enrolled in the SRP. As noted above, the results were startling.
Prior to the policy change, about 20 percent of all eligible employees participated in the SRP. Eight months after the passage of automatic enrollment legislation, 91 percent of new, eligible employees whose units chose to implement automatic enrollment participated in the plan and remained in it. The opt out rate was only 8.7 percent, which is in line with opt out rates in 401(k) plans that institute automatic enrollment.
This study is an example of how other states can dramatically aid their employees in preparing for retirement simply by instituting features such as automatic enrollment—and other features such as automatic escalation. It is yet another piece of evidence that automatic enrollment is one of the simplest things employers are doing to ensure the majority of their employees are planning adequately for retirement costs. And, the small opt-out rates suggest employees recognize they needed the extra push to start saving. More on that in a minute.
It is worth commenting on two aspects of the South Dakota case study—and the lessons that can be drawn from it.
First, the policy was rolled out with no opposition and no push back from employees. This shows the recognition by all parties—legislators, plan administrators and employees—that saving more was a good and necessary course of action.
Numerous studies and first-hand accounts indicate that actual opposition to automatic features tends to be weak or nonexistent, and the majority of employees—new and existing—welcome automatic features in their retirement plans. In addition, a study by Harris Interactive on behalf of Retirement Made Simpler found that 85 percent of workers said that automatic enrollment helped them to start saving earlier than they would have otherwise done. Given this, a best practice is to roll out an automatic enrollment process for all employees, not just those who are newly hired. There is more detail on this below.
Second, the default contribution to the plan is relatively small—only $25. The plan administrators hope that once workers participate in the plan, they will continue to contribute to the SRP throughout their career, and increase their monthly contributions. In fact, many who did participate quickly upped their contributions. Also, the rollout of automatic enrollment was buttressed with a communications program educating employees about the need to save amounts in addition to their defined contribution benefit, which in many cases was not enough to retire in comfort. In our discussions with plan administrators we got the sense that if they were to do this again, they would roll the plan with a more substantial default rate.
The lesson is to “think bigger” when it comes to setting default rates in a plan. A recent analysis from the Principal Financial Group adds credence to this. Client data from Principal shows that plans with an automatic enrollment feature defaulting at 3 percent produce an average deferral of 6.3 percent. This is lower than the average deferral of 6.8 percent for plans without an automatic enrollment feature. In contrast, if a plan rolls out an automatic default rate of 6 percent, the average deferral is 7.1 percent. Furthermore, nearly twice as many participants (61 percent) reach an overall savings rate greater than 11 percent—which many financial experts consider to be a threshold for a secure retirement—when their employers’ plan defaulted them at 6 percent. In contrast, when the default rate was 3 percent, only 32 percent of employees reached that 11 percent savings rate.
Workers Like Automatic Features and Start Saving Earlier
The South Dakota study reinforces the fact that employees want and value the guidance found in automatic features. An October 2009 survey by Prudential Insurance found that 74 percent of American workers would rather be automatically enrolled into a 401(k) plan than use the traditional method; 65 percent support automatic contribution escalation.
Even more telling, a 2007 Retirement Made Simpler study shows that workers who have been automatically enrolled strongly support the mechanism and start saving before they would have otherwise. According to the study, 97 percent of workers who had been automatically enrolled and remained in the plan were satisfied with the procedure, while 90 percent of those who had been automatically enrolled and then opted out due to individual circumstances felt the same way. Eighty-five percent of those participating in the 401(k) plan said that they had started to save for retirement earlier than they had otherwise planned, while 95 percent felt that automatic enrollment made saving for retirement easy. Finally, 98 percent of those who were automatically enrolled and remained in the plan were glad their company offered automatic enrollment, as were 79 percent of those who had opted out. This last number is key, as it shows that even if workers decided that their specific circumstances did not allow them to remain in the company’s 401(k) plan, they still valued automatic enrollment.
403(b) Case Study: North Shore-LIJ Health Systems
The value of automatic features is further shown in a another example, this one a 403(b) case study. The company is North Shore-LIJ Health System, which includes 15 hospitals across Long Island and in New York City, and additional long-term care facilities and home health agencies. North Shore employs nearly 20,000 physicians, nurses and other healthcare professionals.
Despite a matching contribution by the health care provider, in 2007 more than 30 percent of eligible workers did not contribute to its 403(b) savings plan—this despite an employer match of 2 percent on an employee contribution of 6 percent. It was a frustrating situation for the company. In the words of the company’s director of benefits, “Some people believed they couldn't afford to contribute. Others had an expectation that their employer would take care of everything.” In some cases immediate gratification won out over long-term financial security—a new car won out over the need to save for retirement.
The 30 percent nonparticipation rate was of deep concern to the company, which felt it had an obligation to boost participation but was dubious that additional education would improve employees' participation rate. So North Shore turned to automatic enrollment. The benefits staff made its case to the health care system’s senior leadership and the organization worked with its plan administrator to roll out the program.
Interestingly enough, while cost was a consideration, it was not the ultimate determining factor. In fact, automatic enrollment was estimated to increase the system's costs by nearly $3 million, but the company reasoned that since it was providing a competitive retirement program that included participation in a defined contribution plan with an employer match—a plan they hoped everyone would participate in—it made sense to ensure that all employees maximized the availability of the benefit.
This attitude is present in many employer deliberations about automatic plan design. A sense of corporate responsibility is the overriding factor in the decision to automate. It is a refrain I have heard over and over again in discussions of 401(k) plans, and it appears to be no different with respect to the 403(b) decision-making process.
In 2008, automatic enrollment was adopted. Employees—new and existing—who did not voluntarily enroll were automatically enrolled with a contribution amount equal to 3 percent of their salary, with that amount increasing in annual increments of 1 percent of salary for the next three years. Those contributions would be invested in target-date investment funds offered the plan administrator.
The net result? Participation rate in North Shore’s 403(b) plan soared—jumping to, and remaining at, 90 percent.
The company was concerned about negative employee reaction, in spite of a robust educational campaign that went well beyond DOL’s mandatory communications requirements to employees. But North Shore received virtually no calls from employees who didn't understand why they were automatically enrolled in the plan. In fact—and this is another refrain that we hear time and again—employees who called were delighted and said “they never would have done this on their own.” In fact, the director of corporate benefits said employees actually thanked him and the organization for its decision to automatically enroll them.
This case study is more confirmation that rolling out automatic features in a 403(b) environment is quite comparable to doing so in a 401(k) environment:
- The decision to automate is often born out of frustration with current participation rates and the failure of traditional communications programs to significantly change this fact.
- Because organizations are implementing automated features for the first time, they rely heavily on the expertise of their plan administrator.
- Organizations are desperate for information from federal sources such as DOL and the IRS. There appears to be a need for more information by 401(k) and 403(b) providers related to fiduciary rules as they pertain to automatic plans, in spite helpful governmental resources, including DOL’s valuable “Meeting Your Fiduciary Responsibilities” content and The ERISA Fiduciary Advisor, an interactive guide that asks plan sponsors specific questions about their own circumstances and provides fiduciary answers.
- Because of the relative newness of automatic features by 403(b) plans, organizations also seem to be keenly interested in case studies of how other organizations have handled automatic 403(b) implementation.
- There are concerns about cost—but those concerns are often secondary to a sense of corporate responsibility to help employees adequately save for retirement.
- Concerns about negative employee reaction are real—but ultimately unfounded. Unfortunately, they can cause organizations to move too tentatively with automatic plan design best practices. Rolling out automatic enrollment to all employees and escalating up to 10 percent or more remain underutilized strategies.
Given these realities, RMS has several suggestions for ways to improve the use of automatic features by 403(b) plans:
- There seems to be a steep learning curve among 403(b) providers, particularly small and medium-sized organizations about the basics—and especially the best practices—of automatic enrollment. Retirement Made Simpler has produced an Automatic 403(b) Toolkit that seeks to bridge the information gap, including a number of sample employee communications resources. We are not alone in offering helpful resources—and this council might consider issuing its own information and resources specifically focused on automated plan design. For instance, consider a sister publication to the current and valuable Automatic Enrollment 401(k)Plans for Small Businesses (IRS Publication 4674) that focuses on automatic 403(b) plans.
- Just as the growth of automatic features in 401(k) plans experienced the greatest growth after the 2006 Pension Protection Act resolved concerns that implementing these features could conflict with certain state and local laws, there is a need for a comprehensive study of any potential conflicts that could affect the use of them with 403(b) plans. This could serve as the basis for future legislation on the subject.
- Undertake efforts to facilitate greater education and information directed at non-ERISA 403(b) plans. Our discussions with experts who work daily in the 403(b) space note that not all 403(b) plans are created equal. For example, governmental plans, some church plans and 403(b) plans that are not subject to ERISA but are instead governed by state and local laws and rules. Decision-makers for these plans are likely to have their own questions and concerns related to plan automation. As one expert noted, “the devil is in the details” when sifting through the various regulations and requirements—and this is especially true for automatic plan design. These concerns play a direct role in the decisions of several local think tanks not to implement automatic features in their 403(b) plans despite publishing papers supporting the use of them for other types of retirement plans. Again, it is worth considering whether specific steps – legislative or regulatory – could help to encourage non-ERISA 403(b) plans to implement automatic enrollment and automatic escalation.
- Finally, because organizations like to hear from their peers, consider convening a round table of companies that have automated their 403(b). By sharing real-life stories of 403(b) automation—the obstacles and the success stories—and disseminating the proceedings to the plan sponsor, plan administrator and plan advisor communities you provide a valuable service to organizations actively considering automating their 403(b) retirement plan.
Thank you for allowing me to share my thoughts with you today.
Orszag, Peter and Rodriguez, Eric. “Retirement Security for Latinos: Bolstering Coverage, Savings and Adequacy.” RSP Policy Brief No. 2005-7 (July). Retirement Security Project and National Council of La Raza, Washington DC. See also “The Ariel-Schwab Black Paper,” published by Ariel Mutual Funds and Charles Schwab. October 2007
Gerry Geisel, For-profit strategy boosts 403(b) plan enrollment to 90%
, Business Insurance, June 27, 2010
Connie Toth, “And the survey says…”Check your paradigm!
BusinessBenefits.com, published by Robert J. Toth, Jr.