198 M Tanninen, T-K Lehtonen, and M Ruckenstein
the General Data Protection Regulation enforced in the European Union.
In light of our study, the model is inadequate in ensuring trustworthy practices, as it fails to consider people’s everyday realities and hesitation when
engaging with the policies (see also Marelli et al. 2020).
The customers’ lack of knowledge is also related to questions of information asymmetry. The processes of datafication are built on informational
asymmetries, but in the insurance context the concept usually refers to
customers withholding information that is crucial for underwriting, thus
increasing risk for adverse selection, that is, the disproportioned selection
of high-risk individual in the pool (Baker 2003; McGleenan 1999). Social
scientists have, however, pointed out that the asymmetry works the other
way around, as well: insurers have much more information about a given
instrument and the associated population values and averages than the customer (Van Hoyweghen 2007). Behaviour-based life insurance policies are
no exception. We have demonstrated how customers struggle to make sense
of the wider context of the policies and how they lack certainty on precisely
what they are signing up for. Thus, the information asymmetry places policyholders in a vulnerable position, as it is very difficult for them to reliably
estimate the policies’ possible effects. At present, this unequal arrangement
might be partly related to the policies’ experimental nature; even the insurers themselves do not know what will become of the new operations and
thus cannot communicate it clearly to customers (Jeanningros 2020; Meyers
& Van Hoyweghen 2020; Tanninen et al. 2020).
Thus, what is at stake with uncertain data for both the insurance companies and in the data practices is how trust will be maintained or created
under these new conditions. The interviewees wanted to feel secure that
even if insurers (or the information technology and wellness companies that
mediate the insurance practice) controlled their data, they could obtain a
reasonable reward for that fact. Yet, such a transactional logic does not in
and of itself guarantee trustful relations. It was hard for people to evaluate
what the price of their behavioural data should be. Furthermore, customers
wanted to be sure that the data would not be used for inappropriate uses
such as online crime or questionable commercial practices and found it difficult to assess who to trust.
Our case speaks to the need for a careful building of trust as the insurance industry moves onto the terrain of the emerging data economy. The
data relationships that insurers promote need careful planning and following through to become genuinely trustworthy. Otherwise, the industry faces
the risk of raising a new kind of mistrust in people, evidence of which we
can already see in the empirical material presented here. We have demonstrated how people find it difficult – if not impossible – to assess how to trust
insurance, especially in the long run. If digital data is an uncertain, lively,
and messy element, the insurers need to make sure that they can handle that
uncertainty. Otherwise, the insurance industry as we have known it will
no longer be viewed as capable of responsibly managing sensitive personal
information.
the General Data Protection Regulation enforced in the European Union.
In light of our study, the model is inadequate in ensuring trustworthy practices, as it fails to consider people’s everyday realities and hesitation when
engaging with the policies (see also Marelli et al. 2020).
The customers’ lack of knowledge is also related to questions of information asymmetry. The processes of datafication are built on informational
asymmetries, but in the insurance context the concept usually refers to
customers withholding information that is crucial for underwriting, thus
increasing risk for adverse selection, that is, the disproportioned selection
of high-risk individual in the pool (Baker 2003; McGleenan 1999). Social
scientists have, however, pointed out that the asymmetry works the other
way around, as well: insurers have much more information about a given
instrument and the associated population values and averages than the customer (Van Hoyweghen 2007). Behaviour-based life insurance policies are
no exception. We have demonstrated how customers struggle to make sense
of the wider context of the policies and how they lack certainty on precisely
what they are signing up for. Thus, the information asymmetry places policyholders in a vulnerable position, as it is very difficult for them to reliably
estimate the policies’ possible effects. At present, this unequal arrangement
might be partly related to the policies’ experimental nature; even the insurers themselves do not know what will become of the new operations and
thus cannot communicate it clearly to customers (Jeanningros 2020; Meyers
& Van Hoyweghen 2020; Tanninen et al. 2020).
Thus, what is at stake with uncertain data for both the insurance companies and in the data practices is how trust will be maintained or created
under these new conditions. The interviewees wanted to feel secure that
even if insurers (or the information technology and wellness companies that
mediate the insurance practice) controlled their data, they could obtain a
reasonable reward for that fact. Yet, such a transactional logic does not in
and of itself guarantee trustful relations. It was hard for people to evaluate
what the price of their behavioural data should be. Furthermore, customers
wanted to be sure that the data would not be used for inappropriate uses
such as online crime or questionable commercial practices and found it difficult to assess who to trust.
Our case speaks to the need for a careful building of trust as the insurance industry moves onto the terrain of the emerging data economy. The
data relationships that insurers promote need careful planning and following through to become genuinely trustworthy. Otherwise, the industry faces
the risk of raising a new kind of mistrust in people, evidence of which we
can already see in the empirical material presented here. We have demonstrated how people find it difficult – if not impossible – to assess how to trust
insurance, especially in the long run. If digital data is an uncertain, lively,
and messy element, the insurers need to make sure that they can handle that
uncertainty. Otherwise, the insurance industry as we have known it will
no longer be viewed as capable of responsibly managing sensitive personal
information.