Climate Change, Inequality and Migration

Transcription

1 Climate Change, nequality and Migration M. Burzynski, C. Deuster, F. Docquier & J. de Melo Paris, January 23rd, 2018 Climate Change, nequality and Migration

2 n the media... But 3% of int l migrants today, 1% of emig. from poorest countries... Climate Change, nequality and Migration

3 What do we do? Quantitative objectives Estimate the mobility responses to global warming (GW) over 21st century and under current migration laws and policies Assess role of migration policies (inequality, extreme poverty) OLG model of the world economy RUM to model mobility decisions (stay, local, urban, long dist) Accounts for country characteristics (speci c costs) Accounts for direct/indirect e ects, for dynamic aspects Simplifying hypotheses GW = rises in temperature and in sea level only Exogenous GW (no feedback e ects: growth/urban! GW) Two adaptation mechanisms: switching crops (implicit) & migrating Climate Change, nequality and Migration

4 Table of contents 1. Motivation 2. Heterogeneous e ects Climate Change, nequality and Migration

5 Why linking GW to migration? Mean surface temp of the world has increased since 19th, and the process has accelerated since st cent: +1 to +3 C in temp, +0.3 to +1.5m in sea level Many economic implications: TFP, forced displacements, health, drudgery of work, con icts, etc. (Dell et al. 2014) Heterogeneous e ects across countries! Nonlinear e ects of temp on TFP and utility: initial levels matter Exposition to sea-level rise Di erent adaptation capacities, etc. ) Favourable conditions for increasing labor mobility Climate Change, nequality and Migration

6 Literature review What does the empirical literature say? Mix of case studies, cross-country, lit. reviews (Piguet et al. 2011; Millock 2015; Berlemann-Steinhardt 2017; Cattaneo et al. 2018). Wide variety of results but mig responses are usually small Meta-analysis (Beine-Jeusette 2018): ndings depend on... Measurement of dependent and climatic variables Direct vs indirect impact (income/tfp, con icts) Speci cation: unconditional vs conditional e ects Sample choice and identi cation strategies Climate Change, nequality and Migration

7 Empirical approach Cross-country ndings: Marchiori et al. (2012): weather anomalies directly a ects urban. and indirectly a ect int l migration Marchiori et al. (2015): little e ect of weather-driven income volatility on int l migration Coniglio-Pesce (2015), Backhaus et al. (2015): variability in rainfalls a ects mig to OECD Beine-Parsons (2015): natural disasters a ect urbanization + int l migration via income in cities Cattaneo-Peri (2016): rise in temp increases internal and int l mig in middle-income countries, reduces it in low-income countries Cai et al. (2016), Henderson et al. (2017)... Climate Change, nequality and Migration

8 Empirical approach Limitations of empirical studies: Distinguishing btw climate variables & other drivers is tough Mobility responses are context speci c (geography, development, network, political, demog, cultural) GW has not fully materialized yet Our response: use quantitative theory (world economy model) Climate Change, nequality and Migration

9 Alternative approach Quantitative theory is appropriate to multiple mig. options at various spatial scales Account for the context (calibration of migration technologies using dyadic data: very good predictive power when backcasting) Account for static general equilibrium e ects and externalities (direct and rich indirect e ects of GW) Account for dynamic e ects (pop growth, education, etc.) Closest reference: Desmet and Rossi-Hansberg (2015) DRH: continuous space, same price/inputs at a given latitude Here: regions and countries (6= heterogeneity + borders + sea level) Climate Change, nequality and Migration

10 Table of contents Baseline scenario Channels of transmission Alternative scenarios 1. Motivation 2. Heterogeneous e ects Baseline climate scenario Channels of transmission Alternative scenarios Climate Change, nequality and Migration

11 Baseline scenario Baseline scenario Channels of transmission Alternative scenarios Raw data + projections of monthly temperature levels Decreasing relationship btw current mean temp and mean latitude Median CCPK scenario of median emissions: RCP-4.5 Median RCP variant w.r.t. temperature: C after 2010 Link CCPK climatological windows to 2040, 2070 and 2100 Correction for population density Dell et al. (2012): population-weighted temperature in We compute scale correction & use it 8 period/month (T m,r,t ) Climate Change, nequality and Migration

12 Baseline scenario Baseline scenario Channels of transmission Alternative scenarios Mean temp in 2010 and latitude Worldwide mean temp in RCP-4.5 Climate Change, nequality and Migration

13 Baseline scenario Baseline scenario Channels of transmission Alternative scenarios Projections on monthly temperature levels LR variations in temperature do not vary with latitude Link temperature (+2.09 C) to rising sea level Uncertain dynamics of ocean heat uptake and ice sheets/glacier Vermeer-Rahmstorf (2009), DeConto-Pollard (2016): +1.1m Climate Change, nequality and Migration

14 Baseline scenario Baseline scenario Channels of transmission Alternative scenarios Var. temp. & latitude ( ) Var in temp. & sea level Climate Change, nequality and Migration

15 Channels of transmission Baseline scenario Channels of transmission Alternative scenarios Economic implications? Temperature and productivity Desmet and Rossi-Hansberg (2015): n o G r (T ) = max g 0r + g 1r T + g 2r T 2 ; 0 Agr: agronomic studies, envelope of crop-speci c relationships Nonagr: relationship between population density and latitude TFP scale factor: G r,t = m=1 G r (T m,r,t ) Productivity responses are country-speci c: initial temp. matters And more... Climate Change, nequality and Migration

16 Channels of transmission Baseline scenario Channels of transmission Alternative scenarios G a (T) : Agricultural TFP and temp. G n (T) : Nonagricultural TFP and temp. After Gaussian smoothing and standardization Climate Change, nequality and Migration

17 Channels of transmission Baseline scenario Channels of transmission Alternative scenarios G a /G a ( ) and latitude G n /G n ( ) and latitude TFP responses vary with initial temp (i.e., with latitude, as in DHR) Climate Change, nequality and Migration

18 Channels of transmission Baseline scenario Channels of transmission Alternative scenarios Temperature and productivity Temperature and utility Output per worker falls by 2% per 1 C when temp is above 22 C Assume it is due to lower participation rate (`) Quasi-linear model U(c,l): U U = (1 + ϑ) ` = 2(1 + ϑ) T ` Rising sea level Use NASA data to identify share pop by elevation (Θ r,t ) More to come: impact of GW through con icts!!! GW ) frequency of extreme events () temp & short-dist mig.) High frequency may induce tensions over resources and con icts Climate Change, nequality and Migration

19 Channels of transmission Baseline scenario Channels of transmission Alternative scenarios Percentage of population living below 1.1m in 2010 Our model endogenizes forced displacements in non-oecd countries Climate Change, nequality and Migration

20 Baseline scenario Channels of transmission Alternative scenarios Alternative scenarios Use model to simulate 4 scenarios ( ): Baseline scenario C in temp, +1.1m in sea level Optimistic scenario C in temp, constant sea level (unrealistic) Pessimistic scenario C in temp, +1.3m in sea level (RCP-8.5) Worst-case scenario. Pessimistic + direct utility loss More to come: con icts in most adversely a ected countries Climate Change, nequality and Migration

21 Table of contents Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization 1. Motivation 2. Heterogeneous e ects Technology Preferences Dynamics and equil. Parametrization Climate Change, nequality and Migration

22 Structure Motivation Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization World economy with 145 developing countries and the OECD Two age groups: adults (decision makers) and children Two skill groups (s=h,l): college grads & less educated Two regions (r=a,n): agriculture and nonagriculture Two areas (b=f,d): ooded and un ooded The model endogenizes: Mobility: local, rural-urban (short-dist), to OECD (long-dist) Self-selection of migrants Population dynamics: net migration, fertility and education World distribution of income, hum cap, TFP and poverty Migration is the only linkage between regions Climate Change, nequality and Migration

23 Technology Motivation Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization Output is feasible in un ooded areas only: CES technology: Y r,t = A r,t s η,s,t`σr 1 σr σr 1 σr r r,s,t With s = (h, l) = College grads vs. Less educated And r = (a, n) = Agr vs. Nonagr Technological externalities: Aggregate: A r,t = γ t A r G r,t `r,h,t Skill-bias: Γ η r,t η r,h,t η r,l,t = Γ η r `r,l,t ɛr `r,h,t `r,l,t κr These eqs. govern income and productivity disparities Climate Change, nequality and Migration

24 Preferences Motivation Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization Two types of adult by region and by skill group Climate Change, nequality and Migration

25 Preferences Motivation Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization Adults raised in un ooded areas: N d r,s,t = (1 Two-stage random utility model: Outer utility function, r! r = (a, n, F ): Θ r,t )N r,s,t U d r r,s,t = ln v d r,s,t + ln(1 x r r,s,t ) + ξ d r r,s,t nner utility function (warm glow): ln v d r,s,t = ln(1 T r,t ) + ln c d r,s,t + θ ln nr d,s,tpr d,s,t Budget constraint: c d r,s,t = w r,s,t (1 φn d r,s,t) n d r,s,tq d r,s,te r,t Training technology: p d r,s,t = π r + q d r,s,t λ Climate Change, nequality and Migration

26 Preferences Motivation Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization Education and fertility (interior): ( q d r,s,t = λφw r,s,t π r E r,t n d r,s,t = (1 λ)e r,t θ(1 λ) 1+θ w r,s,t ) vr d,s,t(.) φw r,s,t π r E r,t Migration (taste shocks ξ d r r,s,t are EVD(0,µ)): m d r r,s,t Md r r,s,t M d r r,s,t = v d r,s,t v d r,s,t! 1/µ (1 x r r,s,t) 1/µ Eqs. govern consumption, fertility, educ. & mobility Climate Change, nequality and Migration

27 Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization Preferences Adults raised in ooded areas: N f r,s,t = Θ r,t N r,s,t One di erence: they lose a fraction B of their labor earnings if they relocate within the region of birth (no compensation): w r,s,t (1 φn d r,s,t)! (1 B)w r,s,t (1 φn f r,s,t) Decrease in local utility: v f r,s,t (.) < v d r,s,t (.) Di erent migration responses: m f r r,s,t Mf r r,s,t M f r r,s,t = v d r,s,t v f r,s,t! 1/µ (1 x r r,s,t) 1/µ Climate Change, nequality and Migration

28 Rest of the model Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization Access to education: E r,t = ψ r,t w r,h,t Pop & labor supply in un ooded area only (L r,s,t, `r,s,t ) Population dynamics (idem for N r,l,t+1 ): N n,h,t+1 = s " N d n,s,tn d r,s,tp d r,s,t + Nf n,s,tnr f,s,tpr f,s,t 1 + m na,s,t + m nf,s,t 1 + mna,s,t f + mnf f,s,t # + m an,s,tn d a,s,tn d r,s,tp d r,s,t 1 + m an,s,t + m af,s,t + mf an,s,tn f a,s,tn d r,s,tp d r,s,t 1 + m f an,s,t + m f af,s,t Climate Change, nequality and Migration

29 ntertemporal equilibrium Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization De nition For a set fγ, θ, λ, φ, µ, Bg of common parameters, a set of sector-speci c elasticities fσ r, ɛ r, κ r g, a set of region-speci co exogenous characteristics na r, R η r, x r r,s,t, T r,t, Θ r,t, ψ r,t, π r, and a set fn r,s,0 g of predetermined variables, an intertemporal equilibrium is a set of endogenous variables Ar,t, η r,s,t, w r,s,t, E r,t, `r,s,t, N b r,s,t+1, nb r,s,t, q b r,s,t, v b r,s,t, m b r r,s,t satisfying technological constraints, pro t & utility max conditions, and population dynamics in all countries of the world. Climate Change, nequality and Migration

30 Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization Data Calibration for 145 developing countries + OECD single entity Macrodata on VA, population, HC by country/sector for Bilateral migration matrices (DOC), urbanization trends Microdata on fertility, income per HH member, mig. intentions/plans by region and by education level (Gallup World Polls) UN socio-demographic projections for 2040 (pop and HC) Climate Change, nequality and Migration

31 Parameters Motivation Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization Technology Elasticity of substitution: σ n = 2 and σ a = η r,s,t matches w r,h,t w ; A r,l,t r,s,t matches Y r,t in 1980 and 2010 Skill biased extern. (correlation): κ n =.26 and κ a =.00 TFP extern. (correlation): ɛ n =.56 and ɛ a =.64 Externality = halved correlations (κ n =.13, ɛ n =.28, ɛ a =.32) Preferences Common parameters: θ =.2, λ =.6, φ =.1, µ = 1.4 Mig costs x rf,s : match DOC + Gallup data Others (π r, ψ r,t, x an,s ): match pop, educ, urban in (+ in ) Climate Change, nequality and Migration

32 Projections Motivation Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization Estimation of a convergence eq. for ψ r,t (access educ) dentify ψ r,t in 1980 and 2010 and predictions for h i 2 ln (ψ r,t+1 /ψ r,t ) = α r +β 1,r ln(ψ US r,t /ψ r,t ) + β 2,r ln(ψ US r,t /ψ r,t ) Convergence btw middle-income and rich countries Constant migration costs and other parameters Socio-demo outcomes in line with o cial projections: proof of concept that the stylized model is relevant (Burzynski et al. 2017)!!! Climate Change, nequality and Migration

33 Parameters Motivation Technology Preferences: people born in un ooded areas Preferences: forcibly displaced people Parameterization E ect of temperature and rising sea level G r,t and Θ r,t identi ed above Utility loss of increasing temp (health, drudgery of work): Output per worker decreases by 2% per 1 C above 22 C Quasi-linear utility (with LS elasticity of 1/3): U U = 4 l l = 8 T Relocation costs for forcibly displaced people: B =.5 Literature on con icts (Fiala 2015; banez and Moya 2006; Kellenberg and Mobarak 2011) Climate Change, nequality and Migration

34 Table of contents Worldwide implications Country-speci c implications nternational migration Role of migration policies 1. Motivation 2. Heterogeneous e ects World economy Heterogeneous e ects Migration responses Role of mig. policy Climate Change, nequality and Migration

35 World economy responses Worldwide implications Country-speci c implications nternational migration Role of migration policies Worldwide responses: weighted average of rich & poor countries Small e ects: GW increases income per worker... Negligible e ects on population growth and education Larger e ects: GW makes the world distribution of income more unequal Urbanization increases (in the worst-case scenario) Share of international migrants to OECD increases Climate Change, nequality and Migration

36 World economy responses Worldwide implications Country-speci c implications nternational migration Role of migration policies ncome per worker Theil index World population Share of college grads Climate Change, nequality and Migration

37 World economy responses Worldwide implications Country-speci c implications nternational migration Role of migration policies Urbanization Share of int l migrants (OECD) Climate Change, nequality and Migration

38 Country-speci c responses Worldwide implications Country-speci c implications nternational migration Role of migration policies Country-speci c responses to +2 C ncome pc College grads Urbanization Emigration Self-selection E ect by line of latitude? Climate Change, nequality and Migration

39 Country-speci c responses Worldwide implications Country-speci c implications nternational migration Role of migration policies ncome per worker Share of college graduates Equator: -15% in mean income Climate Change, nequality and Migration

40 Country-speci c responses Worldwide implications Country-speci c implications nternational migration Role of migration policies Urbanization Skill bias in internal migration Equator: +5% in urban (e.g to 52.5%: =2.5% of pop) Climate Change, nequality and Migration

41 Country-speci c responses Worldwide implications Country-speci c implications nternational migration Role of migration policies Emigration rate Skill bias in int l migration Equator: +12.5% in emig. (e.g. 2.0 to 2.25%: =0.25% of pop) Climate Change, nequality and Migration

42 Country-speci c responses Worldwide implications Country-speci c implications nternational migration Role of migration policies Country-speci c responses to +2 C: vary with latitude Urbanization: +5% close to equator, less positive selection Emigration: +12.5% close to equator, less positive selection College grads: small negative responses close to equator (income e ect... but urbanization and increasing emigration) ncome pc: -15% close to equator, +15% at high level of latitude Most adversely a ected countries (Sao Tome & P, Gambia, Venezuela, Nepal, Grenada, Nicaragua, Malaysia, Dom R, Ghana, Philippines) Extreme pov. (inc. <2% of world average): GW increases poverty at the extensive (pov headcounts) and intensive margin (pov depth) Climate Change, nequality and Migration

43 Country-speci c responses Worldwide implications Country-speci c implications nternational migration Role of migration policies Pov heacounts Pov depth Ten poorest GW increases extreme poverty at the ext. and int. margins Climate Change, nequality and Migration

44 nt l migration Motivation Worldwide implications Country-speci c implications nternational migration Role of migration policies Under current migration laws and policies, is a juggernaut of climate refugees headed for OECD countries? Climate Change, nequality and Migration

45 nt l migration Motivation Worldwide implications Country-speci c implications nternational migration Role of migration policies Number and share of movers in 2040, 2070 and 2100 Number (10E6) As % world pop Base-Optimistic Total Rural-Urban nternational Local Flooded Pessimistic-Base Total Rural-Urban nternational Local Flooded M/P goes from 2.4 to 3.4% (GW explains 1/5), 85% of ooded move locally Climate Change, nequality and Migration

46 Worldwide implications Country-speci c implications nternational migration Role of migration policies nt l migration Emigration rates by region (as % of native pop 25-64) Baseline No SL Optim Pess LAC SSA MENA Asia OECD Cont of GW (x ) to rising emig (x2) is limited... not due to SL Forced movers: 85-90% move locally, a few emigrate abroad Climate Change, nequality and Migration

47 Worldwide implications Country-speci c implications nternational migration Role of migration policies nt l migration mmigration to OECD countries (as % of resident pop 25-64) Baseline No SL Optim. Pessim USA Canada Australia EU Germany France UK taly LR contr. of GW to rising immig (x2) is limited, and not due to SL Climate Change, nequality and Migration

48 Policy implications Worldwide implications Country-speci c implications nternational migration Role of migration policies ) GW has little e ect on future int l migration pressures Should OECD countries adjust their immigration policy to limit the inequality and poverty e ects of GW? Relaxing (or reinforcing) restrictions countries with greatest pov headcounts vs 20 most heavily a ected countries by GW Policy for all workers vs for low-skilled workers in agriculture Reinforcing restrictions has little e ect: current costs are large!!! Fall in extreme poverty only if the policy targets poorest group Climate Change, nequality and Migration

49 Policy implications: pov headcounts Worldwide implications Country-speci c implications nternational migration Role of migration policies Poorest, all workers Poorest, LS rural GW impact, all workers GW impact, LS rural Climate Change, nequality and Migration

50 Table of contents 1. Motivation 2. Heterogeneous e ects Climate Change, nequality and Migration

51 Concluding remarks GW increases inequality and extreme poverty Mobility responses: Local nterregional > nternational Concerns about international migration pressures? Current policies: small impact on international migration (+0.2pp) Small e ects of relaxing migration policies in GW-a ected countries What is a climate refugee? 85-90% of forcibly displaced people move locally Half of non-local movements... and 95% of international movements are voluntary (indirect economic channel) Climate Change, nequality and Migration

52 Thanks for your attention Climate Change, nequality and Migration