Just So 4 All
Just So 4 All is an international social justice advocacy movement.
Its mission is to defeat the Dark Society and establish a Bright Society where individuals live in peace and dignity.
1. Introduction
The concern that immigrants may become public burdens has always been a central component of the debate over immigration policy in the United States (Borjas, Grogger, and Hanson 2008). The perception that immigration adversely affects crime rates led to a change in legislation in the 1990s that particularly increased punishment of criminal aliens. In fact, immigrants have much lower institutionalization (incarceration) rates than the native born – on the order of one-fifth the rate of natives (Butcher & Piehl 2007). Researchers have unanimously proven that immigrants, due to their potential loss of residency, work authorization and travel privileges, are less likely to get involved in criminal activities. However, the information and data used to draw such conclusions does not always enclose illegal immigrants.
Due to that the limited presence of inclusion of illegal immigrants in previous studies, I will build a cross-sectional model that encompasses both the legal and illegal immigration data and the determinants of crime (poverty, age, gender, races and unemployment). My hypothesis is that among immigrants, the illegal ones are more likely to commit crimes. The ultimate goal of the model is to test the public’s belief that immigrants, as a whole, increase crimes in the United States.
In this academic research paper, I am setting out to accomplish several tasks. First, I will argue why attention should be given to the immigration and crime issue. Second, I will synthesize the overarching debate on immigration and crime in the United States. Third, I will build a cross-sectional model that takes into consideration both the legal and illegal immigration data. Finally, analyze the findings and draw conclusions.
2. Problem Statement
People often associate immigration in the United States with increased crime rates. Despite the empirical evidence that has been brought forward, immigration and crime are perpetually juxtaposed. The media, policy makers, and politicians’ frequent speculations on the issue have forged the perception of a casual link between immigration and crime. However, researchers’ failure to differentiate illegal from legal immigrants has let the users of their findings classify immigrants in a uniformed group with limited skills and vulnerable to crimes (Mears 2001). Currently, a controversy exists among researchers as to whether the estimated illegal immigrants’ data should be taken into consideration or not in further studies.
3. Literature Review
The immigration debate encompasses various issues ranging from displacement of American workers in favor of immigrants, increased government funding to secure the borders, welfare depletion, increasing crimes, etc. However, among all, welfare depletion and increasing crimes are the heart of the debate. According to Mears, the immigration debate engenders two conflicting groups. The first group which comprises the uneducated and unskilled Americans, policymakers, and anti-immigration organizations tends to blame immigrants for all social and economic problems. The second group that consists of scholars and “well-rounded” individuals disagrees with allegations against immigrants. It can’t be denied that it is merely human nature that compels us to be inclined to “point the finger” at the other guy when problems arise.
Whether it is illegal immigration or legal immigration, proponents to curtail immigration in general are attempting to create scapegoats. Scapegoats are a necessary evil in any society that is unable to explain, define, or curtail what society perceives as negative forces within itself. A society will identify a group, a sect, a lifestyle to be the “cause” of the particular “ill” that this society faces.
The Welfare Magnets
There is a concern over the possibility that the relatively generous welfare programs offered by the United States have become a magnet for immigrants (Borjas 2002). The magnet hypothesis has several facets. It is possible, for example, that welfare programs attract immigrants who otherwise would not have migrated to the United States; or that the safety net discourages immigrants who “fail” in the United States from returning to their source countries; or that the huge interstate disparity in welfare benefits affects the residential location choices of immigrants in the United States and places a heavy fiscal burden on relatively generous states. In short, the welfare state creates a magnet that influences the migration decisions of persons in the source countries, potentially changing the composition and geographic location of the immigrant population in the United States in ways that may not be desirable.
Implications of Immigrants in Crimes
Currently, a public perception exists suggesting that “immigration to the United States is a primary cause of increased crime rates” (Mears, 2001). This mindset has had a considerable influence on public policy and lawmaking in the nation (Burns & Gimpel, 2000). It is therefore important to examine the relationship between immigration and crime in this country. Immigration and crime affect all Americans, both economically and socially. Examples typically focus on costs, especially that of public services for immigrants (Butcher & Piehl, 1998), which may lead to a decrease in the social well-being and comfort of U.S. citizens. A sense of frustration among contemporary citizens seems to have blossomed out of what some call a “depletion of welfare resources” (Butcher & Piehl, 1998). This so-called depletion includes unemployment effects and additional funding for public services such as law enforcement, as well as “a host of other social problems,” (Butcher & Piehl, 1998). Consequentially, perceptions and stereotypes from the resulting paranoia such as the ones that connect immigration to crime are born. Even though these opinions vary among generations and political ideologies (Burns & Gimpel, 2000) they tend to influence public policy. Hagan and Palloni support this idea, stating that “prison statistics often are used to make the point that the number of immigrants in prison is large” (1999). Actions such as these contribute to the public perception of immigration and crime despite a lack of supporting empirical data.
This statistically unsupported view of immigrant behavior and the consequences it entails seems to have been furthered by early to mid 20th century literature and studies. These early works were very influential in molding and steering the public perception and policies, which have carried through today. Early literature argues that the foreign-born population (immigrants) caused a multitude of social problems and particularly engaged in more criminal acts than the native-born population (Mears, 2001). However, Hacker (1929), looking at the family situation of immigrants said that statistics show many immigrants are unwed individuals, and that unwed individuals (as a group) have significantly higher rates of criminality than those of married individuals, “for family life tends to diminish criminality.” Thus, he claims that “even in this respect, immigration directly tends to contribute to the increase of criminality.” However, this is where his study becomes flawed; Hacker misinterpreted correlation with causation, which plays a significant role in the mistakes of many empirical studies.
As economic and statistical theory advanced, later work began proposing that immigration does not cause crime in itself. It was suggested that immigrants generally posses the same characteristics as the typical criminal, thereby explaining the public’s perception of many immigrants as criminals (Butcher & Piehl, 1998). The beginning of this change came with the questioning of older methods. For example, Colburn and Pozzetta (1974) found that “Anglo-Saxon names have appeared just as often on the rosters of criminal organizations” as the Irish and Italian. Furthering, this is the finding that sentiment towards immigrants is often determined by the economic climate of the nation (Espenshade & Hempstead, 1996). It is studies like Colburn and Pozzetta’s and Espenshade and Hempstead’s that tend to break stereotypes. Some researchers began to look beyond these acknowledgments to create stronger statistical studies, and in doing so, reduced bias in the results by taking into account many factors beyond immigration itself.
Much of recent research argues that the characteristics of criminals are often the same as those of immigrants (young, male, and nonwhite). Based on that assumption, Butcher & Piehl state that when “demographic characteristics” are controlled, by including the necessary independent variables, immigration appears to have no effect on crime rates (1998). For example, “Legal immigrants in El Paso and San Diego are involved in drug crimes at about the same rate as native citizens” (Hagan & Palloni, 1999). Illegal immigrants were not included in that study however, which is something that I will speak about shortly. In other recent studies, it is argued that immigrants tend to be involved in criminal acts more than native-born citizens, but as victims rather than offenders (Mears, 2001). Furthermore, Mears even goes as far as saying “many immigrant groups consistently demonstrate significantly lower crime rates than do native populations.”
The question that most researchers address is whether or not a higher immigration population would lead to higher crime rates. Butcher and Piehl, following Mears’ advice, performed a cross-sectional study over time, and found no relationship between changes in crime and changes in immigration (1998). Their mistake, however, is that they failed to include estimated illegal immigrant populations. This is likely because it is quite difficult to find statistical data on illegal immigration as no census or government survey asks about one’s legal status (Espenshade, 1995). Nevertheless, the illegal immigration factor still plays a significant role in this debate. Although they did not consider it in their study, Butcher and Piehl (1998) did point out that “research and media accounts frequently fail to distinguish legal from illegal immigrants” as well as “crime committed by immigrants and crime not committed by immigrants but that nonetheless is the direct or indirect result of immigration” (Mears, 2001).
4. Model: Overview and Importance
Due to the likely effect of illegal immigration’s absence in past studies, I have decided that it is necessary to include both the legal and the estimated illegal immigration populations as independent variables within any study on immigration. These forms of immigration more than likely vary wildly characteristically, socially and financially. From these differences, consequences may occur such as a significant difference in earnings between legal and illegal immigrants (Espenshade, 1995). This discrepancy in income may factor into an effect on crime in itself.
Based upon recent research, I hypothesize that a larger foreign-born population (illegal and legal) in a state does not cause more crime in that particular state. Although they include many factors that would suggest criminal vulnerability, I do not believe that immigrants’ effect on crime is significant if existent at all. Reasons behind this hypothesis include the fact that any legal immigrant found guilty of any type of crime, set themselves up to a potential loss of permanent residence benefits, which would automatically activate the deportation procedure. In addition, illegal immigrants would more than likely try to avoid any governmental attention as they can be deported with even less formality. I feel that there is simply too little incentive to commit crimes when something as pertinent as residency (both legal and illegal) is on the line. My objective is to show that immigrants do not contribute to criminal activities any more than the native population through an analysis of cross-section data for crime and immigration within the fifty states and Washington DC in the year 2000.
In my analysis, I consider property crimes and violent crimes separately and examine the effect of legal immigration (foreign-born population) and illegal immigration (estimated unauthorized resident population) on each. By separating property and violent crimes and illegal and legal immigrants I can look at each group’s effect on the specific type of crime. This allows me to look at the relationship between the different types of immigrants and crime separately. As such, I am able to correct some of the empirical mistakes made in former studies. I also consider race, poverty, gender, age, and unemployment as independent variables in my study because of their association with crime, as discovered in past studies (most profoundly stated by Butcher and Piehl). I will then interpret relationships and draw conclusions based on a regression analysis of the data.
5. Data Description
The model for studying the relationship between crime and immigration is a cross-sectional one. The independent variables were chosen based upon previous studies (Butcher & Piehl (1998), Mears (2001)). These are the independent variables: Foreign-Born Population (Legal Immigrants), Estimated Illegal Immigrant Population, Unemployment, Male Population, African-American Pop, Asian Pop, Hispanic/Latino Pop, and Native American Pop.
Heteroskedasticity occurs most often in cross-section data where there are large differences in the size between the observations. To counteract this potential for heteroskedasticity, I divided the variables by the total population of each state. This would later explain the reason why a White Test was not initially performed.
I gathered this data from the Federal Bureau of Investigation database, Immigration Officer Rusty Lee (United States Citizenship and Immigration Services) and the U.S. Census Bureau websites. The model contains fifty-one observations (50 states and Washington DC). All previous literature on this subject only studied the number of recent legal immigrants to the United States. I will look at the overall number of legal and illegal immigrants, not just recent and legal immigrants. I will express each variable as a percentage of the states’ population. It should be noted that the USCIS did not report data for eight states because they estimated the number of unauthorized immigrants residing in those states to be less than twenty-five hundred.
5.1 Table 1: Descriptive Statistics
The percent of the population that is legal immigrants has a higher standard deviation compared to the percent that is illegal (5.66 vs. 1.53). Many factors explain such a significant difference. One such reason is that the number of legal immigrants is significant because of program such as The Diversity Lottery, The Visa Waiver Program (VWP), and other international treaties that give foreigners great incentives to immigrate into the U.S. Another reason why foreign-born population has higher standard deviation is that legal immigrants are not evenly dispersed across the United States. They tend to migrate to bordering States such as California, New Mexico, Texas and Florida who all have a much higher than average immigrant population. By contrast, Northern and land-locked States such as Wisconsin, North Dakota, and Kentucky have legal immigrant populations significantly lower than the mean.
To isolate the effect of immigration on crime, I included crime determinants identified in previous studies. Studies have shown that crime is higher in places of higher poverty rates (Colburn & Pozzetta, 1974). Butcher & Piehl state that “Recent immigrants have lower levels of education, lower wages, and lower employment probabilities than the rest of the population. In addition, immigrants are more likely to be Hispanic, male, and young.” To correct for these factors, I included data for poverty, unemployment, the number of people that are between the ages of 18 and 44, male, Black, Hispanic, Asian, and Native American. I feel that each of these variables would be relevant to the study based on Butcher & Piehl’s earlier analysis of immigrants. I chose not to include the data for education levels because it was strongly correlated with poverty, at a level of over 99%. Similarly, I predicted a positive relationship between poverty and crime as well as unemployment and crime. The data for each of these variables comes from the US Census for the year 2000. I expressed each as a percentage of the states’ population (basically per capita).
In looking at the inmate population in the United States, there are significantly more men than women, as well as men aged 18-44 (Butcher & Piehl, 1998). The age range 18-44 is somewhat arbitrary in that the US Census gives population counts for ages 18-24 years and 25-44 years, but also because the mean age of criminals is 32.0 according to Butcher & Piehl (1998). The number of people aged 18-44 was calculated by adding together the data for ages 18-24 and 25-44 years old for each state for the year 2000.
The percent of the population aged 18-44, percent male, and unemployment rates tend to be fairly consistent across the nation as they are general population characteristics. This is reflected in the table as a low standard deviation for each variable. On the other hand though, the populations of each race (Black, Hispanic, Asian and Native American) are not evenly dispersed throughout the nation. Blacks are more concentrated in the South, while there tends to be a greater Hispanic population in states bordering Mexico such as California, New Mexico, and Texas. Asians are more prevalent in states on the West coast as the main ports of entry from Asia are located there. The Native American population is higher in Alaska and Hawaii, as well as in states that have reserved land for Native Americans. These examples explain why the standard deviation for each of the races is so large.
6. Results & Analysis
6.1 Table 2: Regression Results (VIOLENT_CRIME as Dependent Variable)
| Dependent Variable: VIOLENT_CRIME | ||||
| Method: Least Squares | ||||
| Date: 10/21/08 Time: 19:34 | ||||
| Sample: 1 51 | ||||
| Included observations: 51 | ||||
| Variable | Coefficient | Std. Error | t-Statistic | Prob. |
| C | 2.177656 | 2.255575 | 0.965455 | 0.3401 |
| AGED18_44 | -0.003747 | 0.018063 | -0.207412 | 0.8367 |
| ASIAN | -0.004375 | 0.005107 | -0.856554 | 0.3968 |
| BLACK | 0.013488 | 0.003003 | 4.491894 | 0.0001 |
| HISPANIC | 0.008737 | 0.004466 | 1.956366 | 0.0574 |
| ILLEGAL | 0.007974 | 0.032876 | 0.242552 | 0.8096 |
| LEGAL | 0.002427 | 0.008796 | 0.275923 | 0.7840 |
| MALE | -0.041453 | 0.045761 | -0.905861 | 0.3704 |
| NATIVEAMERICAN | 0.012467 | 0.009434 | 1.321530 | 0.1938 |
| POVERTY | -0.006853 | 0.009449 | -0.725321 | 0.4725 |
| UNEMPLOYMENT | 0.096494 | 0.052194 | 1.848757 | 0.0719 |
| R-squared | 0.774869 | Mean dependent var | 0.441554 | |
| Adjusted R-squared | 0.718586 | S.D. dependent var | 0.241443 | |
| S.E. of regression | 0.128081 | Akaike info criterion | -1.083874 | |
| Sum squared resid | 0.656195 | Schwarz criterion | -0.667206 | |
| Log likelihood | 38.63878 | F-statistic | 13.76745 | |
| Durbin-Watson stat | 2.367220 | Prob(F-statistic) | 0.000000 | |
Using Ordinary Least Squares, the regression with VIOLENT_CRIME as the dependent variable produces insignificant t-statistics for the immigration coefficients; I was unable to determine whether or not there was a relationship between crime and immigration. However, only the variable BLACK came out significant, thereby supporting prison statistics. Running the regression with PROPERTY_CRIME as the dependent variable generates significant results using.
6.2 Table 3: Regression Results (PROPERTY_CRIME as Dependent Variable)
In the end, I excluded age and poverty from the analysis because they had very little effect on the estimation results. In other words, there was little change in the coefficients, t-statistics, and p-values when I ran a regression with these two variables and one without them, so I concluded that age and poverty were not necessary variables.
However, the independent variables ASIAN, BLACK, HISPANIC, MALE, NATIVE_AMERICAN, and UNEMPLOYMENT were all important to the regression, despite the fact that HISPANIC was somewhat highly correlated with ILLEGAL and LEGAL. The coefficients for all of the above variables are positive except for LEGAL and MALE. The negative coefficient for male would be surprising, but it is not significant. Only ASIAN, BLACK, ILLEGAL, and LEGAL came up significant. Because I am mostly interested in ILLEGAL and LEGAL, this is not a big concern as these are both at least significant. These results show that each of the independent variables, except for LEGAL and MALE, cause an increase in PROPERTY_CRIME, implying that minorities in general engage in more property crime.
These results indicate that, in the year 2000, a one percent change in the illegal immigration rate increased the property crime rate about 34% on average, while a one percent change in the legal immigration rate decreased the property crime rate about 11%, provided all other independent variables are held constant.
This negative relationship between legal immigration and property crime can be seen in two different ways. It could be the case that legal immigrants decrease property crime rates by preventing property crimes from occurring somehow. A possible theory for this comes from an article in the New York Times, which says that, the family dynamic characteristic to the legal immigrant population “can prevent even the poorest neighborhoods from spiraling into chaos” (Press, 2006). However, it is could also be the case that legal immigrants do not decrease crime themselves but rather add to the population while not adding to crime, therefore decreasing the crime rate. My argument was that illegal immigrants would be more likely to commit crimes than legal immigrants. Although I did not predict a negative relationship between legal immigrants and property crime, I predicted that the relationship would not be positive. So, for either explanation, the result is consistent with my argument and can be further supported by the presumptions used in making my hypothesis.
However, the relationship between illegal immigration rate and property crime is somewhat what I expected, and it is consistent with public perception. I have four possible theories that could be used to explain the positive relationship between illegal immigration and crime. The first is that illegal immigrants characteristically have illegal tendencies based on the fact that their means of entering the U.S. is a crime in itself. A second theory is that illegal immigrants, unable to get proper documentations such as a social security card, are unable to find reliable work, get proper medical care, attend schools, as proper documentation is needed to secure most jobs, as well as take advantage of public services. Again, another possible theory comes from an article in the New York Times, which attributes family-structure to lower crime rates, so therefore because illegal immigrants usually enter the country by themselves, they would be more likely to engage in criminal activities (Press, 2006). Finally, some studies say that because there is such strong animosity towards illegal immigrants, they become more involved in crimes as victims rather than as perpetrators and attribute the increase in crime to that (Mears, 2001). Whatever the explanation for this positive relationship is, the result is consistent with the public perception.
7. Conclusion
In conclusion, my results show that when taking into consideration all other potential variables related to crime, illegal immigration has a positive relationship with property crime and legal immigration a negative relationship, while for violent crime I could not determine a relationship with either variable. This tells us that immigration as a whole does not necessarily cause crime as society commonly perceives. While higher illegal immigration rates are associated with higher crime rates, I cannot assume illegal immigrants cause more crime. However, the data suggests that the typical illegal immigrant is more likely to commit a crime, or to at least influence the increase in crimes somehow. Perhaps if we are trying to decrease crime rates it would be beneficial to make decreasing illegal immigration rates a priority. On the other hand, it appears that the higher the number of legal immigrants, the lower the crime rate (not necessarily the number of crimes, but the crime rate). Although I implied that legal immigrants wouldn’t cause an increase in crime, I did not take into consideration the fact that they would increase population, which would therefore decrease the overall crime rate.
However, there are several limits to these results, including the fact that there is diversity within states as well as among states. For example, Eastern Washington is characteristically different than Western Washington socially, politically, geographically, etc. in the sense that population statistics differ in each area, overall political-party association is significantly different, and geographic terrain varies. Thus, the independent variables including unemployment, foreign-born population (both legal and illegal), and minority race populations likely vary from one region to another within the same state. City or county-level data would be more useful in this regard, but information on illegal immigration was not available at this level. Future studies would potentially have a greater influence in policy-making if they used illegal immigration data at a lower level.
Furthermore, it is a good thing that I distinguished illegal from legal immigration rates, because it ended up showing that there is difference, and that they do have different effects on property crime rates. While most people feel that all immigrants cause an increase in crime, my results show that illegal immigrants might lead to an increase in property crime, while legal immigrants lead to a decrease in crime rates. This study seems to address the discrepancy between previous studies on immigration and crime and public perception. Policy makers and politicians often tend to base their decisions around public perception in order to please people. Nevertheless, this study and many others before mine show that public perception is not always accurate. However, because I did not see any significant results for violent crime rate, I cannot argue whether or not immigration has a positive/negative relationship with violent crime.
8. References
BORJAS, G. J., J. GROGGER, and G. H. HANSON. “Imperfect Substitution between Immigrants and Natives: A Reappraisal.” NBER Working Paper (2008).
Borjas, George J. “Welfare Reform and Immigrant Participation in Welfare Programs.” International Migration Review 36.4, Host Societies and the Reception of Immigrants: Institutions, Markets and Policies (2002): 1093-123.
Burns, Peter, & Gimpel, James A. (2000) Economic Insecurity, Prejudicial Stereotypes, and Public Opinion on Immigration Policy. Political Science Quarterly. 115, 2, 201-225
Butcher, K.F., & Piehl, A.M. (1998). Cross-City evidence on the relationship between immigration and crime. Journal of Policy and Management. 17, 3, 457-493
Colburn, David R., & Pozzetta, George E. (1974). Crime and Ethnic Minorities in America: A Bibliographic Essay. The History Teacher. 7, 4, 597-609.
Cotin, Susan Bibler. (2005). Contesting Criminality: Illegal immigration and the spatialization of legality. Theoretical Criminology. 9, 1, 5-33.
Espenshade, Thomas J. (1995). Unauthorized Immigration to the United States. Annual Review of Sociology. 21, 195-216.
Espenshade, Thomas J., & Hempstead, Katherine. (1996). Contemporary American Attitudes Toward U.S. Immigration. International Migration Review. 30, 2, 535-570.
Federal Bureau of Investigation. (2000). Table 5. Index of crime by state, 2000. http://www.fbi.gov/ucr/cius_00/x1/00tb105.xls
Hagan, John, & Palloni, Alberto. (1999). Sociological Criminology and the Mythology of Hispanic Immigration and Crime. Social Problems. 46, 4, 617-632
Hourwich, I.A. (1912). Immigration and Crime. The American Journal of Sociology. 17, 4, 478-490.
Mears, D.P. (2001). The immigration-crime nexus: toward an analytic framework for assessing and guiding theory, research, and policy. Sociological Perspectives. 44,1, 1-19.
United States Citizenship and Immigration Service. Table 11. Estimated Unauthorized Resident Population, by State of Residence: 1990 and 2000. http://www.uscis.gov/graphics/shared/aboutus/statistics/Ill_Report_1211.pdf
Kristin F. Butcher & Anne Morrison Piehl, 2007. “Why Are Immigrants’ Incarceration Rates So Low? Evidence on Selective Immigration, Deterrence, and Deportation,” Departmental Working Papers 200605, Rutgers University, Department of Economics.
Lee R. (2008, October 17). Personal interview on illegal immigration data. Seattle, WA.
By Issa Ndiaye | Principal and CVO at OVINDI International Group
