With Social Security, Waiting Can Be the Hardest Part

For retirees and soon-to-be retirees, few financial decisions are as important as when to begin claiming Social Security benefits. Yet despite the fact that waiting to claim these benefits can substantially increase monthly payments as well as lifetime payouts, few take full advantage of this provision. In fact, half of all Americans claim Social Security benefits at age 62 or within two months of stopping working while just 2% wait until 70, the age at which monthly benefits are maximized. Suzanne Shu, an assistant professor at UCLA's Anderson School of Management, has been studying the reasons people file early for Social Security. (Click here to read a paper she co-authored on the topic). Shu recently spoke with Morningstar about this very important topic. 

Can you first explain how each of the main factors you've identified--life expectancy, loss aversion, and a sense of fairness--plays a role in people's decisions as to when to start taking Social Security benefits?
The first factor, life expectations, is simply a question of how long someone thinks they will live. This is an important and purely rational input to the Social Security benefits decision. If you expect to live a long time, you should claim later to get a larger set of benefits, but if you expect to die early, you should claim early. In our work, however, we find two potential sources of bias around life expectation.

First, people's answers depend on how we ask the questions, either as how long you expect to "live to" or when you expect to "die by." We found that people who thought about how long they would "live to" gave, on average, a predicted life span of about 10 years longer than those who thought about when they would "die by." To be more specific, the median life expectancy was 75 years old in the "die by"frame, and 85 years old in the "live to" frame.

Second, we find that even when life-expectation judgments are longer, people do not adjust their claiming as much as they probably should. For example, a 10-year increase in life expectations only causes a delay of around six months in Social Security claiming.

The next two factors, loss aversion and fairness, speak to people's emotional reactions to the benefits-claiming decision. We find that many individuals feel strong ownership toward their benefits, probably because they've been contributing to them for so many years. The stronger the sense of personal ownership toward the benefits, the more unfair it seems to have to wait to claim those benefits. In addition to those feelings of unfairness, the chance that you could fail to get anything because you've delayed too long (that is, from early death) generates feelings of potential loss for that "owned" resource.  Thus, the more averse someone is to potential losses, the more likely [they have the urge] to claim early and avoid that possibility.

How can these factors be countered, so that more people appreciate the advantage of delaying Social Security in order to receive a bigger lifetime payout?
This is one of the directions we are taking the research, but we don't have good answers yet. One of the interventions we tried was to give people a more complete picture of what their cumulative benefits would be under different claiming ages. This is an approach we had previously found to work in the annuities environment, which is similar in principle to Social Security claiming. To our surprise, it had the opposite effect on claiming than it had on annuities--it made people want to claim even earlier! 

We suspect that one reason for the different reactions comes back to the concepts of loss aversion and ownership. For an annuity, the decision is a purchase decision: Do I want to spend money to acquire this stream of lifetime income? In contrast, for claiming, people seem to perceive the income stream as already owned, and not as a purchase. So then the decision feels more like a question of: When can I start taking advantage of this thing I already own?

It's been well-documented that people value an object they own more than an object they're considering buying, a difference called the endowment effect, coined by Nobel-prize winning behavioral scientist Daniel Kahneman and his coauthors Jack Knetsch and Richard Thaler. If individuals feel a strong sense of ownership toward their Social Security benefits, then this same endowment effect could be influencing their decisions of how soon to claim. Understanding the psychology of endowment and methods for addressing it could allow us to better assist individuals facing these difficult trade-offs.

Other possible interventions might try to decrease the feelings of personal ownership of the benefits, such as by pointing out that Social Security is a national shared resource rather than a personal savings account. Another option might be to try to counter the loss aversion by highlighting the potential regret that could come from claiming early and having significantly smaller benefits later in life. All of these approaches are being tested.

Your study mentions additional factors that contribute to people taking Social Security early, including a lack of patience and concerns about the program's solvency. Can you describe some of these other factors and ways they might be addressed?
We know that impatience is a big factor in lots of our daily decisions. For example, lack of patience can cause people to fail to save for the future, or to overindulge in food and drink without thinking about future health consequences. Impatience matters here, too. As you might expect, being impatient causes people to not be willing to delay claiming their benefits; they want the benefits as soon as possible, even if it means they won't get as much money. Self-control tools for helping people counteract impatience, such as precommitments or self-imposed limits (like a diet), might be useful ways to address those factors here.

Program solvency is a factor that has been suggested by many economists as a reason for people to claim early. Basically, if you think the Social Security program is going to disappear in a few years, you want to start claiming now. We do find that people's beliefs about program solvency are a significant input to the claiming decision, but the impact of this factor is much smaller than many of the aforementioned other factors.

Did your research reveal any notable demographic differences? For example, were some groups more likely than others to cite life expectancy or loss aversion as a factor in taking Social Security early?
We found some overall demographic differences. People who reported being in better health were more willing to delay, and older people were less willing to delay. For example, people in their 40s or early 50s are more likely to say they'll wait until full retirement age or beyond, while people in their late 50s are more likely to say they'll claim at 62. We found small gender effects (women were sometimes more willing to delay) and little effect of personal savings or financial literacy. We haven't yet found interactions between these demographic measures and the psychological factors like loss aversion.

Aren't there some instances when taking Social Security early makes sense, such as if the individual is in dire need of income, or if he has reason to assume a shorter-than-average life expectancy? 
Absolutely. Claiming early in both of these situations is certainly reasonable. Our work attempts to understand whether people are making those trade-offs optimally for their situations--for example, are they perhaps overreacting to a slightly shorter life expectancy and claiming much earlier than would be financially best?