Supporting Statement - Final Version

SUPPORTING STATEMENT

AGENCY: Pension Benefit Guaranty Corporation

TITLE: Survey of Newly-Insured Pension Plans

STATUS: Generic ICR under Customer Service Focus Groups and Surveys

1212-0053 (expires 12/31/2008)

CONTACT: Catherine B. Klion (202) 326-4223, ext. 3041 or Thomas H. Gabriel (202) 326-4223, ext. 3898

1. Need for the collection. The purpose of this survey is to improve PBGC’s customer service. This survey will provide information on why sponsors establish new plans and how PBGC can better serve sponsors of newly-insured plans.

This survey will also help PBGC to be more proactive in fulfilling its mandate under ERISA section 4002(a)(1) to “encourage the continuation and maintenance of private pension plans for the benefit of their participants.” Over the past 20 years the number of plans insured by PBGC has declined by almost 75 percent from more than 112,000 plans to under 29,000. However, during this period of time, some employers have continued to adopt new defined benefit plans covered by PBGC’s insurance program. In 2006, 1,950 plans paid premiums to PBGC for the first time.

This survey is designed to provide information on what factors prompt employers to adopt new plans, what employers perceive as potential areas of concern in adopting new plans, actual areas of concerns after new plans are adopted, adequacy of regulatory compliance guidance, ways to ease the burden of administering the plan, and whether employers are satisfied with their decisions to create new plans and if so, why.

Review of premium and Form 5500 reports suggests that a number of recent new premium payers are not covered by PBGC (e.g., plans of small professional service companies). PBGC wants information on the scope of this problem so that it can take action to address the problem if necessary.

Not all of the plans associated with the new premium filers are new to the defined benefit population or to the population of defined benefit plans insured by PBGC. PBGC wants to learn whether the issues faced by new plans vary by type of plan – new defined benefit plan covering previously uncovered workers; new defined benefit plan covering workers previously covered under another defined benefit plan; or defined benefit plans that had been in existence for some time before they became covered by PBGC’s pension insurance (such as an existing professional service plan that recently began covering 26 or more active participants).

2. Intended use. PBGC plans to use the survey responses to improve customer service, including through education and outreach activities. PBGC will protect the confidentiality of the survey responses to the extent provided by law.

3. Description of respondents. The 400 respondents are a sample of the sponsors of the 1,950 plans that began paying premiums to PBGC in 2006.

4. Information collection procedures. The plan sponsor or plan administrator’s contact will first be notified by letter of the survey to encourage the plan to participate in the survey. A PBGC contactor will then conduct the survey by telephone. The contractor is Stuart Shapiro, associated with Wittenberg Weiner Consulting, LLC.

If the respondent indicates he or she did not receive the letter, the interviewer will summarize its content as follows:

The letter requested your participation in a voluntary survey of plans that first paid premiums to PBGC in 2006. The results of the survey will be used to improve PBGC's customer service to plans like yours. PBGC’s efforts to encourage employers to maintain plans rely greatly on feedback from plan sponsors.

The letter also stated that the Office of Management and Budget has approved this survey under OMB Control No. 1212-0053 (expires 12/31/2009) and that an agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB control number.

The interviewer will be free to refer respondents to PBGC’s website, but any questions will be forwarded to PBGC’s Policy, Research, and Analysis Department (PRAD) to handle. In addition, PBGC’s customer service call center will be advised that it might receive questions and will be instructed to forward them to PRAD staff.

5. Expected response rate. Based on the contractor’s experience with response rates for a survey of this kind, PBGC expects 60% of the respondents to complete the survey.

6. Justification for incentive. No incentive will be paid for participating in this voluntary survey.

7. Estimated burden. The telephone survey is expected to take 15 minutes. If all 400 potential respondents participate in the survey, the total burden will be 100 hours.

8. Methodology. Starting with the universe of 1,950 newly-insured plans that first paid premiums to PBGC in 2006, PBGC used an online sample size calculator to determine the optimal sample size with a 5 percent margin of error and a 95 percent confidence level. The key estimate of the survey is what percentage of the sponsors of the newly-insured plans would establish a defined benefit plan if they had it to do over again (question 15). Not knowing what the responses would be for any survey question, including question 15, PBGC assumed that 50 percent would respond "Yes" and 50 percent would respond "No.” This was a conservative means of determining the appropriate sample size to provide the precision and confidence level desired. The calculator indicated 321 plans should be sampled. PBGC increased this number to 400 plans to allow for non-responses. 

Because PBGC is most interested in sponsors of large plans (100 or more participants) and of plans for which the sponsor had made a decision to terminate, PBGC over-sampled these plans (there were 88 of the former and 25 of the latter, for a total of 113 plans), and separately sampled the remaining 1,837 plans. The same online sample size calculator was used to determine the optimal sample size of the two groups using the same criteria as was used above. The calculator indicated that 87 plans should be sampled from the group of greatest interest to PBGC (87 of the 113), and 318 from the remaining 1,837 plans.

Because the target sample size was 400 plans, PBGC selected 85 plans from the smaller group and 315 from the larger. After assigning each plan a unique number (based on the plan’s employer identification number and plan number), PBGC used an online random number generator to select the samples.

PBGC found a few sponsors had more than one plan in the sample. To reduce respondent burden, PBGC used the same random number generator to select which sponsor plan to keep and which ones to delete, and then selected plans from the unselected original groups to replace those plans that were dropped from the sample.

PBGC will combine the survey results with data from the plans' premium and Form 5500 filings. This will allow PBGC to analyze results by more refined plan size levels, industry group, locality, and other plan characteristics.

5