ASSESSMENT OF HISPANIC IMMIGRANTS' OPINIONS TOWARDS IMMIGRATION AND IMMIGRATION POLICY REFORM PROPOSALS GRACE ROCIO MELO

Transcription

1 ASSESSMENT OF HISPANIC IMMIGRANTS' OPINIONS TOWARDS IMMIGRATION AND IMMIGRATION POLICY REFORM PROPOSALS by GRACE ROCIO MELO (Under the direction of Greg Colson and Octavio Ramirez) ABSTRACT Immigration reform is the most polarizing legislative issues in the US. Surprisingly, despite regular polling evidence of the American public's attitudes towards immigration reform proposals, little evidence has elicited the preferences of the group most affected by any policy changes - legal and illegal Hispanic immigrants. This study presents evidence from a survey and choice experiment of Hispanic immigrants who entered the US illegally on their preferences and willingness to pay for different immigration reform proposals. Policy attributes, which are based on current competing US Senate and House bills, include pathways to legal permanent residence, length of temporary work visas, family visitation rights, and access to medical care. The results quantify the value Hispanic immigrants place on different policy attributes and suggest that longer term work visas are valued on par with legal permanent residence. Furthermore, the ability to legally work in the US is substantially more valued than social services such as medical care and social security benefits. INDEX WORDS: Illegal immigration, Immigration reform, Hispanic immigrants, Choice experiment

2 ASSESSMENT OF HISPANIC IMMIGRANTS OPINIONS TOWARDS IMMIGRATION AND IMMIGRATION POLICY REFORM PROPOSALS by GRACE ROCIO MELO B.Sc., Zamorano University, 2010 A Thesis Submitted to the Graduate Faculty of The University of Georgia in Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE ATHENS, GEORGIA 2013

3 2013 Grace Rocio Melo All Rights Reserved

4 ASSESSMENT OF HISPANIC IMMIGRANTS' OPINIONS TOWARDS IMMIGRATION AND IMMIGRATION POLICY REFORM PROPOSALS by GRACE ROCIO MELO Major Professor: Committee: Gregory Colson Octavio Ramirez Maria Navarro Joshua Berning Electronic Version Approved: Maureen Grasso Dean of the Graduate School The University of Georgia December2013

5 DEDICATION Dedicated to God, my lovely family, and my friends who have supported me during this academic experience and to my professors in Zamorano University and The University of Georgia who have challenged me to go beyond what I thought I could do. iv

6 ACKNOWLEDGEMENTS Furthermore, I want to recognize all my professors in the Department of Agricultural and Applied Economics and in other departments who have planted passion in me about Economic sciences. This was a crucial factor that influenced my decision to apply to Ph.D. Thanks to Dr. Wetzstein, Dr. Ramirez, Dr. Colson, and Dr. Navarro, for sharing not only your knowledge but also for being passionate in your lectures. Thanks to all who believed in me and made it possible to achieve my goal and develop my educational and professional skills. Very special thanks to my family for believing in me and giving me constant spiritual support. Thanks to God for wisdom and strength during this challenging and at the same time exciting process. v

7 TABLE OF CONTENTS Page ACKNOWLEDGEMENTS... v LIST OF TABLES... viii LIST OF FIGURES... ix CHAPTER INTRODUCTION Problem Statement Objectives Organization OVERVIEW OF IMMIGRATION REFORM POLICY PROPOSALS Recent Proposals Comparison between Senate and House immigration reform bills IMMGRATION AND THE AGRICULTURAL SECTOR Foreign-Farm Labor Supply and Immigration Policy Immigration and poverty SURVEY AND CHOICE EXPERIMENT DESIGN Data collection Survey design SAMPLE CHARACTERISTICS Immigration status Socio-demographics Employment information MOTIVATIONS AND OPINIONS ABOUT LIVING AND WORKING IN THE US Original and Current Motivations for coming and staying in the US vi

8 6.2 Perceptions of living and working in the US HISPANIC IMMIGRANTS PREFERENCES FOR IMMIGRATION REFORM POLICY ATTRIBUTES Choice experiment design Mixed logit model General estimation results Willingness to pay estimation results CONCLUSSIONS AND POLICY DISCUSSION Conclusions REFERENCES vii

9 LIST OF TABLES Page Table 1. Sample Characteristics Table 2. Choice Experiment Immigration Policy Attributes and Attribute Levels Table 3. Immigration Policy Attribute Preferences: Mixed Logit Estimates Table 4. Willingness to Pay for Immigration Policy Attributes viii

10 LIST OF FIGURES Figure 1. Motivations for Illegal Entrants to Come (Original motivation) and Stay (current motivation) in the US Figure 2. Illegal Entrants' Perceptions of Living and Working in the US Figure 3. Example of Choice Set (English Translation) Page ix

11 CHAPTER 1 INTRODUCTION 1.1 Problem Statement With the unanimous Supreme Court affirmation in 1875 that "The passage of laws which concern the admission of citizens and subjects of foreign nations to our shores belongs to Congress, and not to the States" (Chy Lung vs. Freeman, 1875), immigration policy has fallen under the exclusive purview of the US Federal government and been cemented in national politics. While debate and public outcry over the scale and ethnic makeup of new immigrants entering the US is not a modern phenomenon, legal and illegal immigration has taken center stage in recent local, state, and national elections and policy discussions, often times along strong partisan lines. Currently, the most contentious legislative immigration issue is how to manage the significant number of otherwise lawful illegal immigrants in the US; the predominance of whom originated from Latin American countries. Following the passage of the Immigration and Nationality Act of 1965, which replaced the national-origins quota system with a preference system that prioritized immigrants' skills and family relationships with US citizens, legal and illegal immigration from Latin America increased tremendously. According to the Department of Homeland Security, between 1970 and 2012 it is estimated that 14.6 million Latin Americans obtained legal permanent resident (LPR) status in the US, which accounts for 43% of all immigrants obtaining LPR during that period. Additionally, the US has experienced a substantial influx of illegal immigrants primarily from Mexico and other Latin American countries. While estimates vary, it is roughly estimated that 1

12 there are 11.2 million illegal immigrants currently in the US with 81% being of Latin American origin (Passel and Cohn, 2011). Recent national polling data suggests the populace and business community are divided on how to address the current population of illegal immigrants along a variety of fronts (e.g., see Dugan, 2013; Newport and Wilke, 2013; Suro, 2005). Even among those who support a pathway to citizenship, there is disagreement regarding the conditions of the process such as: (1) what financial penalties should be levied? (2) how long should the waiting period be prior to becoming a legal permanent resident?, and (3) what rights and services should be afforded to the immigrants during the interim period? Differences in opinions on how to deal with illegal immigration among the US electorate, political parties, and the business community are reflected in the wide variety of competing reform proposals that are currently being debated. Despite regular polling, political discussions, and media coverage of immigration reform, surprisingly an important piece of information is missing from the debate: the opinions and preferences of illegal Hispanic immigrants themselves. To our knowledge, very little research or polling has explicitly elicited the attitudes of Hispanic immigrants in general, and of illegal Hispanic immigrants specifically, regarding the current state of immigration policy and the various proposals for reform. Furthermore, there is no research or data quantifying the Hispanic immigrants' ability and willingness to pay for the different components of the policies being considered. For example, there is not information about what preference/economic tradeoff for current illegal immigrants between a long pathway to citizenship vs. a 10-year work visa with no potential route to legal permanent residence. For the current policy debate, understanding illegal Hispanic immigrants' views and reactions to different policy components is critical for finding the right reform program balance. 2

13 In this study we present unique results eliciting the preferences of Hispanic immigrants who entered the US illegally for different immigration reform proposal elements. In lieu of using traditional polling methods we employ a choice experiment approach that allows us to measure and compare the monetary (dollar) value that illegal immigrants place on a variety of proposal attributes such as 3 vs. 5 year visa lengths, eligibility for a green card, and family visitation rights. To our knowledge, this is the first application of choice experiment methods in the realm of immigration policy. 1.2 Objectives The goal of this study is to explore the preferences of Hispanic immigrants for the main attributes of proposed immigration reform policies. Specifically, their individual valuation of those attributes is estimated through the use of stated choice experiments (CE). Even if it is assumed that the policy makers objective is to maximize the welfare of current US citizens, the different potential provisions of an eventual Bill imply social costs and benefits for those citizens. The information afforded by this study could help weight those costs and benefits versus the value that the associated provisions have from the perspective of the immigrants. 1.3 Organization In the remainder of this paper we first provide an overview of the current immigration reform proposals that served as the basis for our choice experiments and highlight critical literature focusing on the impact of immigration on the agricultural sector. Then we present an overview of our survey and the sample characteristics, and discuss the immigrants' motivations for coming and settling in the US and their perceptions of living in the US. Next we outline the details the choice experiment and use regression analysis to assess preferences and values for 3

14 different immigration reform policy attributes. Finally, we end with conclusions and policy implications. 4

15 CHAPTER 2 OVERVIEW OF IMMIGRATION REFORM POLICY PROPOSALS 2.1 Recent Proposals Proposals floating in the public domain for reforming immigration range from full and immediate amnesty and citizenship for all illegal immigrants to deportation of all immigrants unlawfully living in the US. In the intermediary, several competing proposals have gained political traction and contain a number of similarities and distinct differences. Most recent and prominent are the Border Security, Economic Opportunity, and Immigration Modernization Act of 2013 (S.744, 2013) passed by the Senate on June 27, 2013 and a series of smaller individual bills marked up in the House Judiciary and Homeland Security Committees. These include the Agricultural Guestworker Act (H.R. 1773), the Border Security Results Act (H.R. 1417), the Legal Workforce Act (H.R. 1772), the Strengthen and Fortify Enforcement Act (H.R. 2278), and the Skills Visa Act (H.R. 2131). 2.2 Comparison between Senate and House immigration reform bills Both the Senate and House proposals address a wide-range of immigration reform issues including temporary visas for agricultural and non-agricultural workers, paths to legal permanent residence (e.g., green card), and border security. However, there are several key differences between the proposals. For non-agricultural workers who qualify, the Senate bill creates a new registered provisional immigrant status and offers a renewable six-year visa. After at least ten years, those with provisional immigrant status are eligible to apply for legal permanent residence 5

16 and may be able to naturalize three years later. Costs borne by the applicants include a $1000 penalty and application and processing fees. In contrast, the series of House bills would not establish a path to lawful permanent residence, but allow for unauthorized residents currently working in the US to apply for temporary worker programs. The Senate bill would also expand family-based visas, while the House bills make no such provisions. For agricultural workers, the Senate bill would establish a blue card visa and allow holders to move to lawful permanent resident status after at least three years with a penalty of $400 and applicable fees. As well, the current H-2A program would be replaced with three year visas (with one renewal possible) for contract and at-will employment. Under this program, dependents would not be permitted to accompany the workers to the US. The House bills also make special provisions for agricultural workers, but do not offer any path to permanent residence. Under the House bills the H-2A program would be replaced with a new H-2C agricultural worker program. Immigrants would be permitted to engage in contract employment in seasonal and temporary work for 18 months and three years in non-seasonal jobs (with a potential 18 month extension). Like the Senate bill, no dependents would be allowed. Contrasting the Senate and House bills, several key differences in features are apparent. The Senate bill offers a path to lawful permanent residence after 3 to 10 years depending upon the nature of the employment while the House bills do not. The Senate bills establish financial penalties while in the House bills are not clear on these costs. Finally, both bills have specific instructions not allowing dependents for temporary agricultural workers; however, for blue card visa holders who are at some point able to obtain permanent residence under the Senate bill, the door for dependents eventually coming to the US is more open. 6

17 CHAPTER 3 IMMGRATION AND THE AGRICULTURAL SECTOR 3.1 Foreign-Farm Labor Supply and Immigration Policy Among the industries that rely on a significant percentage of foreign born workers such as the accommodation, construction, and food service sectors, agriculture and meat processing are particularly sensitive to current and potential future laws governing immigration. According to the 2010 National Agricultural Workers Survey (NAWS), 78% of all crop farmworkers were foreign born with the vast majority (71%) originating from Mexico. The prominent role of Hispanic workers in rural America and in labor intensive agricultural operations such as fruit and vegetable production has raised a number of concerns regarding the status quo of immigration and potential reforms (see Martin, 2012; Martin and Calvin, 2010 for an overview). These concerns include the relationship between immigration policies and (a) farm labor supply, (b) poverty, (c) farm worker wages, and (d) rural economic impacts. Evidence suggests that although the North American Free Trade Agreement (NAFTA) and the Immigration Reform and Control Act led to an increase in the supply of migrant labor to the US from Mexico (Boucher et al., 2007; Luckstead, Devadoss, and Rodriguez, 2012), it is believed that market forces and the changing agricultural labor market in that country will lead to a smaller supply of farm labor for the US (Taylor, Charlton, and Yúnez-Naude, 2012). Further, evidence suggests that the agricultural (nursery) labor supply is highly elastic (Buccola, Li and Reimer, 2011), farm workers move in and out of agriculture frequently (Tran and Perloff, 2002), 7

18 and immigrant flows increase net outmigration of natives muting wage and employment impacts (Partridge et al., 2008). Looking at the last time the US government provided a pathway to citizenship under the Immigration Reform and Control Act (IRCA) in 1986, research suggests that IRCA increased agricultural labor costs and motivated additional illegal immigration (Martin, 1994; Phillips and Massey, 1999; Taylor and Thilmany, 1993). However, empirical simulation studies estimate that tightening of immigration policies could result in a long run reduction of 2-4% in the acreage of labor intensive crops (Zahniser et al., 2012) and decrease farm exports by 181 million dollars (Devadoss and Luckstead, 2010). Alternatively, an increase in the availability of nonimmigrant farmworkers could have a large positive impact on the supply of labor-intensive products such as fruits, vegetables, and nursery crops (Gunter et al., 1992, Zahniser et al., 2012) Immigration and poverty In terms of poverty and farm wages, there is strong evidence that immigration policies and illegal immigration have an impact. Undocumented workers suffer a significant wage penalty due to their illegal status (Isé and Perloff, 1995; Kossoudji and Cobb-Clark, 2002) and gaining legal permanent resident status would lead to a wage increase (Kandilov and Kandilov, 2010). Looking at the relationship between immigration and poverty (Martin and Taylor, 1998, 2003), evidence suggests that there is a cyclical relationship between immigration and farm employment that led to a decrease in rural poverty in the 1980s but an increase in the 1990s. Further, for rural communities there are clear economic tradeoffs between the benefits and costs associated with industries that attract large populations of Hispanic workers such as meat packing and processing plants (Artz, Jackson, and Orazem, 2010). 8

19 CHAPTER 4 SURVEY AND CHOICE EXPERIMENT DESIGN 4.1 Data collection To elicit legal and illegal Hispanic immigrant opinions and valuations for immigration reform policy features a questionnaire- in Spanish was conducted in the summer of Given the aim of the study to assess preferences of first generation Hispanic immigrants who might have a direct stake on the outcome of immigration reform, participants were limited to native Spanish speakers. Spanish speaking adults of Hispanic origin were invited to participate in the project and surveyed at select local businesses, churches, Hispanic organizations, and farmers markets in areas around Northern Georgia 1. This region of Georgia is home to a substantial population of Hispanics working in agriculture (mainly poultry, ornamental, and meat-packing plants), landscaping, and service professions such as cleaning and food services. Its Hispanic population has rapidly grown during the last two decades and Georgia is estimated to be the state with the sixth highest share of illegal immigrants in the labor force (Passel and Cohn, 2011). Given the potential sensitivity of issues related to immigration for both legal and illegal immigrants alike, several key measures were employed for the survey. In order to minimize any discomfort to the participants and avoid potentially biasing responses, no identifying information was collected. The two surveyors were native Spanish-speaking graduate students born and 1 For the selection of these locations, a variety of non-probability sampling methods were employed. However, we select the first locations based upon a list of contact information of different Hispanic organizations and Hispanics leaders in this region of Georgia, provided by the Latin American and Caribbean Studies Institute (LACSI) of the University of Georgia. 9

20 raised in Latin American countries which allowed them to better relate to the participants and convey the importance of truthful responses in the anonymous survey. Each participant was provided a $10 gift card as compensation for their time. 4.2 Survey design Most critically, survey respondents were not explicitly asked their legal status, past or present. Respondents were simply asked to classify their status upon entering the US from a list of legal alternatives (e.g., permanent resident, border crossing card, non-immigrant visa). To the surprise of the authors, but giving credence to the other sensitivity conscious mechanisms employed in crafting and conducting the survey, a high percentage of respondents voluntarily stated that they had entered the country illegally. While originally the authors had intended to use a variety of survey questions to infer which respondents might have entered the US illegally, the voluntary openness of the survey participants rendered this inference unnecessary. It is important to note that while we have confidence in the veracity of those who voluntarily stated that they entered illegally, we cannot infer that all individuals who did not make such a declaration entered the US legally. As well, for those who affirmed that they entered illegally, we cannot conclusively infer their current status beyond the high likelihood of their continued illegal status or at least lived in the US illegally for many years, considering that is has been 27 years since the last major amnesty. Regardless, the large sample of verified illegal US entrants presents a unique opportunity to elicit information from individuals who have or can easily put themselves in the hypothetical position of having a direct stake in US immigration reform. Of the 209 individuals who completed the full questionnaire lasting approximately 20 minutes, 139 (67%) voluntarily stated that their l immigration status upon entering the US was illegal. Over the course of the survey, participants were asked questions on a variety of topics 10

21 including (1) their socio-demographic and immigration characteristics, (2) their original motivation to come to the US and their current motivations to stay in the US, and (3) their preferences towards various immigration policy features via a choice experiment. The demographic and immigration questions were based on standardized instruments including the National Agricultural Workers Survey conducted by the US Department of Labor. The motivation questions were modeled after previous work by Alvarez (1987), Alsalam and Portes (2005), and Portes and Rumbaut (2006). The immigration policy attributes included in the choice experiments were selected upon the review of the competing Senate and House bills discussed in the previous section. In the following sections we describe each component of the questionnaire and summarize the responses and findings from their statistical analysis. 11

22 CHAPTER 5 SAMPLE CHARACTERISTICS 5.1 Immigration status Table 1 presents summary statistics of socio-demographic information, employment data, and immigration history for (a) the Entire Sample, (b) the respondents who voluntarily stated they entered the US illegally (Stated Illegal Entry), and (c) the respondents who did not state they entered the US illegally (Did Not State Illegal Entry). From a list of five options (Permanent Resident, Border Crossing Card, Temporary Resident, Pending Status, and Other) the vast majority (78%) of respondents stated that their initial status when coming to the US was Other. Of those who responded Other, 85% stated voluntarily that they came to the US illegally (67% of the total sample) and 15% did not volunteer additional information. Of those who did not state that they came to the US illegally, 50% arrived in the US as a temporary resident and 13% as a permanent resident (green card holder). 5.2 Socio-demographics In terms of the socio-demographics of the sample, the majority of participants study were born in Mexico (77%), followed by El Salvador (11%), and Peru (6%). There were no respondents from Nicaragua, Cuba or Haiti, the three nationalities that were granted targeted amnesties after A comparison of the demographics of those who stated that they had entered the US illegally and those who did not, reveals that a statistically higher percentage were born in Mexico (86% vs. 57%). Education and self-rated English abilities are statistically lower 12

23 among the stated illegal entrants (9.93 vs years of school, and 2.85 vs self-rating of current English abilities on a 5 point scale). As well, the average monthly income level and the number of years living in the US were lower but not statistically different ($ vs. $ and years vs years). According to Taylor et al. (2011), the median age of unauthorized immigrant adults was 36 years in 2010 and about two-thirds of this group had lived in the US for at least a decade. Overall, these statistics are similar to the socio-demographic characteristics of the Hispanic population in the U.S. reported by the 2012 Census Bureau American Community Survey (ACS). In the 2012 survey the mean per capita income was nearly $1300 per month, 61% of the population was over 21 years old, only 9% had a bachelor s degree, and 4% worked in agriculture (excluding packing houses). Two other characteristics of the sample reflecting life as an immigrant in the US are important to note. A substantial percentage of the respondents had a spouse, child, or both not residing in the US. Of those who were married, 58% had a spouse not living in the US and 32% of those with children under the age of 18 also had their children not living in the US. 13

24 5.3 Employment information The difference in education, English abilities, and income between stated illegal entrants and non-stated illegal entrants is further reflected in the employment characteristics of the sample. Stated illegal entrants are more frequently employed in the agricultural industry including field operations and meat packing plants. Those who did not state that they entered the US illegally are more likely to be employed in cleaning or food service jobs. This is also further reflected on the first job respondents said they held in the US. About 27% of stated illegal entrants' first job in the US was in the agricultural industry compared to 17% for the rest of the sample. Looking at the entire sample, we find that 24% of individuals' first job in the US was in the agricultural industry, but this percentage declined to 9% in the present as immigrant workers moved to other industries. One of the concerns in the agricultural industry is that immigration reform, and particularly a path to citizenship, could lead to a decrease in the farm labor supply. The survey responses tend to support this view. Thus, unless pay and working conditions improve, an amnesty program could motivate illegal immigrants to leave farm jobs as occurred after IRCA (Martin, 1994). 14

25 CHAPTER 6 MOTIVATIONS AND OPINIONS ABOUT LIVING AND WORKING IN THE US 6.1 Original and Current Motivations for coming and staying in the US Two of the most controversial components of the competing Senate and House proposals for immigration reform are (a) whether a path to legal permanent residence should be offered and (b) the length and legal rights associated with the temporary work visas. Any reforms to immigration policies (or maintaining the status quo) could potentially affect the labor supply and the well-being of immigrants as well as the US-born population. Thus, several questions were included in the survey in order to assess the motivations for illegal Hispanic immigrants to come to the US, how these motivations change with time, and how work conditions in the US are perceived by them. Focusing just on respondents who stated that they had come to the US illegally, Figure 1 presents summary statistics for two questions asking participants to rate on a 5-point Likert scale from unimportant (1) to very important (5) their original motivations for coming to the US and why they would like to stay in the US now. Respondents were asked to rate six different options: Availability of Jobs and Higher Wages, Safety, Better Education Opportunities, Family Reunification, Opportunity to become a Citizen, and Access to Better Medical Care. Clearly, availability of jobs and higher wages was by far the greatest motivation for individuals to first come to the US illegally (average rating 4.21). Surprisingly, opportunity to become a citizen was the second lowest rated motivation (average rating 2.31), thus giving credence to the argument suggestions that illegal immigrants from Latin America mainly come looking for jobs and better 15

26 wages, not citizenship or free medical care. Our findings that better jobs and wages are the driving force behind illegal immigration from Latin America is anecdotally supported by recent statistics of a steep decline in the number of illegal immigrants living in the US that occurred during the last great recession of (Passel, Cohn and Gonzales-Barrera, 2013). More generally, the results suggest that the only other benefit that Hispanic immigrants are keenly interested in when coming to the US is better educational opportunities for their children. However, contrasting these ratings with current motivations to stay in the US presents a markedly different picture. All factors were rated higher and the difference between the highest rated factor (jobs and wages) and the lowest rated factor (medical access) shrank substantially. As might be expected, over time living in the US expands the importance of the different opportunities and benefits that US residents generally enjoy. While statistically lower than availability of jobs and higher wages and not statistically different from the other motivations, citizenship was rated as the second most important motivation to remain in the US. The statistical equivalence among the different motivations (except jobs and higher wages) for wanting to stay in the US points to another important result. Despite the increased risk of harassment and even possible deportation created by the enactment of Senate Bill 160 in 2011 in the state of Georgia, immigrants still do not view citizenship as being substantively more important than a job, living in a safer environment, or having access to educational opportunities for their children. 6.2 Perceptions of living and working in the US In addition to understanding their motivations for migration and remaining in the US, we included several survey questions to assess illegal Hispanic immigrants' opinions about working in the US under current regulations. Figure 2 presents a summary of the responses to those 16

27 questions. Among respondents who stated that they entered the US illegally, approximately 40% feel that they are not paid fairly for their work and 46% feel like they have been discriminated against. Interestingly though, only about 18% disagreed or strongly disagreed with the statement "My employer treats me fairly". Together, this indicates that a substantial share of Hispanic immigrants feel that they are economically penalized and marginalized by society, but not by their employers themselves. The respondents affirmed that the lack of legal immigration status and poor English skills are a primary obstacle to integration in society, which is consistent with the findings of Hakimzadeh and Cohn (2007). Finally, we asked participants about their perception of the idea of becoming a US citizen, and found that close to 90% agree or strongly agree with the idea. This is in line with previous results from Latino Decision polls that indicated that 9 in 10 undocumented Hispanic immigrants would like to become citizens if afforded the opportunity (Barreto and Segura, 2013; Gonzalez-Barrera et al., 2013). However, as we illustrate in the following section, this simple citizenship survey question fails to capture the tradeoff between other immigration reform proposal features and their relative importance to Hispanic immigrants. 17

28 CHAPTER 7 HISPANIC IMMIGRANTS PREFERENCES FOR IMMIGRATION REFORM POLICY ATTRIBUTES 7.1 Choice experiment design In the latter part of the survey, we implemented a choice experiment in order to assess immigrant attitudes for the different attributes that constitute the Senate and House immigration reform proposals. One of the advantages of a choice experiment, in lieu of traditional yes/no polling type questions, is that it allows to quantitatively measure the tradeoff between attributes of interest. While choice experiments are commonly employed in the agricultural (e.g., Lusk, Roosen, and Fox, 2003; Lusk and Schroeder, 2004; Hu, Hünnemeyer, Veeman, and Adamowicz, 2004) and environmental (e.g., Revelt and Train, 1998; Hanley, Wright, and Adamowicz, 1998) research as well as in other social sciences like marketing, to our knowledge, this is the first use of a choice experiment to analyze immigration policies. In the choice experiment, the Hispanic immigrants were each presented with 12 decision scenarios. In each scenario, they were asked to make a choice between two different immigration policy proposals. In each proposal we varied five different key attributes that were selected due to their controversial nature and prominent role in the competing House and Senate bills. These included whether there would be a path to legal permanent residency, family visitation rights, the length of work visas, eligibility for federal benefits, and the cost to the participant. Each of these five attributes was varied according to the three different levels summarized in Table 2. The 18

29 choice scenarios were generated using NGENE software to create an efficient design 2 (an example choice scenario is presented in Figure 3). As previously stated, the scenarios were individually presented and explained to the participants in Spanish by a Hispanic native speaker born and raised in a Latin American country. 7.2 Mixed logit model To model choice experiment responses we estimated a mixed logit model with random and correlated coefficients, an approach that has become increasingly standard in the choice experiment research. The mixed logit model, as opposed to the simpler multinomial logit model, relaxes the assumption that all respondents have the same preferences for immigration policy attributes by allowing for random taste variation among individuals. Formally, the model can be expressed and estimated as follows. Each participant was faced with 12 choice scenarios 3 ( ) consisting of 24 different alternatives indexed by. The utility of individual from alternative in choice scenario is specified as where is a vector representing the attributes of alternative in choice scenario and is a vector of unknown parameters. The error term is assumed to be distributed i.i.d. extreme value over individuals, alternatives, and choice scenarios. Given the distributional assumption, utility maximization results in the probability of individual choosing alternative in choice scenario being the same as in a conditional logit model 2 An experimental design is called efficient if the design yields data that enables estimation of the parameters with as low as possible standard errors. NGENE generated these 12 different choice scenarios according to D-efficient criteria for a panel mixed multinomial logit.. 3 For our experiment of two alternatives, each having five attributes with three attribute levels, the total number of possible choice situations is 3 2x5 =59,049. The number of choice sets has to be at least 5/(2-1)=5, however 5 is not a number divisible for the number of attribute levels (3), thus we have set 12 choice sets, which were selected among the 59,049 choice situations using an algorithm that follows the efficient design criteria. 19

30 (1) ( ) ( ). Preference heterogeneity can be incorporated in this model by assuming that the vector of unknown parameters is random instead of fixed. Specifically, letting ( ), where is a probability distribution function, is the mean vector value of the distribution, is an i.i.d error term vector, and is a parameter covariance matrix, the choice probability becomes (2) ( ) ( ) ( ). Although there is no closed form solution for equation (2), several standard statistical packages (e.g., NLOGIT, STATA, and SAS) now have built in routines to estimate this mixed logit model using simulated maximum likelihood. We estimate our mixed logit model with random and correlated coefficients using STATA 12. All parameters are assumed to be normally distributed except for the cost attribute, which is assumed to be a fixed value. The coefficient estimates are reported in Table 3 for each of the three sample categories: (1) the Entire Sample, (2) respondents who stated they entered the US illegally (Stated Illegal Entry), and (3) respondents who did not state that they entered the US illegally (Did Not State Illegal Entry). 7.3 General estimation results In each of the three models reported in Table 3, the signs of the coefficient estimates fall in line with expectations and the majority of the attributes are statistically significant at 1% level. The cost coefficient is negative which is consistent with expectations that immigrants prefer, holding all other factors constant, immigration policies with lower costs. All other coefficients are positive, indicating that immigrants prefer longer visa lengths, eligibility for a green card, eligibility to possibly bring family for visits, and access to social security and medical care. Interestingly, the only coefficient that is not statistically significant is whether spouses and 20

31 children would be potentially eligible for a visa to visit. However, the corresponding standard deviation coefficient is statistically significant across all models, which indicates that thre is substantial heterogeneity among respondents in regard to their preference for this immigration policy attribute. In all likelihood, this is due to the sample characteristic that only 32% of respondents had a child located outside of the US. While those individuals should be much more likely to value this particular policy attribute, the other 68% would not. The models presented in Table 3 estimate the average effect for all respondents regardless of whether or not they have family living in their home country. In light of the above result, we estimate another set of models that include a slope shifter to separately assess the preferences for this policy attribute of those who do and those who do not have children living outside of the US. We found that although the interaction term is positive, its coefficient and its standard deviation coefficient are not statistically significant across all models. However, when we interact a dummy variable for family (spouse and children not living in the US) with this attribute, it was found that only for the model Stated illegal entry, at least the standard deviation parameter of this positive interaction term coefficient is statistically significant, meaning that there is still substantial preference heterogeneity among illegal entrants who do not live with their family in the US. When create a slope shifter for spouse living outside the US for this sample group we did not found any significant interaction effect, although the effect is positive. Similarly, it is noted that the standard deviation coefficient is significant across nearly all other attributes and models, indicating that there is significant preference variation among Hispanic immigrants, both legal and illegal entrants, for different immigration reform policy attributes. In other words, the value of a particular attribute might be considerably higher for certain individuals than for others. 21

32 7.4 Willingness to pay estimation results While the signs of the coefficient estimates correspond with expectations, the average value that likely illegal Hispanic immigrants place on the different attributes should be of substantial interest to policy makers. Table 4 provides estimates of the immigrants' willingness to pay (WTP) 4 for the different immigration policy attributes calculated using the coefficient estimates from the mixed logit models. Across all policy attributes, except family-related attributes, illegal entrants are willing to pay statistically less than those who did not state that they entered the US illegally. We attribute this difference to the fact that illegal entrants have a lower annual income ($2,056/year less on average). The absence of statistical difference of the WTP estimates for the familyrelated attributes for both groups shows an interesting picture, regardless the difference in income education, and immigration status, Hispanic immigrants display similar WTP for the possibility to bring their close and extended family (spouse, children and parents). Focusing on stated illegal entrants, they place an average monetary value of $925 to each additional year of work visa length that is allowed by the policy. For example, they would be willing to pay an additional $925 to acquire a two-year visa instead of a one-year work visa. In addition, the estimated WTP for green card eligibility after a five-year wait is $10,180, which is a little over twice the value of a five-year work visa. Interestingly, these respondents only discounted that value to $8,482 if the required wait for the green card is extended to 10 years, which is nearly the same as the valuation placed on a 10-year work visa ($925*10=$9,250). The 95% confidence intervals 5 of the WTP estimates suggest that Hispanic illegal US entrants are willing to pay 4 The willingness to pay (WTP) for a marginal change in an attribute across alternatives is defined as the ratio between the coefficient estimate of the attribute to the coefficient estimate on alternative price. Given a fixed price coefficient, the WTP distribution corresponds to the distribution of the attribute coefficient. 5 The WTP standard errors and 95% confidence interval were computed using standard bootstrapping methods. 22

33 anywhere between $5,000 and $16,000 to eventually be eligible for a green card, which on the high side is over years worth of income for these individuals. Looking at the non-work related immigration policy attributes, we find that immigrants in the sample also place a sizable value on these attributes, but they are statistically less than the work related attributes. In particular, eligibility to parents beyond just spouses and children to potentially be able to obtain visitation (temporary non-resident) visas is highly valued. Remarkably for individuals who are on average earning $1,300 per month, Hispanic immigrants are on average willing to pay an estimated $7,665 to give those relatives the opportunity to come to the US. While this may on the surface seem like a high WTP, it is possible that survey respondents viewed the visitation visas as a potential stepping stone for easier legal or illegal opportunities for their relatives to come to the US. In contrast, although statistically significant, the mean willingness to pay for policy provision that allows immigrants to have the same access to medical care as US citizens is relatively small ($1,960). It would appear that the belief by some that access US medical care system is a major issue for the Hispanic community is not supported by our finding. However, the results suggest that the prospect of eventually being able to retire and draw upon Social Security is highly valuable for this community, and gives credence to claims that immigrant workers feel in some ways cheated due to their undocumented status. In this regard, it has been recently established that many undocumented workers pay sales, income, and payroll taxes with no possibility of claiming a refund or receiving Social Security or Medical benefits upon retirement (Dwyer, 2004). 23

34 CHAPTER 8 CONCLUSSIONS AND POLICY DISCUSSION 8.1 Conclusions With an estimated 11.2 million illegal immigrants currently in the US, immigration reform is one of the most polarizing public and political issues in the US with divides along political, industry, and labor groups. For the agricultural sector in particular, any changes to the work visa system or prospects for immigrants obtaining legal permanent residence could have substantial impacts on the agricultural labor supply and ultimate farm viability and profitability. Current US Senate and House immigration reform proposals have distinct differences in terms of how they address the current population of undocumented workers and create a system for future temporary workers. While the American populace, labor groups, and business community is divided on how immigration policies should be reformed, and the competing political parties are far from agreement, it is likely if any reform occurs it will be some compromise in between the current US Senate and House proposals. In order to find that compromise that will remediate immigration policy shortcomings, it is critical to understand the preferences and acceptance of policy features of illegal immigrants themselves. In this study we present the first evidence specifically eliciting the preferences towards immigration reform of Hispanic immigrants with first-hand experience with illegally emigrating, working, and living in the US. Evidence from the survey and choice experiment support several important hypotheses relevant to understanding Hispanic immigration and attitudes towards immigration reform. First, it seems clear that the original motivation to come to this country of likely illegal 24

35 Hispanics immigrants was to seek jobs and better wages. Even though most of them have been here for years and would now like to become citizens, their recollection is that the prospect of eventually achieving a path to US citizenship or the belief that they would have access to better medical care or government benefits was not an important factor in their original decision to relocate. These results are further reflected in the lower economic value that illegal entrants place on access to these federal benefits in comparison to other immigration policy characteristics. The fact that having the opportunity to become citizen is not more important than jobs and other motivating factors presents an interesting contrast when compared to the typical polling question presented to Hispanic Americans asking whether they would like to become a citizen. While in line with previous polls we find that about 90% agree that they would like to become a citizen, this only presents part of the picture. Other rights such as long term work visas and family visitation rights have substantial value as well. Relative to their current assumed illegal status, illegal entrants place a very high value on being able to legally stay and work in the U.S. regardless of whether a path to citizenship is being offered. In addition, they are willing to wait for up to a decade and still pay a substantial fine to become permanent residents and eventually citizens. 25

36 REFERENCES Alsalam, N.A. and R.E. Smith The Role of Immigrants in the US Labor Market. Congressional Budget Office, Washington DC. Alvarez, R.R A Profile of the Citizenship Among Hispanics in the United States. International Migration Review 21(2): Artz, G., R. Jackson, and P.F. Orazem Is it a Jungle Out There? Meat Packing, Immigrants, and Rural Communities. Journal of Agricultural and Resource Economics 35(2): Barreto M., and G. Segura Poll of Undocumented Immigrants Reveals Strong Family and Social Connections in America. Latino Decisions/NALEO/AVEG Poll of Undocumented Latinos. Boucher, S.R., A. Smith, J.E. Taylor, and A. Yúnez-Naude Impacts of Policy Reforms on the Supply of Mexican Labor to US Farms: New Evidence from Mexico. Review of Agricultural Economics 29(1): Buccola, S., C. Li, and J. Reimer Minimum Wages, Immigration Control, and Agricultural Labor Supply. American Journal of Agricultural Economics 94(2): Chy Lung vs. Freeman, U.S. 275 (U.S. Supreme Ct.). 26

37 Devadoss, S. and J. Luckstead Implications of Immigration Policies for the U.S. Farm Sector and Workforce. Economic Inquiry 49(3): Dugan, A Passing New Immigration Laws is important to Americans. Gallup Dwyer, J Illegal immigrants, health care, and social responsibility. Hastings Center Report 34: Gonzalez-Barrera, A., M.H. Lopez, J.S. Passel, and P. Taylor The Path Not Taken: Twothirds of Legal Mexican Immigrants are not US Citizens Washington, DC: Pew Hispanic Center. Gunter, L.F., J.C. Jarrett, and J.A. Duffield Effect of U.S. Immigration Reform on Labor- Intensive Agricultural Commodities. American Journal of Agricultural Economics 74(4): Hakimzadeh, S. and D. Cohn English Usage Among Hispanics in the US. Washington, DC: Pew Hispanic Center, Hanley, N., R.E. Wright, and V. Adamowicz Using Choice Experiments to Value the Environment. Environmental and Resource Economics 11(3-4): HR.1417, Border Security Results Act, 113th Congress. HR.1772, Legal Workforce Act, 113th Congress. HR.1773, Agricultural Guestworker Act, 113th Congress. 27

38 HR.2131, Supplying Knowledge-based Immigrants and Lifting Levels of STEM Visas Act, 113th Congress. HR.2278, Strengthen and Fortify Enforcement Act, 113th Congress. Hu, W. A. Hünnemeyer, M. Veeman, W. Adamowicz, and L. Srivastava Trading off Health, Environmental and Genetic Modification Attributes in Food. European Review of Agricultural Economics 31(3): Isé, S. and J.M. Perloff Legal Status and Earnings of Agricultural Workers. American Journal of Agricultural Economics 77(2): Kandilov, A.M.G. and I.T. Kandilov The Effect of Legalization on Wages and Health Insurance: Evidence from the National Agricultural Workers Survey. Applied Economic Perspectives and Policy 32(4): Kossoudji, S. and D. Cobb-Clark Coming Out of the Shadows: Learning about Legal Status and Wages from the Legalized Population. Journal of Labor Economics 20(3): Luckstead, J., S. Devadoss, and A. Rodriguez The Effects of North American Free Trade Agreement and United States Farm Policies on Illegal Immigration and Agricultural Trade. Journal of Agricultural and Applied Economics 44(1): Lusk, J.L., J. Roosen, and J.A. Fox Demand for Beef from Cattle Administered Growth Hormones or Fed Genetically Modified Corn: A Comparison of Consumers in France, Germany, the United Kingdom, and the United States. American Journal of Agricultural Economics 85(1):