1 Introduction

On May 25, 2020, George Floyd was killed by a police officer in Minneapolis, Minnesota. Throughout the United States, protestors took to the streets. Police responded with substantial surveillance, arrests, and force. While local police departments were the most visible at the time, in some cities Customs and Border Protection (CBP) was involved in policing these protests. For example, on May 29, CBP sent a drone to fly over Minneapolis to surveil protests. CBP engaged in similar aerial surveillance of protests in fifteen cities within the first three weeks of Floyd’s death. In July, members of the Border Patrol Tactical Unit (BORTAC), the U.S. Border Patrol’s top SWAT team, grabbed at least four individuals and dragged them into unmarked vans near protests at a federal courthouse in Portland, Oregon (Jones, 2023).

Policing protests in American cities has little direct connection to border security. Yet personnel, equipment, and powers that were granted to CBP in the name of border security were redirected towards domestic policing of protests. This mission creep illustrates a broader problem with decisions about granting powers and resources to government bureaucracies. In a changing world, policymakers cannot predict ex ante all the ways that bureaucrats will use the powers and tools at their disposal. Once they have been given powers, hardware, training, or personnel, public officials face incentives to devise new ways to deploy the resources at their disposal. This creates conditions where policymakers and voters cannot know ex ante how powers they grant will be used in the future.

The purpose of this paper is to analyze how this mission creep occurs. I argue that both officials and citizens think about border regions in a way that creates greater leeway for experimenting with techniques of social control. This gives rise to border militarization. Once militarized tactics and tools are adopted at the border, officials have incentives to think creatively about other ways to use the tools at their disposal. One reason is to make it clear to policymakers that they are using the resources at their disposal, thereby increasing the chance that they will continue receiving such resources (Niskanen, 1968). Another reason is that creative uses of their powers can enable bureaucrats to aid higher-level officials in achieving policy objectives. This can improve the bureaucrat’s chances of career advancement (Breton & Wintrobe, 1986).

This paper contributes to a broad literature in public choice and political economy on the expansion of government power over time. A substantial literature in public choice and constitutional political economy examines the growth of government over time, focusing on a range of factors that contribute to that growth, including ideology, responses to crises (Goodman et al., 2021; Higgs, 1987), technology (Cowen, 2021), the incentives facing bureaucrats (Breton & Wintrobe, 1986; Niskanen, 1968), and rent-seeking distributional coalitions (Holcombe 2018; Olson, 1982). Salter (2017) analyzes “constitutional drift,” in which the informal constitutional constraints on political officials shift over time as an “unintended result of political bargains among political elites” (p. 569). Coyne and Hall (2014, 2018) argue that during foreign interventions officials experiment with new techniques of state control, which results in the development of organizational structures, physical capital, and human capital that are brought home and deployed domestically. Techniques of social control used in foreign intervention then give rise to domestic police militarization, mass surveillance, and other expansions in state control.

I argue that the dynamics Coyne and Hall identify in the case of foreign intervention apply to border security efforts as well. Just as officials have substantially greater leeway to experiment with coercive power abroad, the border is perceived as a zone in which greater coercive power is permissible than the domestic interior of the country. This creates an expanded scope for experimentation with state control. Just as experimentation with state control during foreign interventions does not remain confined to the realm of foreign policy, experimentation with state control at the border does not remain confined to border regions. Instead, tools and powers developed for use at the border are eventually used for domestic policing. Militarizing the border does not merely militarize the border: it militarizes domestic life.

In addition, border militarization centralizes policing. Activities traditionally associated with local police come to be carried out in partnership with the US Border Patrol. This shifts the “fiscal attention” of police away from local citizen-taxpayers and towards officials in the central government (Boettke et al., 2017; Enninga & Goodman, 2023). This can be understood as part of a broader constitutional drift away from competitive federalism towards cartel federalism (Blankart, 2000; Brennan & Buchanan, 1980; Greve, 2012, 2015; Salter, 2017; Wagner, 2014). While this shift has typically been analyzed in terms of reduced fiscal competition and therefore increased monopoly power when it comes to taxing citizens, the move to a monopolistic or cartelized federalist system could also reduce incentives for police to protect and serve. By reducing opportunities for interjurisdictional competition, centralization may give police relatively unconstrained power to engage in predation against residents of their jurisdictions. If a surveillance practice or militarized police approach is unique to a jurisdiction, citizens who find this practice onerous or unconscionable can move to a new jurisdiction. If it is carried out in partnership with a police force from the national government, both voice and exit become far less effective as checks on police power.

A multidisciplinary literature documents border militarization. This militarization involves use of military hardware, training, and tactics for border security purposes. Some scholars trace the history of border militarization, analyzing the political processes that militarized the U.S.-Mexico border (Andreas, 2009; Dunn, 1996, 2009; Hernandez, 2010). Some analyze the political incentives and interest groups that shape border militarization (Coyne & Goodman, 2020a; Golash-Boza, 2009; Miller, 2019a). Others analyze the consequences of border militarization, particularly the increased migrant deaths that result from diverting migrants to more dangerous routes (Chambers et al., 2021; Cornelius, 2001; Cornelius & Lewis, 2007; Dunn, 2009; Huspek, 2001; Michalowski, 2007). Some focus on the violence entailed by border militarization (Jones, 2016; Slack et al., 2016). Border militarization can also be analyzed in terms of expression or discourse (Andreas, 2009; Palafox, 2000). Closest to my analysis is Jones (2023), who examines the development of the United States Border Patrol as a police force and the way that development has shaped American policing. I argue that this development reflects the incentives bureaucrats face in a world characterized by a combination of political incentives, uncertainty, and novelty. Therefore, the dynamics of border militarization and its impact on domestic institutions has broader implications for the expansion of government power over time. Moreover, I follow Enninga and Goodman (2023) in emphasizing how border militarization renders institutions more monocentric and less polycentric.

This paper’s historical evidence is entirely drawn from a reading of the secondary literature, rather than novel empirical work. The paper uses economic theory to interpret the empirical and historical record laid out in the existing literature, using this to construct an analytic narrative (Bates et al., 1998; Skarbek & Skarbek, 2023). This analytic narrative contributes to the multidisciplinary literature on border militarization by offering a theory about border militarization’s broader institutional implications, the incentives and institutions that drive the ongoing militarization, and the welfare implications of entrepreneurial processes within the context of state-provided border security. It also contributes to the literature on constitutional political economy by illustrating how political entrepreneurship within law enforcement organizations contributes to institutional change and elucidating the likely welfare consequences of such change.

The paper proceeds as follows. Section 2 presents a theory of how bureaucrats operating in an environment characterized by novelty and uncertainty experiment with different techniques for state control and creatively devise new uses for these techniques. Section 3 discusses the history of border militarization and the experimentation entailed in said border militarization. Section 4 discusses how tactics and tools originally granted for border security purposes have been applied for law enforcement tasks in the interior of the United States. Section 5 concludes, discussing broader implications of our analysis for understanding the growth of government and the costs and benefits of public policies.

2 Power, creativity, and mission creep

2.1 Bureaucratic entrepreneurship through time

Rather than thinking of “the state” as a unitary entity, it is worth analyzing the incentives facing individuals interacting within a political process (Buchanan, 1949; Coyne, 2015). This approach empowers us to analyze the incentives facing voters, politicians, and bureaucrats and how those incentives lead individuals to act in a manner that alters the formal and informal rules that structure governmental activity.

Our analysis focuses primarily on the incentives facing bureaucrats, defined here as individuals employed within hierarchically structured government agencies (Tullock, 1965, 1992). There are several plausible ways to model the motivations of bureaucrats. One focuses on bureaucrats as attempting to maximize their budgets (Niskanen, 1968). This view is plausible because a larger budget enables bureaucrats to secure benefits in the form of job security, wages, and benefits for themselves and their colleagues. Moreover, if bureaucrats are public spirited and mission driven, a larger budget provides them with more resources to carry out their mission. Because bureaucrats are not residual claimants and are not accountable to a defined group of residual claimants, they have little incentive to seek out cost savings. However, while the budgets of their bureaus are one important concern for bureaucrats, they are not the only thing that they value. Breton and Wintrobe (1983, 1986) illuminate additional aspects of bureaucratic behavior, such as competition and exchange among bureaucrats. They argue that bureaucrats value career advancement and compete with one another to serve the interests of their superiors within a hierarchy, thus increasing their future opportunities for advancement.

This competitive process creates a role for entrepreneurial creativity on the part of bureaucrats. While entrepreneurs in a private market are alert to opportunities to seize pecuniary profits by using resources in ways that consumers value more than current uses (Kirzner, 1973), bureaucratic entrepreneurs are alert to opportunities to engage in activities that advance the policy goals of their superiors.

Advancing these policy goals often involves the use of coercion and social control. If individuals were already voluntarily acting as political leaders and bureaucrats prefer, policy interventions would be unnecessary. While policy goals can sometimes be obtained through soft power such as persuasion, propaganda, or nudging, policies can also be implemented through more militarized means such as surveillance, raids, arrests, beatings, or the use of arms and weapons. Bureaucrats choose among the means at their disposal, and they may also at times creatively experiment with means that they previously have not deployed.

Experimentation with techniques for state control is limited by formal and informal institutions that limit what is considered possible and acceptable. Some of these informal rules result from prevailing ideological views about the permissible scope of government power and activity (Higgs, 1987). However, ideological views about the permissible scope of government power and political activity can vary geographically. Citizens and politicians may believe that certain powers are impermissible to wield against citizens domestically but permissible to use abroad as tools of foreign policy. This results in a situation where bureaucrats can experiment with more techniques of social control during foreign interventions than they can domestically (Coyne & Hall, 2014, 2018).

A similar dynamic occurs at the border. There are several reasons for this. One is the close ties between “border security” and “national security” in contemporary ideology. Borders define a state’s political boundaries, and the notion of protecting territorial integrity and sovereignty from perceived threats is often seen as a matter of national security. This became even more explicit in the United States when the Border Patrol became part of the Department of Homeland Security in 2003. Another reason officials have broader leeway at the border is that many people believe states have duties to their citizens that they do not have to non-citizens. In practice, immigration restrictions and border security efforts impact the liberties of citizens as well as non-citizens (Coyne & Goodman, 2020a; Kukathas, 2021). However, non-citizens are often the most salient targets of border security efforts, which creates a broader scope for government activity, as the targets are considered “other” rather than members of the polity.

One illustrative example of the special status granted to border security is the so-called “100 mile zone.” Within 100 miles of the U.S. border, including coastlines, the Border Patrol has expanded powers of search and seizure (Anthony, 2020). Critics have referred to this as a “Constitution-free zone,” arguing that the typical requirements of the Fourth Amendment to the U.S. Constitution have effectively been waived (Dorsey & Díaz-Barriga, 2015). The 100-mile zone illustrates the way that judges interpret the border as a special place where a wider scope of state power is permissible. This broader leeway for experimentation with state control is secured not just by the courts, but also by prevailing ideology among policymakers and the public.

Using this broader leeway, officials involved in border security can experiment with a broad range of approaches to state control. This can mean acquiring legal powers, such as powers of search and seizure, that other law enforcement officers do not yet have. It can mean developing new organizational structures. It may mean acquiring physical capital, such as weapons, unmanned aerial vehicles, or surveillance technology. Finally, it may mean developing human capital in the form of skills and knowledge related to the social control techniques being used.

Once each of these are developed, bureaucrats may use the physical capital, human capital, powers, or organizational structures for new purposes. When policymakers authorize an intervention or operation that results in these developments, they do not know ex ante how they will be used in the future. Bureaucrats creatively seek out opportunities to advance the goals of their superiors, as well as potential future employers. There are new policy objectives and perceived security threats that arise over time, and the novelty of these opportunities generate unpredictable dynamics within the system. Something introduced to address a perceived crisis at the border may later be repurposed to respond to high profile protests and riots within the nation’s interior, for instance.

Whenever a bureaucrat chooses the means to advance a policy objective they have been tasked with pursuing, their choice is shaped by the available means at their disposal. The skills and human capital bureaucrats have shape which strategies are within their opportunity set. Likewise, if an organization currently possesses a piece of physical capital, individuals within that organization will be familiar with that equipment’s capabilities and think creatively about additional ways to use it. Similarly, the creation of an organizational structure generates incentives for those who work within that organizational structure to demonstrate their team and organization’s ongoing usefulness to policymakers.

As a result, powers, organizational arrangements, and human and physical capital developed for one purpose will rarely be confined to that single purpose. Moreover, they will rarely remain confined to the geographical area where they were originally deployed. Coyne and Hall (2014, 2018) illustrate this through several case studies where tools of social control developed during foreign interventions are brought back home. For example, surveillance techniques initially used by Captain Ralph Van Deman during the U.S. occupation of the Philippines were techniques that he eventually used to develop domestic surveillance capabilities. Likewise, the first SWAT team was the brainchild of a Vietnam veteran, John Nelson, who built on his experience in a Force Recon unit.

Just as the creative use of powers and tools developed in foreign interventions results in expanded state control domestically, powers and tools developed for border security can be creatively repurposed domestically. Moreover, as the previously mentioned 100-mile border zone illustrates, the distinction between activity at the border and activity within the country is somewhat fuzzy. The paradigmatic case of something that is truly at the border is a checkpoint at a port of entry (POE). But given that anywhere within 100-miles of the border is interpreted as a place for expanded Border Patrol search and seizure powers, the Border Patrol is able to set up internal checkpoints. As Gonzales (2015) explains:

CBP is tasked with maintaining internal traffic checkpoints directed at deterring illegal immigration and smuggling activities that may have gone undetected at official border crossings. Unlike POEs, these checkpoints are situated in the interior United States, and affect domestic as well as international travelers. Several checkpoints in the Southern District are situated on major highways and thus regulate a high volume of traffic. While drivers are usually waved through these checkpoints and only delayed a moment, they can be stopped, questioned, and delayed for much longer if the CBP officer on duty suspects illegal aliens might be in the car. (Gonzales 2015: 204–205)

In this respect, even activities formally aimed at border security within recognized border regions can begin to seriously impact the liberty and privacy of domestic travelers who never even attempt to cross the border. Creative entrepreneurial action by CBP agents can expand the scope of coercive activities within the interior of the U.S. for ostensibly border security purposes. Moreover, entrepreneurship both within and outside of CBP can be used to discover ways to wield powers and tools developed at the border in new domestic contexts. Before using historical evidence to illustrate how this process occurs, it is worthwhile to examine the expected welfare implications of this entrepreneurial process.

2.2 Welfare implications of bureaucratic entrepreneurship

An optimistic view of this entrepreneurial process might argue that it leads bureaucrats to nimbly adapt to novel circumstances, thereby effectively using the tools and powers at their disposal to address new challenges and effectively promote the welfare of citizen-consumers. For instance, by sharing technology with other police forces, they may effectively protect citizens from crime and predation. To examine whether this optimistic view is correct, it is worth looking a bit more closely at the institutional contexts within which entrepreneurship can take place.

Entrepreneurship within a market, characterized by institutions of property, contract, and consent, tends to move scarce resources to their higher valued uses. Market entrepreneurs are alert to profit opportunities that result from disequilibrium prices, and they discover ways to use productive inputs in ways that obtain higher profits (Kirzner, 1973). Such profits reflect the fact that consumers, at least ex ante, value the output more than other consumers value alternative uses of the relevant inputs.

A core difference between political entrepreneurship and market entrepreneurship is that political entrepreneurs are legally empowered to coerce third parties. Entrepreneurs operating in a market characterized by property, contract, and consent may only purchase inputs from willing sellers and sell products to willing buyers. By contrast, political entrepreneurs have the power to legally coerce unwilling persons. A CBP agent can detain or arrest someone, for instance. Where a voluntary exchange can be assumed mutually beneficial, at least ex ante, an act of coercion can be presumed to harm the person being coerced. This may still be beneficial on net. After all, it is possible that the coercion is used to protect citizens from predation. For instance, a CBP agent might arrest a terrorist who plans to harm many citizens. However, coercion could also be wielded in a manner that is purely predatory, harming the coerced person without meaningfully protecting anyone. Regardless, the coercive nature of the interaction means that we cannot necessarily consider it beneficial ex ante. This has significant implications for evaluating the welfare consequences of entrepreneurial activity. Within markets, exchanges are voluntary, and therefore ex ante mutually beneficial. These mutually beneficial exchanges give rise to exchange ratios, or prices, that can be used to assess tradeoffs and engage in rational economic calculation.

By contrast, “public administration begins where economic calculation ends” (Aligica et al., 2019: 17–18). Bureaucrats in this context obtain resources via political means, and their outputs are not sold on the market. In the absence of prices, property, and profit and loss feedback, economic calculation is not feasible, so we cannot expect bureaucratic entrepreneurship to deliver the beneficial welfare consequences typically associated with the entrepreneurial market process (Mises, 1920, 1944).

Despite this, some argue that other feedback mechanisms will guide bureaucratic entrepreneurs to use their creativity in a manner that serves the interests of citizens (Wittman, 1995). Consider the feedback provided by voting. In a representative democracy, bureaucratic competition to satisfy politicians may indirectly induce bureaucrats to act in the interest of voters, as politicians strive to win elections. This may result in bureaucrats creatively acting to satisfy the anticipated preferences of the median voter (Black, 1948; Downs, 1957). However, several issues limit the effectiveness of voter feedback. Due to the duration of time between elections, voters may not be able to hold a politician (or a bureaucrat they appointed) accountable until years after damage has been done (Boudreaux, 1996). Moreover, past malfeasance may no longer be salient by the time of the next election. These issues are exacerbated by bundling of policies during elections. A voter may agree with a politician on healthcare policy but disagree on border policy, and that voter has no way to unbundle their single vote to signal their views on distinct policies (Coyne & Goodman, 2020b). An individual voter faces the further problem that their single vote is exceedingly unlikely to decide an election (Gelman et al., 2012). This means the voter has little incentive to acquire information and will therefore often be “rationally ignorant” about political facts (Somin, 2016). This challenge is further exacerbated by information asymmetries, in which relevant information is available to political insiders but not to outsiders. For instance, information on the number of times CBP lent drones to other law enforcement organizations was initially not available to the general public. It only became available after a Freedom of Information Act lawsuit by the Electronic Frontier Foundation (Lynch, 2013). Such information asymmetries can be especially pronounced regarding national security issues, as much information is classified and therefore only available to individuals with security clearances (Coyne et al., 2019). Together, these issues limit the extent to which voter feedback can limit political opportunism. Bureaucratic entrepreneurs will not necessarily be guided towards serving the interests of the median voter.

In addition to the use of voice by voters, another possible political feedback mechanism comes from exit (Hirschman, 1970). If citizens can easily move between jurisdictions, then they may move to jurisdictions that offer public services that fit their preferences or move away from jurisdictions whose leaders coerce them in ways they find onerous. If leaders’ budgets are largely dependent on citizens remaining within their jurisdiction, then this process of interjurisdictional competition will function like a quasi-market, encouraging political entrepreneurs to satisfy citizens in order to maintain a fiscal base (Tiebout, 1956). When citizens choose to vote with their feet, their choice directly decides where they will reside, which means that citizens have better incentives to acquire relevant information (Somin, 2016).

In practice, however, this process is imperfect. There are significant costs to moving between jurisdictions. Moreover, local officials do not solely rely on their local tax base, and many of the taxes and other forms of coercion citizens face are decided at the national rather than local level. This can turn competitive federalism into cartel federalism, in which local political officials have monopoly privileges and are insulated from interjurisdictional competition (Blankart, 2000; Greve, 2012, 2015; Wagner, 2014). The dynamic entrepreneurial process this paper documents exacerbates both imperfections. To the extent that local police rely on assistance from federal agencies such as CBP, they become relatively less dependent upon local taxpayers. This shifts their “fiscal attention” from their local constituents to national political officials (Boettke et al., 2017). Moreover, to the extent that border militarization renders it more difficult to move between jurisdictions, interjurisdictional competition is hampered (Somin, 2008, 2021). By limiting the ability of migrants to exit the jurisdictions of other governments, border militarization forcibly keeps subjects and taxpayers under the control of governments whose territory they wish to exit. These dual problems mean that border militarization simultaneously restricts ability to exit at the international level and limits the political significance of exit within the borders of the United States.

Voting and interjurisdictional competition are not the only factors that can guide politicians and bureaucrats to serve the interests of citizens. Citizen voice in the form of public participation, input, and social movement activity may encourage leaders to serve the interests of citizens (Novak, 2021; Hegre et al., 2020). Coproduction of public services may align the interests of citizens and bureaucrats while empowering bureaucrats to take advantage of citizens’ local knowledge (Boettke et al., 2016). But just as interjurisdictional competition is undermined by the bureaucratic entrepreneurship discussed in this paper, these feedback mechanisms are similarly hampered by bureaucratic entrepreneurship and mission creep. To the extent that local police have their fiscal attention reoriented towards national decision-makers, they may be less sensitive to concerns revealed through citizen voice. The coercive powers or tools that bureaucrats learn to redeploy may be wielded to surveil, coerce, or repress social movement activity, thereby limiting citizen participation. The surveillance of Black Lives Matter activists discussed in the introduction is one example of this. Moreover, by expanding the scope of the central government and limiting functional polycentricity, bureaucratic entrepreneurs may reduce the scope for citizen coproduction of public services.

This analysis suggests that bureaucratic entrepreneurs, particularly within the central government, are likely to erode the very institutional context that may otherwise guide political entrepreneurs to serve the interests of citizens. As the following sections show, border militarization has involved alert and creative bureaucratic entrepreneurs using their powers and tools in novel ways and centralizing political power in the process.

3 Militarizing the border

Security along the U.S.-Mexico border has been increasingly militarized. This militarization has occurred along multiple margins. This paper focuses on two key aspects of this militarization that have been moved from being simply used in border regions to being used in the domestic interior:first, aerial surveillance technology, and, second, teams within the Border Patrol that adopt military-inspired organizational structure and training programs.

3.1 Aerial surveillance at the border

Aerial surveillance at the U.S.-Mexico border has a long history. For instance, “in late 1945, the military provided the Border Patrol with three surplus Stinson L-5 airplanes” (Hernandez, 2010: 105). Pilots of these aircraft used radios to communicate with Border Patrol agents in cars on the ground, which enabled them to quickly find and apprehend migrant workers identified through aerial surveillance. “By 1953, at least one plane and pilot was assigned to each sector along the U.S.-Mexico border” (Hernandez, 2010: 105).

The Border Patrol’s fleet of surveillance aircraft continued to expand. For instance, the Border Patrol acquired additional small fixed-wing aircraft during the late 1970s (Dunn, 1996: 38). The Reagan administration expanded the Border Patrol’s fleet of fixed-wing aircraft from 28 to 46 planes (Dunn, 1996: 44). The fleet of helicopters was also expanded, from two helicopters to 42 helicopters (Dunn, 1996: 43). Helicopters used infrared radar for surveillance purposes, though they also intimidated migrants using searchlights and loudspeakers (Dunn, 1996: 43–44).

In the twenty-first century, aerial surveillance was increasingly carried out using Unmanned Aerial Vehicles (UAVs), commonly known as drones. “In 2002, the Customs and Border Patrol began using drones along the U.S. border as a way to stop illegal immigration and prevent those with possible terrorist ties from entering the country” (Coyne & Hall, 2018: 130). Starting in “fiscal year 2006, CBP’s Air and Marine Operations has operated Predator B drones, a version of the military’s MQ-9 Reaper drone, along the border” (Bier & Feeney, 2018: 1). The Border Patrol has experimented with additional drone technologies as well. For instance, in “September 2017, CBP began testing smaller hand-launched drones, including AeroVironment’s Raven and Puma small unmanned aircraft systems” (Bier & Feeney, 2018: 1).

Drones carry equipment that enables surveillance of individuals and vehicles. This includes radar systems as well cameras with infrared capabilities, all of which can be used to identify “persons or vehicles and track them on the ground in real-time” (Bier & Feeney, 2018: 4). Predator drones carry this equipment at heights “between 19,000 to 28,000 feet” (Bier & Feeney, 2018: 4).

Smaller hand-launched drones can fly at lower elevations, enabling them to obtain more detailed information on the targets of their surveillance. CBP would like to enable these small drones to use facial recognition technology, which raises a variety of civil liberties concerns surrounding privacy, errors, and racial and gender biases in the distributions of these errors. The error-prone nature of these systems would create “increased risk of detaining law-abiding people as suspected border crossers” (Bier & Feeney, 2018: 4). Combining drones with facial recognition continues the decades long trend of the Border Patrol embracing more advanced and invasive aerial surveillance systems.

As of 2020, CBP had “more than 135 of these [drone] systems in use throughout the country, with 60 more in the procurement process. Plans are eventually to have 460 drones patrol from above” (Davis, 2020). CBP’s aerial surveillance technologies fit into a broader approach to border surveillance that has been dubbed “the virtual wall.” Aerial surveillance is used in conjunction with surveillance towers, ground sensors, on the ground agents with night-vision goggles, and more. In contrast to a physical wall, the “virtual wall” does not involve visible fencing, but instead less observable networks of different surveillance capabilities. However, the virtual wall involves Border Patrol agents using invasive surveillance technologies, purchasing surveillance equipment from politically connected defense contractors, and ultimately undermining the privacy and civil liberties of migrants and U.S. citizens in border regions (Coyne & Goodman, 2020a). As we will see in Sect. 4.1, the technologies of the virtual wall, particularly drones, have been used within the interior of the United States rather than just at along the border.

3.2 The rise of BORTAC

The Border Patrol Tactical Unit (BORTAC) was established in 1984 “to serve a civil disturbance function in response to rioting at legacy Immigration and Naturalization Service detention facilities” (Customs & Border Protection, 2014). The agents assigned to BORTAC “received special training in riot control, counterterrorism, and other paramilitary activities similar to the training provided to U.S. marshals and the FBI Special Weapons and Training [sic] (SWAT) teams” (Dunn 1996: 52). The creation of BORTAC meant the establishment of a new organizational structure, as well as training programs that developed human capital among BORTAC agents. These training programs “deliberately emulated the military, specifically special forces” (Coyne & Goodman 2022: 10). Official CBP materials describe how “BORTAC’s Selection and Training Course (BSTC) was designed to mirror aspects of the U.S. Special Operations Forces’ selection courses” (Customs & Border Protection, 2014).

The initial purpose of BORTAC was to respond to riots at INS facilities. However, BORTAC agents and other Border Patrol decisionmakers quickly began to devise new ways to deploy BORTAC. Many of BORTAC’s early missions were aimed at counternarcotics and drug interdiction, especially in the borderlands. “By 1987 BORTAC was taking part in drug enforcement and crop eradication efforts in the United States” (Dunn, 1996: 52). These efforts often involved collaborating with other teams, including teams from outside of the Border Patrol. For instance, the National Guard worked with BORTAC on “clandestine reconnaissance patrolling operations (dubbed Operation Unity)…in Big Bend National Park in the fall of 1988 and spring of 1989 for one week and two weeks, respectively, as a pilot project to assess the feasibility of such operations” (Dunn, 1996: 129). At times, these counternarcotics operations even extended beyond U.S. borders. For instance, in 1987 BORTAC participated in a counternarcotics operation in Latin America, dubbed “Operation Snowcap” (Jacobellis, 2014; Coyne & Goodman 2022: 11).

Operations outside of U.S. borders became a much larger part of BORTAC’s mission in the aftermath of the 9/11 attacks in 2001. At that point, border security became intimately tied with the broader missions of homeland security and counterterrorism, which were explicitly conceptualized in international terms. As one Border Patrol agent explained, “CBP is expanding into foreign countries to be more effective and keep the bad actors away from U.S. soil” (Seiler, 2017: 5). To further these international security objectives, BORTAC “conducts training and operations both in the United States and in other countries in furtherance of the U.S. Border Patrol’s mission” (Customs & Border Protection, 2014). These operations have included participation in nation-building efforts in Iraq, training efforts in Central American countries such as Guatemala, and trainings in African countries such as Kenya, Tanzania, and Uganda (Coyne & Goodman 2022; Miller, 2019b).

In response to riots, bureaucrats created a specialized SWAT team within the Border Patrol. In the decades that ensued, they creatively responded to new opportunities to use this SWAT team, including opportunities for counternarcotics efforts at U.S. borders as well as opportunities to train security forces abroad. This entrepreneurial approach to expanding BORTAC’s mission has continued within the interior of the United States, which will be explored further in Sect. 4.2.

4 Domestic applications of border patrol powers

The Border Patrol has substantial leeway to experiment with state control. They have used this leeway to construct massive surveillance apparatuses that rely on fleets of unmanned aerial vehicles. They have also built a powerful SWAT team that operates both along U.S. borders and outside U.S. borders. These systems of state control were originally justified by concerns about border security. However, bureaucrats within the Border Patrol have continued to discover creative uses for aerial surveillance and for BORTAC, including uses in the interior of the United States with little to no connection to border security.

4.1 Aerial surveillance in the interior of the United States

CBP’s aerial surveillance programs are sold to the public as an important aspect of border security efforts. For instance, CBP’s Frontline Magazine has published multiple articles touting the use of drones as an indispensable tool for border security (Davis, 2020; Frontline Magazine, 2022). However, Border Patrol agents have increasingly discovered creative ways to use drones for domestic surveillance and law enforcement projects with little connection, if any, to border security. They largely do this by lending their drones to other local, state, and federal law enforcement agencies:

CBP often uses its aircraft to assist non-Border Patrol operations. From 2010 to 2012, CBP operators flew 687 missions on behalf of other agencies. From 2013 to 2016, only about half of CBP drone flight hours were actually in support of Border Patrol. Furthermore, CBP reports that 20 percent of all Predator B flights were not in coastal or border areas. The cooperation between CBP and other federal agencies means that Americans living near the border aren’t the only ones who risk having their privacy infringed by CBP drones. (Bier & Feeney, 2018: 4)

By collaborating with other federal agencies, CBP can undermine privacy far from the border. This is especially concerning because aerial surveillance faces weaker Fourth Amendment requirements than many other types of surveillance. “No law requires CBP to obtain a warrant before using drones for surveillance, and the U.S. Supreme Court has upheld warrantless aerial surveillance in three cases” (Bier & Feeney, 2018: 3).

In addition to collaborating with federal agencies, CBP frequently collaborates with state and local law enforcement agencies and lends these agencies drones for this purpose.

State and local agencies also often request CBP drone assistance for routine law enforcement matters. From 2013 to 2016, CBP drones flew 416 flight hours for state and local police, often failing to record which police departments were requesting CBP drone assistance. The first reported case of a drone aiding a domestic arrest occurred in 2011 when CBP deployed a drone without a warrant in North Dakota to determine whether suspects that local police were seeking to arrest were armed. A district judge rejected one of the suspect’s motion to suppress the warrantless drone surveillance, writing that ‘there was no improper use of an unmanned aerial vehicle’. (Bier & Feeney, 2018: 4)

CBP’s collaborations with state and local law enforcement raise additional concerns about warrantless surveillance. While warrant requirements for CBP are weak, some states and localities have passed warrant requirements for aerial surveillance by state and local law enforcement. It is unclear “whether CBP adheres to state warrant requirements when assisting local law enforcement agencies” (Bier & Feeney, 2018: 4). This raises concerns that “CBP drones could be a loophole allowing local police to evade local democratic scrutiny of drone use or the requirement for a search warrant. Unfortunately, CBP fails to release sufficient data on local support operations to identify exactly how they are used” (Bier & Feeney, 2018: 4).

Drawing on an official DHS statement from 2015, Bier and Feeney (2018: 3) write that “Predator B drones are supposed to avoid urban areas, and CBP states that it does not deploy them to monitor protests and other activities protected by the First Amendment.” However, since then there have been several documented incidents of CBP drones being used for surveillance purposes against protesters. This includes multiple instances, discussed in the introduction to this paper, in which CBP drones were used to monitor Black Lives Matter protests in urban areas (for more details on these incidents, see Jones, 2023: 227–228). This clearly violates the 2015 DHS guidance in multiple ways.

In addition to the surveillance of Black Lives Matter protests in 2020, CBP drones have been lent to various law enforcement agencies that used them to monitor environmentalist activists and indigenous rights activists. In 2016, a CBP drone was used to monitor the Standing Rock Sioux as they protested the construction of the Dakota Access Pipeline. In 2020, CBP drones were used to surveil the planned route of a pipeline being constructed by Enbridge, as well as the homes of Indigenous activists who opposed the pipeline (Jones, 2023: 229).

CBP’s use of drones was initially justified using the rhetoric of the war on terror, and quickly turned towards more ordinary border security efforts aimed at drug interdiction and catching unauthorized immigrants (Coyne & Hall, 2018). However, Border Patrol agents acted creatively to satisfy the desires of political officials for surveillance, often engaging in favors and exchange with bureaucrats in other government agencies by lending officials in those agencies surveillance drones. The result has been a substantial expansion of aerial surveillance in the interior of the United States. This surveillance has been largely unbound by warrant law and has been used to surveil protesters and political activists.

4.2 BORTAC turns its efforts inward

Similar mission creep and activity in the domestic interior has been observed with BORTAC. Members of BORTAC can participate in task forces that enable them to work with other agencies on a wide variety of law enforcement efforts. Jenn Budd, a former Border Patrol agent, argues that many Border Patrol agents prefer this type of policing to their traditional roles in border security. “They don’t want anything to do with immigration anymore,” Budd argues. “That is why they have the BORTAC units. … They intend to be a national police force. That’s why you are seeing them do actual police work in the joint task forces” (Budd, quoted in Jones, 2023: 231).

From 2019 to 2021, Rodney Scott was the chief of the Border Patrol. Jenn Budd alleges that Scott was a major advocate for expanding Border Patrol’s role to cover a wide range of policing and intelligence tasks, and that advocacy began before he was appointed to his high-ranking position in 2019. Budd argues that Scott “really pushed the Border Patrol to go all around the world when he was working in Washington D.C. from 2005 to 2007. He was telling people, if a car bomb goes off, I should know about it. So there is a creep, whether it is a creep into regular policing, or it’s a creep into intelligence. The management of the agency is just gung-ho to get their fingers into everything” (quoted in Jones, 2023: 231).

As president, Donald J. Trump supported an expanded role for BORTAC in policing the American interior. As Jones (2023: 232) explains:

While he was in office, Trump expanded the authority of the Border Patrol and directed its use in the interior of the United States. In February 2020, Trump authorized using BORTAC units on missions to support ICE operations to ‘flood the streets’ with agents in sanctuary cities. As protests grew after the killing of George Floyd, BORTAC units were deployed to cities around the United States to protect federal buildings and monuments.

While sending BORTAC agents to sanctuary cities relates to their immigration enforcement mission, it still entails deploying a team developed for use at the border into the interior. This means that individuals far from the borders that BORTAC was developed to police are subject to BORTAC’s militarized policing. BORTAC’s policing of protests, on the other hand, has no clear connection to border security.

How did BORTAC legally justify policing protests in major cities in the interior of the U.S.? BORTAC’s operations during these protests “were not based on the sections of the U.S. code that traditionally govern the Border Patrol, but rather on little-known and previously unused sections of the U.S. code that give law enforcement authority to the secretary of the Department of Homeland Security. Section 40 USC § 1315 was a post-September 11 provision that allowed the secretary to protect government property” (Jones, 2023: 233–234). As an attorney from the ACLU explained, “These other statutes have a lot of very permissive language, allowing DHS entities like CBP to act like a federal police force” (Ebadolahi, quoted in Jones, 2023: 234). President Trump was able to activate these statutory powers using an executive order (Jones, 2023: 234). BORTAC used these powers to conduct a variety of operations near protests, including operations in Portland that involved grabbing individuals and dragging them into vans without first identifying themselves as officers (Jones, 2023).

The expansion of BORTAC’s role over time illustrates the creative ways that bureaucrats and politicians use the resources at their disposal, often expanding their authority far beyond the purposes it was initially granted to address.

5 Conclusion

The expansion of Border Patrol’s powers over time, and the use of those powers far from the border, illustrates more general lessons about state power. Powers granted at time t1 for a particular purpose in a particular place may be used in unexpected ways and in unexpected places at time t2 or t3. This is partially because novel circumstances will arise that were unanticipated at time t1. It is also because bureaucrats and politicians are creative, entrepreneurial actors who seek ways to advance their ends using the tools at their disposal. Similar processes of mission creep and expanding power occur in other parts of government as well. Border security is simply particularly noteworthy because prevailing ideological presuppositions about the border create a wider scope for experimentation with state control at the border. Moreover, proponents of border policing tend to see their preferred policies as primarily impacting foreigners. Recognizing the expansion of border policing into the American interior helps correct this misconception.

The growth of CBP into a national police force raises a variety of important issues. Local level constitutional and legal constraints on police powers may not apply to the Border Patrol, which means the Border Patrol’s presence in traditionally local law enforcement can create loopholes that empower police to evade these constraints. Moreover, a substantial theoretical and empirical literature in public choice and institutional analysis (see Boettke et al., 2016) finds advantages to relatively polycentric local policing, which can be attributed to Tiebout competition, the use of local knowledge, and opportunities for coproduction. These advantages are undermined by the rise of a national police force. Furthermore, when the Border Patrol acts as a national police force, then federal politicians and bureaucrats have a greater scope of authority to impose their views, which undermines opportunities for pluralism that accommodates value heterogeneity. Bureaucratic entrepreneurs have seized opportunities that expand the power of the central government. In the process, they have undermined the institutional constraints that might otherwise guide political entrepreneurs to discover value-added ways to serve citizens.

The entrepreneurial process that has occurred within the Border Patrol therefore has broader implications for constitutional political economy. Even when formal constitutional constraints are designed to maintain competitive federalism, bureaucrats within the central government will have opportunities to use their powers in new ways that intersect with subnational concerns. As they seize these opportunities, they centralize power and authority, thereby undermining the competitive mechanisms intended to maintain limited government. To properly analyze the potential for institutions to limit government, static analysis is insufficient. We must take the dynamic aspects of novelty and entrepreneurship seriously to properly understand constitutional political economy.

As previously noted, this paper builds on the existing secondary literature on border militarization rather than engaging in novel empirical analysis. Future research could do more empirical work to examine the processes discussed here and test the explanations this paper offers. Qualitative work such as fieldwork and interviews may help illuminate more details about the institutions, beliefs, and practices of Border Patrol agents, other law enforcement officers they work with, and people living and migrating in the borderlands. Survey evidence could be gathered to discern public approval and disapproval of the use of tools such as aerial surveillance in border regions and the American interior. Public records could be used as a dataset to examine the use of surveillance drones, perhaps developing more precise measures of what proportion of aerial surveillance can be attributed to CBP drone lending. Researchers could examine the congressional record to better understand how far the practices discussed in this paper deviate from initial legislative intent. Future research could use either quantitative econometrics or qualitative process tracing to tease out which changes in state and local police practices can be attributed to border militarization and which are better attributed to other factors such as the 1033 Program and citizen-voter demands. More detailed data analysis could also be carried out using records from sources such as the Transactional Records Access Clearinghouse (TRAC), which maintains databases gathered from FOIA requests regarding various aspects of the U.S. immigration system, including Border Patrol activities. In addition, comparative work across jurisdictions could be used to further assess the claims made in this paper and examine the extent to which they generalize. While multidisciplinary research has taught us a great deal about border militarization and its consequences for domestic institutions, there is still much to learn. Going forward, multiple methods can help scholars uncover more about these pressing issues (Poteete et al., 2010).