Labour Mobility Pacific Possible Background Paper No.1. Pacific Island countries face unique development challenges. They are far away from major markets, often with small populations spread across many islands and vast distances, and are at the forefront of climate change and its impacts. Because of this, much research has focused on the challenges and constraints faced by Pacific Island countries, and finding ways to respond to these. This paper is one part of the Pacific Possible series, which takes a positive focus, looking at genuinely transformative opportunities that exist for Pacific Island countries over the next 25 years and identifies the region’s biggest challenges that require urgent action. Realizing these opportunities will often require collaboration not only between Pacific Island Governments, but also with neighbouring countries on the Pacific Rim. The findings presented in Pacific Possible will provide governments and policy-makers with specific insights into what each area could mean for the economy, for employment, for government income and spending. To learn more, visit www.worldbank.org/PacificPossible, or join the conversation online with the hashtag #PacificPossible. PACIFIC POSSIBLE Labour mobility: the ten billion dollar prize Richard Curtain, Matthew Dornan, Jesse Doyle and Stephen Howes ISBN: 978-0-9943520-4-0 February 2017 © 2016 International Bank for Reconstruction and Development / The World Bank 1818 H Street, NW Washington, DC 20433 Telephone: +1 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in the work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818 H Street, NW, Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Acknowledgments Professor Stephen Howes is Director of the Development Policy Centre of the Crawford School of Public Policy, Australian National University. Dr Matthew Dornan is the Deputy Director of the Development Policy Centre. Dr Richard Curtain is a Visiting Fellow at the Development Policy Centre, and a labour market expert. Jesse Doyle is a Labour Migration Specialist in the Social Protection and Labour Global Practice of the World Bank. Manjula Luthria provided supervision within the World Bank for this project. Ashlee Betteridge provided editorial assistance. The authors would like to thank Jehan Arulpragasam, Venkatesh Sundaraman, Robert Utz, Kim Edwards, Manohar Sharma, Leitumai Malaulau and Stephen Close from the World Bank for their contributions and feedback. We are also extremely grateful to those who provided commissioned papers for this report, including: Charlotte Bedford, Richard Bedford, Zulmira Fonseca, Sophia Kagan, Carmen Voigt-Graf and Ann Wigglesworth. The authors would like to express their appreciation to the peer reviewers – Professor Alan Winters from the University of Sussex and Matthias Lundberg of the World Bank. We thank all the participants of the ‘New research on Pacific labour mobility’ workshop held on June 2, 2016 at The Australian National University for their many suggestions and comments. And we thank all those including officials from both the New Zealand and the Australian governments and a number of other organisations and individuals who provided comments and feedback on an earlier draft of this report. This study is one part of a broader ‘Pacific Possible’ research agenda, which explores transformative opportunities that exist for Pacific Island countries over the next 25 years and identifies the biggest challenges that require urgent action. Contents Acknowledgments ...................................................................................................................................... ii Executive Summary ...................................................................................................................................... i 1. Introduction ............................................................................................................................................. 1 2. Labour mobility benefits ......................................................................................................................... 5 2.1 Benefits to migrants and their families.............................................................................................. 5 2.2 Benefits to sending countries ............................................................................................................ 6 2.3 Benefits to labour-receiving countries............................................................................................... 8 3. Receiving-country reforms: an overview ............................................................................................. 11 4. Expanding seasonal worker access ....................................................................................................... 12 5. Expanding medium-term or temporary Pacific labour mobility options (1–5 years) ......................... 15 5.1 Working holiday maker (backpacker) programs .............................................................................. 16 5.2 Korea’s Employment Permit System................................................................................................ 17 5.3 Expanding the Pacific uptake of temporary skilled migrant programs............................................ 18 5.4 Proposal for a Pacific in-home caregivers program for those with high medical-care needs ......... 23 6. Expanding permanent migration: Pacific options ................................................................................ 26 6.1 Permanent skilled migration ............................................................................................................ 26 6.2 Migration lotteries: learning from New Zealand’s Pacific Category ................................................ 27 6.3 Open access: the proposed Australia-New Zealand Atoll Access Agreement ................................. 31 7. Sending country reforms ....................................................................................................................... 34 7.1 Education and training policies to expand labour mobility opportunities while avoiding brain drain ....................................................................................................................................................... 34 7.2 Promoting Pacific schemes and marketing Pacific workers............................................................. 37 7.3 Policies to manage the social impact of labour mobility ................................................................. 38 8. Labour mobility projections .................................................................................................................. 39 8.1 The scenarios: business as usual, medium growth, and high growth ............................................. 39 8.2 Business as usual results .................................................................................................................. 41 8.3 Results from the medium and high growth scenarios ..................................................................... 44 8.4 Economic and social impact of increased mobility .......................................................................... 47 8.5 Receiving country impacts ............................................................................................................... 49 9. Conclusion ............................................................................................................................................. 50 References ................................................................................................................................................. 52 Annex 1 – Methodology for projections................................................................................................... 57 Annex 2 – Detailed projections ................................................................................................................. 68 Executive Summary The ability to access external labour markets varies significantly across the Pacific. Some Pacific island countries have free labour market access to the United States or New Zealand. Others have historically had high levels of labour mobility. Five countries are the particular focus of this report. The three Melanesian countries of PNG, Solomon Islands and Vanuatu have amongst the lowest rates of outward migration in the world due to severely restricted external labour market access. The atoll countries of Kiribati and Tuvalu have low to moderate levels of labour mobility, and both face high levels of climate risk.1 Pacific migrants go mainly to one of three countries: Australia, New Zealand and the United States. The report mainly focuses on reforms for Australia and New Zealand because of the strategic importance of the Pacific to them. The paper also considers Korea, because of one particular Korean employment scheme which could be of significant benefit to the Pacific if it was extended to the countries of the region. This report looks at the issue of labour mobility from both sides of the equation. It seeks to understand the perspectives and concerns of both sending and receiving countries, and it proposes reforms for both sets of countries. The report places particular emphasis on providing opportunities for low-skilled and medium-skilled migration. A broad range of reforms is proposed, but we are particularly interested in providing labour mobility opportunities for those beyond the elite of sending countries. This is to maximize the welfare gains of migration, and to reduce the risk of brain drain. The report is based on extensive new research and presents a menu of options. Not everyone will agree with every idea presented in the paper. Its purpose is to present well-researched proposals to stimulate thinking and discussion. The benefits of labour mobility International migration offers the potential for a ‘triple win’, delivering gains for migrants, sending countries, and receiving countries. Increased international migration for work offers critical opportunities for the Pacific and its people. Given the unique development challenges faced by the Pacific Island countries, there is now broad consensus that expanding labour mobility is vital for their future. Given their youth bulge, unemployment is a pressing problem. Where these countries are unable to bring jobs to the people, the alternative is to bring the people to where the jobs are. For labour-sending countries, remittance flows can be important sources of income and consumption, as well as foreign exchange and investment, often in education and health. More 1 The report only covers World Bank member countries in the Pacific. The Cook Islands, Tokelau and Niue are not World Bank members. Nauru only very recently joined the World Bank and is therefore also excluded from the analysis. i broadly, migration opportunities increase the incentives families have to educate their children, and facilitate knowledge transfer. Australia and New Zealand would also benefit from greater Pacific labour mobility. Both countries have deep interests in a stable and prosperous Pacific, interests which are advanced by increased Pacific labour mobility. Aid dependency in the region is high, and reliance on aid alone is an unbalanced strategy. Both by improving employment prospects and by increasing remittance flows, labour mobility helps stabilize otherwise fragile states. Moreover, in marked contrast to aid, migration offers self-selected individuals and their households the chance to change their economic and social circumstances, as the funds generated go directly to households. Migration, much more than aid, requires and produces major changes in the attitudes of those involved. Encouraging Pacific labour mobility is also important to Australia and New Zealand for domestic reasons. Advanced economies will require high rates of net migration in coming years to address major labour market shortfalls. These are often in sectors – such as construction, health care, and social assistance – where Pacific nationals, with some targeted training, would be well placed to fill the gaps. Working out a Pacific labour mobility regime would help Australia, in particular, move beyond the current ad hoc and unsatisfactory arrangements for importing unskilled labour (via reliance on international students and backpackers in particular). The evidence suggests that Pacific workers do well in metropolitan settings. In summary, while New Zealand has already gone further down this road than Australia, it is in both countries’ interests, for both domestic and international reasons, to expand Pacific labour mobility. The small size of most of these Pacific nations makes this goal a manageable one. Countries have a range of tools at their disposal to increase migration from the Pacific. They can use their aid budget to fund ‘aid for migration’ schemes. This can be considered as a form of ‘aid for trade’, a form that is at least, if not more, relevant for Pacific nations than trade facilitation. Receiving countries can also modify their migration regime to favour citizens of particular countries. It is a myth that these countries run non-discriminatory migration programs. For example, Australia plays host to hundreds of thousands of backpackers. They are nearly all (95 per cent in 2013-14) from developed countries because the relevant backpacker visas for developing countries are capped (typically at very low levels), while those for most developed countries are uncapped. The report proposes a range of seasonal (less than 12 months), temporary (1-5 years) and permanent labour-mobility schemes. Long-term migration brings the greatest benefits, but temporary migration opportunities are often more plentiful and may provide a route to permanent residence. The report also argues that receiving countries can reduce the cost of remittances and other financial barriers to mobility via superannuation reforms. Expanding seasonal worker access Seasonal worker programs are increasingly offering an important pathway for Pacific labour mobility. Almost 12,000 workers from the Pacific arrive in Australia and New Zealand annually ii under the former’s Seasonal Worker Programme (SWP) and the latter’s Recognised Seasonal Employer (RSE) scheme. New Zealand’s RSE has outperformed Australia’s SWP, and is still more than twice the size. The real constraint in Australia is the lack of an aggregate labour shortage due to the large number of backpackers working in agriculture, as well as the number of illegal workers (the two categories partially overlap). The constraint in New Zealand is the cap, which, as of mid-2016, was 9,500. (Australia’s scheme is uncapped.) A number of reforms are needed to expand Australia’s Seasonal Worker Programme. The second-year visa extension provided for backpackers if they work on a farm for three months was put in place in 2005-06 when Australia was unwilling to introduce a seasonal migration scheme. It leads to direct competition between backpackers and seasonal workers, a competition that the backpackers are winning. In 2014-15, there were some 41,000 second-year backpacker visa applicants, about 90 per cent of them on the basis of farm work completed. It is unfortunate that a scheme that benefits the citizens of poor countries is undermined by one that benefits mostly the citizens of the rich. Either the second year visa extension should be done away with, or it should be made available in return for three months of work in any sector. Similarly, there needs to be a crackdown on illegal labour (including by requiring the registration of labour hire companies). Other reforms that would help expand the SWP include: greater publicity; removing the $A500 employer contribution to the international airfare of returning workers to ease cost concerns; and giving all employers participating in the scheme the primary role in the selection of workers from a sending country labour pool to ensure that workers meet their needs. Expanding medium-term or temporary Pacific labour mobility options (1–5 years) Skilled, semi-skilled and low-skilled labour mobility programs of longer but still temporary duration are little utilized by the Pacific, but could become important. The report examines four programs through which medium-term or temporary labour mobility opportunities could be expanded for the Pacific: working holiday maker programs; Korea’s Employment Permit System; Australia and New Zealand’s temporary skill visas; and a proposed new aged care program. Most OECD countries offer working holiday visa programs with unlimited access to young visitors from other OECD countries. As the name of these visas suggest, work is intended to be incidental to travel. However, some OECD countries rely on backpackers to fill low-skilled labour shortages. Australia does this more than any other OECD country, attracting in 2013 some 249,000 backpackers, about half of all working holiday makers in 22 OECD countries in that year. As noted earlier, access to backpacker visas is biased in favour of developed countries. Some developing countries have limited access, and Australia and New Zealand could open up their working holiday maker schemes to their Pacific island neighbours. So far, Australia has signed an agreement (yet to come into effect) with PNG (for 100 places), and has entered into discussions with Fiji. If the working holiday visa is to be made relevant to the Pacific, the eligibility requirements iii should be made less restrictive. For example, a high school certificate should suffice. The minimum requisite level of savings should be reduced, and the requirement for sending government approval abandoned. Although Korea’s Employment Permit System (EPS) is not currently open to Pacific countries, Timor-Leste was included in the scheme in 2009. This program annually grants up to 55,000 workers from 15 countries across Asia access to three years of low-skilled work with scope for a 22 month extension. Timor-Leste has managed to send almost 2,000 workers to Korea over a six year period. A survey of the experience of Timorese workers who participated in the scheme came out with generally positive findings. Overall, the returned workers spoke well of their experience abroad, in terms of both amounts earned and skills gained. The Pacific Island countries should engage with the South Korean Ministry of Employment and Labour which manages the EPS. Given Korea’s growing aid interest in the Pacific, it is quite likely that it would respond positively to a request for participation. There are currently few Pacific islanders migrating annually through temporary skilled migration schemes. From 2010 to 2015, only 2,905 temporary skilled (457) visas were granted to migrants to Australia from the Pacific, less than one per cent of total arrivals under this visa category over that period. Over the same period, a total of 11,777 NZ Essential Skills visas were awarded to Pacific migrants or nine per cent of the total. Pacific workers find it easier to get access to the New Zealand scheme than the equivalent Australian one because the former has no minimum skills threshold, and because the Pacific NZ diaspora is larger. Both visas are expensive, but the New Zealand scheme provides a discount for Pacific islanders. Australia has tried – unsuccessfully to date – to increase skilled and semi-skilled labour mobility through the Australia Pacific Technical College. This was a potentially innovative ‘aid for migration’ initiative. Its failure to deliver on the labour mobility front has undermined the College’s viability, with a recent evaluation finding that it had a negative net present value due to the low number of migrant graduates it has so far produced. There is high demand from APTC graduates to migrate, but inappropriate qualifications, high costs and a lack of links with employers in Australia and New Zealand have prevented their aspirations being realized. APTC’s redesign presents a new opportunity. Australia has signalled its long-term commitment to APTC with new funding through to June 2018. Seven measures are recommended to ensure APTC graduates are able to access Australia’s Temporary Skilled Work and New Zealand’s Essential Skills Visa.  First, APTC graduates should be made eligible for Australian graduate visas. This would allow APTC graduates time to complete work experience certification requirements and to find jobs.  Second, the maximum duration of stay under the SWP should be increased to one year for certain sectors. APTC graduates (and others) could work in Australia through the Seasonal iv Worker Programme in sectors now covered by the scheme, such as accommodation and tourism in Northern Australia, and a range of agriculture industries including cattle, sheep, grain and mixed enterprises. A more extended work period under the SWP would suit employers better and ensure a higher take up. This longer duration would also give APTC graduates more time to find a longer term job by being sponsored by an employer for a 457 visa.  Third, some of the APTC funding could be used to promote APTC graduates to Australian and New Zealand employers. This could be done by, for example, hosting employer visits to APTC campuses to meet instructors and students.  Fourth, cost constraints need to be addressed. The aid program could be used to reduce the costs of skill certification, and increase the frequency and ease of certification opportunities.  Fifth, to allay fears of brain drain it is essential that APTC either train or support the training of fresh workers in the fields it is training in. This will ensure that the Pacific skilled labour stock is replenished rather than depleted. This makes economic sense, and is also key for ensuring sending-country political support for APTC.  Sixth, APTC needs to focus its training on occupations where there is a feasible pathway from training to migration to specific destination countries.  Seventh, better data collection on employment and migration outcomes would support better achievement of a labour mobility objective for APTC. Implementing the above reforms would not only make APTC a viable project, but also usher in a new era of migration for the medium-skilled from the Pacific to Australia and New Zealand. The ageing of the population in OECD countries and the greater longevity of individuals will lead to increasing numbers of people at older ages with a severe disability. In Australia and New Zealand, the share of those aged 80 and over is estimated to nearly double between 2015 and 2040 to 7.4 per cent of the population. As a result, there will be increasing shortages of aged care workers. In Australia, we estimate that the number of aged care workers (excluding managers) in both residential and community care could increase from 201,600 in 2011 to 532,000 or higher in 2040. Australia and New Zealand both lack a reliable system to supply qualified workers who can provide care on a continuous basis to people in their own homes. Home-based continuous care relies mainly on informal care givers for the elderly, usually partners, who are diminishing in number due to the ageing of the population. At the same time, the demand for continuous home- based services is increasing as more older people prefer to stay in their own homes. Reforms giving the elderly more control over the form of delivery of aged care services will reinforce this trend. Canada’s program for in-home caregivers for people with high medical needs provides a good model. It allows persons residing in Canada to employ qualified foreign workers in private residences to provide care for elderly persons or persons with a disability. The program offers a permanent residence pathway for migrants who have completed two years of work in a four-year v residency stint and has a cap of 2,750 per year. The core elements of the Canadian program are the minimum requirements for employment as a caregiver (eg post-school qualification in health care of at least six months and good English language ability), no requirement to live in as a home-based caregiver (since 2014), agency oversight of the employment contract, and a prior labour market assessment to ensure that citizens or permanent residents have first opportunity to apply for the work. Canada’s approach could be used as the basis for the design of a similar program for Australia and New Zealand. A focus on the beneficiaries is essential to justify why a Skill Level 4 occupation should be an exception to the usual Skill Level 3 threshold for migrants in Australia and for access to a permanent residence pathway in both Australia and New Zealand. Limiting the program to Pacific migrants would enable donor funds to be directed to providing Australian & New Zealand recognised qualifications to potential migrants. It would also enable employers to develop a reliable source of carers and to have the primary role in their selection. Expanding permanent migration: Pacific options Long-term migration opportunities for the Pacific currently come through three types of programs: permanent skilled migration, visa lotteries, and open access. The reforms proposed in relation to temporary skilled migration and the APTC above will also lead to more Pacific islanders migrating under the permanent skilled migration regime. This section outlines proposals relating to visa lotteries and open access. Through its ‘Pacific Category’ visas, New Zealand offers 1,750 places for permanent residence each year to citizens of selected Pacific island countries. The Samoa Quota (SQ) allows 1,100 Samoans and the Pacific Access Category (PAC) visa allows 250 Fijians and Tongans and 75 i-Kiribati and Tuvaluans to be granted residence in New Zealand each year. The requirements are that the primary applicant has an offer of an ongoing job that meets a minimum income level, is in good health, speaks reasonable English and is of good character. The numbers applying for the two visa ballots show that there is a high level of interest in migrating to live and work in New Zealand. Overall, the evidence on employment outcomes and incomes earned shows that the Pacific Category visas have been successful. They have enabled a relatively small number of migrants from four Pacific countries to come as families to settle permanently in New Zealand. The minimal requirements such as a prior job offer have produced in general good employment outcomes. Most of both the principal and secondary applicant Pacific Category migrants are in employment and are earning a basic income after they migrate, as shown by tax records over a period of five years. The Pacific Category visas are well designed. They were designed not to ‘cherry pick’ migrants with the highest qualifications to avoid the danger of creating a brain drain from small countries. The use of the ballot to select migrants and putting the onus on those selected to find a job in New Zealand are important features of the program. The process ensures that the sending countries accept that the selection for a small number of places is carried out in a fair and transparent way. vi But two reforms would help improve migrants’ employment rates and lift their annual incomes. First, selection requirements could be toughened. Completion of secondary schooling increases the chances of employment by 8 percentage points for men and by 10 percentage points for women aged 25-64 years compared with the low educated. Completion of secondary school (or at least completion to Year 10) should be a requirement for a successful application. The English test could also be made more stringent, and a numeracy test also imposed. This would provide an incentive for the sending country to improve the quality of its schools. And these selection requirements, even if toughened, are ones that are achievable by prospective migrants, for example, by returning to school if needed, or at least ensuring that one’s children stay at school. The risk of brain drain is therefore minimal; indeed, the tougher requirements could lead to brain gain for the population not migrating. Providing more support for job brokers would also help improve employment outcomes. Heavy reliance is placed on the diaspora when it comes to job-hunting. However, if this diaspora is small and struggling, as some are, the result for new migrants may be an over-reliance on a small number of employers, resulting in high levels of risk and vulnerability. Reliance on informal networks needs to be supplemented by a greater use of job brokers with relevant experience and broader geographical reach. Funding for job brokers could be provided by the government of the sending country, and could be based on a fee levied at departure. With the reforms suggested above, there is a strong case for expanding New Zealand’s Pacific Category and for introducing a similar scheme in Australia. Introducing a ‘Pacific window’ into its permanent migration regime would help balance Australia’s overall immigration regime given the ‘OECD window’ in the much larger working holiday market visa discussed earlier (significant numbers of backpackers become permanent residents). Introducing a Pacific Category visa scheme for Australia would also have the benefit of helping the country move away from its current ad hoc reliance on students and backpackers to meet unskilled job needs. Countries provide open access only under very special circumstances. Climate change could be such a circumstance. The case is particularly strong for the low-lying atoll states. While there is still much uncertainty, the Intergovernmental Panel on Climate Change has highlighted the extreme danger that small island countries face. Both Kiribati and Tuvalu are actively seeking greater access to greater temporary and long- term employment overseas, especially for their burgeoning young adult populations. Without migration channels specifically for them, it is highly unlikely that they would be able to increase their rates of outward migration. With weak agricultural sectors, they struggle to compete in the seasonal worker schemes. Their low skill profiles and extreme remoteness place them at a disadvantage in accessing the temporary and permanent skilled pathways. Current schemes are inadequate. New Zealand’s Pacific Access Category reserves 75 slots each for the two countries. And just last year Australia established the Pacific Microstates–Northern Australia Work Pilot in Australia which provides 250 places (in total) for i-Kiribati, Tuvaluans, and vii Nauruans to work in Australia for up to two years. Much more is needed. Simulations show that migration rates from Tuvalu need to double and for Kiribati increase more than ten-fold just for growing populations to stabilize. One intervention which should be considered is the provision of open labour market access by Australia and New Zealand to Kiribati and Tuvalu, on grounds of their acute climate change risks. Open access would likely result in modest outflows. While surveys show that many households feel that migration will be a necessary strategy as a result of sea-level rise, most lack the financial means to migrate. Under the assumption that real income remains stagnant, we estimate that only about 31,000 i-Kiribati and 2,200 Tuvaluans would have the financial means necessary to migrate. If these numbers were to migrate steadily over a 25 year time horizon, Australia and New Zealand would be looking at an additional average inflow of 1,300 i-Kiribati and Tuvaluans annually or the equivalent of 0.6 per cent of their annual permanent migration program. Given the open access arrangements between their own two countries, it would make sense for Australia and New Zealand to provide open access to Kiribati and Tuvalu together. The total size of the diaspora in Australia from the two countries combined in 2011 was only 625 migrants. For New Zealand in 2013 it was 2,922 migrants. One important factor that might make Australia and New Zealand more popular destinations than the US is the better availability of free health care. Open access could be restricted to those with a certification of good health. To provide incentives for good education, basic educational and prior job requirements could also be imposed. This would effectively result in a program which is a half-way house between a lottery and an open-access scheme: that is, an uncapped, selective, country-specific program. Sending country reforms It takes two to tango. The report also explores what will be required of labour-sending countries, beginning with education and training policies, then marketing, and finally the addressing of social impacts. A major obstacle sending countries face in increasing migration flows is the low education levels of their populations, even among those employed in the formal sector. The small number of workers with post-school qualifications in the Pacific means that even small losses among this group can have serious consequences. There is already a reliance on foreign workers at medium- and higher-skill levels in many Pacific countries. Pacific governments need to increase the quantity and lift the quality of skills supply to expand opportunities for access to more than low-skill work overseas. The low post-secondary education level of the Pacific workforce is due in part to limited secondary education access, and poor quality throughout Pacific education systems. viii Several, particularly smaller Pacific Island countries are worried that more migration will result in a ‘brain drain’ of their skilled workforce. Their concern is that too many of the few with scarce skills and experience will migrate to high-income countries. The available data on emigration rates show that these concerns of small island countries are hardly surprising. Pacific governments’ concerns about brain drain need to be addressed as part of a national skills investment strategy. The first step for a Pacific government in developing a national skills strategy is to work out what skill sets or qualifications the public and private sectors need. The second is to lift the quantity and quality of supply of young people with the required post-school qualifications to meet both domestic and international requirements. This requires improving educational quality at all levels. Migrants can pay for the cost of their education. Resources need to be allocated to enable potential migrants to attain the required education level or qualifications needed for successful migration. A student loan scheme could also be applied to all students pursuing a post-school qualification, only to be repaid if the graduate emigrates for three years or more, or, more broadly, if incomes exceed a certain level. Receiving country policies should also be directed towards reducing brain drain. Any destination country ‘aid for migration’ program should ensure that it does not exacerbate brain drain. This can be done by training up, either directly or indirectly, fresh intakes of workers either to migrate or to replace those migrating. This is essential for both economic and political reasons. The Pacific Island countries should invest in developing marketing strategies for their workers in key receiving countries. Few employers in the main receiving countries are familiar with countries in the Pacific region. Without the volume of workers in-country and strong reputation that some of the regional labour-sending heavyweights have, such as India and the Philippines, the Pacific Island countries will struggle to gain a foothold with employers. As a result, they will be required to invest in creating demand through effective marketing. Some countries already have marketing plans and strategies in relation to the SWP and RSE, but a broader approach is needed. The most effective marketing is delivered by good service for employers by both workers and sending governments. Pacific governments should prioritize labour mobility schemes and monitor their progress at the highest levels. Pacific Islands Trade & Invest (PT&I) should be renamed Pacific Islands Trade, Invest and Work (PTI&W) to reflect a much-needed broadening of its mandate. Labour mobility is more important to many Pacific island economies than most exports. It is anomalous for the Pacific’s trade promotion body not to have a labour-mobility remit. PTI&W, as it might be renamed, could be made responsible for reaching out to neighbouring country governments and employers to explore new labour mobility opportunities for all of the Pacific. Adverse family and community impacts of labour mobility can be minimized by avoiding long- term family separations, as all the recommendations in this report do. In all the schemes we ix consider migrants can bring their families, except for three: seasonal workers, where the duration is short; as well as backpackers and Korean EPS workers, where the migrants are not expected to have dependants. In addition, negative family and community impacts can be addressed by sending countries through a variety of measures. These would include: discussing social impacts in pre-departure briefings; improving international communications infrastructure; strengthening prevention efforts around gender-based violence; improving migrant family support services in sending countries; and funding formal social protection programs for groups that are particularly vulnerable as a result of migration, such as social pensions for the elderly. Sending countries for seasonal workers should encourage the development of codes of conduct with a publicly funded complaint mechanism. Labour mobility projections What would happen if the reforms of the previous sections were implemented? The report illustrates the impacts of the reforms it proposes by a series of projections. Pacific labour mobility is on the rise, but is not keeping pace with population growth. In the absence of reforms, growth in the number of migrants and seasonal workers is projected to be steady but slow. Our ‘business as usual’ projections forecast the stock of Pacific migrants (temporary and permanent, but excluding seasonal workers) to reach 490,000 by 2040 up from the existing 420,000. Even without further reforms, the number of seasonal workers is set to reach about 29,000 by 2040, compared to about 12,000 in 2015, based on an extrapolation of current growth trends in the annual numbers under the uncapped Australian scheme. While these trends are in the right direction, they are inadequate. The ratio of Pacific migrant stock to Pacific population would fall from 4.5 per cent in 2013 to 3.5 per cent in 2040. With reforms, it would be possible to achieve significantly higher flows and stocks of migrants. We consider two scenarios, a medium growth and a high growth scenario, based on quantifications of the reforms presented in the report. The difference between them represents the degree of policy effort put in by both sending and receiving countries. The numbers of migrants and seasonal workers leaving the Pacific every year are 3.6 and 5 times higher in the medium and high growth scenarios than they are currently, and 1.6 to 2.2 times higher than what they would be in 2040 under business as usual. The scenarios are indicative only, but both are feasible, and in fact full implementation of the reforms outlined in this report would achieve an even higher number of labour mobility opportunities for Pacific islanders. The labour mobility gains are targeted at the low mobility and atoll countries. For example, the stock of PNG migrants increases almost nine-fold under the high growth scenario (between 2013 and 2040) while the stock of i-Kiribati migrants increases almost seven-fold. The reforms generate an additional 75,000 to 150,000 migrant job opportunities and $US5 to 10 billion of net income for the people of the Pacific by 2040, relative to business as usual. x The net income gain (measured in 2005 prices and in terms of the expected income increases net of opportunity costs) of both seasonal workers and temporary and long-term migrants is about 5-10 times the current value of Australia and New Zealand’s aid to the Pacific ($A1.2 billion). Increased labour mobility results in income growth doubling (or more) for the poorest Pacific island countries. In the high growth scenario, labour mobility reforms double per capita income growth (between 2013 and 2040) relative to BAU for PNG and the Solomon Islands, triple income growth for Vanuatu, and quintuple income growth for Kiribati. (Income is here defined as Gross National Income and net income from migrants.) Even without further reforms, remittances to the Pacific island countries are forecast to more than double by 2040. They are currently estimated at $US590 million. With average incomes rising in the OECD and the migrant stock increasing, remittances (from non-seasonal workers) and net income (from seasonal workers) are expected under business as usual to reach $US1.5 billion Funds sent home would increase sharply with reforms. Remittances of non-seasonal migrants and the income of seasonal workers increase by $US400 million and $US800 million respectively under the two scenarios. The small size of the Pacific makes reforms that are transformative for the region doable for the metropolitan countries. Under the scenarios, Pacific migrants are projected to account for just one per cent or less of the Australian population, and 6-7 per cent of the New Zealand population. There are real benefits not only for the Pacific island countries but for destination countries such as Australia and New Zealand in terms of filling labour shortages using labour from the Pacific. The positive experiences compiled in this report show what is possible. By enhancing Pacific labour mobility, Pacific isolation can be overcome. xi 1. Introduction Increased international migration for work offers critical opportunities for the Pacific. There is now broad consensus that, given the unique development challenges faced by the Pacific Island countries, expanding labour mobility is vital for their future. The Pacific Island economies tend to be impeded, albeit to differing extents, by their small size, remoteness from major markets, internal population dispersion and weak governance. These barriers to growth push up the costs of private production and public administration and limit job growth. As a result, the economies of the Pacific island countries have not been able to provide sufficient employment opportunities for their people. Employment rates are estimated to be less than 50 per cent of the working age population in most countries. In the four Melanesian countries, an estimated 3 million more people aged 20 to 64 years will be part of the working age population by 2040. Where these countries are unable to bring jobs to the people, the alternative is to bring the people to where the jobs are. These jobs are primarily concentrated in the larger, more densely populated countries around the Pacific Rim. International migration offers the potential for a ‘triple win’, delivering gains for migrants, sending countries, and receiving countries. For Pacific Islanders, migration generates significant employment opportunities. For labour-sending countries, remittance flows contribute to increased income and consumption at the household level, reducing poverty, loosening credit constraints and providing insurance against negative shocks. Remittances are often invested in education and health, with positive flow-on effects for human capital development. They ‘help finance trade deficits and bolster financial reserves at the macroeconomic level’ (World Bank 2014a, 2). Labour-receiving countries in the Pacific Rim benefit from the filling of labour shortages and by using migration access as a major policy lever to lift the prosperity, security and stability of their low-growth neighbours. The benefits for both receiving and sending countries are discussed in greater detail below in Section 2. The ability to access external labour markets varies significantly across the Pacific. Some Pacific islands have free labour market access to the United States or New Zealand. Others have amongst the lowest rates of outward mobility in the world. When it comes to labour mobility, Pacific island countries can be divided into four groups. Marshall Islands, the Federated States of Micronesia and Palau – all of whom have signed the Compacts of Free Association with the United States – are the only Pacific countries in this study with open labour market access. They are referred to in Figure 1 as the ‘open access’ countries. Despite not having the same privilege, Fiji, Samoa and Tonga have high outward migration rates through historical ties as well as bilateral agreements providing limited access to New Zealand. They are referred to as the ‘high mobility’ countries. The three Melanesian countries of PNG, Solomon Islands and Vanuatu have amongst the lowest rates of outward migration in the world, due to severely restricted external labour market access. They are the ‘low mobility’ countries. The two remaining countries in this study, Kiribati and Tuvalu, are isolated, low-lying atolls. Tuvalu but not Kiribati has a relatively high labour mobility rate, but neither has the open access of Marshall Islands, 1 another Pacific atoll country. Due to their physical geography, and their limited labour mobility opportunities, Kiribati and Tuvalu are the Pacific countries most at risk from projected sea-level rise. This requires that they be given a special treatment. They are the ‘atoll’ countries. Figure 1 Country groupings Open access High mobility Marshall Islands Fiji Federated States of Micronesia Samoa Palau Tonga Low mobility Atolls (climate affected, and lacking Papua New Guinea open labour market access ) Solomon Islands Kiribati Vanuatu Tuvalu Table 1 over page provides some basic statistics for these four country groupings. Some stark contrasts are evident. PNG, Solomon Islands and Vanuatu have some of the lowest remittance/GDP ratios in the world, while Samoa and Tonga are at the other end of the spectrum. The stock of Pacific migrants is growing, but is unevenly distributed. Around 1950, there were fewer than 15,000 Pacific-born migrants in developed countries. Throughout the latter half of the twentieth century, the rate of migration from the Pacific expanded greatly (Bedford and Hugo 2012). This occurred largely due to immigration to the USA and New Zealand, facilitated by special arrangements. As of 2013, the total stock of Pacific-born migrants resident in OECD countries (including the US territories of Guam and CNMI) is approximately 420,000 (Table 1). Four-out-of- five (79 per cent) come from the high-mobility countries, one-in-six from the open access countries (15 per cent), a small proportion from the low-mobility countries (5 per cent) and hardly any (1 per cent) from the atoll countries. More Tongans are overseas than all the migrants of the low mobility and atoll countries put together. The poorest Pacific countries have the fewest labour mobility opportunities. For some countries, notably Fiji, a higher GDP per capita reflects higher skill levels, which result in greater labour mobility opportunities. But for most countries, it is simply an accident of history. In particular, PNG, Solomon Islands, Vanuatu and Kiribati are both poor and isolated, with few labour mobility pathways. 2 Table 1 Basic labour mobility and economic indicators for different Pacific country groupings Stock of Resident Emigrants/ Remittances GDP p.c. emigrants population population /GDP (2005 $US) Open labour market access Marshall Islands 11,841 52,786 22.4% 13.8% 3,045 Micronesia (Fed. States of) 40,642 103,718 39.2% 9.1% 2,338 Palau 6,855 20,919 32.8% 5.6% 8,731 High mobility Fiji 189,571 880,487 21.5% 6.2% 3,828 Samoa 87,949 190,390 46.2% 28.2% 2,668 Tonga 53,247 105,139 50.6% 45.8% 2,502 Low mobility PNG 17,464 7,308,864 0.2% 0.2% 1,122 Solomon Islands 1,768 560,685 0.3% 2.8% 1,125 Vanuatu 2,280 253,165 0.9% 5.9% 2,089 Climate affected atoll countries Kiribati 4,324 108,544 4.0% 11.1% 1,109 Tuvalu 1,816 9,876 18.4% 16.2% 2,654 Note: Due to data constraints, emigrants are to OECD countries only. Emigrants to the US territories of Guam and CNMI are included. Migrant stocks for Kiribati, PNG, Solomon Islands and Vanuatu have been adjusted to record only migrants in Australia who claim their ancestry is indigenous to their country of birth. The population used as a denominator in the third column is the resident population only. Remittances are gross not net, and include income from seasonal workers. Figures are for 2013, or for the stock of migrants 2011-2013, depending on the OECD countries concerned. Source: World Development Indicators 2015 for GDP per capita, for the most recent year available. Population statistics are taken from the UN Population Division (2015). Migrant stocks in 34 OECD countries taken from the DIOC Database on Immigrants in OECD Countries 2010-13. For more details, see Arslan et al. (2014). Remittances are from the World Bank Migration and Remittance database, FSM is an exception, with remittance data based on an authors’ estimate. The inflows of remittances are unequally distributed. Remittances remain relatively small for the majority of countries, but significant for a few. Remittance data for the Pacific is unreliable and should be treated with caution. Nevertheless, it is clear from Table 1 that, for the low-mobility countries, low stocks of migrants translate into a trickle of remittances. Remittance inflows are reported to equivalent to 6 per cent of GDP for Vanuatu, but only 3 per cent for the Solomon Islands, and 0.2 per cent for PNG. For the open-access and atoll countries, the value of remittances is between 6 and 17 per cent of GDP. Two of the high-mobility countries sit amongst the highest remittance receiving countries in the world. As Table 1 shows, Tonga receives remittances equal to about 46 per cent of its GDP and Samoa 28 per cent. Fiji has the highest inflows of remittances in absolute terms, but its remittance/GDP ratio is only 6 per cent, reflecting the size of its population and economy. The challenge of overcoming Pacific isolation has to be pursued in a targeted way. Not all of the Pacific is isolated in terms of human movement, but substantial chunks are, and, worryingly, the poorest parts tend be the most isolated. Those countries which have the least access and which are the most vulnerable need to have top priority. From this perspective, the report focuses in particular on expanding labour mobility access for the low mobility countries of PNG, Solomon Islands, and Vanuatu and the atoll nations of Kiribati and Tuvalu. 3 Pacific migrants go mainly to one of three countries: Australia, New Zealand and the USA. The report mainly focuses on reforms for Australia and New Zealand because of the strategic importance of the Pacific to them. The paper also considers Korea, because of one particular Korean scheme which could be of significant benefit to the Pacific if it was extended to the countries of the region.2 Figure 2 The three major destination countries for Pacific Islanders 160 4.0% Thousands 140 3.5% 120 3.0% 100 2.5% 80 2.0% 60 1.5% 40 1.0% 20 0.5% 0 0.0% Australia New Zealand USA Stock of Pacific migrants Pacific migrants/population Source: See Table 1. This report looks at the issue of labour mobility from both sides of the equation. It seeks to understand the perspectives of both sending and receiving countries, it proposes reforms for both sets of countries, and it tries to address the concerns of both. The report places particular emphasis on providing opportunities for low-skilled and medium-skilled migration. A broad range of reforms is proposed, but we are particularly interested in providing labour mobility opportunities for those beyond the elites of the sending countries. This is to maximize the welfare gains of migration, and to reduce the risk of brain drain. It provides an indicative quantification of the impact of its proposed reforms. The report lays out a ‘business as usual’ baseline scenario to project forward labour mobility and remittance flows for seasonal, temporary and permanent migrants. It then sets out a ‘medium growth’ scenario and a ‘high growth’ scenario in which labour mobility opportunities are significantly expanded for the countries that most need them. The benefits are assessed by examination of a broader set of 2 Papua New Guinea and Palau are the two Pacific countries that attract a large number of foreign workers. However, very few of these are from other Pacific countries. In the case of Palau’s population of 15,109 in 2005, one-in-three (32 per cent) were persons born overseas, mostly from Asia and the Philippines in particular. In the case of Papua New Guinea, the total number of foreign workers on work permits in May 2015 was 41,096. This represents 12 per cent of its small formal sector of 340,000 (Voigt-Graf 2015a). Of the total number of foreign work permit holders in May 2015, 413 were from Fiji, 1 per cent of the total. Some 53 work permit holders were from Solomon Islands, 24 from Tonga, 8 from Kiribati, 5 from Samoa and Vanuatu each, 2 from Tuvalu and 1 from FSM (Voigt-Graf 2015b). While in the future PNG might become a more important destination for Pacific island countries, it is not yet. 4 economic outcomes. While this exercise is only indicative in nature, it is revealing, and confirms that the currently unequal and limited distribution of labour mobility opportunities in the Pacific can be addressed. This report is based on extensive and new research. It draws on research undertaken earlier on Australia’s Seasonal Worker Programme (Hay and Howes 2012; Doyle and Howes 2015). It also draws on commissioned work relating to: Pacific migration pathways (Curtain 2016a) Timor-Leste’s experience of providing low-skilled labour to South Korea (Wigglesworth and Fonseca 2016); new analysis of New Zealand’s Pacific Access Category and Samoa Quota (Curtain 2016b); analysis of Kiribati and Tuvalu’s labour migration strategies (Voigt-Graf and Kagan 2016); a study of aged-care labour needs and related labour mobility schemes (Curtain 2016c), and a study of the impact of migration on the demography of Kiribati and Tuvalu (Bedford and Bedford 2016). These background papers are currently being finalized for publication.. The reports presents a menu of reforms. Not everyone will agree with every idea presented in the paper. Its purpose is to stimulate thinking and new approaches. It proceeds as follows. Section 2 sets out the benefits of labour mobility for all the various parties involved. Sections 3 to 6 look at various receiving-country reforms. Section 3 discusses the proposed reforms and their rationale in general, while Sections 4-6 respectively look at seasonal (less than 1 year), medium-term (1-5 years) and long-term (>5 years) labour mobility options. Section 7 looks at what the sending countries need to do. Section 8 quantifies the gains from increased labour mobility. Section 9 concludes. 2. Labour mobility benefits The benefits of international labour mobility represent a potential ‘triple win’ for the migrant and his or her household, for the receiving economy and for the sending country. This section sets out the case for labour mobility from the perspective of all three of these units of analysis, with a particular focus on the Pacific and its major destination countries. 2.1 Benefits to migrants and their families There is ample evidence that for the Pacific the economic benefits of migration for individuals and their families are large. According to one Tongan (quoted by Small 1997), ‘every family needs to have someone overseas. Otherwise the family is to be pitied.’ Gibson and McKenzie (2010) found strong evidence of the economic and social benefits of seasonal migration in Tonga and Vanuatu. Four waves of specially conducted surveys fielded in Tonga and Vanuatu between 2007 and 2010 showed large positive effects on the households supplying seasonal workers to New Zealand. These workers earned an estimated $NZ12,000 each. After deducting costs for accommodation, food, health insurance and contribution to airfares, the net return was about $NZ5,500 per worker (Gibson et al. 2014, 26). In both Tonga and Vanuatu, the per capita income of participating households rose by over 30 per cent relative to the comparison groups, while per-capita expenditure and savings also increased. 5 Migrant households made home improvements. They were more likely to open bank accounts and the school attendance rates for 16 to 18 year olds in migrant households increased. Another survey showed large, sustained benefits to longer-term Pacific migration. A survey of Tongans who were selected to migrate permanently through the Pacific Category visa ballot – compared with those who applied but did not migrate – showed a long-term gain in income to the migrant’s household. Migration resulted in large sustained benefits to the migrant’s immediate family, due to have higher consumption, ownership of more durable assets, more savings and dietary diversity (Gibson et al. 2015). Sometimes there are concerns for the families of migrants, if they are left behind, but in the Pacific the evidence is that the families of seasonal workers are better off, at least in terms of education. Children from the households of seasonal workers in Tonga are 10 to 14 per cent more likely to be attending school than children from other Tongan households (Gibson and McKenzie 2010). Other concerns regarding the social impacts of migration, and measures that can be taken by sending governments, are addressed in Section 7.3. 2.2 Benefits to sending countries The Pacific countries are experiencing a youth bulge. Analysis of the projections of country-level populations by age produced by the UN Population Division shows that nearly half (45 per cent) of the youth population in 2016 in the four Melanesian countries is aged 15-29 years (UN Population Division 2015). This share will remain high to 2025 and then will decrease to 40 per cent in 2040. Samoa and Tonga have even higher youth bulges, despite more opportunities to migrate. In both Samoa and Tonga, the share of the youth population 15-29 years is estimated to increase to 48 per cent and 50 per cent respectively by 2025, before falling to 43 per cent and 42 per cent respectively by 2040. Labour mobility will help address the Pacific’s chronic unemployment problems. The mismatch between labour market entrants and job opportunities in the Pacific is stark, as shown by Table 2 below. Only a very small fraction of those entering the labour force can be employed in the formal sector in the Pacific. As a result, underemployment is rife, and social tensions and poverty can be acute, especially when there are large numbers of unemployed young people. 6 Table 2 Number of labour force entrants and formal sector job creation in various Pacific island countries Country Labour force entrants Formal sector jobs (annually) created (annually) Kiribati 3,200 65 Papua New Guinea 87,000 11,932 Solomon Islands 13,000 2,089 Tonga 5,600 325 Vanuatu 3,800 1,260 Sources and notes: Kiribati: Census (2005; 2010); PNG: Census (2000; 2011); Solomon Islands: Census (1999, 2009); Tonga: Census (2006; 2011); Vanuatu: Census (1999, 2009). These numbers are rough approximations only, and comparability across countries may be limited due to different census definitions. Expanding labour mobility opportunities can help stabilize fragile states. Of the 29 countries rated by the World Bank as fragile in 2016, five are in the Pacific: Federated States of Micronesia, Kiribati, Marshall Islands, Solomon Islands, and Tuvalu. The Asian Development Bank also rated Nauru as fragile. Three Pacific countries, Solomon Islands, Papua New Guinea and Fiji also appear on the 2015 Fragile States Index, compiled by the New York-based The Fund for Peace, used by the OECD. While migration alone will not make a sending country prosperous, by reducing employment pressures, and providing valuable foreign exchange, labour mobility acts as a stabilizer. Remittances reduce poverty. Household income in the poorest quintiles strongly increases when remittance income is included, whilst the poverty headcount drops significantly, as much as 25 percentage points in the case of Tonga (World Bank 2006). Remittances in many cases represent a return on investments in (human) capital, and it is appropriate that they be used for consumption, but the evidence shows that many households do invest remittances in community assets (Gibson and McKenzie 2010). In well-connected Pacific countries, remittances are large enough to be significant at the macroeconomic level. For example, in 2010, the savings of seasonal workers alone injected an estimated total of $NZ5.3 million in Tonga and $NZ9.7 million in Vanuatu which is the equivalent to 42-47 per cent of total annual bilateral aid from New Zealand to these countries (Gibson et al. 2013, 9). The funds were equivalent to almost one half of annual export earnings for Tonga and one quarter of annual export earnings for Vanuatu (Gibson et al. 2013, 9). Migration has a beneficial impact on human capital development. It is no coincidence that in the high-mobility countries, where remittances are highest as a percentage of GDP, development indicators are also high. The causality runs in both directions. The extra income received from working in higher income countries can ‘relax borrowing constraints faced by poor households that previously hindered human capital investments’ (World Bank 2014a, 69). In the case of Tonga, remittances have been shown to have a positive impact on both school attendance and post- secondary education attainment (World Bank 2006). Opportunities for migration generate incentives for human capital development for the remaining population, which can potentially offset 7 the loss of human capital experienced in labour-sending countries (Chand and Clemens 2008). The issue of brain drain is addressed further in Section 7, where we consider policy responses sending and receiving countries can put in place to minimize this risk. Knowledge transfer is another benefit of migration. Both low-skilled and skilled Pacific migrants can transmit knowledge and skills to their compatriots both upon return and while abroad. Gibson and McKenzie (2010) find that 13 per cent of expatriate Tongans from a sample of academic high achievers provided advice to the Tongan government while overseas, and more than 50 per cent provided advice to Tongan residents on study and work opportunities. Migration can also change world-views, and thereby support domestic reform. 2.3 Benefits to labour-receiving countries International migration has a clear set of economic benefits for labour-receiving countries. For receiving countries, the immediate economic benefit of having migrants available to work is to expand the output of goods and services. Migrants often plug labour shortages in important sectors that struggle to attract native workers. At the macroeconomic level, migration can improve a country’s growth prospects and the sustainability of its public finances (Lagarde 2016). Evidence suggests that migration generally leads to an increase in GDP (World Bank 2014a). Existing estimates suggest that the potential net benefits to receiving countries from reducing barriers to international migration are large and much bigger than trade liberalization alone (World Bank 2016, 208). Advanced economies will require high rates of net migration in coming years to address major labour market shortfalls. These are often in sectors – such as construction, health care, and social assistance – where Pacific nationals, with some targeted training, would be well placed to fill the gaps. In Australia, over the five years to 2020, the health care and social assistance sectors are projected to make the largest contribution to employment growth, increasing their number of jobs available by 250,200. Within this sector, hospitals are due to increase employment by an estimated 54,000 jobs, residential care by an estimated 47,000 jobs and allied health services by 36,000 jobs (Department of Employment 2015). In New Zealand’s case, about a quarter of the total employment growth over the next three years to 2018 is forecast to occur in construction and related activities. Construction activities across New Zealand are projected to grow by 31,000 jobs. This is likely to be due to residential construction demand in the Auckland region and the Canterbury reconstruction. Another sector forecast to increase its employment significantly is hospitality by 11,600 jobs. The health and education sectors combined are expected to increase by 17,200 (Ministry of Business, Innovation and Employment 2015). Working out a Pacific labour mobility regime would help Australia, in particular, move beyond the current ad hoc and unsatisfactory arrangements for importing unskilled labour. Australia gives access to migrants below the skill threshold in numbers that far exceed those granted access as skilled migrants. Unskilled job needs are met by holders of visas that are designated for 8 other purposes but with work rights such as working holiday makers, the partners of temporary skilled workers, New Zealand citizens and international students (able to work up to 20 hours a week during term, and without limit at other times). In 2014-15, some 622,000 visas in these categories were approved for entry to Australia. The forms of access and level of scrutiny by authorities of these lower skilled workers make many of these workers vulnerable due to weak bargaining power. The 2016 report of the Australia Senate, A National Disgrace: The Exploitation of Temporary Work Visa Holders, has highlighted the vulnerability of these lower skill workers without more structured arrangements in place. Reforming the existing Pacific Seasonal Worker Programme to support its expansion, and introducing some of the other programs proposed by this report (in particular, introducing an equivalent of New Zealand’s permanent ‘Pacific Category’ visa) would make an important contribution to strengthening the integrity of Australia’s low-skill immigration regime.3 Pacific workers do well in metropolitan settings. Administrative data and various studies of Pacific migrants in New Zealand highlight their value to the economy. The preference of New Zealand horticultural employers for workers from the Pacific, mainly from Vanuatu, Tonga, Samoa, and Solomon Islands, is shown by the large and increasing share of Pacific workers recruited to work under the NZ Recognized Seasonal Employer (RSE) horticultural scheme, despite competition with workers from Thailand, Malaysia, Indonesia and the Philippines in the same scheme. The share of Pacific workers has risen from 74 per cent in 2009-10 to 85 per cent in 2014-15 of all RSE workers. An evaluation of the RSE, based on employer surveys, found that employers appreciated the dependability and enthusiasm of Pacific workers. The evaluation noted that the flow-on effects for employers of having a reliable workforce were reduced recruitment and training costs, increased confidence to expand and invest, and reduced stress (Evalue Research 2011). Evidence based on wage rates and attendance confirms the employer feedback and shows that Pacific workers are more productive than others in terms of higher attendance at work, and higher output as reflected in their higher wages with less variability in wages within the group (Bedford and Bedford 2015, 14). The beneficial impact of Pacific migrants on the New Zealand economy is also confirmed by the high employment rate of recent Pacific Category migrants. The employment rate of recent Pacific working-age migrants to New Zealand under that country’s permanent migration Pacific Category (discussed in more detail in Section 6.2) ranged over the five year period from 2007 to 2011 from 79 to 72 per cent (Merwood 2013a). The fall in the employment rate reflects the impact of the world recession from 2009, but overall this rate confirms the employability of Pacific migrants to New Zealand. Census and related data on employment rates for Pacific migrants for dates between 2011 and 2013 for Australia, Canada, New Zealand, UK and USA show that 77 per cent of Pacific men and 64 per cent of Pacific women aged 25-64 years are in paid work (DIOC 2014).4 The employment rates for 3 New Zealand relies on some of the same sources but already benefits from larger inflows of Pacific migrants to fill its unskilled labour needs. 4 Papua New Guinea and Solomon Islands have been excluded from this analysis due to the large number of migrants in Australia who did not report an ancestry indigenous to their country of birth. 9 Pacific migrants of prime working age with medium and high education levels do not differ in large part from the employment rates of all other migrants in the same OECD destination countries. In the case of the low-educated Pacific migrants of prime working age, women have a higher employment rate than all other migrants in these five main OECD destinations (45 per cent compared with 41 per cent). However, low educated Pacific men aged 25-64 years have a slightly lower employment rate than all other migrants. The economic contribution of Pacific migrants in terms of employment and taxes to their receiving countries is significant and in total differs little from all other migrants. Increasing access to labour mobility would allow Australia and New Zealand to broaden their linkages – cultural, social and economic – with their regional partners to build confidence and trust. Migration is increasingly seen as a mechanism, like trade, of mutual benefit between regional partners. Australia and New Zealand themselves have an open access migration policy called the Trans-Tasman Travel Arrangement, which formalized in 1973 a long-established practice of movement between the two countries. The cross border movement of people is now seen by both governments as one of the four key elements, alongside trade in goods, trade in services, and the free flow of capital, that promote closer economic ties between the two countries. Australia’s free trade agreement with China (December 2015) included a provision to give young people with at least some tertiary education and basic English language ability access to Australia’s Work and Holiday visa with a cap of 5,000 per year. Special migration access is also given for broader geopolitical reasons such as enhancing national security. The Compact of Free Association between the United States and three Pacific countries provides migration access in return for the power of ‘strategic denial’ in the oceans between these countries. Australia and New Zealand both have a strong interest in countering the fragility observed in the Pacific. Australia’s 2013 National Security Strategy highlighted climate change and changing demographics in fragile states as important factors affecting national security: both will impact on Pacific countries in coming years. Better migrant access can greatly extend and deepen the impact of a country’s foreign aid program. Australia has a foreign policy objective of ensuring that neighbouring countries are ‘secure, stable and prosperous’ (DFAT 2015, 15). New Zealand’s has stated a similar objective in the following words: 'Maximize the impact of New Zealand's engagement in improving the prosperity, stability, and resilience of the Pacific Islands region and its people' (MFAT 2015a, 6). Both Australia and New Zealand target large shares of their foreign aid programs at countries in the Pacific.5 Many of the countries in the Pacific are at the top of the list of the most aid dependent countries in the world.6 5 Australia in 2015-16 plans to spend $A911 million or 23 per cent of its total expenditure on overseas development assistance on countries in the Pacific. This amount is nearly half of all the development assistance allocated to country and regional programs. New Zealand has allocated to the Pacific region near to half (46 per cent) of its expenditure of $NZ544 million on overseas development assistance 6 A World Bank indicator of aid dependency using data for 2014 on 129 countries shows that the following Pacific countries in rank order are in the top 15 countries: Tuvalu, Federated States of Micronesia, Kiribati, Marshall Islands, Tonga and Solomon Islands. Vanuatu and Samoa are in the top 25 most aid dependent countries and Palau ranks 28th on the list. 10 Reliance on aid programs alone is an unbalanced strategy. In marked contrast to aid, migration offers self-selected individuals and their households the chance to change their economic and social circumstances. Migration offers the opportunity for households themselves to acquire their own capital to invest in education and business activities. Funds generated by migrants go directly to households. And migration, much more than aid, requires of those involved major changes in attitudes. Greater promotion of Pacific labour mobility can help promote the broader foreign policy objectives of both countries. As a former Secretary of the Australia’s Department of Foreign Affairs has noted: The challenge for Australia is the need ‘to deploy foreign policy, trade, economic and development tools in an integrated fashion’ (Varghese 2016). 3. Receiving-country reforms: an overview What needs to be done to enhance labour mobility opportunities for those Pacific countries that need them the most? The next three sections look at reforms receiving countries could put in place if they wanted to expand labour mobility opportunities for the Pacific, especially for the low- mobility and atoll countries. The Pacific is fortunate to be close to countries of migration. New Zealand, Australia and the US are all countries of migrants. In 2013, Australia and New Zealand both had 28 per cent of their total resident population born overseas. This is close to the highest migrant share of all 36 OECD countries (OECD 2015, 35). The USA, with 13 per cent of its total population born overseas in 2013, is in the middle rank order of OECD countries. The Pacific-born proportion of the total migrant population in the US and Australia is negligible, and it is small in New Zealand (3 per cent in 2013; see Figure 2). Countries have a range of tools at their disposal to increase migration from the Pacific. They can use their aid budget to fund ‘aid for migration’ schemes. This can be considered as a type of ‘aid for trade’, a type that is at least if nor more relevant for many Pacific nations than trade facilitation. A good example of this is using the aid budget to promote technical skills that can be used to qualify for existing skills-based migration systems. The Australia Pacific Technical College is discussed in detail in Section 5.3 of the paper. Receiving countries can also modify their migration regime to favour citizens of particular countries. It is a myth that these countries run non-discriminatory migration programs. For example, Australia plays host to hundreds of thousands of backpackers. As discussed in Section 5.1, nearly all (95 per cent in 2013-14) are from OECD countries because backpacker visas for non-OECD countries are capped (typically at very low levels), while those for OECD countries are uncapped. Migration reforms requires countries to ask where they want their migrants from. This can be an uncomfortable question to ask. However, it is in the interests of both Australia and New Zealand to promote Pacific migration. The small size of the Pacific islands makes the required reforms manageable. 11 A wide range of reforms is proposed. The interventions put forward are separated by duration into the following categories: seasonal worker programs (for less than 12 months); temporary migration schemes (1-5 years); and permanent migration (greater than five years). As noted earlier, the aim is to provide a menu rather than a road-map. Long-term migration brings the greatest benefits, but temporary migration opportunities provide important routes to permanent residence. Long-term migration based on permanent residence has the largest economic impact because the annual inflows of migrants accumulate over time. However, more temporary forms of migration are also important because of the skills and experience they provide, because they promote circular migration, because they are sometimes more open to the less-skilled, and because they provide a route to more permanent forms of migration. The full benefits of migration can only be realized if a variety of migration options is available. While the main focus has to be on increasing labour mobility, reducing the cost of sending remittances is also important. The Pacific countries are some of the most expensive in the world when it comes to sending remittances to them. The average cost of sending remittances to the Pacific from Australia and New Zealand is 11.5 per cent and steady, compared to the global average of 7.5 per cent and falling (Betteridge and Howes 2015). Improving the portability of superannuation to the Pacific Islands, as New Zealand has done, would be of particular importance to temporary migrants. Temporary migrants are not granted access to social security payments in Australia or New Zealand. However, both benefit from compulsory superannuation accumulated from employer contributions (3 per cent in New Zealand and 9.5 per cent in Australia). New Zealand does not have any bilateral social security arrangements allowing reciprocal portability of benefits with any Pacific Island country, but it has unilaterally included a provision to sections of the New Zealand Superannuation and the Retirement Income Act 2011 that provides for enhanced portability of New Zealand superannuation to 22 Pacific countries (Woolford 2009). The provision is known as the Special Portability Arrangement and means that any superannuant, not just former Pacific Island migrants, can receive their full entitlement in any of the 22 countries (including all those in this report). Australia should consider putting in place a similar arrangement. Whilst most workers who leave Australia before reaching the retirement age are able to take their accumulated savings with them, they are required to pay an additional 30 per cent withholding tax on the money they take out of the country (Woolford 2009). 4. Expanding seasonal worker access Seasonal worker programs are an increasingly important pathway for Pacific labour mobility. Almost 12,000 workers from the Pacific arrive in Australia and New Zealand annually under the former’s Seasonal Worker Programme (SWP) and the latter’s Recognised Seasonal Employer (RSE) scheme. Pacific countries (other than the Compact states) also have access to the United States H-2A Temporary Agricultural Visa, but no Pacific workers have participated in it to 12 date. The RSE is focused entirely on horticulture; this is the main sector for the SWP, but seasonal work in tourism in Northern Australia is now also covered by the SWP as are a number of other agricultural low skill jobs. New Zealand’s RSE has outperformed Australia’s SWP. In Australia’s SWP, the number of Pacific seasonal workers has increased since the scheme’s introduction in 2008.7 However, it remains small in comparison both to the New Zealand’s RSE and to the total number of foreign workers operating in the agriculture, accommodation and tourism sectors in the locations where it is permitted to operate. The SWP cap was removed in 2015, but it was non-binding prior to removal. The real constraint in Australia is the lack of an aggregate labour shortage due to the number of backpackers working in these industries, as well as the number of illegal workers (the two categories partially overlap). A survey of employers shows that there is a lack of awareness of the scheme (Doyle and Howes 2015). Most of the growers who are aware of the scheme feel the costs and risk involved with the program are excessive and the scheme’s administrative requirements are considered burdensome. Figure 3 Visas issued to Pacific Islanders under New Zealand’s Recognized Seasonal Employer (RSE) and Australia’s Seasonal Worker Programme (SWP), 2007-08 to 2013-14 8000 7000 6000 5000 4000 3000 2000 1000 0 2007-08 2008-09 2009-10 2010-11 2011-12 2012-13 2013-14 Australia's SWP New Zealand's RSE Source: Administrative data from both countries. New Zealand’s RSE does not suffer from the same demand-side constraints as Australia’s SWP. Continuing to increase the RSE cap would support the expansion of the sector, and improve access. The RSE started with a cap of 5,000 workers in 2007. The cap increased to 8,000 the following year, then to 9,000 in 2014 and then to 9,500 late in 2015. The RSE has consistently managed to reach 95-100 per cent of its cap (mainly with Pacific islanders, but about 15 per cent of RSE workers come from South-east Asia). The biggest constraint on future RSE growth is the cap, which remains in place to protect the jobs of New Zealanders. There is a severe shortage of New Zealanders wanting to work on farms, and there are constant calls by the New Zealand horticultural 7 It was introduced as a pilot in 2008, and converted to a permanent scheme in 2012. 13 industry to increase the caps, and complaints about worker shortages. The sector as a whole plans to double the value of its output by 2020 (Horticulture NZ 2009). It is hard to see how this would be possible without ongoing increases in the number of seasonal workers. As noted earlier (Section 2.3), there is significant evidence from New Zealand of productivity gains from the RSE. Benefits to employers include a more reliable workforce, reduced recruitment and training costs, increased confidence to expand and invest, and reduced stress (Evalue Research 2011), as well as higher productivity (Bedford and Bedford 2015). A number of reforms are needed to expand Australia’s Seasonal Worker Programme. The second-year visa extension provided for working holiday makers or backpackers if they work on a farm for three months was put in place in 2005 when Australia was unwilling to introduce a seasonal work migration scheme. It leads to direct competition between backpackers and seasonal workers, competition which the backpackers are winning. In 2014-15, there were 41,339 second-year backpacker visa applicants, some 90 per cent of them on the basis of farm work completed (DIBP 2015a). It is unfortunate that a scheme that benefits the citizens of poor countries is undermined by one that benefits the citizens of the rich. Either the second year visa extension should be done away with, or it should be made available in return for three months of work in any sector. Advertising the SWP through a targeted group of agricultural and accommodation industry bodies would help raise awareness. Removing the $A500 employer contribution to the international airfare of returning workers would help ease cost concerns. Removing the requirement on employers to make a superannuation contribution would increase the take home pay of seasonal workers and lessen the paperwork for employers. Giving all employers participating in the scheme the primary role in the selection of workers to ensure that they suit their work requirements would address some of the inherent risks with the program (Doyle and Howes 2015). Another barrier to growth for the Australian SWP is the prevalence of illegal labour in the Australian horticultural sector. Greater enforcement of compliance with the law would help address this issue, and growing community concern may also lead to a response from employers. In New Zealand, a key driver for the increased employment of seasonal workers has been the desire of NZ employers to meet the requirements of export markets. European markets require exporters to meet a supplier code of practice called GlobalGAP.8 It places a strong emphasis not only on high product quality but also on the health, safety and welfare of workers involved in the harvest. New Zealand growers seeking to export want a stable, committed and trained seasonal workforce to harvest a high-quality product and to reduce waste. They also want to remove any threat to their export trade through bad publicity in Europe by stopping the use of illegal workers and improving the working conditions of the workforce (Whatman and Van Beek 2008). In Australia, while export markets are much less important, growing media attention in Australia to stories of worker exploitation has the potential to foster a campaign among consumers to put more pressure 8 A European Retailers standard. The aim of GlobalGAP is to ensure integrity, transparency and harmonization of global agricultural standards. This includes the requirements that food is produced respecting worker’s health, safety and welfare, the environment and sustainable land use. 14 on the major supermarket chains. An important practical step in this regard would be to introduce registration of labour hire companies, as recommended by two recent Parliamentary inquiries. A further problem with Australia’s SWP is that few places have been obtained by the low- mobility countries. Vanuatu provides about 40 per cent of RSE labour requirements, but Samoa and Tonga provide about 80 per cent of SWP requirements, with Tonga by far the largest single provider. Of course, this is good for Tonga, and a credit to the country, but it does little to direct remittances to where they are most needed in the Pacific. Vanuatu is the third largest sending country in the SWP. The Solomon Islands and PNG, countries with relatively large numbers of agricultural workers, have sent very few seasonal workers. Ultimately the two problems can only be solved together. It is only if the Australian SWP grows that many more positions can be found for Pacific islanders from low-mobility countries. (It is unrealistic to expect seasonal worker programs to be major employers for the atoll countries.) It will also be important to consider ways that women can benefit from growth in the SWP and RSE. At present, even though Pacific agriculture is heavily reliant on female labour, the participation rate of women in both schemes is less than 20 per cent. This is mainly due to opportunities being highly concentrated in horticulture and viticulture, sectors where women face barriers to participation on both the supply side (through local recruitment practices) and the demand side (from ANZ employers giving preference to male workers). Whilst the seasonal schemes are industry led and demand-driven, all governments need to work together to combat discrimination. As the schemes expand further into sectors with greater gender parity in employment, such as the tourism sector, it is likely that Pacific women will capture a greater proportion of the direct benefits from seasonal employment. How much growth in seasonal work flows can be expected? Despite the constraints on demand in Australia, the SWP is steadily growing. Even without policy reforms (an increase in the cap in New Zealand, and measures to stimulate demand in Australia), we estimate that by 2040 the two countries’ seasonal work schemes will have increased in size to 29,300, up from 11,800 in 2015. With reform, double that number is possible, with about 57,500 workers arriving under the two schemes each year. It is reasonable to assume that New Zealand doubles its cap to 20,000 (in line with plans to double the size of the horticultural sector; most but not all the positions would go to Pacific Islanders). And the World Bank (2014b) presents a scenario in which the Australian SWP attracts 40,000 workers a year. 5. Expanding medium-term or temporary Pacific labour mobility options (1–5 years) Skilled, semi-skilled and low-skilled labour mobility programs of longer but still temporary duration are little utilized by the Pacific but could become important. This section looks at four programs through which medium-term or temporary labour mobility opportunities could be 15 expanded for the Pacific: working holiday maker programs; Korea’s Employment Permit System; Australia and New Zealand’s temporary skill visas; and a proposed new aged care program. 5.1 Working holiday maker (backpacker) programs Most OECD countries offer working holiday visa programs with unlimited access to young visitors from other OECD countries. The original purpose of these schemes was to promote international understanding by enabling youth to experience the culture of another country. As the name of these visas suggest, work is intended to be incidental to travel. However, some OECD countries rely on backpackers to fill low-skilled labour shortages. Australia does this more than any other OECD country, attracting in 2013 some 249,000 backpackers, about half of all working holiday makers in 22 OECD countries in that year (OECD 2015, 26). Access and prerequisites are biased in favour of developed countries. In 2013-14, 95 per cent of Australia’s backpackers came from developed countries (calculated from data presented in DIBP 2015b). This bias denies Pacific migrants access to a migration pathway that could result in permanent residence or a temporary skilled work visa. According to official statistics, for the period 1991 to 2014, close to one-in-five of working holiday makers became permanent residents (Productivity Commission 2015, Figure 11.4). In 2012-13, some 14 per cent of temporary skilled work visas (Subclass 457) were granted to working holiday visa holders (Azarias et al. 2014, Table 5, p. 36). There are no limits on the number of visas that can be issued to the citizens of 19 developed countries under the Working Holiday Maker visa (Subclass 417). Citizens of 16 developing and developed countries are subject to a quota ranging from 100 to 1,500 for the much smaller Work and Holiday visa (Subclass 462). Applicants from these countries must, in addition to the rules already mentioned, be university graduates, pass English tests, and (except for Chinese nationals) have a letter of support from their government.9 The New Zealand Working Holiday Schemes have similarly discriminatory eligibility requirements (Immigration NZ 2015a). Australia and New Zealand could open up their working holiday maker schemes to their Pacific island neighbours. So far, Australia has signed an agreement with PNG (for 100 places). However, the agreement is yet to come into effect. Australia has also entered into discussions with Fiji. If the working holiday visa is to be made relevant to the Pacific, the eligibility requirements should be made less restrictive. For example, a high school certificate should suffice. The minimum requisite level of savings should be reduced, and the requirement for government approval abandoned. The proof of funds for onward ticket should be kept in place in order to avoid a scenario in which workers become stranded in Australia. The requirement that participants have a functional level of English should also be maintained to ensure employment outcomes are positive upon arrival. 9The United States is in the anomalous position of being eligible under the 462 visa but not having any other requirements imposed on it – just like other OECD countries under the 417 visa. 16 If Australia and New Zealand introduced quotas of 100 each for the low-mobility and atoll countries, this would give a combined cap of 1,000 a year. Given that, as discussed earlier, working holiday makers can stay for a second year if they do three months of regional work in the first year, this would provide significant work opportunities to a large number of Pacific islanders. If PNG, by virtue of its greater size, was given a much larger number, say its own cap of 1,000, this would give a combined cap of 2,800. 5.2 Korea’s Employment Permit System Korea’s Employment Permit System annually grants up to 55,000 workers from 15 countries across Asia access to three years of low-skilled work with scope for a 22 month extension. To be accepted for work in Korea, candidates are required to pass two tests. The first is a Korean language proficiency test. The second is a job skills test, which assesses candidates’ physical fitness, interpersonal and basic job performance skills. Although the Korea Employment Permit system is not currently open to Pacific countries, Timor-Leste was included in the scheme in 2009. Timor-Leste has managed to send to Korea almost 2,000 workers over a six year period. This is only a small fraction of the places allocated to it by Korea. A major barrier has been the Korean language proficiency requirement. Language lessons are funded by the Timorese government. The pass rate amongst all participating countries from 2007-2013 was 38 per cent (HEPS 2013). Timor-Leste has achieved a pass rate of 39 per cent. A survey, commissioned as a background paper for this report, of the experience of Timorese workers who participated in the scheme came out with generally positive findings. See Box 1 below for a summary of Wigglesworth and Fonesca (2016). Overall, the returned workers spoke positively of their experience abroad, in terms of both amounts earned and skills gained. The island countries of the Pacific should engage with the South Korean Ministry of Employment and Labour which manages the EPS. Given Korea’s growing aid interest in the Pacific, it is quite likely that it would respond positively to a request for participation. If Korea did open its Employment Permit System to the Pacific, it is likely that, based on the Timorese experience, it would make available many more positions than the Pacific could fill. If the Pacific were to send as many workers per year as Timor-Leste has done on a per capita basis, that would provide work for up to five years for as many as 3,200 Pacific workers per year. Box 1 Timorese workers in Korea Timor-Leste is one of 15 countries in Asia that have access to the South Korea’s Employment Permit System giving foreign workers work visas for up to five years. The Government of Timor-Leste promotes this program to give Timorese young people aged 18 to 39 access to jobs and to learn job skills that they can use on their return home. Between 2009 and June 2015, Timor-Leste sent 1,886 workers to Korea. By the end of that period, 340 had returned and the remaining 1,546 were still working in Korea. To be accepted for work in Korea, candidates have to learn how to communicate in the Korean language and take the Test of Proficiency in Korean, requiring 80 out of 200 points to be accepted. Those with the highest scores are selected first to achieve the designated quota. Over six years to 2015, 8,638 candidates took part in the Korean language course and 3,410 passed, a pass rate of two in five. 17 A case study based on a small survey of 30 returned workers was undertaken in Dili in July 2015. The study sought information about the experiences of Timorese workers in Korea and on return. The results showed that those interviewed generally spoke well of their employers and valued the experiences they gained. In response to the statement ‘Overall I think my experience working in Korea was a good one’, all agreed with most returned workers agreeing strongly and only two saying they ‘somewhat agreed’. There was more caution about responses to the statement ‘I found living in Korea easy to do’ with only 8 (27 per cent) agreeing or strongly agreeing, 18 (60 per cent) neutral and two disagreeing (7 per cent). Importantly, the responses to the statement ‘I am happy to tell others to apply to work in Korea’ showed acceptan ce from all but two interviewees who gave a ‘somewhat disagree’ response. Many workers struggled to adapt to the work demands of their employers in Korea. Nevertheless, many felt they had developed useful work skills and were able to contribute to their f amilies’ wellbeing through their income. The amount of money earned and sent home was significant. 80 per cent of the migrant workers surveyed received between $US1,001-2,000 a month, while 20 per cent of those surveyed received $US500- 1,000. There was no difference between men and women in the pay received. Almost half the workers remitted home $US6,000 or more a year. The female agriculture workers were the most successful. They were more likely to complete their contracts in a single company and to speak positively of their experiences. Most were employed in small family businesses with just a few workers and some were treated as part of the family. The study (Wigglesworth and Fonseca 2016) on which this box is based concludes: In summary then, with a few exceptions, the former workers spoke positively of their experience and/or the Korea bosses they worked for. Overwhelmingly they spoke of how hard work was in Korea, but many felt they had gained skills. Many respondents noted Korea’s potential in contra st to the lack of work opportunities in Timor-Leste. 5.3 Expanding the Pacific uptake of temporary skilled migrant programs There are currently few Pacific islanders migrating annually through temporary skilled- migration schemes. The Australian Temporary Skilled Work Visa (Subclass 457) permits skilled workers (typically Skill Levels 1-3) with their families to travel to Australia and work in their nominated occupation for an approved sponsor for periods of up to four years (DIBP 2015c). From 2010 to 2015, only 2,905 visas were granted to migrants from the Pacific, less than one per cent of total arrivals under this visa category over that period. New Zealand has its equivalent Essential Skills visa. Visa durations vary from one year for jobs at Skill Levels 4-5 (lower skilled), three years for jobs at Skill Levels 1-3 (higher-skilled) and up to five years or longer for some jobs at Skill Level 1, the highest level (Immigration NZ 2015b). Over the same period, a total of 11,777 NZ Essential Skills visas covering Skill Levels 1 to 5 were awarded to Pacific migrants or 9 per cent of the total. These Essential Skills visas are almost exclusively taken up by the high-mobility countries, with Fiji, Samoa and Tonga accounting for 98 per cent of Pacific arrivals. Other temporary skilled visas include the Australian Temporary Graduate visa (Subclass 485) and equivalent graduate visas in New Zealand. New Zealand also offer visas (Silver Fern Visas) which provide opportunities for young or recent graduates including those with trade qualifications who have not been international students in New Zealand. New Zealand’s scheme has no skill threshold. Aged and disability carers, support and personal care workers, livestock farm workers and truck drivers are lower skill occupations for which temporary visas can be issued in New Zealand but not in Australia. Essentially, Australia cuts off its 457 visa at Skill Level 3 while New Zealand allows employers to recruit workers at all skill levels, 18 albeit for different durations (on the grounds that there should be more frequent labour market testing for lower-skilled jobs). The New Zealand skills profile of Pacific migrants shows that one of the major barriers for Pacific migrants to work in Australia is the absence of a medium and lower skills occupational pathway. Figure 3 below compares the skill levels of the occupations of Pacific migrants granted temporary work visas in Australia and New Zealand in recent years. Four out of five Pacific migrants approved to work in New Zealand are concentrated at the lower end of the skills spectrum, mainly at Skill Levels 3 and 4 but also at Skill Level 5. In contrast for Australia, the skills profile of Pacific migrants is concentrated at the other end of the skills spectrum, with the largest number of occupations at Skill Level 1, followed by occupations at Skill Level 3. The operation of the skills threshold in Australia’s case is evident in the cutoff at Skill Level 3 for Pacific migrants approved to work in Australia. The only occupation below Skill Level 3 allowed for a temporary work visa is a Skill Level 4 occupation euphemistically called ‘skilled meat worker’. Figure 4 Temporary work visas held by Pacific migrants: Australia and New Zealand (2010- 11 to 2014-15): per cent of the total for each destination 50 47 40 36 33 30 27 22 20 15 15 10 4 1 1 0 Skill Level 1 Skill Level 2 Skill Level 3 Skill Level 4 Skill Level 5 Skilled Meat Worker Australia New Zealand Source and notes: Authors’ analysis based on DIBP (2016a) and Immigration NZ (2015c). ‘Skilled meat workers’ is provided as a separate category in the Australian data. Reducing Australia’s 457 skills threshold would be good for Pacific migration, but is unlikely to happen. A more targeted approach is needed, and is put forward in the next section in relation to aged care. Prior access is the other reason that New Zealand’s temporary skilled migration program is more accessible to Pacific islanders than Australia’s program. New Zealand’s Pacific island diaspora is larger than Australia’s, and Pacific islanders have more opportunities to visit New Zealand on a short-term basis than Australia (e.g. because the New Zealand RSE is bigger than Australia’s SWP). In-country presence matters because it provides opportunities to network, meet employers directly and to get relevant work experience. In 2014-15, half (49 per cent) of successful primary applicants for 457 visas were already in Australia on other visas (e.g., student, backpacker) before being granted a 457 visa (DIBP 2016a). Getting a temporary skilled visa is expensive. The 457 visa requires applicants to be nominated to work in an approved occupation on the Consolidated Sponsored Occupations List (a list of some 19 650 occupations considered skilled), provide evidence of qualifications, and have relevant work experience confirmed by an assessment authority in Australia (DIBP 2015c). The need to submit to an in-person skills assessment by a Registered Training Organization (RTO) is costly. The Skills Assessment Fee costs $A600-800, and the technical interview $A500. If the applicant has to travel to take the test, that adds to costs. If applicants pass the skills assessment, the 457 visa fee is fixed at $A1,035. As a result, Australian skill recognition can be costly for many Pacific Islanders, even if they have the necessary qualifications. However, there has recently been some progress. From February 2016, it has been possible for carpenters and bricklayers in Fiji seeking to migrate to Australia to get their trade skills assessed for the 457 visa in Suva. Australia has tried – unsuccessfully to date – to increase skilled and semi-skilled labour mobility through the Australia Pacific Technical College. This was a potentially innovative ‘aid for migration’ initiative. Its failure to deliver on the labour mobility front (see Box 2 for details) has undermined the College’s viability, with a recent evaluation finding that it had a negative net present value due to the low number of migrant graduates it has so far produced (Johanson et al. 2014). There is high demand from APTC graduates to migrate, but inappropriate qualifications, high costs and a lack of links with employers in Australia and New Zealand have prevented their aspirations being realized (ibid). APTC’s redesign presents a new opportunity. Australia has signalled its long-term commitment to APTC through an additional $A96 million in funding, which will see the College through to June 2018. What does the redesign need to do to ensure APTC graduates are able to access Australia’s Temporary Skilled Work (457) and New Zealand’s Essential Skills Visa? We offer seven suggestions to help meet APTC’s labour mobility objective. First, APTC graduates should be made eligible for Australian graduate visas. The relevant visa is the Temporary Graduate visa (Subclass 485). This visa requires applicants to have completed a degree, diploma or trade qualification with an Australian educational institution, hold an eligible student visa at the time of application, meet the English language requirement and have skills or qualifications that relate to an occupation on the skilled occupation list (DIBP 2016b). It would be a simple matter to extend this visa to APTC graduates, especially since it is an Australian educational institution. This would allow APTC graduates time to complete certification requirements and to find jobs. Second, the maximum duration of stay under the SWP should be increased to one year for certain sectors. APTC graduates (and others) could work in Australia through the Seasonal Worker Program in sectors now covered by the scheme such as accommodation and tourism and a range of agriculture industries including cattle, sheep, grain and mixed enterprises. The Australian Government announced in May 2016 the Seasonal Worker Programme Northern Australia Tourism Pilot (Department of Employment 2016). Seasonal workers can be employed in 15 occupations. Of these, up to 11 APTC qualifications offered in 2016-17 would be relevant to APTC graduates as an opportunity to obtain relevant work experience in Australia. 20 A more extended work period under the SWP beyond the six months limit now imposed would allow APTC graduates enough time to get relevant experience and, with an appropriate qualification, find a job and be sponsored by an employer through a 457 visa for a more extended period of work. This in turn opens up a pathway to permanent residence through employer sponsorship. Note that workers recruited from Kiribati, Nauru and Tuvalu are already allowed up to nine months of work to recover their higher costs of transport. Third, some of the APTC funding could be used to promote APTC graduates to Australian and New Zealand employers. This could be done by, for example, hosting visits of employers to APTC campuses in order to meet instructors and students. Fourth, cost constraints need to be addressed. The aid program could be used to reduce the costs of skill certification, and increase the frequency and ease of certification opportunities. Fifth, to allay fears of brain drain it is essential that APTC either trainor support the training of fresh workers in the fields it is training in. This will ensure that the Pacific skilled labour stock is replenished rather than depleted. This makes economic sense, but is also key for ensuring sending-country political support for APTC. Sixth, APTC needs to return to its original labour mobility mandate and focus its training on occupations where there is a feasible pathway from training to migration to specific destination countries. Training aged care workers for the New Zealand market could be one example. The Northern Territory Designated Area Migration Agreement might also provide opportunities. Training for jobs at skill levels which do not provide a migration pathway, for example, youth or community work, should be dropped. Seventh, and last, better data collection on employment and migration outcomes would support better achievement of a labour mobility objective for APTC. Data should be regularly collected and published about migration and employment outcomes more broadly. 21 Box 2 Australia-Pacific Technical College (APTC) and labour mobility The Australia-Pacific Technical College was created in July 2007 with a goal of linking skill creation and skill mobility in the Pacific region. Since its inception, it has been financed by the Australian government through its aid program, and implemented by Australian technical education providers (Clemens et al. 2015). APTC has campuses in five countries (Fiji, Samoa, Vanuatu, Papua New Guinea and Solomon Islands). As of June 2014, 6,211 students had graduated with qualifications. Many APTC students enrol with the aim of migrating abroad. An independent evaluation of the APTC conducted in 2014 (Johanson et al. 2014) found that in response to a direct question about plans to migrate to Australia or New Zealand, nearly two-thirds of APTC students interviewed (64 per cent) said they were planning to migrate at some stage in the future. One in five (18 per cent) had taken concrete steps towards this goal. Unfortunately, extremely few APTC graduates have had the opportunity to migrate. According to the evaluation, one in forty (2.5 per cent) have migrated to other countries in the Pacific or to high-income metropolitan countries around the Pacific Rim. Figure 5 APTC migration outcomes (2014) 80% 68% 68% 70% % of APTC students 56% 57% 58% 60% 50% 43% 40% 30% 23% 22% 20% 14% 12% 10% 2% 1% 4% 2% 2% 0% FIJ PNG SAM SOL VAN Planned to migrate Took steps to migrate Migrated Source: Johanson et al. 2014. The APTC independent evaluation conducted in 2014 found that there was a distinct lack of promotion of its labour mobility objective. It also found that that 'less than half of APTC graduates (46 per cent) are in an approved occupation or appropriate qualification level to be eligible to gain skilled migrant entry to Australia or New Zealand' (Johanson et al. 2014, p 17). APTC students and graduates highlighted several other barriers to migration including: a lack of information; difficulties applying for a visa; costs and a lack of finance;complications getting their qualifications recognized; and a lack of job offers from foreign employers (Johanson et al. 2014). Implementing this full set of reforms would not only make APTC a viable project, but also usher in a new era of migration for the medium-skilled from the Pacific to Australia and New Zealand. Some of the reforms discussed above could be applied to all semi-skilled workers from the Pacific, such as subsidizing skill certification fees. Others make sense only for APTC graduates (such as making them eligible for the graduate work visas). An encouraging recent development shows the way forward. The New Zealand Government through the Canterbury Trades Employment Initiative pilot, will bring an initial cohort of around 20 carpenters from Samoa, Fiji and Tonga to work in the Canterbury rebuild starting from April 2016. New Zealand aid will fund the initiative. Support for Pacific Island workers will include assistance with skills screening and testing, completing recruitment and visa applications, review of employment terms and conditions, pre-departure preparation, on-job training plans and pastoral care (MFAT 2015b). Opportunities to work in New Zealand have also been made more attractive with Immigration NZ’s decision to extend the maximum duration of Essential Skills visas for lower 22 skilled occupations from one to three years for work in the Canterbury reconstruction for the period 1 July, 2015 and 31 December, 2016. 5.4 Proposal for a Pacific in-home caregivers program for those with high medical-care needs The ageing of the population in OECD countries and the greater longevity of individuals can be expected to lead to increasing numbers of people at older ages with a severe disability (OECD 2011, 63). A good indicator of the increased future need for the long-term care of the elderly is the growth in the share of the population aged 80 and over. In Australia and New Zealand, this share is estimated to nearly double between 2015 and 2040 to 7.4 per cent of the population. Similar increases the shares of the population aged 80 and over are also evident for Canada, USA and the UK (9.2, 7.5 and 7.6 per cent by 2040 respectively). The same ageing of the population and a smaller working age population in many countries means that the pool of informal carers –such as partners and children – will also decrease, at least in relative terms (OECD 2011, 64). In Australia, the number of aged care workers (excluding managers) in both residential and community care may increase from 201,000 in November 2011 to 532,000 in 2040.10 However, the ratio of care workers to the aged population is due to increase with the introduction of the National Disability Insurance Scheme, and so the forecasted number of care givers will increase further. If, for example, the long-term care worker ratio per 100 people aged 65 and over of the Netherlands in 2013 were applied to Australian population data, the number of care workers needed in 2040 would be an estimated 622,000. Home-based care givers account for 0.5 per cent of the labour force of all OECD countries but this proportion varies greatly by country (OECD 2015, 123). Sweden has the highest proportion of home-based caregivers in the labour force at 5 per cent; followed by the UK (3 per cent) and Canada (2 per cent). Foreign-born workers account for 29 per cent of all home-based caregivers in OECD countries. In Sweden, the proportion of foreign-born home-based caregivers is 22 per cent, in the UK it is 19 per cent and in Canada, it is 28 per cent (OECD 2015, 123). Data are not avaiable for Australia or New Zealand . The opportunities for work as aged or disability carers for migrants from the Pacific depend on employers’ preferred source of workers. Employers may not want to spend the time, effort and cost to recruit directly from overseas if there are temporary migrants already resident in the country and are entitled to work. For Australia, in relation to aged or disability care, over the five years to end June 2015, employers recruited few occupations directly from overseas via temporary skilled work visas.11 However, migrants account for 35 per cent of the aged care workforce in 10 This estimate is based applying the ratio of direct care workers enumerated in the 2012 National Aged Care Workforce Census and Survey to the population aged 65 years and over in 2010. This ratio was then applied to the estimated size of the population in this age group in 2040, based on the UN World Population Prospects: The 2015 Revision 11 At Skill Level 2, they were: residential care officer (196 visas), disabilities services officer (151 visas) and enrolled nurse (206 visas). For the health care sector, the only Skill Level 3 occupation was cook (60 visas). At 23 residential care and 37 per cent of recent hires employed for 12 months or less (NILS 2012, 14). These data show that employers can recruit migrants for aged care work from local labour markets. However this source of workers is often unreliable due to the work duration limitations imposed on international students and backpackers during their stay in Australia. Employers in New Zealand, unlike Australia, are able to recruit migrants directly for work as aged and disability carers, a Skill Level 4 occupation, because a Skill Level 3 theshold does not apply, as it does in Australia. The number of New Zealand work visas for aged and disability carers approved over the five years to 2014-15 was 2,538. This is 2 per cent of all temporary work visas over this period. Some 494 visas approved for this occupation went to migrants from the Pacific. The Pacific share of visas approved for this occupation was one-in-five (19 per cent) of all visas for this occupation.12 However, the New Zealand Essential Skills visas for this skill level are for one year only and visa holders cannot qualify for permanent residency. The visas for jobs at lower skill levels can be renewed each year, provided that for each renewal the job is advertised locally and no suitable applicants are available. This process means that employers have no guarantee that the aged or disability carer recruited directly from overseas can continue to work beyond the one-year period of their visa. This built-in instability in the supply of migrant aged care workers in New Zealand has been criticized for failing to place the needs of the aged care client first and foremost (Callister et al. 2014). Australia and New Zealand both lack a reliable system to supply qualified workers who can provide care on a continuous basis to people in their own homes. Home-based continuous care relies on informal care givers for the elderly, usually partners, but as the population ages reliance on partners becomes increasingly problematic. Formal care services available to the elderly in their homes in both countries are delivered by the staff of community care agencies who are limited, for the most part, to short, carefully timed visits to provide specific services. Where full-time home- based care is available through an agency, it may cost up to $1,000 a day (Brown 2015). Yet demand for home-based services is increasing as more older people prefer to stay in their own homes. Reforms giving the elderly more control over the form of delivery of aged care services will reinforce this trend. Canada’s program for in-home caregivers for people with high medical needs provides a good model. It allows persons residing in Canada to employ qualified foreign workers in private residences to provide care for elderly persons or persons with a disability.13 The program offers a permanent residence pathway for migrants who have completed two years of work in a four-year residency stint. The program stream related to care of the elderly has a cap of 2,750 places per year. The core elements of the Canadian program are the minimum requirements for employment Skill Level 1 the only identifiably relevant occupation was registered nurse (aged care) (1,899 visas) (DIBP 2016e). 12 The data refer to visas approved and not to numbers of migrants as the visa approvals may include renewals for the one year visa. 13 See http://www.cic.gc.ca/english/immigrate/caring-medical/. 24 as a caregiver (eg post-school qualification in health care of at least six months and good English language ability), no requirement to live in as a home-based caregiver (since 2014), agency oversight of the employment contract, and a prior labour market assessment to ensure that citizens or permanent residents have first opportunity to apply for the work. Canada’s approach could be used as the basis for the design of a similar program for Australia and New Zealand for Pacific migrants. The focus on the beneficiaries in the name of the program is essential to justify why a Skill Level 4 occupation should be an exception to the Skill Level 3 threshold for migrants in Australia and for access to a permanent residence pathway in both Australia and New Zealand. Limiting the program to Pacific migrants will enable donor funds to be directed to providing Australian and New Zealand recognised qualifications to potential migrants. It will also enable employers to develop a reliable source of carers and to have the primary role in their selection. Aged or disability full-time care has a number of special characteristics that need to be taken into account in the design of a program to provide a supply of reliable caregivers. Key design features would need to include: applicants having an appropriate skills set, employer selection, government agency vetting of the employment contract, labour market testing, appropriate safeguards and regular monitoring and reporting. To provide continuity of care for the elderly or disabled, it is also important that migrants are able to apply for permanent residency after completing some minimum period of work, say, two years of work within a four year period, as in Canada. Migrants would need to have formal qualifications. These should based on technical requirements and obligations of caring for the elderly or disabled required for any caregiver in a formal employment relationship in Australia or New Zealand. APTC, or other designated colleges, would be responsible for providing the required training. A relevant one-month work placement in aged or disability care should be a requirement of the program. Work in aged or disability care also requires a range of social and emotional skills. These skills include the ability to work as a member of a team, reliability, a sense of responsibility and the ability to relate to others. This placement could be arranged as part of the training, and would enable the instructor to assess the social and emotional capacity of the trainee as a caregiver and provide feedback to the trainee. Employers would select the workers they want. An important role for the training provider is to provide prospective employers or their agents with relevant information about their graduates. Providers should also make it easy for employers or their agents to interview applicants in-country and seek information from the instructor based on the trainee’s assessment. The importance of facilitating this relationship requires a trusted intermediary such as the APTC which runs campuses in five Pacific countries. Safeguards would be needed for both the home-based caregivers as well as for those receiving the care. Sweden, the UK and Canada have well-developed health care systems with 25 mechanisms for setting, implementing and auditing quality standards.14 For example, the UK’s new Care Act 2014 requires each local health authority in England to set up a Safeguarding Adults Board and to conduct Safeguarding Adults Reviews where the Board ‘knows or suspects that the adult has experienced serious abuse or neglect’.15 In relation to home-based caregivers, the legislation allows a local heath authority to decide whether a caregiver working under a contract can be regarded as a ‘carer’. This designation entitles the carer to practical and emotional support as determined by the local health authority’s carer’s assessment.16 Some 2,750 migrant care givers for people with high medical needs are allowed to apply each year for In-home Caregiver visas in Canada (CIC 2014). Comparing Canada’s population with Australia and New Zealand means that an equivalent program in the two countries could accommodate 2,200 places. 6. Expanding permanent migration: Pacific options Long-term migration opportunities for the Pacific currently come through three types of programs: permanent skilled migration, visa lotteries, and open access. These are discussed in turn in the three sub-sections following. 6.1 Permanent skilled migration Employer sponsored and points-based skilled migration account for a large part of the permanent intake. On 2014-15, approximately 3,500 migrants arrived monthly in Australia and 4,000 in New Zealand under these streams. Since the 1980s, the Australian Government has progressively developed policies designed to target migrants with experience in areas where there are skills shortages (Phillips 2006). Australia currently offers both employer-sponsored and points- based permanent migrant entry. In 2014-15, 896 Pacific migrants were granted permanent residence in Australia through the skilled migration program. New Zealand implemented its points-based system in 1991. This provides the basis for the Skilled Migrant Category (SMC), which accounts for approximately half of its total immigration program (OECD 2014b). Approximately 2,000 Pacific Islanders were granted access through the SMC in 2014 (ibid, Table 4.1). The barriers to greater Pacific access to permanent skilled migration opportunities are similar to those discussed in Section 5.3 in relation to temporary skilled migration. Pacific islanders are out-competed by migrants from large countries. They are discouraged by cost barriers, and lack the advantage provided to those nationalities which are better able to access Australia on 14 Canada recently removed the requirement that a migrant live-in carer had themselves to live-in. See Canada Immigration (2014). 15 UK Care Act 2014, Part 1 Safeguarding adults at risk of abuse or neglect Sections 42-47. Abuse includes financial abuse having money or other property stolen, (b) being defrauded, (c) being put under pressure in relation to money or other property, and (d) having money or other property misused. http://www.legislation.gov.uk/ukpga/2014/23/pdfs/ukpga_20140023_en.pdf 16 UK Care Act 2014, Assessment of a carer’s needs for support, Section 10, subsection 10. 26 a temporary basis. Countries with large numbers of tertiary graduates currently dominate the skilled migrant intake. India, China, the United Kingdom and the Philippines account for approximately 50 per cent of Australian long-term skills-based arrivals (DIBP 2015d). The costs involved with migrating include the visa application charge ($A3,060), the skills assessment application fee ($A1,050), the International English Language Testing System (IETLS) test fee ($A310), medical examination fee ($A303), police clearance certificate fee ($A20) and translation and document certification fee ($A500). These costs total $A4,940 and put permanent migration out of reach for many prospective Pacific migrants. Finally, and most importantly of all, the vast majority of employer-sponsored migrants who come through the skilled stream are already in Australia on other temporary work, international student or working holiday visa categories, which the Pacific island countries struggle to access. New Zealand’s Skilled Migrant Category poses similar obstacles, but does reduce costs for Pacific applicants. New Zealand has a dual fee structure which charges $NZ1,620 for Pacific countries and $NZ2,505 for the rest of the world (Immigration NZ 2015b). Pacific migrants are more successful in relation to New Zealand’s temporary skilled programs, and a temporary visa often transitions into successful places in the SMC, given the additional points awarded for in-country work experience (OECD 2014b). The same reforms discussed in Section 5.3 – reducing application costs, and encouraging employer-employee links – would also work to increase Pacific access to the temporary skilled migration programs of Australia and New Zealand. Reforms related to the treatment of Australia Pacific Technical College graduates, also discussed in Section 5.3, are especially pertinent. 6.2 Migration lotteries: learning from New Zealand’s Pacific Category New Zealand offers through its permanent migration ‘Pacific Category’ visas, 1,750 places for permanent residence each year to citizens of selected Pacific island countries.17 In what follows, we examine the New Zealand scheme to see how well it is working, and whether New Zealand should expand it, and Australia replicate it. The Samoa Quota (SQ) is the product of a Treaty of Friendship after Samoa achieved Independence from New Zealand in 1962. The treaty allowed for family migration to New Zealand for permanent residence with an informal quota. In 1970, the Samoan Quota was formally set at 1,100 Samoans (including dependants) to be granted residence each year, provided the primary applicant meets certain basic criteria such as an offer of an ongoing job that meets a minimum income level, is in good health, speaks reasonable English and be of good character. 17The United States Diversity Immigrant Visa program is another permanent migration lottery that benefits the Pacific islands. It is congressionally mandated and is also known as the Green Card Lottery. The program makes 55,000 permanent resident visas available annually to people from countries that are deemed as having low ratesof immigration to the United States. All the Pacific island countries are eligible to apply and 1,056 migrants were granted access in 2015. 27 The Pacific Access Category (PAC) visa, operating since 2002, offers 250 places per year to citizens from Fiji and Tonga, and 75 places each to citizens from Kiribati and Tuvalu. The quotas include dependants, and the scheme uses the same criteria as the SQ. As Fiji has only recently become eligible again to apply for the visa, the following analysis refers only to four Pacific countries – Samoa, Tonga, Kiribati and Tuvalu. The numbers applying for the Pacific Category visa ballots show that there is a high level of interest in migrating to live and work in New Zealand. The ratio across all countries of principal applications registered to those selected is 16 to 1 for the five-year period to 2014-15. The demand is highest in Kiribati (32 to 1). It is also high in Tonga (21 to 1). Samoa’s ratio of 14 applicants to each successful ballot is lower because the number of available places is much higher (1,100 compared with 250 for Tonga). Also, the opportunity to migrate to New Zealand has existed for a much longer time period than for the other Pacific eligible countries. For Tuvalu, the ratio is lower at 8 to 1. Tuvalu is a tenth the size of Kiribati, but enjoys the same PAC quota. The lower demand in Tuvalu may also reflect the fact that a large number of Tuvaluans have migrated already to New Zealand, as shown by the fact that near to one in five (18 per cent) of the total population aged 15 years and over born in Tuvalu are already living in New Zealand. In contrast, only 6 per cent of the population of same age group born in Kiribati are living in New Zealand. A distinctive feature of the Pacific Category visas is the requirement to have a job. However, unlike visas for skilled migrants, this requirement is not limited to jobs of a specific skill level or type of work. The Pacific Category principal applicant has to be aged between 18 and 45 years, and is required to have an acceptable offer of employment in New Zealand, and/or have a partner included in their application who has an acceptable offer of employment in New Zealand. If the principal applicant has dependants, the job offer or job offers have to produce combined income of just above a minimum threshold of $NZ32,046 per year. The principal applicant is also required to meet minimum English language requirements. As well, all persons on the application are required to meet health and character requirements. Applicants are only required to find a job if they are selected through the lottery. This requirement results in some under-utilization of the quotas available, but over the five years to 2014-15 some 90 per cent of quotas were filled. There are two ways to find a job. Extended families and diaspora in New Zealand play a major role in finding jobs for lottery winners and in helping with their early settlement in the country. In addition, the Ministry of Business, Innovation and Employment (MBIE)’s Pasifika Work and Skills Team has two relationship managers who help approximately 40 per cent of successful ballotees find work by contacting a network of 30-40 employers. Pacific Category migrants find unskilled or semi-skilled work. Data from tax records show that the proportion of principal migrants earning wages and salaries ranged from 79 per cent in March 2007 to 72 per cent in March 2011 (Merwood 2013a, 56-66). (Clearly the Global Financial Crisis resulted in worse employment outcomes.) Other data on annual income from all sources for the principal Pacific Category migrant show that self-employment was a source of annual income for 5- 6 per cent of principal Pacific Category migrants over the five years to end March 2011 (Merwood 28 2013a, Table 26). Between 2008/09 and 2012/13, 70 per cent of the wage jobs (not including self- employment) were at the lowest skill level and 16 per cent are at the second lowest skill level (Merwood 2013b). Only 14 per cent of jobs held by Pacific Category migrants were at the three higher skill levels. The nature of their jobs means that these migrants have much lower earning power than skilled workers. Between 15 and 23 per cent of principal Pacific Category migrants between 2007 and 2011 were not in some form of employment. Over the same period, those migrants reliant mainly on government benefits ranged from 4 to 13 per cent. Notably, census data for 2013 show that migrants from Tuvalu have a lower employment rate than migrants from other Pacific Category countries. And migrants from Kiribati and Tuvalu both have a larger share in the lowest income group compared with migrants from other Pacific Category countries (Curtain 2016b). Outcomes vary by educational level. The 2013 New Zealand Census results show that only 44 per cent of migrants from the four Pacific Category countries of Samoa, Tonga, Kiribati and Tuvalu with low education are employed, compared with 54 per cent with a medium level of education, and 66 per cent with a high level of education.18 The New Zealand 2013 Census results also show important differences in employment outcomes by country of origin. Male and female migrants aged 25-64 years from Tuvalu have a lower employment rate than the other three Pacific countries. The weaker employment results for Tuvalu migrants in the New Zealand hold for each level of education attainment . The overall lower employment outcomes for Tuvaluan migrants in New Zealand reflect in part the lack of numbers in a micro-state such as Tuvalu to make post-secondary training facilities viable. The lower employment rates for each level of education, however, may reflect their lower quality of education. This same curse of smallness may also make it difficult to produce education outcomes such as the level of literacy in English required to obtain work in New Zealand. Overall, the evidence on employment outcomes and incomes earned shows that the Pacific Category visas have been successful. They have enabled a relatively small number of migrants from four Pacific countries to come as families to settle permanently in New Zealand. The minimal requirements of a prior job offer, proof of basic English, a clean bill of health and being of good character have produced in general good employment outcomes. Most of both the principal and secondary applicant Pacific Category migrants are in employment and are earning a basic income after they migrate, as shown by tax records over a period of five years. The Pacific Category visas are well designed. They were designed not to ‘cherry pick’ migrants with the highest qualifications to avoid the danger of creating a brain drain from small countries. This has been done by setting a low threshold requirement of a permanent job regardless of skill level and requiring only a rudimentary standard of English as well as evidence of good health and good character. The use of the ballot to select migrants and putting the onus on those selected to find a job in New Zealand are important features of the program. The process ensures that the sending countries accept that the selection for a small number of places is done in a fair and transparent way. 18 Nearly all migrants from these countries gained entry through the Pacific Category. 29 But two reforms would help improve migrants’ employment rates and lift their annual incomes. The first involves the design of the selection process. The second is to provide better access to job brokers in New Zealand. First, selection requirements could be toughened. Completion of secondary schooling will increase the chances of employment by 8 percentage points for men and by 10 percentage points for women aged 25-64 years compared with the low educated. Completion of a non-tertiary post- school qualification offers a 12 and 21 percentage points improvement for men and women of the same age group. We therefore propose that completion of secondary school (or at least completion to Year 10) should be a requirement for a successful application for an expanded quota for New Zealand and for similar visa access to Australia. The English test could also be made more stringent, and a numeracy test also required. This will provide a strong incentive on the sending country to improve the quality of the education provided. These selection requirements are achievable by prospective migrants, for example, by returning to school if needed, or at least ensuring that one’s children stay at school. The risk of brain drain is therefore minimal; indeed, the tougher requirements could lead to brain gain among the population that did not want to migrate. Providing more support for job brokers would also help improve employment outcomes. As noted above, heavy reliance is placed on the diaspora when it comes to job-hunting. However, if this diaspora is small as is the case for the migrants from Kiribati, they will have few resources to help. Similarly, many in the diaspora may be struggling to find work themselves, as is the case especially for migrants from Tuvalu. Over-reliance on a small number of employers may result in high levels of risk and vulnerability. Reliance on informal networks needs to be supplemented by more use of job brokers with relevant experience who may or may not be members of the diaspora. Funding could be provided by the government of the sending country, and could be based on a fee levied at the departure for New Zealand. These brokers should be paid, at least in part, for results with payments made if the migrant has the job for six and 12 months or more. A website could also be developed where applicants with post-school qualifications and prior work experience in New Zealand or Australia could post details on a website and invite employers to select one or two applicants for interview via the Internet or face-to-face. In summary, the PAC and SQ agreements are well suited to manage migration flows when the potential supply of migrants is much greater than the capacity of the receiving society to absorb them. They are also well suited to jobs whose requirements are not defined or certified as specifically as the professions or trades are (OECD 2014a,139-140). These agreements are also a valuable way of controlling illegal recruitment for low-skilled and low-paid jobs (OECD 2014a, 140). With the reforms suggested above, there is a strong case for expanding New Zealand’s Pacific Category and for introducing a similar scheme in Australia. Introducing a ‘Pacific Category’ into its permanent migration regime would be a major shift for Australia, though a small window would help balance the overall immigration regime given the much larger ‘OECD Category’ in the temporary holiday market visa scheme (see Section 5.1). Introducing a Pacific Category into Australia would also have the benefit of helping the country move away from the current ad hoc 30 reliance on students and backpackers to meet unskilled labour needs. If Australia were to introduce its own Pacific Category equivalent, how big should it be? It might grow slowly over time, but ultimately it could become five times as big as the New Zealand scheme based on the differences in population. If New Zealand were to increase the size of its scheme by half, that would provide another 900 slots. Any expansion or replication of the Pacific Category should be targeted where it is needed: at the low-mobility and atoll countries. 6.3 Open access: the proposed Australia-New Zealand Atoll Access Agreement The USA, New Zealand and Australia all have open-access agreements, but only the USA and New Zealand have them with Pacific Island countries. As discussed, the Federated States of Micronesia, Marshall Islands and Palau are granted free access to the US labour market through their respective Compacts of Free Association. New Zealand has similar agreements with the Cook Islands and Niue, and the directly administered territory of Tokelau (none of whom are covered by this study as they are not members of the World Bank). Australia and New Zealand provide open access to each other’s citizens. This benefits a number of Pacific-born migrants who are able to live and work in Australia because they are New Zealand citizens. Countries provide open access only under very special circumstances. Countries generally provide open access only to countries that they are similar to or that are important to them. The open access which Australia and New Zealand grant each other, or that exists within the European Union are examples of similar countries granting open access to each other. That the compact states enjoy open access to the US labour market is an indicator of their strategic importance to the US (which maintains the right to require land for military bases and exclude ocean access to the armed forces of other countries), just as the open access enjoyed by Niue, Tokelau and the Cook Islands is a sign of their historical and ongoing close association with New Zealand. The worsening impacts of climate change have provided a new moral imperative for providing open access. This is especially so given the contribution developed countries around the Pacific Rim have made to greenhouse gases and the disproportionate effects experienced by the Pacific Island countries (Ritter 2015, Marles 2015). The case is particularly strong for the low-lying atoll states. While there are still many uncertainties (Johnson 2015), a 2012 Report of the Intergovernmental Panel on Climate Change (IPCC) highlighted the extreme danger that small island countries face as a result of climate change noting that: ‘the small land area and often low elevation of small island states makes them particularly vulnerable to rising sea levels and impacts such as inundation, shoreline change and saltwater intrusion into underground aquifers’ (IPCC 2012, 185). Half of the populations of Kiribati and Tuvalu live on the atolls of Tarawa (Kiribati) and Funafuti (Tuvalu) in overcrowded urban areas on narrow strips of coral with limited access to water and land to grow food. As a result of this overcrowding, atoll and reef island environments have become degraded. Climate change can make this worse by the increasing erosion of the foreshores where settlement is often concentrated. Global warming has also increased acidification of the oceans resulting in progressive damage to reef ecosystems (see Bedford and Bedford 2016 and the references therein). 31 The various atoll island states and territories in the Pacific have differing levels of international labour market access. Marshall Islands and Tokelau already have open access to the US and New Zealand respectively. Kiribati and Tuvalu do not.19 Both countries have long histories of labour migration primarily through the phosphate mining and seafaring industries (Bedford and Bedford 2016). But both countries have been experiencing difficulties in recent years in obtaining employment opportunities overseas given the termination of large-scale phosphate mining on Nauru and the heightened competition in the maritime freight business from cheaper sources of labour. Only about 100 migrate from Tuvalu and 200 from Kiribati every year (Bedford and Bedford 2016). The combination of restricted external labour market access, youthful populations and high fertility rates mean that the populations of Kiribati and Tuvalu are expected to grow rather than shrink without greater migration opportunities. Bedford and Bedford (2016) show that official projections have the populations of the two countries increasing by 77-83 per cent and 46-73 per cent respectively. The total fertility rates in Kiribati (3.9) and Tuvalu (3.7) are amongst the highest in the East Asia Pacific Region and well above the average for Oceania of 2.4 (UN Population Division 2015). Both countries are actively seeking greater access to greater temporary and long-term employment overseas, especially for their burgeoning young adult populations. An important principle underpinning the recently developed National Labour Migration Policies (NLMPs) of both Kiribati and Tuvalu is the need for increased access to employment and residence opportunities overseas to allow for ‘migration with dignity’ (Voigt Graf and Kagan 2016). The two governments would prefer a slow outward flow resulting from voluntary migration and do not wish their peoples to be treated as ‘refugees’ fleeing a hopeless economic and environmental situation. Without migration channels specifically for them, it is highly unlikely that Kiribati and Tuvalu would be able to increase their rates of outward migration. Without large agricultural sectors, they struggle to compete in the seasonal worker schemes. Their low skill profiles place them at a disadvantage in accessing the temporary and permanent skilled pathways. Meanwhile, their remoteness and the associated cost of travel imposes an additional barrier for prospective migrants, even if there was strong demand in receiving countries. In recognition of this, two schemes have been established specifically targeted at Kiribati and Tuvalu. New Zealand’s Pacific Access Category reserves 75 places each for the two countries. And just last year Australia established the Pacific Microstates–Northern Australia Worker Pilot in Australia which provides 250 places over five years (in total) for i-Kiribati, Tuvaluans, and Nauruans to work in Australia for up to two years. However, much more is needed. One intervention which should be considered is the provision by Australia and New Zealand of open labour market access to Kiribati and Tuvalu, on grounds of their acute climate change risks. 19 Nor does Nauru, but, as noted earlier, Nauru is not covered by this report. 32 How many i-Kiribati and Tuvaluans would migrate if they were provided with open access? A recent United Nations University (UNU) survey revealed that 77 per cent of i-Kiribati and 79 per cent of Tuvaluan households feel that migration will be a necessary strategy as a result of sea-level rise (ESCAP 2015). However, most of these households lack the financial means to migrate. Only 26 per cent across both countries suggested they had the savings required to leave (UNU 2015). This implies that only about 31,000 i-Kiribati and 2,200 Tuvaluans will have the desire and financial means necessary to migrate. If these numbers were to migrate steadily over a 25 year time horizon, Australia and New Zealand would be looking at an additional average inflow of 1,300 i-Kiribati and Tuvaluans annually or the equivalent of 0.6 per cent of their combined annual permanent migration program. Migration of this scale is similar to Samoa’s annual quota for permanent migration to New Zealand which has existed since 1962. It is worth noting that Samoa’s population in 1962 was 113,000, th e same size as the current population of Kiribati (Bedford and Bedford 2016, p 16). More realistically, the inflows would start with a trickle and eventually increase to a higher rate after 10-15 years before levelling out. This has been the case with both Tokelauans migrating to New Zealand since 1951 and the Marshallese migrating to the United States since 1986. Under such a scenario, inflows might peak at about 2,500 per year by 2040 and then either level out or steadily decline. Scenario analysis illustrates the very different prospects of the two countries. Bedford and Bedford (2016) show that with current emigration rates, population sizes of both Kiribati and Tuvalu will continue to increase rapidly, especially the former (by 76 and 15 per cent respectively by 2050 relative to 2015). As Figure 6 shows, doubling outflows in Tuvalu gradually (from 100 currently to 200 by 2030) results in a population decline of 25 per cent by 2050. But even a large increase in outflows by Kiribati (from 200 currently up to 2,400 in 2040) still results in a population increase by 2050 of 23 per cent. In other words, if we take stabilizing the population as a proxy goal (or at least first step) for responding to climate change risks, this task is manageable with respect to Tuvalu but very challenging with respect to Kiribati. Kiribati poses challenges of a different magnitude due to its larger resident population base, its more rapid population growth, its high urban population density, its severely degraded urban environment and its smaller overseas diaspora (Bedford and Bedford 2016, 18). 33 Figure 6 Population growth in Kiribati and Tuvalu under different migratory assumptions 220,000 13,000 200,000 12,000 Kiribati population Tuvalu population 11,000 180,000 10,000 160,000 9,000 140,000 8,000 120,000 7,000 100,000 6,000 2015 2020 2030 2040 2050 Kiribati BAU Kiribati increased access Tuvalu BAU Tuvalu increased access Source and notes: Bedford and Bedford (2016). BAU corresponds to the ‘Net loss-100’ (for Tuvalu) and ‘Net loss-200’ (for Kiribati) scenarios. ‘Increased access’ corresponds to the ‘gradual increase’ (for Tuvalu, with out-migration moving up to 200 by 2030 and 250 by 2040) and the ‘major increase’ (for Kiribati, with out-migration moving up to 2,400 by 2040) scenarios. Given the open access arrangements between their own two countries, it would make sense for Australia and New Zealand to provide open access to Kiribati and Tuvalu together. The total size of the diaspora in Australia from the two countries combined in 2011 was only 625 migrants. For New Zealand in 2013 it was 2,922 migrants. One important factor that might make Australia and New Zealand more popular destinations than the USA is the better availability of free health care. Open access could be restricted to those with a certification of good health. To provide incentives for good education, basic educational and prior job requirements could also be imposed. This would effectively result in a program which is a half-way house between a lottery and an open-access scheme: that is, an uncapped, selective, country-specific program. 7. Sending country reforms It takes two to tango. The reforms required from sending countries are as important as those outlined in the previous section for receiving countries. This section addresses what will be required of labour-sending countries, beginning with education and training policies, then marketing, and finally addressing social impacts. 7.1 Education and training policies to expand labour mobility opportunities while avoiding brain drain A major obstacle sending countries face in increasing migration flows is the low education levels of the population, even among those employed in the formal sector. The small number of workers with post-school qualifications in the Pacific means that even small losses among this 34 group can have serious consequences. There is already a reliance on foreign workers at medium- and higher-skills levels in many Pacific countries. The overseas-born population of Pacific countries ranges from 6 per cent in Marshall Islands to 4 per cent in Nauru, under 2 per cent in Tonga and less than 1 per cent in each of the other countries. But the share of foreign workers in the formal sector of these countries is much higher: for example, 12 per cent of PNG’s formal sector workforce are foreign workers (Voigt-Graf 2015). The answer is not to clamp down on the hire of foreign workers but to improve education standards so that fewer foreign workers are needed. Pacific governments will need to increase the quantity and lift the quality of skills supply to expand opportunities for more than low-skill workers overseas. Post-secondary education is important for both employment in Pacific countries and to work overseas. However relatively few Pacific Islanders have post-school qualifications. Table 3 below shows for nine countries the proportion of the employed workforce in the formal sector with post-school qualifications. The range varies from nearly half (46 per cent) of the urban formal sector in Samoa to only 4-5 per cent of the formal sector workforce in Kiribati, Solomon Islands and Vanuatu. In contrast, in the labour markets of the two main migrant destination countries, the post-school qualification rate of the employed labour force is much higher. For Australia in May 2015, two-thirds of those in paid work aged 15-74 years have completed a post-school qualification (ABS 2015). In New Zealand in 2013, 43 per cent of those in employment aged 15 years and above have a post-school qualification (New Zealand Statistics 2013). Table 3 Share of formal sector labour force aged 15 years and above with a post-school qualification, per cent Percentage of formal sector labour force Total number in Country Source Date with post-school formal sector qualification Fiji 23.5 EUS1 2011 314,382 Kiribati 4.23 Census 2010 28,747 Marshall Islands 14.2 Census 2011 12,312 PNG 28.2 Census 2011 360,732 Samoa4 45.7 Census 2011 22,274 Solomon Islands 4.43 Census 2009 91,282 Tonga 20.9 Census 2011 27,799 Tuvalu 25.02 Census 2012 2,763 Vanuatu 5.03 Census 2009 64,929 Notes: 1. Employment & Unemployment Survey. 2. Described as 'tertiary' only, likely to include college but no degree. 3. Post- school qualification does not include ‘Form 7’, ‘university entrance’ or ‘some college but no degree’. 4. Urban formal sector only. The low post-secondary education level of the Pacific workforce is due in part to continued limited access to secondary education, and quality issues throughout education systems. Most 35 Pacific Island countries still only manage to get fewer than 70 per cent of their school age population to the first year of secondary school, and far fewer progress to complete their secondary schooling. Samoa and Tonga have a high participation rate (80 per cent) in secondary education. However, in Vanuatu, secondary school participation is only 50 per cent of the relevant school age population and it is just only above 60 per cent in RMI. Too many students continue to be pushed out from secondary education, many without adequate literacy and numeracy skills. The poor quality of education in the Pacific undermines the gains made in extending access. Regular standardized assessments of student learning outcomes in the region have consistently reported children failing to meet curriculum standards. Reading levels in the early primary school grades are low and progression towards fluency for comprehension is slow. For Primary Grade 3 students, only three in ten in Tonga, and one in five in Vanuatu are able to read fluently enough for comprehension (World Bank 2012). In Papua New Guinea after three years of primary schooling, only one in ten students in the Western Highlands, two in five students in Madang and half of the students in the National Capital District are fluent readers (World Bank 2014c, d, and e). Several, particularly smaller Pacific Island countries are worried that more migration will result in a ‘brain drain’ of their skilled workforce. Their concern is that too many of the few with scarce skills and experience will migrate to high-income countries. This concern has been behind government and private sector resistance to initiatives by donors to promote regional labour mobility. As discussed in Section 5, the Australia Pacific Technical College (APTC) has as one of its objectives to help its graduates gain ‘improved employment opportunities nationally, regionally and internationally in targeted sectors.’ However, as noted earlier, very few graduates have migrated internationally. APTC has been reluctant to promote international labour mobility in response to the clear indication given by stakeholders in the five countries in which the APTC has campuses that this was not an objective they supported. Emigration rates show that these concerns of small island countries are hardly surprising. Small, remote populations in the tropics in low-growth economies produce the highest emigration rates compared with other developing countries (Docquier 2014, 3). Of all the countries of less than two million people, as many as six Pacific countries are among the top ten with the highest emigration rates. At the top of the list are Tonga and Samoa with 43 and 40 per cent respectively of their populations aged 15 years and above born in these countries living overseas (Arslan et al. 2014, 41; also in Table 1). The Pacific countries next in rank order of high emigration rates are: Federated States of Micronesia (34 per cent), Palau (31 per cent), Marshall Islands (27 per cent) and Fiji (23 per cent). Pacific governments’ concerns about brain drain need to be addressed as part of a national skills investment strategy. The first step for a government in developing a national skills strategy is to work out what skill sets or qualifications the public and private sectors need. The second is to lift the quantity and quality of supply of young people with the required post-school qualifications to meet both domestic and international requirements. This requires improving education quality at all levels. 36 Migrants can pay for the cost of their education. Resources need to be allocated to enable potential migrants to attain the required education level or qualifications needed for successful migration. A student loan scheme could also be applied to all students pursuing a post-school qualification, with loans only to be repaid if the graduate emigrates for three years or more, or, more broadly, if incomes exceed a certain level. Sending governments also need better data. This requires collecting data at several levels: from the individual migrant, from within households based on gender roles, at the community level and at the macro-economic level. The data collection should involve recording the occupation and qualifications of those who emigrate and those who return. The costs to monitor include the cost of training the skilled workers who emigrate and the foregone tax revenue of the emigrant. The remittances sent by skilled emigrants also need to be measured to see to what extent they offset the loss of tax revenue (Gibson and McKenzie 2010, 125). Receiving country policies should also be directed towards reducing brain drain. As discussed in Section 5.3, any destination country ‘aid for migration’ programs should ensure that they do not exacerbate brain drain. This can be done by training up, either directly or indirectly, fresh intakes of workers either to migrate or to replace those migrating. This is essential for both economic and political reasons. 7.2 Promoting Pacific schemes and marketing Pacific workers The most effective marketing is delivered by good service for employers by both workers and sending governments. Pacific governments should prioritize labour mobility schemes and monitor their progress at the highest levels. The Pacific Island countries should invest in developing marketing strategies for their workers in key receiving countries. Few employers in the main receiving countries are familiar with countries in the Pacific region. Without the volume of workers in-country and strong reputation that some of the regional labour-sending heavyweights have, such as India and the Philippines, the Pacific Island countries will struggle to gain a foothold with employers where they are not granted preferential access. As a result, they will be required to invest in creating demand through effective marketing. Some countries already have marketing plans and strategies in relation to the SWP and RSE, but a broader approach is needed. For example, employers’ concerns about the selection and recruitment of workers can be addressed by having dedicated staff in Australia and New Zealand who are able to facilitate and trouble-shoot. Pacific Islands Trade & Invest (PT&I) should be renamed Pacific Islands Trade, Invest and Work (PTI&W) to reflect a much-needed broadening of its mandate. Labour mobility is more important to many Pacific island economies than most exports. It is anomalous for the Pacific’s trade promotion body not to have a labour-mobility remit. PTI&W, as it might be renamed, could be made responsible for reaching out to neighbouring country governments and employers to explore new labour mobility opportunities for all of the Pacific, backed by up-to-date analysis of current 37 occupations in demand and the employment outcomes of migrants from each country (Howes 2015). 7.3 Policies to manage the social impact of labour mobility Whilst the direct and indirect economic benefits of labour migration are clear (Section 2), there are potential negative social impacts. There are concerns over the cohesion of the family unit due to an extended absence by a key household member. Absent spouses can result in psychological stress caused by loneliness or feelings of neglect (Hayes 2009). There are also anecdotal reports of more gender-based violence with increased rates of alcoholism amongst returning workers. Children and the elderly are also affected by the loss of a household member who provided care. This can have broader impacts on social relations as siblings or in-laws are called in to perform functions normally expected of the absent family member. At the community level, the social impact of migration can include the undermining of valued social institutions, such as extended families, churches and patterns of leadership. Adverse family and community impacts can be minimized by avoiding long-term family separations, as all the recommendations in this report do. In the labour mobility schemes we advocate for migrants to be able to bring their families, except for three instances: seasonal workers, where the duration is short; and backpackers and Korean EPS workers, where the migrants are not expected to have dependants. In addition, negative family and community impacts can be addressed by sending countries through a variety of measures. These would include: discussing social impacts in pre-departure briefings; improving international communications infrastructure; strengthening prevention efforts around gender-based violence; improving migrant family support services in sending countries; and funding formal social protection programs for groups that are particularly vulnerable as a result of migration, such as social pensions for the elderly. Sending countries should encourage the development of codes of conduct with a publicly funded complaint mechanism. These codes of conduct need to set out the minimum attributes and abilities that a person must demonstrate to perform as a registered migration or recruitment agent. They should allow clients to make official complaints when the code has been breached or to provide feedback on how to improve the standards of the recruitment process. Generally, they should be formulated based on discussions with trade unions, government and civil society organizations. They should provide for independent monitoring mechanisms with clear criteria and penalties for noncompliance. The public, and interested workers in particular, should also be informed about firms or individual recruiters who are violating the code and what sanctions have been applied (OSCE, IOM and ILO 2006). A public report card should be set up on the web to show the performance of the agency managing the migration arrangements in responding to issues raised by returning migrants. 38 8. Labour mobility projections What would happen if the reforms of the previous sections were implemented? What would happen to migrant numbers, to remittances, to economic activity? What would be the impact on employment, and on women? This section illustrates the impacts of the reforms discussed in Sections 3-7 by a series of projections. 8.1 The scenarios: business as usual, medium growth, and high growth We first project the trends of recent years to estimate levels of labour mobility and associated remittances in the absence of reforms, out to 2040. For this ‘business-as-usual’ (BAU) scenario, we extrapolate the annual flow of non-seasonal migrants in a linear fashion for most countries, reaching a total of 11,100 by 2040 up from 8,600 in the base year (2013). (For full details on this and all other assumptions, see Annex 1.). Flows are converted to Pacific-born migrant stocks by assuming that one-third of non-seasonal migrants are on a temporary visa, and stay for an average of three years, and that the rest are permanent migrants (with an attrition rate of one per cent to account for deaths and returns). In the absence of reforms, growth in the number of seasonal workers from the Pacific is expected to reach 29,300 by 2040, compared to about 12,000 in 2015. This assumes no change in the RSE cap, and growth of 700 per year in the SWP, consistent with growth over three years from 2011-12 to 2014-15. The distribution of these places between Pacific island countries is based on historical trends, with allowance for saturation from countries that already dominate, and greater growth in numbers from the new entrant, Fiji. With reforms, it would be possible to achieve significantly higher growth rates in migrant flows. We consider two scenarios: a medium growth and a high growth scenario. The medium growth scenario envisages a near doubling of migration outflows relative to business as usual. The annual outflow of Pacific islanders for non-seasonal work is projected under the medium growth scenario to increase by 2040 from 8,600 in 2013 to not just 11,100 as under BAU, but to 20,500, an increase of 9,400. This approximate-doubling would be achieved by a number of the reforms proposed in the paper, drawing on the illustrative quantifications presented at the end of the various sub-sections in Sections 4 to 6. For example:  Assume that Australia introduces a Pacific Category equivalent (in terms of Australia’s total population) to only one-half the size to New Zealand’s Pacific Category. This itself would amount to 4,600 places (Section 6.2).  Assume that Australia and New Zealand introduce Pacific caregiver programs for those with high medical needs, along the lines introduced by Canada. Given Canada’s population, an equivalent cap for Australia and New Zealand combined would be 2,200 (Section 5.4).  Assume that Korea opens up its Employment Permit System to the Pacific and that the Pacific sends half as many workers per year as Timor Leste has done on a per capita basis. This would provide another 1,600 places per year (Section 5.2). 39 These three schemes alone would lead to annual non-seasonal employment opportunities of 8,400 workers. Many more than 1,000 would come from the other reforms proposed – for example, by welcoming Pacific backpackers (Section 5.1), and by reducing some of the barriers to Pacific participation in existing skilled and temporary migrant work programs (Sections 5.3 and 6.1). We also need to consider the geographical composition of the increase. It is assumed that the new labour mobility opportunities are directed at, and taken up by, the Pacific island countries that are the most isolated and in need. We assume no increase (relative to BAU) in the outflows from the Compact states, as they already have open access to the United States. And we assume only a modest increase (of 25 per cent) in the high-mobility states and in Tuvalu, an atoll state with already relatively high levels of outward migration. The real increase is in the low mobility states and in Kiribati, where the flow of migrants triples relative to BAU. Under the high growth scenario, annual outflows triple relative to business as usual. The increase in non-seasonal outflows in 2040 relative to BAU is not 9,400 but 18,700: annual outflows reach 29,800 in 2040. This sort of increase is also quite feasible, and could be achieved as follows:  Assume that Australia introduces a Pacific Category equivalent (in terms of Australia’s total population) to the size to New Zealand’s Pacific Category. This would be of a size of 9,200. If New Zealand were to increase the size of its scheme by 50 per cent, that would provide another 900 slots (Section 6.2).  As before assume that Australia and New Zealand introduce Pacific caregiver programs, along the lines introduced by Canada, with an equivalent combined cap of 2,200 (Section 5.3).  Assume that Korea opens up its Employment Permit System to the Pacific and that the Pacific sends as many workers per year as Timor Leste has done on a per capita basis. This would provide another 3,200 places per year (Section 5.2).  Assume that Australia allows open access to Kiribati and Tuvalu, on humanitarian grounds. As estimated earlier, that could result in a peak annual migration of approximately 2,500 migrants (Section 6.3). The above interventions would lead to non-seasonal employment of 18,000 workers. The additional 700 places to achieve the increase of 18,700 would be more-than-achieved by implementing some of the other reforms suggested in the paper (as before, by welcoming Pacific backpackers, and by reducing some of the barriers to Pacific participation in existing skilled and temporary migrant work programs). What of the geographical composition of this much increased number of migrant workers under the high growth scenario? Similarly targeted assumptions are made as under the medium growth scenario. We again assume no increase (relative to BAU) in the outflows from the Compact states, as they already have open access to the United States. We assume a 50 per cent increase in outflows from the high-mobility states and in Tuvalu, an atoll state with already relatively high levels of outward migration, and a 500 per cent increase in outflows from the low mobility states and Kiribati. 40 The number of seasonal workers would also increase with reform. Turning now to seasonal workers (so far we have considered non-seasonal migrants only), as argued in Section 4, with reforms, the number of seasonal workers could expand from 29,300 in 2040 (under BAU) to 57,500. We incorporate this increase into the high growth scenario. The medium growth scenario assumes an annual intake of 43,400, which is the midpoint between BAU and the high growth scenario. Existing shares by nationality of the Pacific seasonal work force are maintained going forward under BAU, with a few exceptions (see Annex 1, sub-section 3.5). Future shares by nationality of the Pacific seasonal work force are assumed to be the same under BAU and the two growth scenarios. The moderate and high growth scenarios are indicative only. However, both scenarios are certainly feasible. Indeed, full implementation of all the reforms outlined in this report would lead to a significantly larger number of migrants. The difference between them represents the degree of policy effort put in by both sending and receiving countries. 8.2 Business as usual results Under business as usual, the stock of Pacific-born migrants including seasonal workers in 2040 reflects modest growth of approximately 20 per cent, relative to the 2013 base year. Figure 7 encouragingly suggests that, based on recent trends, growth will be more rapid where it is most needed, among the low mobility and atoll countries, where the number of migrants increases by 70 and 100 per cent respectively. This reflects the recent growth, from a low base, in labour mobility involving these countries. However, both the absolute numbers from these countries, and their proportion as a percentage of the population resident in the sending countries remains low. The low mobility and atoll countries both have high population growth rates, and the stock of migrants relative to the resident population increases only marginally for both groups. For the low- mobility countries it increases from 0.3 to 0.5 per cent and for the atolls, it increases from 5.4 to 8.4 per cent. The low-mobility countries in particular would remain among the most isolated in the world. More broadly, labour mobility for the Pacific region as a whole fails to keep up with population growth, and the migrants/population ratio falls slightly under business as usual for the Pacific as a whole from 4.5 per cent in 2013 to 3.5 per cent in 2040. 41 Figure 7 The stock of Pacific migrants overseas in 2013 and 2040 under business as usual: total numbers and as a percentage of the resident population 600,000 40% 35% 500,000 30% 400,000 25% 300,000 20% 15% 200,000 10% 100,000 5% 0 0% 2013 2040 2013 2040 2013 2040 2013 2040 2013 2040 Open access High mobility Low mobility Atolls Total Non-seasonal Seasonal Proportion of resident population With more and better paid migrants, there would be significant income gains from labour mobility by 2040, even under BAU. The net income of both non-seasonal and seasonal migrants is defined as the expected value of the income gain from migration net of the foregone domestic earning opportunities associated with migration. It would more than double from $US12 billion in 2013 to $US23 billion in 2040 (Figure 8). (Note all monetary values here and throughout are in constant 2005 prices.) This reflects both more migrants, but also the assumption that migrants would be better paid in line with increases in OECD average income. Non-seasonal migrants earn much more than seasonal workers, and there are many more of them, so their incomes dominate. Under BAU, income from migration does not rise as a percentage of GDP and is still very unequally distributed across the Pacific, with the populations of the low mobility countries sharing in about the same level of migrant income as the much smaller open access countries, for example. 42 Figure 8 Net income of Pacific migrants in 2013 and 2040 under business as usual 25 Billions $US (2005 prices) 20 15 10 5 0 2013 2040 2013 2040 2013 2040 2013 2040 2013 2040 Open access High mobility Low mobility Atolls Total Migrant net income Seasonal worker net income With more income, more funds would be sent home. We assume that non-seasonal migrants remit a portion of their income, and that seasonal migrants take home all their income (after expenses). Under BAU, the Pacific would benefit from approximately $US1.5 billion in remittances and seasonal worker income annually by 2040, up from $US0.6 billion today (Figure 9). However, overall, the remittance/GDP ratio for the region increases only marginally from 4.2 per cent to 4.9 per cent. (Note that a regional distribution for these figures is not given due to unreliable remittance data from some countries. Detailed country estimates are available in Annex 2, however.) 43 Figure 9 Pacific remittances and seasonal worker income in 2013 and 2040 under business as usual: in $US million and as a percentage of GDP 2,000 6% Millions $US (2005 prices) 5% 1,500 4% 1,000 3% 2% 500 1% 0 0% 2013 2040 Remittances Seasonal worker income Remittances and seasonal worker income/GDP 8.3 Results from the medium and high growth scenarios The number of migrants and seasonal workers leaving the Pacific every years are by 2040 3.5 and 5 times higher in the medium and high growth scenarios than they currently are. Figure 10 shows the stock of migrants (including seasonal workers) under the three policy scenarios. Overall, as a result of the reforms there are projected to be 120,000 more Pacific migrants abroad in 2040 in the medium growth scenario than in BAU, and another 120,000 in the high growth scenario, taking the total stock to 750,000. Two-thirds of these increases are from low-mobility countries and 10 per cent from the atolls. Under the high growth scenario, by 2040 the low-mobility country migrant stock has increased nine-fold relative to 2013 and the atoll country migrant-stock has increased by almost six-fold. Figure 10 Stock of Pacific migrants in 2013 and in 2040 under the three scenarios 800,000 700,000 600,000 500,000 400,000 300,000 200,000 100,000 0 2013 2040 2013 2040 2013 2040 2013 2040 2013 2040 Open access High mobility Low mobility Atolls Total BAU Medium High Note: This and the other graphs in this section includes seasonal workers in the stock of migrants. 44 The share of the Pacific migrant stock relative to resident population stays the same as it is now or increases modestly in both reform scenarios, but it increases dramatically where it matters. Under business as usual, the Pacific migrant stock falls as a proportion of the resident population (Figure 11). Under the medium growth scenario, it more or less stays at its current level of 4.4 per cent, and under the high growth scenario it rises to 5.2 per cent. However, there is a dramatic impact for the low mobility and atoll states. In the atoll states, the proportion of the population overseas rises from 5.4 per cent in 2013 to 15 per cent in 2040 under the medium growth scenario and to 23.1 per cent under the high growth scenario. For the low-mobility countries, the ratio rises from 0.3 per cent in 2013 to 1.2 per cent in 2040 under the medium growth scenario and to 1.8 per cent under the high growth scenario. For Vanuatu and Solomon Islands the increase is more marked: for Solomon Islands, from 0.3 per cent to 2.5 or 3.9 per cent (under the two scenarios) and for Vanuatu, from 0.9 per cent to 6.4 or 9.6 per cent. Figure 11 Stock of Pacific migrants in 2013 and 2040 under the three scenarios, as a proportion of the resident population 40% 35% 30% 25% 20% 15% 10% 5% 0% 2013 2040 2013 2040 2013 2040 2013 2040 2013 2040 Open access High mobility Low mobility Atolls Total BAU Medium High There are dramatic gains in net income from these increased number of workers. The reforms generate $US5-10 billion of income for the people of the Pacific relative to BAU. Income gains for low-mobility and atoll countries are doubled or tripled relative to BAU (Figure 12). 45 Figure 12 Net income of Pacific migrants in 2013, and in 2040 under the three scenarios 35 Billions $US (2005 prices) 30 25 20 15 10 5 0 2013 2040 2013 2040 2013 2040 2013 2040 2013 2040 Open access High mobility Low mobility Atolls Total BAU Medium High Most of these net income gains are retained by the migrants, but a significant proportion is sent home, resulting in an additional $US400 million to $US800 million of remittances. Remittances and income of seasonal workers (after expenses) rises from $US0.6 billion in 2013 to $US1.5 billion in 2040 under BAU, to $US1.9 billion under the medium growth scenario and to $2.3 billion in 2040 under the high growth scenario (Figure 13).20 Remittances as a share of GDP rise from 4.2 per cent in 2013 to 4.9 per cent in 2040 under BAU to 6.2 per cent under the medium growth scenario and 7.5 per cent under the high growth scenario. (Again these figures are not shown separately for different regions due to the unreliability of individual country remittance data.) Figure 13 Pacific remittances and seasonal worker income in 2013, and in 2040 under the three scenarios: in $US millions and as percentage of GDP 2,500 8% % GDP Millions Millions $US (2005 2,000 prices) 6% 1,500 4% 1,000 2% 500 0 0% 2013 2040 2013 2040 BAU Medium High BAU Medium High 20It is also assumed in the two reform scenarios that over a ten year period the average cost of sending remittances declines from 12 per cent (the current cost) to 3 per cent, consistent with SDG targets (Betteridge and Howes 2015). 46 8.4 Economic and social impact of increased mobility There are various ways to measure the economic impact of migration. Gross Domestic Product (GDP) per capita is the measure most conventionally used by economists to measure economic activity, but is unsuitable for measuring the impact of migration since it only takes account of domestic production. Gross National Income (GNI) is a better measure since it is income based (adjusting GDP for net factor income) but it is still limited because it takes no explicit account of remittances. A better and increasingly used measure when migration is being considered is Gross National Disposable Income (GNDI), which explicitly adds remittances to GNI (Capelli and Vaggi 2014). However this too fails to fully account for the welfare gains enjoyed by migrants – it only measures financial flows back to the sending country, not the income gains of migrants themselves. In order to better assess these gains, we develop a new indicator here, which we call ‘Gross National Income+’ (GNI+). It adds to GNI the income gains of migrants, not just the remittances that they send home, thereby better accounting for total welfare gains. It thus takes into account the welfare of migrants as well as those who stay behind. We refer to this measure as per capita income throughout the remainder of the report. Increased labour mobility makes a huge difference for several countries, especially the poorer ones. In the high growth scenario, labour mobility reforms double income growth (between 2013 and 2040, measured in US$ at 2005 prices) relative to BAU for PNG and the Solomon Islands, triple income growth for Vanuatu, and quintuple income growth for Kiribati (Figure 14). (Results showing the economic impacts of the reforms in terms of other indicators are provided in Annex 2.) Figure 14 Income per capita (GNI+ per capita) in 2040, setting 2013=1, under the three scenarios 4.0 GNI+ per capita (2013=1) 3.5 3.0 2.5 2.0 1.5 1.0 BAU Medium High 47 Labour mobility reforms would create a significant number of new jobs where these are most needed. Expanding employment is critical for the future of the Pacific. Under the medium growth scenario, about 75,000 job opportunities are created by 2040 relative to BAU, and in the high growth scenario 150,000 are (Figure 15).21 Encouragingly, over half of these jobs go to Pacific islanders from the poorer, low mobility countries, and a disproportionate number go to Pacific islanders from Tuvalu and Kiribati, where employment opportunities are limited. Figure 15 New job opportunities for migrants created by 2040 as a result of labour mobility reforms under the two scenarios 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0 Medium growth High growth High mobility Low mobility Atolls There would be significant opportunities for female migrants from the Pacific. As Figure 16 shows, relative to BAU, an additional 33,000-67,000 female migrants would find non-seasonal work by 2040. At current shares, an additional annual 2,500-5,000 women would be employed as seasonal workers (also relative to BAU). There is considerable scope to increase these numbers further in the case of seasonal work. If 50 per cent of seasonal workers were women by 2040, there would be another 13,000-17,000 female seasonal workers in the two reform scenarios. 21New jobs are defined on an annualized basis. These figures do not net out employment opportunities in sending countries that are lost as a result of migration. 48 Figure 16 Additional female migrant workers in 2040, relative to 2013 Additional female non-seasonal migrants Additional female seasonal workers employed 30,000 100,000 90,000 25,000 80,000 20,000 70,000 60,000 15,000 50,000 40,000 10,000 30,000 5,000 20,000 10,000 0 0 BAU Medium High BAU Medium High Current participation rate 50% participation rate 8.5 Receiving country impacts The impacts of higher migration from the Pacific would be minimal for all of the receiving countries, even under the medium and high growth scenarios. Employed Pacific migrants would account for an insignificant share of the population at about 1 per cent under both reform scenarios in Australia (Figure 17). The share of the New Zealand population would be higher at about 6-7 per cent, but this is only one or two percentage points more than under BAU (the appropriate counterfactual). In other receiving countries, these figures are significantly lower again. The results suggests that concerns around low-skilled migration placing downward pressure on domestic wages and displacing large numbers of native workers are largely unfounded in relation to Pacific migration. Figure 17 Pacific migrants as a percentage of receiving country population: 2013, and 2040 under the three scenarios 8% 7% 6% 5% 4% 3% 2% 1% 0% Australia New Zealand 2013 2040 - BAU 2040 - Medium 2040 - High 49 9. Conclusion The countries of the Pacific faces unique development challenges resulting in a deficit of employment opportunities. There is broad consensus that expanding labour mobility opportunities is important for the future of these countries and their citizens. Section 2 on the benefits of labour mobility argues that labour mobility arrangements also provide labour-receiving countries such as Australia and New Zealand with a means to address labour shortages, and to boost prosperity and stability in low-growth countries in the Pacific. This report explores opportunities for expanding low-skilled and medium-skilled migration from the Pacific. It canvases a number of reforms. In relation to seasonal work, Section 4 proposes increasing the cap on New Zealand’s RSE scheme and removing (or at the very least, reforming) the second-year visa extension provided for backpackers in Australia in order to stimulate demand under Australia’s SWP. Both reforms would increase the number of Pacific islanders coming to New Zealand and Australia for seasonal work, while at the same time addressing problems of uncertain labour supply and workforce quality in agricultural industries. The report also discusses skilled, semi-skilled and low-skilled labour mobility programs of temporary but longer duration (1-5 years) that could be made available to Pacific islanders. Options considered in Section 5 include: extending access to Australia’s working holiday maker (backpacker) programs to the Pacific; opening up the Korean Employment Permit System to Pacific islanders; redesign of the APTC to enable it to achieve its labour mobility mandate, and use of aid funds to reduce the costs of skill certification for Pacific islanders; and establishing a Pacific in-home caregivers high-medical-care-need program, similar to that of Canada, as part of an initiative that would address an emerging labour shortage problem that is facing an ageing Australia and New Zealand. Opportunities for permanent migration are considered in Section 6. New Zealand’s positive experience with its Pacific Category visas provides lessons and offers an interesting model whereby migration is only possible on the applicant securing a job in New Zealand. The value of an Australia- New Zealand Atoll Access Agreement that includes Kiribati and Tuvalu – two countries severely affected by climate change – is also discussed as a means of facilitating orderly and well-planned ‘migration with dignity’. For beneficial migration outcomes to be maximised, sending countries must also implement reforms. Section 7 outlines these reforms, the most important of which is education and training to help Pacific island countries expand the migration opportunities available to their citizens while at the same time addressing the ‘brain drain’ challenge. Other reform challenges include better marketing at both the national and regional level, and policies to help manage the social impact of labour mobility. Section 8 quantifies the impact of reforms through projections associated with business as usual, medium growth, and high growth scenarios. The projections show that increased labour mobility is of significant economic benefit to several countries that have historically not had good access to the 50 labour markets of the OECD. Increased labour mobility makes a huge difference for several countries, especially the poorer ones. In the high growth scenario, labour mobility reforms double income growth (between 2013 and 2040) relative to BAU for PNG and the Solomon Islands, triple income growth for Vanuatu, and quintuple income growth for Kiribati. There are strong employment gains: an additional 75,000 to 150,000 migrant jobs, worth $US5-10 billion, are created as a result of reforms (relative to BAU). If 50 per cent of seasonal work was filled by women, in 2040 there would be 30,000 females employed on a seasonal basis in the OECD each year in the high growth scenario. The small size of the Pacific makes reforms that are transformative for the region doable for the metropolitan countries. Under the scenarios, Pacific migrants are projected to account for just 1 per cent or less of the Australian population or less, and 6-7 per cent of the New Zealand population (and therefore a similarly small share of the labour force). There are real benefits not only for the Pacific island countries but for destination countries such as Australia and New Zealand in terms of filling labour shortages using labour from the Pacific. The positive experiences compiled in this report show what is possible. 51 References ABS (Australian Bureau of Statistics) 2015. Education and work, Australia, May 2015, Catalogue 6227.0, Issued 27 November. Ahsan, A., Abella, M., Beath, A., Huang, Y. Luthria, M and Van Nguyen, T. 2014. International migration and development in East Asia and the Pacific. World Bank, Washington, DC. Arslan, C., Dumont, J.-C., Kone, Z., Moullan, Y., Ozden, C., Parsons, C. and Xenogiani, T. 2014. "A new profile of migrants in the aftermath of the recent economic crisis", OECD Social, Employment and Migration Working Papers, No. 160, Paris: OECD Publishing. http://dx.doi. org/10.1787/5jxt2t3nnjr5-en. Australian Government. 2015. Our North, our future: White Paper on developing Northern Australia. Canberra: Department of Prime Minister and Cabinet. Azarias, J., Lambert, J., McDonald, P. and Malyon, K. 2014, Robust New Foundations: A Streamlined, Transparent and Responsive System for the 457 Programme, Independent Review into Integrity in the Subclass 457 Programme, Canberra. https://www.border.gov.au/ReportsandPublications/Documents/reviews-and- inquiries/streamlined-responsive-457-programme.pdf. Bedford, R. and Bedford, C. 2010. International migration and climate change: A post-Copenhagen perspective on options for Kiribati and Tuvalu. In B. Burson (ed.), Climate change and migration: South Pacific perspectives (pp. 89-134). Institute of Policy Studies Bedford, R, and Bedford, C. 2015. ‘The RSE work policy and worker productivity in the horticulture and viticulture sector: a progress report”, National Institute of Demographic Analysis, University of Waikato, Hamilton. Bedford, R. and Bedford, C. 2016. “Population change in Kiribati and Tuvalu, 2015-2050: some hypothetical scenarios”, paper presented at Workshop on New Research on Pacific Labour Mobility, Development Policy Centre, Crawford School of Public Policy, ANU 2 June. Bedford, R. and Hugo, G. 2012. Population movement in the pacific: a perspective on future prospects. Wellington: Department of Labour and Canberra: Department of Immigration and Citizenship, Betteridge, A. and Howes, S. 2015. “The persistently high cost of Pacific remittances”, Devpolicy Blog, November 17. http://devpolicy.org/the-persistently-high-cost-of-pacific-remittances- 20151117/. Brown, B. 2015. “Home Vs residential care” Third Age Matters, 7 May http://thirdagematters.com.au/home-vs-residential-care/. Callister, P., Didham, R. and Badkar, J. 2014, “Ageing New Zealand: the growing reliance on migrant caregivers, a 2014 update”. Working paper, December, Callister & Associates, Wellington New Zealand. http://callister.co.nz/aged-care-workers-nz-2014.pdf. Capelli, C. and Vaggi, G. 2014. “Why Gross National Disposable Income should substitute Gross National Income”, Pavia: Universita di Pavia. Canada Immigration. 2014. ‘Major reforms to Caregiver Program announced by Canadian Government’ Canada Immigration Newsletter, November. http://www.cicnews.com/2014/11/major-reforms-caregiver-program-announced-canadian- government-114044.html#ISb1jcGW8YAEp1Gz.99. Chand, S. and Clemens, M. 2008, “Skilled emigration and skill creation: quasi-experiment", Working Paper Number 152, Center for Global Development, September, Washington DC p41. 52 Ciammichilli, F., 2014. “Medicaid parity for Pacific migrant populations in the United States” Asian American Policy Review (blog), June 2. http://hksaapr.com/medicaid-parity-for-pacific-migrant- populations-in-the-united-states/. CIC (Citizenship and Immigration Canada) 2014. Live-in Caregiver –findings from the Longitudinal Immigration Database. Ottawa: CIC. Clemens, M., Graham, C. and Howes, S. 2015. “Skill development and regional mobility: lessons from the Australia-Pacific Technical College”, Journal of Development Studies, 51(100): 1502-1517. Curtain, R. 2016a. “Pacific migration pathways and employment outcomes: barriers and opportunities”, paper presented at Workshop on New Research on Pacific Labour Mobility, Development Policy Centre, Crawford School of Public Policy, ANU 2 June. Curtain, R. 2016b “Giving the low-skilled from the Pacific the opportunity to migrate to a better life: evidence from New Zealand’s Pacific Category visa”, paper presented at Workshop on New Research on Pacific Labour Mobility, Development Policy Centre, Crawford School of Public Policy, ANU 2 June. Curtain, R. 2016c. “Aged and disability caregivers: the case for sourcing qualified carers from the Pacific”, paper presented at Workshop on New Research on Pacific Labour Mobility, Development Policy Centre, Crawford School of Public Policy, ANU 2 June. Department of Employment. 2015. Employment outlook to November 2019. Canberra: Government of Australia. Department of Employment. 2016. ‘SWP Tourism pilot fact sheet’, 6 May, Canberra https://docs.employment.gov.au/node/36196. DIBP (Department of Immigration and Border Protection). 2015a. Working Holiday Maker Visa program report. June, Canberra: Government of Australia. DIBP (Department of Immigration and Border Protection) 2015b Working Holiday Maker visa applications granted at 31 December 2014. Canberra. https://www.border.gov.au/ReportsandPublications/Documents/statistics/pv-working-holiday- programme-dec14.xls DIBP. 2015c. Temporary Work (Skilled) Visa (Subclass 457). Canberra: Government of Australia. DIBP. 2015d. 2014-15 Migration programme report (Programme year to 30 June 2015). Canberra: Government of Australia. DIBP. 2016a. ‘Subclass 457 visas granted pivot table 2015-16 to 31 March 2016 - comparison with previous years’ March BP0014-Temporary-Work-skilled-visa-granted-2016-03-31-v100.xlsx. DIBP 2016b. Temporary Graduate visa (subclass 486). https://www.border.gov.au/Trav/Visa- 1/485-. DIOC (Database on Immigrants in OECD and non-OECD Countries) (DIOC). 2014. http://www.oecd.org/els/mig/dioc.htm. DFAT (Department of Foreign Affairs and Trade). 2015 Strategic Direction Statement. Canberra: Government of Australia. Docquier, F. 2014, The brain drain from developing countries, Institute for the Study of Labor - IZA World of Labor 2014: 31 doi: 10.15185/izawol.31 wol.iza.org. Doyle, J. and Howes, S. 2015. Australia's Seasonal Worker Programme: demand-side constraints and suggested reforms. Washington, DC: World. Evalue Research. 2011. Review of the Recognised Seasonal Employer (RSE) worker pilot training programme, for the Ministry of Foreign Affairs and Trade, Government of New Zealand 53 https://www.mfat.govt.nz/assets/_securedfiles/Aid-Prog-docs/Evaluations/Review-of-the- Recognised-Seasonal-Employer-Worker-Pilot-Training-Programme.pdf. Gibson, J., McKenzie, D., and Rohorua, H. 2013. “Development impacts of seasonal and temporary migration: a review of evidence from the Pacific and Southeast Asia”, Asia and the Pacific Policy Studies, vol. 1, no. 1, pp. 18–32. Gibson, J, and McKenzie, D. 2010. “The development impact of a best practice seasonal worker policy.” Policy Research Working Paper 5488, World Bank, Washington DC. Gibson, J. and McKenzie, D. 2013, "Scientific mobility and knowledge networks in high emigration countries: evidence from the Pacific." Discussion Paper, Centre for Research and Analysis of Migration, London. Gibson, J., McKenzie, D. and Stillman, S. 2013a. “How important is selection? Experimental vs. non- experimental measures of the income gains from migration”, Journal of the European Economic Association, 8(4): 913-945. Gibson, J., McKenzie, D. and Stillman, S. 2013b. “Accounting for selectivity and duration-dependent heterogeneity when estimating the impact of emigration on incomes and poverty in sending areas”. Economic Development and Cultural Change 61(2): 247-280. Gibson, J., McKenzie, D. Rohorua, H. and Stillman, S. 2015. “The long-term impacts of international migration: evidence from a lottery”, Policy Research Working Paper 7495, Development Research Group, Finance and Private Sector Development Team, World Bank November. Hay, D. and Howes, S. 2012 “Australia’s Pacific Seasonal Worker Pilot Scheme: why has take-up been so low?” Development Policy Centre Discussion Paper 5. Hayes, G. 2009. “Maximizing the development benefits and minimizing the negative impact in the Pacific Islands sub-region”. Suva: United Nations Economic and Social Commission for Asia and the Pacific. HEPS (Hello Employment Permit System). 2013, “EPS-TOPIC passing rate by country”, viewed 02 July 2014, http://www.helloeps.com/jywqyb4asen2wzht9gwn/. Horticulture NZ. 2009. An overview: Horticulture Industry Strategy: ‘Growing a new future’, http://www.hortnz.co.nz/assets/Uploads/HortNZ-Strategy-Overview.pdf. Howes, S. (2015) “Pacific Islands, Trade, Invest and Work?” Devpolicy Blog, 4 February http://devpolicy.org/pacific-islands-trade-invest-and-work-20150204/. Immigration NZ (Immigration New Zealand) 2015a. New Zealand Working Holiday Scheme. Wellington: Immigration NZ. Immigration NZ. 2015b. Essential Skills Work Category Requirements. Wellington: Immigration NZ. Immigration NZ. 2015c. ‘W3 - Work applications approved by occupation’, Statistics files for download: Work. https://www.immigration.govt.nz/about-us/research-and-statistics/statistics. IPCC 2012. Managing the risks of extreme events and disasters to advance climate change adaptation. A special report of Working Groups I and II of the Intergovernmental Panel on climate Change [Field, C.B et al. (eds. Cambridge University Press, Cambridge. Johanson, R. et al. 2014. Independent review of the Australia-Pacific Technical College: Full Report. Canberra: Department of Foreign Affairs and Trade, Australian Government. Johnson, G. 2015 “So far, most atolls winning the sea level rise battle.” Pacific Institute of Public Policy, 11 May 2015. http://pacificpolicy.org/2015/05/so-far-most-atolls-winning-the-sea-level- rise-battle/. Marles, R. 2015. “Australia must not be afraid of its obligations to Pacific Climate Migrants.” The Guardian (online edition). November 11. 54 http://www.theguardian.com/commentisfree/2015/nov/11/australia-must-not-be-afraid-of-its- obligations-to-pacific-climate-migrants. McKenzie, D. 2006. “Remittances in the Pacific”. Paper for 2005-2006 Werner Sichel Lecture- Seminar. Washington, DC: World Bank. Merwood, P. 2013a. “Labour market outcomes of recent migrants to New Zealand.” Migration Research, Evaluation & Analysis, Strategy & Governance Group, Ministry of Business, Innovation & Employment, June. Merwood, P. 2013b. “Settlement outcomes for Pacific migrants to New Zealand. Presentation to ‘Managing Migration in the Pacific: Agreements, Outcomes, Futures’”, Ministry of Business, Innovation & Employment, Wellington, 12 September. Ministry of Business, Innovation and Employment. 2015. Short-term Employment Forecasts: 2015- 2018. August, Wellington, Government of New Zealand. MFAT (Ministry of Foreign Affairs and Trade). 2015a. Statement of strategic intentions, 2015-19, Wellington, Government of New Zealand. MFAT. 2015b. Canterbury Trades Employment Initiative, November, Wellington, Government of New Zealand. Ministry of Foreign Affairs, Trade, Tourism, Environment and Labour. 2015. Tuvalu National Labour Migration Policy, Government of Tuvalu. NILS (National Institute of Labour Studies). 2012. The aged care workforce, 2012 – Final Report. Flinders University, Adelaide. OECD (Organization for Economic Cooperation and Development). 2011. Help wanted? Providing and paying for long-term care. Paris. OECD (Organization for Economic Cooperation and Development). 2014a. International Migration Outlook 2014. Paris. OECD. 2014b. Recruiting immigrant workers: New Zealand 2014. Paris, OECD. OECD. 2015. International Migration Outlook 2015. Paris: OECD. OSCE, IOM and ILO (Organization for Security and Co-operation in Europe, International Organization for Migration, International Labour Office). 2006. Handbook on establishing effective migration policies in countries of origin and destination. Vienna and Geneva, OSCE, IOM, and ILO. Phillips, J. 2006. Skilled migration to Australia. E-Brief. Canberra: Parliament of Australia, 5 June. PIFS (Pacific Islands Forum Secretariat). 2014. Strengthening economic linkages – financial infrastructure and fiscal space. Suva, PIFS. Productivity Commission. 2015. Migrant intake into Australia, Draft Report. Canberra, November. Ritter, D. 2015. “Climate change and human rights: the imperative for climate change migration with dignity (CCMD)” Global Policy (blog), November 24. http://www.globalpolicyjournal.com/blog/24/11/2015/climate-change-and-human-rights- imperative-climate-change-migration-dignity-ccmd. SEPFOPE (Secretary of State for Professional Training and Employment Policy). 2014. Sending peoples overseas, generating money. Dili: Government of Timor-Leste. Small, C. 1997. Voyages. From Tongan villages to American suburbs. Ithaca, NY, and London: Cornell University Press. UN Population Division. 2015. World Population Prospects: 2015 Revision. New York: United Nations. 55 US DoL (United States Department of Labor). 2015. Section H-2A of the Immigration and Nationality Act (INA). Washington, DC. Varghese, P. 2016. “Opening address to the 2016 Australasian Aid Conference”, Development Policy Centre, The Australian National University, 10 February. Voigt-Graf, C. and Kagan, S. 2015. “Migration and labour mobility from Kiribati”, paper presented at Workshop on New Research on Pacific Labour Mobility, Development Policy Centre, Crawford School of Public Policy, ANU 2 June. Voigt-Graf, C. 2015a. “The changing composition of PNG’s foreign workforce”, Devpolicy Blog, 16 March http://devpolicy.org/the-changing-composition-of-pngs-foreign-workforce-20150316/. Voigt-Graf, C. 2015b “Why are so few Pacific islanders working in PNG?” Devpolicy Blog, 23 September, http://devpolicy.org/why-are-so-few-pacific-islanders-working-in-png-20150923/. Whatman, R. and Van Beek, J. 2008. “The seasonal labour strategy and the role of Recognised Seasonal Employer (RSE) in helping make transformative changes for employers and industry”, Labour, Employment and Work in New Zealand 2008, Victoria University, Wellington. Wigglesworth, A & Fonseca, M. 2016. ‘Experiences of Young Timorese as Migrant Workers in Korea: Lessons for Policy Makers’, paper presented at Workshop on New Research on Pacific Labour Mobility, Development Policy Centre, Crawford School of Public Policy, ANU, 2 June. Woolford, G. 2009. ‘Social protection for migrants from the Pacific Islands in Australia and New Zealand.’ SP Discussion Paper No. 0912. Sydney: World Bank World Bank. 2006. Expanding job opportunities for Pacific Islanders through labour mobility: at home and away. Washington, DC: World Bank. World Bank. 2007. General principles for international remittance services. Washington, DC World Bank. 2012 “Tonga and Vanuatu reading assessment”, Feature Story 10 October. http://www.worldbank.org/en/news/feature/2012/10/10/tonga-vanuatu-reading-assessment. World Bank. 2014a. International migration and development in East Asia and the Pacific. Washington DC. World Bank. 2014b. Well-being from work in the Pacific Island Countries. Washington DC. World Bank. 2014c. “In Papua New Guinea, how well are children learning to read in Madang?” Pacific Early Reading Assessments Series. World Bank. 2014d. “In Papua New Guinea, how well are children learning to read in the National Capital District?” Pacific Early Reading Assessments Series. World Bank. 2014e. “In Papua New Guinea, how well are children learning to read in the Western Highlands Province?” Pacific Early Reading Assessments Series. World Bank. 2016. Global Monitoring Report 2015/2016: development goals in an era of demographic change. Washington, DC. DOI: 10.1596/978-1-4648-0669-8 56 Annex 1 – Methodology for projections Projections are made for the following Pacific Island countries (PICs): Fiji, Kiribati, Marshall Islands, Micronesia (Fed. States of), Palau, Papua New Guinea, Samoa, Solomon Islands, Tonga, Tuvalu, and Vanuatu. In the main report, results are shown for migrants including seasonal workers, but in this annex the term ‘migrants’ excludes seasonal workers. Also, as in the main report, only remittances from non- seasonal migrants are counted as remittances. Net seasonal worker income is incorporated into the measures of national accounts as described below, as income of PIC nationals working overseas for a period of less than one year. Scenarios Three scenarios are examined as part of these projections. (a) A 'business as usual' (BAU) scenario, in which migration from PICs continues as per historical trends. (Further details are provided below: see 3.2 in particular.) (b) A ‘medium growth’ (MG) scenario, in which migration from PICs is significantly higher than under BAU. In the medium growth scenario, annual migration from countries with open labour access (FSM, RMI, Palau) remains the same as in BAU, increases by 25 per cent relative to BAU for countries with high levels of labour mobility (Fiji, Samoa, Tonga, and Tuvalu), and increases three- fold relative to BAU for countries with limited levels of labour mobility (PNG, Solomon Islands, Vanuatu, and Kiribati). With implementation of some reforms, we project that the number of seasonal workers would expand from 29,300 in 2040 (under BAU) to 43,400 in the medium growth scenario (half-way between BAU and the high growth scenario below). Seasonal work shares by nationality of the Pacific seasonal work force remain the same as under BAU. (c) A 'high growth' (HG) scenario, in which migration from PICs is higher still. In the ‘high growth’ scenario, migration flows from PICs remain the same for countries with open labour access (FSM, RMI, Palau) relative to BAU, increases by 50 per cent relative to BAU for countries with high levels of labour mobility (Fiji, Samoa, Tonga, and Tuvalu), and increases five-fold relative to BAU for countries with limited levels of labour mobility (PNG, Solomon Islands, Vanuatu, and Kiribati). The number of seasonal workers increases from 29,300 in 2040 (under BAU) to 57,500. This assumes a doubling of the NZ RSE cap to 20,000 (most but not all of which would go to Pacific Islanders), and a significant expansion of Australia’s SWP to 40,000 places each year by 2040, consistent with a high growth scenario identified by World Bank (2014, 19). Seasonal work shares by nationality of the Pacific seasonal work force remain the same as under BAU. Welfare indicators Four welfare indicators are of particular interest to this study. More information on the calculation of three of these indicators (GDP, GNI, GNDI) is available in the latest System of National Accounts (2008) adopted by the United Nations Statistical Commission (UNSC). Calculation of the fourth (GNI+) is discussed below. The four indicators include: (i) Gross Domestic Product (GDP) - the monetary value of all the finished goods and services produced within a country's borders in a specific time period. GDP is the sum of consumption, investment, government, expenditure, and the external balance (exports minus imports) of the economy. 57 It is expressed by the equation: GDP = C + I + G + (X-M) where C is consumption, I is investment, G is government spending, and (X-M) is net exports. GDP is a poor indicator when analysing the impacts of labour mobility. Migration is likely to diminish GDP by reducing the workforce. But GDP does not directly incorporate the benefits of migration, namely income earned by nationals overseas and remittance income. However, it is influenced indirectly by remittances and foreign income to the extent that these are spent in the economy, generating economic activity (assuming current underemployment of domestic factors of production). (ii) Gross National Income (GNI) – the total domestic and foreign output claimed by residents of a country, consisting of gross domestic product (GDP) plus factor incomes earned by foreign residents, minus income earned in the domestic economy by non-residents. It is expressed by the equation: GNI = GDP + NPI (net primary income) where NPI = remuneration of factors of production (capital and labour) receivable from the rest of the world less those payable to the rest of the world. GNI is a better indicator than GDP when analysing the impacts of labour mobility, as it incorporates the income of nationals working overseas on a short-term basis (convention defines this period as less than one year). GNI therefore incorporates the incomes of seasonal or other workers from the Pacific working overseas for less than twelve months, whereas GDP does not. It also subtracts the income of foreigners working in PICs for short periods (again, for less than a year). However, GNI does not incorporate remittance flows, meaning that the income and remittances of migrants that live overseas for longer than one year are not reflected in GNI. (iii) Gross National Disposable Income (GNDI) measures the income available to the total economy for final consumption and gross saving. It is expressed by the equation: GNDI = GNI + NSI (net secondary income) where NSI = current transfers (aid, remittances) receivable from the rest of the world less those payable to the rest of the world. GNDI (Capelli and Vaggi 2014) is the best of the three official indicators of national output/income outlined here as is includes remittances flows from migrants living overseas. However, it does not capture the welfare gains for migrants that are associated with migration. (iv) Gross National Income+ (GNI+) is a measure created for the purposes of this study, which is designed to better capture the welfare impacts of migration for Pacific islanders. This is considered the measure of wellbeing that most comprehensively captures the economic impacts of migration. As a result, this measure is emphasised in the report, where it is reported in per capita terms as ‘income per capita’. GNI+ is a variation to GNI. It captures the income of Pacific Islanders irrespective of where they reside. As a result, it is better suited to capturing the welfare impact of overseas migration for Pacific Islander permanent migrants. It is expressed by the equation: GNI+ = GNI + income gains of migrants living overseas for more than 12 months. 58 Remittances from permanent migrants are not separately included in measurement of GNI+, as the income of these migrants is captured in full by GNI+, and therefore, income earned that is sent as remittances is already included in the GNI+ measure. Incorporating remittances explicitly would lead to double-counting. The income of seasonal workers is incorporated into GNI+, just as it is in GNI. More concretely, the projections incorporate the impact of labour mobility on economic well-being as follows: (i) GDP - The loss of economic activity undertaken by migrants is taken into account by subtracting GDP per capita in the sending country for every migrant. The indirect impact of all foreign remittances and seasonal worker income on economic activity in PICs is incorporated by using a GDP multiplier, assuming a Keynesian framework whereby factors of production in PICs are currently underemployed. (ii) GNI - Impacts on GDP as outlined above + income of PIC nationals working overseas for a period of less than one year net the opportunity cost of their migration (that is, net seasonal worker income). (iii) GNDI - Impact on GNI as above + remittances from PIC nationals working overseas for more than one year. (iv) GNI+ - Impact on GNI as above + income of PIC nationals working overseas for more than one year. List of indicators (Note: Not all indicators are reported on in this report. All are available on request.) National accounts under BAU scenario 1.1 GDP under BAU scenario 1.2 GNI under BAU scenario 1.3 GNDI under BAU scenario 1.4 GNI+ under BAU scenario National accounts under medium growth/high growth scenarios 2.1 GDP under MG/HG scenarios 2.2 GNI under MG/HG scenarios 2.3 GNDI under MG/HG scenarios 2.4 GNI+ under MG/HG scenarios Population, migration and migrant income under BAU 3.1 Stock of PIC non-seasonal migrants in OECD under BAU 3.2 Annual inflow of PIC non-seasonal migrants to OECD under BAU 3.3 Remittances of PIC non-seasonal migrants in OECD under BAU 3.4 Income and net income of PIC non-seasonal migrants in OECD under BAU 3.5 Annual number of PIC seasonal workers in OECD under BAU 3.6 Income and net income of seasonal workers under BAU 3.7 Resident population in PICs under BAU 59 Population, migration and migrant income under medium growth/high growth scenarios 4.1 Stock of PIC non-seasonal migrants in OECD under MG/HG 4.2 Annual inflow of PIC non-seasonal migrants to OECD under MG/HG 4.3 Remittances of PIC non-seasonal migrants in OECD under MG/HG 4.4 Income and net income of PIC non-seasonal migrants in OECD under MG/HG 4.5 Annual number of PIC seasonal workers in OECD under MG/HG 4.6 Income and net income of seasonal workers under MG/HG 4.7 Population in PICs under MG/HG Employment outcomes 5.1 Employment numbers in PICs under BAU 5.2 Employment to resident population ratio in PICs under BAU 5.3 Labour force (numbers) in PICs under BAU 5.4 Employment numbers in PICs under MG/HG 5.5 Employment to resident population ratio in PICs under MG/HG 5.6 Labour force (numbers) in PICs under MG/HG 5.7 New job opportunities for migrants under MG/HG Gender 6.1 Stock of female migrants under BAU 6.2 Number of female seasonal workers under BAU 6.3 Number of female seasonal workers under BAU, assuming female participation rate of 50 per cent 6.4 Stock of female migrants under MG/HG 6.5 Number of female seasonal workers under MG/HG 6.6 Number of female seasonal workers under MG/HG, assuming female participation rate of 50 per cent Projection of indicators As with any long run forward-looking exercise, these estimates are subject to wide margins of error. Methods are kept as simple as possible given the uncertainty of assumptions made in producing the projections. The approach taken in estimating projections for each variable is detailed below. Where references are made to assumptions, these are detailed in the assumption section below. All projections are in constant prices (2005 $US) unless otherwise stated. National accounts under BAU 1.1 GDP under BAU scenario GDP under the BAU reference scenario has been projected by the PP team at the World Bank. This includes GDP in current and constant prices, GDP growth, and a GDP deflator. GDP in constant prices ($US 2005) is used throughout this exercise (as with all projections made here). 1.2 GNI under BAU scenario There were no existing projections of net primary income to help guide GNI estimates. Projecting future GNI on the basis of historical GNI data led to large discrepancies between BAU GDP (1.1) and BAU GNI, even though the two are related. 60 The approach taken to estimate future GNI was based on GDP projections (1.1) supplied by the World Bank’s Pacific Possible team, assuming the GNI/GDP ratio for 2015 stays the same in the future. As with all projections, GNI is expressed in constant prices (2005 $US). 1.3 GNDI under BAU scenario GNDI is calculated using GNI (1.3) projections, to which remittance income (3.3) is added. 1.4 GNI+ under BAU scenario GNI+ is calculated using GNI (1.3) projections, to which the income of migrants is added. National Accounts under medium growth/high growth scenarios 2.1 GDP under MG/HG scenarios GDP under the MG/HG scenarios is calculated using GDP under BAU (1.1), to which the indirect impacts of net seasonal worker income and migrant remittances is added. This indirect impact is calculated using the GDP multipliers developed by Pacific Possible team for each PIC and described in the assumptions section. The full value of remittances and seasonal worker income is not included in GDP calculations. The loss of economic activity undertaken by migrants is also taken into account by subtracting GDP per capita in the sending country for every migrant. 2.2 GNI under MG/HG scenarios GNI under MG/HG scenarios is calculated using GNI under BAU as a base to which is added seasonal worker income that is additional to BAU (calculated as 4.6 less 3.6) and the indirect impacts of remittances and seasonal worker income on GDP that are additional to BAU (calculated as 2.1 less 1.1). The loss of economic activity undertaken by migrants is also taken into account by subtracting GDP per capita in the sending country for every migrant. 2.3 GNDI under MG/HG scenarios GNDI is calculated using GNI (2.2) projections, to which remittance income under MG/HG scenarios (4.3) is added. 2.4 GNI+ under MG/HG scenarios GNI+ is calculated using GNI under MG/HG (2.3) projections, to which the income of migrants under MG/HG scenarios (4.4) is added. Population, migration and migrant income under BAU 3.1 Stock of PIC non-seasonal migrants in OECD under BAU Calculated on the basis of 2013 OECD stock data (see Table 1 for details and sources), with stock projections based on inflow projections (3.2). Projections of stock data are adjusted to take into account attrition. All migrants resident in destination countries in 2013 are assumed to be long-term migrants, with an annual attrition rate of 1 per cent. Two-thirds of ‘new’ migrants, or migrants that migrate after 2013 are assumed to also be long-term migrants, with the same attrition rate. One- third of ‘new’ migrants are assumed to be medium-term migrants, who stay in the destination country for an average of three years. One-third of these ‘new’ medium-term migrants are assumed to return home each year. 3.2 Annual inflow of PIC non-seasonal migrants to OECD under BAU We utilise inflow data in order to project both future inflows and stocks of PIC migrants to OECD. Projections of inflows are linear projections, or in other words, a continuation of historical trends. There are some exceptions. Where projections produced using this method were not credible, caps were placed on inflow projections. These caps and other alterations include the following: 61  Fiji inflows remain at 4,000 per annum instead of continuing to drop into negative territory (i.e., net immigration to Fiji). While a drop from historical figures is believable given increased political stability and improved economic conditions, pull factors (in the form of higher incomes) in the OECD will continue to attract Fijians, who with high education levels and diaspora networks are in a good position to migrate.  RMI projections are based on linear projections made from 2006 onwards given dubious earlier data.  Kiribati projections are based on linear projection made from 2004 onwards given dubious earlier data (and differences with stock data)  Palau projections are capped at 81 per annum (the figure for 2013), given its small population and good economic conditions. 3.3 Remittances of PIC migrants in OECD under BAU Remittances are calculated on the basis of the stock of non-seasonal migrants (3.1) multiplied by average remittances from each PIC (see assumptions). 3.4 Income and net income of non-seasonal migrants under BAU This is calculated on the basis of the stock of non-seasonal migrants (3.1) multiplied by average income or net income of non-seasonal migrants. GDP per capita in destination (OECD) countries is used as a proxy for non-seasonal migrant income. GDP per capita in source countries (PICs) was used as an estimate of opportunity cost associated with non-seasonal migration (what the migrant would have been expected to earn in the PIC). Subtracting this opportunity cost from income gives the net income of non-seasonal migrants. 3.5 Annual number of PIC seasonal workers in OECD under BAU Under BAU, New Zealand’s latest RSE cap is assumed to remain the same, while Australia’s SWP continues to expand at the same rate as over the three years from 2011-12 to 2014-15 (about 700 per year) to reach 29,300 in 2040. The starting point for estimating the future distribution of these places between Pacific island countries is analysis of historical trends. Allowance is also made for saturation from countries that already dominate these schemes (Tonga, Vanuatu, and to a lesser extent, Samoa), and for greater growth in numbers from the new entrant, Fiji. We assume that RMI, FSM or Palau continue to have no seasonal workers, given other migration opportunities. 3.6 Income and net income of seasonal workers under BAU Calculated as the number of seasonal workers in any given year multiplied by average income or net income of seasonal workers (details for which are in the assumptions section). 3.7 Population in PICs under BAU Taken from UN projections used by the Pacific Possible team in the BAU scenario. Population, migration and migrant income under medium growth/high growth scenarios 4.1 Stock of PIC migrants in OECD under MG/HG Calculated on the basis of 2013 OECD stock data, with stock projections based on inflow projections under MG/HG (4.2). Projections of stock data are adjusted to take into account attrition. All migrants resident in destination countries in 2013 are assumed to be long-term migrants, with an annual attrition rate of 1 per cent. Two-thirds of ‘new’ migrants, or migrants that migrate after 2013 are assumed to also be long-term migrants, with the same attrition rate. One-third of ‘new’ migrants are assumed to be medium-term migrants, who stay in the destination country for an average of three years. One-third of these ‘new’ medium-term migrants are assumed to return home each year. 62 4.2 Annual inflow of PIC migrants to OECD under MG/HG In the medium growth scenario, annual migration from countries with open labour access (FSM, RMI, Palau) remains the same as in BAU, increases by 25 per cent relative to BAU for countries with high levels of labour mobility (Fiji, Samoa, Tonga, and Tuvalu) by 2027, and increases three-fold relative to BAU for countries with limited levels of labour mobility (PNG, Solomon Islands, Vanuatu, and Kiribati) by 2027. In the high growth scenario, migration flows from PICs remain the same for countries with open labour access (FSM, RMI, Palau) remains the same as in BAU, increases by 50 per cent relative to BAU for countries with high levels of labour mobility (Fiji, Samoa, Tonga, and Tuvalu), and increases five-fold relative to BAU for countries with limited levels of labour mobility (PNG, Solomon Islands, Vanuatu, and Kiribati). 4.3 Remittances of PIC migrants in OECD under MG/HG Remittances are calculated on the basis of the stock of migrants (4.1) multiplied by average remittances from each PIC (see assumptions). 4.4 Income and net income of migrants under MG/HG This is calculated on the basis of the stock of migrants (4.1) multiplied by average income or net income of migrants. GDP per capita in destination (OECD) countries is used as a proxy for non- seasonal migrant income. GDP per capita in source countries (PICs) was used as an estimate of opportunity cost associated with non-seasonal migration (what the migrant would have been expected to earn in the PIC). Subtracting this opportunity cost from income gives the net income of non-seasonal migrants. 4.5 Annual number of PIC seasonal workers in OECD under MG/HG In the medium growth scenario, we project that the number of seasonal workers would expand from 29,300 in 2040 (under BAU) to 43,400 in the medium growth scenario (half-way between the BAU and high growth scenario forecasts). In the high growth scenario, the number of seasonal workers increases from 29,300 in 2040 (under BAU) to 57,500. This assumes a doubling of the NZ RSE cap to 20,000 (most of which filled by Pacific islanders), and a significant expansion of Australia’s SWP to 40,000 places each year by 2040, consistent with the reform scenario identified by the World Bank in an earlier analysis (World Bank 2014b, 19). Shares by nationality of the Pacific seasonal work force would remain the same in both scenarios as under BAU. 4.6 Income and net income of seasonal workers under MG/HG Calculated as the number of seasonal workers in any given year multiplied by average income or net income of seasonal workers (details for which are in the assumptions section). 4.7 Population in PICs under MG/HG Population projections developed for the BAU scenario are adjusted in order to account for out- migration from PICs to OECD countries. Migrant numbers, as well as future children of migrants, are accounted for when making this calculation (see assumptions section below for fertility assumptions). Employment outcomes 5.1 Employment numbers in PICs under BAU Projections produced by the World Bank Pacific Possible team. 5.2 Employment to resident population ratio in PICs under BAU Projections produced by the World Bank Pacific Possible team. 5.3 Labour force (numbers) in PICs under BAU Projections produced by the World Bank Pacific Possible team. 63 5.4 Employment numbers in PICs under MG/HG Employment figures in the MG/HG scenarios adjust the BAU figures (5.1) based on: (i) the number of additional long-term migrants * employment rates of Pacific migrants in destination countries (as recorded in OECD data) - the number of additional long-term migrants * employment rates in sending countries; (ii) number of additional seasonal workers (who are assumed to work, given this is a requirement of migration programs) - employment rates of seasonal workers in sending countries; and (iii) Secondary employment impacts, which are calculated based on the indirect economic effects (through spending multipliers) of remittances and seasonal worker income. 5.5 Employment to resident population ratio in PICs under MG/HG Employment to population ratio is based on the population (15+) estimates produced by the PP team. This is consistent with the approach used by the PP team in calculating 5.2. No adjustments have been made due to migration, given that the employment figures encompass both migrants and non-migrants [i.e., the term does not refer to domestic employment only]. This makes adjusting population as a result of emigration inappropriate. 5.6 Labour force (numbers) in PICs under MG/HG Calculated as Labour force in BAU less additional migration in MG/HG * labour force participation rate (calculated by PP Team)) 5.7 New job opportunities for migrants under MG/HG The number of additional long-term migrants * employment rates of Pacific islander migrants in destination countries (as recorded in OECD data), plus the number of additional seasonal workers * 0.5 (given the assumption that seasonal workers are employed for only 6 months in the year). Gender 6.1 Stock of female migrants under BAU Calculated using migrant stock projections under BAU (3.1) and the percentage of migrants that are female (calculated from OECD data, as detailed in the assumptions). Note that the proportion of female migrants from the Pacific is assumed to hold constant over time. 6.2 Number of female seasonal workers under BAU Calculated using seasonal worker projections under BAU (3.5) and the percentage of seasonal workers that are female (calculated using historical data, as detailed in the assumptions). Note that the proportion of female seasonal workers from the Pacific is assumed to hold constant over time. 6.3 Number of female seasonal workers under BAU, assuming female participation rate of 50 per cent Calculated using seasonal worker projections under BAU (3.5) and an assumption that female participation climbs to 50 per cent by 2020, thereafter remaining constant. 6.4 Stock of female migrants under MG/HG Calculated using migrant stock projections under MG/HG (4.1) and the percentage of migrants that are female (calculated from OECD data, as detailed in the assumptions). Note that the proportion of female migrants from the Pacific is assumed to hold constant over time. 6.5 Number of female seasonal workers under MG/HG Calculated using seasonal worker projections under MG/HG (4.5) and the percentage of seasonal workers that are female (calculated using historical data, as detailed in the assumptions). Note that the proportion of female seasonal workers from the Pacific is assumed to hold constant over time. 64 6.6 Number of female seasonal workers under MG/HG, assuming female participation rate of 50 per cent Calculated using seasonal worker projections under MG/HG (4.5) and an assumption that female participation climbs to 50 per cent by 2020, thereafter remaining constant. Assumptions This section highlights some of the key assumptions. Average remittances from migrants Calculation of average remittances of PIC migrant are made on the basis of World Bank annual remittance inflows data, OECD data on the stock of Pacific migrants, and OECD data on seasonal worker numbers. Income from seasonal workers is included in World Bank remittance figures, and must be subtracted from total remittance data in order to estimate the average remittances sent by PIC migrants. Without detailed data on the incomes of seasonal workers, estimates produced by Gibson and McKenzie (2010) are used, and are assumed to be similar across seasonal workers from different countries (see below). Given the very approximate nature of these estimates, and the simplifications detailed above, it was better to base projections on average remittances of PIC migrants across PICs rather than continue to use PIC-specific data, which included very large (and not credible) differences across PICs. To illustrate, using the method outlined above, average annual remittances from PNG migrants were estimated to be $US435, as opposed to those from Solomon Islands which were $US5,718. However, basing all projections on a PIC-wide average remittance figure led to inconsistencies between historical remittance data (from which very large differences in average remittance flows were established) and projected remittances. Country specific average remittances are therefore assumed to converge gradually over a ten year period, which facilitates the use of a PIC-wide weighted average remittance figure from 2023 onwards. Remittance flows are also adjusted upwards on the basis of increasing GDP per capita in OECD countries (projections for which are sourced from OECD Economic Outlook). Increases are proportional to the increase in OECD GDP per capita. Given the very large adjustments involved, country-specific remittance findings should be treated with great caution. Remittances are also adjusted to account for lower remittance costs in the future (see below). Average income and net income of seasonal workers Gibson and McKenzie (2010) estimate that the income (after taxes but before expenses) of SWP workers is between $A12,000 and $A13,000, while for RSE workers it is $NZ12,351 (after the new 10.5 per cent tax), or $A11,170 In order to simplify, it is assumed here that average income after taxes but before expenses from any seasonal work (in Australia or New Zealand) is $A12,000 ($US8,720). However, expenses to participate in seasonal work (which accrue overseas) and the opportunity cost of participation (due to foregone domestic economic activity) also need to be accounted for. Gibson and McKenzie estimate that earnings less expenses are between $A4,000-$6,000 per worker. We call this income for seasonal workers and estimate it to be $A5,000 (equivalent to $US5,000 using exchange rates in place at the time the data was produced). Subtracting the opportunity cost (including farming for 65 those rural residents not employed prior to participation in seasonal work) give a net income (less expenses and opportunity cost), equal to $A2,600 (or $US2,600 using exchange rates in place at the time the data was produced). However, these calculations exclude $400 that seasonal workers are estimated to remit to households other than their own. This is included here, meaning that average net income is approximately $3,000 (again with $US and $A equal at the time these estimates were made). GDP multipliers Long run GDP multipliers are used throughout for each PIC. These were calculated by the World Bank PP Team and are listed in the table below. GDP multiplier Fiji 0.41 Kiribati 0.41 Marshall Islands 0.34 Micronesia (Fed. States of) 0.45 Palau 0.36 Papua New Guinea 0.46 Samoa 0.66 Solomon Islands 0.43 Tonga 0.68 Tuvalu 0.24 Vanuatu 0.57 Fertility of migrants It is assumed that migrants will have 0.5 children each in the future (and would have had the same number had they remained in their PIC). This estimate may seem low, but it accounts for the fact that many migrants already have children, and that those children may not have their own offspring prior to 2040. Remittance costs Fees associated with sending remittances currently average 12 per cent for PICs. It is assumed in this exercise that these fees decline over a 10 year period in order to meet the SDG target of 3 per cent. Employment impacts 62 per cent of seasonal workers are assumed to be unemployed prior to participation in the scheme. This figure is taken from Gibson and McKenzie (2010), who estimate that 38 per cent of seasonal workers are employed prior to participation. The figure is assumed to hold for all PICS. Participants are employed for only part of the year in seasonal work – here, it is assumed employment is of 6 months duration. 66 Gender The participation rate for females in seasonal work programs (RSE and SWP) is assume to remain at its historical level of 17 per cent, unless specified otherwise. The percentage of migrants who are females is assumed to remain the same as is currently the case according to OECD data. These are as follows for different PIC sending countries: Proportion of stock of PIC migrants in OECD who are: Male Female Fiji 0.4703 0.5297 FSM 0.4779 0.5221 Kiribati 0.421 0.579 Marshall Is 0.4823 0.5177 Nauru 0.453 0.547 Palau 0.5195 0.4805 PNG 0.441 0.559 Samoa 0.4902 0.5098 Solomon Islands 0.4797 0.5203 Tonga 0.5179 0.4821 Tuvalu 0.4723 0.5277 Vanuatu 0.5166 0.4834 67 Annex 2 – Detailed projections GDP per capita ($US 2005) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 3828 4376 5206 6195 3828 4383 5238 6256 3828 4389 5270 6321 Kiribati 1109 1186 1219 1308 1109 1203 1284 1430 1109 1221 1358 1580 Marshall Islands 3045 3402 3769 3907 3045 3402 3769 3907 3045 3402 3769 3907 Micronesia (Fed. States of) 2338 2361 2291 2288 2338 2361 2291 2288 2338 2361 2291 2288 Palau 8731 9912 10945 12440 8731 9912 10945 12440 8731 9912 10945 12440 Papua New Guinea 1122 1349 1464 1651 1122 1350 1469 1660 1122 1351 1474 1670 Samoa 2668 2851 3198 3585 2668 2862 3251 3682 2668 2874 3306 3785 Solomon Islands 1125 1292 1543 1821 1125 1295 1556 1848 1125 1299 1570 1877 Tonga 2502 2813 2994 3221 2502 2842 3104 3423 2502 2873 3220 3645 Tuvalu 2654 3019 3397 3987 2654 3032 3461 4127 2654 3045 3531 4288 Vanuatu 2089 2118 2395 2747 2089 2130 2437 2830 2089 2143 2481 2920 GNI per capita ($US 2005) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 3825 4275 5087 6052 3825 4285 5127 6130 3825 4294 5168 6210 Kiribati 1938 1719 1767 1895 1938 1746 1873 2091 1938 1774 1993 2331 Marshall Islands 3616 4161 4610 4778 3616 4161 4610 4778 3616 4161 4610 4778 Micronesia (Fed. States of) 2508 2410 2338 2336 2508 2410 2338 2336 2508 2410 2338 2336 Palau 7993 9215 10175 11565 7993 9215 10175 11565 7993 9215 10175 11565 Papua New Guinea 1078 1162 1262 1422 1078 1163 1265 1430 1078 1164 1269 1438 Samoa 2527 2566 2878 3227 2527 2580 2940 3339 2527 2595 3003 3460 Solomon Islands 1088 1208 1443 1702 1088 1213 1459 1734 1088 1217 1475 1766 Tonga 2623 2785 2964 3189 2623 2836 3146 3524 2623 2888 3340 3894 Tuvalu 3991 4066 4575 5371 3991 4096 4712 5663 3991 4126 4861 6001 Vanuatu 2036 1954 2210 2535 2036 1975 2275 2656 2036 1996 2342 2784 GNDI per capita ($US 2005) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 4062 4650 5548 6581 4062 4663 5611 6703 4062 4677 5675 6829 Kiribati 2055 1838 1887 2049 2055 1892 2090 2437 2055 1947 2321 2913 Marshall Islands 4037 4593 5065 5264 4037 4593 5065 5264 4037 4593 5065 5264 Micronesia (Fed. States of) 2721 2979 3009 3043 2721 2979 3009 3043 2721 2979 3009 3043 Palau 8473 9758 10770 12212 8473 9758 10770 12212 7993 9758 10770 12212 Papua New Guinea 1080 1166 1269 1432 1080 1168 1278 1452 1080 1169 1288 1473 Samoa 3241 3400 3836 4281 3241 3425 3948 4488 3241 3451 4065 4708 Solomon Islands 1115 1224 1458 1725 1115 1234 1489 1789 1115 1243 1521 1855 Tonga 3618 4011 4041 4403 3618 4083 4300 4882 3618 4156 4575 5411 Tuvalu 4372 4580 5213 6274 4372 4626 5433 6758 4372 4673 5671 7318 Vanuatu 2089 1999 2239 2576 2089 2030 2329 2752 2089 2060 2422 2939 GNI+ per capita ($US 2005) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 10606 12080 14192 16493 10606 12145 14510 17111 10606 12210 14828 17729 Kiribati 3193 3475 4133 4936 3193 3832 5808 8107 3193 4188 7483 11278 Marshall Islands 10681 12072 13598 14376 10681 12072 13598 14376 10681 12072 13598 14376 Micronesia (Fed. States of) 14851 15315 15588 16303 14851 15315 15588 16303 14851 15315 15588 16303 Palau 18314 20215 21922 24328 18314 20215 21922 24328 18314 20215 21922 24328 Papua New Guinea 1154 1261 1402 1619 1154 1280 1512 1857 1154 1298 1622 2094 Samoa 17077 19009 21788 24051 17077 19133 22385 25167 17077 19257 22981 26282 Solomon Islands 1187 1384 1738 2146 1187 1440 2037 2765 1187 1495 2335 3384 Tonga 18574 21228 24220 27148 18574 21434 25157 28917 18574 21641 26095 30685 Tuvalu 9782 12688 17180 23191 9782 12886 18233 25426 9782 13084 19286 27661 Vanuatu 2320 2343 2779 3349 2320 2440 3291 4422 2320 2537 3802 5494 69 Remittances of non-seasonal migrants and seasonal worker income (millions $US 2005) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 209 355 486 618 209 362 519 684 209 369 551 751 Kiribati 13 16 21 31 13 20 35 60 13 23 48 89 Marshall Islands 22 23 25 30 22 23 25 30 22 23 25 30 Micronesia (Fed. States of) 22 61 79 88 22 61 79 88 22 61 79 88 Palau 10 12 15 17 10 12 15 17 10 12 15 17 Papua New Guinea 15 35 77 127 15 41 136 274 15 48 195 421 Samoa 143 177 217 262 143 179 229 285 136 182 240 308 Solomon Islands 18 16 22 38 18 20 38 74 15 25 54 110 Tonga 121 164 172 218 121 169 191 257 105 175 211 296 Tuvalu 4 6 8 12 4 6 9 14 4 6 10 16 Vanuatu 31 44 54 78 31 51 78 129 13 58 101 180 Total 610 909 1176 1520 610 946 1353 1914 610 983 1530 2307 Remittances and seasonal worker income as a percentage of GDP BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 6.21% 8.80% 9.51% 9.54% 6.21% 8.99% 10.21% 10.67% 6.21% 9.19% 10.91% 11.81% Kiribati 11.12% 11.32% 12.12% 15.00% 11.12% 13.68% 20.19% 29.03% 11.12% 16.04% 28.39% 43.28% Marshall Islands 13.82% 12.71% 12.08% 12.44% 13.82% 12.71% 12.08% 12.44% 13.82% 12.71% 12.08% 12.44% Micronesia (Fed. States of) 16.97% 24.12% 29.30% 30.91% 16.97% 24.12% 29.30% 30.91% 16.97% 24.12% 29.30% 30.91% Palau 5.49% 5.48% 5.44% 5.20% 5.49% 5.48% 5.44% 5.20% 5.49% 5.48% 5.44% 5.20% Papua New Guinea 0.19% 0.30% 0.52% 0.66% 0.19% 0.37% 0.93% 1.42% 0.19% 0.43% 1.33% 2.18% Samoa 28.20% 31.13% 32.23% 32.04% 28.20% 31.63% 34.10% 35.08% 28.20% 32.13% 35.98% 38.17% Solomon Islands 2.84% 1.94% 1.91% 2.37% 2.84% 2.47% 3.27% 4.65% 2.84% 3.00% 4.64% 6.94% Tonga 45.84% 52.71% 47.54% 51.52% 45.84% 54.23% 52.59% 59.89% 45.84% 55.74% 57.58% 68.06% Tuvalu 16.23% 19.37% 22.13% 26.84% 16.23% 20.26% 25.51% 32.98% 16.23% 21.17% 29.02% 39.58% Vanuatu 5.90% 7.02% 6.39% 6.84% 5.90% 8.11% 9.15% 11.31% 5.90% 9.20% 11.92% 15.78% Total 4.28% 4.90% 4.93% 4.92% 4.28% 5.10% 5.68% 6.22% 4.28% 5.31% 6.44% 7.51% 70 Net seasonal worker income (millions $US 2005) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 5.87 20.03 39.15 8.51 29.55 57.97 11.15 39.08 76.80 Kiribati 0.46 1.12 2.37 4.03 0.46 1.39 3.35 5.96 0.46 1.66 4.33 7.90 Marshall Islands Micronesia (Fed. States of) Palau Papua New Guinea 0.27 1.28 3.42 6.29 0.27 1.71 4.95 9.31 0.27 2.13 6.48 12.33 Samoa 4.32 6.36 9.22 12.82 4.32 7.22 12.34 18.98 4.32 8.09 15.46 25.15 Solomon Islands 1.63 3.37 6.60 10.86 1.63 4.10 9.24 16.08 1.63 4.84 11.88 21.29 Tonga 9.59 17.08 25.10 35.21 9.59 19.45 33.67 52.14 9.59 21.83 42.23 69.07 Tuvalu 0.29 0.43 0.73 1.12 0.29 0.50 1.00 1.66 0.29 0.58 1.27 2.19 Vanuatu 10.69 18.30 26.40 36.54 10.69 20.77 35.29 54.10 10.69 23.23 44.17 71.67 Total 27.25 53.81 93.87 146.02 27.25 63.65 129.39 216.2 27.25 73.51 164.9 286.4 71 Net income of non-seasonal migrants (millions $US 2005) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 5245 6316 7862 9601 5245 6370 8146 10192 5245 6424 8431 10781 Kiribati 131 208 326 472 131 250 556 962 131 292 785 1448 Marshall Islands 337 382 456 552 337 382 456 552 337 382 456 552 Micronesia (Fed. States of) 1185 1300 1476 1668 1185 1300 1476 1668 1185 1300 1476 1668 Palau 156 179 218 257 156 179 218 257 156 179 218 257 Papua New Guinea 530 800 1361 2235 530 953 2440 4940 530 1106 3520 7645 Samoa 2535 3016 3687 4413 2535 3038 3798 4639 2535 3059 3908 4863 Solomon Islands 54 109 215 376 54 142 433 900 54 175 650 1423 Tonga 1544 1884 2389 2957 1544 1900 2477 3141 1544 1916 2565 3322 Tuvalu 52 80 125 183 52 82 135 206 52 83 146 229 Vanuatu 67 108 191 321 67 131 352 725 67 154 514 1128 Total 11838 14381 18304 23034 11838 14726 20488 28182 11838 15070 22668 33317 Population (resident) (thousands) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 880 921 981 1045 880 918 970 1025 880 916 958 1005 Kiribati 109 122 142 159 109 121 133 145 109 119 125 130 Marshall Islands 53 53 56 62 53 53 56 62 53 53 56 62 Micronesia (Fed. States of) 104 108 118 125 104 108 118 125 104 108 118 125 Palau 21 22 25 27 21 22 25 27 21 22 25 27 Papua New Guinea 7309 8413 10057 11699 7309 8406 10018 11617 7309 8400 9980 11536 Samoa 190 199 210 228 190 198 206 220 190 197 202 213 Solomon Islands 561 640 757 878 561 639 749 863 561 637 741 847 Tonga 105 111 121 132 105 110 117 126 105 109 114 119 Tuvalu 10 10 11 11 10 10 10 10 10 10 10 10 Vanuatu 253 294 354 416 253 293 348 403 253 292 342 391 Total 9595 10894 12831 14782 9595 10878 12750 14623 9595 10863 12670 14464 72 Non-seasonal migrants (stock) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 189571 199026 205876 212071 189571 200758 213507 225443 189571 202490 221139 238815 Kiribati 4324 5952 7720 9412 4324 7157 13192 19221 4324 8362 18664 29030 Marshall Islands 11841 11668 11511 11597 11841 11668 11511 11597 11841 11668 11511 11597 Micronesia (Fed. States of) 40642 38521 35922 33921 40642 38521 35922 33921 40642 38521 35922 33921 Palau 6855 6837 6707 6590 6855 6837 6707 6590 6855 6837 6707 6590 Papua New Guinea 17464 23018 32453 44866 17464 27418 58210 99195 17464 31818 83967 153523 Samoa 87949 90678 91718 92165 87949 91355 94606 97081 87949 92032 97494 101998 Solomon Islands 1768 3123 5142 7580 1768 4074 10341 18141 1768 5026 15540 28702 Tonga 53247 56564 59125 61292 53247 57105 61482 65378 53247 57646 63839 69465 Tuvalu 1816 2412 3117 3852 1816 2468 3387 4355 1816 2525 3658 4858 Vanuatu 2280 3165 4649 6581 2280 3843 8606 14913 2280 4522 12564 23246 Total 417758 440964 463940 489927 417758 451205 517473 595835 417758 461447 571007 701744 Non-seasonal migrants (annual flow) BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 3835 4000 4000 4000 3835 4500 5000 5000 3835 5000 6000 6000 Kiribati 298 351 366 380 298 702 1097 1140 298 1054 1828 1899 Marshall Islands 103 124 159 194 103 124 159 194 103 124 159 194 Micronesia (Fed. States of) 84 130 184 238 84 130 184 238 84 130 184 238 Palau 81 81 81 81 81 81 81 81 81 81 81 81 Papua New Guinea 729 1377 1977 2576 729 2753 5930 7729 729 4130 9883 12882 Samoa 1731 1554 1478 1402 1731 1749 1848 1752 1731 1943 2217 2103 Solomon Islands 197 292 386 481 197 583 1159 1443 197 875 1931 2404 Tonga 1282 1248 1225 1202 1282 1404 1531 1502 1282 1560 1837 1803 Tuvalu 95 134 149 164 95 151 186 205 95 167 224 246 Vanuatu 131 212 303 394 131 424 910 1183 131 636 1516 1972 Total 8566 9502 10307 11112 8566 12601 18083 20467 8566 15699 25859 29822 73 Seasonal worker numbers BAU Medium growth High growth 2013 2020 2030 2040 2013 2020 2030 2040 2013 2020 2030 2040 Fiji 1679 4770 7860 2435 7037 11639 3191 9305 15418 Kiribati 150 319 564 808 150 397 797 1197 150 475 1030 1586 Marshall Islands Micronesia (Fed. States of) Palau Papua New Guinea 88 368 815 1262 88 489 1179 1869 88 610 1543 2476 Samoa 1416 1818 2196 2574 1416 2066 2938 3811 1416 2313 3681 5048 Solomon Islands 536 964 1572 2179 536 1174 2201 3227 536 1384 2829 4275 Tonga 3147 4886 5977 7069 3147 5565 8017 10468 3147 6245 10056 13867 Tuvalu 96 123 174 225 96 144 239 333 96 166 303 441 Vanuatu 3506 5236 6286 7335 3506 5942 8402 10862 3506 6647 10518 14389 Total 8940 15393 22353 29313 8940 18212 30809 43406 8940 21031 39265 57500 74 GNI+ per capita Fiji GNI+ per capita Kiribati GNI+ per capita 20000 12000 18000 16000 10000 14000 8000 12000 10000 6000 8000 6000 4000 4000 2000 2000 0 0 BAU High Growth BAU High Growth PNG GNI+ per capita Samoa GNI+ per capita 2500 30000 2000 25000 20000 1500 15000 1000 10000 500 5000 0 0 BAU High Growth BAU High Growth Solomon Islands GNI+ per capita Tonga GNI+ per capita 4000 35000 3500 30000 3000 25000 2500 20000 2000 15000 1500 1000 10000 500 5000 0 0 BAU High Growth BAU High Growth Tuvalu GNI+ per capita Vanuatu GNI+ per capita 30000 6000 25000 5000 20000 4000 15000 3000 10000 2000 5000 1000 0 0 BAU High Growth BAU High Growth 76 77 worldbank.org/PacificPossible