Controlling the Borders

Broad public concern [1] about the flow of illegal immigrants into the United States has fueled demands in Congress for the approximate doubling in the size of the Border Patrol (BP), from about 11,000 to 21,000 agents [2].

Although the BP has grown substantially in the last two decades, the proposed addition of 10,000 new agents in the next five years would be unprecedented. It also would make the BP the largest federal law enforcement agency in the country -- much bigger than the FBI and with over four times more agents than the Drug Enforcement Agency, over seven times more agents than the IRS and almost nine times the number working for the Bureau of Alcohol, Tobacco, Firearms and Explosives. See table below.

Agency Number of Agents
FBI 12,214
Border Patrol 11,106
proposed 21,000
DEA 4,925
IRS 2,829
ATF 2,401

Such an enhancement would also be expensive. At its present size, the current annual cost of the BP is about $1.7 billion. This would more or less double. The overall cost, however, would not be limited only to the agency. If, for example, the new "get tough" policy was coupled with a decision to fully enforce only the laws that are now on the books for the current number of BP apprehensions, there would necessarily be a dramatic rise in the total workload being handled by the federal criminal court system. At that point, also, the makeup of workload would would dramatically shift, with roughly 10 immigration cases for every current non-immigration prosecution. Such a volume clearly would require a substantial increase in funding for the additional prosecutors, judges, detention beds and support staff needed to process them.

Given the dimensions of many of the proposals now under consideration -- the House has passed new legislation and a Senate Committee is currently holding hearings -- the Transactional Records Access Clearinghouse (TRAC) has undertaken a special analysis of government data examining how changes in BP staff in the past have affected its principal job: the apprehension of illegal immigrants.

Curious Outcomes

For all enforcement agencies -- federal, state or local -- the common sense rule-of-thumb is that more agents or more police officers result in more official actions -- more inspections, more arrests, more apprehensions or whatever. (This is thought to be true because the number of offenses -- whether it is people taking illegal drugs or aliens sneaking across the border -- typically far exceeds the capacity of law enforcement ever to handle them all.)

Figure 1. Click for larger image

This theory , however, does not appear to hold true for the BP on either a short or long term basis. From FY 1995 to FY 2005, for example, beginning in the Clinton Administration, the number of full-time BP employees more than doubled, jumping from 4,876 to 11,106. But for the whole period, BP apprehensions went the other way, dropping by over 10 percent from 1,324,202 in 1995 to 1,188,977 in the most recent available year. See Figure 1.

An examination of BP records for a much longer time period reinforces the doubts about the direct link between agents and apprehensions. Looked at from 1947, for example, the data show three peaks in the overall number apprehensions -- in FY 1954, FY 1986 and FY 2000.

Surprisingly, the unusually large number of apprehensions the BP reports making in both 1986 and 2000 -- about 1.6 million of them -- were considerably higher than the 1.2 million recorded in 2004 and 2005, even though the BP now has a lot bigger staff. See Figure 2 and supporting table.

Figure 2. Click for larger image

Because of the spotty availability of figures for border patrol agents -- previously called immigration patrol inspectors -- prior to FY 1975, it is not possible to determine how the number of agents may have changed during the fifties. But contemporary government reports of that period do not mention change in staffing numbers as a cause or consequence of the peak in apprehensions observed in 1954.

Making apprehensions is the primary job of the BP and thus its key performance measure. But once an apprehension is made, a number of various follow up options are possible such as voluntary departure, removal, expedited removal and criminal prosecution. These options are discussed below.

A variety of different political, economic, and social forces -- some of them difficult to spot, some of them impossible to measure -- have contributed to the highly erratic nature of the year-by-year apprehension counts during the last 58 years. From TRAC's knowledge of many other agencies, for example, one of the invisible factors that often affects such counts is the unannounced and generally unnoticed changes in the agency's internal book-keeping rules.

In the case of the Border Patrol, it is assumed that another factor behind the agency's seemingly erratic apprehension counts is the rise and fall in the total number of individuals attempting to slip across the borders of the United States. But precisely estimating this illegal flow is difficult. And documenting the cause and effect relationship between the number of attempted entries and apprehensions is a real challenge. That said, however, it would seem likely that shifts in the economic and political conditions in both the United States and other countries around the world would alter the intensity of interest in coming to this country.

The BP's Many Explanations

For this and many other reasons, the dramatic shifts in the number apprehensions -- sometimes up and sometimes down -- are hard to fathom, even though the annual editions of the BP yearbook have seldom hesitated to offer a wide range of the agency's sometimes conflicting theories.

In the period from 1954 to 1956, for example, there was an astonishing fifteen-fold collapse in recorded apprehensions. The size of this decline was indeed puzzling. But the FY 1955 report of the BP, then published by the now-defunct Immigration and Naturalization Service, did not hesitate to weigh in. The precipitous drop in apprehensions, the report claimed, was the result of a series of vaguely defined enforcement initiatives including "special mobile force units" and "mopping up operations in the interior."

The central result of the agency's new tactics, the 1955 report said, was that "the so-called 'wetback' problem no longer exists." Put another way, the BP analysis continued, the decline of illegal aliens entering the country from Mexico proved that they no longer were a concern. The report concluded with a sweeping judgment that in hindsight would seem somewhat optimistic: The border, it claimed, "has been secured."

By 1980 the number of apprehensions was close to the 1954 peak and the BP was making no grand claims about how its new tactics had secured the border. Instead it pointed out that Border Patrol agents had been diverted from their normal duties because of that year's crisis in Iran and the influx of Cubans and Haitians who were seeking entrance to the United States. And as the years went by, the war on drugs and fighting terrorism also were offered as factors reducing the number of agents available for the primary mission: controlling the border.

The 2003 Yearbook, now published by Office of Immigration Statistics of the Department of Homeland Security rather than the INS, discussed the three-year decline in the annual apprehension counts after the 1986 peak. The report suggested that this time the decline was related to changes in the desire of aliens to come to the United States as a result of the passage of the Immigration and Control Act (IRCA) of 1986 and a substantial increase in the size of the Border Patrol also mandated by the law. It should be noted that both of these conditions were still in effect in the 1989-to-1999 period when apprehensions again reversed direction, generally moving up to yet another peak in 2000.

Map 1. Click for larger image

Changes in Apprehensions at the Sector Level

The nation is divided into Border Patrol sectors. See map [3]. A sector-by-sector comparison of BP staffing and BP apprehensions shows a highly inconsistent pattern, thus further undermining the widely accepted assumption that border security will be automatically improved by the hiring of more agents. While numerous special factors -- a new fence, new lights, new electronic sensing device, etc. -- may sometimes be involved, the inconsistent data at the sector level is puzzling.

Southern Border

The U.S. border with Mexico has been a key focus of the government's immigration control efforts for many decades. Figure 3 and the accompanying tables display data on apprehensions and staffing from FY 1992 through 2005 on the eight most active BP sectors along the southern border. An accompanying table gives the underlying figures for all nine southern border sectors.

Driven by a variety of political pressures, this border to this day soaks up an overwhelming majority of all BP staff. But the question of how well or poorly the agents were spread was rarely debated.

Figure 3. Click for larger image

  • During the last five years in the Rio Grande sector in Texas (also known as McAllen), for example, staffing levels went up and down but in the end were essentially unchanged. The number of apprehensions, however, climbed by about a quarter -- from 107,844 to 134,185.

  • In same period in Tucson, the Arizona sector where the political debate about immigration policy has been unusually intense, the number of BP agents climbed substantially, from 1,686 to 2,220. Despite the increased number of agents, however, the apprehensions they reported -- while showing considerable year-by-year variation -- remained essentially unchanged: 449,567 in 2001, 438,997 in 2005.

  • In San Diego, the BP California sector with a significant border barrier, staffing declined sharply from 2,014 to 1,600 full time employees but apprehensions increased from 110,075 to 126,913.

Northern Border

There is no question that the enforcement activities along the northern border have become a matter of concern. In July of 2004, for example, the 9/11 Commission criticized the government for what it had achieved since the attacks even though the absolute number of agents had been greatly expanded, almost tripling since 9/11. See TRAC Report on BP staffing, including this table.

More recently a Congressional Research Service report in May 2005, echoed these concerns although it said the Border Patrol hoped to improve security in this area by better cooperation with Canadian authorities and enhanced intelligence through the deployment of cameras and remote sensors. The data, however, show that when it comes to apprehensions -- down from 12,305 to 9,959 in the last five years -- the increased staff was not paying off. See Figure 4 and accompanying tables.

Figure 4. Click for larger image

  • In the Buffalo sector in New York , staff more than tripled, 35 to 128, but apprehensions nose dived from 1,434 in 2001 to 400 in 2005.

  • Apprehensions in the BP's Detroit sector declined modestly from 2,106 to 1,912, although full-time employees had jumped from 38 to 135.

  • On other side of the United States in Blaine, the BP sector in Washington state, the same general pattern emerged with staff jumping from 49 to 135 but apprehensions sliding from 2,089 to 998.

The point here is not to suggest that the BP is unnecessary or that the United States should give up on efforts to control its borders. Rather, it is to recommend that the administration, congress and others carefully examine the widely accepted assumption that more agents will automatically result in more effective border control.


The impact of any law enforcement agency, most experts agree, is not limited only to the raw number of law breakers they apprehend. A second factor is deterrence: the manner in which the presence and effectiveness of the officers on a city street or at the nation's borders persuades potential offenders to avoid situations where they may be caught.

In the case of the BP, the 1994 National Strategy followed for over a decade was called a "prevention through deterrence" strategy. The FY 2003 Yearbook's definition of the then current strategy was quite general and sounded like a continuation of strategy from previous years: the deployment of "Border Patrol agents along the border to prevent and deter illegal entry, rather than apprehending undocumented immigrants after they have entered the United States."

The language of the current National Border Patrol Strategy published in September 2004 now seeks "operational control of our nation's border" as its mission, with the aim of "preventing terrorists, ... illegal aliens, smugglers ... from entering the United States." While the language has shifted somewhat in this post 9/11 world, the government describes this new strategy as "build[ing] on the successes of the previous 1994 'prevention through deterrence' strategy."

In the same way that the assumed link between staff size and effective enforcement requires skeptical examination, so do the BP's deterrence claims.

One serious problem is that few Americans have a realistic understanding of the actual working force that any police agency can muster at any one time. At first glance, for example, the BP -- with 11,000 agents -- sounds quite formidable. But when it is recalled that the United States has approximately 7,500 miles of remote and often rugged land border plus extensive additional miles of coastal / Caribbean borders patroled by the BP, expectations about the possible impact of the agency must be considerably scaled back.

The next fact to be considered is that the BP is not just another Monday-to-Friday nine-to-five business, that its agents must conduct their operations for three shifts a day and seven days a week. And not all of a 40-hour week would normally be spent on actual patrol as ancillary duties from meetings, completing reports, and the transport and processing apprehended aliens necessarily take time. A third fact to consider are the vacation, training and sick days that must be subtracted from the agency's available work days and the supervisors who typically are office bound and not on actual patrol. Finally, because of the basic dangers involved, BP patrol actions frequently require two or more officers rather than one.

The bottom line: TRAC's analysis of the numbers suggests that the Border Patrol currently would be hard pressed to have more than perhaps an average of 2,000 agents on duty at any given time, perhaps one agent for every four miles of the nation's total border if they were all working solo. Although the construction of more fences and the use of more remote surveillance devices may somewhat offset the impact of these realities, the actual number of agents that the BP can muster at any one time must raise some doubts about the results that will be achieved by doubling its size.

What Happens After Apprehension?

In considering the deterrence potential of any enforcement agency, a second component that requires reflection is the question of what happens when the individual is arrested by the police officer or the alien is apprehended by the agent. If an arrest or apprehension is probable and the resulting outcome significant, the deterrent effect will be multiplied. But if capture is not certain and the resulting outcome relatively mild, the promise of deterrence is reduced.

For those apprehended illegally crossing the border, the government and the alien have several options that can be followed: criminal prosecution, "voluntary departure," an asylum claim, or an order of removal. Each of these paths results in significantly different consequences for the apprehended alien as discussed below.

Criminal Prosecution

Illegal entry is in itself a federal crime punishable as a "petty" misdemeanor with a jail term of up to six months under Title 8 Section 1325 of the United States Code. In actual practice, those convicted usually receive no prison time or are sentenced to "time served" -- usually the few days it takes between being picked up by the BP and being tried and sentenced before a judge in a U.S. Magistrate Court. Deportation presumably usually follows conviction, although this is an entirely separate administrative process not under the jurisdiction of federal district courts.

Conviction also means that individual aliens, if they again try to illegally re-enter the United States, are subject to more severe penalties including significant prison terms. For example, conviction under Title 8, Section 1326 for re-entry resulted in a median sentence of two years in FY 2004.

Few illegal aliens apprehended by the BP are actually prosecuted. During fiscal year 2004, only 17,100 individuals were reported as convicted as a result of Customs and Border Protection (CBP) enforcement activity. Compared to the 1,160,395 apprehensions, this represents less than 1.5%. And since this number includes not only apprehensions by the Border Patrol but aliens detected of fraud or false statements by CBP Inspectors at ports of entry, even this small percentage is likely to be an overstatement. [Total immigration prosecutions are higher because other agencies outside of CBP are charged with criminal investigation of immigration violators. See August 24, 2005 TRAC Report.]

Voluntary Departure

Available records show that about nine out of every 10 of the individuals who were apprehended are given "voluntary departures." During FY 2004, the latest year for which figures are available, the government reports that there were 1,035,477 "voluntary departures" and more than 99 percent of these were aliens apprehended by the Border Patrol. Figure 5 shows graphically how the number of apprehensions compare with the volume of criminal convictions and voluntary departures.

Figure 5. Click for larger image

Under a voluntary departure, aliens who admit to being illegally in the country and agree to depart voluntarily receive no penalty of any kind. Particularly for Mexicans, this procedure can be quite quick. In 2005 the Congressional Research Service reported that the processing of Mexicans took only 10-15 minutes. Aliens are then typically put on a bus or van and driven back across the border, or flown back to their country of origin. There are no legal restrictions imposed on their freedom to enter the United States in the future.

Indeed, the number of apprehensions is a count of events, and thus recidivists who try to re-enter more than once and get caught are counted as an apprehension each time. According to accounts published in the press, as confirmed by TRAC conversations with Border Patrol offices along the southern border, there are frequently formal or informal policies requiring an illegal alien to be apprehended five to ten times before the BP will seek to have the individual criminally prosecuted for illegal entry.

Remaining Options: Removal, Expedited Removal and Asylum

The government reports that current data systems do not allow it to follow apprehended aliens through to the final disposition of their cases. Thus details of what happens to apprehended aliens -- particularly those who are neither criminally prosecuted nor given voluntary departures -- are unclear.

The border patrol can refer illegal aliens they apprehend to the Immigration and Customs Enforcement (ICE) branch of the Department of Homeland Security for a deportation (removal) proceeding in immigration court. If there is a prior formal removal order, that order can simply be reinstated. (The immigration court is an administrative court which is part of the Department of Justice.) Once the BP does the initial paperwork, ICE has responsibility for detaining the individual and handling the proceeding before the immigration court.

No data are currently available to track how often BP apprehensions end up in a removal proceeding in immigration court or how many actually get removed as a result. While there were 323,749 removal proceedings received by the immigration court during fiscal year 2005, most of these appear to be for aliens picked up by other law enforcement agencies and not by the BP. Also, many are released prior to their court hearings due to lack of detention space, and may fail to show up for their scheduled hearings. A total of 222,360 removal orders were issued by the immigration court during FY 2005 according to court statistics. Thus being placed in a removal proceeding does not automatically mean that the individual ends up actually being removed from the U.S.

The Congressional Research Service reports that, due to lack of detention space, release into the interior of the U.S. after referral to the immigration court for removal is the default choice of the BP for those non-Mexicans who have no criminal history and pose no threat to national security. In addition to deportation, a removal order also generally bars an alien from legal re-entry to the United States for 10 years. "Aggravated felons" are barred for life.

Recently the BP has been given general authority to issue removal orders without going through the immigration court. This procedure -- called "expedited removal" -- results in deportation and prevents an alien from legally re-entering the United States for five years (20 years for repeat offenders). While this procedure has been on the books since 1996, until 2005 expedited removal as a general rule was applied only to persons arriving at ports of entry.

It is unclear how often this procedure will be used by the BP. According to the latest government figures for detention and removal operations, in FY 2004 there were 41,935 expedited removals. This number had grown to 71,875 in FY 2005. These figures, however, cover all sources and are not limited to those originating from the BP. Thus it is difficult to know how much of the increase in FY 2005 was the result of the BP's new authority.

Those apprehended also can ask for a hearing requesting asylum before an immigration judge. The number of successful asylum seekers among BP apprehensions is not reported. However, in FY 2005 the total number of individuals from all sources granted asylum in defensive proceedings, even only conditionally, was only 3,009. Since most apprehensions are Mexicans, while only 34 of the individuals receiving asylum are from Mexico, it appears likely that only a small proportion of the individuals granted asylum originally turn up in the United States as BP apprehensions.

Last but not least some aliens apprehended by the Border Patrol may be released because of a temporary lack of detention beds and "disappear". The BP does not make available any statistics on how often this occurs.

Where Does the United States Stand Today?

More than ten years ago, Congress passed a law requiring each agency to develop processes to plan for and measure mission performance. The Customs and Border Patrol, as a new agency established with the Department of Homeland Security, published its second report looking at achievements in FY 2005.

In this remarkable report published in November of 2005, the CBP outlined a view of the border control situation that starkly contrasted with the government's repeated claims about what had been achieved in years past. One section of the report set out a series of strategic goals and the "performance measure" that would be used to determine whether each goal had actually been achieved. The border patrol's key strategic goal was "preventing terrorism between ports of entry" and the key performance measure was "border miles under operational control." [4]

One CBP official acknowledged that the agency's definition of under operational control "is very stringent." To be included as meeting this standard, the border mile must have sufficient agents, technical equipment and infrastructure to detect the entry of an alien, analyze the threat, respond to the entry and subsequently "apprehend the alien at or as close to the border as possible."

During FY 2005, the CBP's performance and accountability report said, a total of 288 miles of the nation's 8,000 mile border had been brought up to this standard.

"One of the major factors leading to this success is the continuance of the strategy to gain operational control in urban areas along the southwestern border," the report said. The decision to begin in the urban areas -- San Diego, El Paso, and the Nogales region of Arizona -- was dictated by the judgment that in such areas an illegal entry could be completed so quickly and merging into the population was relatively easy. In the more rural areas, the agents have more time to intercept aliens and it is more difficult for them to mingle with the residents.


A plan entitled the Secure Border Initiative was announced on November 2, 2005. According to the press release it was designed to "enable DHS to achieve operational control of both the northern and southern border within five years." Less than two weeks later, however, the CBP's Performance and Accountability report stated that only 288 miles are now under operational control. With this apparently candid assessment, it would seem the agency has a lot of work to do if it is to bring all of the northern and southern borders up to this standard in only five years.

Is achieving this goal realistic? Assuming Congress can find the money, would hiring 10,000 additional agents do it? From the historic record described here, it seems clear that the complex issue of border patrol requires a lot more careful analysis than it so far has received.

[1] A national survey conducted by the Gallup Poll in January 2006 has found that 57% of the respondents said they would like to see a decrease in the flow of immigrants into the United States. According to the poll, the only time in the last few years that a slightly larger proportion of American felt this way was shortly after the events of 9/11.

A second Gallup poll, completed last November, found that two out of three Americans disapproved of the way the Bush Administration was handling immigration matters.

[2] P.L. 108-458, The Intelligence Reform and Terrorism Prevention Act of 2004, was signed into law by President Bush on December 17, 2004. It authorizes adding 2,000 Border Patrol agents per year for five years. Part of the current heated debate in Congress concerns funding for those positions.

[3] There are currently 20 sectors. Livermore sector although still shown on some BP's maps was closed July 30, 2004.

[4] Lanugage referring to "illegal entrants" has undergone numerous changes over the years. In the 1955 INS Annual Report, the term "wetbacks" was used for those illegally crossing the Mexican border. For many years "deportable aliens" was the phrase of choice. Since 9/11 "terrorists" appears in official documents to refer to all "illegal entrants" whatever their purpose — those seeking work, engaging in cross-border smuggling or other forms of criminal behavior, in addition to terrorists per se.