Future labour force in ageing European societies: the role of human capital and migration Michaela Potančoková, Ph.D. Senior Research Scholar, Multidimensional Demographic Modelling group, Population and Just Societies program @ IIASA External lecturer in Global demography program, University of Vienna Text Description automatically generated with medium confidence Objective •Demographic shifts in Europe will affect future population sizes and labor supply, fear of negative consequences of population decline •Population composition changes -> educational expansion, longevity, working lives •Immigration often presented as a tool to (partly) mitigate the expected future labour force decline •How demographic multidimensional population projection methods are being used to explore future changes in populations, labour force (labour supply) and dependency • • What is population ageing and why is it considered a challenge? • Population ageing 4 Demographic transition shift in distribution of a country's population towards older ages Demographic transition & population ageing •Declining birth rates and longevity boost population in productive ages, a window of opportunity to harness benefits of expanding working age population and declining dependency-> demographic dividend (stage 3) •Stage 4: Increase in the elderly due to ageing from the top (increased longevity due to declines in mortality) and from the bottom of the pyramid (low, below-replacement level birth rates) • • Population ageing in age pyramid •Source: Statistics Sweden Demographic transition & population ageing •Declining population and working-age population in European countries and other industrialised countries •Increasing share of elderly and dependency •posing challenges to the social security systems and the welfare state •Posing challenges to public finances through increased health care expenditures •deemed detrimental to prosperity, economic growth, productivity •In common narrative increasing age dependency is something to be mitigated • • • https://economictranscript.wordpress.com/2017/06/22/ageing-population-of-japan/ Demographic shifts in industrialised countries •Longevity -> longer lives (more survival to higher age) •Low birth rates •Population ageing – increase in number and share of older population •-> Smaller populations and labour force • •Increasing human capital – educational expansion, health •-> Qualitatively different populations in the past and future •-> implications for labour force size, population structures (beyond age), labour force participation rates • • • Demographic Metabolism: Change of populations structures through generational replacement cohort turnover -> older generations replaced by better educated younger generations • 10, date Source: Lutz, et al (2018), http://dataexplorer.wittgensteincentre.org/wcde-v3/ Commonly used measures of population ageing •Share of population at age 65+ •Age dependency ratio (ADR) •ADRt = Popt 0-14 + Popt 65+ / Pop 15-64 •OADR = Popt 65+ / Pop 15-64 •YADR = Popt 0-14 + Popt 65+ / Pop 15-64 • •Different age limits can be used as cut offs – age 19, age 60 •Working age population (age 15-64) is not actual labour force, but potential labour force if everyone at that age was active at the labour market -> includes workers and non-workers • •Deficiencies / limitations of ADR: •does not capture qualitative changes in working age population •assumes that everyone at working age participates equally and is equally productive and everyone age 65+ assumed to be inactive and unproductive • • • Alternative dependency ratios •Prospective Age dependency ratio (Scherbov and Sanderson) •POADRt = Popt Popt with less than 15 remaining years of life / Pop 15-64 •Captures increases in longevity – 65 year olds 50 years ago different to those now and to people in the same age 50 years from now •Improved cognitive skills, health -> potential for longer working lives • •Labour force dependency ratio •LFDR = inactive (non-workers)/ active at the labour market (workers) •Takes into account that not everyone in working age equally participates in the LF and that not everyone age 65+ is inactive • •Productivity-weighted dependency ratio •Adds a weighting factor that approximates differences in productivity by through education-specific wage differentials •Takes into account that not everyone is equally productive • • • • • • Do you think all EU countries will see rapidly shrinking labour force? • Projected change in working age population (20-64) 14 Potancokova et al. (2021): short explanation in European demographic data sheet 2022 @ www.populationeurope.org Source: Potancokova et al (2021) Multidimensional population projections •Population projection widely used to inform policy options -> need to go beyond standard population projections by age and sex and to capture population heterogeneity • A useful tool to simulate future scenarios for better informed evidence-based policies to address demographic challenges and assess their consequences of population ageing in broader perspective •WiC Global human capital projections: Education stands out as differentiating factor of demographic behaviours • http://dataexplorer.wittgensteincentre.org/wcde-v3/ • •Extensions to demographic modeling of labour supply •Numerous scenarios and simulations show that reversing or even slowing down population ageing is not a viable policy option •Extension to migration scenarios (CEPAM, QuantMig) and ongoing work on modelling skills (Liknk4Skills) • • Demographic modelling of labour supply •Option 1: Multistate cohort-component population projection by age, sex, educational attainment and labour force participation •Option 2: Dynamic microsimulation population projection model with multiple status variables and stochastically modelled transition rates • •What we take into account in modelling? •Mortality differentials by education -> higher edu associated with longevity (and better heath) •Fertility differentials by education -> higher edu associated with higher childlessness and lower fertility (for women) •Emigration rates by age and immigration volumes as during 2010-2016 •Fixed educational composition of immigrants •Educational expansion modelled using cohort trends in past educational attainment •Differences in labour force participation by age, gender and education • What do we know about gaps in labour force participation? •Depends on age -> lower before age 25 and after age 60 •Depends on gender -> different age patterns due to parental leave •Depends on educational attainment -> lower for those with lowest qualifications and highest for highly-educated •Educational attainment is closely associated with LFP and employment • Labour force participation of immigrants is lower than of EU-born Inequality between men and women among native-born and immigrant women LFPR by age, gender and education, men and women Slovakia 2014-2019, LFS Differences in labour force activity across EU countries 19, date Source: Marois, Belanger and Lutz (2020)Population aging, migration, and productivity in Europe. PNAS 117(14) https://doi.org/10.1073/pnas.1918988117 Highest LFPRs Lowest LFPRs Differences in labour force activity across EU countries 20, date Source: Marois, Belanger and Lutz (2020)Population aging, migration, and productivity in Europe. PNAS 117(14) https://doi.org/10.1073/pnas.1918988117 Highest LFPRs Lowest LFPRs When we consider educational expansion, higher female labour force participation and longer working lives what are the implications for future labour force sizes? •Constant scenario: fixed LFPR by age, sex, education •Equalisation scenario: women reach the same LFPR as men at equivalent age and education •Swedish scenario: by 2040 men and women in all EU countries will reach labour force participation as in Sweden When we consider educational expansion, higher female labour force participation and longer working lives can stabilize labour force size in the EU28 •Constant scenario: fixed LFPR by age, sex, education •Equalisation scenario: women reach the same LFPR as men at equivalent age and education •Swedish scenario: by 2040 men and women in all EU countries will reach labour force participation as in Sweden Projected labour force dependency ratio in EU28 (inactive / workers) When we consider that productivity is associated with educational attainment, future dependency ratios look less daunting for EU-28 •Age dependency ratio (ADR) • • •Labour force dependency ratio (LFDR) • •Productivity-weighted dependency ratio: •estimated productivity weights for Active population are set at 1 for medium education, 1.66 for high education, and 0.62 for low education Source: Marois, Belanger and Lutz (2020)Population aging, migration, and productivity in Europe. PNAS 117(14) https://doi.org/10.1073/pnas.1918988117 What other feasible ways to address the labour force decline can you think of? Policy debate •Increase labour force participation of women •Longer working lives: higher retirement age, active aging •Increase birth rates to stabilize and prevent future decline in the labour force •Activate locally available human capital – activation, reskilling, upskilling •Immigration •Replacement migration – how many immigrants would be needed to stabilize population, working age population, dependency ratios… •Skilled labour – recognition of foreign degrees, matching demand and supply •Access to labour market for different types of migrants beased on residence permits •Curb emigration, incentivize return migration 28 Impact of Migration and Fertility on Population Growth in the UK European Demographic Data Sheet 2020 Scenarios of migration impact on future labour force •Immigration often presented as a tool to improve the expected labour force decline •But shift towards rights-based migration policy implies a need to remove barriers for better integration of (some groups of) immigrants into labour market •What would be the long-term impacts of improved or worsened economic integration of immigrants on labour supply and employment? • • • Source: EPSC (2018) Successful economic integration of immigrants is of high importance as it fosters also their social, linguistic and cultural inclusion and improves the quality of life of migrants, their families and empowers communities. Immigrants' better economic integration is highly needed also against the backdrop of the unavoidable future smaller labour force sizes in the EU overall and in practically all Member States. we focus on supply of labour driven by structural changes in potential labour force and we do not aim to simulate economic context, labour demand or changes in employment due to economic cycles. Also, there are no feedback effects included and labour force activity and employment depend purely on individual characteristics. What do we know about gaps in labour force participation of immigrants? •Labour force participation differs between EU-born and (different groups of) immigrants • Labour force participation of immigrants is lower than of EU-born Inequality between men and women among native-born and immigrant women Labour force participation rates at age 35-39, EU28 Data: EU-LFS 2010-2016 MEN WOMEN What do we know about gaps in labour force participation and employment of immigrants? I will illustrate this with some descriptive results from EU-LFS analysis Labour force participation rates at age 35-39, EU28 Data: EU-LFS 2010-2016 MEN WOMEN Educational attainment is closely associated with LFP and employment Labour force participation rates at age 35-39, EU28 Data: EU-LFS 2010-2016 MEN WOMEN For men, labour force participation rate reaches similar levels to natives after 10 years of stay, while it stays way lower for women. For women, the gap compared to female EU-born is wider and is never completely closed. Among low educated recent immigrants, the participation rates is 23 percentage points lower compared to EU-born women (40% vs. 63%). After 10 years, their participation rates, though improving, are still below the rates of EU-born (59%). For highly educated women born out of the EU the pattern is similar pattern but the gap to the EU-born women is wider after 10 years (81% vs 90%) compared to low-educated women (59% for non-EU-born compared to 63% for EU-born). The effect of education is smaller for immigrants, especially for women. These lower returns on education could be partly explained by lower quality degrees in source countries, as well as by cultural differences in the definition of gender roles (Inglehart and Norris 2003; Antecol 2000). Reasons for immigration may play a role and one could expect higher labour force participation rates among immigrants coming as students or on economic grounds. In contrast, one can expect lower labour force participation rates, at least in the first years after immigration, among female immigrants arriving on family reunification grounds because in their case migration and family formation are directly interlinked. Studies on fertility outcomes of immigrants show elevated fertility rates in the first year after migration particularly for those arriving for family reunification (Rosero-Bixby et al. 2011; Mussino and Strozza 2012). A decision to have a child shortly after the arrival into the host country can also be linked to poor employment prospects in the destination country (Kulu and Milewski 2007). What do we know about gaps in labour force participation of immigrants? •Labour force participation differs between EU-born and (different groups of) immigrants •LFP associated with educational attainment for both natives and immigrants •For immigrants increases with duration of stay -> catching up native levels •Immigrants who arrived as children similar outcomes as native-born •Gaps to natives more pronounced for women •Educational attainment matters but the gap never closes for immigrant women even, even after controlling for education •Women face higher unemployment in addition to lower activity rates • • Labour force participation of immigrants is lower than of EU-born Inequality between men and women among native-born and immigrant women Scenarios of economic integration of immigrants from outside the EU28 •Scenarios to help assess the long-term impact of better or worsened economic integration of immigrants •Not predictions or plausible scenarios but stylized situations (what-if scenarios) representing a range of policy outcomes •Assumptions on LFPR at fixed demographic and educational trends, and at fixed volume and composition of immigrantion flows • • What we take into account in modelling? •Fertility differentials by education, region of birth, migration status and duration of stay in the country •Differences in labour force participation by age, gender, education, region of birth and duration of stay •Differences in migration rates between the native-born, born in the EU+ and born outside the EU+ •Immigrants pulled into destinations along established migration corridors CEPAM scenarios to examine the impact of immigration on future labour force in the EU Source: Lutz et al (2019) 40 Scenarios to examine the impact of immigration on future labour force in the EU Source: Lutz et al (2019) 41 What-if scenarios of improved or failed integration: Component 1-Volume of immigration into EU+UK 2-Educational composition of immigrants 3-Integration of immigrants in the labor force 4-Labor force participation trends for EU-born i. Baseline 10M/5 years Same as recent immigrants Average of 2010-2016 Constant entry and exit rates ii. Baseline/Swedish_LFP 10M/5 years Same as recent immigrants Average of 2010-2016 Rates reach those of Sweden iii. Baseline/high integration 10M/5 years Same as recent immigrants Rates reach those of native-born by 2040 Constant entry and exit rates iv. Baseline/low integration 10M/5 years Same as recent immigrants Rates reach those of immigrants in Denmark by 2040 Constant entry and exit rates v. Canadian 20M/5 years Same as recent immigrants in Canada (53% highly educated) Average of 2010-2015 Constant entry and exit rates •Projected labour force size in EU28 •Projected labour force dependency ratio in EU28 Projected labour force by integration scenarios (constant educational attainment of immigrants as in recent past) 44 Relative change in projected total labour force, Austria Projected labour force dependency ratio, Austria (inactive/active) Labour force by ILO(LFS) definition - includes job seekers Source: CEPAM projections 45 Projection of the productivity-weighted labor-force dependency ratio (inactive / active) for the EU-28 under different scenarios, 2015–2060 Source: Marois, Bélanger and Lutz (2020) PNAS. Labour force dependency ratio in the EU28 – improved economic integration is paramount under high migration flows •Negative impacts of population ageing on labour supply can be mitigated by changes in migration (education-selective & high integration) AND increased labor force participation Age pyramids disaggregated by labor force participation and education for the EU-28 in 2015 and 2060 under different scenarios Source: Marois, Bélanger and Lutz (2020) PNAS. High & selective migration + high LFP ADR=1.0 PWLFDR=1.0 ADR=1.6 PWLFDR=1.1 ADR=1.5 PWLFDR=0.8 •demography is not destiny and policies can make a decisive difference How would high immigration events impact future demographic trends and labour force? §Volatility and unpredictability of migration §Past political events linked to large migration events to Europe §Uncertainty about future migration quickly increases already in short time horizons (Baker and Bijak 2021) §Much of uncertainty about future migration is irreducible and driver environments are complex (Bijak and Czaika 2020) §How to reflect the uncertainty against the backdrop of the past unexpected large immigration, such as from Syria and Ukraine, in scenarios? §What would be long-term demographic impacts of High migration events emerging from different origins ? Data source: Statistik Austria Framework for quantifying rare migration events 48 §We cannot predict but we can simulate what-if scenario §To predict we would need to know/assume: onset, magnitude, duration and origin § §Inspiration from contingency planning -> Bijak (2023) pioneers use of selected quantiles from the upper tails of the heavy-tailed probability distributions that approximate volumes of migration for selected frequency of occurrence of a migration event §The method is based on statistical approaches to modelling rare events, including the extreme value theory (Coles 2001) §Rare / extreme events have certain frequency of occurrence and magnitudes § §In this presentation: migration corresponding to twice-in-a-century event (0.98 quantiles from Pareto distribution) fitted to harmonised origin-destination-specific migration flows estimates between EU+ countries and 8 world regions (by Aristotelous et al. 2022) § § § § § High migration events scenarios 49 Once-in-a-decade (Migration event) Twice-in-a-century (High migration event) short With persistence from: Other Europe North Africa Sub-Saharan Africa West Asia South & South-East Asia East Asia Latin America short } With persistence What-if scenarios simulated one by one as either short or persistent immigration event from each world region Short event: twice-in a century flows in one single year in 2025-2029 period Persistent event: increased immigration from High migration event region follows for 10 years, with gradually declining intensity due to family reunifications, migration networks and other reasons (after 10 yeas same flows from that region as in the baseline scenario) High migration events scenarios 50 Twice-in-a-century (High migration event) X 7 regions of birth outside EU+ = total 28 scenarios Other Europe North Africa Sub-Saharan Africa West Asia South & South-East Asia East Asia Latin America short With persistence What-if scenarios simulated one by one as either short or persistent immigration event from each world region Short event: twice-in a century flows in one single year in 2025-2029 period Persistent event: increased immigration from High migration event region follows for 10 years, with gradually declining intensity due to family reunifications, migration networks and other reasons (after 10 yeas same flows from that region as in the baseline scenario) Rare migration events of immigration into EU+ from 8 world regions 51 Source: Bijak 2023, Deliverable 9.4 Results from twice-in-a-century immigration events •Scenarios simulating Immigration event from different world regions in 2025-2029 •A surplus of immigrants arrives from a specific world region •What would change compared to the baseline? Twice-in-a century migration event with persistence (millions) 2020-2024 2025-2029 2030-2034 2035-39 2040-2044 Baseline immigration form the rest of the world 15.3 11.9 12.2 12.7 13.0 + event from Other Europe 0 +2.7 +2.5 +0.3 0 + event from Sub-Saharan Africa 0 +2.5 +2.3 +0.3 0 + event from North Africa 0 +1.4 +1.3 +0.2 0 + event from West Asia 0 +3.2 +2.9 +0.3 0 +event from South&South-East Asia 0 +2.7 +2.4 +0.3 0 +event from East Asia 0 +1.2 +1.1 +0.1 0 +event from Latin America 0 +3.0 +2.8 +0.3 0 High-migration events projection scenarios key results Projected total labour force – Twice-in-a-century immigration with persistence Key messages and recommendations •Most EU countries will experience sizeable labour force declines in future •Positive effect of educational expansion – better educated LF but also better educated elderly -> potential for longer working lifes •Impacts of population ageing not as grim as when only LF size and age dependency considered •Inclusive policies that nurture human capital and remove barriers to labour force participation are feasible policy options that must be considered vis a vis immigration •Increased LF participation of women (and in particular immigrangt women) to LF would boost LF and can stabilize LFDR •Unrealistically sustained immigration would be needed to slow down population ageing and stabilize age dependency ratios -> not a feasible migration policy target •A lot is at stake if immigration is high but LF integration and inequality in educational achievent among migrant’children must be tackled How to interpret / understand demographic scenarios •Not predictions •Some are forecasts – best guess at most plausible future given the knowledge at the time of creation •What-if scenarios – modify assumptions and provide an outlook given those assumptions •Medium / baseline scenario – business as usual, can be a forecast •Alternative scenarios – usually try to cover scenario space given a rage of possible future developments •Quantitative, narrative-based •Understanding assumptions crucial to their interpretation •Long-term scenarios for better preparedness and assessment of policy options Limitations •Demographic scenarios based on characteristics approach - > how compositional changes in populations affect the outcome indicator •Abstract from the economic cycle •The association between the characteristic and the outcome can change in the future •Focus on labour supply, not linked to labour demand •How to account or automation, changing nature or work, skills matching references •Potančoková, M. et al. (2023) Discussion paper: •Potančoková, M. , Stonawski, M.J. & Gailey, N. (2021). Migration and demographic disparities in macro-regions of the European Union, a view to 2060. Demographic Research 45, 1317-1354. 10.4054/DemRes.2021.45.44. •Marois, G. , Bélanger, A. & Lutz, W. (2020). Population aging, migration, and productivity in Europe. Proceedings of the National Academy of Sciences, e201918988. 10.1073/pnas.1918988117. •Marois, G. & Potančoková, M. (2020). Scenarios of labour force participation and employment integration of immigrants in the EU: demographic perspective. Publications Office of the European Union 10.2760/021884. •Lutz, W., Amran, G., Belanger, A., Conte, A., Gailey, N., Ghio, D., Grapsa, E., Jensen, K., Loichinger, E., Marois, G. et al. (2019). Demographic Scenarios for the EU: Migration, population and education. Publications Office of the European Union , Luxembourg. 10.2760/590301. •Marois, G. , Sabourin, P. & Bélanger, A. (2019). How reducing differentials in education and labor force participation could lessen workforce decline in the EU-28. Demographic Research 41, 125-160. 10.4054/DemRes.2019.41.6. •Loichinger, E. & Prskawetz, A. (2017). Changes in economic activity: The role of age and education. Demographic Research 36 1185-1208. 10.4054/DemRes.2017.36.40. •Sanderson, W. & Scherbov, S. (2019). Prospective Longevity: A New Vision of Population Aging. Cambridge, MA: Harvard University Press. ISBN 978-0-06-7497561-3 •Lutz, W. & K.C., S. (2011). Global human capital: Integrating education and population. Science 333 (6042), 587-592. 10.1126/science.1206964. [USEMAP] Text Description automatically generated with medium confidence potancok@iiasa.ac.at Thank you Microsimulation model dimensions we focus on supply of labour driven by structural changes in potential labour force and we do not aim to simulate economic context, labour demand or changes in employment due to economic cycles. Also, there are no feedback effects included and labour force activity and employment depend purely on individual characteristics. Population-projection model with 13 dimensions that dynamically projects population of 31 EU and EFTA countries Simulations at country-level, preliminary results presented for EU28 Simulation starts in 2011, preliminary results presented for 2015-2060 Population consists of individual simulated actors Life courses are simulated in continuous time and the transitions between the states are determined stochastically When a change occurs to the characteristic of an individual (age, education, duration of stay, etc.), the module determines probabilistically whether or not event happens through a Monte Carlo experiment project the population for EU28 member countries in several socioeconomic and ethnocultural dimensions A person can be economically active between age 15 and 74 When a change occurs to the characteristic of an individual (age, education, duration of stay, etc.), themodule determines probabilistically whether or not he/she participates in the laborforce. The labor force participation status is imputed through a Monte Carlo experimentin which a random number is compared to the probability of being active: a successfultrial means that the simulated individual is active. Microsimulation model dimensions 61 x Native-born, born in other EU+ country, 8 world regions (EU+ = EU27, UK & EFTA) Assumptions on migration in QuantMig-mic §Emigration rates differ by place of birth §Share or emigration to other EU+ country and to the rest of the world differs by place of birth § §Intra-EU+ migration: emigration rates of native-born and EU+ born converge to average of EU15 in 2011-2019 -> declining migration rates from Eastern and Southern EU+ and slight increase for migrants from Western EU+ §Declining volumes of intra-EU+ migrants also because of ageing -> smaller cohorts of young adults § §Immigration from the rest of the world regions to EU+ countries follows scenarios §Destinations follow established migration corridors Baseline scenario key results •A baby of a low educated mother has less chance achieving a university degree, and consequently, to participate in the labor force. • •Labor force participation rates are currently lower for women and recent immigrants, who face multiple obstacles. • •What if this was not the case? • • Source: Marois, Guillaume, Patrick Sabourin, and Alain Bélanger (2019). "How reducing differentials in education and labor force participation could lessen workforce decline in EU28", Demographic Research, 41(6), 125-160. Reducing differentials in education and labor force participation could lessen workforce decline in EU28 First, here are the outputs of the projection regarding the population size. For the baseline scenario, we expect a decline from about 244M to 220M. The scenario equality in education, give about the same result in terms of labour force size, although high educated people are more likely to work. The reason is that when we improve education, we improve labour force rates, but we decline fertility levels, so at the end, we have high participation, but smaller cohorts. Equality for immigrants increases the size by 5M compared to the reference scenario, and finaly, the equality for women increases the size more than 10M, so equality for women is the one having the biggest impact. Overall when the three types of inequalities are equalized to the higher levels of other groups, the expected decline in the labor force size is reduced by 45%. Scenario assumptions To analyse the effect of thse divergent integration trajectories the assumptions on LFPR and unemployment are set at fixed demographic and educational trends, holding the demographic trends same for all of them and at fixed volume and composition of immigrantion flows 1. The baseline scenario presents a business as usual situation, an EU where labour force participation rates and employment rates of both EU-born and foreign-born population stay as they were in the recent past. 2. High integration scenario envisages the best of the worlds – the highest labour force participation and employment rates of immigrants follow the best performing EU countries as described in the parameters section above. This most optimistic scenario can be seen as upper limit, the parameters are explained in the previous section. 3. Low integration scenario follows the worst possible situation and follows the worst labour force participation and employment rates among the EU countries and combines the two into a lower limit scenario. In this scenario, the economic integration of immigrants deteriorates in all EU countries. 4. High employment scenario helps to understand purely the effect of high employment rates of immigrants. The activity (labour force participation) rates remain all EU Member states constant as of the recent years until 2060 but we model a situation of the highest observed employment rates of immigrant in all these countries. Scenarios of economic integration of immigrants 1.Baseline integration - no change in the integration of immigrants in labour force and employment compared to what was observed in recent years. 2. 2.High integration – integration improves and by 2040 immigrants have the same activity and employment rates as their native-born peers. 3. 3.Low integration -deterioration of the labour force participation and employment rates of immigrants. 4. 4.High employment - only employment rates improve to those of native-born peers by 2040, activity rates stay the same as observed in the recent years We built the assumptions on future employment using the parameters from the logit regression presented in equation 3. Our stylized scenarios envision a set of what-if situations The four scenarios described above thus combine a set of three trajectories (variants) of labour force participation rates (baseline, high and low integration) with three variants of assumed trajectories of employment rates for foreign-born population (baseline, high and low integration) at fixed labour force participation rates and employment rates of EU-born population Scenario and variant are not the same. Scenario is a combination of assumptions on several dimensions of the model. We have prepares several sets of assumptions for some of the dimensions – in this case labour force activity and employment. These sets are termed variants (high, low, baseline). These assumptions are formulated with respect to future labour force participation and employment rates as well as the differential in these rates between the immigrants vs. EU-born. Therefore, we specify four assumptions on baseline, high, low 1. The baseline scenario presents a business as usual situation, an EU where labour force participation rates and employment rates of both EU-born and foreign-born population stay as they were in the recent past. 2. High integration scenario envisages the best of the worlds – the highest labour force participation and employment rates of immigrants follow the best performing EU countries as described in the parameters section above. This most optimistic scenario can be seen as upper limit, the parameters are explained in the previous section. 3. Low integration scenario follows the worst possible situation and follows the worst labour force participation and employment rates among the EU countries and combines the two into a lower limit scenario. In this scenario, the economic integration of immigrants deteriorates in all EU countries. 4. High employment scenario helps to understand purely the effect of high employment rates of immigrants. The activity (labour force participation) rates remain all EU Member states constant as of the recent years until 2060 but we model a situation of the highest observed employment rates of immigrant in all these countries. continuation of parameters β[8 ]and β[9] from equation 2 and β[4 ]and β[5] of equation 3 throughout the projection integration variant assumes that parameters for immigrants (β[8 ]and β[9] from equation 2 and β[4 ]and β[5] from equation 3) converge to 0 for all countries by 2040. This assumption thus progressively removes the disadvantage of immigrants in the labour force and employment. By 2040, for a same country, age, gender, and level of education immigrants and natives would have the same lab β[8 ]and β[9 ]from equation 2[ ]are assumed to converge by 2040 for all countries to those of Denmark in 2010-2015, which is the country in the EU where those parameters are the lowest (see parameters in table 4 and table 5). For employment, parameters β[4 ]and β[5 ]from equation 3 converge to those observed in Sweden, which is the high-immigration country with the biggest gap between immigrants and natives. our force participation rates and the same unemployment rates. The four scenarios described above thus combine a set of three trajectories (variants) of labour force participation rates (baseline, high and low integration) with three variants of assumed trajectories of employment rates for foreign-born population (baseline, high and low integration) at fixed labour force participation rates and employment rates of EU-born population Scenario and variant are not the same. Scenario is a combination of assumptions on several dimensions of the model. We have prepares several sets of assumptions for some of the dimensions – in this case labour force activity and employment. These sets are termed variants (high, low, baseline). Preliminary results for EU28 Projected labour force participation rates in EU28, population age 15-74 The trajectories of labour force activity rates for EU population age 15-74 years for individual Member States, depending on country-specific participation rates for native-born and foreign-born populations. Looking at EU average, the labour force participation rates (LFPR) of EU-born population follow in all scenarios the same trajectory and they increase from 64% in 2015 to 67% in 2060. Age-specific age patterns of labour force participation and sizes of older age groups versus younger ones explain the curve seen in Figure 5 (dotted line). The increasing proportion of the older population (65-74) among the working age population (15-74) drives the light decline of the overall LFPR towards 2035 because this age group has low labour force participation. The increase after 2035 is associated particularly with higher labour force participation of women among the 55-74 and the growth of the highly educated – among the younger but as well among the older age groups – because highly educated men and women have the highest labour force participation rates. For foreign-born, the high integration variant (used on High integration scenario) represents an increase from 65% in 2015 (thus 1pp higher than the EU-born), peaking at 72% in 2040 followed by a gradual decrease to 66% in 2060. In the baseline variant (used in Baseline and High employment scenarios) LFPR follows a similar trajectory as in the high variant but at a lower activity level: starting at 65% in 2050, peaking at 67% in 2040 and declining to 61% in 2060. In the low integration variant (used in Low integration scenario) the labour force participation rates gradually decline to 53% in 2060. The initial increase in the first years is explained by the boost of young immigrants of prime working age to the existing stock in the EU. Because labour force participation increases with the duration of stay in the country, the inflow of new young immigrants drives the increase in the overall labour force participation rate of the immigrant population in the EU. The ageing of this immigrant stock is driving the declining overall labour force participation rate of foreign-born and this becomes apparent after 2040. Preliminary results for EU28 Projected employment rates for total population 15-74 Projected employment rates of foreign-born population 15-74 The difference between the Baseline and High employment scenarios illustrates the effect of improved employment of immigrants to the level of the EU-born population The model accounts for structural changes in EU-born and foreign-born populations and is purely labour supply driven. We do not project changes in employment rates within a broader economic context and in relationship to possibly changing demand for labour that may arise due to changing nature of work, digitalization and automation. . Closing of the gap between immigrants' and EU-born employment rates by 2040 thus increases the overall employment rate compared to the Baseline scenario by 1.17% (0.1pp) in 2040 and by 1.32% (0.8pp) in 2060. The very optimistic High integration scenario imagines that not only employment gap between the immigrants and EU-born closes by 2040, but that the gap in labour force participation closes as well. In other words, more immigrants would be willing to participate in the labour force and they will face no disadvantage in finding a job compared to their EU-born peers. In such case, the overall employment rate in the EU would be boosted by 2.8% (1.6pp) in 2040, resp. 3.4% (2pp) in 2060, compared to the baseline scenario. Projected employed population 2015-2060 QuantMig population projections & migration scenarios •Dynamic microsimulation population projection for 31 European countries (EU+) 2020-2060 •System modelling of migration between individual EU+ countries •What would be the long-term demographic and labour force implications of unforeseen migration events? •Scenarios of immigration events of different magnitude, duration and from a different world region • • • Successful economic integration of immigrants is of high importance as it fosters also their social, linguistic and cultural inclusion and improves the quality of life of migrants, their families and empowers communities. Immigrants' better economic integration is highly needed also against the backdrop of the unavoidable future smaller labour force sizes in the EU overall and in practically all Member States. we focus on supply of labour driven by structural changes in potential labour force and we do not aim to simulate economic context, labour demand or changes in employment due to economic cycles. Also, there are no feedback effects included and labour force activity and employment depend purely on individual characteristics. Baseline Persistent-high immigration event from East Asia Persistent-high immigration event from Latin America Persistent-high immigration event from North Africa Persistent-high immigration event from other Europe Persistent-high immigration event from South and South-East Asia Persistent-high immigration event from Sub-Saharan Africa Persistent-high immigration event from West Asia Working age population 80 81 81 81 81 81 81 82 Labour force 86 86 87 87 87 87 87 87 Population born outside EU+ 194 197 201 198 201 200 200 203 Labour force born outside EU+ 172 174 179 175 179 179 179 180 2020 Baseline Persistent-high immigration event from East Asia Persistent-high immigration event from Latin America Persistent-high immigration event from North Africa Persistent-high immigration event from other Europe Persistent-high immigration event from South and South-East Asia Persistent-high immigration event from Sub-Saharan Africa Persistent-high immigration event from West Asia % of population born outside EU+ 8% 17% 17% 17% 17% 17% 17% 17% 17% % of labour force born outside EU+ 15% 26% 26% 27% 26% 27% 26% 27% 27%