How our friends’ spending habits mislead us to undersave for retirement

It’s easier to see what our friends, family, and neighbors buy than what they save. Research co-authored by Berkeley Haas finance professor Johan Walden shows how this “visibility bias” can lead us to undersave for retirement.

Three women with shopping bags taking a selfie in acity.
Image: AdobeStock

Americans are not saving enough for retirement, and new research co-authored by Berkeley Haas Professor Johan Walden shows one reason why: They’re misled by conspicuous consumption in their social networks.

“Our neighbors don’t put out a sign showing how much they’ve set aside for retirement, but it’s easy to see the new car sitting in their driveway,” Walden says. “The fact is, observations of someone consuming are more salient than observations of someone saving, and so we observe more of those signals, and we give them more weight in our own decision-making.”

In an article published in The Journal of Finance and coauthored with Bing Han from the University of Toronto and David Hirshleifer from the University of Southern California, Walden shows how the consumption we see among our friends and family pushes us to consume more—and fall short on retirement goals.

Modeling the effect of visibility bias

A recent estimate in the U.S. suggests that the gap between what consumers are saving and what they’ll need to retire is almost $3.7 trillion. For some, this is not a choice: They don’t have the income they need to set money aside. But that’s not the whole explanation.

Walden and his co-authors show that the psychological phenomenon they call “visibility bias” contributes to this chronic undersaving. People don’t tend to publicly share their savings habits, so consumers’ social networks provide inaccurate signals about peers’ consumption and savings. People tend to think their peers are spending more than they are. This, in essence, makes people believe that it’s safe to consume more.

Unlike the phenomenon of competitive consumption—often known as “keeping up with the Joneses”—visibility bias is driven less by comparing oneself to others by incomplete information about others’ spending habits.

In modeling this dynamic, the researchers find that one important implication is the way this effect feeds back on itself. As people see their neighbors and friends buy more, they may raise their rate of consumption; in turn, their neighbors see this increase and bump up their own levels of consumption, continuously squeezing savings out of the picture.

The researchers also show how denser and more extensive networks are likely to intensify visibility bias. Given this, residents of cities are more likely to undersave for retirement. Our growing connectedness through social media, as well as the window it gives into other people’s purchases, is likely to exacerbate the problem as well. (Walden and his coauthors don’t run any experiments of their own, but they propose several empirical approaches to probing the nature and magnitude of visibility bias.)

Finally, the demographic contours of a person’s social network could naturally reduce the strength of this bias. People who are more financially sophisticated tend to consume less, as do people who are older. If someone’s social network contains more of these types of people, “then we find a dampening effect, as you’re receiving signals of low consumption from these groups,” says Walden, the Mitsubishi Bank Chair in International Business and Finance. “But this doesn’t completely offset overconsumption among younger generations.”

More disclosure could increase savings

The effects of undersaving ripple across society, pushing people to work into older age than they otherwise would. People take on more debt, which, in turn, drives up interest rates. Fortunately, to the degree social networks and visibility bias influence the amount we consume, policy interventions may be relatively straightforward.

The researchers suggest two approaches focused on disclosure. First, policymakers could better publicize information on the risk of economic shocks—the frequency of layoffs, for instance, or the costs attached to illness and hospitalization. However, Walden notes that people generally have a difficult time interpreting and internalizing probability, so this may not be the most effective approach.

More productive could be efforts to reveal information about savings and consumption within social networks. On the one hand, policymakers or marketers could build campaigns designed to publicize how and how much people save, thereby making the unseen seen and reducing the influence of visibility bias. Campaigns could also clarify how much people actually consume. One study from 2020 shows that giving people who consume more than they can afford information about average consumption figures reduced their spending by 3%.

But given people’s propensity to believe that others consume more than they do, it’s important to provide appropriate information, Walden says. Simply publishing aggregate saving and consumption data may not be very helpful, since these are already affected by visibility bias. Policymakers should take social networks and demographics into account—for example, by emphasizing consumption habits of older adults. “The more specific, the better,” Walden says.