WebCT: ECON30631 Labour Economics (004510) - ECON30631 &LEC 2009-10 1st Sem (coursework) : Immigrant Labour Market Assimilation
This page last changed on 18 Dec 2009 by msra7ps2.

Contents
1 Trends
      1.1 Immigrants in the U.K
      1.2 Immigrants in the U.S
2 U.S Literature
      2.1 True Assimilation
      2.2 Assimilation & Cohort-Effects
      2.3 Measuring Assimilation
      2.4 Unemployment & Earnings of Immigrants
      2.5 Second Generation & Assimilation
3 U.K Literature
      3.1 Chiswick, B.(1980)
      3.2 Bell, B.D.(1997)
      3.3 Clark, K. & Lindley, J.(2009)
4 Evidence from other countries
5 Other indicators for assimilation
6 Conclusion
7 References

Assimilation, sometimes known as integration or incorporation, is the process by which the characteristics of members of immigrant groups and host societies come to resemble one another. That process, which has both economic and sociocultural dimensions, begins with the immigrant generation and continues through the second generation and beyond.


This wiki examines the situation of immigrant workers in the United Kingdom and the United States. In keeping with the literature we use the term assimilation to describe the process of convergence between migrants and their native born counterparts in an outcome of interest such as average earnings. It has been widely recognized that immigrant workers tend to face disadvantage in the labour market, both in terms of their employment and earnings experience.

However, this work also shows that the experience of immigrants is very heterogeneous. Immigrants from different ethnic backgrounds also tend to fare differently. But also immigrants who have been in the U.K. and the U.S. for longer also tend to do better. In particular, the work of Borjas has examined the assimilation of immigrants into the U.S. However, there is little work on the assimilation of immigrants into the British labour market. Chiswick was the first to analyse this issue. More recently, Bell examines assimilation of wages of different immigrant groups using retrospective data from the General Household Survey.


UK & USA Immigration Legislations

UK Timeline
1914 British Nationality 
1948 British Nationality Act
1958 British Nationality Act
1962 Commonwealth Immigrants Acts
1962 Immigration Act
1964 British Nationality Act
1968 Immigration Act
1968 Commonwealth Immigrants Act
1971 Immigration Act
1981 British Nationality Act
1990 British Nationality Act


USA Timeline
1906 The Naturalization Act of
1917 Immigration Act
1921 Emergency Quota Act
1924 Immigration Act
1943 Chinese Exclusion Repeal Act
1952 Immigration & Nationality Act
1954 Operation Wetback
1965 Immigration & Nationality Act
1986 Immigration Reform Act
1996 Immigration Reform Act
2005 REAL ID Act

Immigration Trends in the UK and US

United Kingdom

Data from the Lifetime Market Database (LLMDB) is used to examine the assimilation of immigrants into the labour market within the UK. Figure 1 uses our LLMDB data to plot the number of new arrivals each year between 1977/78 and 2003/04. What we see is that in the late 70s and early 80s, there were about 100,000 new immigrants each year. Immigration then rose through the second half of the 1980s to around 200,000. Immigration then rises strongly in the late 1990s to around 300,000. These trends show some major changes in the inflow of immigrants over this period. However, our data does not capture the even larger changes that have occurred since 2003 and the entry into the EU of the accession countries. Figure 1 also plots data from DWP estimates of the number of inflows for 2000/01 to 2004/05 using 100% records data on NI allocations (DWP, 2006). The numbers coincide well with our estimates from LLMDB. However, we see a huge jump in the series so that in 2004/05 over 460,000 immigrants enter Great Britain. Our data on migrants does not cover this later period at present but it is reassuring to see our estimates are close to those of the DWP.

Figure 1: Number of new immigrant arrivals 1977/78-2004


United States

The first humans crossed over to the U.S from northeast Asia over 10,000 years ago. It was not until the 15th century that European immigration to the U.S. started, a few decades after Columbus' discovery in 1492 and was mainly composed of Spaniards. In the early years of the U.S., immigration was only about 6000 people a year on average. In the 16 and the 17 hundreds, immigrants were composed of mainly British and African and other western European countries.

Table 1: U.S. Populations, Immigrants before 1970

The war with Britain slowed down immigration to the U.S and did not really take off till the 1830s. The potato famine in Ireland caused a huge jump in immigration in the 50s. The pattern of European continued until the early 20th century.

Figure 2: Immigration to the U.S. by Decade (in millions)

| To Top |

Trends

Net immigration has added 1.5 million people to the population over the last 10 years. Two-thirds of them have come from the continents of Asia and Africa.


In 2006 the total UK population was 60.6m. If net immigration continues at its current rate, as the government actuary's department calculates, the total population in 2031 will be 70m, and by 2081 it will be 85m.


People born overseas account for about 10% of the population. This is up from just more than 6% in 1981 and just over 8% in 2001. This is compared with 24% in Australia, 23% in Switzerland, 19% in Canada and 13% in America. In France that figure is 8%, and 7% in Denmark.


In 2006 the largest numbers of immigrants were born in Africa and the Middle East (3% of the working age population) and the Indian sub-continent (2.5%).


Those born in the east and central European EU member states - the A8 - made up 1% of the working age population. Today A8 immigrants account for one in three of new immigrants since 2004.


The three most popular sectors for all foreign-born workers in the UK are public administration, education and health (32%), distribution, hotels and restaurants (21%) and banking, finance and insurance (20%). Immigrants are concentrated at the high and low skill end of the occupation distribution.


The U.S. government has forecast a shortage of 20 million workers by 2026, prompting many parties to call for a relaxation of the US immigration laws in order to meet the labor demand.

U.S Literature

True Assimilation

Chiswick (1978) 1 wrote strongly in favour of true assimilation. His findings agreed with the textbook case. He stated that after immigrants arrive in the US they gradually acquire knowledge of the language, customs, and the nature of labour markets in the US. He found earnings increased, although at a decreasing rate, with the number of years an immigrant. But did he agree with the text book case or is he the text book case?

Chiswick (1978) analyzed the earnings of immigrant adult white men in the US. He took a cross section of data, from the 1970 census of population, giving a snap shot of the US population at this time. This allowed Chiswick to compare the current earnings of newly arrived immigrants with the current earnings of immigrants who arrived years before. He analysed these earnings through comparisons with the native born and among the foreign born by country of origin, years in the United States and citizenship. Differences in the effects of schooling and post school training were also explored.

Figure 3


The evidence that Chiswick (1978) found agreed with the assimilationist view. His cross section data set gave three distinct results when looking at the Age-earnings profiles of immigrant and native men.

  1. Immigrant earnings are initially lower than native men.
  2. The immigrant age-earnings profile is steeper than the native age-earnings profile, the earnings of the foreign born are growing at a faster rate.
  3. After 10 -15 years in the US immigrants earn more than natives.

Assimilation theory, supported by Chiswick (1978), offers the following explanations for the results:

  1. As there are aspects of schooling that are country specific, a year of schooling prior to immigration will have a smaller effect on earnings than a year of schooling for the native born.
  2. As there are aspects of labour market that are country specific, a year of experience prior to immigration has a smaller effect on earnings than a year of experience prior to immigration has a smaller effect on earnings than a year of experience for a native person. Therefore upon arriving in the country and gaining more experience in US immirants earnings will rise quickly.
  3. As immigrants initially have less human capital specific to the United States than natives - born persons of the same schooling and age, just after they arrive their earnings are lower than the native born.
  4. After they arrive, as immigrants make investments in post school training informally acquire "experience" living in the United States, the earnings of immigrants rise at a faster rate than the earnings of the native born.
  5. As immigrants have the incentive to make their largest adjustment investments just after they arrive, the absolute decline in the "knowledge gap" between immigrants and the native born is sharpest in these years. The rise in earnings with time in the United States is steepest in the first few years.
  6. Immigrants make investments in post - school training relevant for jobs in the US. These investments initially depress earnings and raise them later on. This can be seen on the age earnings profile.
  7. Becker (1964) has shown that for the same total investment in training, experience earnings profiles are steeper the smaller the proportion that is firm specific and the smaller the proportion of firm- specific training that is financed by the employer. Chiswick(1978) stated that theforeign born were more likely to have training which was not firm specific or financed by the employer because employers have poorer knowledge of them than the native born. The experience immigrants therefore gain has a larger effect on their earnings.
  8. The effect on earnings of time in the United States, holding total labour market experience constant, is weaker for immigrants from countries that more closely resemble the United States.
  9. The foreign born may have more innate ability, are more highly motivated toward labour market success, or self - finance larger investments in post school training. These high earning, Chiswick believed, are a consequence of self - selection explain why eventually immigrants earnings overtake those of natives.

Effect of Schooling

Chiswick explains the smaller effect of pre-immigration schooling by the fact that knowledge acquired at school may be country specific and provide employers in the US with poor information. It could also be explained by the lower possibility of lower quality of foreign schooling.

Chiswick also found a smaller effect of post-immigration schooling for foreign born men. He discussed that this may be due in part to self- selection in migration, stating that maybe only the most able and most highly motivated of those with little schooling migrate while those with higher levels of schooling (or who subsequently acquire) came from a broader ability and motivation spectrum.

Chiswick(1980) 2 analysed the earnings of immigrants in Britain. Of the sample of 5578 men 6.2% were foreign born.

Chiswick found that the annual earnings foreign born men were 4 per cent lower than that of native - born white men.

Importantly he found that when holding all other variables constant (e.g. schooling, location and including post school training), the number of years since the person migrated to Britain has no additional explanatory power. This means that for immigrants in Britain labour market experience in Britain and in the country of origin are equally productive, this does not support true assimilation. Chiswick (1980) explained this by the fact that those who had come to Britain in the 20 years before 1970 were well "anglicised" before they arrived. This was backed up by the fact that a large proportion of the immigrants had come from English - speaking commonwealth countries.

He is suggesting that this wave of immigrants arrived in the UK with a high level of human capital already, with little potential to assimilate. However he did find a small effect of schooling on earnings when looking at non - white immigrants, suggesting they had received a lower quality of schooling.

Assimilation and Cohort-Effects

Chiswick concluded that immigrants had a steeper are-earnings profile when they entered the United States, his studies also revealed an apparent time period in which immigrant wages would overtake that of natives.

The key difference that cohort analysis reveals compared to Chiswick work is that it takes into account the quality of immigrants coming into the United States over time. This is a contrast with Chiswick's work in which he studies cross-sectional data and makes inferences about how earnings of immigrants evolve overtime. Borjas revolutionized the way labour economists thought about immigration in the U.S and his first paper with cohort analysis.

What is a cohort?

A cohort is a group of subjects who have shared a particular experience during a particular time span (e.g., people born in 1950; Mexican women born in 1950). Cohorts may be tracked over extended periods of time in a cohort study. Put simply, it is a group of individuals having a statistical factor in common.
Example - The graph below illustrates a simple example, revealing the logic behind Borjas's studies:

Figure4

Consider a hypothetical situation where there are three separate waves of immigration, the first in 1960, the second in 1980 and the last in 2000. Suppose also that all immigrates enter the U.S at the age of 20. We assume the following:

  1. Assume the first cohort has the highest productivity level of any group and above that of natives. Age-earnings profile PP.
  2. Assume the second cohort has the equal productivity as that of natives. Age-earnings profile QQ.
  3. Finally that the third wave has the lowest productivity of any group. Age-earnings profile RR.

The graph shows that age-earnings profile of the immigrants is parallel with native workers and do not converge like the previous work on assimilation.

Suppose now we have data from the 2000 decennial census (shows each workers wage rate, age, foreign born or not and year of arrival to the U.S). Information from this would enable us to observe the wage of those who arrived in 2000 and are 20, as well as those who arrived in the second wave and are 40, and finally, those who arrived in the 1960s and are 60. Connecting the points reveals a line which has two important properties, cross-sectional data can make it seem like there is a wage convergence between immigrants and natives. First, it is substantially steeper that the age-earning profiles of individual cohorts. Second, the line crosses natives' age-earnings line at the age of 40, misleading when cohorts are not taken into account.

Borjas's studies show that older views about the adaptation process experienced by immigrants can be wrong if exist differences in productivity across immigrant cohorts.3 These differences in skills across cohorts are called cohort effects. 4

Evidence

The example above has striking parallels with the experience of immigrants in the United States. Immigrants in the 1960s were largely from Europe and Canada, and skilled. The productivity of immigrant workers has seen a sharp decline since then, due to the shift in immigration patterns.

Data suggests that there are quite large skill differentials across immigrant cohorts, reflected in their age-earnings profile. The figure below highlights the trend in the entry wage gap between immigrants and natives across successive immigrant waves between 1960 and 2000. In 1960, new immigrants earned 11% less than natives, by 1990 the gap increased to 37%.

Figure5

To determine whether specific cohorts (as opposed to immigrants in general) reached parity with natives, track studies were conducted that followed immigrants after entry. For example, Borjas looked at immigrant wages of those who arrived in the 70s using the 1980 census and tracked their wage increases using the 1990 census then the 2000 census and so on. The figure below illustrates the evidence provided by this type of tracking study.

Figure6

There is a big gap between earnings of those who arrived in the 1960s and those who arrived in 1970, the wage they received upon entry was higher. Those who arrived in the 60s and 70s usually caught up or exceeded native wages. While only a small percentage of those arriving in later cohorts either caught up with natives, or did not catch up at all.

The Data: How Best to Measure Assimilation?

The evolution of theories concerned with assimilation highlights the importance of using the correct data when conducting analysis. Of course, in a perfect world we would also be able to add a variable for immigrant's innate ability but clearly this is not observable. There have been various attempts to model this characteristic but unfortunately there is no reliable source. If we could it could prove or disprove whether or not immigrants 20 or 30 years ago were of a 'higher quality'.

Chiswick (1978) analyzed the earnings of immigrant adult white men in the U.S. He took a cross section of data from the 1970 census of population, giving a snap shot of the US population at this time. Chiswick aimed to compare the current earnings of newly arrived immigrants with that of immigrants who arrived years before. His analysis did not involve tracking or monitoring of a group of immigrants over time, rather he only compared wages of newly arrived immigrants with immigrants already in the country.

Borjas's work sheds light on the importance of having the right data when analysing assimilations patterns. Chicwick's simplistic analysis does not consider the need to assess the evolution of quality/productivity of immigrants across time. The data therefore that is of most importance in showing the actual picture of assimilation is data that shows not only age, earnings and place of birth but also length of time in the country.

Ideally, we would like to compare a sample of immigrants over the years after they migrated into the US and compare their earnings year on year. This time series rather than cross section approach would give us stronger evidence. Obtaining dates concerning entry of immigrants and reliable data about wages is not straight forward.

As many of a third of immigrants into the U.S are illegal and accurate information that reflects true numbers is not available.

Unemployment and the Relative Earnings of Immigrants

Barth - Failure to account for differences between immigrants and natives in their responsiveness to changes in macroeconomic conditions may bias estimates of assimilation effects on immigrant earnings.

Barth et al. (2004) linked individual earnings to the unemployment rate in the local labour market. By allowing such linkages to differ for immigrant and native workers they estimated assimilation effects accounting for differential responsiveness of immigrants and natives to changes in macroeconomic conditions. They found that in periods of increasing unemployment, the standard methodology tends to underestimate both the assimilation process and any growth in cohort changes.5

Studies of immigrants in Canada and the US indicate that their job rates fall more than those of native during economic downturns (McDonald and Worsick (1997) ). Where years since migration have a positive impact on immigrant wages relative to those of natives, higher unemployment has a negative effect.
Work less hours, hampers accumulation of skills, worse job prospects -> earnings are reduced.

  1. The level effect suggests that the earnings gap between natives and immigrants increases with higher unemployment.
  2. The impact of local unemployment on the rate of earnings assimilation i.e the slope effect (on the age earnings diagram) is ambiguous. On the one hand, at least atlow values of years since migration, higher unemployment reduces the accumulation of work experience and therby the learning effect on individual productivity. On the other hand the earnings profile of immigrants may be steeper under high unemployment becuause, under such conditions, job opportunities of immigrants become more sensitive to years in the host country.

Second Generation and Assimilation

Since the work of Borjas and Chiswick assimilation theory has been further developed by looking at the second generation (the US born children of immigrants) of immigrants. David Card (2005) analysed empirical evidence on the immigrants who arrived after the 1965 Immigration Reform Act, focusing on the children of immigrants and the immigrants themselves. Second generation immigrants are growing as a portion of the population, accounting for 10 percent of teenagers nationwide.

The first education model shows that immigrants have 1.2-1.4 fewer years of education then natives. On the other hand, second generation immigrants have 0.3-0.4 years more education than people whose parents were born in the U.S, who are known as the "third and higher" generation. 6

Immigrant men are only slightly less likely to work than members of the third generation (children whose parents were born in the US), however on the other hand second generation men are more likely to work. Among women there is a larger immigrant gap in the probability of working women (13.6 percent lower annual employment rate than natives) but again second generation women are more likely to work than third and higher generation natives. . (David card 2005-Is the New Immigration Really So Bad?

Table 2

The models in columns 3 and 4 present wage models that control for age and geographic location, but not for education. Among men, immigrants have 18-23 percent lower wages than third and higher generation natives, while second generation immigrants have 4-8 percent higher wages. The wage gaps for second generation women are similar for second generation men, whereas for immigrant women the wage gap is smaller than for immigrant men. This could reflect the relative selectivity of labour force participation among immigrant women.

The higher wages of second generation immigrants can be largely explained by their geographic location and their higher education.

The result from Table 2 concludes:

  1. Immigrant workers in the labour market today (over 90 percent of whom arrived after 1965) earn less than natives, however, the magnitude of the wage gaps are not enormous.
  2. Assimilation is tends to be achieved by the second generation, with most of their wage advantage relative to natives attributable to higher education. Despite the lower education of their parents, children born to immigrant parents seem to be on par or even go beyond the levels of children born to U.S. natives.
| To Top |

US Literature

Barry Chiswick

Ph.D., Columbia University, 1967
M.A., Columbia University, 1964
B.A., Brooklyn College, 1962







Approximately 28.4 million foreign-born people live in the United States, representing 10.4 percent of the U.S. population.







Prior to the mid-20th century, immigrants came primarily from northwestern Europe (Great Britain, Ireland, Germany, Scandinavia, France) and from Canada. By 1999, there had been a dramatic shift. Over half of the foreign-born population now originated from Latin America. Europe accounts for only 16 percent, and Canada for 3 percent. Only 2% of the foreign-born originated in Africa.







Relatively few (10%) of the foreign-born are under age 18 compared with 28.3 percent of the native population. The largest percentage of foreign born are in the 35-39 age group.







George Borjas

Ph.D., Columbia University, 1975
M.A., Columbia University, 1974
B.S., St. Peter's College, 1971





The foreign born are more geographically concentrated than the native-born population. Three-quarters of the foreign-born population is concentrated in just eight states (California, New York, Florida, Texas, New Jersey, Illinois, Massachusetts, and Arizona).






The 1990 census tabulated almost 32 million speakers of hundreds of languages from other continents as well as more than 330,000 people speaking more than 130 native North American languages at home.






Borjas published his first paper concerned with cohort effects in the US labour market in 1985

David Card

Ph.D., Princeton University, 1983
B.A., Queen's University, 1978






By the 1990s, immigration from Latin America accounted for half of all immigration, compared to 25% in the 1950s.







More than one in five legal immigrants settle in California, and the 355,600 immigrants who landed there and in New York in 1996 represented 39 percent of U.S. immigration. New York City alone was the planned destination of 15 percent of immigrants.




Interestingly, there was a slight turnaround in the 1990s, in 2000 newly arrived immigrants earned about 31% less than natives.

Erling Barth

Ph.D., University of Oslo, 1993
M.A., University of Oslo, 1985




Failure to account for differences between immigrants and natives in their responsiveness to changes in macroeconomic conditions may bias estimates of assimilation effects on immigrant earnings.
                  -Barth







With such a large share of immigrants heading to just two states, it's no wonder that some states receive very few. Idaho, Kentucky, Nebraska, and South Carolina each claimed only 0.2 percent of immigrants in 1996. The 8,100 people who planned to go to these four states were fewer in number than those thinking of living in El Paso, Texas.






Immigration has a modest but positive effect on the U.S. economy, according to American Demographics, adding about $10 billion a year to America's economic output. Even more important is the contribution immigrants and their children make just by being here to provide workers and leaders for the future. If today's immigration totals hold steady, it will account for about two-thirds of U.S. population growth over the next 50 years.







The achievement of the second generation provides a key measure of the extent to which their parents assimilated into the U.S.
                  -Frey, William H.







Mexico was the country of origin for the largest share, at 18 percent. The Philippines was next, at 6 percent, followed by India, Vietnam, and China, 5 percent each.

U.K Literature

Introduction

According to the Labour Force Survey, some 4.5 million individuals living in Britain 2000 (9% of the population) were born in another country

Over time, they may adjust with respects to their UK-born peers, due to accumulation of skills, collection of information, and adoption of new habits.  

To understand how labour market performance of immigrants differs from UK-born, and from each other, how these differences relate to observed characteristics, and how they change over time is an important pre-requisite for migration policy.  

While adaptation and labour market performance of immigrant populations have been subject of intensive research in the US, Canada, Australia, and also in some European countries, relatively little is known about the absolute and relative performance of the immigrant community living in Britain.  

This section on UK labour market assimilation discusses the key literature in this area.  

Chiswick in 1980 looks at assimilation between non-white immigrants and natives.  

Bell (1997) considers ethnic minorities in more depth to see how labour market assimilation differs between them.  

Clark and Lindley (2009) take a slightly different approach by looking at immigrant labour market entrants and comparing them with immigrants who enter education first. The results for whites and non-whites have implications for UK immigration policy focus.  

Each of these papers analyses labour market assimilation in terms of employment possibilities and wage earnings.

Chiswick (1980): The Earnings of White and Coloured Male Immigrants in Britain

This paper analyses the earnings of adult, white and non-white, immigrant men living in Britain in 1972. Chiswick found that whilst white immigrants earned largely as much as their native counterparts, there was a wage penalty of around 25% for non-white immigrants.

He also found that their relative earnings disadvantage decreases with the number of years of schooling.

Whilst this paper is useful for understanding wage disparities amongst white and non-white immigrants; the small sample size used (214 white immigrants and 127 black immigrants) from a single cross section of data (the 1972 General Household Survey) meant that further investigation into this topic was needed.

It therefore formed the backdrop for the works of Clark and Lindley in 2006 and forthcoming.

Bell (1997): The Performance of Immigrants in the United Kingdom

Bell writes his paper and tries to assess the performance of immigrants in the UK labour market using the results of the General Household Survey (GHS) between 1973-92. 7 Bell splits his study in to three main components focusing on the "changing national-origin mix of immigrants," "Educational attainment among immigrants" and "The wages of immigrants."

While he looks at the differences between whites and non-whites, it does not become the focus unlike the Clark and Lindley paper; instead he differentiates between the origins of the immigrants, and splits their arrival into ten-yearly cohorts.

Table 3: Immigrant-origin


Table 3 shows the distribution of immigrant cohorts by birthplace; the drop in non-European immigration between the 1970-9 and 1980-9 cohorts can be mainly attributed to the 1971 Immigration. Whereas European immigration increased over the two periods, European immigration doubling between 1960-9 and 1980-9, due to Britain joining the European Community in 1973. This change in immigration patterns between cohorts will be useful when explaining the increasing relative levels of education between immigrants and natives.

Educational Differences


||
Table 4
||


Table 5:

For the exact explanation of columns and the regression model used for the table, click here

Columns three and four of Table 5 try to determine the growing gap between relative education of that of natives and immigrants. There are two main theories for this:

1. Immigrants are becoming more educated relative to natives irrespective of their origin of birth.
2. The changing relative education is due to a changing national origin of immigrants. As we have seen from the first table this could have a significant impact due to the drastic changes we have seen in immigrant origin over the successive cohorts.

As seen in column three there is a big jump in relative education of immigrants between the 1960-9 and 1970-9 cohorts, this looks to be an anomaly, as it would be hard to have such a drastic change over one cohort, especially considering the regression from pre-1960 and 1960-9 cohorts.

Column 4 shows the relative education levels of immigrant cohorts weighted by region of birth. The changing differences is indicative of a changing national-origin of immigrants, the large numbers of Caribbean immigrants pre-1960 and in the 1960-69 cohort, combined with the fact that that these immigrants have the lowest average years of schooling (as seen in table 4), correlates with the negative regression of education shown in table 5, the large numbers of Caribbean immigrants could be contributing to the negative regression shown in the pre-1960 and 1960-69 cohorts.

These findings indicate that rather than assimilation with natives the changing relative educational standards are due more to the origin of immigrants as opposed to a process of assimilation over time. This is supported by table 4 which does not show much if any educational assimilation, as the average years of education increases for all groups with no big relative change for any specific one.

Wage Findings

In regards to the wages of immigrants Chiswick (1980) 8 found that the relative wage between migrants and natives did not change over time. Bell found evidence to the contrary, while there often was a difference in relative wages, it assimilated to the native wage over time.

For all immigrants foreign experience and education is valued less than British experience and education. However, for white and Old Commonwealth immigrants the impact is less detrimental than for other immigrants. This is because they are perceived to have an education and work experience similar to that of British natives and the language barrier is less of a problem.

The cohort effects of wage regression have fallen over time, which may be due to a fall in quality of successive cohorts of Indian immigrants. Bell uses table 6 to help assess the impact of immigrant assimilation relative to native wages.

Table 6:

Immigrants from all countries, with the exception of white immigrants with 10 years experience, suffer a penalty for foreign experience (for reasons explained above). West Indians suffered the greatest entry differential, but they also experienced the greatest assimilation effects over the 30 years in the labour market. In all three categories they suffered from a worse entry wage than Indians, but by the end of the 30 years they had a better relative wage (than Indians) in all categories other than those with no foreign experience. The table below shows the assimilation of wages relative to that of similar natives.

Table 7:


White immigrants with no foreign experience benefit from a wage premium  over natives according to Bell it is 30.5%, however over time they suffer from "dis-assimilation".

Bell tries to explain the fact that white immigrants with no foreign experience seem to benefit from a premium over natives in two ways.

1. There was (as Bell noted) a lack of test subjects in this category, nearly 90% of white immigrants having worked in the UK for at least10 years, and 70% for 20. Comparing this to 60% of Indian and 55% of West Indian immigrants having 20 years experience 7 it shows that comparatively there were fewer test subjects, so the wage findings may be skewed by a minority of successful white immigrants.

2. Secondly "this wage advantage may be due to some unobserved ability" 7 Immigrants may go through some process of "self-selection" when migrating here. He explains the "dis-assimilation" over time as these immigrants possibly re-emigrating out of the country, as they have made their money and emigrate back to their home country.

Clark and Lindley - Immigrant Assimilation Pre and Post Labour Market Entry: Evidence from the UK Labour Force Survey

This (2009) paper 9 uses Labour Force Survey (LFS) data from the period 1993-2004 to analyse the effects on earnings and employment for white and non-white immigrants; distinguishing between those that enter the labour market directly and those which first undertake some level of education in the UK.

They classify the latter as 'education entrants' and the former as 'labour market entrants'.

The methods and findings of this paper build directly on their previous study Clark and Lindley, 2006 but also the works of Chiswick, 1980 and Bell, 1997.

The importance of distinguishing between these two types of immigrants stems from work by Kossoudji (1989) who notes both UK labour market assimilation and pre-labour market assimilation (in the UK education system).

See Clark and Drinkwater(2008) for evidence of considerable diversity amongst non-white immigrants in terms of their labour market outcomes.

Why is this work unique?

Previous work in this area ignores the consequences of education entrants and since one third of all UK immigrants enter education; "an important part of the picture is being missed".9 The study also uses a larger sample and more recent data than previous studies.

The Modelling Framework

Clark and Lindley use the following equation to model labour market outcomes for immigrants into the UK: 

The equation controls for years since migration, human capital variables, etc. which aims to explain the differences in cohort quality; an important factor in "explaining immigrant earnings and performance in the UK and US" 9. Further reading on the US labour market can be obtained here.For a more detailed explanation of this model please see
The paper does recognise the limitations of this framework. It notes that an unobservable bias may exist within the sample of UK immigrants: "the frequently analysed situation whereby those in employment are a selected sample of the entire labour force may only be one source of non-random selection on unobservable attributes. Immigrants themselves are likely to be a highly self-selected group (Borjas (1987))4 and the distinction between education and labour market entrants may also introduce further selection bias problems...immigrants may non-randomly remigrate which affects the interpretation of expected assimilation profiles." 9
An example of this is that non-whites in the 1960s and 1970s were found to have an advantage over other immigrants. "A large number of immigrants who arrived in the UK at this time were Asians who were expelled from East Africa [typically entrepreneurs or high status employees]...and it is possible that this finding reflects the arrival in the UK of this highly selected group of "twice immigrants'" 9.

Labour Market Entrants

Important employment findings:

  • Non-whites born in the EU who migrate to Britain experience higher employment and earnings rates than immigrants from other countries (Although this is a small proportion of non-white white immigrants).
  • White immigrants have a declining probability of employment over their working lives.
  • The most noticeable difference is between white an non-white immigrants:
    1. Whilst a white immigrant, on average, experiences a higher probability of employment than a native on entry (which eventually erodes over time);
    2. Non-whites always experience an employment deficit. "By the age of 55, their employment profile is around 13% lower than whites and around 16% lower than natives" 9.
      Important earnings findings:
  • Again, the most noticeable difference here is not between natives and immigrants but between whites and non-whites.
  • On entry, both whites and non-whites earn more than natives, but this advantage is soon eroded.
    Figure 7: Log Earnings Assimilation

Figure 7 shows how the earnings premium for immigrants is eroded over time. It also shows how, compared with whites and natives, non-whites retain some form of penalty over their working lives.

  • The data seems to show that earnings profiles, particularly for non-whites, contradict the textbook model of assimilation.
  •  Why? A possible explanation (suggested by Bell 1997) is that selective out migration underlies a pattern of "dis-assimilation" in the UK.

Education Entrants

Some general points:

  • Highlights the idea that pre labour market assimilation "irons out differences in outcomes determined by cohort quality and origin which are experienced by those who arrive to enter the labour market directly". 9
  • "The enhanced English language ability of those non-whites who have exposure to the UK education system is also likely to be important here". 9
  • The significance of foreign schooling is evident for both ethnicities. However:
    1. Education has a larger effect for whites in the earnings equation;
    2. But a larger effect for non-whites in the employment equation.
    3. Among education entrants highly qualified prime-age non-whites perform as well as their white counterparts.
  • The level of UK qualification gained has more significance for non-whites. For those who gain a degree-level qualification the employment profile is similar to that of whites. However, those who gain no formal qualification, experience a substantial penalty in earnings, compared to natives, during peak working age. This is shown in Figure 8:
Figure 8: Education Entrants Differences
between Immigrants and Natives


There are larger returns to UK education for non-whites compared with whites. Also noted by Clark and Drinkwater(2008).10

  • "The impact of education gradually diminishes over time with immigrants from all qualifications groups having broadly similar employment rates at age of 60".9
  • There is very little difference between white and non-white education entrants at peak working age. "This demonstrates the benefit to non-white immigrants of investment in UK qualifications".
  • Similar to labour market entrants, the earnings of all immigrants tend to decline, compared with similarly qualified natives dis-assimilation. However, whereas the likely explanation for labour market entrants is selective out migration, for education entrants a more plausible explanation may be re-migration; possibly due to temporary residency of education entrants.
  • Earnings differences between the ethnicities are negligible: this is in vast contrast to labour market entrants.

How does education abroad compare?

Clark and Lindley describe labour market experience in this paper as 'potential' experience, since they do not observe work histories or panel data.
Their study found that non-white labour market entrants have more potential labour market experience than whites, although they are fairly indistinguishable in terms of foreign education. It also asserts that, compared to non-whites, white education entrants have:

  • More UK potential experience;
  • More years of UK schooling;
  • Fewer years of foreign schooling;

It is also of particular note that immigrants of all types have more total schooling than natives (on average). However, Clark and Lindley bring into question the quality of this education. The data would suggest that either schooling abroad is below the standard of UK schooling, or that there exists some irrational discrimination in the UK against overseas education standards.

Summary

This study is useful for analysing the contribution of immigrants to the UK labour market because:

  • It addresses the vast diversity amongst immigrants in terms of cohort quality, ethnicity, education, experience etc.
  • It accounts for those immigrants who enter the labour market directly and those who first enter the UK education system (seen as a means of enhancing their formal human capital).
  • It gives a more complete picture of assimilation (both pre and post labour market) in the UK; the importance of which is shown in the results.

The greatest significance of the study is that it highlights the inadequacies of previous labour market assimilation theory. The results show that the distinction between labour market and education entrants is as important as that between white and non-white immigrants.

For whites, education entrants generally perform better than labour market entrants, compared to natives.

For non-whites, education entrants (who achieve degree-level qualifications) experience comparable levels of earnings and employment to whites (native and immigrant).

Non-white labour market entrants' employment and earnings are, on average, lower than whites. This emphasises the importance of UK education for, in particular, non-white immigrants.

Clark and Lindley do, however, point out the potential selection bias of education entrants, for both whites and non-whites, as a contributing factor to their labour market outcomes. Nevertheless, they do acknowledge a value in exposure to the UK education system: "This has implications for policies which are currently under discussion in the UK regarding the need for immigrants to undertake language and other types of training before being granted the right to remain. Our findings also suggest the importance of the UK education system as a passport to success" 9
The study finds "considerable diversity in the patterns of immigrant earning and employment assimilation...The textbook model of assimilation...is not generally supported by these results" and that an important distinction must also be made, within policy focus, between different types of immigrant 9.
Their findings are consistent with Chiswick (1980) who, although based on a very small sample size, theorised that white immigrants had similar earnings to white natives but non-white immigrants earned considerably less. This was mainly due to lower returns to education and experience.

| To Top |

UK Literature


Barry Chiswick

Ph.D., Columbia University, 1967
M.A., Columbia University, 1964
B.A., Brooklyn College, 1962






"In the year 2000, foreign-born individuals constituted about 9 percent of the working- age population in the UK. On average, immigrants had spent 19 years in the UK in 2000, but there were large differences across the different origin groups."







"Immigrants are heavily concentrated in the capital. In 2000, nine percent of British born whites of working age lived in London, compared with 40 percent of the foreign- born, and 45 percent of UK-born ethnic minorities. The concentration of foreign-born individuals in London increased between 1979 and 2000."











The Commonwealth Immigrants Act of 1962 and 1971 Immigration Act helped generate a higher quality of immigrant to the UK. The removal of freedom of immigration restricted the free flow of immigration from Commonwealth countries to the UK, immigrants would now rely more on their qualifications or educational pursuits for entry to the UK.















427,000 workers from eight EU accession states successfully applied for work in the UK from May 2004 - June 2006. This may sound like a lot, but it represents less than one percent of the UK's population.






The UK's population is getting older, meaning that more working age people are needed to prop up the world's fourth largest industrial economy. By 2026 pensioners will outnumber children by two million. Also as immigrants are on average younger than UK-born people, they are less likely to need to claim a pension and will contribute to the economy for a longer period.






There is a current shortage of highly-skilled people in key sectors such as the NHS, public services and the IT industry which is being met by new migrants.













As the UK's population becomes more highly skilled and educated, there will be more and more jobs that people do not want to do because they are viewed as demeaning. All of these factors have come together to create immigration demands.






























The size of the minority ethnic population was 4.6 million in 2001 or 7.9 per cent of the total population of the United Kingdom.

















Indians were the largest minority group, followed by Pakistanis, those of Mixed ethnic backgrounds, Black Caribbeans, Black Africans and Bangladeshis. The remaining minority ethnic groups each accounted for less than 0.5 per cent but together accounted for a further 1.4 per cent of the UK population.





Ethnic group data were not collected on the Northern Ireland Census in 1991. However, in Great Britain the minority ethnic population grew by 53 per cent between 1991 and 2001, from 3.0 million in 1991 to 4.6 million in 2001.






Net immigration has added 1.5 million people to the population over the last 10 years. Two-thirds of them have come from the continents of Asia and Africa.










In 2006 the total UK population was 60.6m. If net immigration continues at its current rate, as the government actuary's department calculates, the total population in 2031 will be 70m, and by 2081 it will be 85m.










People born overseas account for about 10% of the population. This is up from just more than 6% in 1981 and just over 8% in 2001. This is compared with 24% in Australia, 23% in Switzerland, 19% in Canada and 13% in America. In France that figure is 8%, and 7% in Denmark.














In 2006 the largest numbers of immigrants were born in Africa and the Middle East (3% of the working age population) and the Indian sub-continent (2.5%).




























{Half of the total minority ethnic population were Asians of Indian, Pakistani, Bangladeshi or other Asian origin. A quarter of minority ethnic people described themselves as Black, that is Black Caribbean, Black African or Other Black. Fifteen per cent of the minority ethnic population described their ethnic group as Mixed. About a third of this group were from White and Black Caribbean backgrounds.

Evidence from other countries

Table 8: Outline of Readings



Spain

Spain has changed from a traditionally immigrant sending nation to an immigrant receiving country, with immigrants from Europe, Africa and Latin America accounting for 92% of the total immigration.

Results from the paper by Catalina Amuedo-Dorantes and Sara de la Rica in 2006 'Labour Market Assimilation of Recent Immigrants in Spain' broadly shows the labour market performance of the immigrants in Spain widely depends on the originally nationality of the immigrant, with some of the EU15 immigrants reaching parity in occupational attainment with the similarly skilled natives. The data examined supported the notion of an upward occupational mobility and assimilation on the part of non-EU15 and Latino immigrants as their Spanish residencies lengthen. However in the case of the African immigrants, no clear pattern of occupational assimilation was found. In addition to this, another finding of the paper showed that the employability gap varies by gender (from 14% for male and 7% for females).11

Germany

Labour market performance of foreign workers in the German labour market is traditionally poor. Generally, native Germans earn on average 20% more than guest workers in the country, however it seems to be justified by the fact that the natives on average have better education and training.

Data from the paper 'Labour market segmentation and the earning of German guestworkers' by Amelie Constant and Douglas Massey shows that immigrants are initially less able than natives to translate their human capital into occupational status within the German labour market. Results from the paper indicate that immigrants experience significantly lower returns to education and vocational training than Germans. Additionally it was found that immigrants take on very low status jobs when they arrive and they remain in these jobs for some time before moving upward meaning there is a very high difference in the occupational attainment between the immigrants and the native workers. Lastly, by using the rates of return to experience the paper predicts it takes on average 23 years for immigrants to reach parity with the natives.12

Netherlands

It seems that immigrants in the Netherlands have a diverse range of performance in the labour market. Mainly immigrants from western countries perform very strongly in the Dutch labour market, however for non-western immigrants, the performance is not as strong.

Results from the paper 'Employment Assimilation of Immigrants in the Netherlands' by Aslan Zorlu and Joop Hartog in 2002 states that non-western immigrants significantly less likely to be employed that natives and the ones who are employed tend to be working in very low skilled jobs. Additionally, the evidence from the paper showed that the labour market performance of western immigrants did not differ from that of the natives as their skills are more transferable.

However, there was a sign of assimilation for all immigrants, even though it was much slower for non-western immigrants, the labour market performance significantly improved as their residence increases.

Interestingly, this paper also found that the effect of years since migration on employment was not sensitive to the years of education of the immigrants, fundamentally meaning education is insignificant in affecting the position of the immigrants.15

Sweden

Sweden has attracted a substantial amount of immigrants in the aftermath of WW2. Since 1990 where immigrants accounted for 10% of the population, the decline in economic and industrial growth has removed the need for immigrant labour. Meaning the composition of the immigrant population has changed from European immigrants to refugees and tied-movers from non-European countries.

Evidence from the paper 'Declining employment success of immigrant males in Sweden: Observed or unobserved characteristics?' by Pieter Bevelander and Helena Skyt Nielsen shows a strong pattern of assimilation for many of the immigrants, with employment probabilities of the immigrants is often recorded at as little as 10 years. However for some of the immigrants from the ex-Yugoslavian countries, the labour market performance differed vastly due to the decline in industrial activity, many of these immigrants were found unemployed with very few skills leaving the possibilities of employment assimilation scarce.13

Canada

The performance of immigrants in the Canadian labour market is different from a lot of the other countries examined due to the fact that it is based on a 'points system', meaning Canada tries to match the skills of the immigrants to potential shortages in their own labour market.

The analysis of assimilation of the immigrants in Canada shows poor performance in the labour market. The paper 'The Performance of Immigrants in the Canadian Labour Market' by Michael Baker and Dwayne Benjamin in 2004 shows that there is a 20% wage gap in their first years with very modest rates of assimilation in the long term. If the future assimilation carries on in this fashion (similar to earlier cohorts) it looks that convergence/full assimilation of the immigrants to the natives in Canada is unattainable.14
| To Top |

Other indicators for assimilation

The large and increasing presence of immigrants in some of the largest countries in the world highlights the importance of monitoring immigrant health, since immigrant health (and the health of their descendants) has a larger impact on the overall health outcomes of these domestic populations.

Fundamental evidence shows that immigrants tend to arrive in the country healthier than the natives and that their health slowly assimilates towards that of the natives. This phenomenon was found from studies conducted in the US, Australia and Canada, from these results, the term HIE (Healthy immigrant effect) was created to try and explain why these results were found.

There are many reasons that have been suggested as to why this is the case. The paper 'Unhealthy Assimilation: Why Do Immigrants Converge to American Health Status Levels?' by Heather Antecol and Kelly Bedard suggests that immigrants are subject to selection and therefore are more likely to be in better health while the unhealthier immigrants more likely to return home. Another contradicting factor is acculturation. Exposure to a different environment (in this case the US) causes the immigrants to adopt native born behaviours such as diet and exercise, these have important health implications. This is the reason believed to be behind the BMI (body mass index) assimilation that was found in the US.

Lastly, it is generally thought that increasing income should be correlated to an increase in health for the immigrants, but due to relatively poor labour market performances in countries under this study (especially Canada), this is not usually the case and immigrants arrive healthier and their health deteriorates as the duration of stay increases.

| To Top |

Conclusion

United States


The lessons from the U.S. literature provide potentially valuable lessons to researchers in other contexts. Immigration is a major policy concern in many countries around the world. The relative success of immigrants in assimilating has been an important concern to economists and politicians alike. For almost half a century researchers have come up with different theories with varying degrees of success. Over the past two decades economists' perceptions of U.S. immigrants have shifted.

In the 1970s, immigrants were viewed in a mainly positive light. Chiswick (1978) found that immigrant men eventually earned more or equal wages compared with natives, despite having less education. He concluded that investments in on the job training made up for the gap in formal schooling. Furthermore, early theories on assimilation concluded that immigrants would overtake natives wage levels after 10-15 years from arriving in the U.S.

Subsequent research - particularly by Borjas (1985-2006) has chipped away at both conclusions and gradually led to a more negative picture of U.S immigration. The use of cohort analysis has shed new light on the way in which U.S immigration has shifted in the last 50 years and provides credible answers to the question of assimilation.

By analysing and tracking individual cohorts over time, a clearer picture of age-earning profiles reveals that immigrant wages in the U.S have evolved (declined) since the 1960s and this is due to the change in quality/productivity of immigrants. Skill characteristics of immigrants in the U.S. are strongly related to their country of origin. The use of cohort analysis allowed Borjas to point out that Chiswick's analysis will overstate earnings growth if more recent immigrant arrival cohorts have lower unmeasured skill characteristics than earlier arrivals, as is true in 1980 and 1990.

Subsequent research, most notably by Card showed that a narrow focus on immigrant earnings is misplaced. By looking at the experience of the offspring of immigrants after they settle in the U.S, more questions about assimilation can be answered. Few of the immigrants who come to the U.S. without a high school education will ever catch up with the average earnings of natives. However, most of their U.S.-born children will catch up with the children of natives.

The relatively strong educational progress of second generation immigrants, suggest that even the revisionist view of recent assimilation in the U.S may be overly pessimistic.

United Kingdom


 The three key studies in British immigrant assimilation all expanded on one another, Chiswick only looked at one group from a Labour force survey, Bell broke the immigrants into groups, while Clark and Lindley made an important split between education and labour market entrants. 

References

1 Chiswick, B. (1978), The Effect of Americanization on the Earnings of Foreign-Born Men, Journal of Political Economy, 86, 897-921.

2 Chiswick, B. (1980), The Earnings of White and Coloured Male Immigrants in Britain, Economica, 47, 81-87.

3 Borjas, G. (1985), Assimilation, changes in cohort quality, and the earnings of immigrants.

4 Borjas, G. (1987), Self selection and the earnings of immigrants.

5 Barth et al. (2004), Identifying Earnings Assimilation of Immigrants under Changing Macroeconomic Conditions, The Scandinavian Journal of Economic, 106, 1-22

6 Card, D. (2005), Is the New Immigration Really So Bad?

7 Bell, B. D. (1997), The Performance of Immigrants in the United Kingdom, 333

8 Chiswick (1980), The Earnings of White and Coloured Male Immigrants in Britain

9 Clark and Lindley (2009), Immigrant Assimilation Pre and Post Labour Market Entry: Evidence from the UK Labour Force Survey

10 Clark and Drinkwater (2008), The labour-market performance of recent migrants

11 Catalina Amuedo-Dorantes, Labour Market Assimilation of Recent Immigrants in Spain

12 Amelie Constant & Douglas Massey, Labour market segmentation and the earnings of German guestworkers

13 Pieter Bevelander and Helena Skyt Nielsen, Declining employment success of immigrant males in Sweden, Observed or unobserved characteristics?

14 Michael Baker and Dwayne Benjamin (2004), The Performance of Immigrants in the Canadian Labour Market

15 Heather Antecol and Kelly Bedard, Unhealthy Assimilation: Why Do Immigrants Converge to American Health Status Levels?

| To Top |

Attachments:
fact box image.JPG (image/jpeg)
IntroductionFigure1.JPG (image/jpeg)
Clarke and Lindley equation 1.JPG (image/jpeg)
Clarke and Lindley graph 1.JPG (image/jpeg)
Ed table 1.JPG (image/pjpeg)
Ed table 2.JPG (image/pjpeg)
Ed table 3.JPG (image/pjpeg)
Ed table 5.JPG (image/pjpeg)
Ed table 6.JPG (image/pjpeg)
figure4.bmp (image/bmp)
figure4.JPG (image/pjpeg)
figure4.JPG (image/pjpeg)
clark lindley.pdf (application/pdf)
statlib.jpg (image/jpeg)
claire blurry graph.JPG (image/jpeg)
Fig 1.JPG (image/jpeg)
fig2.JPG (image/jpeg)
fig3.JPG (image/jpeg)
us trends table 1.JPG (image/jpeg)
Intro fig2.JPG (image/jpeg)
Ed table 1.JPG (image/jpeg)
Bell table 1.JPG (image/jpeg)
Bell table 2.JPG (image/jpeg)
Bell table 3.JPG (image/jpeg)
Bell table 5.JPG (image/jpeg)
Education entrants differences bet immigrants and natives.JPG (image/jpeg)
2nd generation table 1.JPG (image/jpeg)
the_graph_112355.jpg (image/jpeg)
us lit.JPG (image/jpeg)
Chiswick.jpg (image/jpeg)
borjas.jpg (image/jpeg)
card.jpg (image/jpeg)
barth.jpg (image/jpeg)
unionjacked up.jpg (image/jpeg)
labour econ table.jpg (image/pjpeg)
Document generated by Confluence on 14 Jul 2010 11:33