Occupational credentials and migrants’ return aspirations: Evidence from foreign elderly-care workers in Japan

Nobuyuki NAKAMURA; Aya SUZUKI; Nobuyuki Nakamura; Aya Suzuki

doi:10.16995/ilr.23803

Responsibility for opinions expressed in signed articles rests solely with their authors, and publication does not constitute an endorsement by the ILO.

This article is also available in French, in Revue internationale du Travail 165 (2), and Spanish, in Revista Internacional del Trabajo 145 (2).

1. Introduction

The increasing scale of international labour migration has significant implications for the global economy (World Bank 2023; ILO 2024). More than half of the world’s migrants are temporary, leaving host countries within five years to return to their home countries or relocate elsewhere (Dustmann and Görlach 2016).¹ While policymakers, experts and researchers argue that return migrants contribute to economic growth in their home countries by transferring knowledge, earning income, building skills and importing social norms from host countries (Domingues Dos Santos and Postel-Vinay 2003; Appleton, Morgan and Sives 2006; Dustmann, Fadlon and Weiss 2011; Wahba 2014; Batista et al. 2025), their departure causes human resource losses for labour-intensive industries in host countries that are already facing chronic shortages due to demographic changes. While the impacts of temporary migrants’ return are widely recognized and extend beyond individuals’ career development, the factors affecting migrants’ decisions to return need to be explored in greater depth.

Migrants’ skills and the rate of return on these skills in each country influence their decisions to leave or stay (Borjas and Bratsberg 1996; Rooth and Saarela 2007). Studies conventionally use educational attainment or years of schooling as a proxy for worker skills. However, relying solely on educational attainment as a skill measure may not provide a comprehensive understanding of migrants’ economic behaviour. Educational attainment alone is insufficient to capture potential labour-market performance in either host or home economies, as it is fixed prior to labour-market entry and does not reflect any learning, experience or skill acquisition that occurs after formal education. Gould and Moav (2016) argue that not only general skills, which can be easily transferred to a foreign country, but also country-specific skills, such as language proficiency and country-specific occupational credentials, affect emigration decisions. Patt et al. (2021) provided the first evidence of the importance of occupational skills in international emigrant selection by examining migration from Mexico to the United States. They showed that manual skills, such as those related to physical strength and machinery operation, were more positively selected than cognitive skills like problem-solving and creativity. These findings suggest that occupational or country-specific skills may play a more critical role than traditional skill measures, such as educational attainment, in emigration decisions. If occupational or country-specific skills significantly influence the decision to emigrate, as research has shown, we need to investigate whether these skills also play a critical role in deciding whether to return. Moreover, although occupational credentials² have garnered attention as signalling mechanisms in the labour market (Spence 1973; Kleiner and Krueger 2010, 2013), they are not usually transferable across borders and retain validity solely within the issuing country. Despite this, research on how occupational credentials affect migrant behaviour remains limited.

This study quantitatively explores whether the occupational credentials held by migrant workers in their home and host countries – serving as signals of skills – influence their return aspirations. We examine the case of foreign care workers (FCWs) in Japan. The Japanese Government recently introduced policy measures to facilitate the arrival of FCWs.³ Most FCWs are employed in long-term care facilities, which provide housing, care and support services for elderly individuals who have difficulty living independently, and work as caregivers or nursing assistants (Huynh et al. 2024). Japan’s visa scheme for FCWs grants permanent residency to workers who obtain local occupational credentials. These visas allow flexible renewals based on migration plans rather than fixed expiration dates. This flexibility enables us to evaluate the relationship between workers’ professional qualifications and their intentions to return to their home countries.

The estimation results from our online survey of FCWs suggest that respondents who have care- or nursing-related occupational credentials from their home countries plan to return to their home countries sooner than those who do not. We also find that migrant workers who have occupational credentials only in their home countries tend to expect to leave significantly sooner. By contrast, migrant workers who obtain occupational certification in Japan tend to stay longer in Japan, although this result becomes statistically insignificant after controlling for socio-economic covariates. These results suggest that while other skill-related factors such as educational attainment or language proficiency do not significantly affect return decisions, occupational credentials in home and host countries play a crucial role in shaping whether migrant workers plan to return to their home countries or stay in Japan.

This study contributes to research on migration and labour markets in several ways. First, it contributes to the discussion on the determinants of temporary labour migrants’ return decisions. Gibson and McKenzie (2011), Bijwaard, Schluter and Wahba (2014), Dustmann and Okatenko (2014), de Haas, Fokkema and Fihri (2015), Crescenzi, Holman and Orru’ (2017) and Bekaert et al. (2024) are among the numerous scholars who have studied why labour migrants return to their home countries. However, the link between migrants’ intentions to return to their home countries and their occupational skills, particularly qualifications, remains underexplored. Most studies on migrant workers’ return behaviour have focused on Western destination countries, even though more than 40 per cent of migrant workers originate from Asian countries (IOM 2024). Although the discussion on the global care chain has grown in response to the rising demand for care workers (Sabio, Pandey and Parreñas 2022; Walton-Roberts 2020; King et al. 2021; King-Dejardin 2019), few studies have quantitatively investigated the economic behaviour of FCWs. By analysing return intentions within the specific geographic and sectoral context of FCWs in Japan, this study offers new insights into the factors shaping return decisions.

Second, this study closely examines the roles of occupational credentials in cross-border labour markets, based on the theoretical framework provided by Dustmann and Görlach (2016), which refers to the signalling theory that has been debated since Spence’s (1973) seminal work. While educational attainment has traditionally been viewed as the primary signalling mechanism in job markets, occupational credentials have gained attention for directly reflecting occupational skills, especially among groups facing higher licensing costs (Blair and Chung 2021, 2025). Regarding labour mobility, evidence shows a relationship between interstate movements and occupational licensing in the United States (Deyo and Plemmons 2022; Johnson and Kleiner 2020; Mulholland and Young 2016). Moreover, several studies suggest that occupational credentials function as barriers to labour market entry, while providing higher wage premiums for immigrants in destination countries (Kugler and Sauer 2005; Federman, Harrington and Krynski 2006; Peterson, Pandya and Leblang 2014; Gomez et al. 2015; Cassidy and Dacass 2021; Tani 2021; Koumenta, Pagliero and Rostam-Afschar 2022; Brown 2023; Chung 2024; Kleiner and Xu 2025). On the other hand, Tani (2023) suggests that adjustment to skills mismatch and wage premiums are limited to immigrants in the United States. In Japan, occupational credentials are generally associated with higher wage premiums, but their effects on labour supply vary by occupation (Kawaguchi, Murao and Kambayashi 2014; Ramseyer and Rasmusen 2015; Morikawa 2018). This study also has implications for the existing literature on qualification recognition (e.g. Sweetman, McDonald and Hawthorne 2015). While research shows that recognizing foreign qualifications benefits migrant workers in countries such as Germany (Brücker et al. 2021) and Switzerland (Pecoraro and Tani 2023), no studies have examined the relationship between cross-border return behaviour and host- and home-country credentials. Furthermore, this is the first study in Japan to analyse occupational credentials and labour supply outcomes among immigrants. Our study offers new evidence on the role of occupational credentials in international labour markets through the lens of temporary migration.

The remainder of this article is organized as follows: section 2 describes the contexts of this study, section 3 reports the data, section 4 discusses the estimation strategies employed, section 5 presents the estimation results and section 6 concludes.

2. Contexts

2.1 Theoretical framework

Our theoretical framework is based on Dustmann and Görlach (2016) and focuses on the roles that skills obtained in home and host countries play in determining the migrants’ return to their home countries. In their seminal paper, Dustmann and Görlach (2016) define two components for the log of real earnings Y as:

$l n Y L (S) = α 0 L + α 1 L l n S, L ∈ {d, o}$ (1)

where $α 0 L$ is the log of the rental rate of human capital in location L, with d indicating the destination country and o the origin country, and $α 1 L$ is the rate of return to skills S in location L. Skills are also considered to be accumulated as $S a = S a − 1 + θ S L$ , where a is age and $θ S L > 0$ is the increase in skills in each period at location L. It is important that they differentiate these two components – $α 0 L$ and $α 1 L$ – that contribute to increasing earnings. The former (log rental rates) may differ across countries due to factors such as differences in available technologies, while the latter (returns to skills) may differ due to factors such as “different industry structures and thus different skill demands” (2016, 109) between countries. They further formalize the optimal migration duration among international migrants from the utility gains at each location. The indifference condition of the model is represented as follows:

$μ d (π, Y d (S a) + (1 + r) A a − A a + 1 d) + β V (a + 1, A a + 1 d, S a + θ s d) = μ o (Y o (S a) + (1 + r) x A a − x A a + 1 o) + β V o (a + 1, x A a + 1 o, S a + θ s o)$ (2)

where the left-hand side of the equation represents the value of remaining in the destination country (V^d(Ω)), and the right-hand side denotes the value of returning to the origin country (V^o(Ω)). The value associated with each location consists of: the preference for the host country relative to the home country π; real income as function of skills $Y L (S)$ , where skills are accumulated by age a at each location L ∈ {d, o}; the stock of assets A accumulated by age a; skill accumulation at each location $θ s L$ ; the real interest rate r; and the real exchange rate x. From equation (2), they derive the optimal condition whereby the migrants’ decision to return depends on whether the current utility gain from returning to the origin country during this period exceeds the future gain from remaining in the destination for an additional period.

Among the various reasons for temporary migration considered by Dustmann and Görlach, “temporarily higher earnings in the destination country” is the one that best fits our study. If skilled labourers are scarce in the origin country – due to different social structures and lower demand for such workers – the limited supply implies that the returns to skill are higher than in the destination country (i.e. $α 1 o > α 1 d$ ). On the other hand, the log rental rate of human capital is expected to be higher in the destination country (i.e. $α 0 o < α 0 d$ ) due to differences in available technologies. Moreover, the rate of skill accumulation is likely to be greater in the destination country (i.e. $θ s d > θ s o$ ) due to its more advanced industrial structure and higher capital intensity. In such cases, potential migrants may be attracted to temporary migration because the wages are higher and the accumulation of skills is initially faster in the destination country. However, because the relative scarcity of skills in the origin country increases returns to these specific skills, migrants may become more incentivized to return to their origin country during certain periods. Dustmann and Görlach state that “if differences in skill prices are large,⁴ migrations may be temporary because, at some point, the immigrant’s higher productivity potential in the origin country will overcompensate for the differences in rental rates” (2016, 114). By simulating their model, the authors show how earnings profiles change over the course of immigration in such cases.

This framework applies to FCWs in Japan (the destination country) coming primarily from developing Asian countries (the origin countries). As the care industry in Japan is underpinned by robust infrastructure and policies and is more developed than in other Asian countries due to its ageing population, the returns to these specific care-related skills are lower in Japan than in those countries $(α 1 o > α 1 d)$ . On the other hand, migrant workers expect higher earnings $(α 0 o < α 0 d)$ and faster skill accumulation in Japan $(θ s d > θ s o)$ – for example, from industry-specific training programmes – because of the advanced state of its care industry. Thus, while FCWs are initially drawn to Japan, they may eventually decide to return to their origin countries.

While Dustmann and Görlach’s theory fits our case study well, they do not consider heterogeneous factors across individuals affecting their return decisions.⁵ Although numerous factors influence return decision-making, such as earnings (Reagan and Olsen 2000; Bijwaard and Wahba 2014), risk aversion and patience (Gibson and McKenzie 2011), attachment to the job market or society in the origin and destination countries (Constant and Massey 2003; Bijwaard, Schluter and Wahba 2014; Crescenzi, Holman and Orru’ 2017), social network and integration in the destination country or family ties (de Haas, Fokkema and Fihri 2015) and economic or environmental changes in the origin and destination countries (Bekaert et al. 2024), we focus primarily on the role of occupational credentials in our research. These occupational credentials often serve as objective evidence of workers’ professional skills and competencies in their respective labour markets and are usually valid only in the labour market of the issuing country. We hypothesize that having occupational credentials in the origin or destination country affects temporary migrants’ decision to return home, as these credentials reduce uncertainty in job markets (Spence 1973; Kleiner and Krueger 2010, 2013).

Migrants with occupational credentials from their origin country may be more sensitive to push factors for return migration, such as declining earnings differentials between the destination and origin countries, since these credentials enhance their prospects of finding suitable employment and earning appropriate wages at home, thereby reducing the uncertainty associated with re-entering the home labour market. Returnees face high adjustment costs when reintegrating into the home labour market due to mismatches in skills and reservation wages (Wahba 2014, 2021) and their lack of, or weak, professional connections (Redbird 2017; Kim 2020). However, home-country credentials mitigate these barriers by signalling their high skills. Blair and Chung (2021) find that credentials function as informative signals in the labour market when the obtaining costs are high, and that demographic minorities (i.e. returnees in our case) benefit from licence premiums. The costs of obtaining government certifications are indeed high relative to their benefits, especially in the care or nursing sectors (Kleiner and Krueger 2013; Dill et al. 2022).

In contrast, destination-country occupational credentials act as pull factors that extend migrants’ stay, as employers in destination countries value higher skills S, thereby maintaining earnings gaps between the origin and destination countries for longer. Origin-country occupational credentials are of little help in destination countries because they are not usually regarded as proof of mature occupational skills.⁶ Moreover, workers in the care industry often face greater wage penalties than workers in other industries in most destination countries, including Japan (Lightman 2017; Budig and Misra 2010; Razavi and Staab 2010; England, Budig and Folbre 2002). However, destination-country occupational credentials act as signals of skills and command higher earnings. Thus, the earnings differentials of workers who hold such credentials may take longer to shrink to a level that makes return migration attractive. Based on this framework, we empirically test whether origin and destination-country occupational credentials are associated with the duration of temporary migration.

2.2 Occupational credentials and FCWs in Japan

Japan’s rapidly ageing population has driven the development of the elderly-care industry. The Japanese Government oversees the national certification of care workers, who are known as Certified Care Workers (CCWs). This is a certification that attests to specialized-care knowledge and skills, rather than a regulatory licensure, to ensure quality in the industry. CCWs have better salaries and opportunities for promotion in most facilities. To be eligible for the exam, candidates must have graduated from a government-approved vocational school or have more than three years of care work experience and practical training. The annual written exam has a pass rate of approximately 70 per cent, and the certification, once obtained, does not require renewal. Although foreigners are not subject to additional requirements to take the exam, they may use reading aids and have extra time to complete the exam due to Japanese language requirements. On the other hand, most FCWs in Japan have obtained professional qualifications relating to nursing or elderly-care in their home countries. Each country organizes a qualification system to maintain the quality of its care and nursing systems (see supplementary online appendices for details). These qualifications function as proof of the workers’ skills relating to care and nursing and help to them to secure better salaries and positions, but they do so only in the issuing country due to the absence of recognition systems between Japan and the origin countries.

Japanese elderly-care labour supply remains insufficient and required an estimated 2.43 million additional workers by 2025 (Kakizoe 2023). As the expansion of elderly care is expected to enhance the national labour market and economy (Sugawara and Nakamura 2014; Fu et al. 2017), and the expansion of migrant workers does not appear to have had adverse effects on the employment or wages of non-migrant workers (Nakamura 2010; Tomohara 2022), the Government has implemented policy measures facilitating FCW employment in response to sectoral shortages. There are four types of elderly-care visas in Japan: Economic Partnership Agreement (EPA), Nursing Care (NC), Technical Intern Trainee Program (TITP) and Specified Skilled Worker (SSW) visas. The characteristics of each visa are summarized in table 1, and their details are provided in the supplementary online appendices. EPA and NC are permanent visas, while TITP and SSW are temporary. To be eligible for permanent visas, workers must obtain CCW qualifications. An interesting feature of these visa schemes is that FCWs can move from one visa to another based on their careers. Temporary visa holders who pass CCW exams can switch to NC visas for permanent residence. Likewise, interns on TITP visas can switch to SSW visas without passing exams and stay in Japan for up to ten years without exogenous constraints. These unique and flexible characteristics of the visa scheme allow us to compare the intended length of migrants’ and FCWs’ stay in Japan.

Table 1. Visa categories

	Economic Partnership Agreement (EPA)	Nursing Care (NC)	Technical Intern Trainee Program (TITP)	Specified Skilled Workers(SSW)
Classification	Permanent migrants (high-skilled immigrant workers)		Temporary migrants (low-skilled immigrant workers)
Purpose	To accept FCWs so that they can obtain the national qualification of CCWs under the EPA scheme	To accept foreign specialists in elderly care	To transfer skills from Japan to other countries as part of international cooperation	To accept foreign nationals with specific expertise or skills to address labour shortage
Year introduced	2008	2017	2017	2019
Year introduced	- Japan concluded an EPA with Viet Nam in 2008, the Philippines in 2009 and Indonesia in 2014		- Elderly-care was added to the most extensive training programmes, which had been launched in 1993
Number of visa holders	3 820 (December 2020)	1 714 (December 2020)	12 068 (March 2021)	3 947 (September 2021)
CCW qualification required at arrival?	Not required, but they are expected to obtain it	Required	Not required	Not required
CCW qualification required at arrival?	- EPA visa holders must have graduated from nursing schools or obtained certified nursing or care qualifications in their home countries	- Most NC visa holders have studied in Japan and completed professional school degrees to obtain their qualifications	- TITP holders can switch to NC visas if they pass the CCW exam	- SSW holders can switch to NC visas if they pass the CCW exam
Duration of stay	Permanent after passing the exam	Permanent	Maximum 5 years	Maximum 5 years
Duration of stay	- Must return after 4 years if they fail the exam		- Must take evaluation exams to verify specialization to renew the visa- Can switch to SSW visas without taking exams	- No evaluation exam required for renewing SSW visas
Minimum JLPT level	N3	N2	N4	N5
Family accompaniment	Not permitted (permitted only after passing the exam)	Permitted	Not permitted	Not permitted

Notes: The JLPT is a standardized test that assesses the Japanese proficiency of non-native speakers. It evaluates language knowledge and reading and listening comprehension. The JLPT consists of five levels, ranging from N5 (beginner) to N1 (advanced). Test takers must obtain a certain score to pass each level.

Source: The descriptions of each visa category are based on information from the Japanese Ministry of Health, Labour and Welfare (2019). The EPA, NC and SSW visa data taken are from the Immigration Services Agency of Japan, and the TITP visa data are taken from the Organization for Technical Intern Training.

3. Data

We conducted an online survey in August and September 2021 to gather comprehensive data on the economic behaviour of FCWs.⁷ We recruited respondents through invitation letters sent to more than 4,500 elderly-care facilities (34.5 per cent of facilities nationwide) employing FCWs, based on open data published by the Ministry of Health, Labour and Welfare and information from media sources.⁸ FCWs in these facilities could access the online survey⁹ through the invitation letters delivered to their workplaces. The original survey was drafted in beginner-level Japanese, as all FCWs are required to demonstrate Japanese proficiency for their visa applications. A Japanese-language education expert supervised revisions to the survey. To further support respondents, we added translations of complicated technical terms into several languages, which were reviewed by native speakers, and encouraged respondents to seek assistance from fluent Japanese speakers if necessary.¹⁰ As a result, more than 76 per cent of the respondents reported that the survey was easy or relatively easy to answer.

We collected socio-economic information from respondents, including their visa type, date of entry into Japan, age, income, gender, education, work experience in their home country and any loans taken to finance their migration. We also collected information about their household in their origin country. To address concerns typically associated with online surveys – i.e. that respondents may not answer as diligently as researchers expect – we included three instructional manipulation checks (IMCs) mid-questionnaire (Oppenheimer, Meyvis and Davidenko 2009). Given the random errors from incorrect answers, samples were considered valid only when the respondents correctly answered at least two IMCs. Although we collected data from 218 observations, only 200 met our IMC criteria and were thus included in our research samples.¹¹ It should be noted that approximately 20,000 FCWs were engaged in the care sector in Japan when the survey was conducted in 2021 (table 1).

Table 2 and figures 1 and 2 present summary statistics for the sample.¹² In the sample, 84 per cent of the respondents were women, with an average age of 28 and an educational level equivalent to completing upper secondary school in their home countries. While 33 per cent had obtained CCW qualifications in Japan, approximately 70 per cent had obtained care or nursing credentials in their home country. Among qualified workers, 45 respondents (22.5 per cent) had qualifications both in Japan and in their home country. Sixty per cent worked in metropolitan cities in Japan, such as Tokyo. Seven per cent were accompanied by family members. Moreover, 83 per cent of respondents were from Viet Nam, Indonesia and the Philippines, which have EPAs with Japan, while the remaining respondents were from other Asian countries, such as China and Myanmar. Some 108 workers held EPA or NC visas (i.e. permanent visas), while 92 held temporary visas; thus, the permanent-to-temporary visa ratios were proportional. As the incomes of the respondents and households¹³ were reported in different currencies, we converted them into US dollars using the purchasing power parity-adjusted price.¹⁴ The respondents’ incomes exceeded average household incomes in their home countries.

Table 2. Descriptive statistics

Variable	Obs.	Mean	Std. Dev.
Return index (0~11: return X years later; 12: permanent stay)	200	5.945	4.319
Expected length of stay(no. of days from 1 July 2021 to 31 December 2032 (max.))	200	2263.045	1452.070
Intends to stay permanently (= 1 if yes)	200	0.25	0.434
Has qualifications in the home country (= 1 if yes)	200	0.68	0.468
Has qualifications in the host country (= 1 if yes)	200	0.33	0.471
Has qualifications in both countries (= 1 if yes)	200	0.225	0.419
Has qualifications only in the home country (= 1 if yes)	200	0.455	0.499
Has qualifications only in the host country (= 1 if yes)	200	0.105	0.307
Perceived job-skill score (max: 100 points)^a	200	72.3399	16.899
Language level (min.:1 (N5) to max.: 5 (N1))^b	200	3.315	0.877
Years of education	200	13.32	2.832
Length of stay in Japan(no. of days from entry date to 1 July 2021)	200	956.8	737.693
Period remaining on current visa(no. of days from 1 July 2021 to visa expiration date)	200	399.4	292.542
Gender (= 1 if female)	200	0.84	0.368
Age (in years)	200	27.715	4.433
Visa type (= 1 if temporary visa holder)	200	0.46	0.500
Marital status (= 1 if married)	200	0.175	0.381
No. of children in the household	200	0.225	0.766
Works in a metropolis (= 1 if yes)^c	200	0.59	0.493
Loans taken to finance migration (= 1 if yes)	200	0.44	0.498
Log-converted income	200	7.14413	0.448
Log-converted household income	200	6.458129	1.864
Expected return at the time of arrival (no. of years after arrival)	200	5.82911	3.047
Existing social network in the host country at arrival (= 1 if yes)	200	0.605	0.490
Accompanied by family members in Japan (= 1 if yes)	200	0.075	0.264
No. of days per month contacts family in home country (max.: 28 days))	200	10.82	10.393

Notes:

^a Regarding their perceived job skills, respondents evaluated their daily self-performance on the 15 care-related tasks using a 5-point Likert scale, scoring each task as follows: 0 = “I have never done the task”, 1 = “I cannot do the task”, 2 = “I am relatively unable to do the task”, 3 = “I am relatively able to do the task” and 4 = “I do the task well”. These tasks include cleaning, washing clothes, making beds, assisting with bathing, repositioning for pressure-area care, applying compresses, measuring body temperature, administering eye drops, assisting with eating and drinking, applying ointments, cutting nails, caring for older persons with dementia, assisting with terminal care, assisting with tube feeding and suctioning sputum. The total score was then standardized to a scale with a maximum of 100 points.

^b We used the results of the JLPT – a standardized test that assesses the Japanese proficiency of non-native speakers – as a measure of respondents’ language skills. The JLPT evaluates language knowledge and reading and listening comprehension. It consists of five levels, ranging from N5 (beginner) to N1 (advanced). Test takers must obtain a certain score to pass each level. In this study, we converted the JLPT levels into a numerical scale as follows: N5 = 1, N4 = 2, N3 = 3, N2 = 4 and N1 = 5.

^c We classified the prefectures in three metropolitan areas – Tokyo (Tokyo, Saitama, Chiba and Kanagawa), Osaka (Osaka and Hyogo) and Nagoya (Aichi) – as “the metropolis” in our study.

Source: Own calculations.

Figure 1. Distribution of visa types

Source: Own calculations.

Figure 2. Distribution of origin countries

Source: Own calculations.

4. Estimation strategy

We explored migrant workers’ return expectations, not their actual return behaviours. According to Dustmann and Görlach (2016), the intended duration of migration – as opposed to the completed duration – is vital for measuring the relationship between migration duration and economic behaviour, as migrants continuously reoptimize their migration plans over life cycles, conditioning present choices on future expectations. Citing Manski (1990), Dustmann and Okatenko (2014) suggested that there are discrepancies between migrant workers’ stated intentions and actual behaviours because they learn new information after stating their intentions. Moreover, Tjaden, Auer and Laczko (2019) observed strong correlations between migration intentions and recorded bilateral flows, as well as similar correlations between migration intentions and aggregated out-migration, using official migration flow data from over 160 countries. We primarily investigated how migrant workers’ current occupational skills – including occupational credentials in their home and host countries – shape their return strategies. Migrants’ present intentions are therefore essential for understanding their migration plans. Using unique cross-sectional data consisting of 200 observations, we estimated the following model:

$R e t u r n i = α + OS i β + X i δ + C j + μ i$ (3)

where Return_i is the return intention of the i-th migrant and is the outcome of the model. To measure this, we asked respondents, “When do you plan to return your home country?” regardless of their current visa expiration date. Respondents selected their intended year of departure from the survey year (2021) through 2032, or indicated if they intended to stay in Japan permanently. Based on this question, we constructed three estimation variables. First, we set “the return index” from 0 (intends to stay until 2021) to 11 (intends to stay until 2032) and 12 (intends to stay permanently), which shows the degree of intention to stay permanently. Note that this is not a cardinal index. Second, we created an expected length of stay variable by calculating the number of days between 1 July 2021 and 31 December of each reported target year, assuming a year-end departure. For descriptive purposes, we set 31 December 2032 as the maximum date for this variable, regardless of whether respondents intended to stay in Japan permanently. Finally, we created a dummy variable for permanent stay intention (= 1 if the respondent reported the intention to stay in Japan permanently). We estimate this model using linear probability models for binary outcomes.

OS_i is the vector of the variables related to the occupational skills of respondent i, including home and host country credentials. We asked respondents whether they had credentials related to care or nursing in their home countries, creating a dummy variable for their home country credentials. For CCW qualifications in Japan, NC visa holders were considered qualified (= 1) based on the visa eligibility requirements, while TITP and SSW visa holders were considered not qualified (= 0). EPA visa holders were asked separate questions about their qualifications. Respondents were also asked questions about the care-related tasks that they perform, as elderly care includes basic tasks (e.g. cleaning, washing clothes and making beds) and some medical-related tasks (e.g. tube feeding and sputum suctioning). We asked them about their subjective performance of 15 care-related tasks using a Likert scale from 0 (“I have never done the task”) to 4 (“I do the task well”). We summed the scores of all tasks and standardized them to create a job-perceived score with a maximum of 100 points.¹⁵

Host-language skills are crucial for migrant performance in service industries requiring communication, such as elderly care. We used Japanese-Language Proficiency Test (JLPT) results as language skill proxies. The JLPT is a standardized test that assesses the Japanese proficiency of non-native speakers by evaluating their language knowledge and reading and listening comprehension. The test consists of five levels, ranging from N5 (beginner) to N1 (advanced), and test takers must obtain a certain score to pass. We asked respondents about their most recent JLPT proficiency level. For respondents who were unable to recall their level, we substituted entry-level or minimum visa requirements (SSW: N5, TITP: N4, EPA: N3 and NC: N2). We assigned numerical values based on difficulty (1 for N5, 5 for N1). Moreover, we included years of education as general human capital in HC vectors.

X_i is the vector of the individual socio-economic characteristics of i, including length of stay in Japan (number of days from the date of entry in Japan to 1 Jul 2021), period remaining on current visa (number of days from 1 July 2021 to the visa expiration date), age, gender (1 = female), visa type (1 = temporary (i.e. TITP or SSW)), marital status (1 = married), number of children in the household, workplace location (1 = metropolis¹⁶), loan taken to finance migration (1 = yes), log-transformed monthly average income of individual i, log-transformed monthly average income of i’s household remaining in the home country, and expected return at the time of arrival (years after arrival). As migrants’ social networks play a decisive role in shaping their economic behaviour (Dustmann and Görlach 2016), we included several variables related to social networks: a dummy variable indicating whether the respondent had family or friends in the host country at the time of entry, a dummy for whether any family members currently accompany the respondent in Japan, and the number of days per month the respondent contacts family members in the home country. C_j denotes the fixed effect for home country j and μ_i is the error term.

We regressed this model using ordinary least squares as a benchmark analysis. However, obtaining occupational qualifications is relatively self-selected, potentially causing endogeneity problems. We employed the inverse probability weighted regression adjustment estimator as a robustness check to mitigate potential self-selection bias based on observable characteristics. We also employed negative binomial regression as a robustness check for the count-dependent variable analysis. These robustness checks yielded consistent results. Furthermore, we observed a wage premium associated with credentials issued in the destination country. The results of the robustness checks and the wage-premium analysis appear in the supplementary online appendices.

5. Results

Table 3 presents the main results.¹⁷ Columns 1–3 show the regression results for the return index, measuring the intention to stay in Japan. Foreign workers with home-country care and nursing credentials expressed a stronger desire to return to their home countries sooner, and this pattern remained consistent after controlling for socio-economic covariates. Workers with CCW qualifications showed longer intended stays (column 1), although the effect became insignificant after the inclusion of covariate controls. Columns 4–6 show models of expected duration with similar tendencies. Workers with home-country credentials intended to return 1–1.6 years sooner, with significance at the 5–10 per cent levels, while CCW qualifications extended expected stays by approximately 1.8 years in limited models. The linear probability models for intentions to permanently stay in Japan (columns 7–9) indicate that CCW qualifications increased the likelihood of intending to stay permanently by 17.3 percentage points at 5 per cent significance levels, but only on models including human capital vectors. Other covariates showed no significant effects on human capital variables such as education; instead, proof of occupational skills was more important than traditional skill measures. The expected duration at the time of entry and the period remaining on the current visa are positively associated with migrants’ current expectations of staying. In other words, even when holding constant the differences in the expected duration at arrival and the remaining period on the current visa, occupational credentials still play a significant role in shaping the expected length of migrants’ stay.

Table 3. Estimation results (primary analysis)

	Return index(0~11: return X years later, 12: permanent stay)			Expected length of stay (no. of days from 1 July 2021)			Intends to stay permanently (= 1 if yes)
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
Has qualifications in the home country	–1.775*	–1.183+	–1.055	–602.192*	–394.866+	–353.559	–0.126	–0.102	–0.087
	(0.717)	(0.641)	(0.645)	(238.673)	(213.891)	(214.532)	(0.077)	(0.071)	(0.072)
Has qualifications in the host country	1.997**	–0.526	–0.693	666.278**	–167.754	–220.619	0.173*	–0.067	–0.089
	(0.685)	(0.874)	(0.865)	(229.708)	(297.555)	(294.206)	(0.071)	(0.085)	(0.086)
Perceived job-skill score	0.027	–0.003	–0.003	9.727	–0.782	–0.507	0.001	–0.001	–0.001
	(0.018)	(0.017)	(0.017)	(6.164)	(5.549)	(5.567)	(0.002)	(0.002)	(0.002)
Language level	0.048	–0.246	–0.306	9.467	–87.068	–106.344	0.022	–0.008	–0.015
	(0.381)	(0.339)	(0.337)	(128.069)	(111.415)	(110.335)	(0.040)	(0.041)	(0.041)
Years of education	0.02	–0.031	–0.016	12.204	–4.631	0.541	–0.013	–0.019+	–0.018
	(0.126)	(0.095)	(0.096)	(41.546)	(31.481)	(31.965)	(0.013)	(0.011)	(0.011)
Expected return at arrival		0.607***	0.615***		211.185***	213.965***		0.028*	0.029*
		(0.094)	(0.096)		(31.212)	(31.931)		(0.012)	(0.012)
Period remaining on current visa		0.002*	0.003*		0.816*	0.900*		0	0.000+
		(0.001)	(0.001)		(0.405)	(0.414)		(0.000)	(0.000)
Constant	3.107	8.501+	9.614+	1225.006+	3169.210*	3514.472*	0.254	0.325	0.493
	(2.090)	(4.636)	(4.916)	(703.080)	(1584.110)	(1679.168)	(0.211)	(0.506)	(0.533)
R-squared	0.157	0.418	0.422	0.155	0.425	0.429	0.142	0.286	0.293
Controlled by socio-economic characteristics	N	Y	Y	N	Y	Y	N	Y	Y
Controlled by social network characteristics	N	N	Y	N	N	Y	N	N	Y
No. of obs.	200	200	200	200	200	200	200	200	200

Notes: +, *, ** and *** indicate statistical significance at the 10, 5, 1 and 0.1 per cent levels, respectively. Robust standard errors are reported in parentheses. Columns (1)–(6) are estimated using ordinary least squares, and columns (7)–(9) using a linear probability model. Columns (2), (5) and (8) control for length of stay in Japan, age, gender, visa type, marital status, number of children in the household, workplace location, loans taken to finance migration and the log-transformed monthly average income of the respondent and their household in the home country. In addition, columns (3), (6) and (9) include a dummy variable indicating whether the respondent had family or friends in Japan at the time of entry, a dummy for whether any family members currently accompany the respondent in Japan, and the number of days per month the respondent contacts family members in the home country. All estimations include home-country fixed effects. For brevity, only the relevant and statistically significant variables are shown.

Source: Own calculations.

To examine the effects of qualifications in greater detail, we decomposed respondents’ qualifications into three categories: (i) those holding an occupational qualification in both the host and home countries, (ii) those with home-country qualifications only, and (iii) those with host-country qualifications only. Table 4 shows the estimation results. Respondents holding home-country care and nursing credentials only intended to leave 1–1.5 years sooner than the baseline group (those without any occupation credentials), with the estimates statistically significant at the 5–10 per cent levels (columns 1–6). These results point to several likely scenarios. Given the average expected duration of 6.2 years (table 2), temporary migrants who hold home-country qualifications and migrate under the TITP and SSW schemes (each of which allows a maximum stay of five years) are unlikely to apply for another type of visa or extend their stay. These results suggest a loss of skilled human resources for host-country care facilities, though returning migrants may transfer the five years of elderly-care experience acquired in Japan back to their home countries. Migrants with only host-country qualifications remained in Japan for more than two additional years (columns 1 and 4) and exhibited a 23-percentage-point higher likelihood of permanent settlement compared to non-credentialed workers (p<0.10) (column 7) in limited models. In contrast, dual-qualification holders showed statistically significant negative effects once socio-economic variables were included (columns 3, 6 and 9), suggesting a desire to return to home countries despite holding host-country credentials. Robustness checks using alternative estimation methods are presented in the supplementary online appendices.

Table 4. Estimation results (decomposition analysis)

	Return index(0~11: return X years later, 12: permanent stay)			Expected length of stay (no. of days from 1 July 2021)			Intends to stay permanently (= 1 if yes)
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
Has qualifications in both countries	0.221	–1.732	–1.796+	63.811	–570.861	–589.763+	0.046	–0.169	–0.182+
	(1.000)	(1.062)	(1.030)	(335.188)	(360.430)	(349.371)	(0.102)	(0.106)	(0.105)
Has qualifications only in the home country	–1.609*	–1.241+	–1.171+	–550.652*	–415.270+	–390.932+	–0.101	–0.104	–0.101
	(0.810)	(0.680)	(0.676)	(270.531)	(227.530)	(225.711)	(0.081)	(0.070)	(0.072)
Has qualifications only in the host country	2.373*	–0.685	–1.023	783.130*	–223.133	–326.503	0.230+	–0.074	–0.129
	(1.108)	(1.292)	(1.341)	(365.088)	(438.358)	(452.403)	(0.138)	(0.142)	(0.152)
Perceived job-skill score	0.027	–0.003	–0.002	9.57	–0.704	–0.363	0	–0.001	–0.001
	(0.018)	(0.017)	(0.017)	(6.225)	(5.594)	(5.614)	(0.002)	(0.002)	(0.002)
Language level	0.051	–0.251	–0.319	10.637	–88.905	–110.282	0.022	–0.008	–0.017
	(0.383)	(0.345)	(0.344)	(128.510)	(113.239)	(112.414)	(0.040)	(0.042)	(0.042)
Years of education	0.021	–0.032	–0.019	12.62	–5.005	–0.27	–0.013	–0.019+	–0.018
	(0.126)	(0.095)	(0.097)	(41.669)	(31.514)	(32.090)	(0.014)	(0.011)	(0.011)
Expected return at arrival		0.607***	0.616***		211.370***	214.336***		0.028*	0.029*
		(0.094)	(0.095)		(31.211)	(31.850)		(0.012)	(0.012)
Period remaining on current visa		0.002+	0.003*		0.827+	0.922*		0	0.000+
		(0.001)	(0.001)		(0.424)	(0.440)		(0.000)	(0.000)
Constant	3.000	8.504+	9.648+	1191.782+	3170.374*	3525.555*	0.238	0.325	0.497
	(2.099)	(4.653)	(4.938)	(704.814)	(1590.197)	(1686.586)	(0.213)	(0.508)	(0.534)
R-squared	0.154	0.415	0.419	0.151	0.422	0.426	0.139	0.282	0.289
Controlled by socio-economic characteristics	N	Y	Y	N	Y	Y	N	Y	Y
Controlled by social network-related characteristics	N	N	Y	N	N	Y	N	N	Y
No. of obs.	200	200	200	200	200	200	200	200	200

Notes: +, *, ** and *** indicate statistical significance at the 10, 5, 1 and 0.1 per cent levels, respectively. Robust standard errors are reported in parentheses. Columns (1)–(6) are estimated using ordinary least squares, and columns (7)–(9) using a linear probability model. Columns (2), (5) and (8) control for length of stay in Japan, age, gender, visa type, marital status, number of children in the household, workplace location, loans taken to finance migration and the log-transformed monthly average income of the respondent and their household in the home country. In addition, columns (3), (6) and (9) include a dummy variable indicating whether the respondent had family or friends in Japan at the time of entry, a dummy for whether any family members currently accompany the respondent in Japan, and the number of days per month the respondent contacts family members in the home country. All estimations include home-country fixed effects. For brevity, only the relevant and statistically significant variables are shown.

Source: Own calculations.

6. Conclusion

We examined migrant workers’ return intentions through the lens of occupational credentials as proof of their skills. Our quantitative analysis suggests that home-country occupational credentials lead migrant workers to return to their home countries sooner, while host-country occupational credentials prompt migrants to stay in the host country for longer periods. The results provide new insights into how professional qualifications are recognized and valued differently across national borders, with important implications for understanding international labour mobility patterns.

First, our findings highlight the fact that occupational skills, such as credentials, are non-negligible determinants of economic behaviour among temporary migrants. Much of the existing debate had focused on whether migrants’ decision to migrate or return are shaped primarily by educational attainment, with individual skill levels and skill premiums in home and host countries often used as proxies. However, in our study, educational attainment did not yield statistically significant effects. Instead, occupational skills, in the form of credentials, were a key determinant of migrants’ return intentions. To understand the economic behaviour of temporary migrants, we must fully consider their occupational-specific human capital, including their credentials.

Second, our findings show that home-country occupational credentials play a critical role in migrant workers’ return decisions. Migrants who hold home-country credentials are more likely to consider returning sooner, because they anticipate receiving appropriate wages for their skills and face lower reintegration costs when they re-enter the labour market in their home countries. Moreover, the undervaluation of the potential skills of workers with home-country occupational credentials may further encourage such workers to return sooner (ILO 2018). Despite the high demand and economic opportunities for care workers in Japan, worker retention remains challenging. Research shows that retention of FCWs is strongly influenced by whether such workers feel that their abilities are valued (Kameyama et al. 2022), as well as by their leadership opportunities and relationships with coworkers (Asis and Carandang 2020). Some nurses returning from Japan reported having to perform auxiliary tasks in elderly care and that their limited opportunities made it difficult to maintain their competencies (Efendi et al. 2019). Such constraints can lead to demotivation, hinder skill transfer and professional development, and result in deskilling or “brain waste” (Kurniati et al. 2017). Given that imperfect skill transferability among migrant workers is well-documented (Chiswick and Miller 2009), migrant workers with home-country occupational credentials may opt to return to their home countries sooner to maintain their high level of competencies.

Although many countries have increasingly relied on FCWs, migrant workers in the healthcare sector have been at a disadvantage, mainly due to weak government policies and legal frameworks (Pastor-Bravo and Nelson 2019). In Japan, there are integration complexities stemming from shortcomings in the visa system (Hirano, Tsubota and Ohno 2020). One of the policy measures that the Japanese Government is envisaging is to reform the TITP so as to allow migrant workers with temporary visas to change jobs and work in different workplaces as from 2027 (Japan News 2023), which may enhance mobility but place further strains on retention. In our survey, one FCW on a TITP visa expressed a desire to use her developed Japanese-language skills in other industries if job transfers were permitted. Although our data do not allow us to pinpoint the specific mechanisms behind the credential effects on return decisions, our findings suggest that improving labour conditions in elderly-care facilities to attract skilled FCWs is crucial. While the Japanese Government plans to provide subsidies to care facilities for FCW recruitment and to support a nursing-care education programme in origin countries amid “a global war for talent” (Japan News 2024), our findings underscore the fact that supporting potentially skilled workers, especially FCWs who hold occupational credentials in their home countries, would help to solve the labour shortage in the care industry in Japan. Japan’s advisory panel has proposed mutual qualification recognition to facilitate skill recognition (Immigration Services Agency of Japan 2024). In light of global discussions on the movement of skilled and prospective labour (Sweetman, McDonald and Hawthorne 2015; Abarcar and Theoharides 2024), the establishment of an international certification system for the care sector might help to reduce skill mismatches caused by overqualification in host countries.

Moreover, while our findings suggest that occupational credentials issued in home countries increase return intentions, credential holders may face potential challenges upon their return. Some studies (e.g. Efendi et al. 2021) suggest that FCWs face difficulties in utilizing their skills after returning home. Although the return of successful care workers should reinforce human, financial, and social capital in home-country care industries, returning workers’ skills often remain underutilized due to inadequate supporting infrastructure, such as underfunded health systems and weak frameworks for labour protection and formal employment. Moreover, poor recruitment systems in developing countries create employment barriers for returnees (Kurniati et al. 2017). Traditionally, care work was provided by family members, but economic developments have been reshaping family structures and social norms around care. As institutionalized care systems become essential in developing countries, measures are needed to enable returning workers to fully utilize their acquired skills in their home countries.

Our study is not without limitations. First, the use of cross-sectional data prevents us from being able to establish clear causal relationships. Efendi et al. (2021) review several factors, such as family reasons and communication barriers, that influence migrants’ return behaviours. While we controlled for these household characteristics and language proficiency in our estimations, they also highlight the importance of psychological and emotional dimensions, including career aspirations and workplace discrimination, which may shape migration workers’ acquisition of credentials and return aspirations. Future research should aim to collect information on motivations, workplace experiences and psychosocial dynamics that was unavailable in our survey. Second, Japan’s migration regime in the care sector is still evolving. Temporary visa (TITP and SSW) holders, who began arriving in 2018 and 2020, respectively, were only able to take their first CCW exams in 2022, one year after our data were collected.¹⁸ Furthermore, as noted earlier, the visa system governing FCWs is being reformed. In addition, although this study focuses on return aspirations, the available data do not allow us to observe actual return behaviour. Future studies on determinants for CCWs and actual return behaviours using panel datasets are required. Finally, our analysis focuses on a single industry (elderly care) in one host country (Japan), which restricts external validity. Because credential regulation varies across industries (Kleiner and Krueger 2010), further research on credential generalizability across various destinations is needed. Despite these limitations, we believe that our findings provide important insights into migration policies at a time when global demand for migrant workers, and especially FCWs, is rapidly accelerating.

Acknowledgments

This research was supported by JSPS KAKENHI Grant Numbers JP21J10147 and JP24K16380, the Zengin Foundation for Studies on Economics and Finance and the K. Matsushita Foundation. We would like to thank the anonymous reviewers, the editors of this journal and the participants of the Asian and Australasian Society of Labour Economics 2024 Conference for their insightful comments. We are grateful for the support of Mitsuru Iwaki, Guenwoo Lee, Rogerson Dennis R. Fernandez, Ayu Pratiwi, Nguyễn Thị Thanh Huyền, Lingfeng Zheng, Win Sithu Maung and Daw Wai Soe Zin in editing the questionnaire for this study. We would also like to thank the respondents for taking the time to complete our survey. An earlier version of this article was circulated as JICA Ogata Research Institute Discussion Paper No. 23.

Competing interests

The authors declare that they have no competing interests.

Notes

In 2023, temporary labour migration to OECD countries continued to grow, and more than 2.4 million migrants were granted temporary work permits and authorizations in OECD countries (excluding Poland) (OECD 2024). ⮭
According to the WHO (2024, 14–15) definition, occupational credentials or qualifications under regulatory mechanisms take several forms such as certification, registration, qualification recognition and licensing. Certification is “a mechanism whereby the regulator certifies health practitioners for certain tasks or specialties based on qualification and/or demonstration of competence”. Registration places providers’ names on qualified service lists, entitling them to use reserved professional titles. Qualification recognition “identifies the similarities (and differences) between countries or jurisdictions in training and learning programmes and/or competency requirements for specific occupations”. Licensing restricts practice and specified activities to licence holders beyond professional title use. However, definitions vary across occupations and countries. We use “credentials” or “qualifications” for general regulation forms, changing terminology based on occupational qualification characteristics. ⮭
While the definition and concept of the term “care” are broad (King-Dejardin 2019), in this article the term “care” generally refers to elderly care, which aims to meet the special needs that are unique to senior citizens. ⮭
The term “large” is relative to the term “very large”, which Dustmann and Görlach also discuss and is associated with permanent migration. ⮭
Dustmann and Görlach write: “It is also worth noting that here we have abstracted from individual heterogeneity” (2016, 114). ⮭
King-Dejardin (2019) also notes that care industries in most host countries do not recognize the occupational qualifications of care workers or nurses that are issued in their home countries, and that foreign workers with specific skills are treated as auxiliary workers. In other words, there is an overqualification phenomenon (Shutes and Chiatti 2012). ⮭
The explanation of the data and their collection is largely drawn from Nakamura and Suzuki (2024), which uses the same data. ⮭
The sampling methods are detailed in the supplementary online appendices. ⮭
We conducted surveys through digital platforms to reduce measurement error and operational costs, while also enabling contact with FCWs during the COVID-19 pandemic. As our target participants were relatively younger (73.6–76.3 per cent of FCWs are in their twenties according to data published by Mitsubishi UFJ Research and Consulting), we anticipated with a high level of digital literacy, including frequent use of smartphones and computers. ⮭
We asked respondents whether they completed the survey without any assistance, to which 171 respondents answered yes. Analysis using only these 171 responses yielded consistent results, which are available upon request. ⮭
Some 464 people accessed the survey, of whom only 200 submitted responses that were deemed valid. It is not clear why many respondents exited the survey without completing it, for example whether it was due to the difficulty of the questions or poor internet connection. ⮭
Although micro-level national survey data on migrant workers in Japan are not available, Mitsubishi UFJ Research and Consulting Co. conducted large-scale national surveys of FCWs with support and funding from the Ministry of Health, Labour and Welfare. The data from our sample are very similar to those reported in these large-scale surveys and can be found in the supplementary online appendices. ⮭
Regarding household income left behind, some respondents did not know the exact amount, resulting in missing values. For respondents from Viet Nam, Indonesia, the Philippines, Myanmar and China, we substituted missing values using the sample averages of each country for each period. As for the respondents from Nepal, Sri Lanka, the Republic of Korea and Kyrgyzstan, where there were fewer observations, we used annual household income per capita data produced from CEIC (https://www.ceicdata.com/en/indicator/annual-household-income-per-capita). Respondents from other countries had no missing values. ⮭
We used the purchasing power parity conversion factor for private consumption (local currency units per international US$) from the World Bank’s DataBank (https://data.worldbank.org/indicator/PA.NUS.PRVT.PP) to calculate each income value. ⮭
See notes to table 2 for detailed information on the job-perceived score. ⮭
We classified the prefectures in three metropolitan areas – Tokyo (Tokyo, Saitama, Chiba and Kanagawa), Osaka (Osaka and Hyogo) and Nagoya (Aichi) – as “metropolis”. ⮭
As some outliers or measurement errors were identified in the educational attainment, age at entry and income variables, we excluded respondents who reported fewer than nine years of schooling, entered into Japan before the age of 18 or earned less than JPY 100,000. (Nineteen respondents were excluded under these criteria.) We then re-estimated the same model. The results remain informative and are available in the supplementary online appendices. ⮭
For example, a Sri Lankan SSW visa holder working in Okinawa Prefecture and an Indonesian TITP visa holder working in Nagano Prefecture passed the CCW exam in 2022. ⮭

References

Abarcar, Paolo, and Caroline Theoharides. 2024. “Medical Worker Migration and Origin-Country Human Capital: Evidence from U.S. Visa Policy”. The Review of Economics and Statistics 106 (1): 20–35. http://doi.org/10.1162/rest_a_01131.

Appleton, Simon, W. John Morgan, and Amanda Sives. 2006. “Should Teachers Stay at Home? The Impact of International Teacher Mobility”. Journal of International Development 18 (6): 771–786. http://doi.org/10.1002/jid.1313.

Asis, Edward, and Rogie Royce Carandang. 2020. “The Plight of Migrant Care Workers in Japan: A Qualitative Study of their Stressors on Caregiving”. Journal of Migration and Health 1–2: Article No. 100001. http://doi.org/10.1016/j.jmh.2020.100001.

Batista, Catia, Daniel Han, Johannes Haushofer, Gaurav Khanna, David McKenzie, Ahmed Mushfiq Mobarak, Caroline Theoharides, and Dean Yang. 2025. “Brain Drain or Brain Gain? Effects of High-Skilled International Emigration on Origin Countries”. Science 388 (6749): Article No. eadr8861. http://doi.org/10.1126/science.adr8861.

Bekaert, Els, Amelie F. Constant, Killian Foubert, and Ilse Ruyssen. 2024. “Longing for which Home: A Global Analysis of the Determinants of Aspirations to Stay, Return or Migrate Onwards”. Journal of Economic Behavior and Organization 219 (March): 564–587. http://doi.org/10.1016/j.jebo.2024.02.001.

Bijwaard, Govert E., Christian Schluter, and Jackline Wahba. 2014. “The Impact of Labor Market Dynamics on the Return Migration of Immigrants”. The Review of Economics and Statistics 96 (3): 483–494. http://doi.org/10.1162/REST_a_00389.

Bijwaard, Govert E., and Jackline Wahba. 2014. “Do High-Income or Low-Income Immigrants Leave Faster?” Journal of Development Economics 108 (May): 54–68. http://doi.org/10.1016/j.jdeveco.2014.01.006.

Blair, Peter Q., and Bobby W. Chung. 2021. “A Model of Occupational Licensing and Statistical Discrimination”. AEA Papers and Proceedings 111 (May): 201–205. http://doi.org/10.1257/pandp.20211112.

Blair, Peter Q., and Bobby W. Chung. 2025. “Job Market Signaling through Occupational Licensing”. The Review of Economics and Statistics 107 (2): 338–354. http://doi.org/10.1162/rest_a_01265.

Borjas, George J., and Bernt Bratsberg. 1996. “Who Leaves? The Outmigration of the Foreign-Born”. The Review of Economics and Statistics 78 (1): 165–176. http://doi.org/10.2307/2109856.

Brown, Xin. 2023. “Labor Market Impacts of State-Level Occupational Licensing of Undocumented Immigrants”. Research in Economics 77 (4): 478–496. http://doi.org/10.1016/j.rie.2023.08.001.

Brücker, Herbert, Albrecht Glitz, Adrian Lerche, and Agnese Romiti. 2021. “Occupational Recognition and Immigrant Labor Market Outcomes”. Journal of Labor Economics 39 (2): 497–525. http://doi.org/10.1086/710702.

Budig, Michelle J., and Joya Misra. 2010. “How Care-Work Employment Shapes Earnings in Cross-National Perspective”. International Labour Review 149 (4): 441–460. http://doi.org/10.1111/j.1564-913X.2010.00097.x.

Cassidy, Hugh, and Tennecia Dacass. 2021. “Occupational Licensing and Immigrants”. The Journal of Law and Economics 64 (1): 1–28. http://doi.org/10.1086/709834.

Chiswick, Barry R., and Paul W. Miller. 2009. “The International Transferability of Immigrants’ Human Capital”. Economics of Education Review 28 (2): 162–169. http://doi.org/10.1016/j.econedurev.2008.07.002.

Chung, Bobby W. 2024. “Effects of Occupational License Access on Undocumented Immigrants Evidence from the California Reform”. Journal of Regulatory Economics 65 (1–3): 64–83. http://doi.org/10.1007/s11149-023-09468-9.

Constant, Amelie, and Douglas S. Massey. 2003. “Self-Selection, Earnings, and Out-Migration: A Longitudinal Study of Immigrants to Germany”. Journal of Population Economics 16 (4): 631–653. http://doi.org/10.1007/s00148-003-0168-8.

Crescenzi, Riccardo, Nancy Holman, and Enrico Orru’. 2017. “Why Do They Return? Beyond the Economic Drivers of Graduate Return Migration”. The Annals of Regional Science 59 (3): 603–627. http://doi.org/10.1007/s00168-016-0762-9.

de Haas, Hein, Tineke Fokkema, and Mohamed Fassi Fihri. 2015. “Return Migration as Failure or Success? The Determinants of Return Migration Intentions Among Moroccan Migrants in Europe”. Journal of International Migration and Integration 16 (2): 415–429. http://doi.org/10.1007/s12134-014-0344-6.

Deyo, Darwyyn, and Alicia Plemmons. 2022. “Have License, Will Travel: Measuring the Effects of Universal Licensing Recognition on Mobility”. Economics Letters 219 (October): Article No. 110800. http://doi.org/10.1016/j.econlet.2022.110800.

Dill, Janette, Jennifer Craft Morgan, Jane Van Heuvelen, and Meredith Gingold. 2022. “Professional Certification and Earnings of Health Care Workers in Low Social Closure Occupations”. Social Science & Medicine 303 (June): Article No. 115000. http://doi.org/10.1016/j.socscimed.2022.115000.

Domingues Dos Santos, Manon, and Fabien Postel-Vinay. 2003. “Migration as a Source of Growth: The Perspective of a Developing Country”. Journal of Population Economics 16 (1): 161–175. http://doi.org/10.1007/s001480100117.

Dustmann, Christian, Itzhak Fadlon, and Yoram Weiss. 2011. “Return Migration, Human Capital Accumulation and the Brain Drain”. Journal of Development Economics 95 (1): 58–67. http://doi.org/10.1016/j.jdeveco.2010.04.006.

Dustmann, Christian, and Joseph-Simon Görlach. 2016. “The Economics of Temporary Migrations”. Journal of Economic Literature 54 (1): 98–136. http://doi.org/10.1257/jel.54.1.98.

Dustmann, Christian, and Anna Okatenko. 2014. “Out-Migration, Wealth Constraints, and the Quality of Local Amenities”. Journal of Development Economics 110 (September): 52–63. http://doi.org/10.1016/j.jdeveco.2014.05.008.

Efendi, Ferry, Lisa McKenna, Sonia Reisenhofer, Anna Kurniati, and Eka Misbhahatul Marah Has. 2021. “Experiences of Healthcare Worker Returnees in Their Home Countries: A Scoping Review”. Journal of Multidisciplinary Healthcare 14: 2217–2227. http://doi.org/10.2147/JMDH.S321963.

Efendi, Ferry, Nursalam Nursalam, Elida Ulfiana, and Rista Fauziningtyas. 2019. “Situational Analysis of Career Choices among Indonesian Nurses Returnees”. Indian Journal of Public Health Research & Development 10 (2): 296–299. http://doi.org/10.5958/0976-5506.2019.00305.X.

England, Paula, Michelle Budig, and Nancy Folbre. 2002. “Wages of Virtue: The Relative Pay of Care Work”. Social Problems 49 (4): 455–473. http://doi.org/10.1525/sp.2002.49.4.455.

Federman, Maya N., David E. Harrington, and Kathy J. Krynski. 2006. “The Impact of State Licensing Regulations on Low-Skilled Immigrants: The Case of Vietnamese Manicurists”. The American Economic Review 96 (2): 237–241. http://doi.org/10.1257/000282806777211630.

Fu, Rong, Haruko Noguchi, Akira Kawamura, Hideto Takahashi, and Nanako Tamiya. 2017. “Spillover Effect of Japanese Long-Term Care Insurance as an Employment Promotion Policy for Family Caregivers”. Journal of Health Economics 56 (December): 103–112. http://doi.org/10.1016/j.jhealeco.2017.09.011.

Gibson, John, and David McKenzie. 2011. “The Microeconomic Determinants of Emigration and Return Migration of the Best and Brightest: Evidence from the Pacific”. Journal of Development Economics 95 (1): 18–29. http://doi.org/10.1016/j.jdeveco.2009.11.002.

Gomez, Rafael, Morley Gunderson, Xiaoyu Huang, and Tingting Zhang. 2015. “Do Immigrants Gain or Lose by Occupational Licensing?” Canadian Public Policy 41 (1): S80–S97. http://doi.org/10.3138/cpp.2014-028.

Gould, Eric D., and Omer Moav. 2016. “Does High Inequality Attract High Skilled Immigrants?” Economic Journal 126 (593): 1055–1091. http://doi.org/10.1111/ecoj.12185.

Hirano, Yuko O., Kunio Tsubota, and Shun Ohno. 2020. “Factors Associated with the Recruitment of Foreign Nurses in Japan: A Nationwide Study of Hospitals”. Human Resources for Health 18 (1): Article No. 88. http://doi.org/10.1186/s12960-020-00532-5.

Huynh, Nhien Thi Thuy, Thi Dung Le, Happy Indri Hapsari, Hua-Tsen Hsiao, Mei-Chih Huang, and Chi-Yin Kao. 2024. “The Experiences of Migrant Care Workers in Long-Term Care Facilities: A Scoping Review”. Journal of Immigrant and Minority Health 26 (5): 936–944. http://doi.org/10.1007/s10903-024-01618-1.

ILO. 2018. Care Work and Care Jobs for the Future of Decent Work. Geneva.

ILO. 2024. ILO Global Estimates on International Migrant Workers: International Migrants in the Labour Force. 4th ed. Geneva.

Immigration Services Agency of Japan. 2024. “Advisory Panel of Experts on Ideal Form of Technical Intern Training Program and Specified Skilled Worker System Final Report”. Tokyo: Immigration Services Agency of Japan.

IOM (International Organization for Migration). 2024. World Migration Report 2024. Geneva.

Japan, Ministry of Health, Labour and Welfare. 2019. Guidebook for Care Service Providers on Employment of Foreign Care Workers. https://www.mhlw.go.jp/content/12000000/000526602.pdf.

The Japan News. 2023. “Japan Weighs Future of Contentious Foreign Trainee Program”. 14 October 2023. https://japannews.yomiuri.co.jp/society/general-news/20231014-142807/.

The Japan News. 2024. “Japan to Actively Recruit Nursing Care Staff from Overseas; Government Plans to Subsidize Costs of Finding Workers”. 17 December 2024. https://japannews.yomiuri.co.jp/politics/politics-government/20241217-228221/.

Johnson, Janna E., and Morris M. Kleiner. 2020. “Is Occupational Licensing a Barrier to Interstate Migration?” American Economic Journal: Economic Policy 12 (3): 347–373. http://doi.org/10.1257/POL.20170704.

Kakizoe, Tadao. 2023. “Long-Term Care Insurance Increasingly Insecure”. The Japan News. 11 August 2023. https://japannews.yomiuri.co.jp/editorial/insights-world/20230811-128935/.

Kameyama, Junko, Yumi Hashizume, Yuko Takamura, Shoko Nomura, Tomoki Gomi, and Hisako Yanagi. 2022. “Work Engagement, Well-Being, and Intent to Continue Working Based on Educational Support among Foreign Care Workers in Japan”. Environmental Health and Preventive Medicine 27 (1):1–9. http://doi.org/10.1265/ehpm.21-00248.

Kawaguchi, Daiji, Tetsushi Murao, and Ryo Kambayashi. 2014. “Incidence of Strict Quality Standards: Protection of Consumers or Windfall for Professionals?” The Journal of Law and Economics 57 (1): 195–224. http://doi.org/10.1086/674867.

Kim, Jeounghee. 2020. “Occupational Credentials and Job Qualities of Direct Care Workers: Implications for Labor Shortages”. Journal of Labor Research 41 (4): 403–420. http://doi.org/10.1007/s12122-020-09312-5.

King, Elizabeth M., Hannah L. Randolph, Maria S. Floro, and Jooyeoun Suh. 2021. “Demographic, Health, and Economic Transitions and the Future Care Burden”. World Development 140 (April): Article No. 105371. http://doi.org/10.1016/J.WORLDDEV.2020.105371.

King-Dejardin, Amelita. 2019. The Social Construction of Migrant Work: At the Intersection of Care, Migration and Gender. Geneva: ILO.

Kleiner, Morris M, and Alan B. Krueger. 2010. “The Prevalence and Effects of Occupational Licensing”. British Journal of Industrial Relations 48 (4): 676–687. http://doi.org/10.1111/j.1467-8543.2010.00807.x.

Kleiner, Morris M, and Alan B. Krueger. 2013. “Analyzing the Extent and Influence of Occupational Licensing on the Labor Market”. Journal of Labor Economics 31 (2): S173–S202. http://doi.org/10.1086/669060.

Kleiner, Morris M., and Ming Xu. 2025. “Occupational Licensing and Labor Market Fluidity”. Journal of Labor Economics 43 (3): 937–983. http://doi.org/10.1086/730120.

Koumenta, Maria, Mario Pagliero, and Davud Rostam-Afschar. 2022. “Occupational Regulation, Institutions, and Migrants’ Labor Market Outcomes”. Labour Economics 79 (December): Article No. 102250. http://doi.org/10.1016/j.labeco.2022.102250.

Kugler, Adriana D., and Robert M. Sauer. 2005. “Doctors without Borders? Relicensing Requirements and Negative Selection in the Market for Physicians”. Journal of Labor Economics 23 (3): 437–465. http://doi.org/10.1086/430283.

Kurniati, A., C.-M. Chen, F. Efendi, and R. Ogawa. 2017. “A Deskilling and Challenging Journey: The Lived Experience of Indonesian Nurse Returnees”. International Nursing Review 64 (4): 494–501. http://doi.org/10.1111/inr.12352.

Lightman, Naomi. 2017. “Discounted Labour? Disaggregating Care Work in Comparative Perspective”. International Labour Review 156 (2): 243–267. http://doi.org/10.1111/ilr.12001.

Manski, Charles F. 1990. “The Use of Intentions Data to Predict Behavior: A Best-Case Analysis”. Journal of the American Statistical Association 85 (412): 934–940. http://doi.org/10.1080/01621459.1990.10474964.

Morikawa, Masayuki. 2018. “Occupational Licenses and Labor Market Outcomes in Japan”. Japan and the World Economy 48 (December): 45–56. http://doi.org/10.1016/j.japwor.2018.07.002.

Mulholland, Sean E., and Andrew T. Young. 2016. “Occupational Licensing and Interstate Migration”. Cato Journal 36 (1): 17–31.

Nakamura, Jiro. 2010. “Impacts of International Migration on the Labor Market in Japan”. Japan Labor Review 7 (3): 68–85.

Nakamura, Nobuyuki, and Aya Suzuki. 2024. “How Altruism Works during a Pandemic: Examining the Roles of Financial Support and Degrees of Individual Altruism on International Remittance”. The World Economy 47 (9): 3929–3965. http://doi.org/10.1111/twec.13616.

OECD (Organisation for Economic Cooperation and Development). 2024. International Migration Outlook 2024. Paris.

Oppenheimer, Daniel M., Tom Meyvis, and Nicolas Davidenko. 2009. “Instructional Manipulation Checks: Detecting Satisficing to Increase Statistical Power”. Journal of Experimental Social Psychology 45 (4): 867–872. http://doi.org/10.1016/J.JESP.2009.03.009.

Pastor-Bravo, M., and S. Nelson. 2019. “Migration of Latin American Nurses to Spain 2006–2016: A Case Study”. International Nursing Review 66 (2): 183–190. http://doi.org/10.1111/inr.12511.

Patt, Alexander, Jens Ruhose, Simon Wiederhold, and Miguel Flores. 2021. “International Emigrant Selection on Occupational Skills”. Journal of the European Economic Association 19 (2): 1249–1298. http://doi.org/10.1093/jeea/jvaa032.

Pecoraro, Marco, and Massimiliano Tani. 2023. “Does Certifying Foreign Qualifications Lead to Better Immigrant Skills Utilization?” Social Indicators Research 170 (1): 291–322. http://doi.org/10.1007/s11205-023-03069-x.

Peterson, Brenton D., Sonal S. Pandya, and David Leblang. 2014. “Doctors with Borders: Occupational Licensing as an Implicit Barrier to High Skill Migration”. Public Choice 160 (1–2): 45–63. http://doi.org/10.1007/s11127-014-0152-8.

Ramseyer, J. Mark, and Eric B. Rasmusen. 2015. “Lowering the Bar to Raise the Bar: Licensing Difficulty and Attorney Quality in Japan”. Journal of Japanese Studies 41 (1): 113–142. http://doi.org/10.1353/jjs.2015.0020.

Razavi, Shahra, and Silke Staab. 2010. “Underpaid and Overworked: A Cross-National Perspective on Care Workers”. International Labour Review 149 (4): 407–422. http://doi.org/10.1111/j.1564-913X.2010.00095.x.

Reagan, Patricia B., and Randall J. Olsen. 2000. “You Can Go Home Again: Evidence from Longitudinal Data”. Demography 37 (3): 339–350. http://doi.org/10.2307/2648046.

Redbird, Beth. 2017. “The New Closed Shop? The Economic and Structural Effects of Occupational Licensure”. American Sociological Review 82 (3): 600–624. http://doi.org/10.1177/0003122417706463.

Rooth, Dan-Olof, and Jan Saarela. 2007. “Selection in Migration and Return Migration: Evidence from Micro Data”. Economics Letters 94 (1): 90–95. http://doi.org/10.1016/j.econlet.2006.08.006.

Sabio, Gianne Sheena, Kritika Pandey, and Rhacel Salazar Parreñas. 2022. “Global Care Chains”. In Routledge Handbook of Immigration and Refugee Studies, 2nd ed., edited by Anna Triandafyllidou, 172–183. Abingdon: Routledge.

Shutes, Isabel, and Carlos Chiatti. 2012. “Migrant Labour and the Marketisation of Care for Older People: The Employment of Migrant Care Workers by Families and Service Providers”. Journal of European Social Policy 22 (4): 392–405. http://doi.org/10.1177/0958928712449773.

Spence, Michael. 1973. “Job Market Signaling”. The Quarterly Journal of Economics 87 (3): 355–374. http://doi.org/10.2307/1882010.

Sugawara, Shinya, and Jiro Nakamura. 2014. “Can Formal Elderly Care Stimulate Female Labor Supply? The Japanese Experience”. Journal of the Japanese and International Economies 34 (December): 98–115. http://doi.org/10.1016/j.jjie.2014.05.006.

Sweetman, Arthur, James Ted McDonald, and Lesleyanne Hawthorne. 2015. “Occupational Regulation and Foreign Qualification Recognition: An Overview”. Canadian Public Policy 41 (1): S1–S13. http://doi.org/10.3138/cpp.41.s1.s1.

Tani, Massimiliano. 2021. “Occupational Licensing and the Skills Mismatch of Highly Educated Migrants”. British Journal of Industrial Relations 59 (3): 730–756. http://doi.org/10.1111/bjir.12574.

Tani, Massimiliano. 2023. “Occupational Licensing and Skills Mismatches among Immigrants and Natives in the United States”. The B.E. Journal of Economic Analysis & Policy 23 (3): 673–700. http://doi.org/10.1515/bejeap-2022-0242.

Tjaden, Jasper, Daniel Auer, and Frank Laczko. 2019. “Linking Migration Intentions with Flows: Evidence and Potential Use”. International Migration 57 (1): 36–57. http://doi.org/10.1111/imig.12502.

Tomohara, Akinori. 2022. “Does an Increase in the Number of Immigrant Workers Reduce Unemployment Rates? An Industry Sector Unit Level Analysis in Japan”. Economic Analysis and Policy 74 (June): 464–476. http://doi.org/10.1016/j.eap.2022.03.002.

Wahba, Jackline. 2014. “Return Migration and Economic Development”. In International Handbook on Migration and Economic Development, edited by Robert E. B. Lucas, 327–349. Cheltenham: Edward Elgar.

Wahba, Jackline. 2021. “Who Benefits from Return Migration to Developing Countries?” IZA World of Labor February: 1–10. http://doi.org/10.15185/izawol.123.v2.

Walton-Roberts, Margaret. 2020. “Occupational (Im)Mobility in the Global Care Economy: The Case of Foreign-Trained Nurses in the Canadian Context”. Journal of Ethnic and Migration Studies 46 (16): 3441–3456. http://doi.org/10.1080/1369183X.2019.1592397.

WHO (World Health Organization). 2024. Health Practitioner Regulation: Design, Reform and Implementation Guidance. Geneva.

World Bank. 2023. World Development Report 2023: Migrants, Refugees, and Societies. Washington, DC.