New Survey: Retirement Expectations Don’t Match Reality

Lack of awareness of how Social Security benefits work or concerns that the program will not be fully funded when younger generations retire may drive some of this effect. Without action from Congress, the primary Social Security trust fund is set to be depleted by 2033 but that does not mean the program will become defunct or bankrupt. Rather, without congressional action, all beneficiaries will face an immediate benefit cut of approximately 20%. Even in that worst-case scenario, today’s workers will still receive around 80% of their projected benefits when they retire—a major source of income for all but the wealthiest retirees.

In place of Social Security, non-retired respondents expect their own savings and continued labor income to play much greater roles in retirement than is likely realistic. For starters, as shown in Figure 1, there is a significant disparity between the expectations of today’s workers and the reality for current retirees. In addition, far more non-retired respondents expect retirement savings to be a major source of retirement income than are actively contributing to 401(k)s, IRAs, and the like. While nearly half (48%) of non-retirees anticipate these accounts will be a major income source, only 36% report actively contributing.

The biggest expectations mismatch, in percentage terms, is related to earnings from work in retirement. Nineteen percent of non-retired respondents expect earnings from part-time work to provide significant income in retirement, but only 7% of retired respondents report such earnings. It could be the case that older Americans encounter insurmountable obstacles to working, such as reduced physical abilities or labor market practices that are unfriendly to older workers. It could also be the case that many workers reach retirement without the need to rely on labor earnings. Our survey provides some support for a more optimistic interpretation—43% of retirees are surprised by how much they enjoy retirement.

The State of Retirement Readiness

Our survey suggests that more than half of Americans lack confidence in their ability to retire when they want to and sustain a comfortable life, as shown in Figure 2. Higher incomes correlate with a greater sense of preparedness, but older individuals don’t always feel more prepared than younger ones. Notably, those aged 45-54 feel the least prepared to retire. And although three-quarters (76%) of Americans believe that saving for retirement is important, just 39% of respondents have a plan in place that will allow them to retire when they want to.

Yet many Americans are taking active steps towards investing in their financial health, as shown in Figure 3. In particular, many more respondents making under $75,000 report following a set household budget, paying down debt, and living below their means than report having a financial plan for retirement.

Conclusion

Many Americans are stressed and uncertain when planning for their retirement—an understandable response to headlines suggesting that the average person needs $1.4 million to retire comfortably and the widespread misconception that Social Security will cease to exist if its trust fund runs out of money as projected in 2033.

This survey suggests that a key source of this rampant uncertainty is a poor understanding of what retirement actually looks like—when it will start, how much retirees need, and where that money comes from. This reflects not just incomplete information but also a more fundamental lack of financial literacy and longevity literacy. Policymakers and the financial industry should take note.

---

This survey was conducted online within the United States by Public First on behalf of the American Savings Education Council. The survey polled 2,024 U.S. adults from March 28 to April 7, 2024. All results are weighted using Iterative Proportional Fitting, or ‘Raking.’ The results are weighted by interlocking age & gender, state, ethnicity, and education level to Nationally Representative Proportions. The sample data is accurate to within +/- 2.7 percentage points using a 95% confidence level.