Our days are driven by decisions we often don’t think twice about: Do you take the stairs or the elevator? Should you take this train or wait for the next? Which highway should you choose? Do you get a bagel on our way to work or eat a banana? Do you stop at the stoplight or proceed with caution? We're guided into many of these decisions. For example, an Out-of-Order sign at the elevator or a prompt from a Fitbit might encourage you to look for a stairwell. If a crowded train pulls in, an announcement that there is another behind it may cause you to wait. If you're tracking your caloric intake, you will likely choose the banana. This is not a discussion about the existence of free will. It’s a brief look at the ways in which we make decisions today and how those processes are guided by tools and processes created by other humans that we may not be aware of—and the implications of when that occurs within public policy.
In one way or another, the decisions above and many others are the result of the architecture of our lives. No, not metaphysical architecture, but choice architecture—the careful design of the environments in which people make choices. It’s all around from the signs on doors that indicate whether we should pull or push (or in the absence of signs where there is a handle or not—think about it) to the ways foods are grouped together at a lunch buffet (they don’t often mix the hot entrees with the desserts, do they?). Choice architecture was outlined by behavioral economist Richard Thaler and legal scholar Cass Sunstein. It highlights the power of the an instinctual response that is based on intuition. Known as the Automatic System, this is a gut reaction to the world around us—it’s almost instantaneous, and is not to be classified as traditional thinking. This is contrast to our Reflective System which is deliberate and employs logic and reason to make decisions.
Thaler gives a great example of this with a door in his classroom. This door included two very large vertical handles on the side facing his classroom. This meant his students had to make a decision as they approached it to push or pull the door as a means of exiting the room. This particular door opened outward, meaning it needed to be pushed, however the presence of the handles triggered the students’ automatic system, and they pulled the door to exit. Of course the door would not open and for those trying to make a quiet exit, they attracted the attention of their peers and of Thaler. Eventually, Thaler pointed this out to his students, but even after the pattern of pulling the door when they should push had been discussed, students continued to pull on the handles as they tried to leave. Their instinctual response to the handles, and the believed behavior that should result from interacting with those handles, trumped the information they had been given about how the door actually worked. Thaler maintained this was an example of poor choice architecture.
We don’t rely completely on our reflective system even though it might enhance our decision making skills—we just don’t have time for that! Or we haven’t been primed to ignore our instincts in a world that has categorically been designed to enhance our automatic assumption of the “right” action. Of course, this “right” action can be coaxed in one particular direction. The coaxers are known as Choice Architects—the people who create our environments and shape our experiences, which includes everyone from the person designing the subway platform to the person designing the menu at the restaurant you’re going to this weekend for dinner. Almost anyone can be a choice architect but not everyone should be choice architect.
Universal healthcare is definitely a divisive issue in the United States. And that divisiveness has very likely contributed to the opacity that so firmly grips the Affordable Care Act (ACA). Government programs have a reputation for being mired in red tape and process, but the public dialogue around this service has done little to disentangle the available information. Instead, it has likely prompted responses from program administrators that result in creating additional confusion even as they believe they may be adding clarity with user-friendly tools. This lack of clarity is compounded by the fact that health insurance is something that many people simply don’t understand. While there are definitely complications in the ways the plans are described, part of this confusion is tied to the fact that insurance forces us to consider our mortality. We may get sick. We may be injured. And we may need care in order to survive. This is something we’re not inclined to face. Insurance, as many insurance brokers will tell you, is meant to guard against this need. While it doesn’t prevent mortality from catching up to us, it becomes a reminder of what may happen. So many people are inclined to not want to understand it.
Recent stats indicate that 11.3 million Americans have signed up for health coverage under the ACA. They’re either part of a growing demand for coverage as penalties make it mandatory and the previously uninsured realize a means for obtaining treatment or they were automatically re-enrolled. The issue with these enrollments is that many people have done so blindly. Research suggests that fewer than 40% of individuals who are likely to enroll in a marketplace health plan are confident they understand insurance terminology. These individuals are most likely to pick what they believe is the simplest option on paper which could mean higher premiums and lower coverage. Similarly, for individuals who auto-enrolled, they may have continued on particular plan due to the convenience of auto-enrollment, but that inaction on their part may have left them vulnerable to higher premiums as plan rates increased.
Researchers tested choice architecture in helping to guide better decision making around health insurance. They devised a scenario where participants were asked to choose an insurance plan in year one and then were assigned a status of healthy or sick for the year, which impacted their usage of the plan they had chosen and subsequently impacted their insurance costs for the year. That status carried over to year two, when participants needed to again choose health coverage. The difference for year two was that some individuals were given a recommendation for coverage that maximized expected savings (which was not made explicit) and some were not. They found that recommendations, whether from case studies that illustrate similar experiences, from a trusted physician, or government agency, increased the likelihood that people would choose a plan that would maximize their savings. Basically, people need help in the face of complex, frightening choices—they get that help on basic things (e.g., a red octagonal sign indicates they should stop or stripes on the street show them where it is safe to cross), but on issues such as this where there is so much public discourse, there is little guidance.
While the researchers rightly point out that that the study does not address how preference for providers may impact choice, they’ve highlighted some of the major tenets of Thaler’s choice architecture—features that factor in many of the things we experience on a daily basis:
Defaults: People will opt for the easiest choice that presents itself. During the second annual enrollment period for the ACA, half of the enrollees actively made a plan selection in 2015. The other half opted to keep the plan they already had; they went with the default. While this option maintains enrollment, it doesn’t necessarily offer the enrollees the most cost-savings, which is a major concern for the population that is most likely in need and using this coverage.
Defaults are necessary. They answer the question of what happens if nothing happens, but as in the example above they don’t always serve the interests of the person relying on the option. Other notable example of where the default can be problematic includes magazine subscriptions and software installations. The former can result in a person having a credit card charged repeatedly for magazines they do not read, and the latter can result in additional (and sometimes malicious) software being installed with the desired program. According to Thaler one way around this is to force a choice by requiring the user to either opt-in or opt-out. This simplifies the options that users have to choose from but steers them away from inaction as a choice. This kind of forced choice works best on simpler items.
Errors and messaging: Early research on insurance choices indicated that low income, rural individuals had low insurance comprehension (understanding of what health insurance terminology meant) and a low numeracy (an understanding and ability to calculate probabilities). These two issues could be countered with a more guided experience through the healthcare marketplace.
People will make mistakes, and while the best designs will try to anticipate and correct for these in advance, offering users clues as to how to help correct issues enables their ability to choose. For example, many ATMs today will sound an alarm to indicate that you need to remove your card, and some may not dispense your cash until you have done so. Additionally, people often will not look for information until they need it. Signs that help people find their way to bathrooms or foodcourts or subways or hospitals, or strategically placed tooltips on a webpage, arm people to to be successful and increases their understanding overall.
Structure: The more choices we have, the more people are likely to look for ways to simplify their options. If they’re unable to do so they may give up or resort to the default option. Organization helps counter this. Choosing a paint color is made easier by classifications of warm or cool colors and the grouping of types so that the blues are with the blues and the greens are with the greens. Structure gets to the heart of the resolution that researchers have proposed for ACA comprehension: offering recommendations for plan options may help users navigate and evaluate plans which may offer savings. For example, labeling plan options as Bronze or Silver or Gold helps organize plan benefits and presents the user with cues about cost-benefit implications.
Choice-guiding principles are present in almost every aspect of our lives. The discussion of user issues around the ACA provides a tangible case where the impact of choice architecture may be great. But what’s more, the discussion of choice architecture overall will hopefully make us all aware of the ways in which we’re open to suggestion. One challenge of being guided in this way is that we’re all being guided in the same way—there is a best case scenario that someone has selected for the group. The best of choice architecture are those cases where it enhances or supports our automatic system. This makes it seamless with our lives. The worst is when the choices we are offered support an agenda that is fundamentally selfish in its aims. In this case, it may force a change in our automatic responses or cause general disruption within the overall social network. In general, outliers—those for who do not agree with the best case—may find themselves estranged from consensus and their networks. This may be necessary to the structure of society overall as it moves us toward a normative experience, but as a defining experience, it can be a dangerous tool in the wrong hands. This does not mean that choice architecture is something to be feared, but perhaps should prompt us all to be more aware of how we make the decisions we make.
Barnes AJ, Hanoch Y, Rice T (2016) Can Plan Recommendations Improve the Coverage Decisions of Vulnerable Populations in Health Insurance Marketplaces? PLoS ONE 11(3): e0151095. doi:10.1371/journal.pone.0151095
Thaler, Richard H. and Sunstein, Cass R. and Balz, John P., Choice Architecture (December 10, 2014). The Behavioral Foundations of Public Policy, Ch. 25, Eldar Shafir, ed. (2012). Available at SSRN:http://ssrn.com/abstract=2536504 or http://dx.doi.org/10.2139/ssrn.2536504
You might also like: