People working in insurance know that insurance contracts have a fundamental flaw: their customers don’t read them.
This is a problem because a contract depends on all parties agreeing to its terms. If a customer doesn’t read their policy, they can’t really agree to it. This can cause disputes, which often lead to costly court battles.
Legal scholars call this the ‘no-reading’ problem.
But the no-reading problem has a lesser-known counterpart: the ‘no-understanding’ problem. The no-understanding problem suggests that consumers who do read their contracts can’t reliably understand them.
Now, new research from the University of Michigan sheds further light on just how severe the no-understanding problem is. The authors also offer a few practical solutions.
The study tells us a few important things about how customers read insurance policies.
- Poorly written policies can be worse for readers’ understanding than no policy at all.
- Partially reading or partially understanding is its own challenge, and the standard policy structure makes it worse.
- Readers with a more sophisticated understanding of insurance may struggle just as much as ‘unsophisticated’ readers.
Not reading versus not understanding
They may be unimaginatively named, but the no-reading and no-understanding problems are important to distinguish. That’s because they have different implications.
For example, consumers can shoulder some of the blame for not reading their contract — they are part of the no-reading problem. But, if they try to read a contract and find it too hard to understand, more blame lies with the insurer.
That means the problems have different solutions, too. For example, requiring consumers to scroll to the bottom of an agreement before they click ‘I agree’ may make them more likely to read it (let’s be honest — it probably won’t). But that requirement won’t help consumers understand the agreement if it’s poorly written.
The study: can people tell if a scenario is covered?
To find out just how big the no-understanding problem is, the researchers asked 2,500 American insurance customers to work out whether seven hypothetical scenarios would be covered by a standard home insurance policy.
Respondents were split into two groups.
- The control group had to guess whether an event would be covered without referring to an insurance policy.
- The treatment group had to use an excerpt from a real-world insurance policy to work out whether the event would be covered.
Surprisingly, the treatment group did not fare much better than the control group. In fact, for two out of the seven scenarios, the group without the wording did better.
In other words, having the insurance policy in front of them actually made it harder for respondents to work out whether an event would be covered in some situations. The difficult wording was actively detrimental to their understanding of coverage.
‘While it has long been evident that most consumers do not attempt to read standard consumer forms,’ the authors write, ‘our research indicates that even when they do, they often fail to fully grasp the terms.’
What the study tells us
The researchers found that part of the problem is what they call ‘a partial-reading or partial-understanding problem’.
This problem arises when a consumer reads or understands one part of their policy, but not another part that modifies the first. For example, they might read that their house is covered against flooding, but not read or understand a relevant exclusion for retaining walls.
This incomplete understanding often leads, as you can imagine, to very unhappy customers.
The partial-reading or partial-understanding problem points to a challenge inherent in the standard structure of insurance policies. Benefits are one section; exclusions are another. Customers who give up reading on page five may not realise their home becomes ineligible for cover on page six.
But some readers are more sophisticated, right?
The researchers also broke their findings down by race or ethnicity, income, and ‘sophistication’. Sophisticated respondents were those with a higher education who claimed to have experience reading and understanding insurance policies.
This deeper analysis yielded more surprising results. These ‘sub-populations’ within both the control and treatment group all performed similarly. In other words, even people who thought they had a good grasp of insurance found it hard to tell if a scenario would be covered.
These findings may be a bit of a reality check for some insurers. We often hear that insurance companies know they have a sophisticated or specialist audience. But, according to this study, we can’t assume that more educated or wealthy customers will be more likely to understand a complex policy wording.
What we can do with this information
The authors also suggest a range of possible solutions for lawmakers and regulators. They could impose readability standards on insurers. They could also look to AI. ‘Smart readers’ using generative AI tools like Chat GPT could create simple summaries of cover.
Finally, insurers could make extra effort in the policy wordings themselves to solve understanding problems, like adding contextual ‘warnings’ throughout the text. These would be eye-catching messages designed to:
- address common misunderstandings
- remind readers that they may have an incomplete understanding if they don’t read a whole section.
It may go without saying, but we see the WriteMark® as part of the solution, too. Readability requirements across 25 criteria — from big picture to language and presentation elements — all make insurance policies easier to understand.
For insurers, the overall lesson of this research is simple: there’s no point getting customers to read your policy if they won’t understand it.