By Kim Ferrero, Ph.D. M.Sc.
Assistant Director of Research
At the Academy, we often take a turn while mapping out the many crossroads of insurance and public policy, finding ourselves at the intersection of economic theory and human behavior.
One of the most persistent enigmas the Academy’s Retirement Practice Council has examined in collaboration with the Research team is the “annuity puzzle”— the phenomenon where, despite the clear financial security provided by lifetime income in the form of annuity payments, a significant portion of retirees opt for a lump sum distribution at the start of their retirement instead.
To get to the heart of this issue, our senior research analyst, Julia Goodwin, Ph.D., has been working closely with the Pension Committee. Their recent work provides a fresh, data-driven look at how retirees are actually navigating these high-stakes decisions.
The Challenges of Choosing a Best-Fit Retirement Plan
Modern retirement security often hinges on the choice between an annuity, which provides a guaranteed stream of income for life, and a lump sum, which offers immediate access to the full value of a retirement account. This choice is largely framed by the type of plan an employee holds, usually framed as defined benefit (DB) or defined contribution (DC) plan.
The primary difference between these plans lies in who bears the investment risk and how the final benefit is determined. A DB plan, such as a traditional pension, promises a specific monthly payout for life, based on a formula involving your salary and years of service. Conversely, a DC plan, like a 401(k), provides an individual account where the final retirement balance depends on the total contributions made and how those investments perform in the market over time.
While we know upwards of 95% of participants in DC plans take lump sums, the data for DB plans has historically been much murkier. Past research that utilized survey data, as opposed to administrative data (e.g., regulatory filings, plan sponsor data), often did not distinguish between plans that offer a lump sum option and those that don’t. Survey respondents may be unaware of the type of plan they have and if the plan provisions allow for a lump sum or not.
Without this distinction between plans with and without a lump sum option, we risk underestimating the true rate of lump sum distributions. Including people in DB plans without a lump sum option in an estimation of the lump sum distribution rate drives the average down by including people who did not even have the option. The lump sum research project addresses this gap by utilizing administrative data from recent IRS Form 5500 to create a more precise snapshot of the current landscape in order to better understand DB decision-making.
The research team began by collecting all Form 5500s for private-sector DB plans and leveraged artificial intelligence models to extract lump sum payout rates from the actuarial assumptions section of these administrative forms. With this novel database of plan characteristics and lump sum assumption percentages, the team will be able to estimate average lump sum payout rates across benefit formula type, lump sum option financial limits, and whether the assumptions are based on an experience study.
Why This Matters for Actuaries
Forgoing an annuity—or guaranteed lifetime income—increases retirees’ risk of insufficiently funding their retirement and potentially outliving their assets. While economic theory often favors the security of an annuity, many retirees opt for a lump sum for a number of reasons, including to gain immediate liquidity, manage their own investment portfolios, or facilitate the rollover of assets into an IRA.
This decision-making process is not fully understood. Looking at it through an actuarial lens, it appears to be influenced by the prevailing economic environment. Specifically, interest rates are a critical factor in determining annual annuity payments, as higher rates typically allow for more substantial yearly payouts for the same amount of principal. Understanding these behavioral motivations and the impact of market fluctuations is vital for informing public policy and protecting retirement security.
The research team’s work with our volunteers highlights a critical need for further primary research in order to get a better sense of how frequently people opt for the less secure lump sum payout option. The next steps in the lump sum research project will be to explore why people take a lump sum option in lieu of guaranteed lifetime income by fielding a survey. We look forward to seeing how our volunteers integrate findings from both prongs of the project into our ongoing advocacy for sound actuarial principles.
If you are interested in learning more about the work of the Research Committee, please check out the Academy’s website. You’ll also be able to find more information on the work of the Retirement Practice Council and opportunities to participate as a volunteer for the Academy.