Determinants of immigrant apprehensions: The case of U.S. immigration

Buzurukov, Bilol; Lee, Byeong-Wan Article Determinants of immigrant apprehensions: The case of U.S. immigration Economics: The Open-Access, Open-Assessment E-Journal Provided in Cooperation with: Kiel Institute for the World Economy – Leibniz Center for Research on Global Economic Challenges Suggested Citation: Buzurukov, Bilol; Lee, Byeong-Wan (2015) : Determinants of immigrant apprehensions: The case of U.S. immigration, Economics: The Open-Access, Open-Assessment EJournal, ISSN 1864-6042, Kiel Institute for the World Economy (IfW), Kiel, Vol. 9, Iss. 2015-39, pp. 1-25, https://doi.org/10.5018/economics-ejournal.ja.2015-39 This Version is available at: https://hdl.handle.net/10419/122199 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. http://creativecommons.org/licenses/by/3.0/ Received October 11, 2014 Published as Economics Discussion Paper November 13, 2014 Revised October 19, 2015 Accepted October 27, 2015 Published November 17, 2015 © Author(s) 2015. Licensed under the Creative Commons License - Attribution 3.0 Vol. 9, 2015-39 | November 17, 2015 | http://dx.doi.org/10.5018/economics-ejournal.ja.2015-39 Determinants of Immigrant Apprehensions: The Case of U.S. Immigration Bilol Buzurukov and Byeong Wan Lee Abstract The incidence of illicit behaviors among foreign-born residents in the U.S. reveals significant variations across countries of origin. This paper identifies the factors underlying such cross-country variations in immigrant apprehensions by using cross-sectional data from 104 countries averaged across the 2002–2013 period. The results based on multiple regression techniques highlight the home country’s social ills such as corruption, theft and homicide as key determinants. Also included in the regression models, as control variables, are proxies for economic development and geographic proximity to the U.S. Noteworthy is that, alcohol consumption, another explanatory variable, is negatively related to the apprehension incidence in the host country. These results suggest that chronic socio-economic problems such as corruption, crimes, and poverty in the country of origin are central in explaining the crosscountry variations in alien apprehensions in the U.S. JEL F22 D73 K42 A13 Keywords Immigration; corruption; alien apprehension; social ills Authors Bilol Buzurukov, Department of Economics and Finance, Yeungnam University, Gyeongsan, 712-749, Republic of Korea, [email protected] Byeong Wan Lee, Department of Economics and Finance, Yeungnam University, Gyeongsan, 712-749, Republic of Korea, [email protected] Citation Bilol Buzurukov and Byeong Wan Lee (2015). Determinants of Immigrant Apprehensions: The Case of U.S. Immigration. Economics: The Open-Access, Open-Assessment E-Journal, 9 (2015-39): 1—25. http:// dx.doi.org/10.5018/economics-ejournal.ja.2015-39 www.economics-ejournal.org 2 1 Introduction The U.S. is a country that hosts the largest number of foreign-born individuals from around the world with diverse objectives.1 The history of immigration to the U.S. suggests that some immigrants are more likely to engage in illegal practices than others (e.g., Moore and Vedder, 2000). As explained in the Data section of this paper, there are indeed considerable cross-country variations in the average annual number of alien apprehensions per 100,000 aliens from the same country with a standard deviation as large as 1,108.24 for the 2002–2013 period. Therefore, this paper addresses the question of what country-level factors influence the large alien apprehension variations across countries for the U.S. Many social ills are thought to be at play here, including corruption, crime, traffic safety culture, and alcohol use disorders, among others. For example, the persistence of corruption in a country can make corrupt behaviors a general attitude among citizens, and therefore emigrants from corruption-ridden countries may carry such attitudes to their host countries (Dimant et al., 2013) leading to a more frequent brushes with the legal system. To the extent that immigrants bring their culture, way of life and attitudes with them, it is reasonable to hypothesize that immigrants from countries laden with such social ills are more prone to violating rules and regulations in their host countries. Factors such as levels of economic development and education in the home country are also important factors influencing behavioral patterns of immigrants. Although a review of raw data collected hints at all these factors being likely determinants, few formal empirical tests are found in the literature for determinants of immigrant apprehensions. One previous study examined the effect of home country corruption on parking violations by the diplomats stationed in New York City (Fisman and Miguel, 2007). While inspired by this earlier study, the dependent variable selected in this paper is incidence of alien apprehensions, a measure of illicit behavior that is _________________________ 1 According to “Trends in International Migrant Stock: The 2013 Revision” by the United Nations (2013), the U.S. had the largest number of immigrants (45,785,090) in the world as of 2013, accounting for 19.8% of all immigrants. Almost three quarters of Americans today believe that immigration increases the crime rate in the country (e.g., Spenkuch, 2011), and a rise in the crime rate places considerable burdens on taxpayers’ shoulders in funding immigration enforcement (e.g., National Immigration Forum, 2013). www.economics-ejournal.org 3 wider in scope and coverage and more direct, against a number of explanatory variables for social ills. To focus on the impact of social ills on alien apprehensions, a number of control variables are considered, including indicators of economic development, geographical proximity to the U.S., investment relations, and regional and income-group dummies. The results verify that individuals coming to the U.S. from countries with higher incidences of social ills (e.g., corruption, theft, and homicide) are more prone to apprehensions than those from countries less affected by such social ills. On the other hand, apprehension incidences decrease significantly with increases in levels of economic development and education in the country of origin. The rest of the paper is organized as follows: Section 2 provides a literature review. Section 3 introduces the data and variables and Section 4 presents the econometric strategy and empirical results. Section 5 concludes. 2 A Literature Review Few studies have examined illicit behaviors of individuals in foreign countries in relation to social ills such as corruption in their country of origin (Fisman and Miguel, 2007; Dimant et al., 2013; Alesina et al., 2013). Fisman and Miguel (2007) conducted a unique empirical analysis that relates illicit behaviors of foreign individuals in the host country to the level of corruption in their home country by examining the parking behavior of United Nations diplomats in New York City from 149 countries and showed a significant positive correlation between the number of parking violations and the corruption level of the home country. Diplomats from less corrupt countries behaved remarkably well even in the absence of any legal consequences, whereas those from highcorruption countries committed multiple violations. Their findings also show that diplomats from high-corruption countries were significantly more likely to have unpaid parking violations prior to the 2002 enactment of an enforcement initiative confiscating diplomatic license plates. After the enactment of this strict regulation, their unpaid violations showed a sharp decline.2 _________________________ 2 Parking violations dropped by more than 98% after enactment of the enforcement initiative in 2002. www.economics-ejournal.org 4 Dimant et al. (2013) examined changes in the host country’s corruption level resulting from the arrival of immigrants from high-corruption countries by using annual data for OECD countries over the 1984–2008 period and found that general migration had no significant effect on the host country’s corruption level but that immigration from corruption-ridden countries had a significant positive effect – particularly over the long term – on the corruption level of the host country. Alesina et al. (2013) examined gender treatment attitudes of second-generation immigrants (from different cultural backgrounds) who were born and raised in the U.S. and Europe and found some evidence of cultural norms and beliefs being internal to those individuals, suggesting that when individuals leave their external (corrupt) environments behind, their beliefs and values move with them regardless of where they go. They also presented evidence suggesting that immigrants may export some of their corrupt attributes to their host countries. This study uses as the dependent variable the incidence of alien apprehensions in the U.S. by the country of origin to identify factors that explain cross-country variations in the illicit behavior of aliens. Unlike in Fisman and Miguel (2007), the present study’s dependent variable covers the whole alien population to examine a host of indicators of social ills as candidate determinants by using relevant control variables including the level of economic development, geographical proximity to the U.S., and education as well as regional and income-group dummies. The results provide support for the main findings of Fisman and Miguel (2007). 3 Data and Variables Cross-sectional data covering a total of 104 countries selected for the availability of relevant data for the 2002–2013-period are employed.3 Period averages are used for all variables because of missing values for some variables. Table A1 of the Appendix provides summary statistics for all the variables and Table A2 provides a list of sample countries arranged by the immigrant apprehension ratio for the U.S. _________________________ 3 Territories of colonial status, overseas territories and countries with populations less than 100,000 are excluded. www.economics-ejournal.org 5 3.1 The Dependent Variable: The Apprehension Ratio of Immigrants (APR) Immigration data for the U.S. are obtained from two sources: the U.S. Department of Homeland Security and the Migration Policy Institute. To construct the dependent variable, two sets of data are retrieved: the number of foreign-born individuals (FBI) and the number of apprehended foreign-born individuals (AFBI) for each country. Here the term “foreign-born individuals” (immigrants) refers to people residing in the U.S. who are not U.S. citizens at birth. Included in the foreign-born population are naturalized citizens, lawful permanent residents, certain legal nonimmigrants, those admitted under some refugee or asylum status, and people illegally residing in the U.S.4 Data on FBI and AFBI for each country are obtained from the Migration Policy Institute of the U.S. (MPI, 2014) and the U.S. Department of Homeland Security (DHS, 2011–2013), respectively. The DHS provides annual data for 176 countries and territories, and MPI data cover 109 countries. Cross-country AFBI data are aggregate data that include both criminal offenses and immigration violations and supplied without any breakdown by offense type. Dividing AFBI by FBI and multiplying the result by 100,000 provides, APR(A,t) = [ AFBI(A,t) / FBI(A,t) ] * 100,000 (1) where the APR (apprehension ratio) represents the annual number of apprehensions per 100,000 aliens and (A,t) indicates country A in year t. The ratio is multiplied by 100,000 to keep the dependent variable of the regression model in its log form to reduce the skewness of the distribution while mitigating the potential problem of heteroskedasticity. As shown in Table A1 of the Appendix, the APR reveals considerable cross-country variations (from 17.79 to 7,582.33) with a standard deviation as high as 1,108.24 for the 2002–2013 period. _________________________ 4 Defined by the U.S. Census Bureau. www.economics-ejournal.org 6 3.2 Explanatory Variables Corruption: For the measure of corruption, two sets of data are considered: Corruption Control data from the Worldwide Governance Indicators (WGI) and the Corruption Perception Index from Transparency International (TI). These two indicators of corruption produce similar regression results but ‘Corruption Control’ data are selected (e.g., Kaufmann et al., 2014) for their broader country coverage and data completeness with no missing values after the year 2000.5 The index is scaled from –2.5 to 2.5 with –2.5 representing the highest level of corruption (weakest control) and 2.5, the lowest level of corruption (strongest control). The data are multiplied by –1 for a more intuitive interpretation of the scale such that the higher the value, the higher the level of corruption. Theft: Loss (as a percentage of annual business sales) from theft, robbery, vandalism, and arson is used as a proxy for the crime of theft, and data are collected from World Bank Data Tables (World Bank, 2014b). According to the U.S. Government Accountability Office Criminal Alien Statistics (GAO, 2011), more than 400,000 immigrants were arrested because of theft-related offenses in 2011. Homicide: Data on the homicide rate (per 100,000 people) are obtained from the International Homicide Statistics database (UNODC, 2014a). Captured in the data are unlawful homicides purposely inflicted because of some domestic dispute, interpersonal violence, violent conflicts over land resources, inter-gang violence over turf or control, and predatory violence and killing by armed groups. Gurr (1989) argued that U.S. trends in violent crimes are strongly correlated with the arrival of new immigrants. Some studies (e.g., Spenkuch, 2013; Monkkonen, 1989; Lane, 1989) have also found a significant positive relationship between immigrant populations and their involvement in homicide. Alcohol: Abundant reports and statistics link alcohol consumption to traffic accidents, violent crimes, sexual assaults, child abuse, and domestic violence. Given the habit-forming nature of drinking and the concurrent cross-border migration of immigrants’ culture and way of life, it is reasonable to assume the _________________________ 5 The correlation coefficient between corruption control and the corruption perception index is high across years (ρ=0.97 on the average). However, the corruption perception index has many missing data points requiring extrapolation. It also has a serious data consistency problem because of changes in the data construction methodology over time. www.economics-ejournal.org 7 existence of a relationship between the incidence of alcohol-related crimes and offenses in the home country and the alien apprehension ratio for the host country. Because data on the incidence of alcohol-attributable crimes and anti-social behavior are not available for all countries in the sample, per capita alcohol consumption (average daily intake in grams of pure alcohol for individuals 15 and over) is used as a proxy, and data are obtained from the Global Status Report on Alcohol and Health (WHO, 2014b). However, per capita alcohol consumption has a number of limitations as a proxy measure for the incidence of alcoholattributable crimes and offenses. Being merely a per-person average at the population level, it does not account for the magnitude of alcohol use disorders (including alcohol dependence and the risky use of alcohol) in society, including the prevalence of heavy episodic drinking,6 which varies across countries depending on sociocultural attitudes and legal tolerance toward drinking. Distance: To determine the impact on alien apprehensions in the U.S. resulting from illegal border crossings, data on distances from the U.S. to 104 countries (in kilometers) are collected from the website of Distance From To: Distances between Cities & Places (2015). Individuals from countries near the U.S. are assumed to be more likely to make illegal entry into the U.S. than those residing farther away. The APR values for the whole sample of 104 countries are sorted and ordered (shown in Table A2 of the Appendix) to obtain a list of countries with the highest alien apprehension incidence in the U.S., which is topped by Honduras, Mexico, and Guatemala. These top three are followed by other Latin American countries, including Brazil, Nicaragua, Costa Rica, and Ecuador. This suggests the distance factor should be controlled adequately to reveal the impact of key determinants such as social ills on cross-country APR variations. Income: As a proxy for the home country’s income level, per capita GDP is considered, and data are from the United Nations Statistics Division (UNSD, 2012). In addition, data on energy use (kilograms of oil equivalent per capita) from World Bank Data Tables (World Bank, 2014a) are considered. A number of empirical studies (e.g., Mehlum et al., 2006; Krueger and Maleckova, 2003; Fajnzylber et al., 2002; Ehrlich, 1975; Becker, 1968; Fleisher, 1966) have highlighted a significant positive relationship between poverty and crime. _________________________ 6 The WHO (2014b) defines heavy episodic drinking (HED) as the consumption of 60 grams or more of pure alcohol on at least one occasion monthly. www.economics-ejournal.org 8 Evidence from the aforementioned studies indicates that the level of criminal practices is higher for poorer countries. Between these two proxies for income, per capita energy use is used in the regression analysis because of its lower correlation with the rest of explanatory variables.7 Education: Data on education are obtained from ‘A New Data Set of Educational Attainment in the World’ (Barro and Lee, 2014). As a proxy for immigrants’ literacy, average total years of schooling of people aged 15 and over are used. Lochner and Moretti (2004) find a significant negative relationship between schooling and criminal activity. Using FBI data on arrests, they find significant negative relationships of education to murder, assault, and motor vehicle theft. On the other hand, Krueger and Maleckova (2003) and Ehrlich (1975) find a weak relationship between education and crime. Traffic: The poorer the traffic safety culture in the home country, the more likely its immigrants are to violate traffic rules, leading to a higher incidence of traffic-related arrests in the host country. Data on traffic violations, which are assumed to be a reliable proxy for immigrants’ illicit driving practices in their country of origin, are not available, and therefore data on the estimated number of road traffic deaths per 100,000 people are obtained from the WHO (2014a) database. According to the U.S. Government Accountability Office Criminal Alien Statistics (GAO, 2011), traffic violations account for the third-largest portion of total arrested criminal immigrants. Prostitution: Because of a lack of numerical data on prostitution, a dummy variable is used to account for the legality of prostitution as a proxy for prostitution in the home country.8 The variable is coded 0 (prostitution being illegal) or 1 (otherwise). Two sources are cross-checked to construct this binary measure: Prostitution.Procon.org (2010) and ChartsBin (2010). According to the National Arrests for Prostitution and Commercialized Vice (2001–2010) data, the number of arrests for prostitution is larger in states with larger foreign-born _________________________ 7 The correlation coefficient between GDP per capita and corruption exceeds 0.85, and that between energy use and corruption is about 0.62. The use of GDP per capita causes a multicollinearity problem in the regression, making some key variables insignificant. 8 The practice of prostitution is banned by law in the U.S. except for some rural counties in Nevada. www.economics-ejournal.org 15 capture the magnitude of alcohol use disorders and the prevalence of heavy episodic drinking, which can determine the level of alcohol-related problems in a given society. In addition, the socio-cultural environment surrounding drinking and legal tolerance towards drinking can be unfavorable for hard-drinking immigrants in the U.S. Log (Income) and Education are strongly correlated with each other such that the higher the per capita income (as proxied by per capita energy use), the longer is the average total schooling for people aged 15 and over. In the regression model with the two variables, as in [1-1] and [1-2], both coefficient estimates are not significant at conventional levels. To control for collinearity, Log (Income) is excluded from Models [1-3] and [1-4], and Education from Models [1-5] and [16]. Regression results for Models [1-3] and [1-5] without outliers show that both Log (Income) and Education are moderately significant with negative signs. Estimation results for Model [1-3] indicate that for every 1% increase in living standards (proxied by energy use), immigrants’ apprehension ratio will be 0.2% lower. Similarly, Model [1-5] indicates that every one-year increase in schooling is associated with a 0.07% decrease in apprehensions. As shown in Table 1, among all explanatory variables, Log (Distance) has the greatest impact on the dependent variable with a highly significant negative sign across all models, indicating that the closer the home country to the U.S., the higher the immigrant apprehension ratio for that country. For example, estimation results for Model [1-1] indicate that for every 1% increase in the distance between the U.S. and the home country, the apprehension ratio will go down by 0.55% and vice versa. The distance effect is largely due to frequent illegal border crossings from Mexico and other Central American countries to the U.S. Traffic is not significant in any models except for Model [1-4], in which the coefficient is positive and significant only at the 10% level. This implies that, with all other variables held constant, every one-point increase in road traffic deaths is associated with a 0.01% increase in the APR. To the extent that road traffic accident mortality is not a good proxy for traffic safety culture and illicit driving practices, Traffic is not a robust determinant of the APR, and any interpretation of these estimation results should be made with a caution. www.economics-ejournal.org 16 Table 2: Multiple Regression Results with Regional Dummies Dependent Variable: Log (APR) [2-1] [2-2] [2-3] [2-4] [2-5] Corruption 0.235** 0.251** 0.264** 0.237** 0.229** (0.03) (0.019) (0.02) (0.033) (0.039) Theft 0.162 0.19* 0.179 0.2* 0.188 (0.153) (0.092) (0.133) (0.086) (0.104) Homicide 0.023*** 0.017** 0.021** 0.02** 0.022*** (0.004) (0.035) (0.014) (0.021) (0.009) Alcohol –0.012** –0.01** –0.014** –0.013** –0.012** (0.024) (0.042) (0.01) (0.013) (0.025) Log (Distance) - 0.571*** –0.57*** –0.613*** –0.562*** (0.002) (0.003) (0.007) (0.003) Log (Income) –0.145 –0.115 –0.134 –0.133 –0.185 (0.206) (0.311) (0.257) (0.265) (0.139) Education –0.047 –0.034 –0.038 –0.047 –0.034 (0.313) (0.466) (0.436) (0.326) (0.482) Traffic 0.0008 0.0006 0.006 0.008 0.006 (0.931) (0.949) (0.539) (0.428) (0.549) Prostitution 0.099 0.161 0.174 0.157 0.164 (0.547) (0.311) (0.298) (0.336) (0.315) Log (Tourism) 0.037 0.04 0.032 0.031 0.036 (0.398) (0.357) (0.471) (0.482) (0.415) Unemployment -0.005 –0.0003 –0.009 –0.006 –0.009 (0.682) (0.979) (0.48) (0.626) (0.473) Investment –4.32E-07 –7.25E-07 –9.64E-07 –9.34E-07 –8.98E-07 (0.723) (0.546) (0.436) (0.45) (0.466) Americas 0.66*** - - - - (0.002) Europe - –0.46** - - - (0.021) Africa - - 0.133 - - (0.656) Asia - - - –0.09 - (0.65) Middle East - - - - 0.247 (0.391) Constant 5.98*** 10.98*** 11.19*** 10.92*** 11.3*** (0.0000) (0.0000) (0.0000) (0.0000) (0.0000) Adj-R2 0.56 0.58 0.55 0.55 0.56 F-Statistic 11.86*** 11.74*** 10.69*** 10.68*** 10.8*** (0.0000) (0.0000) (0.0000) (0.0000) (0.0000) # of Obs. 101 101 101 101 101 Notes: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Numbers in parentheses are p values. www.economics-ejournal.org 17 Prostitution is not significant at any level across all models, suggesting that the APR in the host country remains the same regardless of whether prostitution is legal or not in the home country. Tourism is also insignificant across all models regardless of the use of data on tourist arrivals or on tourism expenditures. Unemployment and Investment are also not significant across all models, suggesting that the APR for the U.S. is not influenced by the home country’s labor market conditions and FDI activity from the U.S. Table 2 displays the estimation results for log-linear models excluding outliers with five regional dummies. These dummies are intended to account for possible differences in the APR from region-specific characteristics. As shown in the table, the inclusion of these dummies does not change the main results. Corruption, for example, still remains significant with a positive sign across all models. The regional dummy for Americas in Model [2-1] is positive and highly significant. Log (Distance) is excluded from this model because of its strong correlation with the regional dummy (the correlation coefficient is 0.81). The coefficient estimate of the regional dummy for Americas suggests that the apprehension ratio is 0.66% higher for immigrants from Americas.13 By contrast, the apprehension ratio is likely to be 0.46% lower for immigrants from European countries (Model [2-2]). However, regional dummies for Africa, Asia, and the Middle East are not significant (Models [2-3], [2-4], and [2-5], respectively). Table 3 reports the estimation results for regression models excluding outliers with four income group dummies. The main findings remain largely the same. In Model [3-4], Corruption becomes insignificant because of its strong negative correlation with the high-income group dummy (the correlation coefficient is – 0.77), suggesting the exclusion of one of the two variables. _________________________ 13 Apprehended immigrants from Central American countries account for the largest portion of total apprehensions in the U.S. (DHS, 2011–2013). www.economics-ejournal.org 18 Table 3: Multiple Regression Results with Income Group Dummies Dependent Variable: Log(APR) [3-1] [3-2] [3-3] [3-4] Corruption 0.247**(0.022) 0.218**(0.048) 0.224**(0.044) 0.132(0.285) Theft 0.193*(0.09) 0.174(0.132) 0.202*(0.082) 0.185(0.104) Homicide 0.019**(0.021) 0.019**(0.021) 0.021**(0.014) 0.019**(0.021) Alcohol –0.012**(0.014) –0.012**(0.016) –0.014***(0.008) –0.014***(0.007) Log (Distance) –0.531***(0.005) –0.588***(0.002) –0.528***(0.007) –0.554***(0.003) Log (Income) –0.173(0.134) –0.113(0.335) –0.144(0.212) –0.054(0.659) Education –0.065(0.182) –0.044(0.351) –0.053(0.277) –0.053(0.26) Traffic 0.005(0.574) 0.008(0.392) 0.005(0.587) 0.006(0.559) Prostitution 0.174(0.279) 0.177(0.277) 0.153(0.347) 0.164(0.309) Log (Tourism) 0.007(0.865) 0.02(0.659) 0.031(0.486) 0.024(0.574) Unemployment –0.013(0.326) –0.007(0.568) –0.012(0.386) –0.011(0.387) Investment –6.81E-07(0.577) –9.51E-07(0.438) –8.43E-07(0.494) –9.37E-07(0.44) Low-income –0.548*(0.083) - - - Lower-middle income - 0.239(0.195) - - Upper-middle income - - 0.162(0.355) - High-Income - - - -0.481*(0.068) Constant 11.76***(0.0000) 11.3***(0.0000) 11.01***(0.0000) 10.91***(0.0000) Adj-R2 0.57 0.56 0.56 0.57 F-Statistic 11.26*** 10.99*** 10.82*** 11.33*** (0.0000) (0.0000) (0.0000) (0.0000) # of Obs. 101 101 101 101 Notes: *, **, and *** indicate significance at the 10%, 5%, and 1% levels, respectively. Numbers in parentheses are p values. www.economics-ejournal.org 19 Coefficients for highand low-income group dummies are negative and marginally significant. However, income group dummies for upperand lowermiddle-income countries are not significant at any level. Estimates for Models [31] and [3-4] indicate that, with all other explanatory variables held constant, alien apprehension incidences in the U.S. for immigrants from highand low-income countries are 0.48% and 0.55% lower, respectively, than those from other income groups. Regarding the significant negative relationship between the low-income group dummy and the apprehension ratio, a large income gap may serve as a healthy motivation for poor immigrants to avoid dubious activities and illicit practices. 5 Conclusions Incidences of immigrant apprehensions in the U.S. vary widely across countries of origin, but few quantitative studies of underlying factors are found in the literature. This study fills this gap in the literature by providing cross-country analysis and offering new insights into country-level socio-economic determinants of widely varying immigrant apprehension ratios for the U.S. As expected, geographic proximity to the U.S. and the level of economic development (as proxied by either per capita energy use or years of schooling) have significant effects on the dependent variable. With these as control variables, the study focuses on a set of variables for social ills and tests the effects of other potential explanatory variables. The main findings are summarized as follows: First, corruption in the home country is a significant factor in immigrant apprehensions in the host country. Every one-point decrease in the home country’s corruption level produces a 0.25% decrease in the immigrant apprehension ratio for the U.S. Second, effects of theft and homicide in the home country are significant. Per capita alcohol consumption, a weak and problematic proxy for the incidence of alcohol-attributable crimes and offenses, is negatively related to apprehension incidences. Third, although traffic safety culture, investment and economic relations and exposure to the U.S. through tourism are not significant perhaps because of a lack www.economics-ejournal.org 20 of suitable proxies, some regional (Americas and Europe) and income group (lowand high income) dummies are significant. The results suggest a stratified and targeted approach to the design of immigration policies that can better reflect public security concerns and human capital needs in the host country by considering the determinants identified in this study. Acknowledgments: We are grateful to the three anonymous referees for their valuable comments. We also thank Giovanni B. Ramello and Jan Fidrmuc for their insightful comments and suggestions. Their input has helped to improve this work. All remaining errors and shortcomings are our own. References Alesina, A., Giulina, P., and Nunn, N. (2013). On the Origins of Gender Roles: Women and the Plough. Quarterly Journal of Economics 128(2): 469–530. http://www.econ.northwestern.edu/seminars/Nemmers11/Giuliano.pdf Barro R., and Lee, J. (2014). A New Data Set of Educational Attainment in the World, 1950-2010. Journal of Development Economics 104: 184–198. http://www.barrolee.com/data/dataexp.htm Becker, G. (1968). Crime and Punishment: An Economic Approach. Journal of Political Economy 76: 169–217. http://www.nber.org/chapters/c3625.pdf ChartsBin (2010). ChartsBin Statistics Collector Team 2010. The Legal Status of Prostitution by Country. ChartsBin.com, viewed 1st June, 2015. http://chartsbin.com/view/snb DHS (2011–2013). Aliens Apprehended by Region and Country of Nationality. Yearbook of Immigration Statistics: 2011, 2012, 2013. http://www.dhs.gov/yearbook-immigration-statistics-2011-3 , http://www.dhs.gov/yearbook-immigration-statistics-2012-enforcement-actions , http://www.dhs.gov/yearbook-immigration-statistics-2013-enforcement-actions Dimant, E., Krieger T., and Margarete R. (2013). A Crook is a Crook . . . But is He Still a Crook Abroad? On the Effect of Immigration on Destination-Country Corruption, Working Paper Series, No. 2013-14. https://www.wguth.uni-freiburg.de/forschung/data/data/wgsp_dp_2013_03 Distance From To (2015). The distance from the United States to 104 Countries (in kilometers). Distance Between Cities & Places, 2015. http://www.distancefromto.net/country-distance-from/United+States www.economics-ejournal.org 21 Edmark, K. (2005). Unemployment and Crime: Is there a Connection? Scandinavian Journal of Economics 107(2): 353–373. http://www.jstor.org/stable/3441110?seq=1#page_scan_tab_contents Ehrlich, I. (1975). On the Relation between Education and Crime. In: Juster, F.T. (Ed.), Education, Income and Human Behavior. New York: McGraw-Hill. http://www.nber.org/chapters/c3702.pdf Fajnzylber, P., Lederman, D., and Loayza, N. (2002). What Causes Violent Crime? European Economic Review 46: 1323–1357. http://siteresources.worldbank.org/DEC/Resources/What_Causes_Crime.pdf Farrington, D., Gallagher, B., Morley, L., Ledger, R., and West, D. (1986). Unemployment, School Leaving and Crime. British Journal of Criminology 26(4): 335–356. http://bjc.oxfordjournals.org/content/26/4/335.abstract Fisman, R., and Miguel, E. (2007). Corruption, Norms and Legal Enforcement: Evidence from Diplomatic Parking Tickets. Journal of Political Economy 115(6): 1020–1048. http://eml.berkeley.edu/~emiguel/pdfs/miguel_parking.pdf Fleisher, B. (1966). The Effect of Income on Delinquency. American Economic Review 56: 118–137. https://ideas.repec.org/a/col/000090/005839.html GAO (2011). Criminal Alien Statistics: Information on Incarcerations, Arrests and Costs. United States Government Accountability Office, March, 2011. http://www.gao.gov/assets/320/316959.pdf Gurr, T.R. 1989. Historical Trends in Violent Crime: Europe and the United States. In T.R. Gurr (Ed.), Violence in America, Volume 1: The History of Crime. New York: Sage Publications. https://www.ncjrs.gov/App/Publications/abstract.aspx?ID=119356 Hooghe, M., Vanhoutte, B., Hardyns, W., and Bircan, T. (2011). Unemployment, Inequality, Poverty and Crime: Spatial Distribution Patterns of Criminal Acts in Belgium, 2001-2006. The British Journal of Criminology 51: 1–20. http://www.academia.edu/1369607/Unemployment_Inequality_Poverty_and_Crime_ Spatial_Distribution_Patterns_of_Criminal_Acts_in_Belgium_2001-2006 ILO (2014). Key Indicators of the Labor Market. International Labor Organization, 2014. http://www.ilo.org/stat/lang--en/index.htm Kaufmann, D., Kraay, A. and Mastruzzi, M. (2014). The Worldwide Governance Indicators. The World Bank Group. http://info.worldbank.org/governance/wgi/index.aspx#home Krueger, A. and Maleckova, J. (2003). Education, Poverty and Terrorism: Is there a Causal Connection? Journal of Economic Perspectives 17 (2003): 119–144. http://www.rochester.edu/College/PSC/clarke/214/Krueger03.pdf www.economics-ejournal.org 22 Lane, R. (1989). On the Social Meaning of Homicide Trends in America. In: Gurr, J.,H. (Ed.), Violence in America. Volume 1: The History of Crime. New York: Sage Publications. https://uk.sagepub.com/en-gb/asi/violence-in-america/book2728 Lochner, L., and Moretti, E. (2004). The Effect of Education on Crime: Evidence from Prison Inmates, Arrests, and Self-Reports. American Economic Review 94(1): 155– 189. http://eml.berkeley.edu/~moretti/lm46.pdf Mehlum, H., Miguel, E., and Torvik, R. (2006). Poverty and Crime in 19th Century Germany. Journal of Urban Economics 59(3): 370–388. http://emiguel.econ.berkeley.edu/assets/miguel_research/37/_Paper__Poverty_and_Cr ime_in_19th_Century_Germany.pdf Monkkonen, E.H. (1989). Diverging Homicide Rates: England and the United States, 1850–1875. In: Gurr, J.,H. (Ed.), Violence in America. Volume 1: The History of Crime. New York: Sage Publications. https://uk.sagepub.com/en-gb/asi/violence-in-america/book2728 Moore, S., and Vedder, R. (2000). The Immigration Problem: Then and Now. The Independent Review 4(3): 2000. http://www.independent.org/pdf/tir/tir_04_3_vedder.pdf MPI (2014). Immigrant Population by Country of Birth, 2000-Present. Migration Policy Institute Tabulation of Data from the U.S. Census Bureau's 2006 to 2013 American Community Survey and 2000 Decennial Census. www.migrationpolicy.org National Immigration Forum (2013). The Math of Immigration Detention: Runaway Costs for Immigration Detention Do Not Add Up to Sensible Policies. Practical Solutions for Immigrants and for America, August 2013. http://www.immigrationforum.org/images/uploads/mathofimmigrationdetention.pdf Prostitution.Procon.org (2010). 100 Countries and Their Prostitution Policies, 2010. ProCon.org, Online, viewed 30th March, 2010. http://prostitution.procon.org/view.resource.php?resourceID=000772 Raphael, S., and Winter-Ebmer, R. (2001). Identifying the Effect of Unemployment on Crime. Journal of Law and Economics 44(1): 259–284. http://www.jstor.org/stable/10.1086/320275 Spenkuch , J.L. (2011). Understanding the Impact of Immigration on Crime. University of Chicago. http://www.kellogg.northwestern.edu/faculty/spenkuch/research/immigration_crime.p df U.S. Bureau of Economic Analysis (2014). U.S. Direct Investment Abroad, U.S. Direct Investment Position Abroad on a Historical-Cost Basis. January 23, 2014. http://www.bea.gov/international/di1usdbal.htm www.economics-ejournal.org 23 UN (2013). Trends in International Migrant Stock: The 2013 Revision. Department of Economic and Social Affairs, United Nations, 2013. http://www.isr.umich.edu/cps/pewpa/archive/archive_98/19980019.pdf UNODC (2014a). International Homicide Statistics. United Nations Office on Drugs and Crime, 2014. http://www.unodc.org/unodc/en/data-and-analysis/statistics/data.html UNODC (2014b). Total Drug-related Crimes at the National Level, Number of Policerecorded Offences. Crime and Criminal Justice Statistics, United Nations Office on Drugs and Crime, 2014. http://www.unodc.org/unodc/data-and-analysis/statistics/crime.html UNSD (2012). National Accounts Main Aggregate Database. United Nations Statistics Division, http://unstats.un.org/unsd/snaama/selbasicFast.asp WHO (2014a). Road Traffic Death Data by Country. World Health Organization. http://apps.who.int/gho/data/node.main.A997 WHO (2014b). Global Status Report on Alcohol and Health. World Health Organization. http://www.who.int/substance_abuse/publications/global_alcohol_report/msb_gsr_20 14_3.pdf World Bank (2014a). The Energy Use (Kilograms of Oil Equivalent per Capita). World Bank Data Tables, 2014. http://data.worldbank.org/indicator/EG.USE.PCAP.KG.OE/countries World Bank (2014b). Losses due to Theft, Robbery, Vandalism, and Arson (% Sales). World Bank Data Tables, 2014. http://data.worldbank.org/indicator/IC.FRM.CRIM.ZS World Bank (2014c). International Tourism, Number of Arrivals. World Bank Data Tables, 2014. http://data.worldbank.org/indicator/ST.INT.ARVL World Bank (2014d). International Tourism, Receipts (Current US$). World Bank Data Tables, 2014. http://data.worldbank.org/indicator/ST.INT.RCPT.CD www.economics-ejournal.org 24 Appendix Table A1: Descriptive Statistics Variable Observations Mean SD Minimum Maximum APR 104 435.82 1108.24 17.79 7582.33 Corruption 104 -0.02 1.03 -2.45 1.50 Homicide 104 8.98 12.46 0.40 66.45 Theft 104 0.60 0.73 0 3.40 Alcohol 104 32.23 13.59 2.9 62.7 Prostitution 104 0.47 0.5 0 1 Income 104 2065.55 2209.79 35 13918.16 Unemployment 104 8.41 5.31 0.87 33.61 Tourism 104 6943575.14 12158045 500 78576583.33 Investment 104 23153.47 66192.5 -4.91 430806.16 Education 104 8.23 2.58923 1.85 11.85 Drug 74 117.79 175.93 0 827.63 Traffic 104 15.76 8.91504 3 48.4 Distance 104 8864.74 3640.50 1632.45 15289.17 High-income 104 0.32 0.47 0 1 Low-income 104 0.10 0.30 0 1 Lower-Middle Income 104 0.25 0.43 0 1 Upper-Middle Income 104 0.31 0.46 0 1 Americas 104 0.26 0.44 0 1 Europe 104 0.27 0.45 0 1 Asia 104 0.24 0.42 0 1 Africa 104 0.11 0.32 0 1 Middle East 104 0.09 0.29 0 1