Public Support for Health Coverage for Migrant Populations in The COVID-19 Pandemic: The Roles of Infection Externalities and Impressions of Migrants' Home Place

Background: International population mobility is a signicant challenge for the management of the coronavirus (COVID-19) pandemic. The suspension of ights and other trac has been adopted to avoid more cases of imported infection by many countries, but these measures mean that most migrant populations, including immigrants, migrant and seasonal workers, international students, and others, have to stay in their host places. Therefore, equitable access to health services in the host places is crucial for them. Immigrant-receiving areas may encounter a dilemma when considering whether to extend health coverage of COVID-19 testing and treatment to immigrants. In this context, it is vital to understand public attitudes towards this issue since they explain and validate the policy responses. Methods: The data used in this study came from an survey conducted in March 2020. The survey targeted adults aged 18 and older who were local residents of nine cities in Guangdong province, China. The sample size was 1,040, and STATA 15.0 was used in the statistical analysis. Results: The results show that individuals with higher health security ratings, lower demand for health services, and greater concern about the risk of infectious diseases are more supportive of extending health services to immigrants. In addition, individuals who rate immigrants’ home place as more immigrant-friendly are also more supportive of it. Conclusions: Public support for health for by infection externalities and impression of immigrants’ home countries. This study attempts to contribute to the body of literature related to welfare attitudes and immigration. It also outlines a series of crucial implications for the global task of managing COVID-19.


Background
International population mobility is a signi cant challenge for the management of the coronavirus (COVID-19) pandemic, an ongoing public health crisis spreading around the world. The suspension of ights and other tra c has been adopted as an important measure to avoid more cases of imported infection by many countries during the pandemic, but these measures mean that most migrant populations, including migrates, migrant and seasonal workers, international students, and others, have to stay in their host nations. Therefore, equitable access to health and social services in their migrantreceiving areas is crucial, and the World Health Organization speci cally issued guidance to Member States and partners calling for the inclusion of migrants and the removal of any excessive out-of-pocket payments for COVID-19-related health services. 1 However, host nations may encounter a dilemma when considering their policy responses. On the one hand, these changes would allow migrant populations to bene t from limited and valuable medical resources, even free medical services, designed for local residents. On the other hand, the exclusion of migrant populations from the welfare system and public health services, such as free testing, diagnostics, care, and treatment, would increase the risk of outbreaks in these populations [1] and present an additional threat to local public health due to the externalities of the pandemic [2][3][4]. In this context, it is vital to understand public attitudes towards migrant populations' access to public health services and welfare bene ts, as these attitudes explain policy responses [5][6][7][8].
The current study was conducted in China, where the COVID-19 pandemic rst emerged. To atten the curve of new cases over time, the Chinese government has taken several decisive actions since the end of January 2020. One of the most effective and dependable actions was the provision of free testing for local residents and the removal of excessive out-of-pocket payments for treatment. Treatment for COVID-19 costs an average of 17,000 yuan ($2,400) per patient, and 65% of the cost is covered by basic medical insurance [9]. In addition, in early February, China announced that the other 35% of COVID-19 patients' medical costs would be subsidized by local governments. In other words, local residents would be offered "free treatment" for the disease.
One of the most controversial topics in China is whether migrates should be given the same free treatment as local residents. The number of foreign immigrants in China is relatively small due to the country's strict immigration control policy [10]. Yet, there is a signi cant number of Hong Kong residents living in Mainland China, especially in Guangdong Province, which borders Hong Kong. Based on the special institutional arrangement of "one country, two systems," Hong Kong's nancial system and social security system are independent of those in Mainland China. Therefore, Hong Kong residents are usually  Table 1) [11].
During the current outbreak of COVID-19, some Hong Kong residents were also infected with the virus, and the debate about whether they should get the free treatment has aroused widespread public attention. Given this context, this study will investigate Guangdong residents' attitudes towards Hong Kong people' access to local public health services and welfare bene ts. Self-interest hypothesis is one of the mainstreaming theories that explain an individual's social policy preference [12][13][14]. According to self-interest hypothesis, local residents who are healthy or at a relatively low risk of communicable disease exposure will be more likely to support health coverage for infectious diseases for migrates, since they will have less need for health care than those who are in poorer health or at a high risk of infectious disease exposure. In addition, the exclusion of migrant populations from public health services, such as free testing, diagnostics, care, and treatment, would increase the risk of outbreaks in these populations [1] and present an additional threat to local public health due to the externalities of the pandemic. The negative externalities of infectious diseases may further prompt healthy local residents to support health coverage for infectious diseases for migrates.
After considering the externalities of COVID-19, we derived the following hypothesis: Hypothesis I Individuals with higher health security ratings, a lower demand for health services, and greater concern about the risk of infectious diseases are more supportive of extending coverage for COVID-19 testing and treatment to migrates.

Impressions of migrates' home country or place and attitudes towards extending health services to migrates
The politics of social policy on immigration are also in uenced by the reciprocal relationship between home country and host country [15]. The friendliness/inclusiveness of public policy in migrates' home country or place exerts a feedback effect on the policy preferences of the host country regarding the welfare eligibility of migrates [16,17]. Thus, the impressions of migrates' home country or place may in uence local residents' attitudes towards extending health services to migrates.
From this line of reasoning, this study formulated the second hypothesis: Hypothesis II Individuals who rate migrates' home country or place as more inclusive and friendly are more supportive of extending health coverage of COVID-19 testing and treatment to migrates.

Data collection
The data used in this study came from an online survey conducted in March 2020. The survey targeted adults aged 18 and older who were local residents of nine cities in Guangdong province. These cities included Guangzhou, Shenzhen, Zhuhai, Dongguan, Foshan, Huizhou, Zhongshan, Jiangmen, and Zhaoqing, all of which are part of the Guangdong-Hong Kong-Macao Greater Bay Area. To ensure that the survey sample was a representative sample of the nine cities, we included respondents from all regions and of different demographic characteristics. We also employed weighting to match the characteristics of the survey sample with the characteristics of the general population. Weighting was calculated using data from the Guangdong Statistical Yearbook. The survey was approved by the Sub-Committee on Research Ethics and Safety of the Research Committee of author's a liation. The sample size was 1,040, and STATA 15.0 was used in the statistical analysis.

Dependent variable
The survey included a question on public support for migrant healthcare coverage, which was the dependent variable in this study. The survey asked, "If a resident from Hong Kong is infected with COVID-19 in your city, do you support offering them the same free treatment that locals enjoy?" The options offered were "Yes," "No," and "Don't know." Responses were collected using a two-point scale where "Yes = 1" and "No or Don't know = 0," since this study used logistic regression analysis to test the research hypotheses.

Explanatory variables
To test the rst research hypothesis, respondents' subjective health statuses were collected using a vepoint Likert-type scale ranging from "1 = very bad" to "5 = very good." The second explanatory variable was related to respondents' perceptions of the risk of the local pandemic. The question read, "How do you evaluate the trend of the COVID-19 pandemic in your city?" Responses were collected using an 11-point scale ranging from "0 = very pessimistic" to "10 = very optimistic." To test the second research hypothesis, this survey asked respondents to evaluate the performance of the Hong Kong government. The question read, "How do you evaluate the Hong Kong governments' performance in epidemic prevention and control?" The responses were measured using an 11-point scale ranging from "0 = very bad" to "10 = very good." Another explanatory variable was related to participants' overall impressions of Hong Kong. The survey invited respondents to indicate to what extent they agreed with the following statements: "Hong Kong is an inclusive city," "Hong Kong is an open city," and "Hong Kong is a friendly city." Responses were collected using a ve-point Likert-type scale ranging from "1 = strongly disagree" to "5 = strongly agree."

Control variable
The survey also collected respondents' individual-level characteristics as control variables in the logistic regression analysis. The characteristics were as follows: age group (

Descriptive statistics
The results demonstrated that 63.48% of respondents supported the provision of free treatment to Hong Kong people infected with COVID-19. Approximately half of respondents thought their health statuses were "very good" (10.38%) or "good" (44.13%). The survey revealed that most respondents were optimistic about the trend of the local pandemic, creating a curve that was skewed right of the distribution of their responses (mean = 9.508, SD = 1.428). As mentioned above, this survey also asked participants to evaluate their overall impressions of Hong Kong. Only 16.16%, 29.81%, and 30.96% of the participants agreed (or strongly agreed) that "Hong Kong is an inclusive city," "Hong Kong is an open city," and "Hong Kong is a friendly city," respectively.
The descriptive statistics for the key variables used in this study are outlined in Table 2.

Multiple and logistic regression analyses
To test hypothesis I, the set of health security risk variables was added to Model 1, controlling for age, gender, educational attainment, and occupation. The dependent variable speci ed in the third section was regressed, and the results of this logistic regression analysis are presented in Table 3. The ndings demonstrated that subjective health status was positively associated with support for migrant health coverage. Speci cally, individuals who perceived themselves to be in good health tended to support the provision of free COVID-19 treatment to Hong Kong people, while those who perceived themselves to be in bad health tended to oppose it. In addition, Model 1 also revealed that the degree of optimism about the local pandemic was positively associated with support for migrant healthcare coverage. It implies that individuals who were optimistic about the local pandemic tended to support sharing medical resources with migrants. Therefore, Model 1 supports hypothesis I.
To test research hypothesis II, the set variables about impressions of migrates' home country were added to Model 2. Similarly, Model 2 controlled for age, gender, educational attainment, and occupation. As seen in Table 3, respondents' evaluations of the Hong Kong government's performance were positively associated with support for the provision of free COVID-19 treatment to Hong Kong people. Individuals' overall impressions of Hong Kong also in uenced their attitudes towards this issue. Individuals who considered Hong Kong to be an inclusive or friendly city tended to support the sharing of medical resources with Hong Kong people in their cities. Thus, Model 2 supports hypothesis II.
The dependent variable was regressed on two sets of explanatory variables together in Model 3 (see Table 3). It turned out to be interesting: On the one hand, the effect of health security risk variables on the support for extending health services to Hong Kong people is no longer signi cant, under control over individual impressions of Hong Kong; On the other hand, however, the positive correlation between the support for health coverage for Hong Kong people and individual impressions of Hong Kong is still signi cant, when controlling for health security risk variables. Research hypothesis II received stronger empirical con rmation in this study, with better impressions of a place leading to more supportive of extending health coverage for the migrant populations from that place. Notes: *p < 0.05, **p < 0.01

Discussion
The paper is one of the rst attempts to quantify the drivers of public support for health coverage for migrates in less developed countries during the COVID-19 pandemic. To address the challenge of the pandemic, many countries provide free diagnostic and treatment services to their citizens. However, migrant populations, including migrates, migrant and seasonal workers, international students, and others, are usually excluded from this policy arrangement due to lack of citizenship or coverage by social health insurance in the host nations. If migrant populations are not adequately protected against the nancial consequences of COVID-19-related health expenses, the risk of outbreaks in immigrant populations can increase and present an additional threat to local public health. Given these considerations, the world health organization called on Member States and partners to integrate migrant populations into their health structures and avoid excessive out-of-pocket payments for COVID-19-related health services [1].
The self-interest hypothesis argues that local residents would consider their needs prior to the wellbeing of migrant populations; therefore, the public may not support the extension of welfare services to migrate at a time of crisis when public service and welfare resources are scarce [18]. In fact, a large number of existing literature material have shown that, the public health crisis and the resulting economic crisis would generally undermined public support for international assistance [19][20][21]. However, the attitudes of local residents may be affected by the externalities of the pandemic, which complicates local residents' attitudes towards migrant populations' access to free health service under the public health crisis. Our empirical study nds that due to the negative externalities of infectious diseases, individuals with higher health security ratings, lower demand for health services, and greater concern about the risk of infectious diseases are more supportive of extending health services to migrant populations. In addition, impressions of migrates' home country or place may also alter this prediction created by the self-interest hypothesis. This study nds that the impressions of migrates' home country or place will exert a feedback effect on the welfare attitudes of citizens in the host country. Individuals who rate migrates' home country or place as inclusive and friendly are more supportive of extending health services to migrates.
According to Johns Hopkins Coronavirus Resource Center, the COVID-19 pandemic has spread to 188 countries or regions around the world, while the overall number of infections has surpassing 64 million by 3 Dec 2020 [22]. According to the COVID-19 pandemic prediction model created by the Center for Infectious Disease Research and Policy at the University of Minnesota, the epidemic is likely to last until 2022 if there is no cure for the COVID-19 and the vaccine [23].

Conclusion
Public support for health coverage for migrates in the COVID-19 Pandemic is in uenced by infection externalities and impressions of migrates' home country or place. This study provides preliminary ndings about the characteristics and causes of public support for health coverage for migrants during the pandemic. It also outlines a series of crucial implications for the global management of COVID-19.