Insurance Series, Part 1: Theoretical Optimum and Actual Behavior
The idea behind insurance – as the name itself implies – is to have an opportunity to hedge against risks coming from uncertainty in areas ranging from material damage to life insurance. We can find insurance for our phones, our cars, and even dancers can insure their legs. The general idea would be that if we counter a situation where our essential assets get damaged, we would not get into a helpless state – our insurers got us covered.
Although this kind of general insurance is a recent innovation, variations of it were around for a long time. A perfect example would be the Burial Societies of ancient Rome. One could join a society, where a monthly “membership fee” had to be paid. Upon the death of one of their members, the Society would bear the costs of the burial process – which was not cheap at the time. It made sense to join one of these clubs, as you hedged against a potentially great expense – not to mention the importance of the ceremony of burials back in the day.
In this insurance series, we will examine the nature of consumer behavior in the realm of the insurance industry by explaining the actions of individuals through the lens of behavioural economics. Our goal is twofold: demonstrating the misjudgments of insurance buyers so that they can improve their decisions while also delineating policy providers the underlying consumer motivations and biases so as to allow them to better construct their products.
Insurance Industry: The synonym for suboptimal behavior
There is scarcely any industry that is characterized by more economically irrational behavior than the insurance industry. From a rational perspective, individuals should be interested in avoiding large risks, that is, events such as floods that have low-probability of occurrence, but impose high costs when it occurs. The reasoning arises from the fact that these are the events that are the hardest to cover from personal savings, and therefore will affect personal situations the most. In a similar vein, individuals should not opt to insure small-risk events, such as insurance for phone theft, or flexible airline tickets, that are characterized by high probability and meagre consequences. Yet, the reality appears to be the opposite of what the economic theory predicts. Individuals rarely purchase insurance for disastrous events that impose the highest costs. This is evinced by the unwillingness of farmers in flood-prone areas of the US to purchase flood insurance, even though it is highly subsidized by the government. What is more interesting, however, is that the insurance industry itself displayed a similar behavior. After the 2008 financial crisis, many insurance companies have become insolvent, as they did not hedge themselves against the low-probability, high-consequence event of a global mortgage default.
Under-Insurance: Overconfidence and Narrow Framing
It is well-documented that people tend to be optimistic, when a decision is made under uncertainty. For instance, 81% of entrepreneurs think that their venture will be successful, whereas, in reality, around 80% of startups fail after its inception. Accordingly, people are overconfident and have a tendency to believe that even if an unanticipated disastrous event were to happen, they would not be the victims. This emphasizes that overconfidence creates an effect, in which the objective probability of an event is converted into a subjective and significantly lower probability. Subsequently, since individuals underestimate the risks associated with themselves, a decrease in the need for insurance will ensue.
Alternatively, even if an insurance for low-probability, high-consequence events was bought, the effect of narrow framing, i.e. the behavioral bias resulting in individuals to overlook the overall aspects of an event, can lead to under-insurance phenomenon. Within the context of insurance, narrow framing is the primary factor that leads consumers to cancel their insurance policy after several years without having a claim. Consecutive years of non-damage creates the illusion for consumers that the probability of the occurrence has diminished, which further translates into the perception of insurance coverage as less important and the decision to not to be insured.
Over-Insurance: Endowment Effect and Loss Aversion
There is a tendency of consumers to pay extra premiums to buy insurance for goods, such as insurance for old cars and minor home contents ranging from bicycle theft to furniture, that, in theory, they should not. Endowment effect is the primary behavioral factor causing the over-insurance phenomenon as it creates an illusion for consumers that the charged insurance premiums are fairly priced. In short, the endowment effect makes individuals value more the things that they actually own and have some sort of connection with. A well-known study on this topic comes from Knetsch (1989), who found that 89% of individuals who owned a coffee mug would not have traded it for a candy bar, and similarly, 90% of those who were endowed with a candy bar would prefer to keep it when offered to trade it for a mug. Note that both of the previous items are pretty cheap, meaning that endowment effect can be distinct from one’s financial status. As individuals have a tendency to assign higher values to the objects in their possession, the endowment effect creates the psychological need to insure.
In addition, loss aversion – the individual's preference to avoid losing compared to gaining an equivalent amount – complements and reinforces the endowment effect in the sense that it creates a second-order effect, in which the losses associated with the damage to the good inflicts additional cognitive costs. Accordingly, on the one hand, endowment effect translates into the overestimation of the value of one’s own possession, and, on the other hand, loss aversion leads to overweighting of these losses, creating a substantial motive to insure the goods that are owned. Insurance companies, being aware of this tendency, charge higher prices for these goods/occurrences than the actuarially fair value, that is, the fair value of insurance premium given its probability of occurrence and the magnitude of the resulting loss. Ergo, consumers need to be aware of and take into consideration the effects of these two behavioural biases on their insurance decisions in order to avoid paying extra insurance premiums over the actuarially fair value.
What should we insure against?
The question still stands regarding the theoretical explanation on why individuals should insure the low-occurrence, high-consequence events, compared to the high-occurrence, low-consequence ones. The key to the answer lies in comprehending the difference in the coverage which individuals get for each dollar of insurance premium that they pay. Accordingly, as insurance companies exploit endowment effect and loss aversion, they charge higher premiums than the actuarially fair value for high-occurrence, low consequence events. Subsequently, the coverage received per each dollar spent for these types of insurances is minimal. In contrast, to subsidize the insurance uptake for low-occurrence, high-consequence events, governments effectively pay a significant amount of the insurance premiums, emphasizing that these disaster insurances offer substantially higher coverage per dollar spent on insurance premium. In addition, in terms of affordability, since these events bear the highest cost on the individual, they should be the ones that are insured most. Yet, given the theoretical goal of maximizing insurance received per dollar spent, the behaviour of insurance customers is exactly the opposite. Instead of purchasing insurance against the highest cost events, insurance customers pay an extra premium for the occurrences that they could have self-insured themselves through bearing the small costs that high-occurrence, low-consequence events entail. Self-insuring against such events could create further savings in the sense that the extra premium paid over the actuarially fair value would not be paid, and the arising savings could be used to pay for high-cost insurances that individuals should be most insured against.
The insurance industry, in general, faces a similar problem to what savings and pension payments do, namely: financial myopia. If you were to ask a finance guy to determine the value of a 10-year-long payment method, he would calculate all the payments’ present value using a certain discount rate. With this technique, he would basically allocate a price to time. If we want to simplify things, the general public pretty much does the same thing – except they overvalue their consumption in the short run, and undervalue their well being in the long run. This is the core of a simple savings-problem: I do not wish to save today, because there are so many things I want to buy today!
Insurance industry is built upon the relationship between behavioural tendencies that drive decisions under uncertainty. Accordingly, our goal is to probe into the underlying mechanisms of both the supply and demand side of the industry from a behavioural perspective. This week, we have provided a bird’s eye view of the most important concepts. Next week, however, we will address one of the most important questions that every practitioner faces: How to make people switch their insurance plans?
References and Further Readings
Gottlieb, D. and Mitchell, S. O. (2019). Narrow Framing and Long-Term Care Insurance. The Journal of Risk and Insurance, 86(2), pp. 1-33.
Knetsch, Jack L. (1989), “The Endowment Effect and Evidence of Nonreversible Indifference
Curves,” American Economic Review, 79(5), 1277-1284.
Salamouris, S. I. (2013). How overconfidence influences entrepreneurship? Journal of Innovation and Entrepreneurship, 2(8), pp. 1-6.
Shapira, Z. and Venezia, I. (2008). On the preference for full-coverage policies: Why do people buy too much insurance? Journal of Economic Psychology, 29(5), pp. 747-761.
Sydnor, J. (2010). (Over)insuring modest risks. American Economic Journal: Applied Economics, 2(4), pp. 179-199.
Thaler, R.H., Tversky, A. K. and Schwartz, A. (1997). The effect of myopia and loss aversion on risk taking: an experimental test. Quarterly Journal of Economics, 112(2), pp. 647-661.