Poverty describes a state in which humans lack access to those resources necessary for providing basic material needs ^[1][2][3][6,9,25]. Therefore, researchers have evolved a better understanding of poverty by defining more precisely what those vital-for-human-lives resources are. Baulch summarized the concepts related to poverty in a pyramid-like format ^[3][25], in which the narrowest definition of poverty (the consumption/income approach) emphasizes humans’ private consumption, which is dependent on their income. In comparison, the broadest definition of poverty (the human development approach) extends the concept of poverty to include other essentially important things, such as common property resources, public provisions, and human dignity and autonomy ^[2][4][5][5,9,26]. Other approaches to understanding poverty have also emerged. For example, by emphasizing the accessibility/exclusion of social networks vital for sustaining livelihoods ^[1][6], the social exclusion approach has offered a valuable complement to the conceptualization of poverty.

Poverty has a complex and dynamic nature. First, it exists pervasively yet contextually. Pervasively, the phenomenon of poverty exists in numerous countries, including both lower-income/higher-income and developing/developed countries. Regarding its diversification, poverty occurs in situ, alongside inequality, across population groups in different countries, regions, regimes, and societies ^[4][6][7][3,5,27]. For example, Alkire and Seth found that reducing national poverty does not necessarily mean a uniform reduction in all population groups ^[8][4]. Second, poverty interacts with other social phenomena, not limited to migration, feminist activity, urbanization, and population aging ^{[2][9][10][11][12]}[7,8,9,10,11]. Third, poverty develops dynamically. Microscopically, the deprived, rather than being passive victims waiting for handouts, are agents struggling to cope with poverty with whatever resources they possess ^[1][13][2,6], thus leading to the macroscopic poverty index of a specific population constantly being transformed ^[13][2].

The complex and dynamic nature of poverty has caused the measurement of poverty to evolve from absolute to relative and from unidimensional to multidimensional measures. All poverty measurements are founded essentially on the basis of comparing available resources to basic human needs: a person/family is identified as poor if their resources fall short of the poverty threshold ^[5][7][26,27]. However, a relative poverty threshold (a cutoff level depending on specific resource distribution) is increasingly replacing the conventional absolute threshold (a fixed cutoff level, or the so-called poverty line, being applied across all resource distributions) ^[5][26]. At the same time, the conventional unidimensional measure of poverty (i.e., private consumption/income) has been altered by the multidimensional poverty structure, which could include a variety of dimensions/indicators, including nutrition, clothing, housing, public facilitation, psychological well-being ^[4][8][4,5], etc. Accordingly, several methodologies for such multidimensional poverty measurement have been proposed, grouped broadly into axiomatic and information theory approaches, fuzzy set theories, and latent variable methods ^[14][28].

Migration is usually understood as a spatial separation between the location of a person’s/household’s place of residence and that of the place where they are engaged in activities to sustain their livelihood ^[15][29]. Migration essentially involves the mobility of labor, together with a person’s/household’s experience, skills, education level, and health status ^[12][11]. There are multiple migration types, which play various roles in reducing poverty: while poverty seems to act as a constraint on international migration (i.e., migration across national borders), it acts as a push factor for internal migration (i.e., migration within national borders, of which rural–urban migration for the purposes of working is a typical form) ^[2][15][16][9,29,30].

Regarding internal migration, the urbanization process has promoted large-scale rural–urban worker migration in many developing countries in recent decades. The ranks of rural–urban worker migration (hereinafter referred to as worker migration) have swelled, which has aroused the attention of poverty researchers in several countries ^{[17][18][19][20][21]}[12,15,17,18,19]. This type of migration contributes positively to poor people achieving secure livelihoods and constructing pathways for themselves out of poverty ^[15][29]. However, migrant workers may now constitute “new” poor communities in urban areas ^[20][22][18,31] for several reasons. First, migrant workers are in a relatively weak position in the labor market: for example, migrant workers work longer hours, with much lower hourly wages, than local residents. Moreover, significant differences in the non-income welfare earned by migrant workers and local residents exist, such as housing conditions and access to social insurance programs as a result of inequalities in the distribution of urban infrastructure (both private and public) and basic services across the length of local stay ^[20][18].

Women are more likely to live in poverty than men in various parts of the developing world, as reported by the Key UN Entity Focused on Development (United Nations Development Programme) in 2017. Along with “gender equality and the empowerment of women and girls” being included as a stand-alone goal among the set of Sustainable Development Goals (SDGs) in the United Nations Agenda for Sustainable Development (Agenda 2030), gender-differentiated poverty has attracted much research interest ^{[9][23][24][25]}[7,22,23,24].

Does poverty actually have a “female face”? There are quite a few studies (see, for example, in ten developing countries ^[26][32]; in Nicaragua ^[27][33]) that empirically challenge the universal acknowledgment of the feminization of poverty (which, however, is still being supported by cross-regional data presented at the United Nations ^[28][34] and by data for 26 diverse Luxembourg Income Study (LIS) countries ^[23][22]). Accordingly, researchers have begun to advocate a more nuanced and complex analysis for validating gender-differentiated poverty ^[26][29][32,35]. Nevertheless, despite the fact that researchers have identified “structural poverty,” or a multiplicity of gender-biased social processes and structures ^[9][25][26][7,24,32], as a component of “transitory poverty” as random shocks and shortfalls in social support in times of emergencies (Casper et al., 1994; Gornick and Jäntti, 2010) ^[23][30][22,36], proper tools and methods that would enable more in-depth empirical examinations of the causes of poverty are still lacking ^[9][7].

China is among those countries experiencing issues related to both the relative poverty levels resulting from worker migration and gender-differentiated poverty. Having generally eliminated absolute poverty by the year 2020 (ten years ahead of the target date for achieving the poverty reduction goal set by the UN 2030 Agenda for Sustainable Development), China is now facing some of the most crucial relative poverty issues and has placed the alleviation of relative poverty into its future developing blueprint plans (see, for example, the proposal of the Fourth Plenary Session of the 19th Central Committee of China’s Communist Party, 2019). In addition to the historically urban–rural difference in poverty, newly emerging phenomena of differentiated poverty, such as differences between urban residents and urban migrants and between male and female population groups, now play a major role concerning relative poverty.

Several researchers have explored the poverty levels among Chinese migrant workers ^{[31][32][33][34]}[37,38,39,40], and while some researchers have demonstrated relative poverty inequalities among different subgroups (see, for example, different occupations ^[35][41]; different ages ^[36][42]), gender differences have rarely been examined empirically (with a few exceptions, such as that by ^[37][43]).

To sum up, the gender-differentiated poverty of migrant workers is an important research topic for developing countries with a high degree of population mobility (such as China) but has so far received little empirical investigation.