"Damage, Living Environment, and Reconstruction Under the Great East Japan Earthquake"-The 5th Survey of Nuclear Disaster Evacuees from Futaba, Fukushima, Summary of Results 2019

1――Basic Information

Surveys on the damage, living environment and reconstruction under the Great East Japan Earthquake have been conducted via a research project called the "International comparison of reconstruction of living infrastructure from disasters" (Yasuyuki Sawada, Professor, Graduate School of Economics, The University of Tokyo; Keiko Iwasaki, Researcher, NLI Research Institute) of the University of Tokyo since 2013. The surveys target all household heads of Futaba in Fukushima prefecture, where all residents were forced to evacuate due to the nuclear power plant accident caused by the Great East Japan Earthquake that occurred in 2011. Surveys were conducted in July 2013, December 2014, July 2016, December 2017 and July 2019. This paper reports the summary of results of the fifth survey conducted in July 2019.¹

The survey includes questions about basic attributes such as age and gender, as well as questions related to connections with others (social capital) and health conditions (see the appendix at the end of this report for all the questions included in the survey). The questionnaires were distributed to all households of Futaba (about 2,950 households) that receive the regular town mail from Futaba. We also distributed a simplified version of the questionnaires excluding already available information gathered in previous rounds of the surveys to 534 residents who had filled out their names and addresses in previous surveys. We received responses from 707 Futaba residents who had evacuated nationwide (365 responses from questionnaires included in the town mail, 342 responses from the simplified version of the questionnaires; the response rate was about 24%).

The survey targeted heads of households and Figures 1 and 2 show the distribution of age and gender of the respondents. As we can see from these figures, compared to the age and gender distribution reported in the national census, the age distribution of the respondents is left-skewed, with the majority of respondents in their late 60s. The gender distribution shows that the majority of respondents are male. In addition, since the survey was conducted after the tremendous disaster, it is possible that the distributions of the respondents' characteristics are significantly different from those of general questionnaire surveys. Therefore, it should be noted that the results of this survey do not necessarily indicate the general trend of Futaba residents.

Figure 1: Age Distribution of Respondents/Figure 2: Gender Distribution of Respondents

¹ This research was supported by the following research grants.
　Grant-in-Aid for Scientific Research (15 J09313, 26220502, LZ003), Research Grants of the Japan Center for Economic Research.
This study is approved by the Ethics Committee of the University of Tokyo (19 -73).

2――Health Condition

Regarding the health condition of the respondents, as shown in Figure 3, many rated their own health condition before the earthquake as "good" or "very good," while many rated their health status after the disaster as "bad," "terrible" or "can't say either way." Compared to the 2016 survey results, the portion of respondents who rated their current health status as "good" increased in the 2019 survey, while the portion of respondents who rated their current health status as "bad" decreased slightly, indicating that the overall health status of Futaba residents has been gradually recovering. On the other hand, as shown in Figure 4, when asked about changes in health condition compared to pre-disaster status, many respondents rated their own health condition as worse than that of their pre-disaster status and the distribution has hardly changed since 2013.

Figure 3: Current Health Status/Figure 4: Change in Subjective Health Status

As for mental health, the distribution of K6 score, a clinically validated index for diagnosing the overall stress state, of Futaba residents gradually improved from 2013 to 2019 (see Figure 5). However, the scores are still higher in Futaba compared to those for Japan or in the surveys conducted immediately after the disaster in the other disaster affected areas, indicating that the mental health recovery of Futaba may take a very long time. (K6 is an internationally used measurement for general mental health status that consists of six questions. The higher the total score, the more likely the respondent is stressed.)

Figure 5: Distribution of K6 Scores for Japan, Futaba,
and Other Disaster Affected Areas

In addition, the previous rounds of surveys showed that the K6 score among Futaba residents living in temporary shelters had been increasing year by year, but only a few (6 people) lived in temporary shelters among respondents of the 2019 survey since many who used to live in temporary shelters had moved to public reconstruction housing. Therefore, instead of checking the distribution of the K6 score among temporary shelter residents, we compared the distribution of the K6 score by dwelling type this time. As shown in Figure 6, the distribution of the K6 score among residents of public reconstruction housing tends to be higher than that of residents living in other kinds of dwellings such as their own house or rental housing. This suggests the importance of continuous mental health support even after moving to public reconstruction housing from temporary shelters.

Figure 6: Distribution of K6 by Dwelling Type

However, the results of this survey do not necessarily apply to all residents of Futaba, and a high K6 score does not necessarily mean that one has a mental disorder. Please note that the purpose of our survey is to provide policy implications to the Government or other administrative agencies.

3――Change in Social Capital

Social capital refers to trusting relationships and networks, and is sometimes referred to as "kizuna" in Japanese. Social capital is getting attention as a key notion to achieve successful disaster recovery and has been one of the major focuses of our study. In previous studies, we showed social capital would play an important role in maintaining mental health after the disaster, while it might have been weakened by the disaster among Futaba residents.

There are several indicators that are commonly used to measure social capital, but we focused on three items. First, we checked the level of "generalized trust" using the GSS trust question, which is one of the most commonly used social capital measures. Generalized trust among Futaba residents had been decreasing from 2013 to 2016 (portion of those that think people can be trusted had been decreasing). However, since 2017, the portion of those that think people can be trusted has been gradually increasing (see Figure 7). The overall distribution has almost recovered to the same level as that of Japan in 2019, eight years after the disaster. On the other hand, since Futaba had a very high generalized trust level before the disaster, recovery to its original high level will take much longer.

The second indicator of social capital, "frequency of mutual assistance with neighbors," also shows a gradual recovery trend. In addition, the third indicator of social capital, "trust of neighbors" also shows a gradual recovery trend since 2016, but both trends have been very moderate. It will take a very long time for social capital to recover, and we believe it is important to continue to monitor these changes over the long term.

Figure 8: Frequency of Mutual Assistance with Neighbors/
Figure 9: Trust of Neighbors

4――Building Relationships with Residents at Evacuation Destinations

Some Futaba residents had told us that building a new social relationship with residents in evacuation destinations is a challenge that needs to be addressed in prolonged evacuation periods, in which the policies and understandings of residents of evacuation destinations vary widely. To capture this challenge, since the 2016 survey, the questions on relationships with residents of the evacuation destinations have been included. As shown in Figure 10, the percentage of residents with opportunities to interact with the residents of the evacuation destinations and the percentage of residents that participate in events held in the evacuation destinations are gradually increasing. In addition, the percentage of those that feel it is better to hide that they are from Futaba, those that feel uneasy about taking out the garbage, and those that are called bad names or been misbehaved to because they are from Futaba have been decreasing. However, we can see that about 40% of the residents still feel it is better to conceal that they are from Futaba, and that they do not have any opportunities to interact with residents at the evacuation destinations, indicating that building new relationships with residents at the evacuation destinations is still an important challenge for Futaba residents.

Figure 10: Relationships with Residents in Evacuation Destinations

5――Relationship Between Damage and Present Bias

Figure 11:Disaster,Present Bias and Mental Health

We found exposure to a disaster aggravates an individual’s present bias and this causal relationship can be a key mechanism behind the disaster and mental health deterioration nexus (see Figure 1). Present bias is an individual's procrastination tendency known to be intimately linked to harmful behaviors against health, such as overeating, smoking, and excessive alcohol drinking. In disaster affected areas, various activities and policies to improve health and to encourage communication have been introduced, such as regular lunch programs, radio gymnastics and other events, the installation of free running machines in community centers, and the use of architectural design in public reconstruction housing which encourages interactions among residents. These activities and policies are considered to prevent the deterioration of mental health caused by procrastination tendencies, and our study results support the effectiveness of such activities and policies. However, it should be noted that not everyone faces aggravation of present bias or mental health deterioration after a disaster and the purpose of our study is to provide policy implications.

6―― Summary of Findings from the Five Rounds of Surveys

(1) Futaba residents could be under more serious psychological distress than those in other disaster affected areas. More than eight years have passed since the Great East Japan Earthquake, and their mental health status has been gradually improving, but recovery may take much longer.

(2) In particular, there was a possibility that the mental health of those who had lived in temporary shelters for a long time was in a serious condition. However, currently, even after many of the residents have already moved from temporary shelters to public reconstruction housing, the mental health of residents in public reconstruction housing tends to be in a serious condition, and continuous support is important.

(3) As to changes in income and health condition caused by the disaster, we find that the greater the extent of decrease or deterioration is, the greater the degree of decline in individual well-being tends to be. This implies the necessity of sufficient compensation so that the original state of the survivors’ well-being can be recovered.

(4) The disaster has weakened the social capital of Futaba residents, and recovery may take a very long time.

(5) Keeping in touch with friends from pre-disaster times as well as participating in hobbies and volunteer activities after the disaster may help people maintain good mental health.

(6) Though relationship building with the residents of the evacuation destinations shows some gradual progress, the progress is subtle and it is still thought to be an important challenge that evacuees are facing.

(7) Aggravated present bias (procrastination tendencies) due to disasters can lead to a decline in mental health condition, but policies that promote interaction among residents and encourage good health behaviors can prevent such a decline.

These results have been presented at international and domestic academic conferences. In addition, these results have been published in international academic journals. We intend to continue our analysis and contribute to the improvement of disaster preparation/rehabilitation policies.

Our survey results are based on aggregates and analyses of responses from approximately 24% of the households of Futaba and do not represent all Futaba residents. Since the survey was conducted after a major disaster, the characteristics of respondents may be very different from general surveys and there is a possibility of an overestimation in our results due to the deterioration of physical and mental health conditions. Therefore, special caution is required in interpreting the results, and any definitive judgments based solely on these findings should be avoided.