Moving Up the Ladder? The Impact of Migration Experience on Occupational Mobility in Albania Calogero Carletto and Talip Kilic Development Research Group, The World Bank Prepared for the Fourth IZA/World Bank Conference Employment and Development Bonn Deutsche Post DHL 04/05/2009
OUTLINE Objective Background Literature Review Data Empirical Approach Results Conclusions
OBJECTIVE Analyze the impact of international migration experience on labor mobility of return migrants vis a vis non-migrants by using data on initial & current employment outcomes o Test the hypothesis of upward occupational mobility induced by international migration o Explore the heterogeneity of impact by host country (Greece vs. Italy & Beyond)
BACKGROUND 1945-1990: Communist regime s ban on international migration 1/5 th of population, driven by widespread poverty & unemployment, migrated abroad between 1991 and 2001; mainly to Greece & Italy o 1 in 3 HHs currently have a migrant abroad (50% of these have >1) o Remittances estimated to have exceeded $US 1 billion by 2005 o Impact of migration on poverty (+), productive investments (?) Much of the migration temporary in nature (circular) o Multiple episodes (~4 in lifetime) prior to settlement Growing number of returnees to re-establish residence in Albania o Recent initiatives to encourage the return of the highly skilled Channels of potential development impact: financial & human capital accumulated abroad
LITERATURE REVIEW 1 st Strand: Castano ( 88): Colombia Arif & Irfan ( 97): Pakistan Ilahi ( 99): Pakistan Dustmann & Kirchkamp ( 02): Turkey McCormick & Wahba ( 04): Egypt Mesnard ( 04): Tunisia Woodruff & Zenteno ( 04): Mexico Gubert et al. ( 08): MENA 2 nd Strand: Co et al. ( 00): Hungary Zhao ( 02): China Wahba ( 07): Egypt de Coulon & Piracha ( 05): Albania Kilic et al. (forthcoming): Albania Albania Case Studies Barjaba ( 00) Nicholson ( 01, 02) Labrianidis & Kazazi ( 06) Labrianidis & Hatziprokopiou ( 06)
DATA 2005 Albanian Living Standards Measurement Study Survey (ALSMS05) o Conducted by the INSTAT, with assistance from the World Bank o Stratified into four regions: Coastal, Central, Mountain & Tirana o Total sample: 3,640 HHs in 455 PSUs o HH (w/ extensive migration module) & community questionnaires Data on o 2005 & Initial (1990 or the year individual turned 15) employment outcomes o Migration & international employment histories of all adults Sample of interest: 9,565 Individuals [15,64] years of age o Returnees within the last year are excluded from the sample o 853 return migrants (9%) in the final sample
DATA (2) In comparison with non-migrants, returnees are, on average, o Older o More educated o Better endowed in terms of durable assets & social capital o More likely to Be labor force participants Hold non-agricultural, particularly high-skilled blue collar, jobs Experience upward occupational mobility*
OCCUPATIONAL CLASSIFICATION ALSMS05 occupational outcomes classified according to the ISCO-88 o 3-digit codes 10 major groups 5 broad occupational categories Agriculture Skilled agricultural & fishery workers Low-Skilled Blue Collar Plant & machine operators & assemblers; Elementary occupations High-Skilled Blue Collar Craft & related trade workers Low-Skilled White Collar Clerks; Service workers & ship & market sales workers High-Skilled White Collar Legislators, senior officials & managers; Professionals; Technicians & associate professionals
OCCUPATIONAL RANKING Occupational categories ranked according to the average level of human capital necessary to be in a given category (Sicherman & Galor, 1990) o Run a wage regression on observable covariates, including years of education, a proxy for labor market experience prior to current occupation & tenure at current occupation o Average the individual sums of weighted education & experience levels within each occupational category, where the weights are the coefficients from the wage regression Occupational Category Rankings Rank Category Index Value 1 Not Working -- 2 Agriculture 0.67 3 Low Skilled Blue Collar 0.80 4 High Skilled Blue Collar 0.83 5 Low Skilled White Collar 0.93 6 High Skilled White Collar 1.20
Employment Transition Matrices Non-Migrant Population (Panel 1) 2005 Employment Status Initial Employment Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Status Working Blue Collar Blue Collar White Collar White Collar Not Working 30.84 9.02 1.65 2.00 2.45 2.91 48.86 Agriculture 7.64 16.01 1.23 1.05 1.12 0.29 27.34 Low Skilled Blue Collar 3.17 0.57 2.43 0.40 0.66 0.20 7.42 High Skilled Blue Collar 2.18 0.42 0.51 2.46 0.54 0.38 6.50 Low Skilled White Collar 0.77 0.17 0.12 0.10 0.66 0.25 2.07 High Skilled White Collar 1.87 0.40 0.21 0.16 0.53 4.64 7.80 Total 46.46 26.59 6.15 6.18 5.95 8.67 100.00 Return Migrant Population (Panel 2) 2005 Employment Status Initial Employment Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Status Working Blue Collar Blue Collar White Collar White Collar Not Working 9.33 5.87 4.08 6.96 5.00 4.27 35.51 Agriculture 1.83 13.80 2.81 5.27 2.46 0.47 26.63 Low Skilled Blue Collar 1.65 1.46 3.46 1.20 0.39 0.65 8.81 High Skilled Blue Collar 2.57 1.19 1.43 9.16 1.83 1.72 17.91 Low Skilled White Collar 0.45 0.35 0.20 0.52 0.73 0.17 2.44 High Skilled White Collar 1.12 0.83 0.49 0.71 0.84 4.72 8.70 Total 16.96 23.49 12.47 23.83 11.24 12.01 100.00
Employment Transition Matrices Return Migrant Population from Greece (Panel 3) 2005 Employment Status Initial Employment Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Status Working Blue Collar Blue Collar White Collar White Collar Not Working 8.22 6.35 3.99 6.71 4.13 2.63 32.03 Agriculture 1.77 14.78 3.05 6.01 3.11 0.60 29.32 Low Skilled Blue Collar 1.39 1.85 3.46 1.49 0.47 0.53 9.19 High Skilled Blue Collar 2.70 1.11 1.51 9.98 1.62 1.69 18.61 Low Skilled White Collar 0.38 0.28 0.26 0.61 0.44 0.22 2.19 High Skilled White Collar 1.27 1.05 0.61 0.40 1.04 4.30 8.66 Total 15.73 25.42 12.88 25.19 10.81 9.98 100.00 Return Migrant Population from Italy & Beyond (Panel 4) 2005 Employment Status Initial Employment Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Status Working Blue Collar Blue Collar White Collar White Collar Not Working 13.53 4.04 4.41 7.92 8.25 10.41 48.55 Agriculture 2.04 10.11 1.91 2.50 0.00 0.00 16.56 Low Skilled Blue Collar 2.64 0.00 3.46 0.12 0.10 1.10 7.41 High Skilled Blue Collar 2.09 1.51 1.16 6.05 2.62 1.83 15.27 Low Skilled White Collar 0.72 0.59 0.00 0.20 1.83 0.00 3.34 High Skilled White Collar 0.59 0.00 0.00 1.91 0.08 6.29 8.87 Total 21.61 16.25 10.93 18.70 12.88 19.63 100.00
Employment Transition Matrices for Return Migrants Employment Status in Last Migration Episode (Panel 1) Initial Employment Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Status Working Blue Collar Blue Collar White Collar White Collar Not Working 7.09 8.93 4.06 12.06 2.54 0.65 35.33 Agriculture 3.37 14.30 0.77 7.90 0.37 0.00 26.71 Low Skilled Blue Collar 0.49 2.82 1.22 4.09 0.22 0.00 8.84 High Skilled Blue Collar 1.76 3.30 0.93 11.37 0.65 0.00 18.01 Low Skilled White Collar 0.28 0.64 0.28 0.99 0.26 0.00 2.44 High Skilled White Collar 1.54 1.74 1.22 2.74 1.16 0.26 8.67 Total 14.53 31.72 8.48 39.15 5.21 0.91 100.00 2005 Employment Status (Panel 2) Employment Status in Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Last Migration Episode Working Blue Collar Blue Collar White Collar White Collar Not Working 5.66 2.54 1.31 2.15 1.03 1.73 14.43 Agriculture 3.03 14.63 4.11 5.76 3.02 1.33 31.88 Low Skilled Blue Collar 2.69 0.73 1.78 0.79 1.04 1.73 8.77 High Skilled Blue Collar 4.30 5.60 5.04 14.40 5.01 4.50 38.85 Low Skilled White Collar 1.22 0.00 0.21 0.63 1.04 2.08 5.17 High Skilled White Collar 0.11 0.13 0.00 0.00 0.00 0.67 0.91 Total 17.00 23.64 12.45 23.74 11.14 12.03 100.00
Employment Transition Matrices for Return Migrants from Greece Employment Status in Last Migration Episode (Panel 3) Initial Employment Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Status Working Blue Collar Blue Collar White Collar White Collar Not Working 6.22 9.72 2.85 11.01 1.58 0.47 31.86 Agriculture 2.78 16.81 0.84 8.59 0.37 0.00 29.39 Low Skilled Blue Collar 0.26 3.34 0.96 4.51 0.13 0.00 9.21 High Skilled Blue Collar 1.33 3.69 0.64 12.31 0.68 0.00 18.66 Low Skilled White Collar 0.00 0.62 0.36 1.02 0.21 0.00 2.20 High Skilled White Collar 1.17 2.15 1.55 2.62 0.92 0.27 8.68 Total 11.76 36.34 7.20 40.07 3.89 0.74 100.00 2005 Employment Status (Panel 4) Employment Status in Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Last Migration Episode Working Blue Collar Blue Collar White Collar White Collar Not Working 5.05 1.84 1.14 2.24 0.56 0.88 11.71 Agriculture 3.13 16.92 4.99 6.43 3.40 1.48 36.34 Low Skilled Blue Collar 1.97 0.55 1.46 0.81 1.24 1.42 7.45 High Skilled Blue Collar 4.77 5.89 5.03 15.08 4.64 4.48 39.89 Low Skilled White Collar 0.88 0.00 0.26 0.59 0.87 1.27 3.87 High Skilled White Collar 0.13 0.17 0.00 0.00 0.00 0.44 0.74 Total 15.93 25.37 12.89 25.15 10.71 9.96 100.00
Employment Transition Matrices for Return Migrants from Italy & Beyond Employment Status in Last Migration Episode (Panel 5) Initial Employment Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Status Working Blue Collar Blue Collar White Collar White Collar Not Working 10.38 5.94 8.61 16.02 6.14 1.31 48.39 Agriculture 5.59 4.85 0.50 5.30 0.39 0.00 16.63 Low Skilled Blue Collar 1.36 0.85 2.16 2.52 0.56 0.00 7.44 High Skilled Blue Collar 3.37 1.83 2.02 7.81 0.54 0.00 15.57 Low Skilled White Collar 1.32 0.71 0.00 0.87 0.46 0.00 3.36 High Skilled White Collar 2.92 0.18 0.00 3.20 2.07 0.24 8.60 Total 24.93 14.35 13.29 35.73 10.15 1.55 100.00 2005 Employment Status (Panel 6) Employment Status in Not Agriculture Low Skilled High Skilled Low Skilled High Skilled Total Last Migration Episode Working Blue Collar Blue Collar White Collar White Collar Not Working 7.97 5.17 1.95 1.84 2.76 4.88 24.57 Agriculture 2.65 6.10 0.84 3.27 1.62 0.78 15.27 Low Skilled Blue Collar 5.38 1.38 2.98 0.73 0.32 2.89 13.67 High Skilled Blue Collar 2.53 4.53 5.05 11.86 6.39 4.61 34.97 Low Skilled White Collar 2.49 0.00 0.00 0.80 1.65 5.06 10.00 High Skilled White Collar 0.00 0.00 0.00 0.00 0.00 1.52 1.52 Total 21.02 17.18 10.82 18.50 12.74 19.74 100.00
EMPIRICAL APPROACH Outcome of interest: Degree of Occupational Mobility Definition: Occupational ranking in 2005 MINUS initial occupational ranking (Leigh, 1975 & Chiswick & Lee, 2005); [-5, 5] Model: Ordered Probit Covariates of interest: o Model 1: Return migrant status, i.e. dummy variable = 1 if a surveyed individual has spent at least 1 month abroad for work purposes o Model 2: Dummy variables that separately account for past migration experience in Greece, Italy & Beyond.(ITB)
EMPIRICAL APPROACH (2) Concerns for endogeneity of return migrant status: Past migration/return decision & occupational outcomes may be jointly determined by unobservable individual & household characteristics Solution: (Non-traditional) Instrumental Variable (IV) Approach o Probit regression of return migrant status o Instrumental variables: HH knowledge of Greek or Italian in 90 Annual avg. # of HH shocks prior to the 1 st (For Non-Migrants: Avg. for 90-05) migration episode # of HH members [0,5] in Albania during the last migration episode (For Non-Migrants: # of HH children in 98) o Use the predicted value of return migrant status as an independent variable o Empirical steps to support theoretical arguments for IV Validity?
EMPIRICAL APPROACH (3) D. = if an individual is male Years of age & its squared term Years of education & its squared term # of HH male members [15,60] # of HH female members [15,60] # of HH members [60+] D. = 1 if HH is female-headed HH land owned & its squared term D. = 1 if dwelling is a brick home Economic status in 90 [1,10] D. = 1 if dwelling was a single family home in 90 D. = 1 if HH receives public transfers D. = 1 if HH receives non-farm real estate earnings HH Social Capital Index (PCA) Regional Fixed Effects: Coastal Urban, Coastal Rural, Central Urban, Central Rural, Mountain Urban & Mountain Rural, where the reference category is Tirana
RESULTS Marginal Effects Non-IV Estimates IV Estimates Model 1 Model 2 Model 1 Model 2 Mobility Category Returnee Returnee from Greece Returnee from ITB Returnee Returnee from Greece Returnee from ITB (-5) -0.002** -0.001-0.003*** -0.013*** 0.000-0.046*** (-4) -0.001** -0.001-0.002*** -0.009*** 0.000-0.032*** (-3) -0.004** -0.003-0.007*** -0.025*** 0.000-0.087*** (-2) -0.006** -0.004-0.011*** -0.037*** 0.000-0.130*** (-1) -0.013** -0.009-0.026*** -0.082*** 0.001-0.287*** (0) -0.009* -0.005-0.025* -0.038*** 0.000-0.133*** (+1) 0.012** 0.008 0.024*** 0.072*** -0.001 0.252*** (+2) 0.006** 0.004 0.012** 0.035*** -0.000 0.123*** (+3) 0.006** 0.004 0.013** 0.034*** -0.000 0.121*** (+4) 0.005** 0.004 0.012** 0.031*** -0.000 0.108*** (+5) 0.006** 0.004 0.014** 0.032*** -0.000 0.112***
CONCLUSIONS Migration experience, particularly in ITB, promotes upward labor mobility o Robust across different specifications & sample definitions The result signals a potential positive contribution of migration to economic development in Albania o Particularly important given global economic woes and the projected trends in remittance inflows \ Important to o Build on programs encouraging return migration of the skilled o Recognize heterogeneity in return migrants needs & capabilities to realize the full potential of financial & human capital accumulated abroad