I. Allegations of Falsification of Apprehension Data

One of Bonner's central contentions was that the apprehension figures reported by INS and the Border Patrol understate the number of illegal aliens apprehended in the San Diego Sector. This contention was echoed by No. 1922, who told the Congressional subcommittee that the apprehension figures released by INS were "phony." Apprehension statistics have been the primary measure of success for Operation Gatekeeper, and the basis for justifying its continuance and expansion. Thus, if true, these allegations call into question the wisdom of Operation Gatekeeper and the future of similar programs along the border.

Because these allegations went to the heart of the public claims regarding Gatekeeper's success, the OIG interviewed a broad range of witnesses at every level at each of the stations, all of the personnel in the Sector Intelligence unit, witnesses at every level of the Sector hierarchy, Western Region Intelligence personnel, INS Intelligence personnel, and others throughout DOJ who regularly receive apprehension statistics. In total, the OIG interviewed 307 persons regarding the apprehension statistics and the possibility that they were falsified, 78 of whom actually prepared reports containing apprehension statistics. The OIG also reviewed hundreds of Border Patrol, INS, and DOJ reports containing apprehension statistics, and examined reports from each level of these organizations for inconsistencies. The OIG also looked for patterns of data that were either consistent or inconsistent with the allegations.

We begin by explaining the significance of the apprehension data and the process for gathering it, and then address the specific allegations regarding this data.

A. The significance of apprehension data

If it were possible to calculate how many persons attempt to enter the country illegally, then the percentage of these individuals who are apprehended could also be calculated. This in turn would permit an accurate evaluation of any given Border Patrol operation. Because of the length and varying topography of the border, however, it is impossible to constantly monitor it or to know with certainty how many people attempt illegal entry. As a result, the only concrete measure of illegal alien traffic is the number of apprehensions, and the primary measure of Operation Gatekeeper's success has been apprehension figures.

Apprehension figures without corresponding entry figures reveal little, however. Apprehension data alone cannot distinguish between changes in the volume of illegal alien traffic from changes in enforcement strategies that increase or decrease the percentage of crossers apprehended. Thus, if apprehension figures increase, this may be due to more effective operations, to an increase in illegal alien traffic, or a combination of both. If apprehension figures decrease, this may be due to less effective operations, to a decrease in illegal alien traffic, or both.

The aim of Operation Gatekeeper was to intercept a sufficiently high percentage of illegal alien traffic so that the aliens are eventually deterred from even attempting entry in the area of patrol. INS believed that if effectiveness rates (the percentage of illegal traffic that is apprehended) reached approximately 85 percent, illegal traffic would decrease because the chance of successful entry was so low.

Assuming that traffic levels remain constant, increased effectiveness will result in higher apprehension levels. Becoming so effective that illegal traffic is deterred from entering, however, does not necessarily result in higher or lower apprehension levels. To keep apprehension levels from rising as effectiveness increases, the number of entrants must decrease. For apprehensions to decrease while effectiveness is increasing, the number of entrants must decrease at a rate faster than the increase in effectiveness. Thus, decreasing apprehensions while being highly effective requires achieving high levels of deterrence.

When the Border Patrol instituted Hold the Line in El Paso a year prior to Operation Gatekeeper, apprehensions of illegal aliens in the El Paso Sector dropped 70 percent in the first year of operation. Because most of the illegal crossers in El Paso prior to Hold the Line crossed during daylight hours and remained in the immediate El Paso area, it was obvious that the decrease in apprehensions was due to dramatic decreases in illegal traffic. Although INS and the Border Patrol argued that San Diego and El Paso were different, the El Paso results became the de facto yardstick for measuring the success of future operations. Achieving similar results in San Diego was made even more difficult by the December 1994 devaluation of the Mexican peso, which increased the pressure for northward migration. During Congressional hearings on Operation Gatekeeper in March 1995, however, the subcommittee extracted a promise from Commissioner Meissner that apprehensions in San Diego would fall 70 percent in the coming year, as they had in El Paso.63  Congress made clear that if INS failed to meet this goal, it would pass legislation mandating different strategic deployments in the San Diego Sector.64   This threat placed tremendous pressure on INS and the Sector to fulfill Commissioner Meissner's promise.

Bonner alleged that the Border Patrol could not fulfill that promise and instead chose to falsify apprehension figures. He alleged that this was accomplished in several ways: (1) supervisors altered apprehension reports to reduce the number of recorded apprehensions; (2) apprehension records were destroyed or "mysteriously lost;" (3) apprehended aliens were returned to Mexico without being processed so that there would be no record of their apprehension; and (4) information regarding the location of apprehensions was falsified to make it appear as if the aliens had been apprehended far east of Imperial Beach. These alleged actions, if undertaken, would have materially reduced the apprehension numbers for Imperial Beach, and made it appear that Operation Gatekeeper was working when it was not.

B. The collection and reporting of apprehension data in the San Diego Sector

The publicly reported apprehension data for the San Diego Sector is the result of a long process of data collection. This process begins with the apprehension of an illegal alien, when two INS forms are completed: the I-826 (Notice of Rights) and the I-827 (Request for Disposition).65   These forms are used to record the name of the apprehended individual, the date and time of the apprehension, the location of the apprehension, and the identity of the apprehending agent. Although the process varies from station to station and from time to time, these forms are generally completed in the field by the apprehending agent. The forms are then handed to the agent responsible for transporting the aliens from the field to the station. At the station these forms are handed to the desk officer. The desk officer keeps a tally of these forms and the aliens brought into the station during a shift. This tally records the total number of aliens apprehended, their sex, whether they are juveniles, whether they are Mexican or of another nationality, and other details.

At the end of each shift these tallies are given to the shift intelligence officer, who uses the information in preparing a shift report. The shift intelligence officer also gathers scope reports, drag reports, and information from other sources to complete his report. A midnight shift supervisor or senior agent collects the information from the three shift reports to create a station supplemental report. The consolidated reports are then forwarded to Sector Intelligence.

Sector Intelligence retypes the information contained in the reports on forms it uses to disseminate information outside the Sector. If there are mathematical errors on the reports, Sector Intelligence personnel contact the applicable station to obtain corrected information. This information is then sent in various formats to Western Region and INS headquarters on a daily basis. There the information is collected for Intelligence unit personnel who enter it into databases. The data is then available for public dissemination by INS.

This same process is used for weekly and monthly reports, with the respective stations tallying their data and forwarding it to Sector Intelligence. This information is then retyped into different reports and forwarded outside the Sector.

While this process governs official apprehension counts, other records also contain apprehension information. The IDENT system, which records the fingerprints and photograph of each apprehended adult, contains a count of persons processed at the Station for any given date and period of time. While IDENT figures are not used for the official apprehension count because of reliability concerns, they provide a basis for comparison with the official count.66   Moreover, because the date and time an individual is processed is automatically recorded by the IDENT system - and there is no means for agents at the station to alter these records once they are made - information in the IDENT system cannot be manipulated subsequently by station personnel.

Transport figures provide a second mechanism to check official apprehension statistics. Apprehended aliens are transported from the stations to the ports of entry by Detention Enforcement Officers (DEOs).67 The DEOs prepare logs of how many individuals they transport, whether to a port of entry or a facility for aliens who are not voluntarily returned (VR'd). Because some aliens, such as the intoxicated and those needing medical attention, may be transported by station personnel instead of the Alien Detention and Removal Branch (ADR), ADR transport figures may be slightly lower than reported figures. They should not, however, be higher than reported figures.68

Certain station reports also record a large percentage of apprehensions on a given shift and thus provide a third check for official apprehension numbers. Scope reports record the number of apprehensions from scope operator observations during the swing (3 p.m. to 11 p.m.) and midnight (11 p.m. to 7 a.m.) shifts. Most apprehensions on these shifts will appear on these reports and most apprehensions occur during these two shifts.69   Accordingly, reported apprehensions for the swing and midnight shifts should be at least as high and probably higher than apprehensions recorded on the scope reports. Desk officers also record the number of persons in a cellblock during a shift, and their reports should also reflect most if not all of the persons apprehended at a particular station during a shift.

The significance of the IDENT, transportation, and scope reports here is that if official apprehension figures are fabricated, all of these reports must also be altered or there will be an obvious trail of contradictory reports. Moreover, because each of these reports is prepared by a different person, all of the responsible personnel must agree to fabricate numbers that will correlate.

C. Allegations that apprehension figures have been altered

In the initial June 1996 North County Times article, Bonner and Dassaro were cited as alleging that agents were being ordered to report fewer apprehensions than actually took place. In the July 1996 Harper's article, an unidentified agent alleged that if 500 aliens were apprehended, supervisors reported only 200 apprehensions. And No. 1922 testified before the Congressional subcommittee in August 1996 that he was asked "to cheat" on the numbers he reported.

Despite conducting 307 interviews of relevant personnel, we found no one who claimed to have first-hand knowledge that any apprehension statistic on a report had been altered. We also found no physical evidence supporting this allegation. Although Bonner initially indicated to OIA that "apprehension reports" in general had been altered, in August 1996 he told the OIG that to his knowledge all processed aliens had been reported. None of the other Union officials could provide any direct evidence of altered apprehension figures, either through testimony or documents.

No. 1922 from Imperial Beach did not claim to have any personal knowledge that apprehension reports were altered, but he claimed that another agent had told him that a particular Field Operations Supervisor (FOS) had altered figures on his apprehension and scope reports. We interviewed that agent. He had no knowledge that anyone had changed the apprehension figures he had reported. He did remark that "it would not be above" that FOS to change the numbers "to impress those above him," but the agent had no evidence that the FOS had actually done so.

With respect to the Harper's article, we found no witnesses at Imperial Beach who corroborated the allegation that apprehension statistics were falsified. Nor did we find anyone who admitted being or knowing the magazine's source.

There were several rumors of changed apprehension reports at Imperial Beach. One checkpoint station supervisor said he had heard rumors that unidentified supervisors at Imperial Beach had altered apprehension statistics. He did not know who was allegedly involved in these incidents and he could not recall who had reported this rumor to him. Two Imperial Beach agents and a supervisor at another station also said that they had heard rumors that apprehension figures at Imperial Beach had been altered. None of these individuals could provide specific information, however, and 84 other witnesses at Imperial Beach testified that they had no knowledge that apprehension statistics on the Station's reports had been altered.

We received one hearsay report at Chula Vista from an agent who said he had been told by another agent that his daily reports of apprehensions were returned with figures changed. We interviewed the agent he named. That witness testified, however, that he had no knowledge that his reports of apprehension figures had been changed. Two other agents said they had heard rumors that apprehension statistics had been changed but said they had never seen any evidence of this and could not identify anyone who might have knowledge of such changes.

One Union official from Imperial Beach testified that he had been called by agents from Campo who said that some of that station's apprehension statistics had been falsely reduced. He did not know the identity of these agents, however, and he never saw any documents that supported their claims. We found no agents at Campo who claimed that apprehension figures had been falsified. Moreover, because the purported motive to falsify was to make it appear that illegal traffic had shifted east, efforts to lower apprehension figures in the east would have completely contradicted the underlying theory of the allegations.

Witness after witness at station after station testified that they had no knowledge of apprehension reports being altered. These same witnesses also testified that any alterations of these reports would be detected. On any given night numerous individuals at a station - including apprehending agents, transport agents, the desk officer, the shift intelligence officers who review scope reports, and the agents who process apprehended aliens through IDENT - know how many people have been apprehended. Accordingly, if 500 aliens were apprehended but only 200 were reported, this fabrication would be obvious to numerous station employees. Moreover, intelligence agents review a broad range of reports that - if inconsistent to that magnitude - would obviously indicate problems. As one supervisor wrote in a memorandum to the Patrol Agent-in-Charge (PAIC) at Temecula, "It would take a fool to do 3 or 4 reports and try to falsify all of them."

These witnesses also noted that personnel in a position to notice major inconsistencies were rotated frequently. The desk officer - who receives a form for each apprehended alien and thus is in a position to know if the figures reported by the station accurately report every processed alien - may change daily. Any consistent efforts to underreport aliens would require the cooperation and silence of dozens of agents over the course of a year. Agents testified that if any falsification of the apprehension reports were taking place, it would have to occur after the reports left the station.

Finally, once apprehension information leaves a station, any effort to alter these figures could be easily discovered, because reports containing this data were maintained at each station. A simple comparison of later reports with these earlier reports would reveal any alterations. In sum, each Sector Intelligence witness denied that apprehension figures were altered, and each emphasized that such an alteration could not occur without detection.

Lacking any specific alleged incidents to investigate, we selected a series of reports from the spring of 1995, 1996, and 1997 - at all levels of the Sector - for analysis. Because the spring is a peak period for illegal traffic, this is the time of year when there would be the greatest incentive to falsify apprehension statistics. We obtained transport logs, IDENT records, shift reports, station daily reports, Sector daily reports, Western Region reports, and INS Intelligence reports and compared the reported apprehension figures.

Although we found isolated and minor differences in the numbers set forth in various reports, we found no evidence of altered or falsified apprehension statistics. Where apprehension figures differed, the discrepancy was usually less than five. We occasionally found transposed figures that appeared to be typographical errors. In no instance was there any appearance of an effort to misreport apprehensions. Because of the huge number of apprehensions recorded at the three primary line stations prior to Gatekeeper, any effort to demonstrate a significant decrease in apprehensions would require a consistent underreporting of apprehensions. For example, Imperial Beach alone was reporting more than 165,000 apprehensions per year. To achieve even a 20 percent reduction in apprehensions, the station would need to undercount apprehensions by nearly 100 aliens every day of the year. We found no such discrepancies in our review of documents. Out of 81 days of reports from Imperial Beach, the Sector reported a total of seven more apprehensions than the Imperial Beach station daily reports. Except for minor discrepancies, INS headquarters' apprehension figures matched the Sector Intelligence reports.

At the station level we compared IDENT and transport records to reported apprehensions. Because not every apprehended alien is processed in IDENT, the number of persons in IDENT should not exceed the number of reported apprehensions. Similarly, the number of persons transported from the station should not exceed the number of reported apprehensions. Neither the IDENT nor the transport records in the sample we reviewed provided evidence of underreporting of apprehensions.

Significantly, we found that when major discrepancies in figures appeared in various reports, people asked questions. For example, we found an October 1996 electronic message in which Western Region Intelligence personnel discussed a discrepancy between the apprehension figures reported for San Diego fiscal year 1995 in the official data system at INS Headquarters and a Southwest Border Weekly Operations Situation Report produced by INS Headquarters Intelligence. An error in a spreadsheet formula had led to the exclusion of several months of data.70  Similarly, the PAIC at one station testified that in the spring of 1996 he discovered that the number of reported apprehensions at his station on one day was 30 persons lower than the number of I-826/I-827 forms for that day. He notified the Assistant Chief Patrol Agent (ACPA) for the station and an investigation revealed that these 30 detainees were reported on another shift.

Our investigation also revealed a concern that any numerical discrepancies be explained. For example, a Brown Field supervisor told us that the processing agents' reports of the number of persons processed must agree with the transportation agents' reports at the end of the shift. If these reports do not agree, all of the agents must remain on duty until the discrepancy is resolved.

Other witnesses responsible for compiling apprehension statistics - at all levels of the hierarchy - testified that they regularly check these statistics for discrepancies. Accordingly, if discrepancies appeared in the apprehension data of the magnitude necessary to make a difference in annual apprehension figures, it would almost certainly have been noted by various individuals up the chain of command. The documentary evidence further demonstrated a concern for accuracy. For example, a memorandum from the Western Region Intelligence Office regarding additional reporting requirements noted that "[t]he information requested is for the use of the Attorney General herself. Any information provided must indicate both its source and its credibility. . . . Under no circumstances should we exaggerate the information we actually have on hand."  A memorandum from the PAIC to all agents at Brown Field in October 1995 regarding new reporting requirements told agents to "[b]e as accurate as possible in recording the total apprehensions during the shift. The consolidated figure for the three shifts must match the number of apprehensions recorded on the morning report."  Handwritten notes of a meeting between INS management and the Attorney General discussing Operation Gatekeeper report that the Attorney General told those present that "credibility is the most important thing, don't oversell." Other contemporaneous documents conveyed a similar message.

As noted by numerous witnesses, any attempt to alter apprehension figures would be very obvious to many employees and would leave a paper trail. Our interviews of desk officers, shift intelligence officers, station intelligence officers, and numerous intelligence unit personnel throughout the INS hierarchy produced no factual support for the allegation that apprehension statistics were altered. Because of the absence of any corroborating evidence for this allegation, we find that it is unsubstantiated.

D. Allegations that apprehension records were destroyed

Bonner testified at the summer 1996 hearings that apprehension records are "mysteriously lost" by the Border Patrol, and No. 1922 testified at the Congressional hearing that he has "seen the paperwork documenting the arrest of illegal aliens thrown in the trash." In the hundreds of interviews we conducted, other than No. 1922, no witness claimed to have seen apprehension processing forms destroyed.

The OIG subsequently learned that the Union's allegation that paperwork at Imperial Beach was thrown in the trash was based on second-hand reports from two or three witnesses, one of whom is no longer in the Border Patrol. None of these alleged witnesses was ever identified to the OIG. As a Union official drafted No. 1922's prepared statement for the Congressional hearing that includes this allegation, No. 1922 certainly was one of these persons. The Union provided no additional information about these alleged incidents except to say they occurred sometime in the spring of 1996.

No. 1922 was the only witness we found who claimed to have personal knowledge that apprehension records were destroyed. Although his Congressional testimony implied that he had observed someone else throw apprehension records away, his OIG testimony made it clear that he had personally destroyed such records. He testified that as a desk officer he would receive a large stack of processing papers for apprehended aliens that he was supposed to sort by zone of apprehension. He said, "And after awhile, you're just like does it really matter? So you just throw them in the trash. And the supervisor will say, `Hmm, not a bad idea.' We get to go home early. Because it takes time to break all that stuff down. It's labor intensive." He claimed that apprehension processing forms were thrown away and that aliens were simply put on a bus and returned to Mexico. He also claimed that this was "a common occurrence."

In a subsequent interview he said that this conduct occurred "occasionally." Ultimately, he admitted that the only occasion he was aware of when records were discarded was an incident in which he had discarded processing forms concerning approximately 100 aliens. When asked why he believed it was acceptable to discard apprehension records, he replied, "You've got three or four hundred pieces of paper in front of you, and they are all looking at you to hurry up. And it's not that you have your fellow agents looking at you. They don't give a rat's ass. You've got supervisors looking at you. And never words are mentioned, and that's probably the hardest thing for me to explain, . . . what I mean by `he told us to do it' without so many words. And there was, other encouragement was, `well, we want to get out of here don't we?'"

This agent admitted to the OIG that no supervisor ever saw or had knowledge of records being thrown in the trash. He said that certain supervisors - whom he could not identify - would walk by and say, "Well, pretty big stack there. Well, time to go home." From these comments, No. 1922 said he assumed that it was acceptable to destroy processing forms.

Only two witnesses at Imperial Beach indicated that they had even heard rumors of records being thrown away. One, a supervisor, testified that he had heard that I-826s and I-827s had been shredded. He said he thought if such conduct did occur, it was at another station because he was certain it had not happened at Imperial Beach. A line agent said he had heard a rumor that some records had been thrown away but he did not believe the rumor because he did not believe a supervisor would ever order an agent to destroy records.

Although we interviewed 117 people who worked at Imperial Beach during Gatekeeper, we found no one else who claimed to have observed apprehensions records being thrown away or supervisors "suggesting" that records should be thrown away. No one else interpreted a comment on the size of a pile of documents to mean that the documents should be thrown in the trash.

If I-826 forms were ever thrown away, there still would be transport records, scope reports, IDENT records, and other evidence that would contradict the tally produced by the desk officer. We found no one who claimed to have observed such a discrepancy. Furthermore, none of the sample comparisons we made between reported apprehensions based on the I-826 forms and other records revealed such a discrepancy.

Thus, the only direct evidence in support of this allegation is No. 1922's testimony that on one occasion he destroyed paperwork relating to approximately 100 aliens. He provided no details of any other instances when this occurred, however, and he provided no evidence that any supervisors even knew about it, let alone ordered it done. Furthermore, even assuming that this conduct occurred on his shift "on occasion," it would not be enough to have an impact on apprehension figures for the station.

We found only two witnesses outside of Imperial Beach who believed or had heard that apprehension records had been destroyed. One agent at Chula Vista testified that he had heard that if the apprehension tally sheets prepared by the desk officer did not agree in number with the apprehension forms, some apprehension forms were destroyed so that the number agreed with the tally sheet. He had no other information regarding this allegation, and he had no personal knowledge of this occurring.

This claim is farfetched. The tally sheet merely reflects the count of the apprehension forms. If an employee miscounted the apprehension forms, the employee's next logical step would be to correct the number on the tally form, not to destroy some of the underlying documents to make the number of forms agree with the figure on the tally form. In addition, there is no evidence that the alleged discrepancies were of any consequence, that this occurred on a regular basis, or that this alleged conduct was engaged in at the suggestion of a supervisor.

The only other report regarding destroyed forms was from an agent at Brown Field who testified that on one occasion in 1996 he observed 15 aliens returned to Mexico "without being processed." After further inquiry he explained that the aliens had been processed in the field with I-826/I-827 forms and when the aliens arrived at the station these forms were turned over to the processing unit. He said the aliens were then placed on a bus to an unknown port of entry without being processed in IDENT. Although he did not see what happened to the forms, he assumed the supervisor destroyed the forms because otherwise there was no way the station could explain the discrepancy between the reported apprehensions and the numbers in IDENT. This is the only basis for his speculation that the forms were destroyed and the aliens not counted. As the number of reported apprehensions throughout 1996 was uniformly higher than the number of aliens in IDENT at each of the stations, such a discrepancy would not have been unusual. We found no evidence that a discrepancy of 15 aliens would raise any questions or need explaining. Thus the underlying premise for the agent's speculation is without merit. There is simply no evidence that any processing forms at Brown Field were destroyed.

Due to the absence of any evidence that processing forms were "mysteriously lost" or that anyone other than No. 1922 (who did so at most occasionally and of his own volition) threw away such forms, and in light of evidence to the contrary, we reject this claim as unsubstantiated.

63 It is unclear whether Commissioner Meissner promised such a decrease in the entire San Diego Sector or just the Imperial Beach Station. A careful reading of the transcript suggests that Commissioner Meissner promised a 70 percent decrease in apprehensions only at Imperial Beach where Operation Gatekeeper was already in operation. Taking portions of the testimony in isolation, however, some might believe she had promised a Sector-wide decrease of that magnitude.

64 Under Hold the Line agents were deployed in essentially a single tier directly along the border. Because of the natural barrier created by the Rio Grande river and the flat terrain, the agents were in visible contact with each other and could view the entire border area. Under Operation Gatekeeper the agents were arrayed in a tiered deployment with some agents directly on the border and others in backup positions to facilitate apprehension of individuals who made it past the first tier. The Congressional panel believed that the Border Patrol should deploy its agents in San Diego in exactly the same manner as in El Paso.

65 Previously another form, the I-213 (Record of Deportable Alien), was also completed. With the advent of the IDENT system this form is generally no longer used for persons who are voluntarily returned.

66 Because juveniles are not fingerprinted, the number of persons entered into IDENT for any given period legitimately may be fewer than the reported apprehensions. Historically, as well, due to machine malfunctions, medical needs of the aliens, and other reasons, all adults have not been entered in IDENT. Thus, IDENT figures should not be higher than the reported figures.

67 Prior to May 1995 the transport function was primarily handled by transport officers at the respective stations. In May 1995 this responsibility was transferred to the Alien Detention and Removal Branch (ADR). ADR personnel are not Border Patrol agents although they are employed by the Border Patrol.

68 Because aliens may be transported several hours after they are apprehended, aliens apprehended on the swing shift may not be transported until the next day. Thus, transport records may not necessarily match apprehension records for any given day. Over time, however, cumulative apprehension and transportation numbers should correlate.

69 Although there were allegations that scope reports were altered, there was no claim that apprehension figures on these reports were altered. The claim that "gotaway" figures on these reports were altered is addressed, infra, at 191.

70 The error made it appear as if apprehensions in the San Diego Sector had increased three percent in fiscal year 1996 over fiscal year 1995 when they had actually decreased eight percent. Thus the error was not in the direction of any claim of falling apprehensions.