2.2. Japanese Life/Health Insurance System
We focus on life and health insurance rather than disaster insurance because those who contract with a private insurance company are likely to be risk averse due to the characteristics of Japanese national and private insurance systems. First, disaster insurance is common, and it is natural for Japanese (and non-Japanese) citizens to have disaster insurance plans. Due to the high number of earthquakes, most Japanese housing, car, and private property insurance systems—and even bank loans—automatically incorporate disaster insurance, which covers damages from disasters such as earthquakes, fires, floods, and tsunamis. Each type of disaster is covered based on the damages and damaged property.
Furthermore, starting in January 2007, existing tax deductions for general insurance premiums were amended, and a deduction for earthquake insurance premiums was created to help people's efforts to compensate for damage caused by earthquake disasters. For example, a maximum of 50,000 JPY for the income tax (national tax) and of 25,000 JPY for the inhabitants tax (local tax) can be deducted from gross income (Ministry of Finance, Japan). Such benefits and coverage along with the frequent occurrence of disasters allow people to easily enter disaster insurance plans. Therefore, we simply cannot argue that an insurance enrollee is risk averse because he or she has a disaster insurance plan.
Second, focusing on private life and health insurance allows us to examine risk-averse enrollees and those who engage in precautionary spending. While the Japanese government provides a mandatory national insurance system to both Japanese and non-Japanese citizens that covers hospital care, outpatient care, mental health care, prescription drugs, home health care, and dental care (Arai and Ikegami 1998), private insurance is growing, as it additionally covers chronic conditions and hospitalizations, offering cash payments to the insured in times of cancer or long-term hospitalization. Private health insurance plans are classified into three types: medical insurance obtained separately from life insurance, medical riders attached to a new or current life insurance policy, and supplementary medical insurance that covers copays for services rendered by the public health insurance system (The Life insurance association, Japan). To be included in one of Japan's private insurance plans, one needs to choose an insurance company and plan and pay insurance premiums in addition to those for national insurance. Such efforts take time and require individuals to assume extra costs to prepare for unknown future events. Thus, in this study, we treat those who pay such additional premiums as engaging in precautionary spending.
2.3. The Great East Japan Earthquake
On March 11, 2011, at 14:46:23 local time, a 9.0 magnitude earthquake occurred, causing widespread damage to Japan's eastern coastal area. The earthquake lasted approximately six minutes and occurred at a relatively shallow depth of 24.4 kilometers (km), or 15.2 miles, with its epicenter approximately 130 kilometers (80 miles) east of the city of Sendai in the Tohoku region. The earthquake magnitude was so massive that it displaced Honshu, Japan's largest island, 2.4 meters east and shifted the Earth's axis by 10 to 25 centimeters. After the earthquake, 647 aftershocks occurred (up to August 4, 2011), and most of them were associated with tsunami alerts; following the earthquake, tsunamis destroyed Tohoku and the coast of southern Hokkaido, killing 15,848 people (the officially recorded death toll as of February 10, 2012). The extensive damage caused to eastern Japan led this natural disaster to come to be regarded as the worst in the world's history. In Tohoku, whole cities were swept away by the tsunami, and some towns were reduced to less than half of their pre-tsunami populations.
Figure 1 depicts the number of injured, missing, and dead earthquake victims. The number of missing people continued to grow two weeks after the incident, peaking at 17,541 on March 25, 2011 (Fig. 1). Drownings due to the tsunami caused more than 90% of the deaths, with the elderly bearing the brunt of the toll. In terms of long-term damages, the earthquake completely displaced communities; basic infrastructures for human well-being, such as schools and hospitals, were destroyed, and medical services were not able to operate due to the lack of capacity.
Alongside these direct damages, shortly after hitting Tohoku, the earthquake and a tsunami 14 meters (46 feet) high arrived and swept over the Fukushima nuclear powerplant's seawall, flooding the lower sections of reactors. This resulted in the failure of emergency generators and the loss of control of circulating pumps. Around March 12 and 15, three nuclear meltdowns, three hydrogen explosions, and the release of radioactive radiation occurred due to the failure of reactor core cooling in the so-called Fukushima nuclear energy plant accident. The Fukushima nuclear energy plant accident was identified as a possible global emergency of public safety following the massive earthquake and tsunami, with the International Nuclear Event Scale triggered to its highest level (level 7).
While damages from mass disasters can reduce people’s SWB, current evaluations of disaster loss are unlikely to consider such falls in SWB. Moreover, those who experience mass disasters are more likely than those who have not to start seeking preventive measures to adapt to or mitigate another mass disaster should one occur again. Thus, taking advantage of our insurance spending and SWB dataset, we first examine whether insurance spending increases the SWB of those who experienced the earthquake and then estimate its monetary value to account for the disaster loss.
2.4. Literature Review
As mentioned in Sect. 2.2, experiencing massive natural disasters can impact people’s well-being through various types of damage. These damages can be categorized into tangible and intangible disaster losses (Hudson, Pham, and Bubeck 2019; Hudson et al. 2019). Tangible losses include monetary losses, housing and infrastructure losses, injuries, deaths, and other physical damages. Intangible losses, in turn, include psychological losses that manifest in fear, anxiety, and other mental health problems.
Intangible losses are often ignored despite their long-lasting characteristics, which may cause an underestimation of overall damages from a natural hazard. Oishi et al. (2015) clarify that housing damages caused by the great Hanshin-Awaji earthquake of 1995 had effects on sufferers’ SWB that persisted even 16 years after the earthquake. Lamond, Joseph, and Proverbs (2015) find long-term psychological losses, such as distress and mental damage, from flooding. These effects are especially severe for low-income people and for households that experience relocation.
Recent studies have focused on the intangible losses caused by the great East Japan earthquake by investigating its effects on well-being (“Well-Being Effects of a Major Natural Disaster: The Case of Fukushima” 2015; Ohtake, Yamada, and Yamane 2016). These studies have shown that people living in the affected areas experienced significant SWB losses due to the disaster itself and disaster-related news broadcasts, unlike people not living in the affected areas.
Moreover, previous studies have explored the characteristics of the impacts of disasters on well-being by focusing on several types of natural disasters such as earthquakes (Sibley and Bulbulia 2012; Valenti et al. 2013), flooding (Valentiet al. 2013; Luechinger and Raschky 2009) and hurricane risk (Berlemann 2016). The effects of disasters on SWB differ by socioeconomic characteristics such as gender (Hudson, Pham, and Bubeck 2019; Valenti et al. 2013) and income (Berlemann 2016, Markhvida et al. 2020). In general, low-income people are found to suffer more serious SWB losses from disaster experiences. Carter et al. (2007) discuss the reason for such long-term impacts, particularly on low-income groups. They note that while theoretically, poorer households can cover their losses by utilizing loans and insurance, in reality, the poorest people tend to lack sufficient resources to recover and cannot easily escape the “poverty trap”, making them suffer longer negative impacts and well-being losses.
Despite the serious and long-term impacts of natural disasters, earlier studies have clarified that people who incur damages from disasters can mitigate or recover from well-being losses in certain ways. One is an adaptation to disaster damages. It has been found that some coping strategies, such as purchasing flood protection, can reduce mental health effects (Lamond, Joseph, and Proverbs 2015). In addition to such behavioral adaptation, studies focusing on SWB have clarified the tendency of individuals to adapt to negative events over time, showing that SWB returns to levels close to the reference point approximately 1 year after the event (Frijters, Johnston, and Shields 2011). On the other hand, some researchers are skeptical about complete adaptation to severe damage from negative events (Oswald and Powdthavee, n.d., 2008). Religiosity is also related to the mitigation of the SWB impacts of disasters. Sibley and Bulbulia (2012)describe that espousing religious beliefs after a natural disaster does not improve subjective health. Nevertheless, a loss of faith after a natural disaster leads to an additional loss of subjective health.
An insurance contract is another effective way to mitigate the damage caused by a disaster. Luechinger andRaschky (2009) show that mandatory insurance has mitigating effects that can almost fully compensate for losses from a flooding experience.
Moreover, the disaster risk itself has a negative influence on well-being even among people who did not experience the disaster (Hudson, Pham, and Bubeck 2019). This finding indicates that risk-avoiding behavior should lead to an increase in SWB for an individual who dislikes being exposed to disaster risk.
SWB measures can also be used to monetize the benefits obtained from intangibles. This method is called the life satisfaction approach and has been used in recent research to measure values that cannot be measured in the market (Frey, Luechinger, and Stutzer 2010). The general procedure of this method is as follows: (1) surveying life satisfaction, a targeted variable, income, and other sociodemographic characteristics among a target sample; (2) regressing the life satisfaction measure on the targeted variable, income, and other control variables; and (3) calculating the change in income that could compensate for the life satisfaction gains or losses caused by a change in the targeted variable.
Utilizing this method, previous studies have investigated the monetary value of intangible goods such as air quality (Welsch 2006; Luechinger 2009), airport noise (Welsch 2006), extreme weather events (Möllendorff, vonMöllendorff, and Hirschfeld 2016), green spaces (Tsurumi and Managi 2015), healthy behaviors (Shi et al. 2019) and the cost of terrorism (Frey, Luechinger, and Stutzer 2009). The life satisfaction approach is also useful for valuing the gains and losses from experiencing a disaster and purchasing insurance if such gains and losses include intangible aspects that cannot be revealed by market prices.