Year
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Median Wage in USD
22.259
23.457
24.635
25.872
27.559
28.680
29.017
29.640
30.404
31.522
33.456
34.941
36.477
38.383
40.201
41.446
41.728
41.911
42.823
43.785
45.770
47.599
49.614
49.671
49.158
48.691
49.423
50.396
52.937
53.013
GDP Per Capita
Population growth (annual %)
18.232
0,9
19.078
0,9
20.063
0,9
21.442
0,9
22.879
0,9
23.954
1,13
24.405
1,336
25.493
1,387
26.465
1,319
27.777
1,226
28.782
1,191
30.068
1,163
31.573
1,204
32.949
1,166
34.621
1,148
36.450
1,113
37.274
0,99
38.166
0,928
39.677
0,859
41.922
0,925
44.308
0,922
46.437
0,964
48.062
0,951
48.401
0,946
47.002
0,877
48.374
0,836
49.781
0,764
51.457
0,762
52.705
0,7
54.502
0,8
Unempolyment Rate
7,20%
7,00%
6,20%
5,50%
5,30%
5,60%
6,80%
7,50%
6,90%
6,10%
5,60%
5,40%
4,90%
4,50%
4,20%
4,00%
4,70%
5,80%
6,00%
5,50%
5,10%
4,60%
4,60%
5,80%
9,30%
9,60%
8,90%
8,10%
7,40%
6,20%
Female/male wage
68,10%
69,50%
69,80%
70,20%
70,10%
71,90%
74,20%
75,80%
77,10%
76,40%
75,50%
75,00%
74,40%
76,30%
76,50%
76,90%
76,40%
77,90%
79,40%
80,40%
81,00%
80,80%
80,20%
79,90%
80,20%
81,20%
82,20%
80,90%
Africa America/ White Wage
90%
90%
90%
90%
90%
87%
87%
87%
87%
85%
86%
85%
85%
86%
85%
86%
87%
86%
87%
86%
84%
85%
85%
85%
87%
87%
85%
84%
women/men bechlor's degree or high
72%
74%
75%
76%
77%
76%
75%
75%
75%
76%
74%
74%
74%
74%
75%
76%
75%
75%
74%
73%
74%
75%
73%
75%
76%
Group Project
Globalization and the US:
Immigration and Unemployment in the US
Avenger-CBM600-Page1
Introdution
Key Word: Unemployment Rate, Immigration, Multiple Linear, and Regression Model
The primary purpose of this report is to determine the relationship of immigration and
unemployment in the United States. First of all, we establish the hypothesis that immigration has
a strong impact on the unemployment in the United States. Besides focusing on the immigration,
we are also going to show the other elements that affect unemployment to reveal a completed
concept of unemployment. With the aim of providing a professional report, all the analyses and
conclusions we will make in this report are all based on accurate statistics from government
reports and websites. The three components of the main body of the report are shown as follows:
Introducing the hypothesis and methodology we will use.
Introducing the variables that effect unemployment
The trend of immigration and unemployment in the latest thirty years in the U.S
Using multiple—linear regression model to analyze how the variables affect unemployment,
and determine which the most significant variable is.
We are about to conclude the relationship between the immigration and unemployment and
determine that immigration has a positive or negative impact on unemployment. Last but not
least, we will present some recommendations according to the conclusion. If immigration brings
a positive impact on unemployment, we will suggest that government should encourage
immigration and establish attractive laws and regulations to welcome foreigners. On the other
hand, if immigration has a negative impact on unemployment, the suggestions would be how to
restrict immigration to provide more career opportunities to native residents.
Findings Highlight
U.S. Monthly Unemployment Rate, through Jan. 2005-Oct. 2015 are 16.5%, 15.3%, 15.7%,
18.7%, 24.3%, 25.9%, 24.4%, 24.1%, 22.9%, 19.5%, and 17.2%. The unemployment rates
from 2009 to 2012 are highlighted since they are all above 20%
During the recession from 2009 to 2011, the unemployment rate of people who attained low
Avenger-CBM600-Page2
education was still much higher than the people who had high education level.
The unemployment rate of immigrants was higher than that of Native-born until 2004. And
the unemployment rate of the natives became lower than immigrants again during the period
of 2009 to 2012.
A noticeable increasing of the unemployment rate of both the natives and the immigrants can
be noticed in 2008, and it became more aggressive in the year 2009.
The unemployment rates of the natives from age 25-44 were slightly higher than those of
immigrants during 2010 to 2012, whereas the rates of all natives were actually lower than all
immigrants.
Hypotheses
Before any analysis is launched based on the existing statistics, we can make a conclusion
right behind the previous literature review that the factors may contribute to the changes in
unemployment rate in the U.S. are based on the following elements:
CPI, which is The Consumer Price Index, is a measure of the average change over time in the
prices paid by urban consumers for a market basket of consumer goods and services. In common
view, the CPI may oppositely change with the unemployment rate, which means as the raise of
unemployment rate, the CPI may likely fall because firstly it comes with more difficulty for
employed workers to ask higher wage due to the high unemployment rate. For another, the lower
manufacturing output might be caused due to the higher unemployment in the picture of
economy.
Education levels, according to the unemployment theory, people who have higher education
level are more employable since they carry stronger competition and capacity comparing with
those who had only little education. In other word, in the whole labor force market, higheducated people have less possibility to take lower end jobs that might increase unemployment
rate under the situation of competition.
Age, as survey shows, people among 16-24 years old have significant association with the
status of out-of-labor-force base on the understanding of socioeconomic mental health
disparities. (Mossakowski, 2009)
Avenger-CBM600-Page3
Immigrants Admitted is another factor that affect directly or indirectly the unemployment
rate of a country. As unemployment rate is usually a efficient index to evaluate the economic
condition since good economic condition may provide more available job opportunities to the
labor market. Otherwise, poor economic condition will cause higher unemployment. In the
meantime, a country’s good economic condition will attract mire immigrants to the country,
which could be potentially regarded as a positive relationship with unemployment rate.
GDP Per Capita, as one the most important economic indicators, is gross domestic product
divided by midyear population. The unemployment rate is an issue of the importance that
involved in the employment, which is a dominant factor to affect the domestic income that
closely related to the GDP per Capita.
We assume that all the variables above as a whole as well as individual ones have direct
relationship with the unemployment rate. And the analysis will find out which one of the
variables has the most significant impacts on the unemployment.
Methodology
The analysis adopts the 30-year data from 1985-2014 for the variables we choose. The data
before 1985 is not chosen because there is the lack and discontinuity for some of them, such as
the civilian labor force attained by education. All the data are from three sources, United States
Census Bureau (USCB), Bureau of Labor Statistics (BLS), and the World Bank (WB). Among
them, the unemployment rate and the civilian labor force by age are from BLS, the data of
immigrants admitted is from USCB, and the CPI and GDP per capita are from WB. Since there
is only data for the labor force attained by education from 1992 to 2014 on BLS, the data before
1992 is from USCB. As mentioned above, the unemployment rate is the annual rate of the
unemployment from 1985-2014. The variable immigration means the number of immigrants
admitted by the U.S. government each year. We use the number of the civilian labor force that
have high school and lower education to symbolize the variable education. Same as variable age,
we choose the number of the civilian labor by age from 16 to 24. As for CPI and GDP per capita,
since they are general economic indicator, we use the data directly from WB as the according
variables. In order to determine if the variables relate to the unemployment, we use one of the
statistical analysis methods, multiple linear regression analysis, to build the mathematical model.
Avenger-CBM600-Page4
In the regression model, the unemployment rate is dependent variable and the factors affecting
the unemployment are independent variables. The multiple linear regression analysis is a
statistical method that tries to find if there are the correlations between the dependent variable
and the independent variables as a whole as well as individuals. It can be expressed simply as:
Y = b0 + b1*X1 + b2*X2 + b3*X3···· + bn*Xn
If there is a certain correlation between the dependent and independent variables, the
regression model will determine the coefficient that best suit for the above equation.
Literature Review
We have read through several articles and reports about unemployment. Unemployment is
usually defined as the amount of people who are willing to work but cannot find a job, and it is
essential to understand how unemployment rate can be presented by calculation. The
unemployment rate is defined as the proportion of the citizens who are sixteen years and older in
the United States who out of work but are actually looking for a position (Frumkin, 1998).
However, people who are nursing homes and jails, and serving the army are not included in the
labor force. The unemployment is Expressed as a percentage, by dividing the number of
unemployed workers by the total labor force, and multiplying it by 100%. Unemployment also
comes in different varieties, such as frictional, structural and seasonal. Frictional unemployment
is when workers leave their jobs to find better ones. It usually occurs when workers want to find
better opportunities to go on interviews before they start their new jobs. Structural
unemployment is defined as the mismatch between jobs and skills. And seasonal unemployment
only occurs where there is a limited need for a type of work to be performance during a
particular period during the year based on factors like deadlines or climate.
We also have read some relevant articles, which explain the factors contribute to
unemployment. First factor is location, which has been shown on a larger scale that
unemployment rates are dramatically different in urban and rural areas (Aragon, 2003, Hargrett,
1965, Klasen and Woolard, 1999, Taylor and Bradley, 1997). Second, mobility also affects
unemployment to some extent; for example, immobility causes high unemployment because
Avenger-CBM600-Page5
many positions are not available so that people will be unemployed in a long time. The third
factor is education. There is a negative correlation between education and unemployment. The
people who have received higher education or more professional skills are less likely to be
unemployed than the people with less education and skills. Last factor is wage; especially
minimum wage has many drastic effects on the labor market and unemployment rates (C.
Campbell and R. Campbell, 1969). When minimum wage laws increase, unemployment
increases because if labor costs are higher than they planned so that they will try to limit the
number of employees to control their expenses.
DATA analyze
Regression Model Analysis
After running the multiple linear regression analysis, we find that there does exist certain
relationships between the unemployment rate and the variables we choose. We have the output
result from the first table shown as below. (Please see the detailed data in appendix 1 and 2)
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.8636511
R Square
0.7458932
Adjusted R Square0.6929543
Standard Error 0.0081746
Observations
30
From the above results, the variables we choose as a whole have a high correlation with the
unemployment rate, because the Multiple R is 86.4%, and the R Square is 74.6%, which means
that the dependent variable, unemployment rate, can be well related to and explained by the
independent variables as a whole.
Avenger-CBM600-Page6
However, the correlation between unemployment rate and the variables as a whole does not
mean that it is still true between the dependent variable and the individual variables. In order to
determine the individual relationships between the variables, we need to use other tables from
the regression analysis.
The following table shows the results of relationships between the individual independent
variables and the dependent variable.
Coefficients
Intercept
-0.0522495
Immigration(thousand) -3.13E-06
Lower Educatoin(Labor force)
2.451E-06
age(16-24,Labor force) -7.008E-06
CPI
0.0023268
GDP per capita
-7.164E-06
P-value
0.655052
0.6550885
5.285E-07
0.0806922
0.0031095
0.0067551
The p-value shows the probability of obtaining a result equal to or “more extreme” than
what was actually observed. It means how significant each variable is. We assume that 0.05 is
our significance level. So, the variable with p-value smaller than 0.05 is considered as the
significant one that has impacts on the unemployment rates. In the above table, the variables of
lower education, CPI, and GDP per capita are all significant variables. Since the p-value of the
lower education is 5.285E-7, which is the smallest among the three variables, the lower
education variable has the most possibility affecting the unemployment rate. And the p-values of
CPI and GDP per capita are 0.0031 and 0.0068, respectively. These two variables are also
significant variables that have high correlation with the dependent variable. However, the
variable of immigrants admitted is not related to the unemployment rate based on the regression
analysis, which is not as what we think it should be at the beginning of this paper. Actually, it is
the least significant factor that affects the unemployment rates.
Results
Based on the above analysis, we have the results that the independent variables that we
choose to build this mathematical model as a whole have high correlation with the
unemployment rate from year 1985 to 2014. Among these independent variables, the number of
low education labor force is the most significant factor that has impacts on the unemployment,
Avenger-CBM600-Page7
followed by the variables of CPI and GDP per capita. Since the low education and CPI have the
positive coefficients according to the above table, they have direct relationship with the
unemployment rate, which means that the higher the two variables are, the higher the
unemployment rates. We think that this result is not difficult to understand. According to the
research we read, people with bachelor degree or above is usually much easier to find and hold
the jobs than those with lower education. Thus, the unemployment rates tend to be high if there
are more low-educated workers. The consumer price index (CPI) measures changes in the price
level of a market basket of consumer goods and services purchased by households. It has direct
relationship with Inflation rate which actually reflect how a country’s economy is. A high
inflation rate usually means the economy is not good, which leads to unemployment. Thus, it is
understandable that CPI has a direct relationship with the unemployment rate. As for the GDP
per capita, it is a measure of the total output of a country that takes the gross domestic product
(GDP) and divides it by the number of people in the country. An increase in GDP per capita
means growth in the economy. This explains the result of the relationship between the
unemployment rate and GDP per capita. As showed above in the table, the coefficient of the
variable GDP per capita is negative, which means a higher GDP per capita-good economy-will
reduce the unemployment rate.
Among the independent variables, immigrants and the young labor force are not the
significant ones. Our results show that the variable of immigrants admitted will have the inverse
relationship with the unemployment rate if it is the significant factor, since the coefficient is 3.13E-6. This result is actually same as the literature we read (Frumkin, 1998).
The variable of young age labor force shows an interesting result. Its p-value is 0.00806922,
and based on our preset significance level, it is not considered as a significant factor. However, if
we set the significance level at 0.1 which is usually acceptable, the result will be different. Under
this assumption, it will be a significant variable, and it will affect the unemployment rate
inversely because the coefficient is negative. When the number of young age labor force
increases, the unemployment rate is reduced. This is contradicted to what we have researched
and the most people’s notion because the young workers are much easier to lose jobs compared
to the primary age workers. We think the reason for this contradictory may be from the data we
choose. For the young age labor force, we choose to use the data of labor force from age 16 to
Avenger-CBM600-Page8
24. The age span may be too long. If we use the data from age 16 to 18, the result probably will
be different. The college students are included in this age group of 16 to 24; same does those just
graduates from university. This may be another reason considering the fact that education is a
significant factor that affects the unemployment rate the most.
Although the independent variables we use have relationships with the unemployment rate,
they can be used to approximate the rate in the period of 1985 to 2014 only. It may be
inappropriate to use these data for the period before 1985 or after 2014. And there are other
factors may have impacts on the unemployment rate as well. We hope the relationships we found
in this paper could provide some when forecasting the unemployment rate and making related
policy decisions.
Avenger-CBM600-Page9
Policy recommendations
Attracting high–skilled and well–education immigrants, graduates, and entrepreneurs
For a long time, the strict immigration system in the United States has made it really
difficult to attract the brightest talents. Many intelligent people choose to leave because they
have to wait for years or even decades to get a green card before they start to contribute to
America. However, U.S has cultivated large majority of talented students due to its outstanding
education system. It will be beneficial for us if we retain the talented students we educate before
they leave, especially the most talented students in mathematics, engineering, and technology,
who will bring benefits to the development of science. It is helpful to extend the period of
internship and on—job—training for the graduates from U.S universities, which makes them
realize United States is an ideal country to make the most of their talent and knowledge. Another
recommendation is to make it more possible for talented people to have portable jobs, allowing
them to accept promotions or start new companies when they are waiting for a green card, which
stimulates them to exert more efforts into their career paths in America. Nevertheless, the
extension would cause the loss the jobs of native residents.
Reducing unemployment without causing an inflation
There is a positive correlation between CPI and unemployment. CPI is used as a tool to
measure the level of inflation in an economy. Therefore, reducing the inflation will be helpful to
reduce unemployment. First of all, reducing the money supply by enacting policies will transfer
money from investors ‘pockets to government, which can significantly control inflation. Second,
increasing reserve requirement can be helpful to require bank to have more money and to have
less money to lend to customers. So that customers have less money to spend and the inflation
will be controlled.
Reducing Unemployment by increasing GDP
There is a negative correlation between GDP and unemployment. An increasing GDP
means the economy of the society is moving forward, which includes the decline of
unemployment rate. For example, raising minimum wage will increase job growth, rising profits
and U.S GDP because it increases demand by putting more money into people’s hands.
Avenger-CBM600-Page10
Conclusion
Throughout structure of the whole article, we firstly introduced the subject and the
background investigation of unemployment - the basic profile of immigration, including
immigrant rate of nearly 30 years, as well as the unemployment rate. Then we have had a
literature review to obtain some inspirations for investigation below. Next, we launched a
hypothesis by listing five variables that probably affect the likelihood of the US unemployment
rate with analysis of each one, explaining why they have a direct or indirect impact on the
unemployment rate. Then we thus introduced a linear regression model to prove our hypothesis.
By analyzing the data and models, we have come to the conclusion that class of education has
the greatest impact on the changes of unemployment rate in the U.S.- the lower the education
level, the higher unemployment rate.
Those results based on the linear regression model above might have further functions and
enlightenments for the future political improvement for the United States. The price index
regulation is used to adjust the price level of a nation in order to balance the overall rate of
inflation in the economy and go further step to optimize the output and inputs. The improvement
should aim on education level nationwide especially on the immigration policy. According to the
previous analysis, the government would address stronger rights of competency on making
efforts to attract growing number of immigrants who can be defined as needed personnel.
Avenger-CBM600-Page11
Reference
Adler, W. (1996) Land of Opportunity: One Family's Quest for the American Dream in the Age
of Crack. Retrieved from https://books.google.com
Aragon, Yves, et al. 2003. “Explaining the Pattern of Regional Unemployment:
the Midi-Pyrenees Region.” Papers in Regional Science 82:
The Case of
155-174.
Baeder, J. (2012, October 22) Why U.S Schools Are Simply the Best. Education Week.
Retrieved from http://blogs.edweek.org
Campbell, Colin and Rosemary Campbell. 1969. “State Minimum Wage Laws as
a Cause of Unemployment.” Southern Economic Journal 35:323-32.
Frumkin, N. (1998). Tracking America’s economy. New York: M.E. Sharpe.
Mazzarol, T. (2014, December 29) 6 Ways Governments can Encourage Entrepreneurship.
Agenda. Retrieved from https://agenda.weforum.org
Mossakowski, K. N., PhD. (2009). The influence of past unemployment duration on symptoms
of depression among young women and men in the united states. American Journal of Public
Health, 99(10), 1826-32.
The Saylor Foundation. "Unemployment Rate." pp. 1 [1] Retrieved 20 June 2012
Tong, C. H., Tong, L., & Tong, J. E. (2012). High unemployment in the United States: Causes
and solutions. Competition Forum, 10(2), 74-79.
Avenger-CBM600-Page12
Appendix 1
Raw Data Collection for Regression Model and Analysis
time
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
Unemplo Immigrati
yment
on(thous
rate
and)
7.2%
570
7.0%
602
6.2%
602
5.5%
643
5.3%
1,090
5.6%
1,536
6.8%
1,827
7.5%
974
6.9%
904
6.1%
804
5.6%
720
5.4%
916
4.9%
798
4.5%
654
4.2%
647
4.0%
850
4.7%
1,064
5.8%
1,064
6.0%
704
5.5%
958
5.1%
1,122
4.6%
1,266
4.6%
1,052
5.8%
1,107
9.3%
1,131
9.6%
1,043
8.9%
1,062
8.1%
1,031
7.4%
991
6.2%
979
Lower
Educatoin age(16(Labor
24,Labor
GDP per
force)
force)
CPI
capita
68,375
23,619
107.6 18269.4
68,754
23,367
109.6 19115.0
68,635
22,965
113.6 20100.9
51,800
22,536
118.3 21483.2
54,488
22,134
124.0 22922.5
52,690
21,214
130.7 23954.4
54,693
20,628
136.2 24405.2
51,184
20,454
140.3 25493.0
50,181
20,384
144.5 26464.8
47,191
21,612
148.2 27776.6
46,762
21,453
152.4 28782.2
47,320
21,183
156.9 30068.2
48,516
21,399
160.5 31572.7
48,658
21,890
163.0 32949.2
47,943
22,299
166.6 34620.9
47,446
22,679
172.2 36449.9
47,410
22,690
177.1 37273.6
50,363
22,366
179.9 38166.0
50,573
22,098
184.0 39677.2
50,304
22,268
188.9 41921.8
50,876
22,290
195.3 44307.9
51,112
22,394
201.6 46437.1
50,946
22,217
207.3 48061.5
50,428
22,032
215.3 48401.4
50,332
21,361
214.5 47001.6
50,115
20,935
218.0 48374.1
48,943
20,997
224.9 49781.4
48,099
21,284
229.6 51456.7
47,364
21,381
233.0 52980.0
46,861
21,295
236.7 54629.5
Source: census.gov
Avenger-CBM600-Page13
Appendix 2
Regression output
SUMMARY OUTPUT
Regression Statistics
Multiple R
0.8636511
R Square
0.7458932
Adjusted R Square0.6929543
Standard Error 0.0081746
Observations
30
ANOVA
df
Regression
Residual
Total
SS
MS
F
Significance F
5 0.004707605 0.000942 14.08969436 1.7605E-06
24 0.001603761 6.68E-05
29 0.006311367
Coefficients Standard Error
Intercept
-0.05225 0.115497092
Immigration(thousand)
-3.13E-06 6.91981E-06
Lower Educatoin(Labor
2.451E-06
force) 3.62006E-07
age(16-24,Labor force)
-7.01E-06 3.84282E-06
CPI
0.0023268 0.000707888
GDP per capita
-7.16E-06 2.41688E-06
t Stat
-0.45239
-0.45234
6.771108
-1.8236
3.286923
-2.96423
P-value
0.655051981
0.655088538
5.28507E-07
0.080692246
0.003109517
0.006755082
Lower 95% Upper 95%Lower 95.0%
-0.2906238 0.186125 -0.29062
-1.741E-05 1.12E-05 -1.7E-05
1.704E-06 3.2E-06 1.7E-06
-1.494E-05 9.23E-07 -1.5E-05
0.00086577 0.003788 0.000866
-1.215E-05 -2.2E-06 -1.2E-05
Upper 95.0%
0.186124758
1.11517E-05
3.19833E-06
9.23442E-07
0.003787785
-2.176E-06
Avenger-CBM600-Page14
Appendix 3
U.S. Monthly Unemployment Rate, Jan. 2005-Oct. 2015 (in Percentage)
Source: U.S. Dept. of Labor, Bureau of Labor Statistics, http://data.bls.gov/cgi-bin/surveymost
SUMMARY OUTPUT
Regression Statistics
Multiple R 0.8636511
R Square
0.7458932
Adjusted R Squar 0.6929543
Standard Error 0.0081746
Observations
30
Coefficients P-value
Intercept
-0.0522495 0.655052
Immigration(thousand -3.13E-06 0.6550885
Lower Educatoin(Labo 2.451E-06 5.285E-07
age(16-24,Labor force -7.008E-06 0.0806922
СРІ
0.0023268 0.0031095
GDP per capita
-7.164E-06 0.0067551
Purchase answer to see full
attachment