Migration, gender and economic integration: international scholarship (2006–09) and an Aotearoa New Zealand research agenda

Abstract

This paper follows from the findings of the 2006 review of research on women, gender and migration published in International Migration Review. We begin by discussing three international trends in contemporary migration flows: diversification, bifurcation and feminisation; and examine their significance for New Zealand. We then review the research on gender and economic integration of migrants in developed countries in relation to three aspects: the characteristics of migrants; the strategies migrants use during settlement; and the contexts of reception in receiving communities. We identify insights and omissions in this scholarship relevant to New Zealand policy-oriented migration research. We argue that the gendered nature of migration cannot be ignored, and that while human capital approaches to economic integration are important, they are insufficient for understanding complex migrant outcomes. Rather, comprehensive, integrated and local research is required to understand migrant experience and outcomes and to assess the effectiveness of immigration policy settings.

Keywords:

• economic integration
• gender
• migration
• New Zealand
• policy implications

Introduction

Although migration has long been an important part of human history (Castles & Miller 2009), an increase in the volume and significance of migration flows has made international migration a defining characteristic of the late twentieth and early twenty-first centuries (Piper 2008b). Global economic inequality and demographic imbalances between the aging populations of the developed industrialised countries (including New Zealand) and large numbers of working age people in many of the less industrialised countries of Asia, Africa, and South and Central America mean that, in the longer term, high levels of international migration are likely to continue (Castles 2009; OECD 2009), despite the decreased levels of mobility, declining migrant quotas and greater protectionism associated with the current global economic crisis (Papademetriou & Terrazas 2009). The United Nations estimates that the number of migrants worldwide reached 191 million in 2005, more than doubling since the 1970s (UNDESA 2005), and predicts that by mid-2010 there will be just under 214 million migrants across the globe (United Nations Population Division 2009). Alongside this proliferation of migrants is a burgeoning literature researching the drivers, flows and outcomes of migration at local and global levels.

In 2006, the International Migration Review (IMR) produced a special issue on gender and migration that reviewed, by selected academic discipline,1 the ‘veritable tidal wave’ of scholarly work on women, gender and migration published since the advent of second wave feminism in the 1970s (Donato et al. 2006, p. 7). This ‘tidal wave’ comprises three distinct phases: the studies that incorporated gender by including the variable of sex in data collection or focused exclusively on female migrants; the research that examined the household economy as a critical site for revealing the relationship between women and migration; and work that considers gender a constitutive element of the entire migration process (Curran et al. 2006; Donato et al. 2006; Mahler & Pessar 2006). The latter was the particular concern of the IMR special issue, in which the editors argued that ‘[i]ncreasingly, the entire migration process is perceived as a gendered phenomenon’ (Donato et al. 2006, p. 6). Gender analysis, the editors argued, adds theoretical value and new insights into the processes and impacts of migration. Even so, gender analysis remains underdeveloped in migration research in Aotearoa New Zealand (see, for example, Badkar et al. 2007a, 2007b). The aim of this paper then is to look forward at the most recent international literature in the gender and migration field, with an eye to what we can learn to apply to future New Zealand research.

Aotearoa New Zealand, as a relatively small island state in the south Pacific, both exemplifies and confounds a number of trends in international migration patterns and pathways (Badkar et al. 2007a; Callister et al. 2006). As a white settler jurisdiction, New Zealand more closely fits the demographic and economic profile of a ‘developed’ than ‘less developed’ migration destination. Uncharacteristically for many developed countries, however, New Zealand experiences a ‘reverse diaspora’ where ‘people born in New Zealand migrate to live elsewhere, but people born elsewhere also migrate to live in New Zealand’ (Bryant & Law 2004, p. 5). New Zealand's high levels of out-migration as well as the increased numbers of permanent long-term arrivals since the late 1980s fits the general trend towards growing numbers of migrants worldwide (Castles & Miller 2009; see also Gibson & McKenzie 2009).

Within this context, the objective of New Zealand's immigration policy is ‘to contribute to economic growth through enhancing the overall level of human capability in New Zealand, encouraging enterprise and innovation, and fostering international links, while maintaining a high level of social cohesion’ (Moody 2006, p. 3). Our focus in this review on the interaction of gender and economic integration reflects these local policy settings. The review is part of, and reflects upon, a programme of research that seeks to describe and understand the migration landscapes created by these policy settings. Specifically, it was instigated to provide a context for work being undertaken by the Integration of Immigrants Programme (IIP), a five-year research programme funded by the New Zealand Foundation for Research, Science and Technology.2 This policy-oriented focus leads us away from more theoretical literature on gender and migration, such as postcolonial, poststructural or feminist readings of migration's effects; or work that challenges the politics of migration in relation to issues of identity and belonging, civic participation and citizenship, health and well being, family relationships, or other dimensions of social life. Nonetheless, our concern with the neglect of gender in New Zealand migration research provides a platform for critical review, as does the pivotal part played by economic experience (labour, consumption, investment) in migrant's lives. Just as economic calculation is always simultaneously social and political, our review inevitably follows the economic lives of migrants into social realms.

The policy environment settings for migration in New Zealand set the tenor for much of the local scholarly engagement with migration issues. The literature has turned attention to policy concerns such as the demographics, numbers and mix of migrants and to such questions as: ‘are too many people coming or too few’; ‘do we have the right mix of skills and incentives for those skills’; ‘have we set the entry bar too high or too low’; and ‘is the language requirement too lax or too onerous’? In most of these debates and discussions, migrants are assumed to be genderless. If the policy will work for men, it is presumed to also work for women (and children).

Our common interest in ‘gender’ as a political issue leads us to question this gender neutrality and consider the gender implications of a number of questions in relation to the empirical work scheduled in the IIP. In particular, we were interested to know how the economic experience of migration to New Zealand differs for women and men; how migration differentially affects the careers and family lives of male and female migrants; and perhaps most importantly, how best to integrate a gendered focus into this New Zealand-specific project we are working on. It is from this position, and for a policy-centric audience, that we have undertaken this review.

This paper is divided into three sections. In the first section, we identify three emerging trends in contemporary migration flows in which both gender and economic integration issues are significant, linking these trends in the New Zealand context. The second section reviews the post-2006 literature on gender and economic integration, recognises that much of this literature is based on a ‘human capital’ approach and identifies, in particular, discussions about migrant characteristics, strategies and contexts of reception. In the final section, we consider the contours of a research agenda for New Zealand that calls for a comprehensive and integrated approach that instantiates gender analysis as a necessary component of policy-centric research.

Diversification, bifurcation and feminisation in migration flows

Diversification

The concept of diversification describes the complexity of contemporary migration flows. Migrants from a growing number of countries are moving in multiple directions and with increasingly different motivations (Bakewell 2009). Internationally, there is a trend towards greater differences within and between birthplace groups. Mexican migrants in the United States today, for example, are more diverse in terms of skill, region of origin and mode of entry than ever before (Rojas Wiesner & Angeles Cruz 2008). There is also greater diversity with respect to the temporal component of migration; migrants from the same source country are moving both permanently and temporarily to the same destination (Piper 2008b).

In the New Zealand context, a focus on skilled migration has facilitated the entry of growing numbers of migrants from an increasing range of countries, thus reflecting international diversification trends (Spoonley 2005; Larner 2006). In New Zealand, data show differences both within and between birthplace groups (Nana & Sanderson 2008, p. 19) and point to increased diversification in sending countries. Internationally, the number of sending and receiving nations has grown, as has the number of countries, such as New Zealand, that function as both in global migration circuits. Piper (2008b) suggests that temporary and circular migrations are increasingly significant in these flows.3 This temporal diversification has become part of the experience of traditional settler societies. In New Zealand (Khoo et al. 2008), Pacific immigration, for example, involves both permanent migrants (through the past provisions of policies such as the Special Samoan Quota Places and Pacific Access Categories) and temporary flows through the Recognised Seasonal Employer (RSE) programme (Department of Labour, n.d.). Nana and Sanderson (2008, p. 21) also note the pronounced patterns of re- or out-migration of migrants to New Zealand.

Bifurcation

Bifurcation (or polarisation) refers to the growing difference in the ease of international mobility between the highly skilled and the unskilled. Most OECD countries, for example, have introduced policies in recent years which facilitate the entry of the highly skilled (Chaloff & Lemaitre 2009), while attempting to limit the entry of lesser-skilled migrants. Such an approach, according to the International Migration Outlook (OECD 2009, p. 10), has ‘contributed to a climate in which irregular migration and employment have found fertile ground’. This bifurcation also occurs at regional or local as well as national levels.

Trends toward bifurcation are evident in New Zealand economic integration literature. In the New Zealand Residence Programme, 60% of places are reserved for the Skilled/Business stream, while unskilled workers are targeted via a range of temporary work visas. Income inequality is seen in New Zealand in the recruitment of highly gendered, skilled immigrant labour in both the health (Dumont et al. 2007; Zurn & Dumont 2008) and education (Tamásy 2008) sectors (see also North 2007). A number of studies that focus on the unskilled end of the spectrum also highlight the inequalities associated with bifurcation. Income inequality is the focus of Gibson et al.'s (2008) study of the pro-poor dimensions of seasonal work from Tonga; Lobo and Wilkinson's (2008) research on construction workers; and Badkar et al.'s (2009) study of care services. Information on the gendered implications of the temporary migration of less skilled cohorts is emerging through the research and evaluation of the Recognised Seasonal Employer Scheme (see, for example, Gibson et al. 2008; Ramasamy et al. 2008; Bailey 2009; J. Williams 2009).

From a border control and migrant entry perspective, Callister et al. (2009, p. 41) point out that New Zealand ‘can easily control its borders’ so issues of illegal migration into essential but low-paid and often gendered work is not a significant issue. McMillan (2008) intimates that New Zealand, in part because of its remote location, has relatively small volumes of irregular migration. J. Williams's (2009) exploration of low-skill temporary migration and human rights law begins to look at the legal implications of bifurcated inequalities in migration streams.

Feminisation

In terms of the feminisation of international migration, the share of women has risen in most, if not all, migration streams and in migration flows to all world regions (Piper 2008b, p. 3). At the turn of last century, women constituted 50% of global migrants (Pfeiffer et al. 2008); and in the Philippines, Indonesia and Sri Lanka the number of women emigrating now exceeds men (Oishi 2008, p. 55). Increasingly also, women are migrating independently of men and family and as skilled, economic migrants (Carling 2005). This trend towards the feminisation of migration has received considerable attention in the global policy-making community (ILO 2004; United Nations 2004; Oso Casas & Garson 2005; UNRISD 2005; UNFPA 2006; Dumont et al. 2007), and is likely to continue given the demand for cheap labour, the ageing of populations in wealthy nations and the increased participation of women in the labour markets of both sending and receiving countries (Ahmed 2008).

The feminisation of migration flows is less explicitly documented in New Zealand economic integration literature. Although we do not know as much about the implications of gender-differentiated migration, we do know that it occurs. Badkar et al. (2007a, p. 32) note that while there is a gender imbalance in the population of the working aged in favour of women, and that ‘preliminary research indicates some reasons behind female-dominated migration from particular countries (for example a large proportion of female nurses migrating from the Philippines that creates a strongly gendered flow from that country), little is still known about the drivers of gendered migration to New Zealand from specific source countries’. Callister et al. (2009, p. 41) explore the possibility of a ‘formal scheme’ (not unlike new Zealand's current registered seasonal employment (RSE) scheme for mostly male agricultural workers) for domestic worker and caregiver migration (likely to also be highly gendered). Work by Tipples (2006) on European Union migration to New Zealand and Dumont et al.'s (2007) work on the brain drain also address feminisation issues.

Overall, while gender is theorised as an important variable in much international research, it remains largely invisible in New Zealand economic migration research. As stated earlier, this is in part because the emphasis of much migration research in New Zealand is implicitly on migration policies, and in part because such policy-focused approaches are based on aggregate data at a national level. Gender is not routinely used as a variable in the analyses that are undertaken.

In the next section, we review some of the more recent findings from the international literature in terms of what it can tell us about emerging issues and relevant directions in relation to gender and the economic integration of migrants. The field of migration research internationally is extensive and tight parameters were needed to contain this review. As stated earlier, two of these parameters were set by our decision to focus solely on empirical research related to developed countries and, within that field, on research with a clear focus on the economic experiences and activities of migrant groups. Our focus was also restricted in other key ways. Firstly, we used Web of Science databases to identify the literature. While this was not an exhaustive approach, we considered it would produce a manageable and accessible record of current international work through which we could identify any emerging trends in gender and migration research. A combination of terms related to ‘employment’, ‘migration’ and ‘gender’ were used as key words, with strict date limits and data type specifications focusing the search. Beyond the limitations already mentioned, we excluded a significant number of papers exploring the employment of male migrants that made no use of gender in the analysis, and the extensive literature on migration and sex work (largely, but not exclusively focused on women). This final limitation was largely pragmatic. Although we recognise that the migration of sex workers is an issue in Aotearoa New Zealand, the dynamics and issues that structure this field differ from those across the rest of the workforce.4

Ultimately, some 95 research papers published in a wide range of journals from 2006 to early 2009 formed the corpus for our review. Additional sources from New Zealand and international agency literature (that were not located through the Web of Science) were also included to amplify particular points and provide the New Zealand focus. The international literature on the economics of gender and migration appeared to be organised around three key foci: how various migrant characteristics contribute to, or correlate with, particular economic outcomes; analysis of specific strategies migrants employ in their search for work and settlement; and how the context of reception in the receiving country constrains or facilitates migrant employment. In response to this finding, we have used these three themes to organise the material that follows in section two. Significantly, a focus on migrant characteristics is frequently stand-alone and this work comprises the first sub-section below. The second combines a focus on migrant strategies and features of the context of reception as these are more regularly researched in tandem.

Characteristics, strategies and the context of reception

Migrant characteristics

Under the overarching theme of economic integration, the focus on the impact of migrant characteristics on employment outcomes was the most common in the literature and comprised research almost exclusively reliant on quantitative methods. Migrant characteristics were often the single focus of these papers, although some looked at these in conjunction with the context of reception. Many of these papers reported on analyses of large-scale census or panel survey data, examining correlations between a range of characteristics and outcomes. Research from a range of Western, developed nations suggests that, while migrants do get a return on their human capital, it is less than it would be in their home country, less than the native-born population (for example, Adsera & Chiswick 2007) and women migrants tend to suffer the greatest disadvantage (see, for example, Clark & Huang 2006; Trzcinski 2006; Valdez 2006; Banerjee et al. 2007; Taylor 2007; Thrane 2007; Foroutan 2008; Rebhun 2008). Researchers focused on human capital characteristics at both the individual and aggregate (family) level.

Disaggregating components of ‘human capital’ reveals a more complex picture. There is evidence that both male and female migrants with higher education suffer fewer labour force participation and earnings deficits than those with lower education levels. However, gender differences remain within both these education cohorts, although the evidence is mixed. Amuedo-Dorantes and de la Rica's (2007) research on migrants in Spain, for example, found that men gained a higher return on education, while Adsera and Chiswick's (2007) European Union-wide study identified women as achieving a higher return overall. Some studies, such as Amuedo-Dorantes and de la Rica (2007), Foroutan (2008), Foroutan and McDonald (2008), Ozden and Neagu (2008) and Read and Cohen (2007) explore the gendered returns on education in specific nation states. Looking at women migrants in Australia, for example, Foroutan (2008) identified a lower return on education qualifications obtained in non-English-speaking countries; whereas Read & Cohen (2007, p. 1723) found that Puerto Rican women in the United States received a higher return on their college degrees than white or Asian women migrants. Suto's (2009) qualitative study reiterated the ‘compromised careers’ (p. 423) of highly skilled women migrants to Canada.

Language skill, regardless of educational achievement, is another key ‘human capital’ factor in gender and migration research, with non-native speakers or those with less language skill having lower labour force participation and earnings (see Foroutan 2008; Foroutan & McDonald 2008; Adsera & Chiswick 2007). Adsera and Chiswick (2007), Hamilton et al. (2008) and Mora and Davila (2006) all found that men suffered less from limited language skills than women. A number of studies pointed to a convergence between the labour force participation and wage rates of migrants and natives over time (Adsera & Chiswick 2007; Amuedo-Dorantes & de la Rica 2007; Foroutan 2008; Foroutan & McDonald 2008). Valdez's (2006) study of Mexicans in the US Southwest showed a somewhat different picture, however, with the more highly skilled having improved incomes over time, while the low-skilled earned less over time (‘downward assimilation’). Again, the evidence in these studies suggests that women fare less well than men.

In sum, this research links individual characteristics, such as education level, country of origin and language skill, to economic outcomes. The evidence presented in these studies indicates that there are gendered differences in outcomes, with women by and large accruing less advantage from their individual characteristics than men (see Boyd 2006, for a detailed breakdown of individual gendered characteristics).

Moving beyond the characteristics of individual migrants to those of migrant families, research suggests that parenthood and childcare responsibilities impact strongly on the experiences of women migrants. Andersson and Scott's (2007) comparison of the labour force participation of native-born and migrant populations governed by the Swedish welfare state is the exception here, with little difference being identified between men and women of both groups. Beyond the Swedish example, however, there are country of origin differences reported in the literature in relation to labour force participation rates and earnings of women with young children (Bevelander & Groeneveld 2006; Clark & Huang 2006; Adsera & Chiswick 2007). Foroutan (2008) reports a higher participation rate amongst Asian women with young children than other women (although Foroutan and McDonald 2008 report a lower participation rate for Asian women overall). In the United Kingdom, Dale et al. (2006) note the lower participation rates of Pakistani and Bangladeshi mothers compared to those from the Caribbean.

Bevelander and Groeneveld (2006), p. 800) suggest that a focus on the migrants themselves leads to explanations for these differences which centre on gender role beliefs from the country of origin (see also Salway 2007). Ho (2006), however, points to the limitations of this ‘migrant-only’ focus and suggests that such analyses may hide the impact of gender-segregated labour markets and a gendered division of paid and domestic work in explaining women's labour market outcomes. Ho's (2006) study of the experience of Chinese women in Australia shows how the move from China and Hong Kong results in a ‘feminisation’ of roles for many Chinese women, as the lack of affordable childcare in Australia combines with gender role expectations (both Chinese and Australian), resulting in the women sacrificing their careers to remain at home with children (see also Chiang 2008 on Taiwanese women in Canada; Muszynska and Kulu 2007 on the link between migrant wives’ labour force participation and relationship breakdown in Russia).

Another family characteristic that impacts on co-habiting women's labour force participation rates is their spouse's income. An unanticipated finding of Foroutan's (2008) study was that women whose spouses have high incomes are more likely to work than those whose spouses have low incomes, although determining why this should be the case was beyond the scope of Foroutan's study. The complexities of spousal income, childcare and work opportunities are also spelled out in Zaiceva's (2010) study on women in Germany. While not strictly research on international migration (it looks at migration within post-unification Germany between East and West), Zaiceva identifies the tensions that exist between labour market and family models in explaining migration outcomes.

On the whole, this body of research points to the general underutilisation of migrant human capital and to women's greater disadvantage in the labour market. We would argue, on the strength of this and earlier literature, that such human capital approaches to studying the economics of migration are limited in at least four ways. First, many researchers in human capital studies cannot adequately explain the differences in economic outcomes for male and female migrant workers (for example Stone & McQuillan 2007; and Zimmermann 2007). Brekke and Mastekaasta's (2008) study of native and migrant employment outcomes in Norway took the economic income gap as their starting point and, in their study of highly skilled migrants who had been educated in Norwegian universities, held human capital characteristics constant across their sample. The results showed complex native/migrant and gendered differences in labour force participation and earnings, including unexpected evidence of increasing disadvantage in earnings over time for male migrants. Whereas explanations for such differences often focus on migrant characteristics, Brekke and Mastekaasta's approach exposed a range of possible contributing factors including the cumulative effects of small differences in human capital over time, differences in social networks or differential experiences of discrimination, as well as the impact of the gendered division of labour on migrant women's labour force participation (Brekke & Mastekaasta 2008, pp. 520–521; see also Hall et al. 2010).

Brekke and Mastekaasta, along with Ho (above), focus on the interaction of human capital characteristics and aspects of the context of reception. Another good example of such work is Salaff and Greve's (2006) study of 50 Chinese couples in Canada over the first 2.5 to five years post-migration. They found that human capital theory alone provided imperfect explanations for migrants’ employment outcomes. It could not, for example, explain why migrants with masters-level qualifications did worse than those with lower education (Salaff & Greve 2006, p. 96). Nor could it explain why women engineers failed to find appropriate-level employment while some of their male compatriots did (Salaff & Greve 2006, p. 98). In this instance the explanation lies in the gendered structure of the Canadian labour market, in which women engineers are largely excluded. On the whole, their study points to the failure of human capital theory to predict migrant economic outcomes (see also Suto 2009). The reality of human capital is that it is to some degree culture-, profession- and gender-specific, and what is recognised in one culture, profession or gender is not identical to that recognised in another (Salaff & Greve 2006, pp. 87–88).

Second, the human capital approach focuses on migrants as individuals. Such a focus typically ignores or downplays the socially embedded nature of the individual and their migration trajectories. Here we emphasise the value of Williams's recent conceptualisation of migratory movements as ‘enfolded mobilities’, which seeks to capture the way that ‘individual migrations are directly enfolded with those of other individuals, either through associated or contingent movements, or through consequential migration at later stages in the life course’ (A.M. Williams 2009, p. 309). Thirdly, and relatedly, in addition to treating migrants as individuals, human capital theory treats them solely as economic actors—or more accurately, resources, since their actions (strategies) are also ignored by this approach. In reality, the evidence points to a mix of reasons and factors that influence individual and family decisions to migrate (see, for example Kobayashi & Preston 2007, p. 153; Asis 2008; N. Williams 2009) and then influence the outcomes of migration.

Finally, much of this literature has a focus on an ‘attribute deficit’ model in which an exclusive focus on migrant ‘attributes’ results in arguments for more highly skilled migrants and/or migrant ‘up skilling’ (Docquier et al. 2009, p. 2). Individual migrants and their families are seen as the ‘problem’ in need of improvement, again resulting in a lack of attention being given to a wide range of significant factors that contribute to migrant outcomes, some of which are raised in the following section.

Consequently, we argue that exploration of the gendered outcomes of migration requires research that, in addition to being concerned with the individual and family characteristics of migrants, investigates the impacts of migrant strategies as well as features of the ‘context of reception’, such as the gendered labour market, employer hiring practices, the shape of immigration policy, the geographical dimensions of migrant settlement and employment, the gendered social order generally (as in Ho 2006; and Zaiceva 2010) and the impact of racism and discrimination. Some of the complexity that such work adds to our overall understanding of migrant economic integration is evident in the scholarship discussed in the following section.

Migrant strategies and contexts of reception

Research on migrant strategies commonly also considers the context of reception, so we examine these studies together in this section. Compared with the mostly quantitative human capital reports, research into migrant strategies and contexts was most commonly qualitative in nature.

The literature that focused primarily on migrant strategies generally looked at what migrants did as they searched for work: how they negotiated their gendered and national identities in the workplace (McKay 2007; Kambouri 2008; Karjanen 2008; Datta & Brickell 2009); their migration stories (Broughton 2008); and the factors that shaped their migration decisions (Kobayashi & Preston 2007; Ryan 2008; Ryan et al. 2009). Research into specifically economic strategies focused on the use of social networks (Livingston 2006; Shih 2006; L. Ryan 2007), on entrepreneurship (Light 2007) and on concentrations of ethnic employees, or niche economies (Ceccagno 2007; Schrover et al. 2007).

In this literature on strategies and contexts, migrants are represented as active agents, rather than as the seemingly passive bearers of characteristics reductively presented in the human capital approach. As noted earlier, the focus on migrant agency frequently also brought issues of the ‘context of reception’ more clearly into view. Immigration policies, labour market structure and gendered segregation, employer hiring practices and the impact of discrimination frequently come into play in accounting for migrant strategies. Pio's (2007) New Zealand-based study exemplifies this qualitative approach. She looked at the entrepreneurship of Indian women migrants as, at least in part, a strategy deployed in response to labour market discrimination.5

Research into gendered identities highlights the ways in which gender differences are reproduced and even exaggerated in the labour market location and segregation of migrant workers. Karjanen's (2008) study of Mexican service workers in the USA highlights the role of employers’ gendered and ethnic stereotypes of ‘good’ Mexican workers in reproducing their labour force location. Interestingly, he also shows how the workers themselves play to these stereotypes and the contradictions and tensions that this creates for them (Karjanen 2008, pp. 56–61). Some of these insights are also reflected in Ramirez and Hondagneu-Sotelo's (2009), p. 86) study of Mexican gardeners in the United States as they negotiate and strategise contradictory positioning in informal and formal economic transactions centred on gendered ‘ethnic entrepreneurship and subjugated service work’. Similarly, Datta and Brickell (2009) argue that the Polish builders in London use their masculine and national identity to position themselves competitively in the labour market, while Batnitzky et al. (2008) explore the intersection of gender and class identity for middle-class Indian men in the British hospitality industry.

McKay's (2007) study of Filipino seamen highlights the role of the Philippine state in constructing seafarer masculinity as a strategy to maintain their niche position and to discipline the workforce to tolerate their harsh working conditions. In these papers, gender identities play a part in securing and maintaining a labour market location. Kambouri (2008), in contrast, studying Albanian women domestic workers in Greece, argues that in this female-dominated employment, women redefine themselves in increasingly masculine terms (c.f. O'Neill 2007 on migration as empowering for Nepali women and Solari 2006 on the interplay of religious and gendered identities).

L. Ryan's (2007, 2008 , 2009) research into Irish women and Polish migrants in the United Kingdom emphasises the need for research that examines the role of migrant networks in migration decisions and support. She also notes the importance of looking beyond nuclear family networks to transnational and local networks (L. Ryan 2007, p. 309). Kobayashi and Preston (2007) explored the transnational links and movements of Chinese families between Hong Kong and Canada. They highlight the need to see migration as a family strategy and not a purely economic one (Kobayashi & Preston 2007, p. 153; Gubhaju & De Jong 2009), a significant point for New Zealand given the importance of lifestyle factors as motivations for migrants and their families (Department of Labour 2009). One of the key reasons for migration to Canada from Hong Kong, for example, is educational opportunities for children (Kobayashi & Preston 2007, pp. 158–159). However, difficulty in finding appropriate work—for both parents and Canadian-educated children—frequently results in a return to Hong Kong (Kobayashi & Preston 2007, p. 166).

Other research investigated the role of social networking in job searching. Livingston's (2006, p. 62) study of Mexican migrants in the USA suggested that women migrants are disadvantaged by their reliance on friendship networks which keep them in informal, gender-segregated employment (also see Bastia 2007). Overall, Livingston's key finding is that the use of social networks was detrimental to women and beneficial to men. The result for women was more likely to be work in the informal sector with poor wages and limited benefits, an outcome which Livingston says needs the insights offered in qualitative research to explain (Livingston 2006, pp. 61–62). She cites a number of studies that show how the utilisation of social capital in a gender-segregated labour market serves to perpetuate this segregation, reinforcing the channelling of women into low-paying jobs In contrast, Shih (2006), studying highly skilled Silicon Valley engineers, found that both men and women successfully used their social networks and adopted job-hopping strategies to escape discrimination in the workplace. She acknowledges that the normality of high turnover and gender and ethnic diversity in this particular workforce makes successful networking an accessible strategy to both genders and to different ethnic groups.

Entrepreneurship was also examined as a strategy against discrimination and under-employment. A special issue of the Journal of Ethnic and Migration Studies (JEMS) in 2007 drew together significant work in this area. These studies point to the role of entrepreneurship in the creation and maintenance of gendered and ethnic niches (Light 2007; Rangaswamy 2007) and to the role that family labour and ethnic networks play in the success of these niche businesses (Rangaswamy 2007; Foner 2009). In the editorial of the JEMS special issue, Schrover et al. (2007) highlight the interaction between niche development and gendered labour market segregation. They argue that the evidence shows that women's participation, as workers or as entrepreneurs, strengthens the niche and ensures its continuity (for example Rangaswamy 2007). Women's participation can also lead to demands for highly flexible child-care and thus the development of an additional niche (for example Ceccagno 2007).

Furthermore, while entrepreneurship and niche development can lessen but not eliminate women's earnings disadvantage (Light 2007), the interaction of niches and labour market segregation tends to keep women and migrants in the secondary labour market where work is flexible and there is no career progression (for example, Bastia 2007; Da Roit 2007; Lazaridis 2007; McGregor 2007). Finally, Gratton's (2007) comparative analysis of Bolivian migration to the USA and Spain points to the role of immigration policy settings and of who migrates first (men or women) and with what network support, in whether or not migrants end up in niche employment (beyond the JEMS special issue also see Stone et al. 2006; Foner 2009; Peixoto 2009).

Summarising this overview of the 2006–09 international literature, we agree with Donato et al.'s (2006, pp. 11–12) argument that ‘gender’ is a migrant characteristic but is also an explanatory variable that is capable of helping researchers develop a more nuanced understanding of the strategies migrants deploy in settlement, employment and upward mobility. Sensitivity to gender difference also highlights aspects of the contexts within which migrant endeavour takes place. Across these areas of research there is support for distinctly gendered outcomes for migrants. Evidence from a wide range of countries, migrant groups and employment contexts point to women gaining fewer returns on their human capital and, in particular, being disadvantaged by family and childcare responsibilities. However, the evidence is not uniform across all groups, professions and contexts; there is no evidence of universal female disadvantage and male advantage on any dimension. What this suggests, we believe, is the need for ongoing research based in the particularities of specific origin and destination countries or regions, their labour market structure, labour force composition and particular migrant groups.

Towards a New Zealand research agenda

The international literature on the economic integration of migrants that has been examined thus far provides insights helpful to the development of a policy-centric research agenda for Aotearoa New Zealand. We suggest that research into economic integration needs to be comprehensive, integrated, and needs to clearly instantiate the relevance of gender to policy making. In this final section, we draw these insights together under these three themes. First, we argue that a research agenda needs to be comprehensive: that is, it needs to facilitate the investigation of the range of migrant economic experiences and activities across New Zealand society, across multiple sites (nationally and regionally), looking at different cohorts, and from different perspectives. Second, we suggest that this agenda needs to be integrated, by which we mean that new research needs to highlight that both men and women (as well as children and families) are migrants, and that their experiences are gendered in complex, paradoxical and locally inflected ways. The new research also needs to emphasise that migrant characteristics are not separate from either the strategies migrants use in order to survive and thrive, or the contexts into which they are received and accommodated. Exploring the nuanced character of these lived experiences alongside large-scale empirical studies that capture one picture of characteristics provides insights that are necessarily richer. Such integrated approaches also mean that we need to rely on and integrate more diverse and innovative research methods. Finally, we address the question of gender's relevance to policy. We argue that gender matters in migration research, not just as a scholarly exercise, but because it is critical to the quality and relevance of policy decision-making: to better understand and respond to the economic and social impacts of integration on both migrants and their ‘host’ societies.

Comprehensive research

The trends identified in the literature toward diversified, bifurcated and feminised migration flows are a useful starting point for a comprehensive framework. The characteristics of these flows are global and are as relevant to Aotearoa New Zealand as anywhere else. In terms of diversification, and following the insights of Badkar et al. (2007b), while we now know quite a bit about the characteristics of migrants to New Zealand, there is much that is not known about the drivers (including gender) of migration from specific source countries, to specific local destinations for particular (economic or other) purposes. One response to this challenge is embedded in the work being undertaken by the IIP on communities from the Peoples Republic of China, Korea, South Africa, India and Britain. In the IIP research programme, gender-specific questions are being asked in both surveys and interviews. This work will be reported between 2010 and 2012.

In relation to temporary and circular flows under the RSE policy, understanding how these flows differentially affect men, women, children and family and transnational networks is important, but still to be clearly addressed. While the holistic model developed by the evaluators (Department of Labour 2010) of the RSE identifies that the RSE workers are embedded in families and communities (p. 5), there is no reference to the gendered implications of these policies. McKenzie et al. (2008), writing about Vanuatu, note that men predominate in the RSE streams and leave spouses and children behind in the islands, but again, there is no analytical lens that frames the implications of these gender differences. In relation to the RSE, then, there is scope for careful scholarship, not just on the characteristics of these migrants, but also on the different strategies men and women deploy in short-term, circular labour migration, both at home and in receiving countries.

New Zealand's current policy objectives of securing skilled migrants on the one hand, and temporary migrants for seasonal work on the other, create bifurcations between these groups of people, and between both of these groups and earlier migrant and settlement streams. We know little about the characteristics and impacts of these bifurcations. Picking up on Pfeiffer et al.'s (2008) suggestions, there is also further work to be done in relation to the gendered implications of migration decision-making, social networking and the costs and benefits of migrating for individuals, families and sending and receiving communities.

The International Migration, Settlement, and Employment Dynamics (IMSED) research arm of the New Zealand Department of Labour, through the Longitudinal Immigration Survey: New Zealand (LisNZ), has a wealth of important empirical data but the reporting of gendered characteristics from these data is negligible. One rare reference records that:

overall, female migrants slightly outnumbered male migrants (52 percent compared with 48 percent), but this result varied by immigration approval category. While the proportion of females was higher for Skilled secondary (70 percent), Family Partner (64 percent), and Family Parent (57 percent) migrants, Skilled principal migrants were more likely to be male (67 percent). (Department of Labour 2009, p. 33)

The lack of any comprehensive analysis of these gendered data means that the implications of a preponderance of women being recruited into service and care industries (the feminisation of flows) are overlooked, as are the implications of skilled principal migrants being more likely to be male. The LisNZ data provide multiple statistical starting points that need to serve as leverage for a raft of further research questions that have not yet been pursued in IMSED studies.

Finally, while work on the economic integration of migrants often eschews irregular migration, this is an important and largely un-researched field in New Zealand and one which has both gender and economic implications, albeit on a smaller scale relative to the data focus on skilled and unskilled migration. As J. Ryan (2007, p. 15) noted in her discussion of Chinese migration to Australia, ‘women arrive as migrants as part of the work categories, but they also arrive as refugees … asylum seekers, and trafficked as outworkers (garment workers etc.); sex workers, mail order brides, domestic workers, service industry employees and unskilled laborers’. We know little or nothing about the ‘commodity chains’ of such irregular migration in New Zealand, but the media reports that such chains do exist (see, for example, Tan 2010). Chui (2008, p. 29) also points out that perceptions exist that small business Chinese migrant entrepreneurs may encourage ‘illegal low-wage overstayers’, but such perceptions are difficult to counter without detailed research.

Integrated research

The findings from the international review point to a wide range of researchable topics about the impact of gender on economic integration that have yet to be addressed in Aotearoa New Zealand. Such topics include: the differential returns on human capital; the underutilisation of female capital; the differences that education, language skills, and child care responsibilities make to the outcomes for men and women; the differential effects of labour market segregation; the different ways in which men and women deploy familial and transnational networks; among just a few that have been mentioned in the review. A further point has also emerged; while migration may be a global process, it is the local effects that are particularly significant in policy terms. Therefore, whatever happens in Aotearoa is unique to this setting and we need to integrate the knowledge that is being developed about migration flows, economic integration and settlement processes in the local context. While international research can inform our understandings, it cannot replicate what we need to know about what is happening here. The local particularities of migration research can provide us with specific New Zealand answers to the debates that produce conflicting findings on the international stage.

Local research can provide the complementary richness that derives from research that considers the interaction of the characteristics of the migrants who come to New Zealand, with the strategies they use in this place as well as a range of factors shaping the context of reception. Further, migrants are never solely individuals who just happen to end up here in some ‘quest for a better life’: they are embedded in families and communities, and have complex motivations and expectations. As Meares (2010), for example, suggests in relation to women migrants to Aotearoa from South Africa:

the changes brought about by the international migration of skilled women can be better understood by examining the dynamic, iterative relationship between the public sphere of paid work and the private sphere of the home. Paying attention to this nexus reveals both the disparate ways women navigate this terrain and the complex ways they make sense of the challenges that arise from their altered circumstances. (p. 3)

Local research that develops an integrated picture of the interaction of migrant characteristics and strategies, and of contexts of reception, must also combine quantitative and qualitative modes of inquiry. In the introduction to the IMR special issue with which we began this review, Donato et al. (2006, p. 22) concluded that the disciplinary unevenness evident in the field was linked to the acceptance or not of qualitative methods and methodologies. While they argued for the importance and place of both quantitative and qualitative research, they also suggested that it was qualitative research that sought to explain the gendered nature of migration. Indeed, across the post-2006 literature that we have surveyed, it is apparent that both quantitative and qualitative methods are crucial to providing a well-rounded account of the interplay between new migrants and the society and workplaces within which they seek work and settlement.

Quantitative research can provide the ‘big picture’ data of patterns of migrant characteristics and employment across locations. The patterning such data make visible can identify sites of disadvantage in labour force participation rates, occupational status and earnings, and longitudinal data collection results in rich data sets with which to investigate change over time. We would suggest (as have others) that it is essential that migration research extends beyond a focus on migrants as bearers of human capital, and here qualitative methods in particular show their worth. The gaps identified in quantitative research are the result of social action on the part of migrants and employers, action shaped by the specifics of the context of the local labour market and the destination society. Beyond human capital, explanations for these gaps and insights into strategies that can reduce and eliminate them require the detailed and nuanced insights of qualitative data. Schwenken and Eberhardt (2008) make similar recommendations in their review of the concept of ‘gender knowledge’ as it is deployed in economic migration theories. They suggest that recent reviews of the migration literature point to the need for more gender analysis in quantitative data, greater recognition of the value of qualitative analysis, and for greater reflexivity in methodological approaches to migration issues.

Relevance to policy

In 1991, Wendy Larner pointed to the dearth of research into women and migration in New Zealand. A scan of recent policy-related research papers in New Zealand exposes the continuing lack of attention paid to gender in this field. Take consideration of the skilled migrant category as one example of significant interest for New Zealand policy researchers: analysis of the trajectories from migration through to settlement of this group (through LisNZ, for example) does not account for their gendered identities and can only provide a very distorted and partial picture of the experiences, strategies and achievements of these prized individuals. Some evidence of the gendered nature of their trajectories appears in IMSED research when it is reported that ‘the estimated median annual income for males was $42,500 compared with $29,100 for females’ and that ‘males were more likely to be in paid and full-time work than were females’ (Department of Labour 2009, p. 105). These statistics point to a pattern of gendered disadvantage for women migrants in New Zealand, which, in itself points to a necessary research agenda to explain such discrepancies and identify strategies to ameliorate or overcome them.

While there are few data on the gendered nature of migration in New Zealand, there is some recognition of this gap. For example, a recent in-house review of research findings from the New Zealand Department of Labour (Grealish 2008) identified the key fields being researched in policy-centric migration research in New Zealand as: the economic and social impacts of immigration; the inter-relationship between migration and employment; the factors associated with global mobility; and the monitoring of the settlement trajectories of skilled migrants and work to residence pathways. Of these key current interests of New Zealand immigration policy researchers, in only one (the inter-relationship between migration and employment) was gender raised as a significant issue. The summary of the findings from this research stream included the following statement:

Despite a growing significance of the global feminisation of migration, including the feminisation of labour market related migration, this area has attracted little research or policy attention in the New Zealand context. Patterns of gendered migration from our main source countries have received little attention in the New Zealand context, despite the growth in migration from these countries. (Grealish 2008, p. 13)

To give another example, Ghosh (2009), p. 10), in identifying short-term or seasonal migration as one of five generic forms of migration, suggests that within such migration patterns it is important to identify what is specific about women's [and men's] long-term, short-term and seasonal migration patterns, and the needs and challenges faced at ‘every stage of the migration process, including return’. Spoonley and Bedford (2008) and Bedford (2008) briefly acknowledge gender differences in labour market engagement amongst Pacific peoples, as do Ramasamy et al. (2008). However, there is no suggestion in any of these papers that gender analysis is being given any prominence. Gender analysis in migration research is relevant to New Zealand researchers and policy makers and remains a perplexing gap in the local literature over 30 years after gender began to appear as a significant factor in the international literature.

Conclusion

What is evident from this review is that gender still matters. As one of the major principles shaping the unequal power dynamics of the social world (Cheng 1999), gender fundamentally impacts on the interpersonal relationships and social institutions that organise the migration experience (Curran et al. 2006). Gender influences decisions about who migrates and who stays at home (Anthias & Lazaridis 2000); shapes the labour markets (Kofman & Raghuram 2006) and migration policies (Piper 2006) of sending and receiving countries; and affects negotiations between women and men over the division of paid and unpaid labour (Espiritu 2002). In short, as Cheng (1999, p. 40) notes, ‘this important organising principle of power shapes the particular contours of the migration process as well as the diverse experiences particular to people situated in different locations along social hierarchies’.

New Zealand is a small country in which migrants are both highly and less visible, comprise a large proportion of the overall population, and are the focus of significant policy and media attention. Ideally, a research agenda focusing on the economic integration of migrants will take comprehensive account of universally identified features of gendered migration, both as currently constituted and as they continue to be identified and theorised in international literature. It will integrate these global perspectives with local insights into the particularities of Aotearoa New Zealand (in place and time) and will bring together the description of migrant characteristics with analysis of the implications of gendered differences in motivations, strategies and contexts of reception. It will deploy sophisticated methodologies that exploit the complementarities of quantitative and qualitative approaches, particularly emphasising projects that use mixed methods in complex thematic and interdisciplinary analyses.

As the funding for social research has retrenched in the last few years, the capacity of research to speak to these complex intersections of migration experiences and economic integration has contracted markedly. Policy decision-makers therefore have to work harder to find an evidence base and are more likely to rely on the ‘default’ descriptions that emerge from administrative or departmental research, in which gender is seldom profiled as a significant variable. Even so, we argue that migration research that does not foreground and analyse gender dynamics has limited value.

Notes

1. Donato et al.'s study (2006, p. 4) looked at anthropology, geography, history, law and society, political science, psychology, sociology and sexuality studies.

2. http://integrationofimmigrants.massey.ac.nz/.

3. See also Michael Ardovino and Marcia Brown's (2008, pp. 3–5) literature review on the gendered impact of circular migration.

4. We argue that this field is another under-researched area in New Zealand migration studies that deserves attention.

5. Note that Pio's was the only New Zealand-based study we found in this international literature survey.