Vulnerability or Flexibility: An Examination of Socioeconomic Status, Employment Changes, and Individual Income in Japan’s 3.11 Earthquake

Vulnerability or Flexibility: An Examination of Socioeconomic Status, Employment Changes, and Individual Income in Japan’s 3.11 Earthquake | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Vulnerability or Flexibility: An Examination of Socioeconomic Status, Employment Changes, and Individual Income in Japan’s 3.11 Earthquake Maoxin Ye, Daniel Aldrich This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7161036/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 5 You are reading this latest preprint version Abstract This study investigates the impact of the 2011 Tohoku earthquake on income inequality in Japan. Using a quantitative analysis of survey responses from nearly 4,000 respondents, we find that individuals with lower socioeconomic status were disproportionately affected by the disaster. During the immediate aftermath, those with lower education and precarious employment experienced significantly higher rates of job loss. In the recovery phase, education emerged as a key factor influencing re-employment, while the impact of prior employment status was less clear. Furthermore, the disaster led to a significant decline in income for regularly employed individuals compared to those with irregular employment. This suggests that the disaster not only affected employment opportunities but also had a significant impact on income distribution. Our findings underscore the complex mechanisms through which natural disasters can exacerbate income inequality. Beyond social vulnerability, the interplay of factors such as education, employment status, and income sources shapes individuals' experiences and outcomes in the aftermath of such events. Earth and environmental sciences/Environmental social sciences Scientific community and society/Geography Social science/Geography Earth and environmental sciences/Natural hazards Employment change employment flexibility income inequality social vulnerability socioeconomic status earthquake natural hazard Figures Figure 1 Figure 2 1 Introduction How disasters impact employment and income remains a critical question in disaster research (Reid, 2013 ). Individuals with lower socioeconomic status (SES) are particularly vulnerable to the shocks caused by disasters (Fothergill & Peek, 2004 ; Wisner et al., 2004). As a result, their employment becomes more unstable during and after disasters, hindering their recovery efforts. This, in turn, contributes to increased income inequality (Elliott & Howell, 2016; Howell & Elliott, 2019 ). Previous research suggests that individuals with lower socioeconomic status (SES) are more likely to experience job loss during disasters (Hallegatte et al., 2020) and that disasters can exacerbate income inequality (Milijkovic & Milijkovic, 2014; Yamamura, 2015 ; Fang et al., 2017 ; Keerthiratne & Tol, 2018 ). However, these studies have limitations. First, few studies have directly examined whether individuals with lower SES experience slower and less effective employment recovery compared to those with higher SES. Secondly, it remains unclear whether individuals with lower SES are less likely to be re-employed or return to their previous employment status after job loss due to a disaster. While some argue that their flexibility in the labor market might facilitate re-employment or return to previous positions (Amuedo-Dorantes & Serrano-Padial, 2010; Hallegatte et al., 2020), further research is needed to definitively demonstrate the impact of SES on post-disaster work recovery. Second, existing research has not adequately explored the connection between post-disaster employment changes and income inequality among individuals with different levels of SES. Social vulnerability theory posits that disparities in employment outcomes between individuals with varying SES during and after disasters are a primary driver of increased income inequality (Fothergill & Peek, 2004 ; Wisner et al., 2004). However, previous studies have not directly examined this causal pathway between employment changes and post-disaster income inequality. To address these research gaps, this study utilizes the 3.11 earthquake in Japan as an intensive case study, employing quantitative methods to: 1) examine the differential impact of the disaster on employment across various socioeconomic statuses (SES); 2) investigate disparities in employment recovery among individuals with different SES after the disaster; and 3) demonstrate the causal mechanism through which employment changes contribute to income inequality among individuals with different SES in the post-disaster period. 2 Literature Review Social vulnerability theory suggests that individuals with different socioeconomic backgrounds experience varying employment outcomes during disasters. These disparities can lead to increased income inequality. This study reviews existing literature on these differences across three stages: (1) employment changes during the immediate response phase (first 6 months to 1 year after the disaster), (2) employment changes during the long-term recovery phase (starting 1 year after the disaster), and (3) the potential for widening income inequality following the event. Disasters have four stages: mitigation, preparedness, response, and recovery. This study focuses on the response and recovery stages. The response stage is considered the short-term period following the disaster, typically lasting from 6 months to 1 year. The recovery stage is the long-term phase, taking place for at least one year after the event. This research adopts these definitions, considering the "short-term phase" as the period from 6 months to 1 year after the disaster and the "long-term phase" as starting 1 year after the event 2.1 SES and Unemployment in the Disaster Response Phase Understanding unemployment patterns following disasters is crucial for accurate economic modeling during disaster response. This section examines existing literature on the relationship between socioeconomic status (SES) and job loss during this critical phase. Two primary mechanisms explain how SES might influence unemployment risk in disaster scenarios. The first mechanism centers on the concept of employment flexibility. Studies suggest that individuals with higher SES often occupy positions with greater occupational and employment security, such as managerial or professional roles. These positions typically offer less flexibility, translating to lower layoff probabilities. Conversely, those with lower SES tend to hold more flexible jobs in unskilled or service sectors (Shavers, 2007). While this flexibility allows for quicker adjustments within the workforce, it also signifies a more precarious employment situation. The second mechanism relates to the impact of disasters on businesses. Disasters can severely disrupt business operations through property damage, supply chain disruptions, and customer displacement. These disruptions often lead to increased costs and decreased profits, forcing businesses to downsize their workforce. Layoffs become a necessary consequence, disproportionately affecting those already facing job insecurity, particularly individuals with lower SES (Tierney, 2007). In essence, the inherent flexibility of lower-SES jobs makes them more susceptible to elimination during economic downturns triggered by disasters. This pre-existing vulnerability is further exacerbated by disaster-induced business disruptions, potentially leading to a significant unemployment burden for low-SES populations during disaster response. Furthermore, lower SES can negatively impact health outcomes following disasters. Studies indicate that preparatory behaviors, such as disaster awareness, training, and information-seeking, can significantly reduce injury and death (Paton, 2003; Frahm et al., 2014; Teo et al., 2018; Metaxa-Kakavouli et al., 2018). However, research suggests that individuals with lower SES are less likely to engage in these preparatory actions compared to their higher-SES counterparts (Fothergill & Peek, 2004). Additionally, those with lower SES are more likely to live in older or more vulnerable housing, which is more susceptible to damage during disasters, potentially leading to injuries (Fothergill & Peek, 2004). Poorer health can lead to lower job performance and increased risk of job loss (Gómez & Nicolás, 2006). As disasters can disproportionately impact the health of lower-SES populations, this can translate to a higher likelihood of unemployment due to poorer work performance in the aftermath of a disaster. Based on these mechanisms, we believe that people with lower SES status are more likely to lose their jobs in disasters than people with higher SES; accordingly, this study proposes Hypothesis 1 as follows: Hypothesis 1 : Individuals with higher socioeconomic status (SES) are less likely to experience job loss during disasters, while those with lower SES are more likely to become unemployed. 2.2 SES and Employment Recovery in the Post-disaster Recovery Phase The differences in employment recovery between individuals with lower and higher SES have long-term effects on income inequality in the post disaster recovery phase. This section reviews these differences. Disasters can exacerbate existing income inequality, leading to long-term socioeconomic disparities. This section explores the differential job recovery experiences of individuals from varying socioeconomic backgrounds (SES) in the aftermath of disasters. Two primary mechanisms contribute to these disparities: human capital and social capital. Human capital, encompassing an individual's skills and knowledge acquired through education, training, job experience, and even health improvements (Becker, 1964; Schultz, 1961), plays a crucial role in job acquisition. Studies by Bloch and Smith (1977) demonstrate that individuals with a broader human capital base are demonstrably more adept at searching for jobs, translating to a higher probability of securing new employment. The link between human capital and SES is well-established. Individuals with higher levels of education and job training tend to occupy higher SES positions, while those with less human capital typically fall into lower SES brackets (Bloch & Smith, 1977). Social capital, another key factor influencing job search success, refers to the social networks of individuals and the resources embedded within those networks (Lin, 2001; Mouw 2003; Aldrich, 2012). As Lin et al. (2001) point out, social capital encompasses various definitions, but can be broadly understood as the connections people have with others and the resources available through these relationships. Social networks, particularly those with diverse connections (bridging ties, cf. Aldrich 2019), provide valuable job information and substantial support during a job search (Lin et al., 2001). Research suggests that individuals with higher SES have more opportunities to build broader social networks, consequently accumulating greater social capital (Lin et al., 2001). Conversely, those with lower SES have fewer chances to expand their networks, resulting in limited social capital. Bloch and Smith's (1977) work on human capital and Lin's (2001) research on social capital highlight that both significantly influence job search success. Following job losses due to disasters, individuals with higher SES are more likely to find new employment due to their richer human and social capital resources. In contrast, those with lower SES face a double challenge: they are more likely to lose their jobs initially and have fewer resources to find new ones quickly. This disparity ultimately widens the income gap during disaster recovery. Based on capital theory, since, compared to people with higher SES, people with lower SES are less likely to be re-employed after disasters, this study proposes Hypothesis 2a as follows: Hypothesis 2a : Individuals with lower SES are less likely to be re-employed after disasters compared to those with higher SES. The second mechanism potentially involved in disasters involves employment flexibility. However, this approach provides another way to frame the possibility that people with higher SES status may see worse recoveries after disasters. As mentioned above, the employment status of people with low SES is more flexible (Amuedo-Dorantes & Serrano-Padial, 2010; Hallegatte et al., 2020), and most of the jobs provided by the labor market involve flexible labor (Hippel et al., 1997). As a result, people with low SES status may be better able to find jobs after disasters. However, as people with high SES status originally had relatively stable jobs, should they lose their jobs, they may have more difficulty in finding jobs after the shock. This is because post-disaster labor markets cannot easily provide such high-income, stable jobs, and the requirements may be higher (Hippel et al., 1997). Based on this theory, we propose Hypothesis 2b as follows: Hypothesis 2b : People with lower SES status are more likely to be re-employed, while those with higher SES are less likely to be re-employed. 2.3 SES, Employment Recovery and Income Inequality As changes in employment are the main cause of income drops (or rises), in this section, we explore the impact of disasters on income inequality based on assumptions about employment changes for people with different SES status. First, during disaster response, according to Hypothesis 1, as people with lower SES are more likely to lose jobs when facing disasters, and as income generally flows from employment, people with higher SES will be more likely to obtain consistent income, while people with lower SES will be less likely to do so. Through this mechanism, income inequality may increase in the short term. Second, in terms of the phase of post-disaster recovery, there are two hypotheses. Based on Hypothesis 2a, as people with lower SES status may see worse recoveries than people with higher SES, the inequality in employment recovery among people with different SES expands after a disaster in the long term, and thus, income inequality increases. While based on Hypothesis 2b, because people with higher SES may see worse recoveries than people with lower SES, it seems like inequality in employment may decrease, and income inequality may also appear to be reduced between them. However, for society, inequality will likely still increase. As people who originally had high SES drop to lower SES status and people who originally had lower SES experience little (or negative) change in status after a disaster, the group of people with high SES shrinks, while the group with low SES grows. Thus, societal inequality grows. 3 Data and Methods 3.1 Data This study uses Japan’s 11 March 2011 triple earthquake as a natural experiment for studying the interaction between SES, income, employment, and inequality. At 2:46 pm on 11 March 2011, a massive, 9.0-magnitude earthquake struck Japan’s northeast region. That earthquake set off a series of tsunamis and a nuclear meltdown, which resulted in the direct deaths of more than 20,000 people across the Tohoku and Kanto regions (National Police Agency, 2018). After this disaster, the University of Tokyo implemented an internet survey named the Questionnaire Survey on Work and Hopes Following the Earthquake seeking to understand the changes in people’s work and lifestyle as of 2014, three years after the earthquake (cf. Aldrich 2019). We selected surveys completed in the Tohoku and Kanto regions in Japan which were greatly affected and damaged by the earthquake and tsunamis. These data contain individual information not only from the surveyed year but also from 2011 when the disaster occurred. In other words, this dataset uses the 3.11 triple disasters as a quasi-natural experiment and supports exploration of their impact on income inequality during and after the events. The survey population sampled residents aged 20 to 59 years (except students) in the Tohoku and Kanto regions, with a total of 10,466 respondents. For more accurate analyses, we set two restrictions on the sample size. The first is based on regions. The data were collected from the Tohoku and Kanto regions in 2014. In 2011, when the earthquake occurred, not all people were in the same place that they were in 2014, and some lived outside these regions. As only the Tohoku and Kanto regions were strongly affected by the disaster, this study is limited to people who lived in one of these regions in 2011. The second limitation involves unemployment. As we investigate the impact of disasters on income inequality from the perspective of individuals’ employment, the employment situation of people who were already unemployed before the disasters would not be affected by the disaster. Therefore, their situation does not conform to the mechanism we assumed and should be excluded from this dataset. Based on these limitations, after removing the missing values of the variables, the sample size for the analyses of this study drops to 3,820 but still remains large enough for drawing quantitative inferences. 3.2 Measurement Our core independent variable is socioeconomic status. To measure SES, scholars typically rely on individuals’ education and occupation (Shavers 2007). Education is captured by the question “What is your highest education?” and the options are “Junior high school,” “Senior high school,” “College of technology or Junior college,” “Vocational school,” “Undergraduate school,” “Graduate school,” and “Other.” We assign values according to respondents’ answers. Regarding occupations, we use employment status to measure people’s SES. There are two benefits to using employment status. First, it can indeed represent people’s SES, especially in the social context of Japan, where the special employment system called seniority and lifetime employment in organizations divides people’s employment into internal and external (Gordon 2017; Keizer 2008; Sato 2010; Sato and Imai 2011; Weathers 2009). People employed in the internal labor market are deemed regular employees and they have better salaries, benefits, and job stability (Sato 2010; Sato and Imai 2011). Therefore, people who are regularly employed could be categorized as having higher SESs in Japan. In contrast, people employed outside of the internal labor market are deemed to have irregular employment, and they have less income, no benefits and job instability (Sato 2010; Sato and Imai 2011). Thus, they can be categorized as lower SES in Japan. The gap in SES between individuals with these two types of employment status is extremely large in Japan—larger than those in other advanced industrial countries (Sato 2010; Sato and Imai 2011). Employment status can serve as an accurate indicator of the inequality between individuals with high SESs and low SESs during disasters. Second, compared to occupation, employment status can better represent job stability. One of the most important mechanisms by which disasters influence unemployment is through job stability. Occupation, however, indicates the content of people's work but cannot directly be traced to the stability of the job. Employment status, on the other hand, directly reflects the stability of employment. Regarding these two variables for measuring SES, people with a higher level of education or regular employment are categorized as having a higher SES, while people with a lower level of education or have irregular employment are categorized as having a lower SES. There are three major dependent variables in this study: disaster-caused unemployment, employment recovery and individual income. Disaster-caused unemployment is measured by the binary question “How was your work affected by the disaster—resignation?” with yes / no answers. Regarding employment recovery, two kinds of measurements can be applied here. The first one is reemployment after a disaster (Zottarelli 2008). Obtaining or reobtaining a job after a disaster is crucial because returning to the labor market means continuing to obtain income, while continued unemployment results in no (private sector) income after the shock. This is a key point in determining income inequality in the long term. The variable of reemployment is measured by the question “What is your current (2014) employment status?” This question has the same options as “What was your job in 2011 before the earthquake occurred?” This variable is only used in the analyses for people who lost their jobs in the 3.11 earthquake, and those who already had jobs in the survey year (2014) indicated that they had been reemployed, while those who still had no jobs in the survey year indicated that they had not yet been reemployed. Accordingly, if the respondents answered any type of employment status except “Unemployed”, meaning that they were reemployed in 2014 and are categorized as 1 “Employed”, while those who answered “Unemployed”, meaning that they were not reemployed in 2014, are categorized as 0 “Unemployed”. The second one concerns the return to predisaster employment (Zottarelli 2008). A change in employment after a disaster leads to a change in SES, especially in terms of income. Whether individuals with higher and lower SESs can reach the same level of employee wages after a disaster is also an essential issue influencing income inequality in the long term after a shock. Return to the same employment is generated by both variables of employment status in 2011 (before the disaster) and current employment status (2014). If people’s current employment status was the same as or higher than their 2011 employment status, they are categorized as returning to the same or a higher employment status, and the value is 1. If their current employment status was lower than their 2011 status, they are categorized as not returning to the same or a higher employment status, and the value is 0. Individual income is measured by the question “Please tell me about your individual income in the last year (2013)”. The 13 answer options range from “No income” to “JPY 15 million”. The median of each category is used as the income of that response. Because several confounding demographic variables may simultaneously affect the independent and dependent variables, these variables are also included in the analyses as controls. They are sex, age, industry in 2011, and residence in 2011. Detailed information on the variables is presented in Table 1. Table 1 Descriptive Statistics of Variables Variable N Mean/Percentage Standard deviation Min Max Disaster-caused unemployment 3,820 Yes 3.740 No 96.260 Reemployed in 2014 143 (1) Yes 79.020 No 20.980 Back in the same employment in 2014 143 (1) Yes 62.240 No 37.760 Education 3,820 14.623 1.958 9.000 18.000 Employment status in 2011 3,820 Regular employment 70.580 Irregular employment 24.840 Self-employment 4.580 Current individual income (Ten thousand ¥) (2) 3,820 461.531 331.678 0.000 1750.000 Sex 3,820 Male 60.080 Female 39.920 Age 3,820 40.645 9.306 20.000 59.000 Industry in 2011 3,820 Primary 0.650 Secondary 27.510 Tertiary 71.830 Residence in 2011 3,820 Tohoku region 17.960 Other regions 82.040 Note: (1) For the sample size of reemployment and return to the same employment status, as the analyses regarding the post disaster recovery phase focus only on people who were unemployed due to the disaster, the sample of these two variables used for the analyses in post disaster recovery is 143 rather than 3,820. (2) The distribution of individual income in the selected sample is already the normal distribution; thus, it does not need to be changed as a logarithm. 3.3 Methodology There are three steps in the analysis, dividing the analysis into three parts. The first part explores which type of SES status was more likely to lead to unemployment due to the disaster. Because disaster-caused unemployment is a binomial variable, the logistic regression method is most appropriate for analysis. The second part explores which type of SES is more likely to have led to reemployment after the disaster, and it is limited to people who became unemployed due to the disaster. As whether an individual was reemployed is a binomial variable, we used the logistic regression for analysis. The third part attempts to demonstrate the different impacts of disaster-caused unemployment on the individual incomes of people with different SES statues. Generally, when assessing the impact of disasters on income, if we have both pre- and post-disaster data and areas that are affected and unaffected by the disaster, we can use the Difference-in-Differences (DID) method to evaluate the impact of the disaster on income. However, since the data in this study does not include pre-disaster income variables, it is not possible to assess the disaster's impact on income using DID. Therefore, we only employ a general linear regression for the analysis. Current individual income is used as the dependent variable, and the ordinary least squares (OLS) equation is applied in this part. To examine regional differences in the impact of the disaster, we also conducted a multilevel linear regression for result validation. We controlled for regions at the macro level and performed the analysis accordingly. Since the results from the multilevel linear regression were consistent with those from the logistic and linear regression analyses, we have chosen to present only the results from the logistic and linear regressions in this paper to save space. 4 Results Following the steps of the analysis, the results are also presented in three parts. The first part shows which employment status and education level were more likely to lead to unemployment due to the disaster. The second part attempts to show the results regarding the relationship between SES and reemployment after the disaster and the results regarding the relationship between SES and returning to the same employment status after the disaster. The third part shows the results regarding the different impacts of disaster-caused unemployment on the individual incomes of people with different SES. The results of the first two parts are summarized in the three models in Table 2 . Table 2 Results of Disaster-caused Unemployment, Reemployment and Return to the Same Employment Status Model 1 Model 2 Model 3 Disaster-caused unemployment Reemployed in 2014 Back in the same employment in 2014 Education -0.148* 0.354* 0.269* (0.064) (0.153) (0.130) Employment status in 2011 (Ref: Regular) Irregular 0.724*** -0.025 1.718*** (0.201) (0.523) (0.491) Self-employment -0.534 -0.569 1.136 (0.598) (1.009) (0.999) Male (Ref: Female) -0.126 0.956† 1.069* (0.203) (0.539) (0.479) Age -0.018† -0.026 -0.017 (0.010) (0.024) (0.022) Industry in 2011 (Ref: Primary) Secondary / -0.009 / (0.580) Tertiary -0.048 / 0.292 (0.210) (0.459) Tohoku region (Ref: Other regions) 0.835*** -0.496 -0.041 (0.186) (0.467) (0.412) Constant -2.329*** 1.011 -1.040 (0.477) (1.215) (1.004) Observations 3,820 143 143 Nagelkerke R 2 0.060 0.123 0.178 AIC 1171.386 151.125 185.520 BIC 1221.318 174.828 209.223 Standard errors are in parentheses. *** p < 0.001, ** p < 0.01, * p < 0.05, † p < 0.1 Model 1 in Table 2 shows the results regarding disaster-caused unemployment. For education level, the coefficient of the result is significantly negative, meaning that more educated people had a lower probability of losing their jobs due to the disaster. Regarding employment status, only irregular employment before the disaster has a positive and significant relation to unemployment in the disaster, while self-employment does not. As the reference category is regular employment, this result means that compared with people who were regularly employed, irregularly employed people were more likely to lose their jobs in the disaster. Since the results regarding both education and employment status show a similar tendency of people being unemployed, it can be concluded that people with lower SESs are more likely to lose their jobs in disasters. The results provide empirical evidence supporting Hypothesis 1 . The variables of age and education are significantly and positively associated with disaster-caused unemployment, meaning that people who were older and had higher education were less likely to lose their jobs in the disaster. However, the variable of the Tohoku region is positively and significantly associated with disaster-caused unemployment, suggesting that people who lived in that region were more likely to lose their jobs in the disaster. Regarding the result for industry in 2011, as the sample size of people who worked in the secondary industry is relatively small in the dataset, there were no people in it who became unemployed because of the disaster. Accordingly, there is no coefficient in the category of secondary industry. Model 2 in Table 2 shows the results regarding reemployment. The results regarding education and employment status are mixed. Education shows a positive and significant result, meaning that people who had a higher level of education were more likely to be reemployed after the disaster. Regarding employment status in 2011, irregular employment has no significance for being reemployed, meaning that both regularly employed and irregularly employed people had the same probability of being reemployed after the disaster. Model 3 in Table 2 shows the results regarding returning to the same employment status after the disaster. The results regarding employment status and education are opposite. Education shows a positive and significant coefficient, meaning that people with a higher level of education were more likely to return to the same employment status after the disaster. However, for employment status, irregular employment is significantly and positively associated with returning to the same employment status, meaning that people who were regularly employed before the disaster were more likely to find jobs with the same employment status before the disaster. Based on the results in Model 2 and 3, it is hard to determine whether Hypothesis 2a or 2b is supported. For the variable of education level, the results are consistent in both Model 2 and 3, indicating that people with lower level of education have lower possibility to be reemployed and return to some employment status. These results supported Hypothesis 2a. However, for the variable of employment status, the result in Model 3 indicated that people regularly employed are less likely to return to the same employment status, which supported Hypothesis 2b. The results of the third part of the analyses are summarized in Table 3 below. This part aims to determine which factor is ultimately related to income inequality. Table 3 Results for Individual Income Model 4 Model 5 Current individual income Current individual income Disaster-caused unemployment (Ref: Nonaffected) -103.967*** -126.694*** (21.116) (22.399) Education 32.498*** 32.360*** (1.946) (1.945) Employment status in 2011 (Ref: Regular) Irregular -245.715*** -247.514*** (10.054) (10.068) Self-employment -221.963*** -221.352*** (18.839) (19.011) Interaction term Disaster-caused unemployment ×education -0.819 (10.286) Disaster-caused unemployment ×irregular 113.656** (40.872) Disaster-caused unemployment ×self-employment 64.387 (139.611) Male (Ref: Female) 177.863*** 177.422*** (21.116) (9.082) Age 9.007*** 8.980*** (0.433) (0.433) Industry in 2011 (Ref: Primary) Secondary 11.721 11.986 (48.308) (48.284) Tertiary 33.511*** 33.756*** (8.770) (8.763) Tohoku region (Ref: Other regions) -66.599*** -66.238*** (10.067) (10.059) Constant 58.516 -7.876 (17.667) (17.293) Observations 3,820 3,820 Adjusted R 2 0.498 0.499 Standard errors are in parentheses. *** p < 0.001, ** p < 0.01, * p < 0.05, † p < 0.1 Table 3 has two models: Model 4 presents the results of the interaction term between disaster-caused unemployment and education, and Model 5 reveals the results of the interaction term between disaster-caused unemployment and employment status in 2011 to provide a clearer demonstration of the different impacts of the disaster on the individual incomes of people with different SES. In Model 4, the interaction term between disaster-caused unemployment and education shows no significant effect on individual income, meaning that the influences of disaster-caused unemployment do not differ among people with different education levels. In Model 5, the interaction term between disaster-caused unemployment and employment status in 2011 reveals that disaster-caused unemployment × irregular employment is positively and significantly associated with individual income, and the main effect of disaster-caused unemployment is negatively and significantly associated with it. As the reference category of employment status in 2011 is regular employment, the impact of disaster-caused unemployment on regular employment is -126.694 (p < 0.001), while it becomes − 13.038 (-126.694 + 113.656) for irregular employment. This means that disaster-caused unemployment impacted individual income more for regularly employed people than for irregularly employed people before the disaster. To further clarify this difference, Figs. 1 and 2 visualize the results in Models 4 and 5, respectively. To clarify the different impacts of the disaster among people with different education levels on individual income, Fig. 1 simplifies education into two categories: people whose education level was equal to or below technical or junior college are categorized as having lower education, and people whose education level is higher than these degrees are categorized as having higher education. The lefthand part of Fig. 1 shows the difference in individual income between unaffected people and unemployed people following the disaster among people who had lower education, while the righthand part shows the difference in individual income between unaffected people and unemployed people among people who had higher education. This figure suggests that there were no differences in the disaster’s impact on individual income between lower- and higher-educated people. Therefore, it provides evidence that although education caused the differences in being reemployed and returning to the same employment status, it did not ultimately lead to differences in individual income. Regarding Fig. 2 , the lefthand part shows the difference in individual income between unaffected people and unemployed people following the disaster among those who were regularly employed before the disaster, while the righthand part shows the difference in individual income between unaffected people and unemployed people among those who were irregularly employed before the disaster. The difference in individual income between unaffected people and unemployed people was much larger among regularly employed people than irregularly employed people. In fact, the 95% confidence interval in the irregular employment part overlaps, meaning that there may have been no significant difference in income between unaffected people and unemployed people among those with irregular employment before the disaster, while there was a significant difference for those with regular employment. This figure reveals that disaster-caused unemployment impacted individual income more for regularly employed people than for irregularly employed people before the disaster. The results of these two interactions suggest that the difference in individual income may be caused by the gap regarding the return to the same employment status between those with regular and irregular employment. As regularly employed people have a lower probability of returning to the same employment status after a disaster, their income decreases more, while irregularly employed people have a higher probability, so their income decreases less. 5 Discussion According to previous research, disasters can exacerbate income inequality through changes in employment patterns. Our findings support this theory. During the disaster response phase, individuals with lower socioeconomic status (SES), characterized by lower education levels and irregular employment, were more likely to experience job loss. This suggests that income inequality can widen as a result of these differential employment outcomes between individuals with higher and lower SES. These findings align with the mechanisms of employment flexibility and health-related vulnerabilities, and they confirm the accuracy of social vulnerability theory in explaining how individuals with lower SES are more susceptible to the negative impacts of disasters. However, the results for the post-disaster recovery phase were mixed. The two measures of SES - education level and employment status - yielded different outcomes, necessitating further discussion. Regarding education, the findings indicated that individuals with lower levels of education were less likely to be re-employed and return to their previous employment status. These results align with capital theory. However, concerning employment status, the results showed no significant difference in re-employment rates between regularly and irregularly employed individuals after the disaster. Moreover, irregularly employed individuals even had a slight advantage in returning to their previous employment status compared to regularly employed individuals. These findings are consistent with the mechanism of employment flexibility. To reconcile these divergent findings and determine their impact on individual income, we analyzed income data. The analysis revealed no significant change in the income gap between lower- and higher-educated individuals affected by the disaster. In contrast, it showed a reduction in the income difference between regularly and irregularly employed individuals affected by the disaster. These findings suggest that education is not the most crucial factor contributing to income inequality, and changes in employment status played a more significant role in shaping income disparities during the disaster. However, based on the result of the decreasing income gap, can we say that income inequality between people with regular and irregular employment was reduced by the disaster? The answer is no. This is because individuals who originally had high SES (regular employment) dropped to lower SES, while those who originally had low SES (irregular employment) experienced no significant change after the disaster. As a whole, the number of people with high SES decreased, and the number of people with low SES increased; thus, societal inequality expanded. This demonstrates how status changes, rather than a poverty trap, can exacerbate inequality after a disaster. Furthermore, as mentioned in the literature, organizations tend to hire people with lower costs (Hippel et al. 1997 ). Irregular employment costs less than regular employment and is thus favored by organizations. Since the specific human assets obtained from previous regular employment are useless, to cut costs, employers are less likely to hire people who were regularly employed for regular employment. Thus, people who were regularly employed before a disaster are less likely to return to the same employment status, while irregularly employed people are more likely to be reemployed in irregular employment after the disaster. This is why in the results, people with irregular employment had a higher probability of returning to the same employment status than people with regular employment. Education is different. Education level represents general rather than specific human capital or assets in both work and study (Becker, 1964 ; Schultz, 1961 ). Therefore, individuals with higher levels of education are likely to possess greater work abilities, while those with lower levels of education are likely to have lower abilities. In labor markets, organizations tend to prefer hiring individuals with higher levels of education due to these ability differences (Bloch & Smith, 1977 ). Consequently, individuals with higher levels of education had a higher probability of being re-employed after the disaster compared to those with lower levels of education. Moreover, individuals with higher levels of education are more likely to be employed in higher-status jobs, while those with lower levels of education are more likely to be employed in lower-status jobs (Esteban-Pretel & Fujimoto, 2020 ). Thus, during the re-employment process, individuals with higher levels of education are more likely to secure higher-status employment, even if they were previously employed in lower-status positions. Conversely, individuals with lower levels of education are more likely to be employed in lower-status positions, even if they were previously employed in higher-status positions. This explains why, in the results, individuals with higher levels of education had a higher probability of returning to the same or a higher employment status compared to those with lower levels of education after the disaster. Based on these research findings, the policy implications for Japan include enhancing post-disaster support for lower socioeconomic status (SES) groups, particularly through job retraining, social security, and job stability measures. Additionally, implementing flexible labor market policies can facilitate quicker employment recovery. Moreover, policies should ensure that higher SES groups are not disproportionately hindered in their post-disaster employment recovery. Globally, these results suggest that other disaster-prone countries should prioritize protecting socially vulnerable groups by adopting comprehensive social safety nets, vocational training, and flexible employment policies to mitigate the exacerbation of income inequality during disaster recovery. Although this study makes several contributions to the research on disasters and income inequality, it has limitations. First, there are still disadvantages in measuring SES. Generally, occupation remains the most efficient way to do so. However, the dataset used in this study contained no questions concerning respondents' occupation. Therefore, we were forced to use employment status instead of occupation to measure SES. Should scholars seek to expand the application of the results obtained in this study, data on occupation are needed. Therefore, in future research, we suggest researchers collect data with measurements of occupation and confirm the results obtained in this study. Second, the sample size of those with disaster-caused unemployment is small. The data include only 143 samples for analysis, which is not sufficient to accurately estimate the coefficient of the regression. Even if the analyses were broadened to the full sample size of the data, which is 10,466, there are only 154 cases of disaster-caused unemployment, meaning that only 1.5% of people in the sample lost their jobs in the 3.11 earthquake. To check the representativeness of these data and numbers, we also calculated the disaster-caused unemployment rate in the 3.11 earthquake using the data of the 2012 Employment Status Basic Survey (Shugyo Kozo Kihon Chosa), one of the censuses in Japan (Statistics Bureau of Japan, 2012 ). These data show that in 2012, the population over age 15 was approximately 110 million, and the cases of disaster-caused unemployment in the 3.11 earthquake amounted to 2.3 million. Thus, the disaster-caused unemployment rate across all of Japan was 2%, which is very close to the rate in the data utilized by this study. Therefore, it can be said that the disaster-caused unemployment cases and results in this study accurately represent the broader universe of cases. Third, the income difference between people with regular and irregular employment in the short term after the disaster was not directly investigated. This study estimated only the difference in disaster-caused unemployment between people with regular and irregular employment and used this point to speculate about the income difference rather than directly demonstrating the effect. This is because the survey did not contain a question regarding individual income in the short term after the 3.11 triple disasters. However, as unemployment is closely related to decreasing income, especially individual income, the estimation of employment can be considered a reliable proxy for income changes. 6 Conclusion This study utilized survey data collected after the 3.11 triple disasters in Japan to dynamically demonstrate the changes in employment caused by the disaster and its consequences on income inequality between people with higher and lower socioeconomic status (SES) from both the disaster response and disaster recovery phases. The analyses yielded four main results. First, it found that in the disaster response phase, individuals with lower SES, characterized by lower education and irregular employment, were more likely to face job loss than those with higher SES. Second, the results regarding post-disaster reemployment were mixed. Employment status showed no significant effect on reemployment, while education presented a positive effect. Third, the results on the relationship between SES and returning to the same employment status were also mixed. Employment status showed a negative effect on returning to the same employment status, while education presented a positive effect. Finally, the results on individual income showed that the income decrease caused by the disaster was greater among individuals who were regularly employed than those who were irregularly employed before the disaster. In summary, the disaster indeed amplified income inequality; however, the mechanism was not solely based on social vulnerability but also involved employment flexibility. Accordingly, for effective post-disaster social governance, policymakers should not only target recovery policies at people with low socioeconomic status but also consider the needs of wealthier individuals. Declarations Funding Declaration This paper is supported by National Social Science Fund of China (Grant No.:22CSH095). Ethics Approval Statement This study involved secondary analysis of publicly available data provided by the Social Science Japan Data Archive (SSJDA), under dataset ID 1036 ("Survey on Work and Hope after the Great East Japan Earthquake, 2014"). The original survey was designed and supervised by Professor Yuji Genda (Institute of Social Science, University of Tokyo) and implemented by INTAGE Research Inc., in accordance with established ethical guidelines for social science research. Access to this dataset was officially granted by SSJDA. The SSJDA requires researchers to submit a formal application for dataset use, and approval of this application serves as an equivalent process to Institutional Review Board (IRB) review, ensuring compliance with ethical principles such as confidentiality of participants and informed consent. All research activities based on this dataset were conducted in accordance with relevant regulations and ethical standards, including the principles of the Declaration of Helsinki. Approval Body : Social Science Japan Data Archive (SSJDA), Institute of Social Science, University of Tokyo Approval Number / ID : SSJDA Dataset ID 1036 Dataset DOI : 10.34500/SSJDA.1036 Date of Approval (Dataset Release Date) : March 10, 2016 Scope of Approval : Secondary analysis of anonymized, publicly available survey data from the 2014 “Survey on Work and Hope after the Great East Japan Earthquake.” This approval covers the use of data solely for academic research purposes, with participant confidentiality strictly protected. Supporting Information : Details on dataset access procedures and ethical compliance are available on the SSJDA official website (https://csrda.iss.u-tokyo.ac.jp/). Statement on Informed Consent The data utilized in this study were obtained from the Social Science Japan Data Archive (SSJDA), specifically dataset ID 1036 ("Survey on Work and Hope after the Great East Japan Earthquake, 2014"). According to the original data collection documentation, informed consent was duly obtained from all individual participants included in the original survey conducted by Professor Yuji Genda of the Institute of Social Science, University of Tokyo, and administered by INTAGE Research Inc. Participants explicitly agreed to participate after being fully informed about the purpose, confidentiality measures, and voluntary nature of their involvement. All procedures followed during data collection complied with ethical standards ensuring participants' consent and anonymity. References Aldrich, D. P. 2012. Building Resilience: Social Capital in Post-Disaster Recovery. Chicago: University of Chicago Press. Aldrich, D. P. 2019. Black Wave: How Networks and Governance Shaped Japan’s 3/11 Disasters. Chicago: University of Chicago Press. Amuedo-Dorantes, Catalina and Ricardo Serrano-Padial. 2010. “Labor Market Flexibility and Poverty Dynamics.” Labour Economics 17: 632–642. Becker, G. S. 1964. Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. Chicago: The University of Chicago Press. Bloch, F. E., and S. P. Smith. 1977. Human Capital and Labor Market Employment. Journal of Human Resources 12(4): 550–560. Esteban-Pretel, J., and J. Fujimoto. 2020. Non-regular Employment over the Life-cycle: Worker Flow Analysis for Japan. Journal of The Japanese and International Economies 57: 1–20. Fang, L., J. Wu, and T. Miljkovic. 2017. Modeling Impact of Natural Hazard-Induced Disasters on Income Distribution in the United States. International Journal of Disaster Risk Science 8(4): 435–444. Fothergill, A., and L. A. Peek. 2004. Poverty and Disasters in the United States: A Review of Recent Sociological Findings. Natural Hazards 32: 89–110. Frahm, K. A., P. J. Gardner, L. M. Brown, D. P. Rogoff, and A. Troutman. Community-Based Disaster Coalition Training. Journal of Public Health Management and Practice 20(5): S111–S117. Gómez, P. G., and A. L. Nicolás. 2006. Health Shocks, Employment and Income in the Spanish Labour Market. Health Economics 15 (9): 997–1009. Gordon, Andrew. 2017. New and Enduring Dual Structures of Employment in Japan: The Rise of Non-Regular Labor, 1980s–2010s. Social Science Japan Journal 20(1): 9–36. Hallegatte, Stéphane, Adrien Vogt-Schilb, Julie Rozenberg, Mook Bangalore and Chloé Beaudet. 2020. “From Poverty to Disaster and Back: A Review of the Literature.” Economics of Disasters and Climate Change 4:223–247. Hippel, C., S. L. Mangum, D. B. Greenberger, R. L. Heneman, and J. D. Skoglind. 1997. Temporary Employment: Can Organizations and Employees Both Win? Academy of Management Executive 11(1): 93–104. Howell, J., and J. R. Elliott. 2019. Damages Done: The Longitudinal Impacts of Natural Hazards on Wealth Inequality in the United States. Social Problems 072018(0): 1–20. Keerthiratne, S., and R. S. J. Tol. 2018. Impact of natural disasters on income inequality in Sri Lanka. World Development 105: 217–230. Keizer, Arjan. 2008. Non-regular Employment in Japan: Continued and Renewed Dualities. Work, Employment and Society 22(3): 407–425. Lin, N. 2001. Social Capital: A Theory of Social Structure and Action. England: Cambridge University Press. Lin, N., K. Cook, and R. S. Burt. 2001. Social Capital: Theory and Research. New York: Editors Press. Metaxa-Kakavouli, D., P. Maas, and D. P. Aldrich. 2018. How Social Ties Influence Evacuation Behavior. Proceedings of the ACM on Human-Computer Interaction 2(CSCW):1–16 Miljkovic, T., and D. Miljkovic. 2014. Modeling Impact of Hurricane Damages on Income Distribution in the Coastal U.S. International Journal of Disaster Risk Science 5: 265–273. Mouw, T. 2003. Social Capital and Finding a Job: Do Contacts Matter? American Sociological Review 68(6): 868–898. Paton, D. 2003. Disaster Preparedness: A Social-cognitive Perspective. Disaster Prevention and Management 12(3): 210–216. Reid, M. 2013. Disasters and Social Inequalities. Sociological Compass 7(11): 984–997. Sato, Y. 2010. Stability and Increasing Fluidity in the Contemporary Japanese Social Stratification System. Contemporary Japan 22: 7–21. Sato, Y. and J. Iami. 2011. Japan’s New Inequality: Intersection of Employment Reforms and Welfare Arrangements. Tokyo: Trans Pacific Press. Schultz, T. W. 1961. Investment in Human Capital. American Economic Review 51(1): 1–17. Shavers, V. L. 2007. Measurement of Socioeconomic Status in Health Disparities Research. Journal of The National Medical Association 99(9): 1013–1023. Statistics Bureau of Japan. 2012. 2012 Employment Status Basic Survey (Shugyo Kozo Kihon Chosa). https://www.stat.go.jp/data/shugyou/2012/index2.html. Teo, M., A. Goonetilleke, A. Ahankoob, K. Deilami, and M Lawie. 2018. Disaster Awareness and Information Seeking Behaviour among Residents from Low Socio-economic Backgrounds. International Journal of Disaster Risk Reduction 31: 1121–1131. Tierney, K. J. 2007. Businesses and Disasters: Vulnerability, Impact, and Recovery. in Handbook of Disaster Research, ed. H. Rodríguez, E. L. Quarantelli, and R. R. Dynes, 275–297. New York: Springer. Weathers, Charles. 2009. Nonregular Workers and Inequality in Japan. Social Science Japan Journal 12(1): 143–148. Yamamura, E. 2015. The Impact of Natural Disasters on Income Inequality: Analysis using Panel Data during the Period 1970 to 2004. International Economic Journal 29(3): 359–374. Zottarelli, L. K. 2008. Post-Hurricane Katrina Employment Recovery: The Interaction of Race and Place. Social Science Quarterly 89(3): 593–607. Additional Declarations No competing interests reported. Cite Share Download PDF Status: Under Review Version 1 posted Reviewers invited by journal 07 Feb, 2026 Editor invited by journal 24 Oct, 2025 Editor assigned by journal 12 Oct, 2025 Submission checks completed at journal 09 Sep, 2025 First submitted to journal 08 Sep, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7161036","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":512260066,"identity":"27164f36-0f53-4c04-9e9b-c3c44753dab1","order_by":0,"name":"Maoxin Ye","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA3klEQVRIie3PMQrCMBSA4SfF5/LUtUVFjxAIKILoVSqCUw/QsSA4ibvoIXqEQIcsOUBLBy1CXRwcBREtTk5p3RzyQyDD+0gegMn0xxE0gs+lFlQnJH4kALZbkTAZ1MO7H3VHziU7E0x6obDyk5YogclGLWm89zgnWPJQ4IhpSexi3FxPiKUedgiieSgIbS05njB5rm1iibw8CF4VSAyYfl6JYWgRiHLiKBfTbrELUx53DmzBdxEOtaQli/WvfjRjUma3qz/tbeUq15KBaN++/1kcSzdf1A9KBkwmk8kEb3VhSHlZpBlGAAAAAElFTkSuQmCC","orcid":"","institution":"Southeast University","correspondingAuthor":true,"prefix":"","firstName":"Maoxin","middleName":"","lastName":"Ye","suffix":""},{"id":512260067,"identity":"10b980fb-aa29-47ec-8a28-b8bcddc31eed","order_by":1,"name":"Daniel Aldrich","email":"","orcid":"","institution":"Northeastern University","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Aldrich","suffix":""}],"badges":[],"createdAt":"2025-07-19 00:38:08","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7161036/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7161036/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":91003923,"identity":"5a356044-dc37-4870-afc4-5aa7ae19363d","added_by":"auto","created_at":"2025-09-10 14:13:36","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":46397,"visible":true,"origin":"","legend":"\u003cp\u003eImpacts by Education on Income Differences for Individuals Affected and Unaffected by the Disaster\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7161036/v1/d24ea0b95e2ec0249e07ef04.png"},{"id":91004176,"identity":"56575218-928b-4b99-ae4e-3d4af146a995","added_by":"auto","created_at":"2025-09-10 14:21:36","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":57082,"visible":true,"origin":"","legend":"\u003cp\u003eImpacts by Employment Status on Income Differences for Individuals Affected and Unaffected by the Disaster\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7161036/v1/9b83747064524737cc227200.png"},{"id":91005394,"identity":"876bb042-92b1-4a90-8c1e-6958112910c3","added_by":"auto","created_at":"2025-09-10 14:29:36","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":864994,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7161036/v1/a15cd4a1-eca7-40e9-8dcf-3baee2ac1d59.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eVulnerability or Flexibility: An Examination of Socioeconomic Status, Employment Changes, and Individual Income in Japan’s 3.11 Earthquake\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1 Introduction","content":"\u003cp\u003eHow disasters impact employment and income remains a critical question in disaster research (Reid, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Individuals with lower socioeconomic status (SES) are particularly vulnerable to the shocks caused by disasters (Fothergill \u0026amp; Peek, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Wisner et al., 2004). As a result, their employment becomes more unstable during and after disasters, hindering their recovery efforts. This, in turn, contributes to increased income inequality (Elliott \u0026amp; Howell, 2016; Howell \u0026amp; Elliott, \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2019\u003c/span\u003e).\u003c/p\u003e\u003cp\u003ePrevious research suggests that individuals with lower socioeconomic status (SES) are more likely to experience job loss during disasters (Hallegatte et al., 2020) and that disasters can exacerbate income inequality (Milijkovic \u0026amp; Milijkovic, 2014; Yamamura, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Fang et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2017\u003c/span\u003e; Keerthiratne \u0026amp; Tol, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). However, these studies have limitations. First, few studies have directly examined whether individuals with lower SES experience slower and less effective employment recovery compared to those with higher SES. Secondly, it remains unclear whether individuals with lower SES are less likely to be re-employed or return to their previous employment status after job loss due to a disaster. While some argue that their flexibility in the labor market might facilitate re-employment or return to previous positions (Amuedo-Dorantes \u0026amp; Serrano-Padial, 2010; Hallegatte et al., 2020), further research is needed to definitively demonstrate the impact of SES on post-disaster work recovery.\u003c/p\u003e\u003cp\u003eSecond, existing research has not adequately explored the connection between post-disaster employment changes and income inequality among individuals with different levels of SES. Social vulnerability theory posits that disparities in employment outcomes between individuals with varying SES during and after disasters are a primary driver of increased income inequality (Fothergill \u0026amp; Peek, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Wisner et al., 2004). However, previous studies have not directly examined this causal pathway between employment changes and post-disaster income inequality.\u003c/p\u003e\u003cp\u003eTo address these research gaps, this study utilizes the 3.11 earthquake in Japan as an intensive case study, employing quantitative methods to: 1) examine the differential impact of the disaster on employment across various socioeconomic statuses (SES); 2) investigate disparities in employment recovery among individuals with different SES after the disaster; and 3) demonstrate the causal mechanism through which employment changes contribute to income inequality among individuals with different SES in the post-disaster period.\u003c/p\u003e"},{"header":"2 Literature Review","content":"\u003cp\u003eSocial vulnerability theory suggests that individuals with different socioeconomic backgrounds experience varying employment outcomes during disasters. These disparities can lead to increased income inequality. This study reviews existing literature on these differences across three stages: (1) employment changes during the immediate response phase (first 6 months to 1 year after the disaster), (2) employment changes during the long-term recovery phase (starting 1 year after the disaster), and (3) the potential for widening income inequality following the event.\u003c/p\u003e\n\u003cp\u003eDisasters have four stages: mitigation, preparedness, response, and recovery. This study focuses on the response and recovery stages. The response stage is considered the short-term period following the disaster, typically lasting from 6 months to 1 year. The recovery stage is the long-term phase, taking place for at least one year after the event. This research adopts these definitions, considering the \u0026quot;short-term phase\u0026quot; as the period from 6 months to 1 year after the disaster and the \u0026quot;long-term phase\u0026quot; as starting 1 year after the event\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.1 SES and Unemployment in the Disaster Response Phase\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnderstanding unemployment patterns following disasters is crucial for accurate economic modeling during disaster response. This section examines existing literature on the relationship between socioeconomic status (SES) and job loss during this critical phase. Two primary mechanisms explain how SES might influence unemployment risk in disaster scenarios. The first mechanism centers on the concept of employment flexibility. Studies suggest that individuals with higher SES often occupy positions with greater occupational and employment security, such as managerial or professional roles. These positions typically offer less flexibility, translating to lower layoff probabilities. Conversely, those with lower SES tend to hold more flexible jobs in unskilled or service sectors (Shavers, 2007). While this flexibility allows for quicker adjustments within the workforce, it also signifies a more precarious employment situation.\u003c/p\u003e\n\u003cp\u003eThe second mechanism relates to the impact of disasters on businesses. Disasters can severely disrupt business operations through property damage, supply chain disruptions, and customer displacement. These disruptions often lead to increased costs and decreased profits, forcing businesses to downsize their workforce. Layoffs become a necessary consequence, disproportionately affecting those already facing job insecurity, particularly individuals with lower SES (Tierney, 2007).\u003c/p\u003e\n\u003cp\u003eIn essence, the inherent flexibility of lower-SES jobs makes them more susceptible to elimination during economic downturns triggered by disasters. This pre-existing vulnerability is further exacerbated by disaster-induced business disruptions, potentially leading to a significant unemployment burden for low-SES populations during disaster response.\u003c/p\u003e\n\u003cp\u003eFurthermore, lower SES can negatively impact health outcomes following disasters. Studies indicate that preparatory behaviors, such as disaster awareness, training, and information-seeking, can significantly reduce injury and death (Paton, 2003; Frahm et al., 2014; Teo et al., 2018; Metaxa-Kakavouli et al., 2018). However, research suggests that individuals with lower SES are less likely to engage in these preparatory actions compared to their higher-SES counterparts (Fothergill \u0026amp; Peek, 2004). Additionally, those with lower SES are more likely to live in older or more vulnerable housing, which is more susceptible to damage during disasters, potentially leading to injuries (Fothergill \u0026amp; Peek, 2004). Poorer health can lead to lower job performance and increased risk of job loss (G\u0026oacute;mez \u0026amp; Nicol\u0026aacute;s, 2006). As disasters can disproportionately impact the health of lower-SES populations, this can translate to a higher likelihood of unemployment due to poorer work performance in the aftermath of a disaster.\u003c/p\u003e\n\u003cp\u003eBased on these mechanisms, we believe that people with lower SES status are more likely to lose their jobs in disasters than people with higher SES; accordingly, this study proposes Hypothesis 1 as follows:\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHypothesis 1\u003c/em\u003e: Individuals with higher socioeconomic status (SES) are less likely to experience job loss during disasters, while those with lower SES are more likely to become unemployed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2 SES and Employment Recovery in the Post-disaster Recovery Phase\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe differences in employment recovery between individuals with lower and higher SES have long-term effects on income inequality in the post disaster recovery phase. This section reviews these differences.\u003c/p\u003e\n\u003cp\u003eDisasters can exacerbate existing income inequality, leading to long-term socioeconomic disparities. This section explores the differential job recovery experiences of individuals from varying socioeconomic backgrounds (SES) in the aftermath of disasters. Two primary mechanisms contribute to these disparities: human capital and social capital.\u003c/p\u003e\n\u003cp\u003eHuman capital, encompassing an individual\u0026apos;s skills and knowledge acquired through education, training, job experience, and even health improvements (Becker, 1964; Schultz, 1961), plays a crucial role in job acquisition. Studies by Bloch and Smith (1977) demonstrate that individuals with a broader human capital base are demonstrably more adept at searching for jobs, translating to a higher probability of securing new employment. The link between human capital and SES is well-established. Individuals with higher levels of education and job training tend to occupy higher SES positions, while those with less human capital typically fall into lower SES brackets (Bloch \u0026amp; Smith, 1977).\u003c/p\u003e\n\u003cp\u003eSocial capital, another key factor influencing job search success, refers to the social networks of individuals and the resources embedded within those networks (Lin, 2001; Mouw 2003; Aldrich, 2012). As Lin et al. (2001) point out, social capital encompasses various definitions, but can be broadly understood as the connections people have with others and the resources available through these relationships. Social networks, particularly those with diverse connections (bridging ties, cf. Aldrich 2019), provide valuable job information and substantial support during a job search (Lin et al., 2001). Research suggests that individuals with higher SES have more opportunities to build broader social networks, consequently accumulating greater social capital (Lin et al., 2001). Conversely, those with lower SES have fewer chances to expand their networks, resulting in limited social capital.\u003c/p\u003e\n\u003cp\u003eBloch and Smith\u0026apos;s (1977) work on human capital and Lin\u0026apos;s (2001) research on social capital highlight that both significantly influence job search success. Following job losses due to disasters, individuals with higher SES are more likely to find new employment due to their richer human and social capital resources. In contrast, those with lower SES face a double challenge: they are more likely to lose their jobs initially and have fewer resources to find new ones quickly. This disparity ultimately widens the income gap during disaster recovery. Based on capital theory, since, compared to people with higher SES, people with lower SES are less likely to be re-employed after disasters, this study proposes Hypothesis 2a as follows:\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHypothesis 2a\u003c/em\u003e\u003cstrong\u003e:\u003c/strong\u003e Individuals with lower SES are less likely to be re-employed after disasters compared to those with higher SES.\u003c/p\u003e\n\u003cp\u003eThe second mechanism potentially involved in disasters involves employment flexibility. However, this approach provides another way to frame the possibility that people with higher SES status may see worse recoveries after disasters. As mentioned above, the employment status of people with low SES is more flexible (Amuedo-Dorantes \u0026amp; Serrano-Padial, 2010; Hallegatte et al., 2020), and most of the jobs provided by the labor market involve flexible labor (Hippel et al., 1997). As a result, people with low SES status may be better able to find jobs after disasters. However, as people with high SES status originally had relatively stable jobs, should they lose their jobs, they may have more difficulty in finding jobs after the shock. This is because post-disaster labor markets cannot easily provide such high-income, stable jobs, and the requirements may be higher (Hippel et al., 1997). Based on this theory, we propose Hypothesis 2b as follows:\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eHypothesis 2b\u003c/em\u003e\u003cstrong\u003e:\u003c/strong\u003e People with lower SES status are more likely to be re-employed, while those with higher SES are less likely to be re-employed.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3 SES, Employment Recovery and Income Inequality\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAs changes in employment are the main cause of income drops (or rises), in this section, we explore the impact of disasters on income inequality based on assumptions about employment changes for people with different SES status.\u003c/p\u003e\n\u003cp\u003eFirst, during disaster response, according to Hypothesis 1, as people with lower SES are more likely to lose jobs when facing disasters, and as income generally flows from employment, people with higher SES will be more likely to obtain consistent income, while people with lower SES will be less likely to do so. Through this mechanism, income inequality may increase in the short term.\u003c/p\u003e\n\u003cp\u003eSecond, in terms of the phase of post-disaster recovery, there are two hypotheses. Based on Hypothesis 2a, as people with lower SES status may see worse recoveries than people with higher SES, the inequality in employment recovery among people with different SES expands after a disaster in the long term, and thus, income inequality increases. While based on Hypothesis 2b, because people with higher SES may see worse recoveries than people with lower SES, it seems like inequality in employment may decrease, and income inequality may also appear to be reduced between them. However, for society, inequality will likely still increase. As people who originally had high SES drop to lower SES status and people who originally had lower SES experience little (or negative) change in status after a disaster, the group of people with high SES shrinks, while the group with low SES grows. Thus, societal inequality grows.\u003c/p\u003e"},{"header":"3 Data and Methods","content":"\u003cp\u003e\u003cstrong\u003e3.1 Data\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study uses Japan\u0026rsquo;s 11 March 2011 triple earthquake as a natural experiment for studying the interaction between SES, income, employment, and inequality. At 2:46 pm on 11 March 2011, a massive, 9.0-magnitude earthquake struck Japan\u0026rsquo;s northeast region. That earthquake set off a series of tsunamis and a nuclear meltdown, which resulted in the direct deaths of more than 20,000 people across the Tohoku and Kanto regions (National Police Agency, 2018). After this disaster, the University of Tokyo implemented an internet survey named the Questionnaire Survey on Work and Hopes Following the Earthquake seeking to understand the changes in people\u0026rsquo;s work and lifestyle as of 2014, three years after the earthquake (cf. Aldrich 2019). \u0026nbsp;We selected surveys completed in the Tohoku and Kanto regions in Japan which were greatly affected and damaged by the earthquake and tsunamis. These data contain individual information not only from the surveyed year but also from 2011 when the disaster occurred. In other words, this dataset uses the\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e3.11 triple disasters as a quasi-natural experiment and supports exploration of their impact on income inequality during and after the events.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe survey population sampled residents aged 20 to 59 years (except students) in the Tohoku and Kanto regions, with a total of 10,466 respondents. For more accurate analyses, we set two restrictions on the sample size. The first is based on regions. The data were collected from the Tohoku and Kanto regions in 2014. In 2011, when the earthquake occurred, not all people were in the same place that they were in 2014, and some lived outside these regions. As only the Tohoku and Kanto regions were strongly affected by the disaster, this study is limited to people who lived in one of these regions in 2011. The second limitation involves unemployment. As we investigate the impact of disasters on income inequality from the perspective of individuals\u0026rsquo; employment, the employment situation of people who were already unemployed before the disasters would not be affected by the disaster. Therefore, their situation does not conform to the mechanism we assumed and should be excluded from this dataset. Based on these limitations, after removing the missing values of the variables, the sample size for the analyses of this study drops to 3,820 but still remains large enough for drawing quantitative inferences.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 Measurement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eOur core independent variable is socioeconomic status. To measure SES, scholars typically rely on individuals\u0026rsquo; education and occupation (Shavers 2007). Education is captured by the question \u0026ldquo;What is your highest education?\u0026rdquo; and the options are \u0026ldquo;Junior high school,\u0026rdquo; \u0026ldquo;Senior high school,\u0026rdquo; \u0026ldquo;College of technology or Junior college,\u0026rdquo; \u0026ldquo;Vocational school,\u0026rdquo; \u0026ldquo;Undergraduate school,\u0026rdquo; \u0026ldquo;Graduate school,\u0026rdquo; and \u0026ldquo;Other.\u0026rdquo; We assign values according to respondents\u0026rsquo; answers.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRegarding occupations, we use employment status to measure people\u0026rsquo;s SES. There are two benefits to using employment status. First, it can indeed represent people\u0026rsquo;s SES, especially in the social context of Japan, where the special employment system called seniority and lifetime employment in organizations divides people\u0026rsquo;s employment into internal and external (Gordon 2017; Keizer 2008; Sato 2010; Sato and Imai 2011; Weathers 2009). People employed in the internal labor market are deemed regular employees and they have better salaries, benefits, and job stability (Sato 2010; Sato and Imai 2011). Therefore, people who are regularly employed could be categorized as having higher SESs in Japan. In contrast, people employed outside of the internal labor market are deemed to have irregular employment, and they have less income, no benefits and job instability (Sato 2010; Sato and Imai 2011). Thus, they can be categorized as lower SES in Japan. The gap in SES between individuals with these two types of employment status is extremely large in Japan\u0026mdash;larger than those in other advanced industrial countries (Sato 2010; Sato and Imai 2011). Employment status can serve as an accurate indicator of the inequality between individuals with high SESs and low SESs during disasters.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSecond, compared to occupation, employment status can better represent job stability. One of the most important mechanisms by which disasters influence unemployment is through job stability. Occupation, however, indicates the content of people\u0026apos;s work but cannot directly be traced to the stability of the job. Employment status, on the other hand, directly reflects the stability of employment.\u003c/p\u003e\n\u003cp\u003eRegarding these two variables for measuring SES, people with a higher level of education or regular employment are categorized as having a higher SES, while people with a lower level of education or have irregular employment are categorized as having a lower SES.\u003c/p\u003e\n\u003cp\u003eThere are three major dependent variables in this study: disaster-caused unemployment, employment recovery and individual income. Disaster-caused unemployment is measured by the binary question \u0026ldquo;How was your work affected by the disaster\u0026mdash;resignation?\u0026rdquo; with yes / no answers.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eRegarding employment recovery, two kinds of measurements can be applied here. The first one is reemployment after a disaster (Zottarelli 2008). Obtaining or reobtaining a job after a disaster is crucial because returning to the labor market means continuing to obtain income, while continued unemployment results in no (private sector) income after the shock. This is a key point in determining income inequality in the long term. The variable of reemployment is measured by the question \u0026ldquo;What is your current (2014) employment status?\u0026rdquo; This question has the same options as \u0026ldquo;What was your job in 2011 before the earthquake occurred?\u0026rdquo; This variable is only used in the analyses for people who lost their jobs in the 3.11 earthquake, and those who already had jobs in the survey year (2014) indicated that they had been reemployed, while those who still had no jobs in the survey year indicated that they had not yet been reemployed. Accordingly, if the respondents answered any type of employment status except \u0026ldquo;Unemployed\u0026rdquo;, meaning that they were reemployed in 2014 and are categorized as 1 \u0026ldquo;Employed\u0026rdquo;, while those who answered \u0026ldquo;Unemployed\u0026rdquo;, meaning that they were not reemployed in 2014, are categorized as 0 \u0026ldquo;Unemployed\u0026rdquo;.\u003c/p\u003e\n\u003cp\u003eThe second one concerns the return to predisaster employment (Zottarelli 2008). A change in employment after a disaster leads to a change in SES, especially in terms of income. Whether individuals with higher and lower SESs can reach the same level of employee wages after a disaster is also an essential issue influencing income inequality in the long term after a shock. Return to the same employment is generated by both variables of employment status in 2011 (before the disaster) and current employment status (2014). If people\u0026rsquo;s current employment status was the same as or higher than their 2011 employment status, they are categorized as returning to the same or a higher employment status, and the value is 1. If their current employment status was lower than their 2011 status, they are categorized as not returning to the same or a higher employment status, and the value is 0.\u003c/p\u003e\n\u003cp\u003eIndividual income is measured by the question \u0026ldquo;Please tell me about your individual income in the last year (2013)\u0026rdquo;. The 13 answer options range from \u0026ldquo;No income\u0026rdquo; to \u0026ldquo;JPY 15 million\u0026rdquo;. The median of each category is used as the income of that response.\u003c/p\u003e\n\u003cp\u003eBecause several confounding demographic variables may simultaneously affect the independent and dependent variables, these variables are also included in the analyses as controls. They are sex, age, industry in 2011, and residence in 2011. Detailed information on the variables is presented in Table 1.\u003c/p\u003e\n\u003cp\u003eTable 1 Descriptive Statistics of Variables\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eN\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eMean/Percentage\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eStandard deviation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eMin\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003eMax\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eDisaster-caused unemployment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3,820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3.740\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e96.260\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eReemployed in 2014\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e143\u003csup\u003e(1)\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e79.020\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e20.980\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eBack in the same employment in 2014\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e143\u003csup\u003e(1)\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e62.240\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e37.760\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eEducation\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3,820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e14.623\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1.958\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e9.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e18.000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eEmployment status in 2011\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3,820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Regular employment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e70.580\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Irregular employment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e24.840\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Self-employment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e4.580\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eCurrent individual income (Ten thousand \u0026yen;) \u003csup\u003e(2)\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3,820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e461.531\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e331.678\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1750.000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eSex\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3,820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Male\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e60.080\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e39.920\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3,820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e40.645\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e9.306\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e20.000\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e59.000\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eIndustry in 2011\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3,820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Primary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.650\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Secondary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e27.510\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Tertiary\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e71.830\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003eResidence in 2011\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3,820\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Tohoku region\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e17.960\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; Other regions\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e82.040\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e \u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003eNote: (1) For the sample size of reemployment and return to the same employment status, as the analyses regarding the post disaster recovery phase focus only on people who were unemployed due to the disaster, the sample of these two variables used for the analyses in post disaster recovery is 143 rather than 3,820. (2) The distribution of individual income in the selected sample is already the normal distribution; thus, it does not need to be changed as a logarithm.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Methodology\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThere are three steps in the analysis, dividing the analysis into three parts. The first part explores which type of SES status was more likely to lead to unemployment due to the disaster. Because disaster-caused unemployment is a binomial variable, the logistic regression method is most appropriate for analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe second part explores which type of SES is more likely to have led to reemployment after the disaster, and it is limited to people who became unemployed due to the disaster. As whether an individual was reemployed is a binomial variable, we used the logistic regression for analysis.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe third part attempts to demonstrate the different impacts of disaster-caused unemployment on the individual incomes of people with different SES statues. Generally, when assessing the impact of disasters on income, if we have both pre- and post-disaster data and areas that are affected and unaffected by the disaster, we can use the Difference-in-Differences (DID) method to evaluate the impact of the disaster on income. However, since the data in this study does not include pre-disaster income variables, it is not possible to assess the disaster\u0026apos;s impact on income using DID. Therefore, we only employ a general linear regression for the analysis. Current individual income is used as the dependent variable, and the ordinary least squares (OLS) equation is applied in this part.\u003c/p\u003e\n\u003cp\u003eTo examine regional differences in the impact of the disaster, we also conducted a multilevel linear regression for result validation. We controlled for regions at the macro level and performed the analysis accordingly. Since the results from the multilevel linear regression were consistent with those from the logistic and linear regression analyses, we have chosen to present only the results from the logistic and linear regressions in this paper to save space.\u003c/p\u003e"},{"header":"4 Results","content":"\u003cp\u003eFollowing the steps of the analysis, the results are also presented in three parts. The first part shows which employment status and education level were more likely to lead to unemployment due to the disaster. The second part attempts to show the results regarding the relationship between SES and reemployment after the disaster and the results regarding the relationship between SES and returning to the same employment status after the disaster. The third part shows the results regarding the different impacts of disaster-caused unemployment on the individual incomes of people with different SES. The results of the first two parts are summarized in the three models in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults of Disaster-caused Unemployment, Reemployment and Return to the Same Employment Status\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"4\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eModel 1\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eModel 2\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e\u003cp\u003eModel 3\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eDisaster-caused unemployment\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eReemployed in 2014\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003eBack in the same employment in 2014\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.148*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.354*\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.269*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(0.064)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(0.153)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.130)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEmployment status in 2011 (Ref: Regular)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIrregular\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.724***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.025\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.718***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(0.201)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(0.523)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.491)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSelf-employment\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.534\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.569\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.136\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(0.598)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(1.009)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.999)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale (Ref: Female)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.126\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.956\u0026dagger;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e1.069*\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(0.203)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(0.539)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.479)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.018\u0026dagger;\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.026\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.017\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(0.010)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(0.024)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.022)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIndustry in 2011 (Ref: Primary)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSecondary\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e/\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.009\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e/\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(0.580)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTertiary\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.048\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e/\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.292\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(0.210)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.459)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTohoku region (Ref: Other regions)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.835***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-0.496\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-0.041\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(0.186)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(0.467)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(0.412)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConstant\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-2.329***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e1.011\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e-1.040\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(0.477)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(1.215)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e(1.004)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObservations\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3,820\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e143\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e143\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eNagelkerke R\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.060\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.123\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e0.178\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAIC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1171.386\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e151.125\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e185.520\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eBIC\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e1221.318\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e174.828\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e\u003cp\u003e209.223\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e\u003cp\u003eStandard errors are in parentheses.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"4\" nameend=\"c4\" namest=\"c1\"\u003e\u003cp\u003e*** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, * p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, \u0026dagger; p\u0026thinsp;\u0026lt;\u0026thinsp;0.1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eModel 1 in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows the results regarding disaster-caused unemployment. For education level, the coefficient of the result is significantly negative, meaning that more educated people had a lower probability of losing their jobs due to the disaster. Regarding employment status, only irregular employment before the disaster has a positive and significant relation to unemployment in the disaster, while self-employment does not. As the reference category is regular employment, this result means that compared with people who were regularly employed, irregularly employed people were more likely to lose their jobs in the disaster. Since the results regarding both education and employment status show a similar tendency of people being unemployed, it can be concluded that people with lower SESs are more likely to lose their jobs in disasters. The results provide empirical evidence supporting Hypothesis \u003cspan refid=\"FPar1\" class=\"InternalRef\"\u003e1\u003c/span\u003e.\u003c/p\u003e\u003cp\u003eThe variables of age and education are significantly and positively associated with disaster-caused unemployment, meaning that people who were older and had higher education were less likely to lose their jobs in the disaster. However, the variable of the Tohoku region is positively and significantly associated with disaster-caused unemployment, suggesting that people who lived in that region were more likely to lose their jobs in the disaster. Regarding the result for industry in 2011, as the sample size of people who worked in the secondary industry is relatively small in the dataset, there were no people in it who became unemployed because of the disaster. Accordingly, there is no coefficient in the category of secondary industry.\u003c/p\u003e\u003cp\u003eModel 2 in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows the results regarding reemployment. The results regarding education and employment status are mixed. Education shows a positive and significant result, meaning that people who had a higher level of education were more likely to be reemployed after the disaster. Regarding employment status in 2011, irregular employment has no significance for being reemployed, meaning that both regularly employed and irregularly employed people had the same probability of being reemployed after the disaster. Model 3 in Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows the results regarding returning to the same employment status after the disaster. The results regarding employment status and education are opposite. Education shows a positive and significant coefficient, meaning that people with a higher level of education were more likely to return to the same employment status after the disaster. However, for employment status, irregular employment is significantly and positively associated with returning to the same employment status, meaning that people who were regularly employed before the disaster were more likely to find jobs with the same employment status before the disaster. Based on the results in Model 2 and 3, it is hard to determine whether Hypothesis 2a or 2b is supported. For the variable of education level, the results are consistent in both Model 2 and 3, indicating that people with lower level of education have lower possibility to be reemployed and return to some employment status. These results supported Hypothesis 2a. However, for the variable of employment status, the result in Model 3 indicated that people regularly employed are less likely to return to the same employment status, which supported Hypothesis 2b.\u003c/p\u003e\u003cp\u003eThe results of the third part of the analyses are summarized in Table\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e below. This part aims to determine which factor is ultimately related to income inequality.\u003c/p\u003e\u003cp\u003e\u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e\u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e\u003cdiv class=\"CaptionContent\"\u003e\u003cp\u003eResults for Individual Income\u003c/p\u003e\u003c/div\u003e\u003c/caption\u003e\u003ccolgroup cols=\"3\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e\u003cp\u003eModel 4\u003c/p\u003e\u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e\u003cp\u003eModel 5\u003c/p\u003e\u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003eCurrent individual income\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003eCurrent individual income\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDisaster-caused unemployment (Ref: Nonaffected)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-103.967***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-126.694***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(21.116)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(22.399)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEducation\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e32.498***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e32.360***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(1.946)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(1.945)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eEmployment status in 2011 (Ref: Regular)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIrregular\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-245.715***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-247.514***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(10.054)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(10.068)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSelf-employment\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-221.963***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-221.352***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(18.839)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(19.011)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003e\u003cb\u003eInteraction term\u003c/b\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDisaster-caused unemployment \u0026times;education\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-0.819\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(10.286)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDisaster-caused unemployment \u0026times;irregular\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e113.656**\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(40.872)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eDisaster-caused unemployment \u0026times;self-employment\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e64.387\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(139.611)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eMale (Ref: Female)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e177.863***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e177.422***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(21.116)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(9.082)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAge\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e9.007***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e8.980***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(0.433)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(0.433)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eIndustry in 2011 (Ref: Primary)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eSecondary\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e11.721\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e11.986\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(48.308)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(48.284)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTertiary\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e33.511***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e33.756***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(8.770)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(8.763)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eTohoku region (Ref: Other regions)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e-66.599***\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-66.238***\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(10.067)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(10.059)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eConstant\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e58.516\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e-7.876\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e(17.667)\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e(17.293)\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eObservations\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e3,820\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e3,820\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e\u003cp\u003eAdjusted R\u003csup\u003e2\u003c/sup\u003e\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e\u003cp\u003e0.498\u003c/p\u003e\u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e\u003cp\u003e0.499\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e\u003cp\u003eStandard errors are in parentheses.\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colspan=\"3\" nameend=\"c3\" namest=\"c1\"\u003e\u003cp\u003e*** p\u0026thinsp;\u0026lt;\u0026thinsp;0.001, ** p\u0026thinsp;\u0026lt;\u0026thinsp;0.01, * p\u0026thinsp;\u0026lt;\u0026thinsp;0.05, \u0026dagger; p\u0026thinsp;\u0026lt;\u0026thinsp;0.1\u003c/p\u003e\u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/colgroup\u003e\u003c/table\u003e\u003c/div\u003e\u003c/p\u003e\u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab3\" class=\"InternalRef\"\u003e3\u003c/span\u003e has two models: Model 4 presents the results of the interaction term between disaster-caused unemployment and education, and Model 5 reveals the results of the interaction term between disaster-caused unemployment and employment status in 2011 to provide a clearer demonstration of the different impacts of the disaster on the individual incomes of people with different SES.\u003c/p\u003e\u003cp\u003eIn Model 4, the interaction term between disaster-caused unemployment and education shows no significant effect on individual income, meaning that the influences of disaster-caused unemployment do not differ among people with different education levels. In Model 5, the interaction term between disaster-caused unemployment and employment status in 2011 reveals that disaster-caused unemployment \u0026times; irregular employment is positively and significantly associated with individual income, and the main effect of disaster-caused unemployment is negatively and significantly associated with it. As the reference category of employment status in 2011 is regular employment, the impact of disaster-caused unemployment on regular employment is -126.694 (p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), while it becomes \u0026minus;\u0026thinsp;13.038 (-126.694\u0026thinsp;+\u0026thinsp;113.656) for irregular employment. This means that disaster-caused unemployment impacted individual income more for regularly employed people than for irregularly employed people before the disaster.\u003c/p\u003e\u003cp\u003eTo further clarify this difference, Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e visualize the results in Models 4 and 5, respectively.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003e\u003c/p\u003e\u003cp\u003eTo clarify the different impacts of the disaster among people with different education levels on individual income, Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e simplifies education into two categories: people whose education level was equal to or below technical or junior college are categorized as having lower education, and people whose education level is higher than these degrees are categorized as having higher education. The lefthand part of Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e shows the difference in individual income between unaffected people and unemployed people following the disaster among people who had lower education, while the righthand part shows the difference in individual income between unaffected people and unemployed people among people who had higher education. This figure suggests that there were no differences in the disaster\u0026rsquo;s impact on individual income between lower- and higher-educated people. Therefore, it provides evidence that although education caused the differences in being reemployed and returning to the same employment status, it did not ultimately lead to differences in individual income.\u003c/p\u003e\u003cp\u003eRegarding Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, the lefthand part shows the difference in individual income between unaffected people and unemployed people following the disaster among those who were regularly employed before the disaster, while the righthand part shows the difference in individual income between unaffected people and unemployed people among those who were irregularly employed before the disaster. The difference in individual income between unaffected people and unemployed people was much larger among regularly employed people than irregularly employed people. In fact, the 95% confidence interval in the irregular employment part overlaps, meaning that there may have been no significant difference in income between unaffected people and unemployed people among those with irregular employment before the disaster, while there was a significant difference for those with regular employment. This figure reveals that disaster-caused unemployment impacted individual income more for regularly employed people than for irregularly employed people before the disaster.\u003c/p\u003e\u003cp\u003eThe results of these two interactions suggest that the difference in individual income may be caused by the gap regarding the return to the same employment status between those with regular and irregular employment. As regularly employed people have a lower probability of returning to the same employment status after a disaster, their income decreases more, while irregularly employed people have a higher probability, so their income decreases less.\u003c/p\u003e"},{"header":"5 Discussion","content":"\u003cp\u003eAccording to previous research, disasters can exacerbate income inequality through changes in employment patterns. Our findings support this theory. During the disaster response phase, individuals with lower socioeconomic status (SES), characterized by lower education levels and irregular employment, were more likely to experience job loss. This suggests that income inequality can widen as a result of these differential employment outcomes between individuals with higher and lower SES. These findings align with the mechanisms of employment flexibility and health-related vulnerabilities, and they confirm the accuracy of social vulnerability theory in explaining how individuals with lower SES are more susceptible to the negative impacts of disasters.\u003c/p\u003e\u003cp\u003eHowever, the results for the post-disaster recovery phase were mixed. The two measures of SES - education level and employment status - yielded different outcomes, necessitating further discussion. Regarding education, the findings indicated that individuals with lower levels of education were less likely to be re-employed and return to their previous employment status. These results align with capital theory. However, concerning employment status, the results showed no significant difference in re-employment rates between regularly and irregularly employed individuals after the disaster. Moreover, irregularly employed individuals even had a slight advantage in returning to their previous employment status compared to regularly employed individuals. These findings are consistent with the mechanism of employment flexibility.\u003c/p\u003e\u003cp\u003eTo reconcile these divergent findings and determine their impact on individual income, we analyzed income data. The analysis revealed no significant change in the income gap between lower- and higher-educated individuals affected by the disaster. In contrast, it showed a reduction in the income difference between regularly and irregularly employed individuals affected by the disaster. These findings suggest that education is not the most crucial factor contributing to income inequality, and changes in employment status played a more significant role in shaping income disparities during the disaster.\u003c/p\u003e\u003cp\u003eHowever, based on the result of the decreasing income gap, can we say that income inequality between people with regular and irregular employment was reduced by the disaster? The answer is no. This is because individuals who originally had high SES (regular employment) dropped to lower SES, while those who originally had low SES (irregular employment) experienced no significant change after the disaster. As a whole, the number of people with high SES decreased, and the number of people with low SES increased; thus, societal inequality expanded. This demonstrates how status changes, rather than a poverty trap, can exacerbate inequality after a disaster.\u003c/p\u003e\u003cp\u003eFurthermore, as mentioned in the literature, organizations tend to hire people with lower costs (Hippel et al. \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1997\u003c/span\u003e). Irregular employment costs less than regular employment and is thus favored by organizations. Since the specific human assets obtained from previous regular employment are useless, to cut costs, employers are less likely to hire people who were regularly employed for regular employment. Thus, people who were regularly employed before a disaster are less likely to return to the same employment status, while irregularly employed people are more likely to be reemployed in irregular employment after the disaster. This is why in the results, people with irregular employment had a higher probability of returning to the same employment status than people with regular employment.\u003c/p\u003e\u003cp\u003eEducation is different. Education level represents general rather than specific human capital or assets in both work and study (Becker, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e1964\u003c/span\u003e; Schultz, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e1961\u003c/span\u003e). Therefore, individuals with higher levels of education are likely to possess greater work abilities, while those with lower levels of education are likely to have lower abilities. In labor markets, organizations tend to prefer hiring individuals with higher levels of education due to these ability differences (Bloch \u0026amp; Smith, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e1977\u003c/span\u003e). Consequently, individuals with higher levels of education had a higher probability of being re-employed after the disaster compared to those with lower levels of education.\u003c/p\u003e\u003cp\u003eMoreover, individuals with higher levels of education are more likely to be employed in higher-status jobs, while those with lower levels of education are more likely to be employed in lower-status jobs (Esteban-Pretel \u0026amp; Fujimoto, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). Thus, during the re-employment process, individuals with higher levels of education are more likely to secure higher-status employment, even if they were previously employed in lower-status positions. Conversely, individuals with lower levels of education are more likely to be employed in lower-status positions, even if they were previously employed in higher-status positions. This explains why, in the results, individuals with higher levels of education had a higher probability of returning to the same or a higher employment status compared to those with lower levels of education after the disaster.\u003c/p\u003e\u003cp\u003eBased on these research findings, the policy implications for Japan include enhancing post-disaster support for lower socioeconomic status (SES) groups, particularly through job retraining, social security, and job stability measures. Additionally, implementing flexible labor market policies can facilitate quicker employment recovery. Moreover, policies should ensure that higher SES groups are not disproportionately hindered in their post-disaster employment recovery. Globally, these results suggest that other disaster-prone countries should prioritize protecting socially vulnerable groups by adopting comprehensive social safety nets, vocational training, and flexible employment policies to mitigate the exacerbation of income inequality during disaster recovery.\u003c/p\u003e\u003cp\u003eAlthough this study makes several contributions to the research on disasters and income inequality, it has limitations. First, there are still disadvantages in measuring SES. Generally, occupation remains the most efficient way to do so. However, the dataset used in this study contained no questions concerning respondents' occupation. Therefore, we were forced to use employment status instead of occupation to measure SES. Should scholars seek to expand the application of the results obtained in this study, data on occupation are needed. Therefore, in future research, we suggest researchers collect data with measurements of occupation and confirm the results obtained in this study.\u003c/p\u003e\u003cp\u003eSecond, the sample size of those with disaster-caused unemployment is small. The data include only 143 samples for analysis, which is not sufficient to accurately estimate the coefficient of the regression. Even if the analyses were broadened to the full sample size of the data, which is 10,466, there are only 154 cases of disaster-caused unemployment, meaning that only 1.5% of people in the sample lost their jobs in the 3.11 earthquake. To check the representativeness of these data and numbers, we also calculated the disaster-caused unemployment rate in the 3.11 earthquake using the data of the 2012 Employment Status Basic Survey (Shugyo Kozo Kihon Chosa), one of the censuses in Japan (Statistics Bureau of Japan, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). These data show that in 2012, the population over age 15 was approximately 110\u0026nbsp;million, and the cases of disaster-caused unemployment in the 3.11 earthquake amounted to 2.3\u0026nbsp;million. Thus, the disaster-caused unemployment rate across all of Japan was 2%, which is very close to the rate in the data utilized by this study. Therefore, it can be said that the disaster-caused unemployment cases and results in this study accurately represent the broader universe of cases.\u003c/p\u003e\u003cp\u003eThird, the income difference between people with regular and irregular employment in the short term after the disaster was not directly investigated. This study estimated only the difference in disaster-caused unemployment between people with regular and irregular employment and used this point to speculate about the income difference rather than directly demonstrating the effect. This is because the survey did not contain a question regarding individual income in the short term after the 3.11 triple disasters. However, as unemployment is closely related to decreasing income, especially individual income, the estimation of employment can be considered a reliable proxy for income changes.\u003c/p\u003e"},{"header":"6 Conclusion","content":"\u003cp\u003eThis study utilized survey data collected after the 3.11 triple disasters in Japan to dynamically demonstrate the changes in employment caused by the disaster and its consequences on income inequality between people with higher and lower socioeconomic status (SES) from both the disaster response and disaster recovery phases. The analyses yielded four main results.\u003c/p\u003e\n\u003cp\u003eFirst, it found that in the disaster response phase, individuals with lower SES, characterized by lower education and irregular employment, were more likely to face job loss than those with higher SES. Second, the results regarding post-disaster reemployment were mixed. Employment status showed no significant effect on reemployment, while education presented a positive effect. Third, the results on the relationship between SES and returning to the same employment status were also mixed. Employment status showed a negative effect on returning to the same employment status, while education presented a positive effect. Finally, the results on individual income showed that the income decrease caused by the disaster was greater among individuals who were regularly employed than those who were irregularly employed before the disaster.\u003c/p\u003e\n\u003cp\u003eIn summary, the disaster indeed amplified income inequality; however, the mechanism was not solely based on social vulnerability but also involved employment flexibility. Accordingly, for effective post-disaster social governance, policymakers should not only target recovery policies at people with low socioeconomic status but also consider the needs of wealthier individuals.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding Declaration\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis paper is supported by National Social Science Fund of China (Grant No.:22CSH095).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics Approval Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study involved secondary analysis of publicly available data provided by the Social Science Japan Data Archive (SSJDA), under dataset ID 1036 (\u0026quot;Survey on Work and Hope after the Great East Japan Earthquake, 2014\u0026quot;). The original survey was designed and supervised by Professor Yuji Genda (Institute of Social Science, University of Tokyo) and implemented by INTAGE Research Inc., in accordance with established ethical guidelines for social science research.\u003c/p\u003e\n\u003cp\u003eAccess to this dataset was officially granted by SSJDA. The SSJDA requires researchers to submit a formal application for dataset use, and approval of this application serves as an equivalent process to Institutional Review Board (IRB) review, ensuring compliance with ethical principles such as confidentiality of participants and informed consent. All research activities based on this dataset were conducted in accordance with relevant regulations and ethical standards, including the principles of the Declaration of Helsinki.\u003c/p\u003e\n\u003cul type=\"disc\"\u003e\n \u003cli\u003e\u003cstrong\u003eApproval Body\u003c/strong\u003e: Social Science Japan Data Archive (SSJDA), Institute of Social Science, University of Tokyo\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eApproval Number / ID\u003c/strong\u003e: SSJDA Dataset ID 1036\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDataset DOI\u003c/strong\u003e: 10.34500/SSJDA.1036\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eDate of Approval (Dataset Release Date)\u003c/strong\u003e: March 10, 2016\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eScope of Approval\u003c/strong\u003e: Secondary analysis of anonymized, publicly available survey data from the 2014 \u0026ldquo;Survey on Work and Hope after the Great East Japan Earthquake.\u0026rdquo; This approval covers the use of data solely for academic research purposes, with participant confidentiality strictly protected.\u003c/li\u003e\n \u003cli\u003e\u003cstrong\u003eSupporting Information\u003c/strong\u003e: Details on dataset access procedures and ethical compliance are available on the SSJDA official website (https://csrda.iss.u-tokyo.ac.jp/).\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003e\u003cstrong\u003eStatement on Informed Consent\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe data utilized in this study were obtained from the Social Science Japan Data Archive (SSJDA), specifically dataset ID 1036 (\u0026quot;Survey on Work and Hope after the Great East Japan Earthquake, 2014\u0026quot;). According to the original data collection documentation, informed consent was duly obtained from all individual participants included in the original survey conducted by Professor Yuji Genda of the Institute of Social Science, University of Tokyo, and administered by INTAGE Research Inc. Participants explicitly agreed to participate after being fully informed about the purpose, confidentiality measures, and voluntary nature of their involvement.\u003c/p\u003e\n\u003cp\u003eAll procedures followed during data collection complied with ethical standards ensuring participants\u0026apos; consent and anonymity.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eAldrich, D. P. 2012. Building Resilience: Social Capital in Post-Disaster Recovery. Chicago: University of Chicago Press.\u003c/li\u003e\n\u003cli\u003eAldrich, D. P. 2019. Black Wave: How Networks and Governance Shaped Japan\u0026rsquo;s 3/11 Disasters. Chicago: University of Chicago Press.\u003c/li\u003e\n\u003cli\u003eAmuedo-Dorantes, Catalina and Ricardo Serrano-Padial. 2010. \u0026ldquo;Labor Market Flexibility and Poverty Dynamics.\u0026rdquo; Labour Economics 17: 632\u0026ndash;642.\u003c/li\u003e\n\u003cli\u003eBecker, G. S. 1964. Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education. Chicago: The University of Chicago Press.\u003c/li\u003e\n\u003cli\u003eBloch, F. E., and S. P. Smith. 1977. Human Capital and Labor Market Employment. Journal of Human Resources 12(4): 550\u0026ndash;560.\u003c/li\u003e\n\u003cli\u003eEsteban-Pretel, J., and J. Fujimoto. 2020. Non-regular Employment over the Life-cycle: Worker Flow Analysis for Japan. Journal of The Japanese and International Economies 57: 1\u0026ndash;20.\u003c/li\u003e\n\u003cli\u003eFang, L., J. Wu, and T. Miljkovic. 2017. Modeling Impact of Natural Hazard-Induced Disasters on Income Distribution in the United States. International Journal of Disaster Risk Science 8(4): 435\u0026ndash;444.\u003c/li\u003e\n\u003cli\u003eFothergill, A., and L. A. Peek. 2004. Poverty and Disasters in the United States: A Review of Recent Sociological Findings. Natural Hazards 32: 89\u0026ndash;110.\u003c/li\u003e\n\u003cli\u003eFrahm, K. A., P. J. Gardner, L. M. Brown, D. P. Rogoff, and A. Troutman. Community-Based Disaster Coalition Training. Journal of Public Health Management and Practice 20(5): S111\u0026ndash;S117.\u003c/li\u003e\n\u003cli\u003eG\u0026oacute;mez, P. G., and A. L. Nicol\u0026aacute;s. 2006. Health Shocks, Employment and Income in the Spanish Labour Market. Health Economics 15 (9): 997\u0026ndash;1009.\u003c/li\u003e\n\u003cli\u003eGordon, Andrew. 2017. New and Enduring Dual Structures of Employment in Japan: The Rise of Non-Regular Labor, 1980s\u0026ndash;2010s. Social Science Japan Journal 20(1): 9\u0026ndash;36.\u003c/li\u003e\n\u003cli\u003eHallegatte, St\u0026eacute;phane, Adrien Vogt-Schilb, Julie Rozenberg, Mook Bangalore and Chlo\u0026eacute; Beaudet. 2020. \u0026ldquo;From Poverty to Disaster and Back: A Review of the Literature.\u0026rdquo; Economics of Disasters and Climate Change 4:223\u0026ndash;247.\u003c/li\u003e\n\u003cli\u003eHippel, C., S. L. Mangum, D. B. Greenberger, R. L. Heneman, and J. D. Skoglind. 1997. Temporary Employment: Can Organizations and Employees Both Win? Academy of Management Executive 11(1): 93\u0026ndash;104.\u003c/li\u003e\n\u003cli\u003eHowell, J., and J. R. Elliott. 2019. Damages Done: The Longitudinal Impacts of Natural Hazards on Wealth Inequality in the United States. Social Problems 072018(0): 1\u0026ndash;20.\u003c/li\u003e\n\u003cli\u003eKeerthiratne, S., and R. S. J. Tol. 2018. Impact of natural disasters on income inequality in Sri Lanka. World Development 105: 217\u0026ndash;230.\u003c/li\u003e\n\u003cli\u003eKeizer, Arjan. 2008. Non-regular Employment in Japan: Continued and Renewed Dualities. Work, Employment and Society 22(3): 407\u0026ndash;425.\u003c/li\u003e\n\u003cli\u003eLin, N. 2001. Social Capital: A Theory of Social Structure and Action. England: Cambridge University Press.\u003c/li\u003e\n\u003cli\u003eLin, N., K. Cook, and R. S. Burt. 2001. Social Capital: Theory and Research. New York: Editors Press.\u003c/li\u003e\n\u003cli\u003eMetaxa-Kakavouli, D., P. Maas, and D. P. Aldrich. 2018. How Social Ties Influence Evacuation Behavior. Proceedings of the ACM on Human-Computer Interaction 2(CSCW):1\u0026ndash;16\u003c/li\u003e\n\u003cli\u003eMiljkovic, T., and D. Miljkovic. 2014. Modeling Impact of Hurricane Damages on Income Distribution in the Coastal U.S. International Journal of Disaster Risk Science 5: 265\u0026ndash;273.\u003c/li\u003e\n\u003cli\u003eMouw, T. 2003. Social Capital and Finding a Job: Do Contacts Matter? American Sociological Review 68(6): 868\u0026ndash;898.\u003c/li\u003e\n\u003cli\u003ePaton, D. 2003. Disaster Preparedness: A Social-cognitive Perspective. Disaster Prevention and Management 12(3): 210\u0026ndash;216.\u003c/li\u003e\n\u003cli\u003eReid, M. 2013. Disasters and Social Inequalities. Sociological Compass 7(11): 984\u0026ndash;997.\u003c/li\u003e\n\u003cli\u003eSato, Y. 2010. Stability and Increasing Fluidity in the Contemporary Japanese Social Stratification System. Contemporary Japan 22: 7\u0026ndash;21.\u003c/li\u003e\n\u003cli\u003eSato, Y. and J. Iami. 2011. Japan\u0026rsquo;s New Inequality: Intersection of Employment Reforms and Welfare Arrangements. Tokyo: Trans Pacific Press.\u003c/li\u003e\n\u003cli\u003eSchultz, T. W. 1961. Investment in Human Capital. American Economic Review 51(1): 1\u0026ndash;17.\u003c/li\u003e\n\u003cli\u003eShavers, V. L. 2007. Measurement of Socioeconomic Status in Health Disparities Research. Journal of The National Medical Association 99(9): 1013\u0026ndash;1023.\u003c/li\u003e\n\u003cli\u003eStatistics Bureau of Japan. 2012. 2012 Employment Status Basic Survey (Shugyo Kozo Kihon Chosa). https://www.stat.go.jp/data/shugyou/2012/index2.html.\u003c/li\u003e\n\u003cli\u003eTeo, M., A. Goonetilleke, A. Ahankoob, K. Deilami, and M Lawie. 2018. Disaster Awareness and Information Seeking Behaviour among Residents from Low Socio-economic Backgrounds. International Journal of Disaster Risk Reduction 31: 1121\u0026ndash;1131.\u003c/li\u003e\n\u003cli\u003eTierney, K. J. 2007. Businesses and Disasters: Vulnerability, Impact, and Recovery. in Handbook of Disaster Research, ed. H. Rodr\u0026iacute;guez, E. L. Quarantelli, and R. R. Dynes, 275\u0026ndash;297. New York: Springer.\u003c/li\u003e\n\u003cli\u003eWeathers, Charles. 2009. Nonregular Workers and Inequality in Japan. Social Science Japan Journal 12(1): 143\u0026ndash;148.\u003c/li\u003e\n\u003cli\u003eYamamura, E. 2015. The Impact of Natural Disasters on Income Inequality: Analysis using Panel Data during the Period 1970 to 2004. International Economic Journal 29(3): 359\u0026ndash;374.\u003c/li\u003e\n\u003cli\u003eZottarelli, L. K. 2008. Post-Hurricane Katrina Employment Recovery: The Interaction of Race and Place. Social Science Quarterly 89(3): 593\u0026ndash;607.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"humanities-and-social-sciences-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"palcomms","sideBox":"Learn more about [Humanities \u0026 Social Sciences Communications](http://www.nature.com/palcomms/)","snPcode":"41599","submissionUrl":"https://submission.springernature.com/new-submission/41599/3","title":"Humanities and Social Sciences Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Employment change, employment flexibility, income inequality, social vulnerability, socioeconomic status, earthquake, natural hazard","lastPublishedDoi":"10.21203/rs.3.rs-7161036/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7161036/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThis study investigates the impact of the 2011 Tohoku earthquake on income inequality in Japan. Using a quantitative analysis of survey responses from nearly 4,000 respondents, we find that individuals with lower socioeconomic status were disproportionately affected by the disaster. During the immediate aftermath, those with lower education and precarious employment experienced significantly higher rates of job loss. In the recovery phase, education emerged as a key factor influencing re-employment, while the impact of prior employment status was less clear. Furthermore, the disaster led to a significant decline in income for regularly employed individuals compared to those with irregular employment. This suggests that the disaster not only affected employment opportunities but also had a significant impact on income distribution. Our findings underscore the complex mechanisms through which natural disasters can exacerbate income inequality. Beyond social vulnerability, the interplay of factors such as education, employment status, and income sources shapes individuals' experiences and outcomes in the aftermath of such events.\u003c/p\u003e","manuscriptTitle":"Vulnerability or Flexibility: An Examination of Socioeconomic Status, Employment Changes, and Individual Income in Japan’s 3.11 Earthquake","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-10 14:13:31","doi":"10.21203/rs.3.rs-7161036/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewersInvited","content":"","date":"2026-02-07T17:47:09+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-10-24T18:17:25+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-12T06:34:41+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-09-09T05:33:27+00:00","index":"","fulltext":""},{"type":"submitted","content":"Humanities and Social Sciences Communications","date":"2025-09-09T01:33:57+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"humanities-and-social-sciences-communications","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"palcomms","sideBox":"Learn more about [Humanities \u0026 Social Sciences Communications](http://www.nature.com/palcomms/)","snPcode":"41599","submissionUrl":"https://submission.springernature.com/new-submission/41599/3","title":"Humanities and Social Sciences Communications","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Nature AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"b9c55cce-a523-42f1-837a-b62de9a39bb0","owner":[],"postedDate":"September 10th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":54514172,"name":"Earth and environmental sciences/Environmental social sciences"},{"id":54514173,"name":"Scientific community and society/Geography"},{"id":54514174,"name":"Social science/Geography"},{"id":54514175,"name":"Earth and environmental sciences/Natural hazards"}],"tags":[],"updatedAt":"2026-02-07T17:53:55+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-10 14:13:31","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7161036","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7161036","identity":"rs-7161036","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}