Learning from Rhode Island’s Pension Reform: What Did it Teach Us About Worker Retention?

Answering the question of how pensions influence employee recruitment and retention is important for fully understanding the implications of retirement policies. In a recent paper from the Center for Retirement Research (CRR) at Boston College, researchers examined how changes to retirement benefits affect the retention of public workers. The unique reform design has provided for an opportunity to evaluate the impact of changes on existing employees. In this commentary, I will review the findings of this research and provide recommendations for further investigation.

This line of research is particularly relevant at a time when public employee pension plans across the nation are facing major solvency challenges, prompting some to make changes to the current benefits structure. If reduced pension benefits prompted highly-skilled workers to separate sooner and choose other employment options, it would be an issue for policymakers to evaluate among the potential tradeoffs. It is important to consider, however, that it might not be true for all professions across the board, notably, teachers might have limited options for employment outside of their profession or the public sector.

The paper looks at the 2005 reform of the Employees’ Retirement System of Rhode Island (ERSRI), which greatly affected public employees who were not yet vested in the pension plan by June 30, 2005 (vesting required 10 years of government service) by raising the normal retirement age (NRA) from 60 to 65; decreasing the benefit multiplier; and shifting from a 3 percent compounded cost-of-living adjustment (COLA) to one capped at inflation, along with a three-day waiting period for the COLA to take effect.

The reform’s cumulative impact caused an estimated 48 percent reduction in pension wealth for a typical ERSRI employee, according to the paper. Simply put, Rhode Island public employees saw their benefits drastically reduced and their retirement age pushed further away.

The actuaries estimated a total savings of $243 million from the adopted changes. However, when evaluating the full fiscal impact of such a reform, it is important to also consider the cost of losing trained workers and needing to hire new employees. According to the authors of the CRR paper, the 2005 pension adjustment caused a one-time increase in direct turnover costs ranging between $1.8 and $8.1 million.

The authors find that the benefit change caused a 2.4 percentage point increase in the rate of separation for ERSRI. Since the lifetime pension wealth for the affected workers was reduced by 48 percent, the labor supply of public employees in Rhode Island can be described as inelastic in response to pension cuts (its elasticity is around 0.25). Interestingly, teachers were less responsive to benefit cuts than other public employees. Teachers affected by the change were just 1.7 percent more likely to leave during the studied timeframe.

It is important to understand that measures taken in Rhode Island were very specific to the particular pension plan and fiscal situation. In this particular case, benefit cuts affected pension plan members that were already promised a very specific set of benefits. It is plausible to assume that a very specific set of retirement wealth expectations affected their decision to join and remain in their particular position, and, therefore, it is no surprise that the number of separations increased relative to the no-reform scenario.

Pension reforms are often prospective in nature, thus influencing only the new employees. Changing benefits to existing employees is pretty unusual. Therefore, these findings have to be interpreted with caution, because they might not descriptive of any of the reforms that are changing benefits for new hires only.

Another important observation from this study is that the impact of benefit changes on retention is not the same across different occupations in the public sector. Specifically, the authors of this piece found that teachers were less responsive to pension cuts than other public employees. This is important, because it highlights that research findings on one particular type of public employee are not necessarily applicable across all occupations. Therefore, we should be cautious in extrapolating the results of these and future studies that involve certain occupations to the whole universe of public sector occupations.

Furthermore, the effects of this particular pension reform are not representative of other pension reforms. In particular, this tells us little about the effects of prospective benefit changes to new hires and the retention of these employees. Feelings of spite and increased uncertainty about future benefits are most definitely going to affect employees that have already been promised a certain benefit package. New hires that had a more modest benefits package in the first place wouldn’t be influenced by this same line of thought, and may not quit at the same rates. Therefore, the retention effects from benefit changes on new hires is another area of research that could further shed light on the impact of such reforms on public sector labor market.

Another interesting direction of research would be to look at the skills of new hires joining ERSRI in comparison to other Rhode Island public sector employees—such as Municipal Employees’ Retirement System (MERS)—before and after the change to benefits. This could show the impact that benefit cuts have on the quality of labor force entering public sector professions. Both retention and the quality of the labor force could be important measures to better understand the impact of retirement benefit changes.

Low levels of retention of public employees could be costly to the public for two main reasons. First, it would require additional resources for hiring new people. It is therefore important to consider the fiscal impact of reform not only in terms of reduced risk or savings in pension contributions but also in terms of expenses for new talent acquisition.

As discussed by the authors, in the case of Rhode Island’s pension reform, savings in pension contributions outweighed the expenses associated with hiring new employees.  A second factor that should be considered, that has not been discussed in this paper, is the quality of newly hired employees. This is an important topic that deserves further investigation because its impact can go beyond a state’s financials and affect all aspects of public services that those employees are tasked with providing.

The example of ERSRI is a cautionary tale for all plans that fail to adjust their underlying assumptions and funding policies according to noticeable changes to investment returns, demographic expectations and other influential variables. Drastic benefit cuts can affect labor supply, highlighting the importance of preventative measures, such as adjusting the assumed rate of return, keeping amortization schedules short and paying required contributions in full every year in order to keep public pension funds in good fiscal health.

If measures are taken in advance, then creating a new retirement plan for new hires—without altering the benefits that have already been promised to existing workers—could be an option. Conducting regular risk assessments and maintaining a set of realistic assumptions behind the benefit calculations are key to keeping the promises made to employees.

This research is a starting point in assessing the impacts of pension reform on the recruitment and retention of employees. Although the CRR study finds “pension cuts for current workers encourage mid-career civil servants and teachers to leave their government jobs,” it doesn’t tell us anything about the impact of reforms that have a prospective character and only apply to future hires. Therefore, more research is needed to understand the implications of pension reform on the public employee labor market.