How to Increase Flood Insurance Take-Up Using Behavioral Economics Insights

Proper application of behavioral economics can improve take-up of flood insurance and narrow the protection gap. Sounds farfetched? Let’s dig deeper.

Behavioral economics is the study of psychology, economics and the decision-making process of humans. It helps us understand why people consistently deviate from rational action when making decisions. In other words, why would someone who lives in a flood-prone region choose not to buy flood insurance?

Behavioral economics can enhance customer centricity and yield quantitative and qualitative insights, all of which are indispensable to insurance. Historically, it has been used to make small changes to alter how someone makes a decision — like getting more people to pay their taxes on time by telling them about the tax paying habits of their neighbors, or increasing organ donation rates by changing the default option.

These small, incremental changes that produce disproportionate results are called “nudges,” and they’re being employed in the insurance space with increasing frequency and success. Most of the time the environment is already established — like an online quoting engine — and behavioral scientists study what drives the consumer in that environment and make small adjustments, like editing some wording or changing the order in which the questions are presented. These tweaks are then tested to see what impact they had, if any.

A few examples of effective nudges include:

• Increasing sales from a direct mail campaign by focusing on who the message was coming from and adding a clear call to action.
• Increasing completion of a claims form by anchoring the policyholder to tangible actions.
• Increasing honesty and disclosure of health conditions on an underwriting form by playing to a consumer’s perspective of themselves. Increasing email open rates by personalizing the subject line.
• Increasing the number of doctors who return forms promptly by adding a personalized, handwritten post-it outlining the benefits for them and their patient.

Beyond Nudges

Nudges are good, but behavioral economics can take us so much further in areas like product development. What if we could develop more relevant products by better understanding our customers and how they perceive and understand risk?

Natural catastrophes are an intriguing area to start. Hurricanes, tornadoes and earthquakes are difficult for humans to comprehend because they are events with a low probability and high consequence. So, people tend to think and behave in irrational ways, which is evidenced by the growing protection gap. A rational person would buy just as much insurance as they need, but considerations like price, education, brain shortcuts and decision-making obscure rational thinking.

What’s Old Is New

In order to understand the mind and behavior of the flood insurance customer and develop new strategies for increasing take-up, there’s a wealth of historical experience to draw on. Behavioral economists examine existing research, run new analyses on public and private data, and perform a variety of tests. Here are a few notable examples:

1. Leveraging existing research.

Howard Kunreuther, a professor at the Wharton School of the University of Pennsylvania, has conducted extensive research on consumers and flood insurance, specifically at this intersection of low-probability events and decision-making. One such study involving New York City homeowners after Hurricane Sandy yielded two recommendations for changing perceptions and encouraging purchase of insurance. The first is to express the timeframe of a risk differently: for example, saying “a greater than 1-in-5 chance of flooding in the next 25 years” instead of “a 1-in-100-year flood.” The former makes it real; the latter makes the consumer say, “It won’t happen in my lifetime.”

The second is to highlight the financial consequences a homeowner faces if a flood occurs and they don’t have the right insurance.

2. Using public and private data.

Analyzing FEMA data from 1980-2007, Justin Gallagher discovered an 8% increase in flood insurance take-up after a significant event. However, that level is sustained for only nine years, and then it reverts to the level seen before the flooding event occurred.

The federal government has released a new dataset, and insurers are using it to better understand flood risk. At Swiss Re we’ve hosted hackathons to brainstorm ideas, and while some of these ideas focus solely on using the NFIP data by itself, there are also ways to incorporate the data into existing models and datasets.

3. A variety of testing opportunities.

As more insurance companies launch flood products, more opportunities are created for testing and disruption.

Swiss Re’s Behavioral Research Unit focuses on running live tests to see how people act and make decisions in the real world.

For example, we’re currently working with an insurer that’s expanding its flood insurance marketing, testing different methods for communicating about this peril to existing customers. We’re reviewing multiple channels and methods including agent scripts, email campaigns and marketing materials.

Another test looks at agents who are now responsible for selling this relatively new private insurance product. Humans generally tend to maintain the status quo and stick to what they know, so asking agents to sell a new product they might not be familiar with can be difficult. In this study we examined what agents truly value and what motivates them.

But What If?

Sometimes it’s impossible to conduct a test in the real world, so we simulate the insurance buying process. One such online study was aimed at gaining a better understanding of how people prefer to purchase flood insurance: either bundled with their homeowners policy or as standalone coverage.

When examined through the lens of behavioral economics, bundling is seen as the better option. It’s easier and helps set the expectation for what the customer needs in order to be fully insured. However, behavioral economics has taught us that often perception is everything. If flood insurance is automatically bundled with homeowners insurance, the price will immediately increase, sometimes dramatically, creating an urge for the homeowner to start shopping around. Because of these opposing views, it’s something insurers must continue to study.

A Swiss Re survey of 1,500 homeowners in the United States revealed some misconceptions of how comprehensive a homeowners policy is, namely when it comes to flood damage.

• More people perceive their flood risk as high than report living in a 100-year floodplain.
• Of the participants who believe they live in an area that is at high risk for floods, only 26% claim to have flood insurance.

This is where customer centricity is key. Insights gleaned from tests, research and data help inform communication, and it becomes more important to customize and tailor each communication based on customer demographics, their existing insurance coverages, their location and their personal experiences.

Relying solely on perceptions or the word of agents and brokers isn’t enough. Similarly, quantitative data doesn’t reveal why people do what they do. To gain the full picture of consumer behavior one must blend these insights with qualitative data. This is especially true when it comes to natural catastrophes, where humans have consistently acted irrationally.

This article was first published in Carrier Management.

Wiese is a senior solutions innovation designer and vice president at Swiss Re.