Immigrant Skills and Immigrant Outcomes under a Selection System: The
Canadian Experience*
Aneta Bonikowska
Social Analysis Division
Statistics Canada
David A. Green
Department of Economics
University of British Columbia
W. Craig Riddell
Department of Economics
University of British Columbia
October 2010
* For helpful comments we thank Charles Beach, David Card, Doug Miller, Pat Grady
and seminar participants at UC Davis, the Goldman School of Public Policy, UC
Berkeley and the Canadian Economics Association meetings.
1
1. Introduction
In many developed countries immigration policy ranks as one of the most important, as
well as most hotly debated, economic and social policies. The importance attached to the
topic reflects, at least in part, the demographic challenges – aging populations and slowly
growing labour forces – that most high-income countries face, challenges that
immigration may help ameliorate. The salience of immigration also reflects the rise in the
fraction of the population that is foreign-born in high-income countries that traditionally
had not regarded themselves as major immigrant-receiving countries. Indeed, many
European countries now have foreign-born populations similar in size to those of
traditional immigrant-receiving countries such as Australia, Canada and the United
States. Intense debate over alternative approaches to immigration reflects both this
increase in the foreign-born population and the challenges that numerous societies have
faced in integrating first, second and third generation immigrants.
The growing significance of these issues has led many high-income, immigrantreceiving countries to develop more systematic immigration policies. One approach is a
formal assessment or “points” system such as that adopted by Canada in the 1960s and
Australia in the 1980s. The UK introduced a points system in 2002 and several other
European countries (e.g. Germany, Denmark and Spain) are considering doing so. This
approach has also been suggested by US immigration specialists (Chiswick, 1981; Borjas,
1999a) and was seriously considered by the previous U.S. congress.
Whether or not it involves an explicit points system, developing a more
systematic approach to immigration requires articulating the policy goals and the means
of achieving these objectives. The possible goals have many dimensions, especially if one
considers the well-being of individuals in both immigrant-sending and immigrantreceiving countries. Nonetheless, two objectives that typically receive considerable
weight are (i) maximizing the net economic benefits to the pre-existing residents of the
receiving country, and (ii) minimizing any adverse effects on the well-being of low
income residents of the receiving country. According to several economic models of the
consequences of in-migration, these objectives imply that receiving countries should
select high-skilled rather than low-skilled immigrants (Borjas, 1999b; Boeri and Brucker,
2005; Ruhs, 2008).
2
Canada represents the leading example of a country that has employed a formal
immigrant selection system for several decades. Although the nature of the Canadian
points system has evolved since its introduction in the late 1960s, the central objective of
selecting immigrants with characteristics appropriate for the Canadian labour market has
remained intact. From the outset -- and increasingly in recent decades -- the points system
has focused on selecting skilled immigrants. Furthermore, immigration policy has shifted
toward admitting more of the “economic class” (those selected under the points system)
and less of the family unification and refugee classes.
Despite the emphasis on selecting skilled immigrants, together with devoting
substantial public resources to the selection and admission systems, recent immigrant
cohorts have performed poorly in the Canadian labour market. After an initial period of
adjustment, immigrant cohorts arriving in the 1950s and 1960s typically did better in
terms of employment and earnings than their native-born counterparts. However, more
recent arrival cohorts have experienced larger earnings gaps on arrival relative to
otherwise comparable native-born workers, and often also slower rates of “catch-up” to
the Canadian born (Baker and Benjamin, 1994; Green and Worswick, 2002; Frenette and
Morissette, 2003; Aydemir and Skuterud, 2005). In addition, poverty has increased
substantially among the immigrant population, in contrast to the trends experienced by
natives, where poverty has generally declined (Picot and Hou, 2003). A common
explanation – one that drives much current policy development – is that immigrants have
the appropriate skills but their credentials are not recognized in Canada’s labour market.
Another explanation is that domestic employers discriminate against immigrants – that is,
pay immigrants less than equally productive native-born workers (Oreopoulos, 2009).
Although credential recognition obstacles and discrimination are the most
commonly offered explanations for the deteriorating performance of immigrants to
Canada, another possibility is that the actual skills of immigrants are limited, despite the
assessment system. The quality of formal education varies substantially across countries,
and even when the quality is high the content of schooling may lack relevance to other
societies and their labour market institutions. The quality and nature of work experience
in the country of origin may also produce skills that have limited relevance to employers
in the host country. Lack of host country language proficiency may result in immigrants
3
with skills in their home language that are not usable to the same extent in Canadian
workplaces. Investigating these issues would be straightforward if we had access to direct
measures of skill. Typically, however, researchers only observe inputs into skill
production like formal schooling and work experience – the same human capital inputs
that are observed by those selecting immigrants. In this paper, we take advantage of the
rich data provided by the Canadian component of the International Adult Literacy and
Skills Survey (IALSS) which includes both standard demographic and labour market
information for the native born and immigrants and results from tests of literacy,
numeracy and problem-solving skills. Using these direct measures of basic cognitive
skills, we provide new insights into the performance of immigrants in a country with
substantial experience with a “points-type” selection system. The combination of a large
sample size (over 24,000 observations) together with a sample drawn from a country with
one of the world’s largest proportions of foreign born allow analysis that would be less
informative, perhaps even infeasible, if applied to other countries that participated in the
IALSS survey.
The paper examines several issues related to immigrants' skills and labour market
outcomes. First, we compare the literacy and numeracy skills of immigrants to those of
the native born -- essentially asking whether “skilled immigrants” selected by a points
system are indeed highly skilled. Next, we investigate whether immigrant – native born
skill differences depend on country of origin as well as whether immigrant human capital
was acquired before or after arrival. Our data contain rich information that allow us to
distinguish foreign from Canadian work experience as well as to distinguish between
those who completed their education prior to arrival and those who completed their
highest level of education in Canada. The third issue examined is whether immigrants
receive different returns to these skills than observationally similar native-born workers.
This question addresses the hypothesis that low immigrant earnings reflect discrimination
in Canada’s labour market. Finally we ask whether differences in levels and returns to
basic literacy and numeracy skills can account for observed differences in earnings
between immigrant and native-born workers.
The results raise doubts about the extent to which a formal points system designed
to select “skilled immigrants” achieves its objectives. Literacy, numeracy and problem
4
solving skills of immigrants are substantially below those of observationally equivalent
native born Canadians. We also find strong evidence that skill differences depend on
country of origin as well as where human capital was acquired. Immigrants who
completed their education prior to arrival in Canada have significantly lower skills than
otherwise similar immigrants who obtained some or all of their education in Canada.
Regardless of these differences in skill levels and acquisition, however, we clearly reject
the hypothesis that immigrants receive lower returns to these basic cognitive skills than
the native born. Indeed, an important group of immigrants benefit more than do natives
from higher skill levels. This evidence argues against discrimination-based explanations
for differences in earnings between immigrant and native-born workers.
Our earnings analysis supports findings in earlier papers that returns to both
foreign-acquired education and experience for immigrants are lower than returns to
education and experience obtained in Canada by either immigrants or native-born
workers. Basic cognitive skills themselves exert a statistically significant and
quantitatively large effect on earnings. This estimated return to skills, together with the
lower skill levels of immigrants, explains a large part of the immigrant earnings
differential. We estimate that raising immigrants’ basic literacy and numeracy skills to
the native born level would almost eliminate the earnings disadvantage of high school
educated male immigrants relative to similarly educated native born men, and would
produce an earnings advantage among high school educated female immigrants. Among
the university educated, raising immigrant skills to the native born level would reduce the
earnings gap by about 75 percent.
The paper is organized as follows. The next section provides additional
background on the evolution of Canada’s immigration policy and on the economic
performance of immigrants to Canada. In the third section, we discuss our data and report
summary statistics. Section four examines whether immigrants have different skill levels
than the native born and, if so, why. The fifth section presents a framework for
considering what we might learn from introducing cognitive skills measures into a
standard earnings equation. Section six contains the analysis of the impacts of skills on
the earnings of immigrants and native-born Canadians. The final section concludes.
5
2. Immigration policy and immigrant outcomes: Canada’s experience
Prior to the 1960s the focus of Canada’s immigration policy was principally on unskilled
migrants. 1 Between the 1980s and World War I the stated goal was to secure farmers,
farm workers and female domestics to help settle the West. During the early post-World
War II period the desired immigrant was also essentially unskilled, needed to meet
growing labour demand in the resource sector, especially forestry and mining.
Throughout this time there was also a clear racist feature of immigration policy;
admission to Canada was mainly restricted to those from traditional source countries such
as the U.S., UK and Northern European countries. Both of these salient characteristics of
immigration policy began to change in the 1960s. In 1962 Canada abandoned the longstanding policy of preferred and non-preferred source countries in favour of a new regime
in which admission was to be based on the individuals’ personal characteristics,
especially their education and skill qualifications. Nonetheless, considerable discretion in
selecting immigrants remained in the hands of overseas immigration officers.
Subsequently, in 1967, a formal points system was adopted. The points system provided
an explicit scale for admission decisions based on factors such as age, education and
language, as well as the predicted demand for workers in the applicant’s intended
occupation and destination. This new policy regime represented a major step toward
reducing the powers of immigration officers and replacing these with more objective
criteria.
From the outset the points system applied to those seeking admission as part of
the “independent” or “economic” class, and not to those admitted as refugees or for
family unification reasons. Indeed, at the time the points system was introduced, the
refugee and family classes were given top priority for processing; thus admissions via the
assessed class were a residual category. During normal economic conditions this residual
category feature did not represent a serious constraint on the flow of economic migrants;
however, it did limit inflows in periods such as recessions when the overall level of
immigration was significantly reduced.
1
Green and Green (1999) provide a detailed account of the evolution of Canada’s immigration policy, with
emphasis on the economic goals.
6
Several noteworthy trends are evident in immigration policy over the past four
decades. First, although there is some year-to-year variation, there has been greater
emphasis over time on the selected independent class and less on the refugee and family
classes. For example, during the 1975-79 period 37% of arrivals were economic migrants
versus 43% in the family class, whereas during the 1995-99 period 54% of new
immigrants were selected under the points system (Picot, 2008). Second, the bar for
admission under the points system has increased over time – from 50 out of 100 points in
1967 to 76 points since the early 1990s. Third, there has been a move away from
assigning points for skills in specific occupations predicted to be in short supply and
toward assigning more points for human capital characteristics associated with long-term
labour market success such as education, experience and language proficiency. As a
consequence of these policy changes, recent immigrants to Canada (especially those
arriving in the past two or three decades) have increasingly been selected on the basis of
their expected economic contributions. New arrivals are also much more likely to be well
educated. For example, 19% of the 1975-79 arrival cohort had a university degree
whereas twenty years later 42% of the 1995-99 arrival cohort were university graduates
(Picot, 2008).
Despite the policy shift toward greater selectivity, immigrants’ labour market
outcomes have worsened. Indeed, when compared to native-born Canadians with similar
levels of education and experience (or age), the gap in earnings has been growing with
each successive arrival cohort, both at entry and after many years in Canada. Much
empirical research has been devoted to trying to understand the relatively poor
performance of immigrants in Canada’s labour market. One consistent finding is that the
change in the immigrant experience coincides with a major shift in source countries –
away from the U.S., the UK, and Western European countries and toward Eastern
Europe, Asia and Africa (Baker and Benjamin, 1994; Frenette and Morissette, 2003;
Aydemir and Skuterud, 2005). Associated with this shift is a large decline in the fraction
of immigrants with English or French as their home language or mother tongue. Recent
studies attribute about one-third of the decline in entry-level earnings of immigrants to
factors associated with the changing source country composition of immigrants (Frenette
and Morissette, 2003; Aydemir and Skuterud, 2005). However, there remains much to be
7
learned about the factors associated with source country – such as culture, language
proficiency, school quality and visible minority status – that are responsible for
deteriorating economic outcomes. Another noteworthy finding of recent research is the
declining contribution of foreign work experience to immigrant earnings (Green and
Worswick, 2002; Frenette and Morissette, 2003; Aydemir and Skuterud, 2005). For
reasons that are not well understood, labour market experience prior to arrival is more
heavily discounted for recent arrivals than was the case in the past. Green and Worswick
(2002), Frenette and Morissette (2003) and Aydemir and Skuterud (2005) conclude that
the decline in returns to foreign experience is a major factor contributing to lower
earnings among recent arrival cohorts – accounting for approximately one-third of the
growth in the earnings gap at entry. Finally, earnings of new entrants to the labour market
– both Canadian and foreign born – deteriorated during the 1980s and 1990s, especially
among males. Thus the earnings gap at entry is not as large when immigrants are
compared to Canadian born new entrants as when they are compared to the workforce as
a whole. Green and Worswick (2002), Frenette and Morissette (2003) and Aydemir and
Skuterud (2005) conclude that this factor can account for up to 40% of the growth in the
immigrant-native born earnings gap. However, unlike the other potential contributors
discussed above, this factor was concentrated in the 1980s and 1990s, especially during
the 1980s.
The Canadian and international literature also points to language proficiency as a
key component of the immigrant assimilation experience. Using a variety of approaches
to address potential endogeneity and measurement error issues, papers by Chiswick
(1991), Chiswick and Miller (1995), Dustmann and Fabbri (2003), and Berman, Lang,
and Siniver (2003) find substantial effects of host country language acquisition on
immigrant earnings. Fluency in the host country language can be viewed as a skill in its
own right and/or as an input to the generation of other skills. In the latter case, employers
care only about the usable amounts of each skill that a worker possesses. Thus, an
engineer who is well trained but cannot communicate with his or her employer or fellow
employees would be counted as having zero usable engineering skills. Language ability
in French or English may thus enter as an input into the production of usable skills, with
greater language ability leading to higher usable skills for any given level of other inputs.
8
Both Chiswick (1991) and Dustmann and Fabbri (2003) include self-reported reading
skills along with host country fluency in earnings regressions, interpreting the reading
and speaking fluencies as separate skills. Chiswick (1991), using a sample of illegal
immigrants to the US, finds that reading fluency has a much stronger effect on earnings
than speaking fluency when both are included. Dustmann and Fabbri (2003), using UK
immigrant data, find that reading fluency is a more important determinant of employment
but speaking fluency is a more important determinant of earnings. Following these and
other authors, we control for language proficiency in our analysis.
This paper also builds on an earlier analysis of the role of literacy skills to the
earnings of immigrants and native born (Ferrer, Green and Riddell, 2006). In that paper
we combined data on the Canadian born from the 1994 International Adult Literacy
Survey with data on immigrants from the 1998 Ontario Immigrant Literacy Survey to
address some of the issues examined here with much better data. The way in which the
current paper extends and improves upon our earlier work is discussed subsequently.
3. Data and Summary Statistics
The dataset we use in this investigation is the International Adult Literacy and
Skills Survey (IALSS), the Canadian component of the Adult Literacy and Life Skills
Survey (ALL). Statistics Canada carried out this survey in 2003 to study the skills of
Canadians. The IALSS data contain the results of literacy, numeracy and problem-solving
tests as well as information on labour market variables such as income, education and
labour force status. The survey covers individuals age 16 and over, and this is also the
age range we focus on in our analysis. We drop individuals who list their main activity as
“student” in order to focus on the effect of completed schooling and what happens
subsequently to individual skills and labour market outcomes. We also drop aboriginals,
whose schooling, skills and labour market outcomes are distinctly different from those of
other Canadians. Because our key outcome measure is individual earnings (wages or
salary), we restrict the sample to those employed at the time of the survey. We also
exclude the self-employed and workers with weekly earnings that are less than or equal to
$50 and over $20,000. The latter restriction cuts out a small number of workers with
earnings that are significant outliers relative to the rest of the sample. We exclude the
9
self-employed because we wish to assess the way skills are rewarded in the labour
market, and earnings from self-employment reflect both returns to skills and returns to
capital. Finally, we drop observations with missing information on age at arrival or
earnings. Although much of the immigration literature focuses on males, we analyse both
male and female immigrants. The final sample size is 9,555, of which 1,787 are
immigrants. We use the sample weights in our analysis, so all summary statistics and
regression estimates are nationally representative. In the regression analysis, when
computing standard errors for the coefficient estimates we also use the Jackknife replicate
weights provided in the data set to account for the complex survey design.
Our principal labour market outcome is weekly earnings. In the IALSS
respondents are first asked about their standard pay period and then asked about typical
earnings in that pay period. Using these responses we construct a weekly earnings
measure for each paid worker. Thus, for example, in the case of individuals who report
that they are paid monthly we divide their usual monthly earnings by 4.333.
The main objective, and major advantage, of the IALSS survey is to provide
measures of four skills: Prose literacy, Document literacy, Numeracy and Problem
Solving. The test questions do not attempt to measure abilities such as those in
mathematics and reading but rather try to assess capabilities in applying skills to
circumstances that arise in everyday life. Thus, the Document questions, which are
intended to assess capabilities in locating and using information in various forms, range
from identifying percentages in categories in a pictorial graph to assessing an average
price by combining several pieces of information. The Numeracy component ranges from
simple addition of pieces of information on an order form to calculating the percentage of
calories coming from fat in a Big Mac based on a nutritional table. Thus, the questions
are related to implementation and use of skills in the real world and are intended not just
to elicit current capacities but also adaptability to answering questions in other contexts
(Murray, Clermont and Binkley, 2005). 2 This is an important point for the interpretation
of our results since we interpret the test results as revealing job relevant skills at the time
2
The IALSS builds on the IALS survey that was carried out in several countries during the period 1994 to
1998. Two of the skill domains – prose literacy and document literacy – are defined and measured in the
same manner in IALS and IALSS.
10
of the interview rather than inherent abilities. It is worth emphasizing that these skills are
essentially cognitive in nature.
A salient feature of the data is the strong correlation among the various cognitive
skill measures. The correlation between the Prose literacy and Document literacy scores
is 0.96, that between Prose literacy and Numeracy is 0.90, and the correlation between
Prose literacy and Problem Solving is 0.93. Further, a principal components analysis
indicated only two principal components with the first being vastly more important and
placing equal weight on all four scores. Thus a simple average of the four scores captures
much of the information available in the skill measures. This is the skill measure that we
use in much of the analysis.
The other main variables in our analysis are standard human capital measures plus
variables related to language ability in English or French. The survey asked respondents
their number of years of completed schooling, so we are able to construct the standard
Mincer measure of potential experience (i.e., age – years of schooling – 6). Since we
know the age at which immigrants entered Canada, we are able to divide experience of
immigrants into foreign experience and Canadian experience. 3 We measure educational
attainment using the respondent’s years of completed schooling. 4 As mentioned earlier, a
major advantage of the IALSS data is its detailed questions on where immigrants
obtained their education. In particular, respondents are asked about their total years of
completed schooling as well as their years of schooling completed outside of Canada.
The survey also asks whether the respondent completed his/her highest level of education
in Canada. We make use of this information in what follows by dividing our analysis
between immigrants who completed their highest level of schooling in Canada versus
those who completed it abroad. This turns out to be an important distinction and is one
that cannot be made very precisely in data sets that are widely used for analysis of
immigration such as the Census.
Our regression models include several other variables. We create a dummy
variable that equals one if the first language spoken is other than English or French. We
3
Foreign experience = age at arrival – foreign years of schooling – 6 if positive, zero otherwise.
Canadian experience = age – age at arrival – (total years of schooling – foreign years of schooling) if
positive and age at arrival >=6. Canadian experience = age – total years of schooling – 6 if age at arrival<6.
4
Use of highest level of education attained, which is also available in the IALSS data, produces very
similar results.
11
also include dummy variables corresponding to the region of origin. One variable
corresponds to immigrants from the U.S. or UK, a second to continental Europe, and a
third to immigrants from Asia, with the rest of the world forming the omitted category. 5
Much of the earlier literature on immigrants indicates that there are strong source country
effects and that immigrants from the U.S. and UK adapt particularly well to the Canadian
economy. The IALSS survey also asked respondents about their mother’s and their
father’s education, and we include parents’ highest education as a covariate in some of
our regressions. Finally, we include two variables as proxies for innate ability. These are:
a dummy variable equalling one if the respondent agreed or strongly agreed with the
statement that they received good grades in math when they were in school and a dummy
equalling one if the respondent agreed or strongly agreed with the statement that math
teachers often went too fast and the person often got lost.
Table 1 displays summary statistics for the main variables of interest. Both male
and female immigrants are, on average, three years older and, correspondingly, have
three more years of labour market experience than their native-born counterparts.
Immigrant men report one more year of schooling than do native-born men, but
immigrant women report the same years of schooling as Canadian born women. This
gender difference in the immigrant – native born educational attainment gap is also
evident when we look at the highest level of education attained. Among males, the
distribution of formal education among immigrants is very different than, and generally
superior to, that for the native born. The fraction of native-born men with no
postsecondary education is 53%, versus 41% among immigrants. Additionally, a much
larger fraction of male immigrants has a university degree (35%) compared to native born
men (18%). For women the patterns of educational differences between immigrants and
natives are similar but the magnitudes of the gaps are much smaller. The fraction of
native-born women with no post-secondary education is 49%, somewhat higher than that
for immigrant women (44%) although the fraction of immigrant women that did not
complete high school (14%) is higher than that of native-born women (12%). At the top
of the educational distribution, a larger proportion of immigrant women has a university
5
We also examined finer breakdowns of the source country but these had little effect on the results.
12
degree (30%) than is the case for native-born women (22%), but the gap between
immigrants and natives is much smaller than that for males.
Parental education of immigrant and native-born Canadians is more similar than
the education levels of the sons and daughters. The similarity is especially striking for
men. Female immigrant women are more likely than Canadian born women to have
parents that are high school dropouts, although the likelihood of having at least one
parent with a university degree is about the same. Although immigrants overall have
parents with broadly similar educational attainment to parents of the Canadian born, the
parental education of immigrants educated abroad is substantially below that of
immigrants who completed their education in Canada.
Another evident difference between immigrants and natives is their levels of basic
cognitive skills, as assessed in English or French. The average skill levels of immigrant
men range from 251 to 265, whereas these range from 285 to 293 for Canadian-born men.
The largest gaps between immigrants and the Canadian-born are in prose literacy and
problem solving and the smallest are in numeracy. Across the four skill domains, male
immigrant skill levels are about 10% lower than those of native-born males. The skill
gaps are even larger for females. As is true for men, the largest gaps are in prose literacy
and problem solving and the smallest in numeracy. Across the four domains, female
immigrant skill levels are about 12% below those of natives.
An interesting fact arising from Table 1 is the substantial fraction of immigrant
workers who acquired their highest completed education level in Canada. Columns 1 and
2 separate immigrants between those who report obtaining their highest level of
schooling in Canada and those who acquired it abroad. It is immediately apparent that
these two groups have very different characteristics. Both male and female immigrants
with Canadian education are much younger, have less work experience, but more
experience in the Canadian labour market. They arrived in Canada at a younger age, and
have been in the country longer, despite being considerably younger than immigrants
educated abroad. In terms of educational attainment, immigrants with Canadian education
have similar years of completed schooling to immigrants without Canadian education, but
a different distribution of educational attainment. Those who completed their schooling in
Canada are less likely to be high school dropouts, but also less likely to be university
13
graduates. Relative to immigrants without Canadian education, they are more represented
in the middle of the educational distribution, making their distribution of educational
attainment more similar to that of native-born Canadians.
Perhaps the most striking differences are those relating to cognitive skills.
Immigrants with Canadian education have skill levels somewhat below those of the
native born, but much higher than those of immigrants without Canadian education. The
skills gap between foreign-educated immigrants and natives is especially large for
women. Relative to native-born men, the gap in average skills for those educated outside
of Canada ranges from 14% for numeracy to 17% for prose literacy and problem solving,
whereas for male immigrants with some Canadian education the average skills gap ranges
from 3% for numeracy to 6% for problem solving. Among female immigrants educated
in Canada the skills gap is somewhat larger than for men – ranging from 6% for
numeracy to 9% for prose literacy and problem solving. Among female immigrants
educated prior to arrival the skills gap is about two times as large – between 14% and
18% across the four skill domains. These differences suggest that controlling for the
origin of education may indeed be important for understanding immigrant labour market
outcomes. They also suggest that there may be gender differences in immigrant outcomes
relative to those of natives.
The apparent advantage in education and experience among immigrant men does
not translate into higher income. Average annual and weekly earnings of native-born men
are higher than those of immigrants without Canadian education, and the gap in median
earnings is even greater. In contrast, mean earnings of immigrants with Canadian
education exceed earnings of the native born, although median earnings are modestly
lower than those of natives. Among women both mean and median earnings of foreigneducated immigrants are below those of natives, whereas mean and median earnings of
Canadian-educated immigrants exceed those of natives.
One possible explanation for the puzzle of lower earnings of immigrants without
Canadian education, despite their generally reporting more experience and education, is
that the Canadian labour market may place a different value on the experience and
education obtained outside of Canada than on that obtained after arrival in Canada.
Another possible explanation for lower earnings is that the skills of immigrants educated
14
abroad are much lower than those of native Canadians, despite their higher levels of
educational attainment and greater amount of total labour market experience. We explore
both of these explanations further in what follows.
4. Immigrant – Native Born Skill Differences
(a) Are “skilled” immigrants high-skilled?
Figures 1(a) and 1(b) plot the kernel density functions of the individual averages
of the four cognitive skill scores for males and females, respectively. 6 For both men and
women the cumulative distribution function (CDF) for native born scores lies to the right
of the immigrant CDF throughout the sample range and stochastically dominates the
immigrant CDF at conventional significance levels. 7 The differences between immigrants
and natives are especially large at the low end of the skill distribution. For example,
among women the immigrant-native born gap is over 50 points at the 10th percentile, 38
points at the median, and 25 points at the 90th percentile. The skill gaps for men are
smaller but follow the same pattern – a differential of 42 points at the 10th percentile, 30
points at the 50th percentile, and 14 points at the 90th percentile.
The remaining panels of the figure show the skills distributions for immigrants
with and without Canadian education relative to that of the native born. The skill
distributions of both immigrant groups are inferior to those of the Canadian born, and the
difference between the respective distributions is largest for immigrants educated outside
of Canada. 8 There is also less dispersion in the cognitive skills of immigrants who
completed their education in Canada than is the case for those educated outside of
Canada. In addition, for immigrant men educated in Canada the upper tail of the
distribution is similar to that for Canadian-born men, whereas this is not the case for male
immigrants who were educated prior to arrival in Canada, nor is it the case for either
6
We estimate the kernel density functions with the kdensity function in Stata, using the Epanechnikov
kernel and Stata’s default bandwidth formula.
7
More precisely, we cannot reject the null hypothesis that the native born CDF first order stochastically
dominates the immigrant CDF at conventional significance levels, using the test for first order stochastic
dominance described in Barrett and Donald (2003).
8
For example, the median skill score of female immigrants without Canadian education is 54 points below
that of natives, while that of female immigrants with Canadian education is 16 points below that of natives.
The story for males is similar: a differential in the median skill score of 45 points between foreign-educated
immigrants and native Canadians versus a differential of 17 points between Canadian-educated immigrants
and natives.
15
group of female immigrants. For both men and women a noteworthy difference between
immigrants who obtained some or all of their education in Canada and those that did not
is the relative absence of individuals with high skill levels in the latter group.
Plots of the distributions of the four individual skills provide further evidence of
differences in the distribution of skills among native Canadians and the two immigrant
groups. 9 For both men and women the immigrant distributions are clearly inferior to
those of the native born. The immigrant-native born skill gaps are most evident for prose
literacy and least evident for numeracy. This latter pattern may reflect the tendency for
numeracy to be less language dependent. As was the case with the average scores,
separating immigrants into two sub-samples delineated by where they obtained their
education reveals substantial differences between the two groups. For each of the four
cognitive skills and both genders the distributions for immigrants educated in Canada
have lower dispersion than those for foreign-educated immigrants and proportions of
individuals with high skill levels that are closer to those observed among the native born.
This similarity at the top of the skill distribution is especially evident for males. Both
immigrant groups have larger proportions of their respective distributions with low skill
levels (scores below 200) than is the case for the Canadian born. This concentration in the
lower tail of the distribution is especially pronounced for immigrants who completed
their education prior to arriving in Canada.
(b) Why are immigrant skills so low?
The IALSS data allows us to investigate potential sources of low immigrant skills.
To do so we examine the determinants of individual literacy, numeracy and problem
solving skills, with particular focus on the relationship between these basic skills and
human capital variables like education, language proficiency, parental characteristics and
work experience. Table 2 reports the results from regressions of the average skill score on
these human capital variables plus source country (for immigrants) and province of
schooling for the Canadian born (not reported).10 The dependent variable is the log of the
9
These figures, which are not shown to conserve space, are available on request.
Education falls under provincial jurisdiction in Canada, and there are important differences in the quality
of elementary and secondary schooling across provinces. For Canadian born the IALSS survey provides
information on the province in which the respondent attended secondary school. When this information is
missing we use province of residence.
10
16
average skill score, so the coefficients can be interpreted as showing the percentage effect
of a unit change in the variable of interest on the average score.
Because the determinants of basic skills are remarkably similar for males and
females in all three groups, regression results are reported for the pooled male and female
samples. 11 We allow the effects of the human capital variables to differ across the three
groups – native born, immigrants with Canadian education, and immigrants who
completed their education prior to arrival. For immigrants the regressions also allow the
effects of foreign and Canadian work experience on skills to differ. In the regression
reported in column 1 the omitted category consists of native-born Canadians whose first
language is English or French. Relative to these natives, the skills of immigrants with
Canadian education are 9% lower, after controlling for other factors, although the
estimate associated with this immigrant dummy is imprecise and not significantly
different from zero. Foreign-educated immigrants have much lower skills than the
reference group – in the order of 29% lower. This estimate is statistically significant.
The experience profiles reported in column 1 are similar for all three groups. In
each case the partial correlation between experience and skills is zero or even slightly
negative. The same result holds if we use Age rather than experience as a covariate. Note
that this finding continues to hold when we add additional controls in columns 2 and 3.
These results suggest that among adults (i.e. after about age 20) there is essentially no
relationship between experience and the individual’s skills. This conclusion is consistent
with Green and Riddell (2003) who find that years of experience are essentially
uncorrelated with the individual’s skill level across various specifications in the IALS
data that is predominantly made up of adult workers.
On the other hand, there is a strong relationship between education and cognitive
skills for all three groups. All groups also display diminishing returns to additional
schooling, which is not unexpected given the basic nature of the skills measured in the
IALSS. The impact of formal schooling on literacy and numeracy skills is very similar
for the Canadian-born and Canadian-educated immigrants – gains, respectively, of 4.8%
and 4.4% associated with each additional year of schooling, beginning at low levels of
11
For all three groups (Canadian born, immigrants with Canadian education, foreign-educated immigrants)
we can’t reject the hypothesis that the impacts of mother’s and father’s education are the same (p-values of
0.5 or higher).
17
education. The returns to additional education diminish somewhat more rapidly for the
native born. For example, at 12 years of schooling the marginal impact of an additional
year of schooling equals 2.8% for immigrants with Canadian education versus 2.6% for
the Canadian born. In contrast, the educational gradients are much steeper for immigrants
who obtained their education abroad – estimated skill gains of over 7.5% associated with
each additional year of schooling at low education levels, falling to a marginal impact of
3.6% for an extra year at 12 years of schooling.
Among the Canadian born, having one’s first language being other than English
or French has no impact on literacy, numeracy and problem solving skills. Similarly, for
immigrants with Canadian education the effect of first language being other than English
or French is small and not statistically significantly different from zero. However, among
immigrants who completed their education prior to arrival, having a first language other
than one of Canada’s two official languages has a large negative effect on these basic
skills. Such differences could arise because exposure to English and French in the school
system and in daily activities offsets any disadvantage for those who spend all or part of
their lives in Canada.
Noteworthy differences in basic skills are evident among immigrants from
different source countries even after controlling for observed human capital variables
(education, experience, language). Those who emigrated from the US or UK have
average literacy and numeracy skills that are about 9% higher than the omitted source
region (rest of the world), while those who arrived from continental Europe have 6%
higher skill levels. Immigrants from Asia have somewhat higher skills (+3%) but the
difference between their skills and the omitted category is not statistically significant.
OLS estimates of the relationship between basic cognitive skills and educational
attainment are likely to suffer from omitted variables bias due to unobserved factors such
as innate ability that influence both education and skills. We deal with this issue in two
ways. First, we control for parental education, a potentially important variable that is not
available in many data sets. Parental education plays two potential roles. First, it controls
to some extent for unobserved ability due to the correlation between parent’s ability
(reflected in parental education) and child’s ability. In addition (and perhaps more
important quantitatively) parental inputs may directly influence the child’s basic
18
cognitive skills, and well-educated parents may provide more and/or higher quality inputs
into skill production as well as exert an influence on the child’s education. Because the
father’s and the mother’s education are highly correlated we use as a control the
education level of the parent with the highest educational attainment. 12 These results are
reported in the second column of Table 2. In addition, the column 3 estimates include the
two proxies for unobserved ability discussed previously.
Parents’ education is positively associated with the child’s literacy and numeracy
skills, controlling for the child’s own education, but the estimated impacts differ across
the three groups and are not continuous over the distribution of parental educational
attainment. One key factor is having a father or mother with at least a high school
education (the omitted category). When both parents did not complete high school, basic
skills are 11% lower among foreign-educated immigrants and 4% lower for the Canadian
born. The estimated effect for immigrants with Canadian education is also negative but
small and not statistically significant. Having at least one parent with post-secondary
education is associated with higher skill levels in some cases, but the gains are modest
and vary across the three groups. Interestingly, not knowing (or not reporting) both
parents’ education is associated with substantially lower skills for all groups. While we
included this variable in order to not lose those observations and keep the sample size the
same as elsewhere, this variable may be picking up real influences – such as being
orphaned, not being raised by one’s parents, or not having a close relationship with one’s
parents. Whatever the underlying reason, not knowing one’s parents’ education is
associated with basic skills that are about 7% lower for the Canadian born and 15%-20%
lower for immigrants.
The addition of controls for parental education has some noteworthy effects on
the estimates in column 1. The most striking change is that to the immigrant dummies.
The coefficient associated with immigrants educated prior to arrival drops by almost 40%
and is no longer statistically significant. Differences in parental education between these
immigrants and the Canadian born evidently account for a large part of the skills gap
12
Including both mother’s and father’s education produces similar results to those reported in Table 2, but
the coefficients are less precisely estimated. We also tested the hypothesis that the impacts of mother’s and
father’s education on the child’s skills are the same, and could not reject this hypothesis for all three
groups.
19
between these groups. At the same time, with the addition of parental education controls
the coefficient on the dummy for immigrants with Canadian education becomes smaller
(more negative) and similar in size to that associated with immigrants without Canadian
education. With parental education controls the coefficient on years of schooling
declines, as expected, for the native born. However, the drop is modest in size – about
7%. In contrast, the coefficient on years of schooling declines by about 12% for
immigrants without Canadian education, while that for the other immigrant group
actually rises. Also noteworthy is the fact that the estimated coefficients for gender and
experience are not affected by the addition of parental education controls. However, those
for source region and language change decline, albeit modestly.
The final column adds the proxies for ability. Both coefficients have the expected
signs, and the combined impact is moderately large (about 7%). Adding these ability
proxies results in further small declines in the coefficients on years of schooling, as well
as those for source region, for all three groups. The language impacts, however, are as
large in column 3 as in column 1.
In order to provide additional insight into the relative importance of these
influences we performed an Oaxaca decomposition of the skills gap between the native
born and the two immigrant groups. The raw skills differential is approximately 7% for
immigrants with Canadian education and 20% for those without. For both immigrant
groups the differences in estimated coefficients account for most of the skill differential.
Indeed, for foreign-educated immigrants the coefficient differences account for 99% of
the gap. Key coefficients are those on language and the immigrant dummy, which
together account for more than the 20% skills differential, while higher returns to
schooling for foreign-educated immigrants operates in the opposite direction. In the case
of immigrants with Canadian education, differences in coefficients account for more than
100% of the skills gap, while differences in individual characteristics contribute to
narrowing the differential. For this group, both higher educational attainment and having
parents with higher education contribute to narrowing the gap, while the immigrant
dummy is a key coefficient difference operating in the opposite direction.
In summary, for all three groups there are substantial gains in skills associated
with higher educational attainment. The skill gains associated with higher education are
20
greatest for immigrants without Canadian education, and similar in size for the native
born and immigrants with Canadian education. Combining the negative immigrant
dummies with the positive effects of higher education, additional education narrows the
skills gap between foreign-educated immigrants and the Canadian born. Parental
education also exerts an important influence on basic literacy and numeracy skills, and
differences in the education levels of parents help to understand the immigrant – native
born skills gaps. Indeed, once we control for parental education the skills gaps between
natives and the two immigrant groups are similar in size and are not statistically
significant. A key factor accounting for the much larger skills differential associated with
foreign-educated immigrants is language. Having one’s first language other than English
or French does not impact negatively on the skills of the native born or immigrants with
Canadian education, but it has a large negative impact on literacy and numeracy skills of
foreign-educated immigrants. Notice that in obtaining these results we control for region
of origin and that immigrants from the U.S. or U.K. as well as continental Europe do not
face as large a skills disadvantage. These differences in skills could arise because of
differences in the quality of education across source countries or because of other
differences across source countries that are not fully captured in our covariates.
4. A Framework for Discussing Earnings Generation
This section outlines a simple framework for analysing the role of basic cognitive
skills in earnings generation. A key distinction is that between skills (attributes that are
acquirable) and abilities (those that are innate). In this taxonomy, skills include cognitive
skills such as literacy and numeracy and non-cognitive skills such as persistence and
conscientiousness. Each worker potentially possesses a range of skills and can possess
each of them in varying amounts. To simplify the exposition the discussion is couched in
terms of three sets of skills. Individual earnings are determined according to some
function of the skills an individual possesses and puts into use, as follows:
1) E i = f ( G1i , Gi2 , Gi3 ) + ε i
21
where Ei are earnings for individual i, Gik is the amount of skill k that person i supplies in
the market, and εi is a disturbance term that is independent of the skills. The disturbance
term captures either idiosyncratic events that are independent of the skill levels or
measurement error in earnings. The earnings generation function f(.) can be interpreted as
being derived from a production function that is separable in other (non-skill) inputs. This
function provides information about the relative importance of the various skills in
production and whether the skills are complements or substitutes. To focus ideas, we will
think of G1 as cognitive skills of the type measured in literacy tests, G2 as other (perhaps
manual) skills that are not captured in such tests and might be acquired through work
experience, and G3 as non-cognitive characteristics such as persistence that might be
partly acquired through schooling.
Based on 1), we can construct a set of skill price functions given by:
2) r k =
∂f
( G1i , G i2 , G i3 ),
∂ Gk
k = 1,...,3
Note that these implicit prices may depend on the complete bundle of attributes the
individual supplies.
Characterizing either 1) or 2) would be straightforward if we observed the skills,
Gik. Typically, of course, researchers do not observe them. What they often do observe
are some of the inputs used to generate skills. To see how they enter our framework,
consider a set of production functions for generating the skills:
3) Gik = hk ( yrsi , expi , θ i )
where k indexes the attribute type, yrs corresponds to years of formal schooling, exp is
years of experience in the work force and θ is a vector of innate abilities. The vector of
abilities may include both cognitive and non-cognitive elements. For example, noncognitive abilities such as persistence could help generate both non-cognitive and
cognitive skills.
22
If we do not observe the Gik's directly, we can obtain an estimating equation by
substituting equations 3) into 1). This yields a reduced form specification for earnings
given by:
4) Ei = g( yrsi , expi , θ i ) + ε i
In this hierarchical model covariates commonly used in wage regressions are inputs into
skill production and these skills (plus an independent error term) determine earnings.
Now, let us examine the partial derivatives of earnings with respect to each of the
skill production inputs. The partial derivative associated with one of the inputs, x, can be
expressed as:
5)
∂E ∂f
∂ h1
∂f
∂ h2
∂f
∂ h3
=
*
+
*
+
*
1
2
3
∂x ∂ G
∂x
∂G
∂x
∂G
∂x
where the i subscript is suppressed for simplicity. Thus, if x corresponds to years of
schooling, equation 5) says that the observed effect of an additional year of schooling
reflects the effects of an extra year of education on the production of each skill times the
price paid for that skill. It is apparent from equation 5) that with measures only of
earnings and observable inputs used in producing skills, we cannot make any statements
about skill production or how skills combine in production. However, if we have
individual observations on one or more skills, we can potentially say much more.
With G1 observed, our quasi-reduced form earnings function becomes:
6) E i = g * ( G1i , yrsi , expi , θ i ) + ε i
The derivative of this function with respect to G1 corresponds to the implicit price
function, r1 - though, now the implicit price is a function of yrsi, expi, and θi :
7) r 1 = χ ( G1i , yrsi , expi , ai )
With the price function given in 7), we cannot fully specify the interactions of G1,
G2 and G3 in production but we can learn more about them. In particular, the derivatives
23
of r1 with respect to the skill production input, x, is equal to (again suppressing the i
subscript):
8)
∂ h3
∂ r1 ∂ r1
∂ h2
∂
=
*
+ r 13 *
2
∂x ∂ G
∂x
∂G
∂x
Further, we can consider the derivative of g* in equation 6) with respect to x:
9)
∂g *
∂f
∂ h2
∂f
∂ h3
=
*
+
*
2
3
∂x ∂ G
∂x
∂G
∂x
With observed values for the derivatives, δr1/δx and δg*/δx, we may be able to place
restrictions on the f (.) function. Thus, the fact that δh2/δx and δh3/δx appear in all these
functions raises the possibility of putting restrictions on the production functions for the
non-observed skills based on sign and significance patterns in the observed derivatives.
With these restrictions in hand, we may also be able to place restrictions on the δr1/δGk
terms. These latter terms reflect interactions in production of the non-observed skills with
G1.
We can also learn something about the production of G1 from the differences
between the derivatives 5) and 9). These derivatives (e.g., the derivative of earnings with
respect to schooling first not conditioning and then conditioning on G1) differ by the term
δf/δG1 * δh1/δx. Thus, the difference between these observed derivatives reflect the
extent to which the coefficient on, for example, schooling in a standard earnings
regression reflects the channel of added schooling generating added earnings through
added cognitive skill creation. Given that we observe G1 directly, we can go further and
derive insights into the production of G1 (as reflected in the δh1/δx terms) through direct
estimation. This is what we do in the next section of the paper.
Note that, as expressed in equation 7), the skill price facing an individual will be a
function of the ability vector θi. The way we deal with these unobserved ability factors is
described in the next section.
6. The Effect of Education and Cognitive Skills on Earnings
24
In this section, we estimate earnings regressions with and without controlling for
individual skills. The dependent variable is the log of weekly earnings. As a first step, we
estimate a specification that includes years of schooling, a quadratic in experience, a
dummy for immigrant status, a quadratic in years since entering Canada for immigrants,
and a dummy for first language other than English or French. 13 Because cognitive skills,
schooling and earnings are likely to be jointly determined we also include the two proxies
for unobserved ability discussed previously. Apart from the ability proxies this
specification is similar to immigrant – native born earnings equations estimated with
cross-sectional data that have been reported in previous studies. 14
Results without cognitive skill variables
The first column in Tables 3a (males) and 3b (females) presents these results.
They reflect commonly observed patterns. In particular, male returns to experience are
approximately 7% per year just after leaving school but decline to zero after 30 years. As
is typically found to be the case, female returns to experience are lower, 5.6% per year
early in the career, declining to zero 27 years later. There are also substantial returns to
education that are on the order of those found in earlier studies, with women experiencing
much higher returns to schooling than men. Male immigrants receive weekly earnings
that are 47% less than earnings of native-born workers with the same level of total
experience and education. For female immigrants the magnitude of this negative entry
effect is lower, but the gap is still substantial – over 30 percent. Immigrant earnings then
rise at rates of approximately 2.0% (males) and 2.2% (females) more per year compared
to similar native-born workers in the years just after the immigrant enters Canada. As
indicated by the negative coefficients on the years-since-migration (YSM) squared
variables, this rate of catch-up to the native born diminishes over time. If male and female
immigrant earnings actually follow these "years since migration" profiles, then their
earnings would equal those of a comparable native-born worker after 25-30 years in
Canada. This, however, is a big “if”. As Borjas (1985) points out, if immigrants arriving
in different years (i.e., in different cohorts) face different entry earnings and/or years
since migration earnings profiles, then a cross-sectional years since migration profile will
13
We also estimated this equation including a quadratic term in years of schooling, and found no evidence
of increasing or decreasing returns to education.
14
Although not reported, all earnings regressions include controls for province of residence.
25
represent a combination of actual profiles and the effects of shifts across cohorts. Thus,
the cross-sectional profile is not necessarily the relevant earnings assimilation profile for
any set of immigrants. With only a single cross-section of IALSS data, there is no way to
address this problem. The immigrant dummy variable and years since migration profile
summarize a combination of cohort effects and assimilation profiles rather than a profile
that bears behavioural interpretation. Since our focus is on effects of cognitive skills
rather than cohort patterns, this is not a central concern. It is only important that we
control for the combination of cohort and assimilation effects, not that we can separately
identify them. 15
Having first spoken a language other than English or French does not have a
statistically significant effect on earnings of men or women. Both of the proxies for
unobserved ability have the expected sign although only the “teachers went too fast”
variable is statistically significant. This variable suggests that low ability males earn
about 15% less than others and low ability females about 5% less. Including these
variables has a small impact on the other estimated coefficients, reducing the estimated
impact of schooling on earnings by about 5%.
The specification in column 1 imposes equal returns to education and experience
for immigrants and the native born but allows immigrants to have separate entry earnings
and an earnings progression with years since arrival. However, the latter YSM effects can
be difficult to interpret even in the absence of the cohort effect complication just
described. For individuals arriving in Canada after they have completed their education,
YSM corresponds to experience in the Canadian labour market. For individuals
completing their education in Canada, YSM will equal years of experience in the
Canadian labour market plus the number of years between arrival and entry into the
labour market. Since the latter years may include time when the migrant is quite young,
their impact on earnings is likely quite different from that of labour market experience.
For that reason, we implement an adjusted specification (reported in column 2) that
allows the immigrant entry effects and Canadian experience effects to differ between
immigrants who arrive after completing their education and immigrants who obtain some
15
See Green and Riddell (2007) for an analysis of cohort and aging effects among native-born Canadians
using the 1994 IALS data and the 2003 IALSS data, and Willms and Murray (2007) for an analysis of skill
loss and gain over time.
26
or all of their education in Canada. Differences between these two groups of immigrants
in the coefficients on Canadian experience variables could represent some combination of
differential returns to experience and differential cohort effects.
The adjusted specification in column 2 includes both the separate immigrant
experience variables described above and dummy variables for source regions discussed
previously. We include these variables because previous studies have placed a great deal
of emphasis on region of origin effects in explaining immigrant earnings patterns (e.g.,
Baker and Benjamin, 1994). In interpreting the estimates reported in column 2 note that
the various experience coefficients are reported so that they can be read directly rather
than as comparisons to, say, the Canadian experience variables.
Although the results for males and females share many common features, there
are also some noteworthy gender differences. We therefore discuss the male results
(Table 3a) and female results (Table 3b) separately. The estimated coefficients relating to
the Canadian experience of male immigrants who obtained some of their education in
Canada and the overall male experience coefficient (which corresponds mainly to the
experience effects for the native born) are similar in size, and tests of the hypothesis that
they are equal to each other cannot be rejected at conventional significance levels. In
contrast, male immigrants without Canadian education receive significantly greater
returns to Canadian work experience. The intercept coefficients for the two groups of
immigrants are both negative and significantly different from zero. The implication from
these coefficients is that immigrants who complete their education abroad have earnings
that are almost 60% lower than comparable native-born workers, whereas those with
some Canadian education receive earnings that are about 24% lower than otherwise
comparable natives. These estimates apply to the base category, those whose first
language spoken was French or English and who are not from the US, the UK, Europe, or
Asia. For those whose first language was other than English or French, average weekly
earnings are not significantly lower. Finally, the source region coefficients suggest that
immigrants from continental Europe have earnings that are 27% higher than those of
other immigrants. Immigrants from the US/UK also receive higher earnings than the base
group, although these estimated effects are smaller than for Europe (about 15%) and are
not precisely estimated. Immigrants from Asia have earnings that are slightly higher than
27
the base group, although this effect is also not statistically significant. As was the case in
column 1, those who report that “teachers went too fast” when in high school have
earnings that are 14% lower than otherwise comparable workers.
Table 3b reports the results for females. As was the case for men, the negative
entry effect experienced by foreign-educated immigrants is much larger than that for
immigrants with Canadian education. However, both entry effects are smaller (in absolute
value) than the corresponding estimates for male immigrants, and neither coefficient is
significantly different from zero. Another feature common to both male and female
immigrants is the result that immigrants experience higher returns to Canadian experience
than do natives. However, in contrast to the results for men, female immigrants with
Canadian education obtain higher returns to Canadian experience than those obtained by
their foreign-educated counterparts. The differences between the two immigrant groups,
however, are not statistically significant, even at the 10% level. Having a first language
other than English or French does not exert a statistically significant impact on earnings,
as was the case for men. However, the earnings gaps associated with alternative source
countries are very different. After controlling for other factors, female immigrants from
the US/UK, Europe and Asia do not experience higher earnings than those from other
source regions, in contrast to the situation for male immigrants.
The adjusted basic specification is still, potentially, too restrictive. In particular, it
restricts the returns to foreign experience (in terms of earnings in Canada) to be the same
as returns to Canadian experience for the native born. The specification in the third
column of Table 2 permits a separate return to foreign experience. This is important
because Friedberg (2000) finds, using Israeli data, that negative immigrant entry earnings
effects can be completely explained by a lower return to foreign experience than native
experience. For immigrants from some countries, she found that foreign experience was
worth zero in the Israeli labour market. These results are replicated for Canada by Ferrer,
Green and Riddell (2006). Green and Worswick (2002) study this further and show that
this is a recent phenomenon for Canada since immigrant cohorts in the early 1980s
earned returns on foreign experience that were similar to the returns the native born
earned for Canadian experience. Similar to results in those papers, when we introduce
foreign experience variables in column 3, the immigrant entry effect coefficients are no
28
longer significantly different from zero; indeed, three of the four estimated coefficients
are positive. At the same time, for both men and women the returns to Canadian
experience for the two immigrant groups are not significantly different from those for the
native born (although the returns differ between native-born and foreign-educated
immigrant women at the 10% level). Finally, note that introducing the foreign experience
effect does not change the returns to education, language impacts, and country of origin
effects.
Among both male and female immigrants the return to foreign experience itself is
essentially zero. It is this low rate of return on foreign experience that is the source of the
negative immigrant entry effects in the first two columns of the table. Comparing
immigrant earnings to those of native-born workers with the same total number of years
of experience shows that immigrant earnings are significantly lower. This occurs because
the immigrants are obtaining zero returns to some of those years of experience. Once we
control for foreign experience, we are effectively comparing immigrants to native born
workers with the same number of years of Canadian experience and it turns out that
immigrant and native-born workers have earnings that are much more similar when
compared on that basis. This does not negate the fact that immigrants have lower
earnings. However, it does help us understand that a major source of those lower earnings
is an inability to transfer human capital acquired in a foreign labour market to Canada. It
is worth noting, as well, that foreign experience does not suffer from the same
interpretation difficulties as Canadian experience for immigrants. That is, there is no
cohort dimension to the number of years an immigrant worked before arriving.
Immigrants arriving in recent cohorts and cohorts from decades ago could all have the
same distribution of foreign experience before arriving. The same is not true of Canadian
experience: those arriving in earlier cohorts necessarily have more. This means that we
can give the coefficient on foreign experience a standard human capital acquisition
interpretation much as we have given to Canadian experience. 16
16
However, Green and Worswick (2002) point out that native born earnings can also be organized in a
cohort format and that doing so provides insights into the cross-cohort patterns in immigrant cohorts. In
particular, they find that approximately 60% of the cross-cohort decline in immigrant earnings in the 1980s
can be attributed to general declines across cohorts of new entrants of all kinds into the Canadian labour
market.
29
The final column of Table 2 contains our preferred specification which we reach
by first allowing a complete set of interactions among all immigrant, experience and
education variables and then eliminating sets of interactions where testing indicates it is
appropriate. Thus, for example, we allowed for different returns to education for
immigrants who obtained their highest level of education in Canada. For both males and
females we could not reject the restriction that the differences between these returns and
those for the native born were zero at any conventional significance level. However, we
do find that returns to education are significantly lower for male and female immigrants
without Canadian education. We also allowed for the possibility that each type of
experience (whether foreign or Canadian-acquired) might interact with each type of
education. We do find evidence of significant interactions of Canadian experience with
education for the native born and immigrants who obtained their highest level of
education after arrival. These interaction coefficients are negative for both men and
women. Thus among the native born and immigrants with Canadian education the
positive impact of experience on earnings diminishes as educational attainment rises.
However, there is no evidence of similar interactions between experience and education
for immigrants educated before arrival.
Results with cognitive skill variables
In Table 4, we use the preferred specification from Table 3 but include the
average skill score. A comparison of the first column in Tables 4a (males) and 4b
(females), where we simply add the skill variable, and the last column in Tables 3a and
3b respectively reveals the direct impact of cognitive skills and their indirect impacts on
other returns. The returns to skills are substantial, with a 100-point increase in cognitive
skills raising earnings by almost 30 percent. 17 The impact of skills on earnings is
remarkably similar for men and women. As in Green and Riddell (2003), there is little, if
any, change in the experience effects or experience interactions when we control for
skills. However, estimated returns to education for natives and Canadian-educated
immigrants decline to a significant extent, indicating that an important component of
conventional estimates of the return to schooling arises from the impact of education on
17
However, these estimated returns to skills are lower than those obtained in our previous research for
native-born workers with the IALS data (Green and Riddell, 2003) and for immigrants with the OILS data
(Ferrer, Green and Riddell, 2006). These differences warrant further investigation.
30
skills and the value placed on skills in the labour market. Because of the interaction with
experience, the magnitude of the decline depends on the amount of Canadian experience.
At low levels the coefficient on years of schooling drops by over 15% for men and 10%
for women. For native-born Canadians with substantial experience the decline is much
larger in percentage terms; at 20 years of experience, the return to an extra year of
schooling falls by 34% for men and 14% for women. With the inclusion of controls for
cognitive skills, estimated returns to education for foreign-educated immigrants decline
even more than was the case for the native born. Indeed, for immigrant men returns to
education fall by almost 40%; for immigrant women the decline is even greater, and, after
controlling for skills, the remaining returns to education are no longer significantly
different from zero. Thus, basic literacy and numeracy skills constitute a significant
amount of what foreign education seems to deliver – at least in terms of skills that are
valued by Canadian employers.
As discussed earlier, a major question of interest is whether returns to skills are
lower for immigrants. To investigate this issue we report in column 2 estimates based on
a specification that allows the returns to skills to differ between immigrants and natives,
but does not allow interactions between skills and human capital inputs like education
and experience. In the absence of such interaction effects differences in the coefficient on
the average skill measure between immigrants and the native born can be interpreted as a
measure of discrimination. The estimates in column 2 provide no evidence of
discrimination in the sense of immigrants receiving a lower return to cognitive skills.
Indeed, male immigrants receive a return that is greater than that experienced by nativeborn men, although the differences are not statistically significant. Among women
immigrants receive returns similar in size to (and not significantly different from) those
of natives.
In column 3 we report a more general specification that allows returns to skills to
differ across the three groups. Among males, foreign-educated immigrants receive the
largest returns to skills, followed by immigrants with Canadian education, and natives
receive the lowest (although still substantial) returns. However, a test of the hypothesis
that all three coefficients equal each other cannot be rejected at normal confidence levels
(indeed, even at the 10% level). Among females, foreign-educated immigrants also
31
experience the greatest returns, followed by native Canadians, with Canadian-educated
immigrants receiving the lowest returns. The three coefficients are also not significantly
different from each other at the 10% level.
Controlling for cognitive skills, and allowing the impact of skills on earnings to
differ across the three groups, has a noteworthy effect on the parameter estimates for
foreign-educated immigrants. For both men and women this group has the highest
estimated returns to skills (according to the point estimates), and clearly receives returns
to skills that are at least as great as those received by the other groups. However, their
estimated returns to formal schooling remain much smaller than for others. Indeed, for
female foreign-educated immigrants the coefficient on years of schooling is small in
magnitude, negative and not significantly different from zero. The implication is that
education acquired abroad produces cognitive skills such as literacy, numeracy and
problem solving skills (since the educational attainment coefficients change substantially
with the introduction of the skill variable) but produces little (in the case of men) or
nothing (in the case of women) in the way of other skills that are valued in the Canadian
labour market (since the educational attainment coefficients are small or not significantly
different from zero once we control for skills).
In summary, when we allow the impacts of skills on earnings to differ across the
three groups, we find no evidence that the returns to skills for natives exceed those for
Canadian-educated or foreign-educated immigrants. Indeed, among foreign-educated
immigrants the earnings gains associated with additional skills are larger in size (although
not statistically significantly different) than those for native Canadians and Canadianeducated immigrants. Thus these estimates provide no evidence of discrimination in the
sense of employers paying immigrants less for the same skills as native-born workers. It
is worth emphasizing that this result refers to what we call “usable” skills. Immigrants
may have higher cognitive skill scores if tested in their native language and one could
argue that those skills are being undervalued. But immigrants are receiving returns to
skills as measured in English or French that are no worse than those obtained by nativeborn workers.
One interesting question arising out of these estimates is the relative importance
of lower immigrant skill levels in explaining immigrant-native born earnings
32
differentials. To investigate this, we constructed a series of counterfactual estimates, all
based on the last column in Table 4. We first estimate average log earnings for
immigrants and the native born separately using the estimated coefficients in conjunction
with the appropriate average values for the regressors. 18 Those estimates imply an overall
average immigrant earnings disadvantage of 14 log points over the native born among
high school educated men (12 years of schooling) and essentially no difference between
native-born and immigrant high school educated women. The corresponding estimates
among those with university education (16 years of schooling or more) imply an
immigrant disadvantage of 16 log points for men and 15 log points for women. We next
repeated this exercise but gave immigrants the same return to foreign experience as the
native born receive for their Canadian experience. The result is a shift from the
immigrant disadvantage of 14 log points to an advantage of 52 log points for high school
educated men – a net change of 66 percentage points in the earnings gap – and a shift
from no difference to an immigrant advantage of 34 log points for high school educated
women – a net change of 34 percentage points. Among the university educated the
immigrant disadvantage changes from 16 log points to an immigrant advantage of 48 log
points for men – a net change of 64 percentage points in the earnings gap, and from an
immigrant disadvantage of 15 log points to an advantage of 14 log points for women, a
net change of 29 percentage points. These estimates fit with results in earlier papers,
described above, indicating that lower returns to foreign experience play an important
role in understanding immigrant-native born earnings differentials, especially for men.
The importance of low returns to foreign experience is much more important for men
than for women, but for each gender is similar across education groups.
In our next counterfactual, we set the returns to foreign experience back to their
original values but gave immigrants the average skill scores observed for native-born
workers with the same level of education. For high school educated men, this turns the
immigrant disadvantage from the 14 log points mentioned above to an advantage of 2 log
points, a net change of 16 percentage points, and among high school educated women it
creates an immigrant advantage of 10 log points, a net change of 10 percentage points.
18
We constructed fitted average earnings separately for the two immigrant groups; the estimates for
immigrants as a whole are weighted averages of the fitted earnings for immigrants with and without
Canadian education.
33
For the university educated, it reduces the immigrant disadvantage from 16 log points to
3 log points for males and from 15 log points to 4 log points for females, net changes of
13 and 11 percentage points respectively. Again, the changes in the earnings differential
are similar across the two education groups. Low skills thus appear to be an important
factor for understanding male immigrant earnings differentials, though not as important
as low returns to foreign experience. To a smaller extent this is also true for women.
6. Conclusions
Immigration ranks as a major economic and social policy issue in many high
income countries. The growing importance of immigration is leading many countries to
adopt more systematic immigration policies. One such approach is a formal points system
as pioneered by Canada beginning in the 1960s and adopted by Australia in the 1980s.
Although the design of immigration policy clearly depends on the objectives society
wishes to achieve, most economic models imply that such policy should aim to select
highly skilled workers. Canada represents a leading example of this approach. Since the
introduction of the points system in the 1960s, Canadian immigration policy has focused
on selecting skilled workers. Furthermore, during the past two decades there has been
increased emphasis on immigration via the selection system vis-à-vis other avenues
(refugees and family unification). Despite these trends, and the resources devoted to
selecting individuals with attributes suitable for the Canadian labour market, recent
immigrants to Canada have been performing poorly relative to native born Canadians.
The earnings gap on arrival between immigrants and otherwise comparable Canadian
born workers has been growing, and this larger initial earnings differential is not being
offset by a higher rate of “catch-up” to earnings levels of comparable Canadian born.
This paper takes advantage of the rich data provided by the Canadian component
of the International Adult Literacy and Skills Survey (IALSS) to provide new insights
into the performance of immigrants in a country with substantial experience with a
“points-type” selection system. The combination of direct measures of basic literacy,
numeracy and problem solving skills, a large sample size, and a sample drawn from a
country with one of the world’s largest proportions of foreign born allow analysis that
would be less informative, perhaps even infeasible, in most other countries. Our data also
34
contain rich information that allow us to distinguish foreign from Canadian work
experience as well as to distinguish between those who completed their education prior to
arrival and those who completed their highest level of education in Canada.
The analysis yields several findings related to immigrants' skills and labour
market outcomes. First, basic literacy, numeracy and problem solving skills of
immigrants are substantially below those of native born Canadians. The average raw
skills gap is about 7% for immigrants with Canadian education and 20% for foreigneducated migrants. Furthermore, these differences are evident throughout the skills
distributions. The native-born skill distributions first-order stochastically dominate the
distributions for immigrants.
Second, we find strong evidence that skill differences depend on where human
capital was acquired. Immigrants who completed their education prior to arrival in
Canada have significantly lower skills than otherwise similar immigrants who obtained
some or all of their education in Canada. Indeed, the latter group is in many respects
more similar to the native born than to foreign-educated immigrants.
These skills differences are not a reflection of differences in education levels or
impacts of education on skill formation across natives and the two immigrant groups.
Indeed, for all three groups there are substantial gains in skills associated with higher
educational attainment. The skill gains associated with higher education are greatest for
immigrants without Canadian education, and similar in size for the native born and
immigrants with Canadian education. Thus additional education narrows the skills gap
between foreign-educated immigrants and the Canadian born. Parental education also
exerts an important influence on basic literacy and numeracy skills, and differences in the
education levels of parents help to understand the immigrant – native born skills gaps.
Indeed, once we control for parental education the skills gaps between natives and the
two immigrant groups are similar in size and are not statistically significant.
A key factor accounting for the much larger skills differential associated with
foreign-educated immigrants is language. Having one’s first language other than English
or French does not impact negatively on the skills of the native born or immigrants with
Canadian education, but it has a large negative impact on literacy and numeracy skills of
foreign-educated immigrants. In obtaining these results we control for region of origin
35
and immigrants from the U.S. or U.K. as well as continental Europe do not face as large a
skills disadvantage. These differences in skills could arise because of differences in the
quality of education across source regions or because of other differences across
countries that are not fully captured in our covariates.
Regardless of these differences in skill levels and acquisition, however, we clearly
reject the hypothesis that immigrants receive lower returns to these basic cognitive skills
than the native born. Indeed, an important group of immigrants benefit more than do
natives from higher skill levels. This evidence argues against discrimination-based
explanations for differences in earnings between immigrant and native-born workers.
Our earnings analysis supports findings in earlier papers that returns to both
foreign-acquired education and experience for immigrants are lower than returns to
education and experience obtained in Canada by either immigrants or native-born
workers. Basic cognitive skills themselves exert a statistically significant and
quantitatively large effect on earnings. This estimated return to skills, together with the
lower skill levels of immigrants, explains a large part of the immigrant earnings
differential. We estimate that raising immigrants’ literacy and numeracy skills to the
native born level would almost eliminate the earnings disadvantage of high school
educated male immigrants relative to similarly educated native born men, and would
produce a substantial earnings advantage among high school educated female
immigrants. Among the university educated, raising immigrant skills to the native born
level would reduce the earnings gap by about 75 percent.
36
References
Aydemir, A. and M. Skuterud “Explaining the deteriorating entry earnings of Canada’s
immigrant cohorts, 1966-2000” Canadian Journal of Economics 38 (May 2005) 641-671.
Baker, M. and Benjamin, D. "The Performance of Immigrants in the Canadian Labor
Market" Journal of Labor Economics 12 (July 1994) 369-405.
Barrett, G. and S.G. Donald. “Consistent Tests for Stochastic Dominance” Econometrica
71 (January 2003) 71-104.
Berman, E., K. Lang and E. Siniver. "Language-skill complementarity: returns to
immigrant language acquisition" Labour Economics 10 (June 2003) 265-290.
Bloom, D., Grenier, G. and Gunderson, M. "The Changing Labour Market Position of
Canadian Immigrants" Canadian Journal of Economics 12 (November 1995) 987-1005.
Boeri, T. and H. Brucker. “Why are Europeans so Tough on Migrants?” Economic Policy
(October, 2005) 629-703.
Borjas, G.J. "Assimilation, Changes in Cohort Quality and the Earnings of Immigrants".
Journal of Labor Economics 3 (October 1985) 463-89.
Borjas, G.J. "Assimilation and Changes in Cohort Quality Revisited: What Happened to
Immigrant Earnings in the 1980s?" Journal of Labor Economics 13 (April 1995a) 201245.
Borjas, G.J. “The Economic Benefits from Immigration” Journal of Economic
Perspectives 9 (No. 2, 1995b) 322-.
Borjas, G.J. Heaven’s Door: Immigration Policy and the American Economy. Princeton:
Princeton University Press, 1999a.
Borjas, G. J. “The Economic Analysis of Immigration” in Handbook of Labor Economics
Volume 3A, edited by O. Ashenfelter and D. Card. Amsterdam: Elsevier, 1999b.
Bowles, S., H. Gintis and M. Osborne. "The determinants of earnings: a behavioral
approach" Journal of Economic Literature 34 (December 2001) 1137-1176.
Chiswick, B.R. "The Effect of Americanization on the Earnings of Foreign Born Men",
Journal of Political Economy 86 (1978) 897-921.
Chiswick, B.R. “Guidelines for the Reform of Immigration Policy” in W. Fellner (ed.)
Essays in Contemporary Economic Problems. Washington: American Enterprise
Institute, 1981.
37
Chiswick, B.R. "Speaking, Reading, and Earnings among Low-skilled Immigrants,”
Journal of Labor Economics 9(2) (1991) 149-170.
Chiswick, B.R. and P.W. Miller "The Endogeneity between Language and Earnings:
International Analyses", Journal of Labor Economics 13(2) (1995) 246-288.
Dustmann, C. and F. Fabbri. "Language Proficiency and Labour Market Performance of
Immigrants in the UK" Economic Journal 113 (July 2003) 695-717.
Ferrer, A. and W. C. Riddell. "Education, Credentials and Immigrant Earnings" Canadian
Journal of Economics 41 (February 2008) 186-216.
Ferrer, A., D. Green and W.C. Riddell. “The effect of literacy on immigrant earnings”
Journal of Human Resources 41 (Spring 2006) 380-410.
Frenette, M. and R. Morissette. "Will they ever converge? Earnings of immigrant and
Canadian-born workers over the last two decades" Statistics Canada, Analytical Studies
Research Paper No 215, 2003.
Friedberg, R. "You can't take it with you? Immigrant Assimilation and the Portability of
Human Capital", Journal of Labor Economics 18 (April 2000) 221-251.
Grant, M. "Evidence on new immigrant assimilation in Canada", Canadian Journal of
Economics 32 (August 1999) 930-955.
Green, A.G. and Green, D.A. “The Economic Goals of Canada’s Immigration Policy:
Past and Present” Canadian Public Policy 25 (December 1999) 425-51.
Green, D.A. and W.C. Riddell. “Literacy and Earnings: An Investigation of the
Interaction of Cognitive and Unobserved Skills in Earnings Generation,” Labour
Economics 10 (April 2003) 165-84.
Green, D.A. and W.C. Riddell. Literacy and the Labour Market: The Generation of
Literacy and Its Impact on Earnings for Native Born Canadians. Ottawa: Statistics
Canada, 2007.
Green, D.A. and C. Worswick. “Earnings of Immigrant Men in Canada: The Roles of
Labour Market Entry Effects and Returns to Foreign Experience” Study prepared for
Citizenship and Immigration Canada, 2002.
Murray, T. Scott, Yvan Clermont and Marilyn Binkley. Measuring Adult Literacy and
Life Skills: New Frameworks for Assessment. Ottawa: Statistics Canada, 2005.
Oreopoulos, P. “Why Do Recent Immigrants Struggle in the Labour Market? A Field
Experiment with 3,000 Resumes” Working Paper, 2009.
38
Picot, G. “Immigrant Economic and Social Outcomes in Canada: Research and Data
Development at Statistics Canada” Statistics Canada, Analytical Studies Research Paper
No 319, 2008.
Picot, G. and Hou, F. “The Rise in Low-Income Rates Among Immigrants in Canada”
Statistics Canada, Analytical Studies Research Paper No 198, 2003.
Picot, G. and A. Sweetman. "The Deteriorating Economic Welfare of Immigrants and
Possible Causes” Statistics Canada, Analytical Studies Research Paper No 262, 2005.
Ruhs, M. “Economic research and labour immigration policy” Oxford Review of
Economic Policy 24 (No. 3, 2008) 403-426.
Schaafsma, J. and Sweetman, A. "Immigrant Earnings: Age at Immigration Matters".
Canadian Journal of Economics 34 (November 2001) 1066-99.
Statistics Canada. Reading the Future: A Portrait of Literacy in Canada. Ottawa, 1996.
.Willms, J.D. and T.S. Murray. Gaining and Losing Literacy Skills Over the Lifecourse.
Ottawa: Statistics Canada, 2007.
39
Figure 1
Distribution of Average Skill Score among Immigrants and Canadian Born
Women
.008
Density
.004 .006
.002
0
0
.002
Density
.004 .006
.008
.01
(b) Native born and immigrants
Men
.01
(a) Native born and immigrants
0
100
200
300
Average Skill Score
Native Born
400
500
0
100
Immigrants
200
300
Average Skill Score
Native Born
400
500
Immigrants
Women
.008
Density
.004 .006
.002
0
0
.002
Density
.004 .006
.008
.01
(d) Native born and immigrants with Canadian education
Men
.01
(c) Native born and immigrants with Canadian education
0
100
200
300
Average Skill Score
Native Born
400
500
0
100
Immigrants
200
300
Average Skill Score
Native Born
400
500
Immigrants
Women
.008
Density
.004 .006
.002
0
0
.002
Density
.004 .006
.008
.01
(f) Native born and immigrants with no Canadian education
Men
.01
(e) Native born and immigrants with no Canadian education
0
100
200
300
Average Skill Score
Native Born
400
Immigrants
500
0
100
200
300
Average Skill Score
Native Born
400
Immigrants
500
40
Table 1(a) Summary Statistics for Immigrant and Native Born Workers – Men
Immigrant
Annual Earnings
Mean
Median
Weekly Earnings
Mean
Median
Hours
Weeks
Age
Experience
Canadian
Foreign
Years of Schooling
% Less than HS
% HS
Foreign
Canadian
% Non-Univ PS
Foreign
Canadian
% University
Foreign
Canadian
% with Highest Parental Education
Less than HS
High School
Non-University Post-Secondary
University
Unknown
Years Since Migration
Age at Immigration
% with First Language not English or French
% with US or UK Origin
% with European Origin
% with Asian Origin
% Did Not Complete Main Skill Tasks
% Good Math Grades
% Teachers Too Fast
Prose Literacy
Document Literacy
Numeracy
Problem Solving
Observations
Native Born
Cdn Educ
No Cdn Educ
All
49,460
40,000
39,147
31,200
44,060
35,100
44,484
41,000
973
795
41
49
38
17
14
4
15
11
32
32
27
27
31
31
780
626
42
50
44
24
13
12
14
14
25
25
22
22
38
38
-
872
702
42
50
41
21
13
8
14
12
29
13
15
24
12
13
35
20
15
895
808
42
49
38
18
13
18
35
29
18
-
26
22
24
25
4
23
15
64
15
22
27
17
75
21
273
282
280
267
39
27
15
17
2
13
31
79
9
18
40
29
75
22
241
249
250
236
33
25
19
21
3
18
23
72
12
20
34
23
75
21
256
265
264
251
30
29
19
18
3
5
13
73
23
289
293
289
285
416
501
917
3,634
Sample: Men, age 16 and older, excluding students and observations with missing information on age at
immigration and with weekly earnings greater than $50 and less than or equal to $20,000.
41
Table 1(b) Summary Statistics for Immigrant and Native Born Workers – Women
Immigrant
Annual Earnings
Mean
Median
Weekly Earnings
Mean
Median
Hours
Weeks
Age
Experience
Canadian
Foreign
Years of Schooling
% Less than HS
% HS
Foreign
Canadian
% Non-Univ PS
Foreign
Canadian
% University
Foreign
Canadian
% with Highest Parental Education
Less than HS
High School
Non-University Post-Secondary
University
Unknown
Years Since Migration
Age at Immigration
% with First Language not English or French
% with US or UK Origin
% with European Origin
% with Asian Origin
% Did Not Complete Main Skill Tasks
% Good Math Grades
% Teachers Too Fast
Prose Literacy
Document Literacy
Numeracy
Problem Solving
Observations
Native Born
Cdn Educ
No Cdn Educ
All
33,091
29,900
27,545
25,000
30,401
27,040
31,178
27,196
747
577
37
49
40
20
16
4
14
12
33
33
29
29
27
27
546
485
36
49
43
24
14
9
14
16
26
26
24
24
34
34
-
649
525
37
49
41
22
15
6
14
14
30
13
17
27
12
15
30
16
14
641
540
35
49
38
19
14
12
37
29
22
-
41
17
21
18
3
25
14
63
16
18
27
15
70
29
274
273
260
262
42
24
14
18
2
15
28
77
12
21
41
31
69
23
246
247
239
237
41
20
18
18
2
20
21
70
14
20
34
23
69
26
260
260
250
250
33
26
22
17
2
6
11
68
33
300
294
278
289
437
433
870
4,134
Sample: Women, age 16 and older, excluding students and observations with missing information on age at
immigration and with weekly earnings greater than $50 and less than or equal to $20,000.
42
Table 2 Log Skills Score Regressions
Cdn Exp (Native Born)
Cdn Exp2/100 (Native Born)
Immigrant (Educ after arrival)
Cdn Exp (Immig. Cdn Educ)
Cdn Exp2/100 (Immig. Cdn Educ)
Foreign Experience (Immig. Cdn Educ)
Foreign experience squared/100 (Immig. Cdn Educ)
Immigrant (Educ before arrival)
Cdn Exp (Immig. Foreign Educ)
Cdn Exp2/100 (Immig. Foreign Educ)
Foreign Experience (Immig. no Cdn Educ)
Foreign experience squared/100 (Immig. no Cdn Educ)
Years of schooling (Canadian born)
Years of schooling squared/100 (Canadian born)
Years of schooling (immigrant, Canadian education)
Years of schooling squared/100 (immigrant, Canadian education)
Years of schooling (immigrant, no Canadian education)
Years of schooling squared/100 (immigrant, no Canadian education)
First language other than Eng/Fr
(Canadian born)
First language other than Eng/Fr
(immigrant, Canadian education)
First language other than Eng/Fr
(immigrant, no Canadian education)
Female
US or UK origin
European origin
Asian origin
(1)
-0.0004
(0.0006)
-0.0022
(0.0014)
-0.0892
(0.1327)
0.0016
(0.0017)
-0.0059
(0.0042)
-0.0105**
(0.0038)
0.0161
(0.0148)
-0.2871***
(0.0960)
-0.0066**
(0.0029)
0.0137*
(0.0077)
-0.0019
(0.0036)
-0.0014
(0.0109)
0.0477***
(0.0060)
-0.0009***
(0.0002)
0.0448***
(0.0134)
-0.0007*
(0.0004)
0.0765***
(0.0077)
-0.0017***
(0.0002)
-0.0030
(0.0128)
-0.0140
(0.0302)
-0.1347***
(0.0284)
-0.0068
(0.0044)
0.0880***
(0.0243)
0.0643**
(0.0234)
0.0280
(0.0245)
(2)
0.0010
(0.0006)
-0.0039**
(0.0014)
-0.1477
(0.0975)
0.0034**
(0.0016)
-0.0077*
(0.0038)
-0.0094***
(0.0034)
0.0150
(0.0123)
-0.1752
(0.1036)
-0.0055*
(0.0031)
0.0122
(0.0075)
-0.0009
(0.0037)
-0.0047
(0.0120)
0.0444***
(0.0058)
-0.0008***
(0.0002)
0.0465***
(0.0085)
-0.0008***
(0.0003)
0.0672***
(0.0073)
-0.0015***
(0.0002)
0.0092
(0.0119)
-0.0107
(0.0302)
-0.1294***
(0.0304)
-0.0056
(0.0039)
0.0796***
(0.0245)
0.0627***
(0.0218)
0.0213
(0.0236)
(3)
0.0011
(0.0007)
-0.0043***
(0.0015)
-0.1466
(0.0925)
0.0034**
(0.0016)
-0.0075*
(0.0038)
-0.0089**
(0.0035)
0.0134
(0.0135)
-0.1625
(0.1010)
-0.0055*
(0.0030)
0.0120
(0.0073)
-0.0010
(0.0037)
-0.0042
(0.0122)
0.0431***
(0.0057)
-0.0008***
(0.0002)
0.0452***
(0.0082)
-0.0008***
(0.0003)
0.0637***
(0.0074)
-0.0015***
(0.0002)
0.0086
(0.0117)
-0.0202
(0.0292)
-0.1354***
(0.0290)
-0.0008
(0.0039)
0.0784***
(0.0240)
0.0580**
(0.0221)
0.0198
(0.0232)
43
Table 2 continued
Highest parental education - native born
Less than high school
Non-university post-secondary
University
Education unknown
Highest parental education - immigrants - Canadian educated
Less than high school
Non-university post-secondary
University
Education unknown
Highest parental education - immigrants - foreign educated
Less than high school
Non-university post-secondary
University
Education unknown
-0.0368***
(0.0068)
0.0103*
(0.0058)
0.0328***
(0.0070)
-0.0672***
(0.0156)
-0.0356***
(0.0063)
0.0112**
(0.0053)
0.0360***
(0.0065)
-0.0692***
(0.0152)
-0.0198
(0.0280)
0.0790**
(0.0316)
0.0539*
(0.0284)
-0.1479***
(0.0481)
-0.0188
(0.0271)
0.0833***
(0.0289)
0.0532*
(0.0270)
-0.1358***
(0.0468)
-0.1089***
(0.0241)
-0.0745*
(0.0392)
0.0080
(0.0244)
-0.1928***
(0.0394)
-0.1060***
(0.0253)
-0.0731*
(0.0381)
0.0044
(0.0251)
-0.1814***
(0.0399)
0.0355***
(0.0063)
-0.0313***
(0.0052)
9555
0.48
Good math grades
Teachers too fast
Observations
R-squared
9555
0.44
9555
0.47
Note: Regressions include controls for the province where the native born attended high school. The sample consists of
individuals age 16 and older, with weekly earnings greater than $50 and less than or equal to $20,000, excluding
students and observations with missing information on age at immigration.
(***) Coefficient significant at 1% significance level, (**) Coefficient significant at 5% significance level,
(*) Coefficient significant at 10% significance level.
44
Table 3(a) Earnings Regressions without Skill Effects - Men
Immigrant
Basic 1
-0.466***
(0.078)
Immigrant (with Canadian education)
Immigrant (with no Canadian education)
Years since migration
Years since migration sq. / 100
Experience
Experience sq. / 100
0.020***
(0.004)
-0.014*
(0.008)
0.072***
(0.003)
-0.122***
(0.007)
Basic 2
Basic 3
Preferred
-0.242**
(0.115)
-0.565***
(0.125)
-0.065
(0.144)
0.065
(0.123)
-0.081
(0.150)
1.130***
(0.275)
0.121***
(0.008)
-0.158***
(0.009)
0.108***
(0.017)
0.073***
(0.003)
-0.124***
(0.007)
Canadian experience (native born)
Canadian experience
(immigrants with Canadian education)a
0.078***
(0.010)
0.077***
(0.004)
-0.132***
(0.008)
0.068***
(0.013)
Canadian experience sq. /100
(immigrants with Canadian education)a
-0.118***
(0.024)
-0.107***
(0.030)
-0.109***
(0.037)
Canadian experience
(immigrants with no Canadian education)a
0.092***
(0.014)
0.052***
(0.012)
0.046***
(0.012)
Canadian experience sq. /100
(immigrants with no Canadian education)a
-0.135***
(0.042)
-0.088**
(0.033)
-0.081**
(0.033)
0.007
(0.009)
-0.012
(0.027)
0.061***
(0.004)
0.004
(0.009)
-0.008
(0.027)
Canadian experience sq. /100 (native born)
Foreign experience
Foreign experience sq. /100
Years of schooling
0.060***
(0.005)
0.061***
(0.005)
Years of schooling
(native born and immigrants with Canadian education)
0.112***
(0.010)
Years of schooling
(immigrants with no Canadian education)
0.044***
(0.009)
Canadian experience * schooling
(native born and immigrants with Canadian education)
-0.003***
(0.000)
First language not English or French
Good math grades
Teachers too fast
-0.012
(0.045)
0.057
(0.041)
-0.146***
(0.035)
-0.007
(0.054)
0.059
(0.042)
-0.142***
(0.035)
-0.003
(0.057)
0.067
(0.045)
-0.138***
(0.034)
-0.037
(0.060)
0.082*
(0.046)
-0.116***
(0.034)
45
US or UK origin
0.145
0.149
(0.102)
(0.113)
European origin
0.274***
0.273***
(0.087)
(0.083)
Asian origin
0.038
0.059
(0.059)
(0.054)
Observations
4551
4551
4551
R-squared
0.37
0.37
0.38
(***) Coefficient significant at 1% significance level, (**) Coefficient significant at 5% significance
level, (*) Coefficient significant at 10% significance level.
Regressions include indicators for province of residence.
a
In specification 2 coefficients on Canadian experience variables for immigrants were estimated as
differences from the experience of the native born. In this table they are reported so that they can be read
directly as returns to Canadian experience of the two immigrant groups rather than as comparisons to the
overall Canadian experience variables.
0.149
(0.117)
0.249***
(0.080)
0.065
(0.052)
4551
0.40
46
Table 3(b) Earnings Regressions without Skill Effects - Women
Immigrant
Basic 1
-0.315***
(0.109)
Immigrant (with Canadian education)
Immigrant (with no Canadian education)
Years since migration
Years since migration sq. / 100
Experience
Experience sq. / 100
0.022**
(0.008)
-0.028**
(0.014)
0.056***
(0.006)
-0.102***
(0.013)
Basic 2
Basic 3
Preferred
-0.063
(0.096)
-0.224
(0.136)
0.039
(0.093)
0.020
(0.168)
0.087
(0.089)
1.663***
(0.339)
0.090***
(0.009)
-0.125***
(0.015)
0.094***
(0.008)
0.056***
(0.006)
-0.101***
(0.014)
Canadian experience (native born)
Canadian experience
(immigrants with Canadian education)a
0.075***
(0.008)
0.060***
(0.006)
-0.110***
(0.014)
0.067***
(0.007)
Canadian experience sq. /100
(immigrants with Canadian education)a
-0.137***
(0.020)
-0.125***
(0.017)
-0.132***
(0.016)
Canadian experience
(immigrants with no Canadian education)a
0.063***
(0.017)
0.040**
(0.018)
0.035**
(0.015)
Canadian experience sq. /100
(immigrants with no Canadian education)a
-0.093**
(0.040)
-0.061
(0.043)
-0.067*
(0.033)
0.019
(0.011)
-0.017
(0.035)
0.091***
(0.007)
0.021**
(0.010)
-0.051
(0.031)
Canadian experience sq. /100 (native born)
Foreign experience
Foreign experience sq. /100
Years of schooling
0.089***
(0.007)
0.090***
(0.007)
Years of schooling
(native born and immigrants with Canadian education)
0.132***
(0.010)
Years of schooling
(immigrants with no Canadian education)
0.026**
(0.012)
Canadian experience * schooling
(native born and immigrants with Canadian education)
First language not English or French
Good math grades
Teachers too fast
-0.002***
(0.000)
-0.055
(0.048)
0.046
(0.035)
-0.054*
(0.031)
-0.042
(0.061)
0.046
(0.035)
-0.058*
(0.030)
-0.045
(0.060)
0.055
(0.035)
-0.050
(0.031)
-0.061
(0.061)
0.065*
(0.036)
-0.034
(0.033)
47
US or UK origin
-0.085
-0.092
(0.103)
(0.116)
European origin
-0.123
-0.114
(0.085)
(0.080)
Asian origin
-0.071
-0.070
(0.084)
(0.083)
Observations
5004
5004
5004
R-squared
0.31
0.31
0.31
(***) Coefficient significant at 1% significance level, (**) Coefficient significant at 5% significance
level, (*) Coefficient significant at 10% significance level.
Regressions include indicators for province of residence.
a
In specification 2 coefficients on Canadian experience variables for immigrants were estimated as
differences from the experience of the native born. In this table they are reported so that they can be read
directly as returns to Canadian experience of the two immigrant groups rather than as comparisons to the
overall Canadian experience variables.
-0.054
(0.120)
-0.124
(0.080)
-0.098
(0.080)
5004
0.33
48
Table 4(a) Earnings Regressions with Skill Effects - Men
Average skill score /100
(1)
0.280***
(0.027)
Average skill score /100 (native born)
Average skill score /100 (immigrants)
(2)
(3)
0.256***
(0.042)
0.332***
(0.054)
0.255***
(0.041)
Average skill score /100
(immigrants with Canadian education)
0.303***
(0.094)
Average skill score /100
(immigrants with no Canadian education)
0.353***
(0.055)
Immigrant (with Canadian education)
Canadian experience
(immigrants with Canadian education)
0.024
(0.144)
1.198***
(0.264)
0.120***
(0.008)
-0.154***
(0.009)
0.106***
(0.017)
-0.190
(0.289)
1.040***
(0.365)
0.120***
(0.008)
-0.155***
(0.009)
0.105***
(0.017)
-0.117
(0.373)
1.015***
(0.357)
0.120***
(0.008)
-0.155***
(0.009)
0.106***
(0.017)
Canadian experience sq. /100
(immigrants with Canadian education)
-0.102***
(0.037)
-0.101***
(0.037)
-0.102***
(0.037)
Canadian experience
(immigrants with no Canadian education)
0.051***
(0.011)
0.052***
(0.012)
0.052***
(0.012)
Canadian experience sq. /100
(immigrants with no Canadian education)
-0.088***
(0.030)
-0.089***
(0.030)
-0.090***
(0.030)
Foreign experience
0.006
(0.008)
-0.009
(0.025)
0.094***
(0.010)
0.007
(0.008)
-0.010
(0.025)
0.095***
(0.010)
0.006
(0.008)
-0.008
(0.026)
0.095***
(0.010)
Years of schooling
(immigrants with no Canadian education)
0.028***
(0.009)
0.025***
(0.008)
0.024***
(0.008)
Canadian experience * schooling
(native born and immigrants with Canadian
education)
-0.003***
(0.000)
-0.003***
(0.000)
-0.003***
(0.000)
First language not English or French
-0.012
(0.057)
0.049
(0.041)
-0.090**
(0.036)
-0.009
(0.058)
0.049
(0.041)
-0.090**
(0.035)
-0.008
(0.058)
0.050
(0.041)
-0.090**
(0.035)
Immigrant (with no Canadian education)
Canadian experience (native born)
Canadian experience sq. /100 (native born)
Foreign experience sq. /100
Years of schooling
(native born and immigrants with Canadian
education)
Good math grades
Teachers too fast
49
US or UK origin
0.048
0.028
0.028
(0.113)
(0.106)
(0.105)
European origin
0.189**
0.179**
0.181**
(0.080)
(0.080)
(0.081)
Asian origin
0.033
0.025
0.025
(0.058)
(0.059)
(0.059)
Observations
4551
4551
4551
R-squared
0.42
0.42
0.42
(***) Coefficient significant at 1% significance level, (**) Coefficient significant at 5% significance level,
(*) Coefficient significant at 10% significance level.
Regressions include indicators for province of residence.
50
Table 4(b) Earnings Regressions with Skill Effects -Women
Average skill score /100
(1)
0.294***
(0.046)
Average skill score /100 (native born)
Average skill score /100 (immigrants)
(2)
(3)
0.297***
(0.056)
0.286***
(0.054)
0.291***
(0.056)
Average skill score /100
(immigrants with Canadian education)
0.175*
(0.092)
Average skill score /100
(immigrants with no Canadian education)
Immigrant (with Canadian education)
0.402***
(0.054)
Canadian experience
(immigrants with Canadian education)
0.148
(0.096)
1.832***
(0.308)
0.094***
(0.009)
-0.125***
(0.014)
0.097***
(0.008)
0.179
(0.234)
1.855***
(0.340)
0.094***
(0.009)
-0.125***
(0.014)
0.097***
(0.008)
0.459
(0.316)
1.721***
(0.305)
0.093***
(0.009)
-0.124***
(0.014)
0.098***
(0.008)
Canadian experience sq. /100
(immigrants with Canadian education)
-0.132***
(0.019)
-0.132***
(0.019)
-0.135***
(0.018)
Canadian experience
(immigrants with no Canadian education)
0.037**
(0.014)
0.036**
(0.014)
0.038**
(0.014)
Canadian experience sq. /100
(immigrants with no Canadian education)
-0.071**
(0.031)
-0.071**
(0.031)
-0.077**
(0.030)
Foreign experience
0.027***
(0.009)
-0.055*
(0.028)
0.118***
(0.010)
0.027***
(0.009)
-0.055*
(0.028)
0.118***
(0.010)
0.024**
(0.009)
-0.047
(0.029)
0.120***
(0.010)
0.007
(0.012)
0.008
(0.012)
-0.002
(0.013)
-0.002***
(0.000)
-0.002***
(0.000)
-0.002***
(0.000)
-0.041
(0.060)
0.041
(0.034)
-0.010
(0.034)
-0.041
(0.060)
0.041
(0.034)
-0.010
(0.035)
-0.040
(0.058)
0.042
(0.034)
-0.010
(0.034)
Immigrant (with no Canadian education)
Canadian experience (native born)
Canadian experience sq. /100 (native born)
Foreign experience sq. /100
Years of schooling
(native born and immigrants with Canadian
education)
Years of schooling
(immigrants with no Canadian education)
Canadian experience * schooling
(native born and immigrants with Canadian
education)
First language not English or French
Good math grades
Teachers too fast
51
US or UK origin
-0.108
-0.106
-0.114
(0.113)
(0.116)
(0.116)
European origin
-0.131
-0.130*
-0.127
(0.077)
(0.076)
(0.078)
Asian origin
-0.075
-0.074
-0.065
(0.074)
(0.073)
(0.072)
Observations
5004
5004
5004
R-squared
0.35
0.35
0.35
(***) Coefficient significant at 1% significance level, (**) Coefficient significant at 5% significance level,
(*) Coefficient significant at 10% significance level.
Regressions include indicators for province of residence.