Foreign-Born Voting Behavior in Local Elections

Abstract

Approximately half of immigrants to the United States are now settling directly in cities and towns with little prior history of immigration. Because this dispersed settlement pattern is so recent, we know little about the political behavior of naturalized citizens in these new immigrant destinations. This article begins to fill this gap by exploring the determinants of foreign-born voting in municipal elections using a new dataset that combines official voting information from the state of Utah with demographic information about Utah residents from the Utah Population Database (UPDB). We hypothesize that in addition to individual-level predictors of prior experience with democratic politics and community attachment, the size of cities and their form of government will also affect the likelihood that foreign-born citizens will turn out to vote in local elections. We use multilevel modeling techniques to test these hypotheses and find that prior experience with democratic politics, whether in the United States or in their home country, along with the city-level characteristics of city size and form of government, are powerful predictors of foreign-born voting in local elections. Moreover, we find that while large cities experience lower levels of turnout for all citizens, the negative effect on participation is strongest for foreign-born citizens.

Keywords

immigrant voting behavior local elections new immigrant destinations political participation

Immigrant settlement patterns in the United States have changed dramatically during the past two decades. The pull of economic opportunities, affordable housing, and immigrant networks are motivating millions of new immigrants to settle in cities and towns with little or no prior history of immigration. Although traditional immigrant gateway cities like New York, Miami, and Los Angeles continue to attract the most immigrants, immigrant populations in metropolitan areas such as Atlanta, Charlotte, Las Vegas, Salt Lake City, and Seattle have more than tripled in the past two decades. More immigrants are also settling directly in suburban areas, small towns, and even rural areas all across the United States (Fortuny, Chaudry, & Jargowsky, 2010; Marrow, 2011; Singer, 2008). Because the change in settlement patterns is relatively recent, most immigrants living in these new gateway communities have not yet naturalized or received much attention from local political parties and campaigns. However, in the coming decades, many of these immigrants will become citizens, vote in local and national elections, and transform the political dynamics of their communities.

This geographic dispersion of immigrants makes understanding the voting behavior of foreign-born citizens outside of traditional gateway cities more important than ever. However, we still know relatively little about the voting behavior of naturalized immigrants living in new destinations. The vast majority of the research on immigrant political participation has focused on foreign-born voting in national elections (Bass & Casper, 1999; Pantoja, Ramírez, & Segura, 2001; Ramakrishnan, 2005) or on immigrant political participation in large cities within traditional gateway states (e.g., Barreto, Villarreal, & Woods, 2005; Mollenkopf, Olson, & Ross, 2001; Mollenkopf & Sonenshein, 2009; Ramakrishnan & Bloemraad, 2008; Wong, 2006). Although we have learned much from this research, we should not assume that the same factors that explain immigrant voting at the national level explains foreign-born voting at the local level. Nor should we assume that the findings from studies of local elections in cities that have huge immigrant populations, co-ethnic politicians in positions of power, and organizations with the capacity and willingness to mobilize immigrants to the polls, apply to new gateway cities and towns that do not have these characteristics. Thus, the core findings from existing studies about immigrant voting patterns may not apply to local elections in new destinations, or may ignore factors that are important outside of traditional gateway cities and states.

To begin to fill this gap, this article analyzes the voting behavior of naturalized citizens in local elections across the state of Utah, which has had one of the fastest growing immigrant populations in the country since 1990. Our explanation focuses on three factors. First, prior experience with political parties and elections is an important determinant of turnout. Prior political experiences may be especially important for foreign-born citizens who have more to learn about politics in the United States, and who tend to face more obstacles to participation than the native-born. Accordingly, we test whether membership in a political party, prior experience voting in a federal election, or migration from a democratic country increase the likelihood that immigrant citizens will vote in local elections. Second, studies of the general population have found that residents with stronger ties to their communities are more likely to vote in local elections. Length of residence in a locality and homeownership are two variables that measure how rooted individuals are in their communities and are likely to have positive influences on immigrant turnout. Third, in addition to immigrants’ demographic traits, studies of voting in local elections have demonstrated that city-level characteristics are also important in shaping local-level turnout. In particular, these studies demonstrate that city size and the form of government are powerful predictors of turnout in local elections. We draw on this literature to test whether living in larger cities or in municipalities with appointed managers, as opposed to elected mayors, affects the turnout of immigrants.

To analyze those factors that enable or hinder foreign-born voting in local elections, we combined information from the Utah Population Database (UPDB), official voter records, and information from over 240 cities and towns in Utah into a unique new dataset. Because of the multilevel structure of our data, and because we want to explore whether city-level factors affect foreign-born citizens differently than native-born citizens, we use multilevel models to analyze turnout in municipal elections. We find that immigrants’ prior experience with democratic politics, whether in the United States or in their home country, is one of the most important predictors of foreign-born turnout in local elections. In addition, we find that immigrant turnout rates are lower in larger cities and in municipalities with manager forms of government. Moreover, in the case of city size, this negative impact on voting is stronger for foreign-born citizens compared with U.S.-born citizens, suggesting that the large metropolitan areas where most immigrants settle create obstacles that slow down their political incorporation.

Foreign-Born Voting in Local Elections

Why is it important to study the voting behavior of foreign-born citizens in local elections? The majority of elections and political activities in the United States are held at the local level (Trounstine, 2009), which means that local politics provide immigrants with many opportunities to practice democratic politics, have a voice in their local communities, and influence government on issues that have a direct impact on their lives (Hajnal & Lewis, 2003; Trounstine, 2010). For the foreign-born, positive democratic experiences at the local level may accelerate and deepen the process of political incorporation by spurring participation in additional local and national political activities (Jones-Correa, 2001a; Wong, 2006). Alternatively, if local politics discourages immigrants from political activity, it may also hinder their political incorporation into the national polity.

During the past several decades, the number of immigrants settling in the United States has increased dramatically, affecting the demographic and political composition of the electorate (Ramakrishnan, 2005).¹ One of the key features of this wave of immigration is the dispersion of immigrants to states, cities, and towns with little or no prior history of immigration. Although the country’s six traditional receiving states—California, Florida, Illinois, New Jersey, New York, and Texas—continue to attract the most immigrants, the rates of growth of immigrant populations in states such as Georgia, North Carolina, Tennessee, Nevada, and Utah are now approximately 3 times higher than the national average. Immigrants are also settling outside of the traditional metropolitan gateway cities like Los Angeles, New York, Miami, and Chicago and moving directly to new metropolitan gateways, to suburbs, and even to small towns and rural areas (Fortuny, Chaudry, & Jargowsky, 2010; Marrow, 2011; Singer, 2008).

Much has been written about the potential influence of naturalized citizens in national elections. However, the rapid growth of eligible foreign-born voters, and their dispersal to cities and towns with little prior history of immigration, means that foreign-born voters may also have a profound impact on the politics of the cities and towns in which they settle. While it is well established that foreign-born citizens vote at lower rates than the native-born in national elections, we know less about their patterns of participation in local politics. If the same disparity that exists at the national level also exists in local elections, then immigrants’ representation in local government is at risk. This disparity would, therefore, have important implications for electoral results, public policy, and minority representation in city councils and other local offices (Hajnal, 2009).

Local Voting Turnout and Previous Political Experience

Virtually, all research that examines the voting patterns of immigrants has found that foreign-born citizens vote at lower rates than the native-born (see, for example, Bass & Casper, 1999; Jones-Correa, 2001b). The participatory deficit for the foreign-born is the result, in part, of their relative lack of experience with the American political system, and in the case of immigrants from authoritarian countries, with democratic politics in general. Because local contests are often non-partisan affairs with few partisan cues and little targeted mobilization of immigrants, a lack of exposure to electoral politics may have a stronger negative effect on immigrant turnout in local elections compared with native-born citizens (Trounstine, 2010).

Naturalized citizens can acquire experience with elections in a number of ways. According to Black, Niemi, and Powell (1987), it is immigrants’ actual practice of political participation in their home countries that encourages political activity in their new one. Wals (2011) also finds that home country political experiences are an important predictor of immigrants’ political behavior once they arrive in the United States. A citizen who comes from a country with a long electoral history may have acquired meaningful political experiences with elections, campaigns, and political parties that facilitate his or her political incorporation into the United States. Once in the United States, this prior democratic experience becomes a participatory resource that reduces the costs and increases the perceived benefits of voting (Bueker, 2005). Although research on this variable has produced ambiguous results (Portes & Mozo, 1985; Portes & Rumbaut, 1996; Ramakrishnan & Espenshade, 2001), we expect to find that, all else equal, immigrants from democracies are more likely to turn out to vote in local elections than immigrants from countries governed by authoritarian regimes.

Once in the United States, the most relevant kind of democratic experience a citizen can have is voting in a prior election, as voting in one election is a powerful predictor of voting in subsequent ones (Niven, 2004). Another way immigrants can gain political experience is by affiliating with a local political party. Joining a party is often relatively easy to do² and can lead to many new opportunities for political mobilization and participation. Historically, political parties played a key role in mobilizing immigrants into local and national politics (Ramakrishnan, 2005; Wong, 2006). Immigrants who have registered with a political party are more likely to be targeted for mobilization by campaigns and other political and civic organizations than immigrants who have not registered with a political party (Kaufmann & Rodriguez, 2011). Moreover, the act of registering with a political party suggests that the individual has achieved a certain level of political engagement and a growing identification with the political system of the United States (Lien, Collet, Wong, & Ramakrishnan, 2001), two important predictors of political participation. Thus, our first three hypotheses are that naturalized citizens who emigrated from a democratic country, who voted in a prior election in the United States, or who affiliated with a state chapter of a political party are more likely to vote in local elections than those who did not.

Local Voting Turnout and Community Attachment

In the United States, people frequently move from one location in the country to another within the United States (U.S. Census Bureau, 2011). This geographic mobility has implications for citizens’ civic and political engagement. Studies show, for example, that citizens with stronger ties to their community are more likely to vote in local elections (Alford & Scoble, 1968; Wood, 2002). Findings from recent studies of immigrant political incorporation show that immigrants who stay in the United States for longer periods of time are more likely to naturalize (Ong & Nakanishi, 1996), register to vote (Bass & Casper, 1999), and have a higher level of political involvement (Ramakrishnan & Espenshade, 2001).

Homeownership is another factor that can bolster immigrant political incorporation (Alford & Scoble, 1968). Homeowners tend to be more rooted in their neighborhoods, and so are more likely to become involved in community and political activities (Barreto, Marks, & Woods, 2007). Homeowners are also more likely to be asked by political parties and campaigns to participate during election times (Ramírez, 2005) and are able to find politically relevant information, such as where to vote, more easily (Jackson, 2003). Therefore, we also expect that turnout in local elections will be higher among foreign-born citizens who own homes and have a longer length of residence in their community.

Local Contextual Factors

Although individual-level factors are important predictors of voting behavior, they tell only a part of the story. Researchers increasingly recognize the important role that local political and institutional contexts play in shaping immigrant political activity (Hochschild & Mollenkopf, 2009; Jones-Correa, 2001b). City-level contexts may be especially important for naturalized citizens because local-level issues related to housing, economic development, education, and health care draw them into politics, give them experience with American political institutions, and may motivate them to vote in elections (Jones-Correa, 2001a). Like any member of a community, immigrants care about the quality of schools, safety in their neighborhoods, clean parks, and other services that are under the control of local governments. Indeed, many immigrants begin experimenting with political activity locally even before they naturalize or register to vote.³

When scholars have studied the political behavior of immigrants in local elections, they have typically focused on large metropolitan gateway cities (Mollenkopf & Sonenshein, 2009; Wong, 2006). However, the insights from this literature do not easily apply to newer settlement contexts. Few of the conditions identified as important for the political mobilization of immigrants in large gateway cities—a large concentration of co-ethnics, co-ethnic candidates in top-of-the-ticket elections, political parties interested in mobilizing immigrant voters, and a density of mobilizing resources such as Spanish-language media and immigrant service organizations—are likely to exist in new immigrant destinations that have little prior history of international migration and small numbers of naturalized citizens (Ayón, 2006; Barreto, Segura, & Woods, 2004; Barreto, Villarreal, & Woods, 2005; Michelson, 2003; Ramakrishnan & Espenshade, 2001; Wong, 2006). A fuller understanding of immigrant voting in local elections requires examining their voting behavior in both small and large towns, in urban and rural areas, and in places with a variety of political institutions.

We test the effects of two city-level variables that the urban studies literature has shown to matter for local-level turnout, city size, and the type of municipal government. Because studies of immigrant turnout have focused almost exclusively on the country’s largest metropolitan areas, it is unclear how city size affects immigrant voting behavior. One possibility is that larger metropolitan areas, because they have bigger populations of immigrants, more mobilizational resources, and receive more attention by local and national get-out-the-vote campaigns, support higher levels of foreign-born turnout in local elections (Andersen, 2008; Kelleher & Lowery, 2009; Ramakrishnan & Bloemraad, 2008). Another possibility is that turnout is lower in larger cities because they have more complex bureaucracies, greater spatial distance between city offices and citizens, and elected officials who represent more people. All of these institutional features of larger cities tend to raise the costs of participation (Oliver, 2000). In contrast, residents of smaller towns are more likely to be familiar with candidates and local government officials, which makes gathering information about issues and candidates easier than in large cities. Oliver (2000) also found that residents of smaller localities are more interested in local politics, are more likely to be asked to participate in political acts by friends and neighbors, and are much more likely to participate in a variety of non-electoral civic activities than are people in large cities. Face-to-face recruitment may be particularly important for the political mobilization of the foreign-born, who are often bypassed by traditional get-out-the vote campaigns (Michelson, 2003; Niven, 2004).

The effect that municipal government structure has on local-level voting is clearer. Studies have confirmed that on average, cities with a council–manager form of government have lower turnout levels than cities with elected mayors (Caren, 2007; Karnig & Walter, 1983; Wood, 2002). Cities with appointed managers as their chief executive experience lower turnout for two reasons. Many cities create and fill the position of a city manager without direct citizen input, which makes it harder for citizens to perceive a central political figure with whom they may identify or hold accountable (Hajnal & Lewis, 2003; Wood, 2002). Manager forms of government also make it harder for voters to gather relevant information, for political parties to mobilize voters, and for residents to find incentives to vote in local elections.

We test two hypotheses related to the potential effects of local context on the voting behavior of foreign-born citizens. Given that manager forms of government erect significant barriers to local-level political engagement, we expect turnout to be lower in cities and towns with appointed managers rather than elected mayors. Our expectations about city size are less clear-cut, but overall, we expect that as city size increases, immigrant turnout in local elections will decrease.

Foreign-Born Versus Native-Born Voters

We are also interested in the relative effect of all these variables on the voting behavior of naturalized citizens compared with the native-born. Naturalized citizens tend to vote at lower rates in presidential elections compared with native-born citizens because they are newer to the political system, have less experience with U.S. political institutions and processes, and are, as a group, rarely, if ever, targets of political mobilization by parties and campaigns. Because of this deficit of experiences, we expect foreign-born citizens to turn out at lower rates in municipal elections. However, also because of this deficit, it is possible that any experiences that socialize people into participation, increase their familiarity with local politics, or make them more likely to be targeted for mobilization, will have a stronger effect on naturalized citizens compared with the native-born. In other words, we might expect that prior experience with voting, becoming a member of a political party, longer length of residency, and homeownership, will have a stronger positive impact on the turnout of naturalized citizens compared with the native-born.

An alternative hypothesis is that naturalized citizens are not necessarily at a deficit in their exposure to the American political system. The naturalization process itself requires immigrants to become familiar with U.S. institutions and must be undertaken to become eligible to vote. Native-born citizens, in contrast, are becoming less efficacious and appear to be increasingly turned off from electoral politics, as evidenced by declining turnout in both national and local elections during the past three decades. If this alternative perspective is correct, then factors like prior political experiences, longer residencies, and homeownership should not have a differential impact on immigrants. Our analysis will test these competing perspectives.

We will also test whether contextual factors such as city size and manager forms of government have a differential effect on the voting behavior of naturalized citizens. According to the local elections literature, large cities and manager forms of government erect barriers and reduce incentives for political mobilization and civic engagement. Naturalized citizens may bring new enthusiasm to the political system. However, the bulk of the available evidence still suggests that, in the absence of mobilization efforts and other incentives to vote, foreign-born citizens have fewer participatory resources, such as income, education, and English-language proficiency, with which to overcome institutional obstacles to participation (Bass & Casper, 1999; Tam Cho, 1999). If this is the case, naturalized citizens should suffer the negative effects of large city size and manager forms of government more heavily than native-born citizens.

Utah’s Foreign-Born Population

Utah is one of about a dozen “new immigrant destinations,” where the immigrant population has grown at 3 times the national rate (Singer, 2008). Although we want to be modest when generalizing to other cases, Utah’s “context of reception” shares key features of the foreign-born population and the limited immigrant-specific resources with other new immigrant destinations, making it a good choice for focused study.⁴ Similar to other immigrant gateways, the majority of Utah’s foreign-born are Hispanic (64%). Of the remaining foreign-born, 18% are from Asia, 11% are from Europe, 3% are Northern American, 3% are from Oceania, and 2% are from Africa (Perlich, 2008). Between 1990 and 2010, Utah’s immigrant population has increased nearly 300%, the 10th fastest growth rate in the country. The latest census population estimates indicate there were 223,533 foreign-born people in Utah, or about 8% of the population.⁵ Also, as with other new immigrant destinations, the foreign-born residing in Utah tend to be less proficient in English than immigrants living in other places.⁶

Although the vast majority of immigrants used to settle in large metropolitan areas, during the current wave of immigration, immigrant settlement patterns have become much more dispersed. As in other new immigrant destinations, immigrants to Utah are settling in city centers, suburbs, small towns, and rural areas. These new settlement patterns make it possible to study immigrant voting behavior in a variety of local contexts and draw some conclusions about the effect of city characteristics on voting behavior.

Because this growth has occurred very recently, Utah lacks many of the local conditions that foster the political mobilization of immigrants in traditional gateways. With support networks provided by the Church of Jesus Christ of Latter Day Saints, educational opportunities for legal and illegal immigrants, and laws that allow undocumented immigrants to drive legally once they received a Driver’s Privilege Card, Utah initially provided a relatively welcoming environment for immigrants. More recently, however, Utah implemented a number of anti-immigration bills that created a threatening environment for immigrants. For example, in 2008, the Utah State Legislature passed a comprehensive immigration bill (SB 81) that made citizenship a requirement for state identification documents, instituted the federal E-Verify program to ensure that all new hires were legally present and allowed to work in the United States, required verification of legal status for adults who apply for public benefits, and required local law enforcement officials to train as federal immigration officials.⁷

Data and Variables

Our dataset consists of individual-level demographic variables from the UPDB and official voter records from Utah’s Lieutenant Governor’s Office. The UPDB is typically used for genetic, epidemiological, and public health studies and contains demographic information on all Utah residents. The Voting and Elections database is publicly available and provides official voting records for all registered voters in the state, including approximately 30,000 foreign-born citizens.⁸ We merged the UPDB with the Voting and Elections database using an individuals’ name, address, and age resulting in a dataset that identifies a majority of registered voters in the state, those who are affiliated with a political party, and key measures of voting in both local and national elections.⁹ We augmented our dataset with measures of socioeconomic status from the 2000 U.S. Census, along with city-level information regarding the municipal form of government and population size of each of Utah’s municipalities. The result is a powerful dataset that allows us to analyze individual-level and contextual determinants of immigrant voting behavior.

Local turnout studies are often hampered by data difficulties and small sample sizes (Kaufmann & Rodriguez, 2011). Even studies of foreign-born voting behavior that rely on Current Population Survey data consist of small sample sizes that produce large margins of error or ignore new destination states (Bueker, 2005). Because we are working with actual voting records instead of relying on citizens’ recollections of their voting behavior, we are able to avoid the pitfalls associated with self-reported voting data, including the overreporting of turnout. The size of our dataset is ample enough to avoid large margins of error and to support the analysis of a new immigrant destination.

Our dataset also has the advantage of having relatively large sample sizes of immigrants living in all kinds of towns—large and small, rural and urban. This dataset allows us to examine not only how turnout in local elections is affected by both individual- and contextual-level variables but also whether the causal effect of individual-level predictors is moderated by the city-level factors of size and form of government. We also note that our data are ordered hierarchically with two levels of analysis in which individuals are nested within cities. Our observations of individuals are naturally clustered within cities, and our statistical models need to take into account the clustered nature of our data (Kreft & de Leeuw, 1998; Luke, 2004; Primo, Jacobsmeier, & Milyo, 2007). Therefore, we employ multilevel modeling (MLM) strategies that take into account both the two-level nature of our data as well as the clustering of standard errors at the individual level (Bryk & Raudenbush, 1992; Kreft & de Leeuw, 1998; Steenbergen & Jones, 2002).¹⁰ In addition, MLM allows us to combine individual-level and city-level variables into a single comprehensive model to determine whether the causal effect of individual-level predictors are conditioned by contextual factors (Guo & Zhao, 2000; Steenbergen & Jones, 2002).

Our dependent variable is voting in the November 6, 2007, municipal elections in Utah,¹¹ and it is coded 1 if an individual voted in the election and 0 otherwise. There are two levels to our data. At the individual level (Level 1), we assess the impact of political experience, community attachment, and birth country on turnout. We use information about an individual’s birth country contained in the UPDB dataset to construct the variable foreign-born, which is coded 1 if an individual was born outside of the United States and 0 if he or she was born in the United States.¹² We used the same birth country variable together with the Freedom House classification of regime types to construct the variable free, which is a dichotomous variable designating whether an immigrant came from a democratic or an authoritarian country.¹³ We also test for the effect of prior voting experience in the United States by including the variable voted in a prior federal election, which is coded 1 if an individual voted in either the 2004 or 2006 presidential or midterm elections and 0 if the individual did not vote in either election. Political party affiliation is a third measure of previous political experience. It is coded 1 if the citizen is registered with any of Utah’s political parties and 0 if he or she is unaffiliated.¹⁴

The two measures of community attachment in our models are length of residence and homeownership. Typically, studies of immigrant voting behavior include a measure of immigrants’ length of residence in the United States, but we are specifically interested in individuals’ familiarity with local politics. We measure length of residence, therefore, as the number of years that an individual has held a Utah driver’s license, which is a more direct measure of their level of familiarity with and attachment to their local communities. Homeownership is a continuous block-group-level variable, garnered from the 2000 U.S. Census, that indicates the percentage of the residences within a block group that are owner-occupied.

One of the advantages of our dataset is our ability to analyze immigrant voting behavior in 243 cities and towns, which represent a variety of local contexts that include cities and towns of different size and with different forms of government. We used the residence variables contained in the UPDB to identify the city that a person lived in during the 2007 municipal elections. We then used population information from the Utah League of Cities and Towns to create the Level-2 variables of city size and manager. Cities in Utah range in size from some very small rural towns (under 500 residents) to large cities embedded in a metropolitan area of more than 1 million residents.¹⁵ We divide city size into three categories for analysis: less than 10,000 residents, 10,000 to less than 50,000 residents, and 50,000 residents or more. Throughout, small cities (under 10,000) are the reference category. We are particularly interested in the effects of the manager form of government, because the existing literature demonstrates that cities with manager forms of government experience lower levels of turnout in local elections. The bulk of Utah municipalities are governed either by an elected mayor and city council or by a council–manager form of government in which the city council is elected and the mayor/manager is appointed. The variable manager is coded 1 for those citizens living under a council–manager or manager by ordinance form of government and 0 for those living in towns with other forms of government.

Finally, the models include several individual-level and contextual control variables. We include a variable indicating whether or not an immigrant is Latino because Latinos are typically less likely to vote than other ethnic groups, and foreign-born Latinos frequently turn out at lower rates than native-born Latinos (Barreto et al., 2005; DeSipio, 1996). Also, there has been a large influx of immigrants from Latin America settling in the United States (Levin, 2013), particularly in new immigrant destinations. We used information about individuals’ ethnicity contained in the UPDB to construct the variable Latino, which is coded 1 if the voter is Latino or Hispanic and 0 if he or she is not.

We also control for age, measured in years, and gender, with male coded 1 and female coded 0. Other control variables are measured at the block-group level.

We used census data and guidelines to construct the variable urban, which is coded 1 if someone lives in a block group with a population density of greater than 1,000 people per square mile and 0 if he or she does not. We include urban in our models to accommodate the geography within Utah, where even within the confines of a large city there are sparsely populated areas.

Studies have also demonstrated that the density of immigrants within a neighborhood may influence the turnout of immigrants and non-immigrants differently (Barreto et al., 2004; Ramakrishnan & Espenshade, 2001). Accordingly, we include a continuous measure of foreign-born density, which is a measure of the percentage of foreign-born living in each block group within the state.

The models also include block-group measures of income and education, which serve as proxies for individual-level measures of socioeconomic status.¹⁶ College education is the percentage of residents in a block group who have at least some college education. Income is the median household income at the block-group level and is broken down into three categories of less than US$35,000, US$35,000 to less than US$50,000, and US$50,000 or more.

Analysis and Results

Table 1 shows the descriptive statistics and frequency distributions of key variables for individuals in our dataset who are eligible to vote in the 2007 municipal elections. Overall, 39% of citizens who are registered to vote turned out to vote in 2007. Table 2 compares mean values of our dependent and independent variables for native-born citizens, and foreign-born citizens who emigrated from democratic countries, and foreign-born citizens who emigrated from authoritarian ones.¹⁷ Foreign-born citizens, in particular those who emigrated from authoritarian countries, voted at much lower rates, a result that parallels the findings of voting behavior in national elections. As expected, there is also a 16% difference in turnout rate between immigrants from democracies compared with immigrants from countries with authoritarian regimes, suggesting that prior experience with democratic politics in one’s home country may be a powerful predictor of voting in the United States.

Table 1.

Descriptive Statistics and Frequency Distributions for Key Variables for Voters Eligible to Vote in 2007.

Variable	n/%	Minimum	Maximum	M	SD
Voted in 2007 municipal election	1,089,049 (100%)	0	1	39%	0.49
Yes	39%
No	61%
Foreign-born	913,217 (100%)	0	1	3%	0.18
Yes	3%
No	97%
Affiliated political party	1,096,472 (100%)	0	1	48%	0.50
Yes	49%
No	51%
Voted in prior federal election	1,085,809 (100%)	0	1	68%	0.47
Yes	68%
No	32%
Length of residence	1,093,928 (100%)	0	65	14.1	7.26
Homeownership^a	701,249 (100%)	0	100	78%	0.20
Manager	935,071 (100%)	0	1	35%	0.48
Yes	35%
No	65%
City size	935,071 (100%)	26	178,810	63,860	57022.62
Urban	701,257 (100%)	0	1	74%	0.44
Yes	74%
No	26%
Latino	1,096,482 (100%)	0	1	3%	0.18
Yes	3%
No	97%
Foreign-born density^a	701,257 (100%)	0	53	6%	0.06
Median income^a	701,249 (100%)	0	US$135,852	US$53,031	17898.23
At least some college education^a	701,257 (100%)	0	1	66%	0.15
Age	1,096,475 (100%)	18	110	46.2	17.11
Male	1,096,404 (100%)	0	1	51%	0.50
Yes	51%
No	49%

Source. Authors’ dataset.

Cells indicate census block-group measures.

Table 2.

Comparison of Mean Values for Key Variables: Native-Born Versus Foreign-Born.

		Foreign-born
	Native-born	All foreign-born	From democracy	From authoritarian
	M	M	M	M
Voted in 2007 municipal election	41%	33%	36%	20%
Affiliated political party	49%	42%	44%	34%
Voted in prior federal election	70%	62%	64%	50%
Length of residence (years)	14.9	14.3	14.2	13.4
Homeownership^a	79%	76%	76%	73%
Manager	35%	37%	36%	36%
City size	62,466	82,660	78,940	106,318
Urban	73%	83%	82%	89%
Latino	3%	26%	33%	27%
Foreign-born density^a	6%	8%	8%	13%
Median income^a	US$52,959	US$52,583	US$51,624	US$49,703
At least some college education^a	66%	65%	64%	60%
Age	45.3	48.6	51.7	47.8
Male	52%	50%	50%	55%

Source. Authors’ dataset.

Cells indicate census block-group measures.

Table 2 also shows that naturalized citizens who emigrated from democracies have additional participatory advantages over immigrant citizens from authoritarian countries. The former are more likely to be affiliated with a local political party, are more likely to have voted in a previous federal election, and have resided longer in their communities. Foreign-born citizens from democracies also enjoy advantages in income and education. In turn, native-born citizens enjoy participatory advantages compared with all foreign-born citizens. We will test whether differences in local voting behavior between the native-born and foreign-born citizens and between immigrants from free and notfree countries persist after controlling for all of these factors. Finally, the table indicates that foreign-born citizens, in particular immigrants from countries with authoritarian regimes, are much more likely to reside in larger cities. In the next sections, we explore whether city-level factors account for a part of the participatory gap between native-born and foreign-born citizens.

Our main theoretical interest is in identifying the individual-level and contextual factors that influence the likelihood that immigrants will vote in local elections. As our dependent variable is dichotomous, we use a logit link function to define our models of voting behavior. We begin by defining an individual-level model of the probability that an individual i voted in the municipal election of city j:

\begin{array}{l} V o t e_{i j} = logit (π_{i j}) = log \frac{π_{i j}}{(1 - π_{i j})} = β_{0 j} + β_{1 j} P A_{i j} + β_{2 j} P V_{i j} + β_{3 j} f r e e_{i j} + β_{4 j} L R_{i j} \\ + β_{5 j} H O_{i j} + \sum_{k = 1}^{7} β_{k j} X_{i j}, \end{array}

where π_ij is the probability that the value of the dependent variable Y_ij = 1, PA_ij, PV_ij, and free_ij are the dummy variables for party affiliation, prior vote in a federal election, and whether the person migrated from a democratic country; LR_ij is a person’s length of residency; HO_ij is the variable measuring homeownership; and X_ij represents our control variables that might influence the likelihood of voting.¹⁸

As we expect voting behavior to vary across cities as a result of city size and form of government, we introduce city-level variables in our Level-2 models to explain the Level-1 intercept. In addition, we assume that the effect of our individual-level predictors is fixed, except for free, which we allow to vary across contexts, so that

β_{0 j} = γ_{00} + γ_{01} M e d i u m C i t y_{j} + γ_{02} L a r g e C i t y_{j} + γ_{03} M a n a g e r_{j} + μ_{0 j},

β_{3 j} = γ_{30} + μ_{3 j} .

Here MediumCity_j and LargeCity_j refer to dummy variables for city size, where small cities are the reference category, and Manager_j is a dummy variable for cities governed by managers as opposed to elected mayors. By substituting Equations (2) and (3) in Equations (1), we obtain our initial mixed-effects model of voting:

\begin{array}{l} V o t e_{i j} = γ_{00} + γ_{01} M e d i u m C i t y_{j} + γ_{02} L a r g e C i t y_{j} + γ_{03} M a n a g e r_{j} + β_{1 j} P A_{i j} + β_{2 j} P V_{i j} \\ + γ_{30} f r e e_{i j} + β_{4 j} L R_{i j} + β_{5} H O_{i j} + \sum_{k = 1}^{7} β_{i j} X_{i j} + μ_{3 j} \times f r e e_{i j} + μ_{0 j} . \end{array}

Because we are interested in explaining the voting behavior of the foreign-born, we first estimate this model using the subsample of foreign-born citizens who were registered to vote in time for the 2007 municipal elections. We show the results in Table 3, Model 1. The results show substantial support for all of our hypotheses, particularly that prior experience with democratic politics influences voter turnout. The coefficients for political party affiliation, prior voting experience in federal elections, and free were all strongly positive and significant, indicating that prior experience with democratic political processes, whether that experience was acquired in the United States or in their home country, increases the likelihood that naturalized citizens will vote in local elections.

Table 3.

Base Logit Multilevel Models of Voting in Municipal Elections.

	Model 1	Model 2
	B (SE)	B (SE)
Fixed effects: Level 1
Political party affiliation	0.647** (.046)	0.700** (.036)
Prior federal election vote	1.989** (.074)	2.174** (.066)
Free	0.507** (.062)
Foreign-born		−0.153** (.028)
Length of residence	0.014** (.003)	0.019** (.001)
Homeownership	−0.007 (.134)	0.356** (.067)
Level 2
Manager	−0.176* (.076)	−0.139^† (.072)
Medium city size^a	−0.200* (.092)	−0.272** (.065)
Large city size^a	−0.278** (.105)	−0.376** (.082)
Intercept	−4.114** (.245)	−4.163**(.127)
Level-1 controls
Latino	−0.139 (.078)	−0.397** (.026)
Urban	0.089 (.062)	0.098** (.026)
Foreign-born density	−0.371 (.306)	0.171 (.357)
US$35,000 less than US$50,000^b	0.148 (.079)	0.060^† (.031)
US$50,000 or more^b	0.178 (.107)	0.029 (.052)
College education	0.619*(.298)	0.591* (.239)
Age	0.019** (.002)	0.026** (.001)
Male	−0.120** (.033)	−0.131** (.009)
Variance components
Intercept	0.056** (.021)	0.260** (.031)
Free	0.000 (.015)	0.013* (.006)
Foreign-born		0.013* (.006)
Level-1 n	13,539	592,893
Level-2 n	175	241
−2 Log likelihood	64512.278	2777022.747

Source. Authors’ dataset.

Note. Cells report the maximum likelihood estimates of coefficients with robust standard errors in parentheses using a diagonal covariance structure.

The reference category for the city size variable is small cities with populations of less than 10,000 residents.

The reference category for the income variable is less than US$35,000.

†

p < .10. *p < .05. **p < .01.

Our two measures of community attachment provide less consistent results. The coefficient for length of residence is positive and significant, which supports our hypothesis that longer residencies increase immigrants’ familiarity with local affairs, making them more likely to vote in local elections. However, contrary to our expectations, the effect of homeownership on foreign-born turnout is non-existent. We suspect the coefficient for homeownership does not achieve statistical significance because homeowners tend to be more affluent, better educated, and older; once we controlled for these factors, the variable homeownership lost explanatory power. To see if this is the case, we ran a model without controls for income and education. In that model, the coefficient for homeownership was positive and significant (p < .05), as we expected. In short, we do not find evidence that the act of owning a home by itself increases the likelihood of voting in local elections.¹⁹

The regression model also shows support for our hypotheses that city-level characteristics have a powerful effect on immigrant voting above and beyond individual-level factors. Compared with cities with elected mayors, manager forms of government appear to discourage foreign-born citizens from voting in local elections. This finding is consistent with other studies that have found that manager forms of government raise the costs and decrease the incentives for potential voters to engage with local politics. In addition, the coefficients for medium and large city size are both negative and statistically significant, indicating that the turnout of naturalized citizens is depressed in larger cities compared with small ones. We did not have strong expectations about the effects of city size on immigrant voting. However, the results support the conclusion that any participatory advantage enjoyed by the foreign-born who live in larger cities—where they might receive political information in their native language, or be mobilized to vote by political campaigns or community organizations—is outweighed by the greater cost that living in large cities can impose on civic and political engagement. The results also support the argument made by Oliver and others that residents of small towns are more likely to be asked to participate in politics by friends and neighbors and are much more likely to participate in civic activities than people living in larger cities. Although Oliver’s study does not distinguish between native-born and foreign-born citizens, the dynamics of small-town life he identified may be especially important for the electoral mobilization of immigrants, who are, on the whole, much less likely to be the targets of mobilization by political parties and campaigns (Oliver, 2000). Together, these results provide strong evidence that the places where immigrants settle matter for their political behavior and political incorporation.

Local Political Behavior of Foreign-Born Versus Native-Born Citizens

To examine whether foreign-born citizens are less likely than native-born citizens to turn out to vote in local elections, we estimate the same mixed-effects model as above, except we substituted Bforeign_ij for free_ij, which is a dummy variable that indicates whether a respondent is a foreign-born citizen. The Level-2 predictors for this model are identical to those included in Equation (4), as are the other Level-1 predictors. The last column of Table 3 reports the estimates of fixed effect and variance components of this model.

\begin{array}{l} V o t e_{i j} = γ_{00} + γ_{01} M e d i u m C i t y_{j} + γ_{02} L a r g e C i t y_{j} + γ_{03} M a n a g e r_{j} + \\ β_{1 j} P A_{i j} + β_{2 j} P V_{i j} + γ_{10} B f o r e i g n_{i j} + β_{4 j} L R_{i j} + β_{5} H O_{i j} + \\ \sum_{k = 1}^{7} β_{i j} X_{i j} + μ_{1 j} \times B f o r e i g n_{i j} + μ_{0 j} . \end{array}

As the final column of Table 3 shows, the coefficient for Bforeign is negative and significant, indicating that foreign-born citizens turn out to vote at lower rates in local elections than the native-born, a result consistent with studies of immigrant voting in national elections (Bass & Casper, 1999; Jones-Correa, 2001b; Portes & Rumbaut, 1996; Tam Cho, 1999).²⁰ These results, which used the full sample to produce estimates of fixed effects and random components, parallel closely the results of the model that used only the subsample of foreign-born citizens who were born in free countries. There were only two exceptions. The coefficient for Manager does not quite achieve significance at the p < .05 level (p = .054), and in this model, the estimate of the coefficient for Homeownership is positive and significant, as we had expected. The significance of the coefficient for Homeownership is not an artifact of a larger sample size; the estimated effect is also much larger and substantively important.

There are theoretical reasons to expect that prior political experiences, community attachment, and city-level context will have differential effects on naturalized citizens. We hypothesized that if naturalized citizens are at a deficit in their exposure to the American political system or are on average less engaged and less integrated with the U.S. political system compared with their native-born counterparts, then political experiences that teach individuals about U.S. politics or that socialize them into participation should have stronger positive effects on the likelihood naturalized citizens will vote in local elections relative to the native-born. If, however, immigrant citizens have learned a great deal about American politics through the naturalization process or are simply already enthusiastic about the prospect of voting, then prior exposure to electoral politics or a deeper attachment to their communities should not have a differential effect on their turnout. To test these competing perspectives, we added interaction terms to our core model that test the relative influence of each of our key Level-1 predictors on the voting behavior of foreign-born and native-born citizens. The results are reported in Table 4.

Table 4.

Logistic Regression of Voting in Municipal Elections Models With Interactions.

	Model 2a	Model 2b	Model 3a	Model 3b	Model 4
Fixed effects: Level 1
Foreign-born	−0.127** (.036)	−0.050 (.045)	−0.134**(.027)	−0.155** (.028)	−0.007 (.059)
Party affiliation	0.702** (.036)	0.700** (.036)	0.700** (.036)	0.700** (.036)	0.700** (.036)
Voted in prior election	2.174** (.066)	2.183** (.067)	2.174** (.066)	2.174** (.066)	2.174** (.066)
Length of residence^a	0.019** (.001)	0.019** (.001)	0.019** (.001)	0.019** (.001)	0.019** (.001)
Homeownership^b	0.356** (.067)	0.356** (.067)	0.356**(.067)	0.354**(.066)	0.356** (.067)
Level 2
Manager	−0.139^† (.072)	−0.139^† (.072)	−0.138^† (.072)	−0.139^† (.072)	−0.127^† (.074)
Medium city^c	−0.272** (.065)	−0.272** (.065)	−0.272** (.065)	−0.272** (.065)	−0.266** (.066)
Large city^c	−0.377** (.082)	−0.377** (.082)	−0.376** (.081)	−0.375** (.082)	−0.347** (.075)
Intercept	−4.164** (.126)	−4.17** (.127)	−3.908**(.128)	−3.886**(.125)	−4.169** (.127)
Democratic experience
Foreign-born × Party affiliation	−0.049 (.038)
Foreign-born × Voted in prior election		−0.236** (.047)
Community attachment
Foreign-born × Length of residence^a			−0.011** (.003)
Foreign-born × Homeownership^b				0.038 (.084)
Cross-level interactions
Foreign-born × Manager					−0.092 (.057)
Foreign-born × Medium city					−0.135* (.068)
Foreign-born × Large city					−0.179* (.072)
Control variables
Latino	−0.397** (.026)	−0.397** (.026)	−0.400** (.026)	−0.397** (.026)	−0.397** (.026)
Urban	0.098** (.026)	0.098** (.026)	0.098** (.026)	0.098** (.026)	0.098** (.026)
Foreign-born density	0.171 (.357)	0.169 (.356)	0.171 (.357)	0.171 (.357)	0.171 (.357)
US$35,000 less than US$50,000^d	0.060^† (.031)	0.060^† (.031)	0.060^† (.031)	0.060^† (.031)	0.060^† (.031)
US$50,000 or more^d	0.029 (.052)	0.029 (.052)	0.029 (.052)	0.029 (.052)	0.029 (.052)
College education	0.591* (.239)	0.591* (.239)	0.591* (.239)	0.591* (.239)	0.591* (.239)
Age	0.026** (.001)	0.026** (.001)	0.026** (.001)	0.026** (.001)	0.026** (.001)
Male	−0.131** (.009)	−0.131** (.009)	−0.130** (.009)	−0.131** (.009)	−0.131** (.009)
Variance components
Intercept	0.259** (.031)	0.261** (.031)	0.261** (.031)	0.260** (.031)	0.265** (.031)
Foreign-born	0.013* (.006)	0.012* (.006)	0.012* (.006)	0.013* (.006)	0.009^† (.005)
Level-1 N	592,893	592,893	592,893	592,893	592,893
Level-2 N	241	241	241	241	241
−2 Log likelihood	2776990.842	2776649.28	2776849.934	2777038.246	2776649.28

Source. Authors’ dataset.

Note. Cells report the maximum likelihood estimates of coefficients with robust standard errors in parentheses using a diagonal covariance structure.

Length of residence is centered around its mean.

Home ownership is centered around its mean.

The reference category for the city size variable is small cities (populations less than 10,000).

The reference category for the income variable is less than US$35,000.

†

p < .10. *p < .05. **p < .01.

The coefficient for the interaction term foreign-born × party affiliation in Model 2a is not significant, which means that affiliation with a political party does not affect the voting behavior of foreign-born citizens differently. Rather, it has a more or less equally positive effect on the local turnout of both naturalized and native-born citizens. In contrast, we find that the positive effect of voting in a prior federal election and length of time residing in a city are muted for naturalized citizens compared with the native-born (Models 2b and 3a). In other words, the familiarity with local and democratic politics that results from living in a place for a longer period of time or from voting in a federal election still boosts the turnout of immigrants, but native-born citizens appear to benefit more from these experiences. In contrast, we did not find a differential effect of homeownership on the likelihood that foreign-born citizens would vote (Model 3b).

Overall, then, we find no evidence that foreign-born citizens benefit more than the native-born from prior participatory experiences or from stronger community attachment. Immigrants who have gone through the naturalization process and have registered to vote are already quite familiar with U.S. political institutions, including institutions at the local level. From this perspective, if a participatory deficit exists relative to native-born citizens, it is not because of lack of socialization into American politics or because naturalized citizens are politically inert. Rather, the participatory deficit is the result of other factors.

We hypothesized that the characteristics of cities and towns where naturalized immigrants live might be a reason for the participatory deficit of naturalized citizens. Large cities and cities with a manager form of government impose greater participatory costs that dampen the turnout of foreign-born citizens relative to native-born citizens, who have on average more participatory resources and are more likely to be mobilized by political campaigns. Our hierarchical models already allow for the possibility that the effect of Bforeign on voting in local elections varies across cities. We can further explore this variation by building a Level-2 model to explain the variation in the coefficient of Bforeign in terms of our Level-2 variables.

\begin{array}{l} V o t e_{i j} = β_{0 j} + β_{1 j} P A_{i j} + β_{2 j} P V_{i j} + β_{3 j} B f o r e g i n_{i j} + \\ β_{4 j} L R_{i j} + β_{5 j} H O_{i j} + \sum_{k = 1}^{7} β_{i j} X_{i j}; \end{array}

β_{0 j} = γ_{00} + γ_{01} M e d i u m C i t y_{j} + γ_{02} L a r g e C i t y_{j} + γ_{03} M a n a g e r_{j} + μ_{0 j};

β_{3 j} = γ_{30} γ_{31} M e d i u m C i t y_{j} + γ_{32} L a r g e C i t y_{j} + γ_{33} M a n a g e r_{j} + μ_{3 j} .

Through substitution, we obtain a single mixed-effects model:

\begin{array}{l} V o t e_{i j} = γ_{00} + γ_{01} M e d i u m C i t y_{j} + γ_{02} L a r g e C i t y_{j} + \\ γ_{03} M a n a g e r_{j} + β_{1 j} P A_{i j} + β_{2 j} P V_{i j} + γ_{30} B f o r e i g n_{i j} + \\ γ_{31} M e d i u m C i t y_{j} \times B f o r e i g n_{i j} + γ_{32} L a r g e C i t y_{j} \times \\ B f o r e i g n_{i j} + γ_{33} M a n a g e r_{j} \times B f o r e i g n_{i j} + β_{4 j} L R_{i j} + \\ β_{5 j} H O_{i j} + \sum_{k = 1}^{7} β_{i j} X_{i j} + μ_{3 j} \times B f o r e i g n_{i j} + μ_{0 j} . \end{array}

This model contains three fixed-effects cross-level interactions, MediumCity_j × Bforeign_ij, LargeCity_j × Bforeign_ij, and Manager_j × Bforeign_ij, which should all have negative coefficients if city-level factors affect the two groups differently. Model 4 in Table 4 shows the maximum-likelihood estimates of this multilevel model. Consistent with our expectation, both the cross-level interactions with medium and large cities are negative and statistically significant. This evidence indicates that larger cities have greater participatory costs for residents that disproportionately demobilize foreign-born citizens. In fact, the coefficient for Bforeign_ij, which is essentially equal to zero (the coefficient is −0.007, p = .9), indicates that in small towns, naturalized citizens are just as likely to turn out to vote in local elections as the native-born (i.e., when MediumCity and LargeCity = 0). However, the coefficient for the cross-level interaction between Foreign-Born and Manager was not significant, which is evidence that turnout is depressed more or less equally for both groups in cities with manager forms of government.

Although individual-level factors such as length of residency, level of education, and English-language proficiency are no doubt core factors that explain the participatory gap between native-born and foreign-born citizens, it appears that immigrant settlement patterns contribute to this gap and may also exacerbate it. Immigrants to Utah and to the United States as a whole tend to settle in larger cities compared with native-born citizens (see Table 2). According to Wilson and Singer (2011), 85% of immigrants live in one of the 100 largest metropolitan areas and 38% have settled in one of the five largest immigrant gateway cities. In Utah, 58% of foreign-born citizens live in a large city, while only 12% call a small town home. Settlement contexts in large immigrant gateway cities (e.g., Los Angeles, Houston, and New York) are very different than the settlement contexts in new immigrant gateways, so we are cautious about generalizing our findings to traditional immigrant gateways. Nonetheless, our findings suggest that the concentration of immigrants in large urban areas may be a part of the explanation for relatively low turnout rates by immigrants.

Discussion

This study sets out to assess the effect of individual and contextual factors on the voting behavior of foreign-born citizens in municipal elections. Our analysis reveals that, all else equal, naturalized citizens are less likely to vote than native-born citizens in local elections, a result that parallels findings for national elections. We find that previous democratic experience is the most powerful predictor of immigrant voting in a municipal election, whether the democratic experience was acquired here or in a home country. Naturalized citizens who have affiliated with a state-level political party or have prior voting experience in the United States are substantially more likely to vote in local elections than those who have not. All of this suggests that naturalized citizens respond quite enthusiastically to political mobilization and recruitment into politics, even in new immigrant destinations where stimuli for political engagement are generally scarce. We also find that immigrants from democratic countries arrive with a distinct participatory advantage over their counterparts from authoritarian regimes, evidence that any kind of exposure to democratic politics, whether here or in their home country, can encourage naturalized citizens to become more involved in local politics. We expected that deeper attachments to communities that result from longer residencies and from homeownership would also boost immigrant voting, but the evidence supporting these hypotheses was mixed. Longer residency in a city does boost immigrant political participation but homeownership does not.

City-level contextual factors also affect immigrant turnout in local elections. All else equal, foreign-born citizens are less likely to vote in cities with manager forms of government compared with cities with elected mayors. The likelihood that foreign-born citizens will vote in local elections also decreases as city size increases, with the lowest turnout rates occurring in the state’s largest cities. We also tested cross-level interactions to assess whether city contexts affect the political behavior of immigrants differently than native-born citizens. Overall, we find mixed support for our hypotheses that naturalized citizens are more sensitive to the effect of local contexts on turnout. On the one hand, we did not find evidence that manager forms of government have a different effect on the native-born than they do on foreign-born citizens. On the other, the negative coefficients of the cross-level interaction terms for city size and nativity indicate that the negative impact of living in larger cities is stronger among immigrant citizens than native-born ones. In fact, as Figure 1 shows, once we account for other factors, there is no difference in the turnout levels of foreign-born citizens and native-born citizens who live in small towns. Even more striking, turnout rates for immigrants who live in small towns are higher on average than turnout rates for native-born citizens who live in medium and large cities. It appears, then, that small towns rather than large cities create the most welcoming environment for immigrant political incorporation.

Figure 1.

Cross-level interactions showing the predicted probability of voting for foreign and native-born citizens by city size (with 95% confidence intervals).

Theoretically, the results demonstrate the importance of going beyond immigrant-specific factors to explain the voting behavior of foreign-born citizens. Immigrant-specific factors are important but so, too, are local political institutions, political dynamics, and contexts.²¹ From a policy perspective, we find that new immigrant destinations do not necessarily provide inhospitable settings for immigrant political incorporation. Recent immigrants are settling in cities and towns that are smaller in population and are located in southeastern states, where political parties are more active in local politics. This new settlement pattern may encourage rather than discourage the political incorporation of new arrivals. Similarly, in recent elections, both Democratic and Republican presidential candidates have aggressively courted the vote of naturalized citizens in swing states with rapidly growing immigrant populations like Nevada, Colorado, and Virginia. Our evidence suggests that if these recruitment efforts continue and immigrants gain in political experience, they will also become more involved in local politics. Despite the barriers associated with local-level participation that surely exist in new immigrant destinations, the results of this study give us reasons to be optimistic about the prospects for the political incorporation of immigrants who have entered the United States during the past two decades and have settled directly in places with little prior history of immigrant settlement.

Footnotes

Acknowledgements

The authors would like to thank the Russell Sage Foundation for generous financial support, as well as Tom Maloney, David Goldsmith, three anonymous reviewers, and Brian Gaines, all of whose comments on earlier drafts refined our thinking and our writing.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Russell Sage Foundation.

Notes

Author Biographies

Melissa M. Goldsmith earned her Ph.D. in political science from the University of Utah. Her research interests include the political behavior of citizens on the local and national levels. She is currently working as a research analyst.

Claudio A. Holzner is associate professor of Political Science and Associate Director of the Center for Latin American Studies at the University of Utah. His research focuses on the political behavior of poor people in Mexico and Latin America, and more recently on the political and civic engagement of legal and undocumented immigrants in the United States.

References

Alford

R. R.

Scoble

H. M.

(1968). Sources of local political involvement. American Political Science Review, 62, 1192-1206.

Andersen

(2008). Parties, organizations, and political incorporation: Immigrants in six U.S. cities. In Ramakrishnan

S. K.

Bloemraad

(Eds.), Civic hopes and political realities: Immigrants, community organizations, and political engagement (pp. 77-106). New York, NY: Russell Sage Foundation.

Ayón

D. R.

(2006). Spanish-language media and Mexican migrant civic participation. In Bada

Fox

Selee

(Eds.), Invisible no more: Mexican migrant civic participation in the United States. Available from www.wilsoncenter.org

Barreto

M. A.

Marks

M. A.

Woods

N. D.

(2007). Homeownership: Southern California’s new political fault line? Urban Affairs Review, 42, 315-334.

Barreto

M. A.

Segura

G. M.

Woods

N. D.

(2004). The mobilizing effect of majority-minority districts on Latino turnout. American Political Science Review, 98, 65-75.

Barreto

M. A.

Villarreal

Woods

N. D.

(2005). Metropolitan Latino political behavior: Voter turnout and candidate preference in Los Angeles. Journal of Urban Affairs, 27, 71-91.

Bass

L. E.

Casper

L. M.

(1999, February). Are there differences in registration and voting behavior between naturalized and native-born Americans? (Population Division Working Paper No. 28). Available from http://www.census.gov

Black

J. H.

Niemi

R. G.

Powell

G. B.

Jr. (1987). Age, resistance, and political learning in a new environment: The case of Canadian immigrants. Comparative Politics, 20, 73-84.

Bryk

A. S.

Raudenbush

S. W.

(1992). Hierarchical linear models: Applications and data analysis methods. Newbury Park, CA: SAGE.

10.

Bueker

C. S.

(2005). Political incorporation among immigrants from ten areas of origin: The persistence of source country effects. International Migration Review, 39, 103-140.

11.

Caren

(2007). Big city, big turnout? Electoral participation in American cities. Journal of Urban Affairs, 29, 31-46.

12.

DeSipio

(1996). Counting on the Latino vote: Latinos as a new electorate. Charlottesville: University of Virginia Press.

13.

Fortuny

Chaudry

Jargowsky

(2010). Immigration trends in metropolitan America, 1980-2007 (Urban Policy Institute Brief No. 1). Available from http://www.urban.org

14.

Guo

Zhao

(2000). Multilevel modeling for binary data. Annual Review of Sociology, 26, 441-462.

15.

Hajnal

Z. L.

(2009). America’s uneven democracy: Race, turnout, and representation in city politics. New York, NY: Cambridge University Press.

16.

Hajnal

Z. L.

Lewis

P. G.

(2003). Municipal institutions and voter turnout in local elections. Urban Affairs Review, 38, 645-668.

17.

Hero

Garcia

F. C.

Garcia

Pachon

(2000). Latino participation, partisanship, and office holding. Political Science & Politics, 33, 529-534.

18.

Hochschild

J. L.

Mollenkopf

J. H.

(Eds.). (2009). Bringing outsiders in: transatlantic perspectives on immigrant political incorporation. Ithaca, NY: Cornell University Press.

19.

Jackson

R. A.

(2003). Differential influences on Latino electoral participation. Political Behavior, 25, 339-366.

20.

Jones-Correa

(Ed.). (2001a). Governing American cities: Inter-ethnic coalitions, competition, and conflict. New York, NY: Russell Sage Foundation.

21.

Jones-Correa

(2001b). Institutional and contextual factors in immigrant naturalization and voting. Citizenship Studies, 5, 41-56.

22.

Karnig

A. K.

Walter

B. O.

(1983). Decline in municipal voter turnout: A function of changing structure. American Politics Research, 11, 491-505.

23.

Kaufmann

K. M.

Rodriguez

(2011). Political behavior in the context of racial diversity: The case for studying local politics. PS: Political Science & Politics, 44, 101-102.

24.

Kelleher

C. A. A.

Lowery

(2009). Central city size, metropolitan institutions and political participation. British Journal of Political Science, 39, 59-92.

25.

Kreft

de Leeuw

(1998). Introducing multilevel modeling. London, England: SAGE.

26.

Levin

(2013). Political inclusion of Latino immigrants: Becoming a citizen and political participation. American Politics Research, 41, 535-568.

27.

Lien

Collet

Wong

Ramakrishnan

S. K.

(2001). Asian Pacific-American public opinion and political participation. Political Science & Politics, 34, 625-630.

28.

Luke

D. A.

(2004). Multilevel modeling. Thousand Oaks, CA: SAGE.

29.

Marrow

H. B.

(2011). New destination dreaming: Immigration, race, and legal status in the rural American south. Stanford, CA: Stanford University Press.

30.

Michelson

M. R.

(2003). Getting out the Latino vote: How door-to-door canvassing influences voter turnout in rural central California. Political Behavior, 25, 247-263.

31.

Migration Policy Institute. (2012). MPI data hub: Migration facts, stats, and maps. Retrieved from http://www.migrationinformation.org/datahub/state2.cfm?ID=UT

32.

Mollenkopf

Olson

Ross

(2001). Immigrant political participation in New York and Los Angeles. In Jones-Correa

(Ed.), Governing cities (pp. 17-70). New York, NY: Russell Sage Foundation.

33.

Mollenkopf

Sonenshein

(2009). The new urban politics of integration: A view from the gateway cities. In Hochschild

J. L.

Mollenkopf

J. H.

(Eds.), Bringing outsiders in (pp. 74-92). Ithaca, NY: Cornell University Press.

34.

Niven

(2004). The mobilization solution? Face-to-face contact and voter turnout in a municipal election. Journal of Politics, 66, 868-884.

35.

Oliver

J. E.

(2000). City size and civic involvement in metropolitan America. American Political Science Review, 94, 361-373.

36.

Ong

P. M.

Nakanishi

D. T.

(1996). Becoming citizens, becoming voters: The naturalization and political participation of Asian Pacific immigrants. In Hing

B. O.

Lee

(Eds.), Reframing the immigration debate (pp. 275-305). Los Angeles: LEAP Asian Pacific American Public Policy Institute and University of California, Los Angeles, Asian American Studies Center.

37.

Pantoja

A. D.

Ramírez

Segura

G. M.

(2001). Citizens by choice, voters by necessity: Patterns in political mobilization by naturalized Latinos. Political Research Quarterly, 54, 729-750.

38.

Perlich

P. S.

(2008). Utah’s demographic transformation: A view into the future. Utah Economic and Business Review, 68. Available from http://bebr.utah.edu

39.

Portes

Mozo

(1985). The political adaptation process of Cubans and other ethnic minorities in the United States: A preliminary analysis. International Migration Review, 19, 35-63.

40.

Portes

Rumbaut

R. G.

(1996). Immigrant America: A portrait (2nd ed.). Berkeley: University of California Press.

41.

Primo

D. M.

Jacobsmeier

M. L.

Milyo

(2007). Estimating the impact of state policies and institutions with mixed-level models. State Politics & Policy Quarterly, 7, 446-459.

42.

Ramakrishnan

S. K.

(2005). Democracy in immigrant America: Changing demographics and political participation. Stanford, CA: Stanford University Press.

43.

Ramakrishnan

S. K.

Bloemraad

(2008). Civic hopes and political realities: Immigrants, community organizations, and political engagement. New York, NY: Russell Sage Foundation.

44.

Ramakrishnan

S. K.

Espenshade

T. J.

(2001). Immigrant incorporation and political participation in the United States. International Migration Review, 35, 870-909.

45.

Ramírez

(2005). Residential mobility and Latino political mobilization. Institute of Government Studies, University of California Berkeley. Retrieved from http://escholarship.org/uc/item/1086c1v4

46.

Singer

(2008). Twenty-first century gateways: An introduction. In Singer

Hardwick

S. W.

Caroline

B. B.

(Eds.), Twenty-first century gateways: Immigrant incorporation in suburban America (pp. 11-29). Washington, DC: Brookings Institution.

47.

Steenbergen

M. R.

Jones

(2002). Modeling multilevel data structures. American Journal of Political Science, 46, 218-237.

48.

Tam Cho

(1999). Naturalization, socialization, participation: Immigrants and (non) voting. Journal of Politics, 61, 1140-1155.

49.

Trounstine

(2009). All Politics is Local: The reemergence of the study of city politics. Perspectives on Politics, 7, 611-618.

50.

Trounstine

(2010). Representation and accountability in cities. American Review of Political Science, 13, 407-423.

51.

U.S. Census Bureau. (2010). American Community Survey 5-Year Estimates 2006-2008. Retrieved from http://www.census.gov/acs/www/data_documentation/2009_release/

52.

U.S. Census Bureau. (2011, November). Lifetime mobility in the U.S.: 2010 (American Community Survey Briefs). Retrieved from http://www.census.gov/prod/2011pubs/acsbr10-07.pdf

53.

Wals

S. C.

(2011). Does what happens in Los Mochis stay in Los Mochis? Explaining postmigration political behavior. Political Research Quarterly, 64, 600-611.

54.

Wilson

J. H.

Singer

(2011). Immigrants in 2010 metropolitan America: A decade of change. Washington, DC: Brookings Institution.

55.

Wong

J. S.

(2006). Democracy’s promise: Immigrants & American civic institutions. Ann Arbor: University of Michigan Press.

56.

Wood

(2002). Voter turnout in city elections. Urban Affairs Review, 38, 209-231.