Recognizing that internal migration is an important source of error in population projections, we evaluate 21 forecasts from five classes of methods: (1) average of the most recent flows; (2) time-series econometric models with and without external factors; (3) machine-learning-based gradient boosting models; (4) multiplicative component and gravity-type models; and (5) average of all forecasts. We forecast bilateral interstate migration flows in Australia for the five years to mid-2016 and to mid-2023. No single model consistently outperforms the others in terms of bias, accuracy, and empirical coverage. Simple models perform as well or better than complex models. ARIMA models and their Bayesian equivalent perform well in both periods. Including control variables is no panacea, because their relationship with migration is flow and period specific. Similarly, globally trained machine learning models are not clear-cut alternatives, particularly when flow sizes vary widely. In contrast, the multiplicative component model suits contexts where the spatial structure of migration is stable, suggesting that internal migration forecasting requires context-specific approaches.

According to the United Nations' World Population Prospects 2024, global population is projected to peak at about 10.3 billion in the mid-2080s. However, rapidly declining fertility suggests an earlier peak, possibly one billion lower. Natural decrease is becoming widespread in high - and middle-income countries, driven by sustained sub-replacement fertility and population ageing. Economic consequences include slower productivity growth, labour market adjustments, shifting consumption, fiscal pressures, and intergenerational tensions. Pro-natalist policies show limited impact, while migration offers only partial and politically constrained relief. Technological adaptation - especially automation, robotics, and artificial intelligence - is likely to expand beyond early adopters such as East Asia. Urbanisation remains dominant, though post-COVID counter-urbanisation and digital nomadism generate localised change. Aotearoa New Zealand illustrates these dynamics, shaped by its distinctive geography, population mobility, and high ethnic diversity. Economic perspectives highlight the risks and costs for policy and planning of ignoring predictable demographic change.

New Zealand has one of the highest immigration rates in the developed world, resulting in a high share of foreign-born residents. Its population is highly urbanized, spatially uneven, ethnically diverse, and includes a significant indigenous Māori population. This paper synthesizes two decades of research on migrant social capital in New Zealand, drawing on data from multiple waves of the New Zealand General Social Survey and the Adult Literacy and Life Skills Survey. These datasets provide insights into community participation, volunteering, perceptions of safety and inclusion, and electoral engagement. We draw two main conclusions from our synthesis. First, although migrants arrive with limited local social capital, they gradually build bonding, bridging, and linking, social capital over time. In addition, reported experiences of discrimination decline with longer residence. However, pooled survey data reveal variation in these patterns across time. Second, social capital investment is shaped by the spatial distribution of ethnic groups. Migrants are more likely to engage in bridging social capital in regions where their group is underrepresented, and in bonding social capital in communities where ethnic clustering occurs.

In this paper we focus on modelling and forecasting gross interregional migration in a way that can be embedded within multiregional population projections. We revisit, and apply, a family of spatial interaction models first formulated during the 1970s. The classic gravity model—in which migration is positively related to the populations of sending and receiving areas, but inversely related to various types of spatial friction associated with migrating between them—is a special case that is nested within this family of models. We investigate which member of the family of models gives the best fit when modelling five-year migration flows between the 66 Territorial Authorities (TAs) of Aotearoa New Zealand, using 2013 and 2018 census data. We find that predicting migration between two TAs can be improved by taking into account, firstly, an index of the ‘draw’ from all other TAs when modelling out-migration of any TA and, secondly, an index of the ‘competitiveness’ of a TA vis-à-vis all other TAs when modelling in-migration of any TA. We highlight the properties of the statistically-preferred model by simulating the impact on internal migration of an exogenous increase in Auckland’s population. In this model, such a population change affects not only migration flows from and to Auckland, but also migration between other TAs. The usefulness of this approach for population projections is assessed by forecasting the 2013–18 migration matrix by means of 2013 census data only. In this specific case, the model outperforms the classic gravity model in terms of forecasting gross migration, but not net migration.

In this paper, we present evidence from quantitative research over the last decade on how the social capital of individuals in Aotearoa New Zealand is associated with birthplace and, for migrants, years since migration. We also consider the effects of spatial sorting and ethnic diversity on social capital formation. Aotearoa New Zealand has one of the highest rates of immigration in the OECD and, consequently, one of the highest shares of foreign-born individuals in the population. Additionally, the population is characterized by high ethnic diversity and a large indigenous population, with Māori representing 17 percent of the population. Using several data sources, we measure social capital by focusing on participation and volunteering in a range of community activities, perceptions of safety and inclusion, and voting in elections. Regression modelling shows that, as expected, migrants have little local social capital upon arrival. However, differences between their social capital and that of native-born individuals reduce considerably as the duration of residence in Aotearoa New Zealand increases. When the migrant share in a region is larger than the national average, migrants invest less in bridging social capital. Migrant clustering within a region increases their investment in bonding social capital. Bridging activities are associated with better employment outcomes. Less than one in five respondents in the utilized survey data report discrimination, and for migrants, discrimination declines with years of residence. However, the trend in discrimination has been upward over time and particularly affects non-European migrants and persons identifying with Māori and Pacific Peoples ethnicities. Residential location matters. Greater ethnic diversity is associated with the perception of a less safe neighbourhood, but individuals in ethnically diverse regions experience relatively less discrimination. Additionally, there is more involvement in elections in such regions. In contrast, greater ethnic polarisation in regions is associated with less civic engagement and more discrimination.

One of the main challenges facing non-metropolitan regions is the attraction and retention of highly-educated young people. A loss of the brightest can lead to reduced business creation, innovation, growth and community well-being in such regions. We use rich longitudinal microdata from New Zealand to analyse the determinants and geography of the choice of destination of recent university and polytechnic graduates 2 years and 4 years after graduation. Rather than considering a range of location-specific consumption and production amenities, we assume spatial equilibrium and calculate, following Chen and Rosenthal (J Urban Econ 64:519–537, 2008), ‘quality of life’ and ‘quality of business’ indicators for urban areas that encompass all amenities that are utility and/or productivity enhancing (or reducing, in the case of disamenities). Specifically, we test whether students locate in places that are regarded as good to live or good to do business; and how this differs by field of study. Our estimates are conditional on students’ prior school (home) location and the location of their higher education institution. We find that graduates are attracted to locate in urban places that have high quality production amenities. High quality consumption amenities have heterogeneous effects on the location choice of students. Creative arts and commerce graduates are relatively more likely to locate in places that are attractive to business, consistent with a symbiosis between bohemians and business. Decision makers can leverage their existing local strengths, in terms of production and/or consumption amenities, to act as drawcards for, or to retain, recent graduates in specific fields.

In this paper we introduce a measure of cultural diversity that takes ‘social difference’ between country of birth and ethnic groups into account. We measure social difference using exploratory factor analysis of subjective identity, attitude and value responses in Aotearoa New Zealand’s 2016 General Social Survey. We examine the level of, and change in, our social difference-based measure of cultural diversity in 31 urban areas between 1976 and 2018, using census data. We compare these patterns with those derived from a standard fractionalisation measure of diversity based on population composition by country of birth and ethnicity. We find that the two diversity measures are highly correlated across the urban areas. Diversity increased everywhere between 1976 and 2018, whether social difference is taken into account or not. However, the social difference-based measure increased much faster than the standard measure in all but one of the urban areas. This suggests that growth in the fractionalisation measure of diversity is likely to have underestimated the trend in experienced social difference. Both measures also show evidence of spatial convergence in diversity: urban areas with low diversity in 1976 – which tended to be in the South Island – exhibited faster increases. Population diversity increased strikingly in Queenstown, which was the 19th most diverse urban area in 1976, in terms of social difference, but second only to Auckland in 2018.

Aotearoa New Zealand has been identified, by several measures, as being one of a few developed countries that have weathered the COVID-19 pandemic in the best possible way. This outcome is generally attributed to strict but effective public health measures that included - besides very high vaccination rates - national and regional lockdowns, as well as total closure of the border except for returning citizens (who were subject to mandatory quarantining). Concurrent fiscal and monetary policies contributed to economic outcomes that remained remarkably buoyant. In this paper we assess the importance of public interventions in New Zealand triggered by the pandemic relative to the mitigating effects of the country being an island nation with a small population scale, low population density and remote location. We summarize the recent international literature, estimate simple but representative cross-country regression models, and provide a qualitative evaluation of the public policy response. We find that the favourable effects of low average population density, remoteness and the absence of land borders have indeed been of great benefit. Geography assisted in the effectiveness of the elimination strategy which was only abandoned in favour of a mitigation strategy once the less severe but highly contagious Omicron variants arrived in early 2022. Hence, while a remote and peripheral location is generally seen as economically disadvantageous, during a pandemic it delays the spread of a viral disease and provides the opportunity to focus on interventions to maintain economic activity, develop effective public health responses and learn from the experience of less remote nations.

This paper focuses on the spatial variation in the uptake of social security benefits following a large and detrimental exogenous shock. Specifically, we focus on the Global Financial Crisis (GFC) and the onset of the COVID-19 pandemic. We construct a two-period panel of 66 Territorial Authorities (TAs) of New Zealand (NZ) observed in 2008-09 and 2020-21. We find that, despite the totally different nature of the two shocks, the initial increase in benefit uptake due to the COVID-19 pandemic was of a similar magnitude as that of the GFC, and the spatial pattern was also quite similar. We link the social security data with 146 indicator variables across 15 domains that were obtained from population censuses that were held two years before each of the two periods. To identify urban characteristics that point to economic resilience, we formulate spatial panel regression models. Additionally, we use machine learning techniques. We find that the most resilient TAs had two years previously: (1) a low unemployment rate; and (2) a large public sector. Additionally, but with less predictive power, we find that TAs had a smaller increase in social security uptake after the shock when they had previously: (3) a high employment rate (or high female labour force participation rate); (4) a smaller proportion of the population stating ethnicities other than NZ European; (5) a smaller proportion of the population living in more deprived area units. We also find that interregional spillovers matter and that resilient regions cluster.

This paper examines the impact of homogamy on the distribution of household income in New Zealand at the national level and across different sized cities. We focus on homogamy by age, education, hours worked, employment status, and migration status. We present a new index of homogamy that takes account of maximum potential homogamy. Our index is less sensitive to categories with small population shares than the commonly used concentration ratios. We compare the inequality impact of actual matching with that of randomized matching by means of the additional randomization method. Contrary to public perception, homogamy of the highly educated has declined relative to random matching. Nonetheless, homogamy has had an inequality-increasing impact on the distribution of income and this effect has grown over time: from around 5% of the mean log deviation (MLD) measure of income inequality in 1986 to 16% in 2013. Allowing for simulated labor supply responses reduces this effect by less than 1%. Spatially, the effect of homogamy is larger and increases more in metropolitan areas than in other urban areas. In Auckland, New Zealand's largest city (around a third of the population), homogamy accounted for a fifth of MLD inequality by 2013, up from 6% in 1986. Educational homogamy plays the biggest role in this process, but the effects of hours worked, employment status and migration status are relatively important too. Homogamy by age has little effect on income inequality. Using the Gini index as an alternative inequality measure yields similar results.

We revisit in this chapter a common issue with popular indices used for measuring residential sorting, i.e. the extent to which a subgroup of the population is spatially distributed (sorted or segregated) differently from the remainder of the population. Specifically, we show that three common measures of residential sorting (viz. the Index of Segregation, the Index of Isolation and the Entropy Index of Segregation) are affected by group size, i.e. the expected values of the indices are positive rather than zero under random sorting, and the size of this positive bias is related to group size. This is an important issue because it is common to compare sorting indices across groups of rather different sizes, both cross-sectionally and over time. Using New Zealand data, we demonstrate group-size impact on bias in measures of residential sorting by means of scatter plots and regression in four different ways: (1) investigating the relationship between group size and each residential sorting measure calculated with actual data; (2) randomly allocating individuals across area units, calculating the resulting residential sorting measures, and again investigating the relationship between index values and group size; (3) showing that normalised/systematic indices of sorting are also related to group size; and (4) calculating the bias for each sorting index and investigating the relationship with group size. Our empirical illustration uses microdata on the self-reported ethnicity of individuals (with multiple responses possible) from five New Zealand Censuses of Population and Dwellings (1991–2013) for the Auckland region, selected due to its high ethnic diversity. Our results demonstrate that the Entropy Index of Systematic Segregation measure of residential sorting is the measure that is least affected by group size variation. As a result, we strongly recommend using this index of sorting as a preferred measure.

This study examines whether the perception of difference in institutional quality between OECD destination countries and Vietnam, and the stated importance attached to such difference, influences Vietnamese migrants’ intention to return home. We use data from a web-based survey (N = 159) that we conducted in 2016. The countries covered capture about 90% of the Vietnamese diaspora in the world. We find, by means of weighted logistic regression analysis with a range of measures of institutional quality, that migrants who perceive a larger institutional quality difference are less likely to have the intention to return. However, there is considerable heterogeneity by gender. Women are, if they attach importance to institutional quality, particularly concerned about control of corruption, while the between-country difference in government effectiveness and regulatory quality matters to men. Concerns about a lack of voice & accountability; and about political instability & the presence of violence/terrorism deter return migration of both genders.

In this paper we describe the development, calibration and validation of a dynamic spatial microsimulation model for projecting small area (area unit) ethnic populations in Auckland, New Zealand. The key elements of the microsimulation model are a module that projects spatial mobility (migration) within Auckland and between Auckland and the rest of the world, and a module that projects ethnic mobility. The model is developed and calibrated using 1996-2001 New Zealand Linked Census (i.e. longitudinal) data, and then projected forward to 2006. We then compare the results with the actual 2006 population. We find that in terms of indexes of overall residential sorting and ethnic diversity, our projected values are very close to the actual values. At a more disaggregated spatial scale, the model performs well in terms of the simulated normalised entropy measure of ethnic diversity for area units, but performs less well in terms of projecting residential sorting for each individual ethnic group.

This paper conducts a robustness analysis of the impact of diasporas on institutional quality in home countries. Using the Database on Immigrants in OECD Countries (DIOC), we attest that the home country institutional development role of diasporas found in the literature is robust to modifications in terms of the dataset, the measurement of diasporas and the instrumental variable procedures used in our cross-sectional and panel analyses. The novelty of this paper is that we take the heterogeneity of diasporas into account in terms of their distribution across host countries, their duration of stay, and the level of development of their home countries. As in earlier literature, we find robust evidence that diasporas enhance institutional quality in home countries. We also find that the intensity of the diffusion of advanced institutions from developed host countries to home countries through the international migration channel is weaker with diasporas characterized by shorter average duration of stay and with diasporas from developing home countries. However, this diffusion effect is not significantly related to the distribution of diasporas across OECD host countries.

A wide range of push and pull factors have been shown to influence the decisions to migrate to another country, but the effect of the quality of institutions on the origin and destination countries has received less attention than it deserves. This chapter reviews the theoretical framework and the extant empirical evidence for the importance of political and economic institutions to international migration decisions. Key aspects of this body of literature are illustrated further with an analysis of the return migration decisions of Vietnamese migrants.

This chapter provides an introduction to the discussion of demographic changes, issues and models in the Asia-Pacific region that are the focus of this book. The Asia-Pacific region represents 4.9 billion people or roughly 63% of the world’s population. This region is hugely diverse: the countries and subregions vary demographically, geographically, economically, culturally and institutionally. We examine salient features of population change over the last three decades and draw conclusions on the underlying factors which influence population dynamics. We also consider population projections until the middle of the twenty-first century and briefly review some salient impacts of demographic change. Population growth is declining and populations are ageing, both numerically and structurally, everywhere in the region. Fertility rates are converging to replacement levels in many countries or remain below replacement. Large differences in life expectancies, urbanisation and international migration remain. Given the large spatial variations, we argue for a greater emphasis on subnational and multiregional population analysis and projections.

One of the main challenges facing non-metropolitan regions is the attraction and retention of highly-educated young people. A loss of the brightest can lead to reduced business creation, innovation, growth and community wellbeing in such regions. We use rich longitudinal microdata from New Zealand’s integrated administrative data infrastructure to analyse the determinants and geography of the choice of destination of tertiary educated (university and polytechnic) graduates. We address the question of post-student location choice in the context of the approach of Chen and Rosenthal (2008) who introduced a methodology for calculating ‘quality of life’ and ‘quality of business’ indicators for urban areas reflecting consumption and productive amenities respectively. Specifically, we test whether students – of different characteristics (e.g. institutional type and field of study) – locate in places that are regarded as good to live or good to do business. Our estimates are conditional on students’ prior school (home) location and the location of their higher education institution. We find that graduates are attracted to locate in places that have high quality production amenities. High quality consumption amenities have heterogeneous effects on the location choice of students. Creative Arts and Commerce graduates are relatively more likely to locate in places that are attractive to business, consistent with a symbiosis between bohemians and business. Places can leverage their existing (productive or consumption amenity) strengths to act as drawcards to recent graduates. The results are important for local decision-makers who wish to know which factors can attract and retain young qualified people.

Auckland, New Zealand, is among the most ethnically diverse cities in the world. Like most large cities, its population is also quite youthful. In this paper, we focus on the dynamics of self-declared ethnic identities of adolescents in Auckland, by using New Zealand Linked Census data for four inter-censal periods between 1991 and 2013. Our dataset links the same young person across two consecutive Censuses (that is, those aged 13-17 in one Census are aged 18-22 in the following Census five years later). We aim to capture the first conscious ethnic identity affiliation of adolescents, assuming that their ethnic identities are initially recorded by their parents, but subsequently determined by the adolescent themselves when they transition to adulthood. We classify our predictor variables into individual, family and neighbourhood-level variables. We find that an adolescent’s ethnicity stated at the previous census, parents’ ethnicity, and the ethnic makeup of the neighbourhood are all major determinants of ethnic-identity choices among adolescents in Auckland.