How State Pension Reforms Can Affect Employee Retention

After the Great Recession took a toll on state pension funds, a growing number of states considered and adopted reforms to their public sector retirement plans to put them on a more secure footing. Although changes to traditional pension plans can alter associated costs and the level of fiscal risk to state systems, little is known about how they affect individual employee decisions and a state’s ability to recruit and retain high quality workers.

Plan design reforms, such as shifting from traditional defined benefit (DB) pension plans to hybrid plans, eliminating or reducing cost of living adjustments (COLAs), or increasing employee contributions, may affect the way that state workers value their employment when compared with private sector jobs. Estimates of these impacts could help policymakers make more informed decisions when considering plan design reforms.

With support from The Pew Charitable Trusts, the Rand Corp. recently used its established dynamic retention model to simulate worker retention in three states that have made changes: Pennsylvania, South Carolina, and Tennessee. The Rand approach uses personnel data from each state to model how decisions by state workers and teachers on holding onto their jobs, or retention, over the life of their careers may be affected by plan design reforms.

Rand looked at the impact of several common changes to plan design, such as moving workers from a traditional pension plan to a hybrid plan that combines a defined benefit with a separate savings account such as a 401(k) (what is known as a defined contribution, or DC, plan). In addition, Rand modeled what happens when certain adjustments were made to an existing pension plan, such as increasing employee contributions or reducing COLAs in retirement. The model examines the career trajectory of new teachers who would spend their entire careers under the new policies, not the impact of the policy change on current teachers who are early-, mid-, or late-career.

The analysis looks at only three states, but it shows some consistent changes, including different effects at different career stages. Although there are clear differences across states, including differences in pay, employee preferences, policies, and access to private job markets, the results proved qualitatively similar in all three.

“Changes to retirement plan design have substantive effects on the retention of new hires over their entire career,” the authors wrote. “Blending defined benefit and contribution plans leads to limited changes in retention over the early career, larger declines over mid-career than under a system with only a defined benefit, but more working years for career teachers.”

Rand produced two research papers, the first looking at state workers and teachers in South Carolina and the second examining teachers in South Carolina, Pennsylvania, and Tennessee.

Overall, moving from a defined benefit to a hybrid plan had a small impact on retention: Changing to a hybrid plan reduced the average number of years of service by 3.3% in Pennsylvania, 0.7% in Tennessee, and had no net effect in South Carolina. However, when retention is examined throughout a worker’s career, Rand found larger effects at different stages. Although the changes reduced mid-career retention, they appear to have increased retention later in workers’ careers.

For example, Rand found that transitioning from Tennessee’s legacy defined benefit plan to the hybrid plan would increase average years of service by a negligible 0.2% for teachers in years 1 through 10 of service, decrease average years of service by 4.2% for in years 11 through 30, and increase average years of service by 20.9% for teachers in years 31 through 44.

“Reducing retirement benefits reduces the value of staying for early- and mid-career teachers, as this reduces the option value associated with staying and later leaving to collect retirement,” the authors wrote. They also noted that teachers are more likely to stay longer in late career under a less generous retirement policy because the value of their retirement benefit is less than it would’ve been under a more generous policy.

Table 1

Expected Teacher Retention When A Hybrid Retirement Plan Replaces a Pension

In 3 states studied, retention drops mid-career but increases in later years

Notes: Baseline plans are Pennsylvania’s Class T-E plan (a defined benefit plan for members hired between 2011 and 2019), South Carolina’s Class 3 plan (a DB plan for members hired on or after July 1, 2012), and Tennessee’s legacy DB plan (a DB plan for members hired before July 1, 2014). Hybrid plans are Pennsylvania’s Class T-G plan (hybrid plan for new members hired after July 1, 2019), a hypothetical South Carolina hybrid plan designed in part based on the state’s existing DB and optional DC plans, and Tennessee’s existing hybrid plan, for new members hired after July 1, 2014. Retention results shown for years 1-10, 11-30, and 31-44 are conditional on hire, 10 years of service, and 30 years of service, respectively. Retention is measured as the change in average years of service.

Source: The Rand Corp.

In addition, the impact of moving from a defined benefit to a hybrid plan has a smaller impact on worker retention than the impact of moving from a defined benefit to a defined contribution plan.

For teachers in South Carolina, shifting from a traditional pension to a hybrid plan has minimal overall predicted effects on retention—average years of service does not change. But changing to a defined contribution plan reduces retention by a more significant 2.4% decrease in average years of service. The impact is more pronounced mid-career, with a reduction of 3.3% for the hybrid plan versus 8.9% under the defined contribution plan.

For South Carolina state workers, the impact of shifting from a traditional pension to a hybrid or defined contribution plan is even more pronounced. Changing to a hybrid plan reduces average years of service by 3.0% and changing to a defined contribution plan reduces average years of service by nearly 10%. The reduction in retention is significant for both early- and mid-career state workers, which is different than teachers, who are more likely to leave mid-career.

Table 2

What Happens to Expected Worker Retention in South Carolina With Changes to Hybrid or Defined Contribution Plan

Larger reduction in mid-career when the switch is from defined benefit to defined contribution plan

Notes: Retention results shown for years 1-10, 11-30, and 31-41 are conditional on hire, 10 years of service, and 30 years of service, respectively. The results in the table reflect the expected change in average years of service for South Carolina teachers and state workers moving from a defined benefit plan to a hypothetical hybrid plan designed in part based on the state’s existing defined benefit and optional defined contribution plans) and the separate South Carolina Optional Retirement Plan offered by the state.

Source: The Rand Corp.

Many state retirement systems have cut pension benefits or increased employee contributions over the past decade. Rand’s analysis looked at the impact of these changes on retention as well.

In all three states, the modeling shows that increasing the employee contribution rate decreases teacher retention throughout their careers. Raising the baseline employee contribution rate by 3 percentage points in Pennsylvania, South Carolina, and Tennessee reduced average years of service by 8.3%, 6.1%, and 4.6%, respectively.

Table 3

How Would a 3% Increase in the Employee Contribution Rate Affect Retention?

Raising the required contribution rate has largest impact on mid-career retention

Notes: Baseline plans are Pennsylvania’s Class T-E plan, South Carolina’s Class 3 plan, and Tennessee’s legacy DB plan. Retention results shown for years 1-10, 11-30, and 31-44 are conditional on hire, 10 years of service, and 30 years of service, respectively. Retention is measured as the change in average years of service.

Source: The Rand Corp.

Rand also modeled teacher retention under a less generous pension plan. Reducing benefits, for example by eliminating COLAs, has minimal impact on retention for early-career teachers, decreases retention for mid-career teachers, but increases retention for late-career teachers—a similar impact to retention as adopting a hybrid plan. Eliminating COLAs reduces the value of retirement benefits, thus reducing the incentive for late-career teachers to leave. Rand found that late-career teachers with less valuable retirement benefits were more likely to continue working beyond retirement eligibility.

Table 4

What Happens to Teacher Retention if Cost of Living Adjustments Are Eliminated 

Slight reduction mid-career but those in later years more likely to stay

Notes: Baseline plans are Pennsylvania’s Class T-E plan, South Carolina’s Class 3 plan, and Tennessee’s legacy DB plan. Hybrid plans are Pennsylvania’s Class T-G plan, a hypothetical South Carolina hybrid plan based on Class 3 and the SORP DC plan, and Tennessee’s hybrid DB/DC plan. Retention results shown for years 1-10, 11-30, and 31-44 are conditional on hire, 10 years of service, and 30 years of service, respectively. Retention is measured as the change in average years of service.

Source: The Rand Corp.

Overall, Rand’s analysis helps shed light on the impact of retirement plan reforms on employee retention discussions and consequently state employers’ ability to manage their workforces.

Aleena Oberthur is a project director and Mollie Mills is a principal associate with The Pew Charitable Trusts’ state fiscal policy project.