view pdf - Integration of Immigrants Programme - Massey University

Assimilation, Ethnic Capital, and the
Earnings of Immigrants in New Zealand: A
Spatial Approach
Paper presented at Pathways Conference, Massey University (Albany),
Friday 26 th October 2012.
Xingang (Singa) Wang & Sholeh A. Maani
Economics Department
The University of Auckland
© K.V. Fernandez

Outline
• Introduction
--background
--motivation
• Ethnic Capital
• Models: Basic and Spatial Models
• Data
• Empirical Evidence
• Conclusion

Introduction-- Background
• Assimilation is an important indicator which measures
the economic performance of immigrants.
• The word “assimilation” refers to the processes along
which an immigrant’s earnings converge to a
comparable native level, after residing in the host
country for a certain period of time.

Introduction-- Background
• Disadvantages of immigrants in the local labour market
(Chiswick, 1978):
--Lack of language skills,
--Absence of social networks,
--Less knowledge of customs,
--Lack of information about the local labour market,
--Lack of firm-specific training

Introduction-- Background
• However, economists (e.g. Chiswick,
1978; Borjas, 1991, 1994) hypothesise
that immigrants would be eventually
assimilated.
Because immigrants are:
-- more able and self-motivated
-- work harder and longer
--easy for them to transfer skills
Thus, immigrants’ income grows faster.
Source: Borjas (1994)

Introduction-- Motivation
• New Zealand is a country of immigrants.
• Before 1987: “Occupational Priority List” (OPL)
• 1987, “traditional source countries” was replaced by
stream approach
• Immigration Amendment Act 1991: “Point System”
• In 2003, it was supplanted by the “Skilled Migrant
Category” policy
• 2006 Census:
• 22.9% of usually residents are foreign born
• 18.9% of immigrants are Asian
• 14.4% of immigrants are Pacific Islanders

Introduction– New Zealand Case
(hourly earning)
Source: New Zealand Income Survey (2001-2010)

Introduction-- Motivation
• Assimilation processes of different ethnic groups happen in
diverse patterns and time ranges (e.g. Chiswick, 1978; Borjas,
1994; Poot, 1993; Winkelmann & Winkelmann, 1998; Card,
2005).
For example, Asian and Pacific Islander immigrants in
New Zealand experience a substantial income disadvantage.
Their assimilation process takes a longer time (e.g. Poot,
1993; Stillman and Mare, 2009; Stillman and Velamuri,
2010).

Introduction-- Motivation
• In addition, previous studies have assumed that
individuals’ labour market performance data is
independent and identically distributed (i.i.d.).
• However, one may consider whether individuals within
ethnic groups influence on each other and their labour
market performance is correlated to some extent.

Introduction-- Motivation
• Core questions:
• Why and how ethnicity is important for immigrants’
assimilation?
• How immigrants’ economic performances are linked
with each other?
• In order to answer these questions related to “ethnicity”
we use “ethnic capital” as a key concept to answer it and
to fill the literature gap.

Ethnic Capital
• Ethnic capital is a form of social capital, and it was
first put forward by Borjas (1992) in the context of
economics of immigration.
• Ethnic capital is the inherent trust and advantages
which stem from, and belong to, a certain ethnic group.
As a result, ethnic capital can offset those
disadvantages of immigrants to a certain extent.
• It can refer to the average skill level, the language
proficiency, the average income level, network and the
geographical concentration (ethnic enclave) for an
ethnic group.

Ethnic Capital-network effects
• Proxy for network for immigrants in previous studies:
• Ethnic Concentration/Enclave (e.g. Aguilera, 2009;
Damm, 2009; Edin, Fredriksson, and Aslund, 2003;
Toussaint-Comeau, 2008).
• Linguistic Concentrations (Bertrand, Luttmer, &
Mullainathan, 2000; Chiswick & Miller, 2002).

Ethnic Capital-network effects
• In this study, we construct a spatial network variable--
the endogenous “ethnic spatial lag” ;
• “Ethnic spatial lag” represents the individual’s network of
economic resources in addition to ethnic concentration
• To separate the spatial network specific effect
• We hypothesize that both ethnic networks and ethnic
concentration influence immigrants’ economic
performance.

Ethnic Capital-Ethnic Network
• Living in a common environment
• Producing shared experiences, knowledge, information
and other products through networks.
• Social networks -“the most profitable avenue of job
search” for immigrants (Frijters, Shields, & Price, 2005).
• Therefore, we hypothesize social networks may act
positively on the process of immigrants’ assimilation.

Ethnic Capital-Ethnic Concentration
• More jobs could be generated (“Protected Market”)
• Lowering barriers to employment for immigrants
• So, employment within the ethnic enclave very attractive; but reduces
the bargaining power of immigrants
• “Complements” vs. “Substitutes”
• Complement: higher demand for immigrant labour would be generated
• Substitute: compete for scarce employment opportunities
• Previous studies observed both positive (e.g. Edin, Fredriksson, &
Aslund, 2003) and negative (Lazear, 1999) effects of ethnic
concentration.

Models-Basic model
• Wi: earnings of individual i ;
• Xi: a vector of explanatory variables (e.g. years of
schooling completed, marital status, and years of
labour market experience);
• ti: years since migration to the host country;
• Ii: a dummy variable set to 1 if person i is an immigrant.
(Borjas, 1985).

Ethnic network- example
Waikato

Ethnic Spatial Matrix

Models-Spatial model (SAR)

Models-Spatial model (SAR)
• Under OLS estimation:
• Anselin (1990), Kelejian and Prucha (1998), and Kelejian
and Robinson (1993) proposed a Spatial Two-Stage Least
Squares (2SLS) approach to control the endogeneity of
spatial lag and estimate the parameter of spatial lag.
• The full set of instrumental variables is:

Data
New Zealand Income Survey
• The first release of NZIS was on 11 December 1997.
• NZIS contains income respondents’ information on selfemployment,
wages and salaries.
• A merged cross-sectional data set created from 2001 to
2010 NZIS data sets was produced for and adopted in this
study.
• Only observations of employed male immigrants and
natives aged 20 and 55 years old have been tested.
• New Zealand-born: 42,259
• Immigrants: 12,163

Data summary

Empirical Results
Table 4: OLS Estimates of Log Hourly Wage: Employed Male in New Zealand, age 20-
55, 2001-2010, NZIS
Pooled Sample
Natives
Foreign-Born
Human Capital
Potential Experience 0.033*** 0.033*** 0.037***
(0.001) (0.001) (0.002)
Potential Experience -squared -0.001*** -0.001*** -0.001***
(0.00002) (0.00002) (0.00004)
High School Qualification 0.179*** 0.181*** 0.174***
(0.009) (0.01) (0.025)
Vocational Qualification 0.193*** 0.193*** 0.191***
(0.004) (0.005) (0.011)
University Degree 0.492*** 0.524*** 0.423***
(0.005) (0.006) (0.011)
Other Post School Qualification 0.174*** 0.174*** 0.162***
(0.009) (0.011) (0.017)
Personal Characteristics
Years Since Migration (YSM) 0.006*** / 0.012***
(0.001) / (0.002)
YSM-squared -0.0001*** / -0.0002***
(0.00003) / (0.00003)
Married 0.116*** 0.125*** 0.079***
(0.004) (0.005) (0.01)
Arrived 1992-2003 -0.007 / 0.053***
(0.014) / (0.019)
Arrived 2003-2010 0.017 / 0.111***
(0.019) / (0.028)
Asia -0.226*** / -0.09***
(0.021) / (0.013)
Australia 0.029 / 0.163***
(0.022) / (0.014)
UK -0.038* / 0.087***
(0.02) / (0.012)
Pacific Islands -0.208*** / -0.084***
(0.02) / (0.011)
Rest of World -0.132*** / /
(0.02) / /
Observations 54422 42259 12163
Adjusted R-square 0.23 0.22 0.24
Note: (1) Standard errors in parentheses
(2) * p

Empirical Evidence-OLS estimations
• Immigrant assimilation in New Zealand is confirmed by OLS
estimations.
• The effect of YSM observed in this study is consistent with
previous New Zealand study (Winkelmann and Winkelmann
1998)
• But lower than the same effect in Australia (e.g. Chiswick
and Miller, 2002).
• The return of potential labour market experience is around
0.03 for natives and 0.04 for immigrants.
• Asian and Pacific Islander immigrants experience a
substantial income disadvantage.

Table 6: GMM Estimates of Log Hourly Wage: Employed Male Immigrants, age
20-55, 2001-2010, NZIS
Natives
Foreign-Born
(1) (2)
Ethnic Capital
Ethnic Network Effect 0.59*** 0.41*** 0.40***
(0.04) (0.04) (0.04)
Ethnic Concentration / *** -0.01
/ () (0.01)
Human Capital
Potential Experience 0.03*** 0.03*** 0.03***
(0.00088) (0.00189) (0.002)
Potential Experience -squared -0.001*** -0.001*** -0.001***
(0.00002) (0.0005) (0.00005)
High School Qualification 0.19*** 0.18*** 0.17***
(0.01) (0.03) (0.03)
Vocational Qualification 0.19*** 0.18*** 0.18***
(0.01) (0.01) (0.01)
University Degree 0.49*** 0.4*** 0.4***
(0.01) (0.01) (0.01)
Other Post School Qualification 0.18*** 0.17*** 0.17***
(0.01) (0.02) (0.02)
Personal Characteristics
Years Since Migration (YSM) / 0.01*** 0.01***
/ (0.002) (0.002)
YSM-squared / -0.0002*** -0.002***
/ (0.00004) (0.0004)
Married 0.13*** 0.07*** 0.07***
(0.01) (0.01) (0.01)
Arrived 1992-2003 / 0.06*** 0.06***
/ (0.02) (0.02)
Arrived 2003-2010 / 0.1*** 0.1***
/ (0.03) (0.03)
Asia / -0.06*** -0.05***
/ (0.02) (0.02)
Australia / 0.1*** 0.08***
/ (0.02) (0.02)
UK / 0.05*** 0.06***
/ (0.02) (0.02)
Pacific Islands / -0.04*** -0.04***
/ (0.01) (0.01)
Observations 42259 12163 12163
R-squared 0.24 0.27 0.27
Note: (1) Standard errors in parentheses
(2) * p

Table 7: GMM Estimates of Log Hourly Wage by Country of Origin: Employed Male
Immigrants, age 20-55, 2001-2010, NZIS
UK Australia Asia Pacific Islands
Ethnic Capital
Ethnic Network Effect 0.51*** 0.28** 0.32*** 0.27***
(0.16) (0.14) (0.09) (0.09)
Ethnic Concentration 0.01 0.15 -0.02 -0.01
(0.04) (0.12) (0.02) (0.01)
Human Capital
Potential Experience 0.04*** 0.05*** 0.03*** 0.03***
(0.004) (0.005) (0.004) (0.004)
Potential Experience -squared -0.001*** -0.001*** -0.001*** -0.001***
(0.0001) (0.00015) (0.0001) (0.0001)
High School Qualification 0.28*** 0.09 0.15*** 0.13***
(0.08) (0.09) (0.05) (0.04)
Vocational Qualification 0.14*** 0.15*** 0.13*** 0.23***
(0.4) (0.04) (0.03) (0.02)
University Degree 0.4*** 0.43*** 0.4*** 0.41***
(0.03) (0.04) (0.03) (0.03)
Other Post School Qualification 0.15*** 0.13*** 0.14*** 0.22***
(0.04) (0.06) (0.05) (0.04)
Personal Characteristics
Years Since Migration (YSM) 0.01*** -0.006 0.03*** 0.01**
(0.005) (0.007) (0.005) (0.004)
YSM-squared -0.0003*** 0.0001 -0.0004*** -0.0001
(0.00008) (0.00012) (0.0001) (0.0001)
Married 0.12*** 0.13*** 0.03 0.04
(0.02) (0.03) (0.03) (0.03)
Arrived 1992-2003 0.07 -0.04 0.05 0.1**
(0.05) (0.07) (0.05) (0.04)
Arrived 2003-2010 0.09 -0.1 0.26*** 0.08
(0.07) (0.11) (0.08) (0.06)
Observations 2464 1417 1466 2921
R-squared 0.25 0.20 0.27 0.27
Note: (1) Standard errors in parentheses
(2) * p

Conclusion
• The empirical findings of immigrant employed males confirmed the
effects of ethnic capital on their labour market performance in New
Zealand.
• Ethnic network play a positive role on any individual’s wage growth in
most cases.
• A stronger social network and linkage help immigrants to achieve
better economic performance and more successful assimilation.
• It seems that immigrants may need a much stronger social network
and linkage for their assimilation.
• We cannot observe the same significant and positive effect of ethnic
concentration on immigrants after controlled for ethnic network.

Conclusion
• Summarized and developed the theory of Ethnic Capital.
• New approach to construct ethnic network
• Separate the effect of network from ethnic concentration
• First trial in Labour aspect
• Examine the effect of ethnic capital in New Zealand
• Estimate the coefficient of ethnic network (spatial lag) by
large sample (54,422 observations); while current Stata
programme only able to estimate the sample with 3,000
observations.

Future research plan
• From this work, two more papers are proposed to be
generated:
• (1) Estimate the direct and indirect effect of ethnic
network on immigrants’ earning.
• (2) Estimate the Spatial Cliff-Ord Type Model by
Generalised Spatial Two Stage Least Square (GS2SLS)
Estimation.