Inspection of the Secure Electronic Network
for Travelers' Rapid Inspection
Report Number I-2000-019
In evaluating SENTRI's effectiveness at both Otay Mesa and the Peace Bridge, we determined that SENTRI has met its dual mission of reducing commuter wait times for enrollees without compromising border integrity. However, we also determined that SENTRI's impact on both commuter wait times and border integrity in the general inspection lanes has not been adequately assessed. We identified several key areas - funding, performance evaluation, and long-range planning - where INS needs to make improvements for SENTRI to remain viable.
Impact on Commuter Wait Times
Commuter wait times in the SENTRI lane are lower compared to the general inspection lanes
As part of SENTRI's overall mission, SENTRI is intended to provide enrolled commuters with lower wait times compared to the general inspection lanes. Commuter wait times include the length of time that commuters must wait in line before reaching the inspection booth (referred to as queue times) and the length of time needed by inspectors to conduct a primary inspection of each commuter (referred to as primary inspection processing times). At both Otay Mesa and the Peace Bridge, SENTRI has lowered commuter wait times for those using the SENTRI lane, although SENTRI's impact on reducing commuter wait times appears to be much greater at Otay Mesa than at the Peace Bridge.
At Otay Mesa, SENTRI has lowered commuter wait times for those using the SENTRI lane in two ways. First, by automating the primary inspection process and by limiting the use of the SENTRI lane to only pre-approved enrollees, inspectors in Otay Mesa's SENTRI lane conduct a typical primary inspection in less time than they otherwise would in the general inspection lanes. In FY 1998, primary inspection processing times averaged 14 seconds in Otay Mesa's SENTRI lane and between 30 and 45 seconds in Otay Mesa's general inspection lanes.14
Second, by lowering primary inspection processing times, inspectors at Otay Mesa move commuters through the SENTRI lane faster and, in doing so, reduce queue times. In FY 1998, queue times in Otay Mesa's SENTRI lane averaged about one minute compared to nearly 14 minutes in Otay Mesa's general inspection lanes.
At the Peace Bridge, SENTRI has led to lower primary inspection processing times for those using the SENTRI lane. In FY 1998, primary inspection processing times at the Peace Bridge averaged 12 seconds in the SENTRI lane and between 30 and 45 seconds in the general inspection lanes.
Overall, however, SENTRI has had only a limited impact in reducing commuter wait times at the Peace Bridge primarily because commuter wait times in the general inspection lanes were already low prior to SENTRI's implementation. From October 1998 through April 1999, commuter wait times in the general inspection lanes at the Peace Bridge averaged approximately one minute. INS noted as part of the Commissioner's FY 2000 Implementation Plan that "the benefit of SENTRI to travelers [at the Peace Bridge] is much smaller." The INS has also expressed similar concern regarding SENTRI's limited benefit to commuters at the Ambassador Bridge, where commuter wait times in the general inspection lanes averaged less than two minutes from October 1998 through April 1999.
SENTRI's impact on commuter wait times in the general inspection lanes has not been adequately assessed
Neither INS nor Customs has adequately assessed SENTRI's impact on commuter wait times in the general inspection lanes.15 Such an assessment is necessary for several reasons. First, Section 235.7 of the Code of Federal Regulations, "Automated Inspection Services," requires that SENTRI cannot "significantly inhibit normal traffic flow" at any of SENTRI's pilot sites. Second, SENTRI's impact on commuter wait times in the general inspection lanes must be assessed to fully consider SENTRI's cost effectiveness as a strategy for reducing commuter wait times. Finally, one of SENTRI's intended benefits is to "show a time savings for non-[SENTRI] lane users."16
An adequate assessment of SENTRI's impact on commuter wait times in the general inspection lanes would have to include data such as wait times in the SENTRI and general inspection lanes, the total number of inspection hours devoted to conducting primary inspections, the total number of primary inspections conducted, the traffic volume in both the SENTRI and general inspection lanes, the percentage of overall traffic volume processed in the SENTRI lane, inspection processing times for SENTRI and general inspection lanes (i.e., the length of time that it takes inspectors to conduct primary inspections), and traffic patterns in both the SENTRI and general inspection lanes. Fully assessing SENTRI's impact on commuter wait times in the general inspection lanes requires isolating SENTRI's impact, distinct from the impact of other port strategies that are intended to reduce commuter wait times. Neither INS nor Customs has assessed these data to understand SENTRI's impact on commuter wait times in the general inspection lanes. While we reviewed available INS and Customs data, because of the lack of data on the processing times in the general inspection lanes and the need to assess SENTRI's distinct impact on commuter wait times, we were unable to draw definitive conclusions regarding SENTRI's impact.
The failure by INS and Customs to adequately assess SENTRI's impact on commuter wait times has important implications as INS and Customs begin to weigh SENTRI against other strategies aimed at reducing commuter wait times, such as increasing inspection staff at the port, expanding port hours, and increasing the number of open inspection lanes.
From FY 1997 through FY 1998, average commuter wait times in Otay Mesa's general inspection lanes were reduced by four minutes, from almost 18 minutes in FY 1997 to nearly 14 minutes in FY 1998. A number of factors likely impact commuter wait times, including the number of open inspection lanes, traffic volume, and a port's hours of operation. Because these factors have not been fully considered by INS, it is unclear what role SENTRI may have played in reducing commuter wait times for those in Otay Mesa's general inspection lanes.
From October 1998 through April 1999, average commuter wait times in the general inspection lanes at the Peace Bridge remained almost unchanged, from 81 seconds in October 1998 to 93 seconds in April 1999. However, the fact that SENTRI appears to have had no real impact on average commuter wait times in the general inspection lanes at the Peace Bridge raises concerns about SENTRI's cost effectiveness there as a strategy for reducing commuter wait times.
At both Otay Mesa and the Peace Bridge, INS has taken existing general inspection lanes out of service and dedicated these lanes as SENTRI-only lanes. In addition, INS has also reassigned inspectors at both these sites from the general inspection lanes to the SENTRI-only lanes. For these reasons, SENTRI's impact on commuter wait times in the general inspection lanes must be assessed to ensure that SENTRI results in an efficient allocation of inspection resources at both Otay Mesa and the Peace Bridge.
Impact on Border Integrity
At all sites, border integrity in the SENTRI lane has been maintained
By limiting enrollment to pre-screened low-risk commuters, conducting random secondary inspections, and using security features such as tire shredders, automatic gates and pop-up bollards to prevent the unauthorized use of the SENTRI lane, SENTRI is intended to maintain the "security and integrity of the United States border."17 At both Otay Mesa and the Peace Bridge, SENTRI has maintained border integrity in the SENTRI lane. No major violations have been reported in any of the existing SENTRI lanes. There have been no reported incidents of either alien or drug smuggling in any of the existing SENTRI lanes.
At Otay Mesa, just 80 of the 3,300 individuals who have participated in the pilot project at any point from November 1995 through January 1999 have had their SENTRI lane privileges permanently revoked because of border violations committed in the SENTRI lane. These revocations all resulted from what INS and Customs inspectors characterized as "minor" violations of SENTRI's rules of participation, including transporting unauthorized passengers, use of an unauthorized vehicle, and use of the SENTRI lane for commercial purposes. As of January 1999, 287 out of 3,500 applicants at Otay Mesa were denied enrollment in SENTRI due to disqualifying criminal backgrounds, including previous arrests and convictions for alien and drug smuggling, welfare fraud, driving under the influence (DUI), and grand theft (see Appendix VII).
At the Peace Bridge, only 4 permanent revocations occurred among the more than 1,600 individuals enrolled in SENTRI since FY 1998. These revocations all resulted from Customs-related violations of SENTRI's rules of participation and were characterized as "minor" by the Customs inspectors we interviewed. As of January 1999, 38 out of 1,638 applicants at the Peace Bridge were denied enrollment in SENTRI due to disqualifying criminal backgrounds, including previous arrests and convictions for DUI, rape, prostitution, and theft (see Appendix VIII).
At both Otay Mesa and the Peace Bridge, the low incidence of border violations among SENTRI enrollees is due primarily to two factors: a pre-screened enrollment base and random secondary compliance inspections. Because applicants know they are subject to a criminal background investigation as a condition of their participation in SENTRI, there is an apparently high degree of self-selection.
The following example illustrates some of the types of violations reported in the SENTRI lane as well as SENTRI's ability to maintain border integrity:
On August 10, 1998, INS inspectors arrested a Mexican citizen for unauthorized use of Otay Mesa's SENTRI lane. Using a vehicle and electromagnetic identification card belonging to an authorized SENTRI enrollee, the woman attempted to illegally enter the United States through Otay Mesa's SENTRI lane. The woman was returned to Mexico after INS inspectors noted that her appearance did not match with the photographic image linked by SENTRI's GES database to the SENTRI-registered vehicle and identification card that she was using. The woman later told INS inspectors that she was running an errand for her supervisor - who was enrolled in SENTRI - and that she had borrowed her supervisor's SENTRI-registered vehicle and identification card. The vehicle the woman was using was impounded. In addition, INS inspectors permanently revoked the enrollment status of the woman's supervisor for allowing someone other than herself to use her vehicle and identification card for entering the United States through the SENTRI lane.
The frequency of random secondary compliance inspections also helps maintain border integrity in the SENTRI lane. SENTRI enrollees are informed by INS and Customs when they first apply for participation in the pilot project that they are more likely to encounter random secondary compliance inspections than they are in the general inspection lanes.
Additional quality control measures are needed to ensure the integrity of SENTRI's enrollment data
SENTRI's ability to maintain border integrity is related to the strength and quality of its enrollment data. This data, which includes biographical information as well as employment history, criminal history, and other relevant information, is used by INS and Customs inspectors to ensure that only eligible, low-risk border crossers are allowed to enter the United States using the SENTRI lane. For each applicant, both a hardcopy file and an electronic record within the GES database are created to store enrollment information.
Because of the information collected from applicants during SENTRI's enrollment process, inspectors know a great deal more about border crossers in the SENTRI lane than they know about border crossers in the general inspection lanes. Nevertheless, problems with data integrity have limited the extent to which inspectors at Otay Mesa are able to rely on information collected from applicants as a tool for ensuring that border integrity in the SENTRI lane is maintained.
In an April 1998 management report on SENTRI's overall operations at Otay Mesa, Customs found that there was a "high percentage of enrollment documents with incomplete information." Our review found that data integrity continues to be a problem at Otay Mesa. For example, inspectors at Otay Mesa have not consistently documented reasons for denying applicants in either the hardcopy enrollment files or in the GES database.18 This lack of documentation makes it difficult for inspectors to re-evaluate previously denied applications that are later re-submitted by applicants for consideration. In addition, inspectors at Otay Mesa have not consistently documented reasons for revoking enrollment status. Our review found no major problems with data integrity at the Peace Bridge.
In ensuring the integrity of SENTRI's enrollment data, several audit procedures have been developed for sites along the Southwest border (see Appendix III, page 46). These procedures include: conducting weekly checks to verify that information collected from new enrollees is both accurate and complete, as well as conducting quarterly reviews of selected enrollment records to verify their accuracy. Although efforts have been undertaken at Otay Mesa to review the quality of SENTRI's GES database and the hardcopy enrollment files, similar audit procedures have not been developed for SENTRI's sites along the Northern border. Following our review, inspectors at all sites are now required to verify the accuracy of information maintained in both the GES database and in hardcopy enrollment files for new applicants. To ensure the integrity of SENTRI's enrollment information, existing enrollment data, as well as new applications, should be systematically reviewed.
Required administrative procedures designed to ensure border integrity are not regularly performed at the Peace Bridge
To ensure that border integrity is maintained in the SENTRI lane, inspectors at each of the sites are required to perform several administrative procedures. For example, inspectors are required to perform IBIS queries on a daily basis for each SENTRI enrollee as well as update SENTRI's GES database with any new information produced from these queries.19 These queries enable inspectors to continually verify the eligibility status of enrollees. At the Peace Bridge, however, inspectors are not performing daily IBIS queries for SENTRI enrollees, as required. Instead, inspectors are performing these queries on a sporadic basis.
SENTRI's rules of participation also require that enrollees renew their enrollment status on an annual basis. As part of the annual renewal process, inspectors are required to perform electronic criminal history checks for each enrollee, verify citizenship status of each enrollee, and ensure that each enrollee has not violated any of SENTRI's rules for participation. Despite these requirements, enrollment status is currently renewed automatically for enrollees at the Peace Bridge without any level of review by inspectors.
Limited staffing resources may be principally to blame for the Peace Bridge not performing required administrative procedures. At the Peace Bridge, just one INS inspector has been assigned to SENTRI's enrollment center, and only as a collateral duty. As a result, the Peace Bridge has been unable to staff the SENTRI enrollment center on a regular basis since April 1998. Port officials at the Peace Bridge estimate that four full-time inspectors are needed to operate the SENTRI enrollment center on a full-time basis. At Otay Mesa, in comparison, nine INS inspectors, two Customs inspectors, and three INS immigration assistants are assigned full-time to SENTRI, with inspectors from both agencies rotating between the SENTRI lane and the enrollment center.
Although SENTRI's staffing levels are determined at the local level under the 1994 MOU between INS and Customs, these staffing levels are based on overall port-wide staffing levels, which are determined by headquarters officials. For this reason, local and headquarters officials should work together to ensure that SENTRI's sites have adequate staffing resources for performing required administrative procedures. While border integrity has been maintained in the SENTRI lane at each of the sites thus far, these administrative procedures provide additional safeguards against the illegal or unauthorized use of the SENTRI lane and therefore should be performed consistently by inspectors at each pilot site.
SENTRI's impact on border integrity has not been adequately assessed for the general inspection lanes
By removing pre-screened, low risk commuters from the general population of border crossers and allowing INS to focus inspection resources on higher-risk crossers, SENTRI is intended by INS to strengthen port-wide border integrity. In fact, as part of its application for the 1997 Innovations in American Government Award, INS wrote that SENTRI's "most important achievement has been to strengthen border security." One of SENTRI's goals for sites along the Southwest border is to enhance border integrity in the general inspection lanes by "enabl[ing] the higher-risk non[SENTRI] traffic to be more intensively examined [by INS and Customs inspectors]." This goal is measured in terms of the number of individuals referred for a secondary inspection and the total number of border violations reported in the general inspection lanes both prior to, and following, SENTRI's implementation at each of SENTRI's sites along the Southwest border.
Based on the available data at Otay Mesa, it is unclear whether SENTRI has achieved the goal of enhanced border integrity in the general inspection lanes. To fully assess SENTRI's impact on border integrity in the general inspection lanes, data must be analyzed for periods both prior to, and following, SENTRI's implementation. Because INS reported data only on a district basis until FY 1996, we were unable to obtain specific data for Otay Mesa prior to SENTRI's implementation there.20 Although data for Otay Mesa should be available to INS prior to FY 1996, we found no evidence that INS or Customs has assessed SENTRI's impact on border integrity in the general inspection lanes using this data.
Regardless of whether specific data for Otay Mesa is available prior to FY 1996, performance benchmarks should be developed that specify SENTRI's expected performance in enhancing border integrity in the general inspection lanes. These benchmarks are necessary for providing a context to SENTRI's reported performance data and allow for useful comparisons to be made between SENTRI's expected and actual performance.
However, just as with commuter wait times, a number of factors likely impact border integrity in the general inspection lanes, including the number of individuals inspected, training levels among INS and Customs inspectors, the length of time spent by inspectors conducting an individual inspection, and trends in illegal border activities. For this reason, SENTRI's existing performance measures should be reviewed to ensure that these are both adequate and appropriate for assessing SENTRI's impact on border integrity in the general inspection lanes.
Specific goals regarding border integrity have not been developed for SENTRI's sites along the Northern border. Specific goals and benchmarks should be developed for all sites to evaluate SENTRI's impact on border integrity.
San Ysidro will be the first major test for SENTRI's enrollment process
In both maintaining and enhancing border integrity, one of SENTRI's goals for sites along the Southwest border is to develop a thorough enrollment process that "rigorously examin[es] each applicant's background to determine verifiable residence and employment status, lack of a disqualifying criminal history or prior border violations, admissibility to the United States, and adequately explained reasons for frequent border crossings." By identifying applicants who may be appropriately considered "low-risk," this enrollment process is the cornerstone of SENTRI's ability to ensure border integrity in the SENTRI lane as well as enhance border integrity in the general inspection lanes.
While our review found no major problems with SENTRI's enrollment process, the first major test for this process will come once SENTRI becomes operational at San Ysidro. As the largest and busiest land port-of-entry in the United States, San Ysidro offers INS and Customs a unique and considerably less-controlled environment in which to test SENTRI's capabilities compared to any of the existing sites. In FY 1998, for example, nearly 41 million individuals were inspected by either INS or Customs at San Ysidro and 130,000 total border violations were reported in San Ysidro's general inspection lanes. This includes 396 individuals who were involved in smuggling narcotics and 8,500 individuals who were intercepted in the course of being illegally smuggled into the United States.
Aside from sheer volume, San Ysidro's anticipated enrollment base may make it more difficult for inspectors to verify information submitted by non-U.S. applicants during SENTRI's enrollment process. The INS and Customs project that 12,000 individuals may eventually enroll in SENTRI at San Ysidro. Unlike at Otay Mesa, where U.S. citizens make up more than 60 percent of SENTRI's enrollment base (see Appendix IX) and at the Peace Bridge, where they make up nearly 80 percent (see Appendix X), INS officials project that non-U.S. citizens may comprise 60 percent or more of San Ysidro's enrollment base.
Because INS expects to enroll a large number of non-U.S. citizens in SENTRI at San Ysidro, the U.S. Attorney for the Southern District of California expressed concern to us that SENTRI's current enrollment process may be inadequate for ensuring that only low-risk commuters are allowed to enroll in SENTRI at San Ysidro. Although INS and Customs inspectors are able to run electronic criminal history checks on crimes committed in the United States by SENTRI applicants, they are usually unable to access information on crimes committed in other countries by both U.S. and non-U.S. citizens alike.
As a result, INS and Customs inspectors must rely on criminal history checks performed by other countries that may not be as comprehensive or complete as those performed in the United States. For example, because Mexico does not have a national criminal database, INS inspectors must rely on letters submitted by applicants from local Mexican police officials that state whether an applicant has a criminal history. These letters only provide information on criminal histories within a particular local jurisdiction within Mexico and do not include national criminal histories. Canadian criminal history checks, on the other hand, can be accessed directly by U.S. law enforcement officials through Canada's national electronic criminal database, the Canadian Police Information Center.
In reviewing enrollment applications submitted by non-U.S. citizens, INS and Customs inspectors must also rely on supporting documentation that may be more difficult to verify than similar documentation submitted by U.S. citizens. This supporting documentation, used primarily to verify residence and employment status, often includes a letter from an employer, a recent payment stub, or a vehicle's registration.
Because of the unique challenges posed by San Ysidro's potential enrollment base, INS and Customs should consider whether additional safeguards are needed at San Ysidro to ensure that SENTRI's goals regarding border integrity are consistently met.
SENTRI's current user fee is inadequate for recovering full operating costs
Because SENTRI offers enrollees unique benefits unavailable to border crossers in the general inspection lanes, OMB Circular A-25 requires that INS recover the full cost of SENTRI's operation through the collection of a user fee when or if SENTRI becomes a permanent program. For this reason, all land border pilot projects, including SENTRI, are intended by INS to be self-supporting through the collection of user fees.
In FY 1997, INS first began collecting an annual user fee from SENTRI's enrollees at Otay Mesa. In September 1999, INS also began collecting an annual user fee from SENTRI's enrollees at the Stanton Street Bridge. At both sites, SENTRI's user fee is set at $129 per enrollee. INS currently has plans to collect SENTRI's $129 annual user fee from enrollees at all future sites along the Southwest border, including San Ysidro. INS does not collect a user fee from SENTRI's enrollees at either the Ambassador Bridge or the Peace Bridge and there are no immediate plans by INS to begin doing so.21
User fee revenue generated by SENTRI and other land border pilot projects is deposited in the Land Border Inspections Fee Account and allocated separately to each pilot project, including SENTRI. At Otay Mesa, SENTRI's operating costs were approximately $595,000 for FY 1998, but only $352,170 in user fee revenue was generated, resulting in a shortfall of $243,000.22 For this reason, SENTRI's current $129 annual user fee is inadequate for recovering full operating costs. Unrealized enrollment projections used by INS in setting SENTRI's current user fee level are primarily to blame.
As early as FY 1997 concerns were raised to INS that SENTRI's current user fee level was insufficient for recovering full operating costs. In June 1997, a report prepared by Electronic Data Systems concluded that in order to recover SENTRI's full operating costs, INS would need to increase SENTRI's user fee to nearly $320 per enrollee, assuming SENTRI reached its full enrollment capacity of 6,000 individuals, and to nearly $600 per enrollee if SENTRI's enrollment reached only 2,500 individuals. A similar report prepared by PriceWaterhouseCoopers (PWC) in June 1999 also concluded that SENTRI's current user fee was insufficient for recovering SENTRI's full operating costs.
Based on enrollment levels at Otay Mesa at that time, PWC concluded that INS would need to increase SENTRI's user fee to $309 per enrollee to recover the full costs of SENTRI's operation. While results from Otay Mesa demonstrate that some commuters are willing to pay a user fee to enroll in SENTRI, it is likely that increasing this fee would discourage at least some enrollment.
Along the Northern border, commuter resistance remains strong to paying a user fee of any amount
Along the Northern border, there are indications that commuters may resist paying a user fee altogether. As part of the Commissioner's FY 2000 Implementation Plan, INS noted that " . . . by charging a fee in Buffalo or Detroit, we would probably eliminate the use of the [SENTRI] program in those locations." The port officials we interviewed at the Ambassador Bridge and the Peace Bridge also shared this view. Like SENTRI, previous DCL programs at both the Ambassador Bridge and the Peace Bridge offered commuters the benefit of an expedited inspection process but at no charge. Combined with the fact that SENTRI appears to offer little timesavings to commuters along the Northern border, INS should continue to expect commuters there to resist paying SENTRI's user fee.
Moreover, commuter resistance along the Northern border to SENTRI's user fee will likely grow once Canada's new CANPASS program, which does not charge enrollees a user fee, becomes operational at both the Ambassador Bridge and the Peace Bridge in FY 2000. SENTRI and CANPASS are expected to draw from the same enrollment pool. At the Peace Bridge, Canadian officials we interviewed project that as many as 35,000 American citizens will initially enroll in CANPASS.
Regardless of commuter sentiment, OMB Circular A-25 requires that INS collect a user fee from SENTRI's enrollees at each pilot site if SENTRI becomes a permanent program. For this reason, INS should assess the impact that SENTRI's user fee will have on enrollment levels at both existing and proposed sites. Without performing such an assessment, INS cannot be assured of SENTRI's financial viability at specific sites.
Funding shortfalls have hampered many of SENTRI's basic operations
Because of the shortfall in user fee revenue generated by SENTRI, INS has had to rely on a number of other funding sources to recover SENTRI's operating expenses, particularly along the Northern border where no user fee is collected. These funding sources have included the INS Commissioner's Contingency Fund as well as a grant from the Ford Foundation. In FY 1997, INS also received a $9.5 million congressional appropriation to jointly develop automation systems with Customs, including $2.8 million for SENTRI. These funding sources have often proven unpredictable from year-to-year, forcing SENTRI to "live hand-to-mouth," as one INS official told us. As a result, SENTRI has experienced several operational challenges since first being implemented at Otay Mesa in November 1995.
In FY 1999, for example, a funding shortfall prevented SENTRI's purchase of additional radio frequency transponders for Otay Mesa.23 These transponders are used by SENTRI-registered vehicles as they approach the inspection booth to transmit a unique signal to SENTRI's automated operating system. This signal then initiates the inspection process. Without this necessary equipment, new applicants were unable to enroll at Otay Mesa from October 1998 through October 1999.24 As a result, SENTRI has lost at least $50,400 in user fee revenue at Otay Mesa that could otherwise have been collected from the 630 individuals whose names were placed on a waiting list for transponder installation as of April 1999.25 Port officials at Otay Mesa also told us SENTRI's shortage of transponders may damage SENTRI's credibility with local commuters interested in participating in SENTRI.
A similar funding shortfall has meant that equipment used in the SENTRI lane at the Peace Bridge can no longer be serviced for repairs. For example, the electromagnetic identification card reader was out of service during our visit to the Peace Bridge. This card reader helps INS and Customs inspectors verify the identity of SENTRI enrollees. Port officials told us they could not afford to pay for a regular service contract. At the Ambassador Bridge, telephone service was cut off to the SENTRI enrollment center during the first week of the site's operation due to late payments.
Project Evaluation and Planning
Critical issues regarding SENTRI's future have not been resolved
In FY 1997, Congress authorized INS and Customs to develop land border pilot projects, including SENTRI, through September 30, 2000. Under Section 286(q) of the Immigration and Nationality Act, Congress may extend the period of authorization for these pilot projects, as it did once before in FY 1994. Although the INS, Customs, and JPR officials we interviewed told us they expect that Congress will reauthorize these pilot projects, long-range plans regarding SENTRI's future have not been developed.
In addition, SENTRI's status as a reinvention laboratory is scheduled to terminate in August 2001. Typically, JPR's reinvention laboratories have life cycles of 18 to 24 months; SENTRI has been a laboratory since March 1995. Prior to the scheduled termination of each reinvention laboratory, JPR requires that a final report be issued that identifies various "closeout options."26 These options may include terminating the reinvention laboratory, extending the reinvention laboratory by one additional year, or establishing the reinvention laboratory as a permanent program within a particular agency. Although the INS, Customs, and JPR officials we interviewed all believe that SENTRI will eventually become a permanent program, no formal plans have been developed for SENTRI's anticipated transition from a reinvention laboratory.
The absence of any long-range planning leaves many critical issues regarding SENTRI's future still unresolved. For example, it is unclear whether SENTRI's sites along the Northern border will be integrated with the Southwest border sites that are currently affiliated with the reinvention laboratory program. Furthermore, the ability of INS to identify a permanent funding source and allocate adequate resources for the program to meet its objectives remains unresolved.
Also yet to be resolved is how SENTRI will be managed if SENTRI becomes a permanent program and what agency - INS or Customs - will assume primary responsibility for SENTRI's management. SENTRI's dual status as both a reinvention laboratory and a separately managed INS pilot project has led to confusion among INS, Customs, and JPR officials over which agency has primary responsibility for developing long-range plans regarding SENTRI's anticipated transition from a reinvention laboratory to a permanent program. For this reason, greater coordination is needed between INS, Customs, and JPR to ensure SENTRI's success.
SENTRI's overall operations have not been fully evaluated
For each of SENTRI's sites, the 1994 MOU between INS and Customs calls for SENTRI:
. . . to be evaluated semiannually to monitor effectiveness and the need for modifications. In addition to the statistics relating to participation, the evaluation shall address the areas of facilitation, compliance, impact on overall processing times, staffing, and other operational concerns.
As of December 1999, the only evaluation that INS has performed of SENTRI was a review of computer security completed in March 1996.27
In April 1998, Customs completed a review of SENTRI's operations at Otay Mesa, which concluded that, "from an enforcement point of view, SENTRI has been a complete success." The Customs review evaluated SENTRI's enrollment process as well as SENTRI's impact on border integrity and included 25 "suggestions" for improvement to operating procedures, such as the information included on enrollment applications and the type of uniform Customs inspectors should wear. Although the Customs review of SENTRI is the most comprehensive to date, this review did not consider SENTRI's effectiveness in reducing commuter wait times, financial viability, impact on the general inspection lanes, and the appropriateness of SENTRI's current performance goals and measures.
For sites along the Southwest border, SENTRI's Performance Measurement Plan also calls for an annual audit of SENTRI to be performed either by SENTRI's project team or by an outside contractor that includes "a review of the goals and the measurement collection methods/records of the entire [SENTRI] lane." As of December 1999, no annual audits of SENTRI have been performed by either SENTRI's project team or by an outside contractor. As a result, neither SENTRI's performance goals, performance measures, or data collection methods have been adequately reviewed since these were first developed. As noted earlier, INS has performed no evaluation of the impact of SENTRI on wait times or border integrity in the general inspection lanes.
Although regular reports prepared by INS, Customs and JPR detail the extent to which SENTRI has achieved specific performance goals, such as reduced commuter wait times or enhanced border integrity, these reports do not fully evaluate SENTRI's overall operations. Among other things, these reports do not consider SENTRI's cost effectiveness, strategic planning efforts, or cross-agency coordination efforts. Unless SENTRI's overall operations are fully and regularly evaluated, neither INS, Customs, nor JPR can be assured of SENTRI's continued effectiveness.
A more comprehensive framework for evaluating proposed SENTRI sites is needed
The 1994 MOU between INS and Customs allows either agency to propose new SENTRI sites, subject to the approval of the other agency. Congress may also propose new SENTRI sites either in coordination with, or independently of, INS and Customs. INS and Customs must be jointly authorized by Congress to establish a SENTRI site at a particular port-of-entry.
As of December 1999, Congress had authorized INS and Customs to implement SENTRI at nine sites, including SENTRI's four current sites: Otay Mesa, the Stanton Street Bridge, the Peace Bridge, and the Ambassador Bridge.28 Unless funding is specifically appropriated by Congress, sites authorized by Congress are regarded as non-binding "recommendations" for INS and Customs to jointly consider.29 Both agencies must report to Congress regarding their final site selection decisions as well as provide justification for why specific sites authorized by Congress were not selected.
Based on criteria outlined in Section 235.7 of the CFR, SENTRI's sites may be selected by INS and Customs only if:
These criteria are quite broad, however. Section 235.7 does not establish specific thresholds that must be met in order to satisfy each of the criteria required for selection as a SENTRI site. For example, while SENTRI's sites must have a "sufficient" number of personnel on hand to perform primary and secondary inspection functions, the word "sufficient" is broadly stated and therefore open to interpretation. In addition, Section 235.7 does not consider other criteria that may also be critical to SENTRI's success at a particular pilot site, including current commuter wait times and incidences of border violations.
Selecting SENTRI's Pilot Sites: A Case Study
The following example illustrates the many complexities that often underlie the process for selecting SENTRI's pilot sites and demonstrates why developing an objective and comprehensive framework for evaluating proposed pilot sites is critical to SENTRI's long-term success:
In FY 1997, Congress authorized INS and Customs to jointly develop a SENTRI pilot site in El Paso, Texas. In response, a joint INS-Customs team assessed five ports-of-entry within the El Paso area for their suitability as SENTRI pilot sites and reported its findings to Congress in February 1997. Based on a number of considerations, including daily traffic volume, existing physical infrastructure, and community support, the team concluded that Ysleta was the most suitable pilot site.
Among the ports-of-entry not recommended by the team was the Stanton Street Bridge, which the team concluded lacked adequate building facilities and posed a major traffic flow problem by leading directly into El Paso's congested business district. Despite the team's recommendations, INS and Customs jointly selected the Stanton Street Bridge as a SENTRI pilot site after city officials and the El Paso Chamber of Commerce agreed to provide INS and Customs with building facilities for SENTRI's enrollment center at the Stanton Street Bridge.
Because of its close proximity to El Paso's business district and central location, the Stanton Street Bridge was heavily favored by El Paso's local community, which hoped that SENTRI would help ease work commutes as well as stimulate local economic growth. Whether the Stanton Street Bridge was the most suitable pilot site from among those considered remains to be seen.
Expanding on the criteria required by Section 235.7, INS and Customs jointly developed a framework for evaluating proposed SENTRI sites that considers a number of additional factors, including community support, existing physical infrastructure, and daily traffic volume. Using this framework, a joint INS-Customs team assessed six ports-of-entry along the Southwest border for their suitability as SENTRI sites in February 1997. These ports-of-entry, which were all authorized by Congress in FY 1997, included: San Ysidro, California; El Paso, Hidalgo, and Laredo, Texas; and Nogales and San Luis, Arizona.
Although a number of factors were considered, the INS-Customs assessment team relied primarily on observational and qualitative data. Actual commuter wait times, incidences of border violations, potential enrollment levels, and local INS and Customs staffing levels were not fully considered.30
A more comprehensive framework for evaluating proposed SENTRI sites, which includes an analysis of both qualitative and quantitative data, would provide INS and Customs with a stronger and more objective basis for selecting sites. Moreover, by linking this framework to SENTRI's specific performance goals, INS and Customs would be better able to project SENTRI's impact at proposed sites as well as to enhance the quality of site selection decisions that are made by both agencies. Finally, improvements to this framework would help INS and Customs in providing Congress with richer and more complete suitability assessments of SENTRI's congressionally-authorized sites.
Current methods are inadequate for projecting enrollment levels at proposed SENTRI sites
Projecting enrollment levels at proposed sites is critical to evaluating each pilot site's overall suitability. Enrollment plays a major part in determining SENTRI's level of success. For example, low enrollment may limit SENTRI's role in reducing commuter wait times for those in the general inspection lanes and may even lead to increased commuter wait times if enrollment falls too low. Enrollment is also critical to SENTRI's cost effectiveness. If, for example, enrollment is too low, SENTRI may become too costly for INS and Customs to operate. Finally, because SENTRI's user fee is based on SENTRI's operating costs per enrollee, varying enrollment levels may mean the difference between a prohibitively high user fee and a user fee that is considered by commuters to be affordable.
Although a number of ports-of-entry, including Otay Mesa and the Peace Bridge, have attempted to survey local commuters regarding their interest in participating in SENTRI, none of these surveys have been helpful in accurately projecting enrollment. In December 1997, for example, INS and Customs jointly distributed 35,000 surveys at the McAllen-Hidalgo-Reynosa International Bridge near Hidalgo to assess local commuter interest for participating in SENTRI.31 Although 78 percent of those who responded to the survey indicated they had an interest in participating in SENTRI, less than two percent of those who received surveys responded. Despite the survey's results, such a low response rate makes it difficult to draw any meaningful conclusions regarding SENTRI's potential enrollment level at Hidalgo, particularly because no analysis was done to determine why such a large number of those who received surveys did not respond. The port concluded from the survey that the bridge had as many as 53,000 potential SENTRI enrollees, a rather high projection given the two percent response rate and the enrollment levels at other SENTRI sites.32
Aside from surveys, INS and Customs have relied mostly on best guesses and rough approximations for projecting enrollment levels. Not surprisingly, these methods have often proven unreliable. At Otay Mesa, for example, INS and Customs originally projected there would be 6,000 SENTRI enrollees by FY 1999. As of September 1999, SENTRI's enrollment at Otay Mesa was just under 3,300 individuals - the highest enrollment level since SENTRI first became operational at Otay Mesa.
Given the important role that enrollment plays in determining SENTRI's level of success at each of SENTRI's sites, current methods used for projecting enrollment are inadequate and should be strengthened to ensure the reliability and accuracy of future projections.
to the USDOJ/OIG Home Page
Return to the Table of Contents