County Sheriffs and Immigration and Customs Enforcement: The Will of the People or a Lack of Accountability?

Introduction

The creation of the sheriff’s office began in England in the ninth century, with duties including law enforcement, tax collection, executing writs, custody of prisoners, and holding court for the Crown (Tomberlin, 2018). Sheriffs have been responsible for “serving process papers, maintaining law and order, collecting taxes, and maintaining jails” (Tomberlin, 2018, p. 120). After the American Revolution, the office of sheriff grew to an elected and constitutional status in most states (Evans, 2019), though this move was not without controversy (DeHart, 2020). Over time, the role of sheriffs became more formal while the importance and power of the office has varied depending on geographical location (Tomberlin, 2018).

Historically and today, sheriffs in the United States have held a wealth of power, including oversight of serving eviction notices and domestic violence calls, and oversight of county jails. While city police forces have law enforcement and patrol powers, sheriffs are in charge of jail maintenance, prisoner transport, executing court orders, process serving, courtroom security, property seizures, fee, and tax collection, as well as other administrative tasks (Falcone & Wells, 1995). Also notable, the chief of police is usually appointed, while majority of sheriffs are elected.

There are 3142 counties across the nation, and 3083 of those counties across 46 states elect their sheriff (Thompson, 2020). This excludes Alaska, Connecticut and Hawaii who do not have sheriff’s offices and Rhode Island whose sheriffs are appointed by the Governor. Of the remaining 46 states that hold sheriff elections, five states and some counties hold non-partisan elections and 41 states (over 2700 counties) hold partisan sheriff elections (Thompson, 2020). As an elected position, sheriffs must campaign for office. This mechanism of democratic accountability assumes voters evaluate performance and policy decisions of elected officials, then use their ballots to show approval or disapproval. However, lack of competitiveness in sheriff elections (Zoorob, 2019) raises genuine concerns about the lack of an effective accountability measure and a check on their power. As elected officials, sheriffs serve as representatives for their constituents and the general public.

Despite the political power they hold, there is a substantial lack of research on county sheriffs. We know very little about the influences on their decision-making, or about how they use their power. In more recent years with a move toward devolution, there has been an increased interest in sub-state governance (Trounstine, 2009; Warshaw, 2019), as well as interest in law enforcement specifically; however, there is still a dearth of research on sheriffs. This study aims to determine if sheriffs make decisions based on the ideological preferences of their constituents, and if not, explores what factors could potentially be driving their decisions. We add to the literature by expanding the understanding about sheriffs overall. Specifically, we add to the understanding of whether sheriffs make policy decisions that are in congruence with public opinion by looking at sheriff compliance with U.S. Immigration and Customs Enforcement (ICE).

Sheriff Accountability

Theoretically speaking, local elections are the mechanism through which voters are able to hold sheriffs accountable in their position. If the county sheriff has been implementing policy that does not align with the voters, we should see this reflected by the sheriff being voted out of office. However, in practice, this is not what we observe. In fact, in some cases we are not even seeing sheriffs voted out of office when they break the law (Weill, 2017). Sheriff misconduct is something that is not rare, but rather seen across the nation (Greenblatt, 2018; Pishko, 2019). Between 2015–2016 there were 97 incidents reported about misconduct in the sheriff’s office (Zoorob, 2019). A few examples include a 2021 California civil rights investigation of the L.A. County Sheriff’s Department (Romero, 2021) and the conviction of theft and abuse of power for Alabama’s longest serving sheriff Mike Blakely (Remkus, 2021). Maricopa County Sheriff Joe Arpaio inhumanely had inmates sleeping outside in the midst of a heat wave (Weill, 2017). There are numerous examples of how some sheriffs have used their office to further their own personal interests. The presence of unchecked sheriff power over time is partly due to the political landscape surrounding sheriff’s offices. In many states, the position of sheriff is a county position that is elected at the county level, and yet has little oversight at the county or state level beyond an election. This means that if a sheriff acts illegally or unethically, elections (or conviction and prison) are the only way to remove them from their position. This is in contrast with local police departments who get their authority from state statutes at the local level (Zoorob, 2019). Without term limits and some level of oversight, sheriffs have an immense amount of discretionary power.

Sheriff Partnership with ICE

Immigration policy generally falls under the scope of the federal government’s control. However, increasingly, the federal government has turned to sheriffs for assistance (Jaggers et al., 2014). In 1996 Congress passed the Illegal Immigration Reform and Immigrant Responsibility Act (IIRIRA) which started the devolution of immigration policing powers to the states and localities. This legislation gave states and localities the ability to opt into enforcing federal immigration laws (Vaughan & Edwards Jr., 2009). Section 287(g) of the Immigration and Nationality Act (INA) provides state and local law enforcement with the opportunity to forge a partnership with the federal government (Coleman, 2012). While the INA was enacted in 1952 and amended in 1990, the first 287(g) contract was not signed until after the terrorist attacks on September 11, 2001. Since then, there has been an increase in the number of 287(g) agreements; (American Immigration Council, 2021; Capps et al., 2011) this is a result of devolution of immigration policy to the states and undocumented immigration.

Since 1970, the immigrant population has quadrupled to reach a record of 44.8 million in 2018 (Budiman, 2022). In addition to legal immigration, there has been a growth in the number of undocumented immigrants as well from around two million in 1980 to 11.5 million in 2011 (Hoefer et al., 2012). The population growth combined with the devolution of immigration policy to the local level has resulted in sheriffs calling for an expanded role in making immigration policy as well as enforcing it (National Sheriffs’ Association). Immigration enforcement at the county level presents itself as a working relationship between the sheriff and ICE. Sheriffs routinely run federal background checks on the individuals in their custody; these background checks are then shared with the Department of Homeland Security (DHS) which runs their own immigration checks on those in sheriff custody. Once DHS identifies someone who is undocumented, ICE then sends a detainer request to the sheriff asking them to hold the undocumented person in custody to give ICE time to collect them (Vaughan and Edwards, 2009). The sheriff then decides to either comply with the request or ignore it. While not a perfect full picture, this compliance can be used as indicator of the how sheriffs view their role in immigration policy. It can shed some light on how they view their discretionary power. Not only does the sheriff’s discretionary power allow them to make the decision to partner with ICE, but they also have control over other aspects of immigration policy:

“Sheriffs have some degree of control over how they want to engage their offices and deputies in the enforcement aspects of immigration law. These decisions include (but are not limited to) choices about generating formal policies on interacting with immigrants, when to check the immigration status of a suspect, witness, or an individual prior to booking, whether to accept foreign identifications and provide translation services, the degree to which the office will actively seek out opportunities for immigration enforcement (including engaging in raids and doing identification checks), and how much the office will cooperate with the federal and state governments in immigration enforcement actions” (Farris & Holman, 2017, p. 144).

There have been sheriffs across the nation who have been vocal supporters of compliance with ICE as well as those who oppose it. Some sheriffs view partnerships with ICE as beneficial to their communities by maximizing resources and enhancing public safety (Huennekens, 2018). Participating sheriffs have credited this partnership as a significant factor in reduced local crime rates, smaller jail populations, and reduced criminal justice costs (Vaughan and Edwards, 2009). Others believe this is a national issue and should not be handled at the county level. Overall, immigration is a salient issue and the highly politicized conversations surrounding it puts pressure on sheriffs (Capps et al., 2011). What factors sheriffs use to navigate this complex decision of cooperation is unknown.

Potential Factors Influencing Sheriff Decision-Making

Most of the literature about decision-making of law enforcement has largely depended on data that focuses specifically on police officers (Coyne & Bell, 2011; Schwartz, 2010). Relatively little is known about sheriffs as decision makers despite, as mentioned, their immense amount of discretionary power (and lack of oversight). The majority of sheriffs are elected officials, and some scholars have argued that elected officials at the local level are becoming increasingly nationalized, and make decisions based on their national party agenda (Hopkins, 2018). Nationalization occurs when voters refer to national cues such as presidential vote choice or party ID in order to make voting decisions in the state or local context (Hopkins, 2018; Jacobson, 2015). Nationalization lends to the argument that sheriffs could be making decisions based on the agenda of their national party since the majority of sheriffs run in partisan elections (Thompson, 2020).

This is especially pertinent to the issue of immigration and cooperation with ICE. Immigration has traditionally been an issue with decisions primarily centering at the national level. However, with this shift to county-level decision-making, potential nationalization of this position is something to consider. Farris and Holman (2017) found that ideology plays a role in shaping immigration attitudes and policies. Those who identify as liberal tend to oppose stricter immigration policies, while more conservative individuals tend to be more supportive (Walker & Leitner, 2011). County support or opposition for restrictive immigration policies would influence the decision making of a sheriff who is serving as a delegate to their constituents.

Another potential influence on a sheriff’s decision to partner with ICE could stem from potential economic benefits that accompany a relationship with ICE. Historically, curbing immigration has often been tied to economic interest (Burns & Gimpel, 2000). When citizens feel that the national economy is not doing well, there is a heightened sense of ‘restriction sentiment’ (Citrin et al., 1997). Further, during the rise in incarceration in the late 1990s and early 2000s, many economists believed that prisons could be an economic boom, bringing increased revenue and jobs to depressed areas (Cherry & Kunce, 2001; Littman, 2021) especially in rural locations (Tootle, 2004). Kang-Brown and Subramanian (2017) stated “some rural jails have built out capacity far in excess of what they need locally in order to maximize this opportunity for revenue, understanding that the more non-local people housed, the more a locality and its jail earns” (p. 22). Varsanyi et al. (2012) found that sheriffs who hold federal immigration detainees may receive some reimbursement from the federal government. In 2008 California jails received upwards of $55 million to hold immigration detainees in their facilities, with $34 million of that going to the Los Angeles Sheriff Department which had the largest contract nationally (Clarke, 2009). Economic influences like inflation, interest rates, and economic downturns influence the decision-making of local governments and those who make decisions within them (Bland, 1997). When asked about their partnership with ICE, Santa Ana Police Chief Paul Walters stated, “We treat [the jail] as a business, [budget] cuts could have been much deeper if it weren’t for the ability to raise money there.” (Clarke, 2009). In 2015, ICE paid jails to hold immigration detainees per diem rates from $30 to $168.84 per bed (TRAC Immigration, 2017). For sheriffs who are looking to increase revenue for their county, especially in rural areas, partnering with ICE may seem advantageous.

Outside of political and economic motivations, a sheriff’s decision to cooperate with ICE may also be based on their professional commitments. Law enforcement agencies work within jurisdictional confines, even though offenders are not constrained to these geographical boundaries (Pickering & Fox, 2022). Offenders’ mobility and the highly fragmented American law enforcement system suggest that inter-agency collaboration is needed in order to improve public safety. Instituting more collaboration among law enforcement was a priority of the Bush administration following the tragic 9/11 terrorist attacks. The Bush Administration created a communication plan for federal, state, local, and tribal law enforcement agencies to collaborate by sharing information (Kapucu, 2006). This intelligence-sharing initiative created a professional norm of interagency partnership and collaboration to aid in public safety efforts. Recently, in February 2024, Laken Riley, a 22-year-old nursing student in Georgia, was killed by an undocumented immigrant. This tragedy has sparked outrage and has given momentum to calls for stricter immigration reforms at the national, state, and local levels, including increased sheriff collaboration with ICE. The Georgia Sheriffs’ Association Executive Director, Terry Norris, supported House Bill 1105 which would require sheriffs to report to ICE when they have an inmate from another country. Georgia HB1105 states, “to require the commissioner of corrections to report certain information regarding the immigration status, offenses, and home countries of persons who are confined under the authority of the Department of Corrections” (House Bill, 1105).

The Patriot Act shifted the organizational culture of law enforcement agencies by authorizing more efficient information sharing between law enforcement and intelligence agencies to aid in safeguarding American communities (Peters & Woolley, 2005). Organizational culture is a significant factor that influences the degree to which agencies collaborate and can influence sheriff decision-making; shared belief systems, mutual interests, and a common purpose all play key roles in interagency collaboration (Cohen, 2018; Mitchell et al., 2015; Yang & Maxwell, 2011). Organizational culture can also result in sheriff resistance to interagency collaboration and information sharing. This relationship across law enforcement agencies can play a part in sheriffs’ decision to collaborate with ICE.

Sheriffs may face pressure from ICE themselves and some political party pressure to cooperate, while on the other side, if they cooperate, they may face intense public backlash. Immigration legal and advocacy groups across the nation, like the ACLU, have worked to make sure that immigrants know their rights and have worked to fight 287 g agreements. In some cases, because of decisions rendered like Clark County in 2019 (Rindels, 2019) sheriffs have had to resend or backtrack on their agreements.

The authority of the county sheriff is a combination of electoral behavior, policymaking, and administration (Farris & Holman, 2015). Sheriffs are, in most cases, elected officials and while there is not a vast literature on sheriffs’ receptiveness to public opinion, there is a dearth of studies that focus on how responsive other elected officials are to public opinion. Existing evidence asserts that elected officials want to be aligned with their constituents’ opinions, and even further, this influences officials’ behavior (Butler & Nickerson, 2011; Caughey & Warshaw, 2022; Erikson, 2013). However, these previous studies are not perfectly applicable to sheriffs, whose position is more insulated than other elected officials.

Sheriffs are given wide latitude in their decision-making regarding the management of the county jail. As the undocumented immigration population increases, sheriffs must decide how and when to cooperate with ICE. The decision about whether to cooperate with ICE is a complicated one that potentially involves weighing political preferences of their own as well as those of the county. This may involve thinking of this decision from a political perspective and whether cooperation is something that will be politically expedient for them in the next election cycle. As democratic representatives, if they are taking the will of the people into account, then they would look to the political opinion on immigration from their county constituents. Additionally, they may also consider the economic impact of a partnership with ICE and how it could potentially impact their jail revenue as well the county economy. Though the data about this is unclear, there is often the belief that cooperation would be economically beneficial for a county. Considering the previous sheriff literature demonstrating a lack of democratic accountability in sheriff elections, we hypothesize that sheriffs are not basing their decision-making on the opinions of the people in their county on immigration issues.

Data and Methodology

The unit of analysis for this study is the county. Because we are investigating the decision making of county sheriffs, we manipulated all variables to be county-level. For variables (like opinion data) that were at the individual level, we used the county mean for the county-level variable.

Independent Variables

To understand the county perspective on immigration, we used a set of 4 questions taken from the Cumulative CCES Policy Preferences dataset to create immigration positionality. This dataset combines a subset of CCES scores on specific issues from 2006–2020 (Cooperative Election Study, 2020) and standardizes the scores across years. We chose this dataset due to its overrepresentation of rural populations.

This data set includes four sub-questions about immigration asked over time (see Table 1). (The set contains five questions, but one was removed from our dataset because it was not asked in all the relevant years.) The main question was “What do you think the US Government should do about immigration?” With 4 possibly policy options asked about over time (see Table 1). These questions were re-coded so that all moved in the same direction and are considered individually in the model. Because the questions were asked of individuals, the mean positionality for each question was calculated at the county level to allow us to conduct county-level analysis.

OPEN IN VIEWER

Table 1.

Coding for Immigration Positionality Variable.

Variable Name	Direction coded	Question: What do you think the US government should do about immigration?
Legalize	+	Grant legal status to all illegal immigrants who have held jobs and paid taxes for at least 3 years, and not been convicted of any felony crimes.
Border	−	Increase the number of border patrol on the US-Mexican border.
Police	−	Allow police to question anyone they think may be in the country illegally.
Employer	−	Fine US businesses that hire illegal immigrants.

The next variable we considered is Political Ideology. The ideology variable was obtained from the CCES and scales political ideology from 1 strong liberal to 5 strong conservative. Like other individual level variables, mean values were calculated at the county-level to enable county-level analysis.

For the Rural/Urban variable we used the USDA Rural-Urban continuum codes, which scales counties by population, and gives distinction to each county as Metro or Nonmetro. For this variable we used a dummy variables with zero being metro/urban counties and 1 being rural/nonmetro. (USDA ERS - Rural-Urban Continuum Codes, 2021). We also included county demographic variables such as total population, population density, percent of the population that is white, percent of the population that is male, education level, percent of the population that is unemployed, median household income, foreign-born population, and the percent that is not a citizen. In order to measure the inequality rate, we used the GINI coefficient from the U.S. Census which is a common measure used by economists and political scientists alike (Brandolini & Smeeding, 2006; Midlarsky, 1988; Trump & White, 2018). ¹ County level economic data and demographic data came from the US Census American Community Survey data.

The Arrest Rate for the county came from the FBI’s Uniform Crime Reporting program and subsequent reporting, Crime in the United States. (USDOJ, 2003) which details different kind of arrests and violent crime rates for counties. We used the overall arrest rate for the county as a measure of crime within the county. Demographic data about the sheriff, such as their party affiliation, race and gender came from the Reflective Democracy Campaign. This dataset gathered data from voter files and other publicly accessible data on 3036 elected sheriffs in 46 states (Reflective Democracy Campaign, 2020).

Results

We conducted initial tests to see the relationship between the share of times the sheriff in a county cooperated with ICE and the political ideology of the county and the positionality of all the immigration policies combined. As you can see in Figure 1, there was no significant difference in sheriff-ICE compliance based on county level ideology or immigration positionality.OPEN IN VIEWER

We then tested the relationship between the county sheriff’s cooperation with ICE criminal justice and economic factors such as the arrest rate and the unemployment rate and found that there was a significant relationship. Both were significant and have a positive relationship, meaning that as the arrest rate and the percent of unemployed in the county increase, the sheriff’s cooperation also is predicted to increase.

To test this relationship further, we conducted a regression that included demographic variables about the sheriff (race, gender, political party), positionality on immigration, county census data, and the county arrest rate. This full model gives us a better idea of what possible influences exist on the sheriff’s cooperation with ICE.

The findings in the full model, displayed in Table 2 show that neither the individual sheriff demographics nor political ideology of the county have a significant relationship with the cooperation of the sheriff, demonstrating that neither the sheriff’s personal political ideology, nor the county ideology significantly affect the decision making of the county sheriff cooperation with ICE.

OPEN IN VIEWER

Table 2.

Regression for Share of Sheriff Cooperating With ICE.

	Estimate	Std. Error
Sheriff demographics
Sheriff’s race	0.03	0.03
Sheriff’s political party	−0.01	0.01
Sheriff’s gender	−0.06	0.06
County demographics
County arrest rate	0.00	0.00	**
Urban/Rural county	−1.54	0.92	.
County population	0.00	0.00	.
County population density	0.00	0.00
Percent of white population	0.00	0.00	*
Percent of male population	0.01	0.01
Percent with a Bachelor’s degree or more	0.01	0.00	**
County percent unemployment	0.02	0.01	***
County median household income	0.00	0.00
County GINI inequality score	−2.27	0.79	**
County percent foreign born population	0.02	0.01	.
County percent not a citizen	−0.01	0.02
County political variables
Mean county political ideology	0.03	0.04
CCES Imm question: Police Questioning	−0.21	0.11	.
CCES Imm question: Path to Legalization	0.05	0.09
CCES Imm question: Employer Duty	−0.09	0.12
CCES Imm question: Border	0.14	0.13
GINI Index*Urban/Rural	3.41	2.08

There were, however, several significant relationships between sheriffs’ cooperation with ICE and county demographics. First, there is a positive relationship with the percent of the county that is White, demonstrating that as the proportion of White population increases there is a higher likelihood of the sheriff cooperation with ICE.

In addition to county demographic data, there are significant relationships with county economic demographics. Both the education of the county and the unemployment rate are significantly related to the sheriff’s cooperation. It is interesting to consider both of these findings simultaneously. As the percent of those with a bachelor’s degree or more increases the sheriff cooperation increases, so as the mean education level increases, the more likely the sheriff is to cooperate with ICE. Additionally, as the unemployment rate increases the cooperation also increases (Figure 2), so as the more unemployment there is in the county, this also makes the sheriff more likely to cooperate with ICE. When thinking about these two together, this could indicate that as there are higher levels of education and high levels of unemployment rate together, the population may feel more threatened by the immigration population, increasing the cooperation with ICE.OPEN IN VIEWER

Further, the cooperation rate is significantly related to the inequality score. As inequality increases in the county, sheriffs are 2.27 times more likely to cooperate. There is also a significant relationship with the arrest rate. As the arrest rate increases the cooperation rate with ICE also increases (Figure 3). There are several theoretical reasons this could be including that in counties where there are more arrests, the jails may be at higher capacity, urging the sheriff to cooperate more with ICE. On the other hand, the high arrest rate could be indicative of the type of law enforcement presence in the county, indicating that the Sherriff is more likely to cooperate with ICE.OPEN IN VIEWER

We also find a differences between urban and rural counties. Figure 4 demonstrates the relationship between the GINI Inequality rate and the sheriff’s compliance for urban and rural counties. While this interaction is not statistically significant, the figure demonstrates that there is a marked difference in the interaction between inequality and the sheriff’s compliance rate in each area. Ironically in urban areas as the inequality rate decreases, the compliance rate increases, whereas in rural areas, as the inequality increases, the compliance with ICE increases.OPEN IN VIEWER

Conclusion and Discussion

In this study we examined the relationship that county sheriffs have developed with ICE, specifically wondering whether what might be related with the sheriff’s being more likely to cooperate. As a democratically elected official, we wondered whether the decisions about how the sheriff cooperates with ICE was driven by citizen ideology on issues of immigration. We acknowledge as researchers that these limited set of policy questions around immigration from CCES do not capture the full perspective of a person’s view on immigration, they are the best proxies we have to investigate this question. We find that citizen’s ideology and positionality on immigration issues do not affect the cooperation rate. Further, neither do the demographics of the sheriff, including the sheriff’s political preferences. ²

What seems to matter more than the sheriff’s personal political ideology or the county’s mean ideology, is the demographics and economics of the county. This reinforces the idea that decisions around the criminal/legal system may not be based on politics or political ideology around crime and justice, rather law enforcement leaders are basing these decisions on economics (Zoorob, 2019). There is an assumption that that this relationship will be an economic boost to the county and the jail and enable the sheriff to bring in additional revenue. So, while the public may see this as a public safety issue, this model suggests perhaps revenue and the economic variables more likely to be the driver.

These findings call into question the idea of democratic accountability being a mechanism to hold sheriffs accountable for their decisions. As mentioned previously, the sheriff is a precarious position because the only mechanism of accountability is the voter. In most states there is not another way to remove a sheriff from their position. Because of this, testing and ensuring democratic accountability is of utmost importance for this role. Further, it points to a possible democratic deficit on this issue, that sheriffs are not basing their decisions about cooperating with ICE on the feelings of the citizens in their own county. Perhaps citizens are not aware of the ways that sheriffs are cooperating with ICE. While it is public information, unless citizens are seeking out this information it may not be something that they are paying close attention to. As Zoorob (2019) points out, sheriffs often run unopposed, keeping citizens from having a true point of competition or choice should they disagree with the actions of the sheriff.

In future research there needs to be more investigation of the power that sheriffs hold. This can be a difficult task due to the lack of open data around this position. However, as we indicated, this position is ripe with power that is largely going unchecked. Just as there has been renewed attention to politics at the subnational level, our hope is that there will be attention paid to sheriffs. They sit at the unique intersection between local politics and law enforcement, both of which are salient topics, and yet this position is largely overlooked in the literature. We believe that because of the expansive nature of the sheriff’s office there are many ways to explore issues of democratic accountability and we look forward to thinking more about this powerful, often overlooked, position within local government.