Robert L. Ashbaugh
Acting Inspector General, U.S. Department of Justice
House Judiciary Committee
Subcommittee on Immigration and Claims
Visa Waiver Pilot Program
February 10, 2000
Mr. Chairman, Congresswoman Jackson Lee, and Members of the Subcommittee on Immigration and Claims:
I appreciate this opportunity to appear before the Subcommittee to discuss the work of the Office of the Inspector General (OIG) with respect to the Visa Waiver Pilot Program (VWPP). The VWPP, a joint responsibility of the Immigration and Naturalization Service (INS) and the Department of State, waives visa requirements for visitors from 29 countries who travel to the United States for business or pleasure. In FY 1997, 14.5 million visitors entered this country under the VWPP.
The OIG examined the potential for fraud in the VWPP and assessed the INS's efforts at air ports of entry to minimize illegal immigration and national security threats posed by abuses of the VWPP. We reported our findings in an inspection report issued in March 1999. During our inspection, we focused on VWPP issues for which the INS has direct responsibility, although we recognize that issues outside of the INS's control can affect the INS's operation of the program.
My testimony today will summarize the major findings in our inspection and report on the status of the INS's efforts to implement our recommendations.
A. VWPP Requirements
The VWPP was created by the Immigration Reform and Control Act of 1986 and implemented two years later. Congress created the VWPP to enhance U.S. foreign relations, to promote effective use of government resources, and - without posing a threat to the welfare or security of the United States - to facilitate international travel, thereby increasing the number of foreign visitors to this country.
Many foreign travelers who wish to enter the United States must first obtain a visa. Obtaining a visa requires that individuals submit written applications and may require interviews and background checks by U.S. Department of State personnel. These interviews and checks are designed to minimize national security threats and prevent illegal immigration.
In contrast, visitors traveling for business or pleasure under the VWPP do not have to obtain a visa to request entry into the United States. Instead, VWPP applicants present their passports and completed I-94Ws (Nonimmigrant Visa Waiver Arrival/Departure Forms) to INS inspectors at U.S. ports of entry. The INS inspectors observe and question the applicants, examine the passports, and conduct checks against computerized databases to make decisions on whether to allow applicants entry into the United States. This review by the INS inspectors is the principal and, in many cases, the only means of preventing illegal entry; the INS inspectors have, on average, less than one minute to check and decide on each applicant.
Under the VWPP, visitors may remain in the United States for up to 90 days per visit but may not work while in this country. In addition, they must possess a roundtrip transportation ticket and must maintain a foreign residence. Finally, VWPP visitors waive their rights to appeal immigration officers' determinations of admissibility or contest any deportation actions.
B. Non-Immigrant Overstays
A country seeking to qualify for VWPP status must meet statutory requirements for a low refusal rate for its citizens who apply for U.S. nonimmigrant visas and it must have a machine-readable passport or be in the process of developing one.
The Secretary of State is responsible for nominating countries to the VWPP and the Attorney General is responsible for determining whether nominated countries should be allowed to participate in the program. The Attorney General must certify that a country nominated for the VWPP does not compromise the law enforcement interests of the United States, and she also must monitor the continued eligibility of VWPP countries.
The INS has several responsibilities related to the VWPP, the most important being the day-to-day operation of the program, including inspection of VWPP visitors at ports of entry. In addition, once countries become participants in the VWPP, the INS must provide the Attorney General with information regarding the countries' continuing eligibility.
However, we found that INS was unable to perform this responsibility because it could not accurately report the number of VWPP visitors who overstayed their authorized periods of admission. Consequently, the Attorney General cannot make informed decisions about countries' continued eligibility for participation in the VWPP using this overstay data.
In a report issued in September 1997, we examined the INS's monitoring of non-immigrant overstays and assessed the Non-Immigrant Information System (NIIS) - the INS's primary data collection system for non-immigrant overstay information. What we found was disturbing.
We reported that the INS's Statistics Division began seriously questioning the quality of NIIS data in FY 1993 when overstay rates for certain classes of visitors - such as visa waiver entrants - rose significantly compared to the preceding year. This finding, coupled with unusual and unexplained fluctuations in overstay rates by class, month of admission, and country of admission, led INS to conclude that NIIS data after FY 1992 was unreliable for estimating overstay rates.
Our report also found that the INS cannot identify individuals who have overstayed their visas nor can it adequately describe the characteristics of the overstay population residing in the United States, such as countries of origin, occupations, and work sites where overstays are employed. This has significantly hindered the INS's efforts to develop an overstay enforcement plan. It also hampers Congress's ability to make informed immigration policy. Two factors contribute to these shortcomings: serious inadequacies in NIIS and a failure to tap potentially rich data developed by INS benefits and enforcement operations.
After our report was issued, the INS formed a team to develop a plan to revise, reload, and analyze NIIS data for fiscal years 1995-1997. In May 1998, INS informed us that their reexamination of the data continued to raise significant concerns about its validity. The INS did not specify what was wrong with the data. We have yet to receive formal notification of their findings, although this is an outstanding recommendation for which we expect to receive a report of corrective action. However, informal contacts with INS staff suggest that the data continues to be unreliable and therefore unusable for accurately estimating overstay rates.
Given the unreliability of NIIS data, the INS cannot provide the Attorney General with the information she needs to certify that the countries enrolled in the VWPP are fully complying with the program's mandates. The disturbing conclusions of our 1997 report caused us to decide to dig deeper into the VWPP, which is the genesis for our 1999 report.
C. VWPP Inspection Process
To facilitate the inspection process at air ports of entry, airlines supply blank I-94Ws to all VWPP applicants en route to the United States. VWPP applicants complete and present these forms, along with their passports, for INS inspectors' review during the inspection process. The INS sends the I-94Ws to a data entry center where clerks enter selected information into NIIS, the INS database that tracks arrivals and departures of non-immigrant foreign nationals. As discussed earlier, if NIIS operated as envisioned it could provide the INS information on whether visitors are overstaying their periods of admission. However, we found during this inspection and other reviews that shortcomings with this system undermine its potential.
During the inspection process, the primary tool available to INS inspectors to prevent applicants from making fraudulent entry into the United States is the Interagency Border Inspection System (IBIS), which is maintained by the U.S. Customs Service (Customs) and commonly referred to as the lookout system. This computerized system allows access to various automated databases that contain records of individuals who should not be allowed entry into the United States. The INS and other federal agencies, such as Customs and the U.S. Department of State, have access to and can enter data into this lookout system.
During primary inspection, an INS inspector examines the applicant's passport and checks the applicant against the IBIS lookout system. If the visitor's passport is machine-readable, the INS inspector scans it into a reader and the applicant's first name, last name, date of birth, and passport number are electronically checked against the lookout system. When a passenger does not have a machine-readable passport, the INS inspector types in the applicant's first name, last name, and date of birth - the only elements the INS requires be checked during primary inspection - and this biographical information is checked against the lookout system. INS inspectors told our inspection team that they manually enter an applicant's passport number that is not machine-readable only if they are suspicious of the applicant.
A. Threat From Passport Theft and Alteration
We found that as the VWPP has grown, so has the potential for its abuse. Because a passport is the only document required for a VWPP applicant to enter the United States, passports from VWPP countries are in high demand on the black market. For example, the INS reported that the price of Slovenian passports increased tenfold on the black market when Slovenia joined the VWPP in September 1997.
We found that the theft of passports, especially blank passports, is a serious problem associated with the VWPP. According to the INS, few countries reported significant problems with thefts of passports prior to implementation of the VWPP. However, at the time of our review, INS estimated that more than 100,000 blank passports from VWPP countries had been stolen in the previous few years. Of those passports, approximately 80 percent were not machine-readable, which means that they would not have been checked automatically against the lookout system.
In addition to stolen blank passports, altered passports and counterfeit passports are used by impostors who falsely claim to be citizens of VWPP countries. For example, we found that citizens of Poland (a non-VWPP country) have presented themselves as Germans (a VWPP country); Albanians have posed as Slovenians; Nigerians have falsely claimed to be citizens of the United Kingdom. Other impostors present genuine passports - often stolen - that belong to citizens of their own countries and attempt to pass themselves off as those individuals.
This evidence from the INS indicates that the VWPP is targeted for abuse and could pose threats to U.S. national security and increase illegal immigration. We also found evidence that a variety of groups, including terrorists, criminals, and alien smugglers, are abusing the VWPP to fraudulently enter the United States. The INS inspectors we interviewed told us that terrorists and criminals apprehended while attempting to enter the United States believed they would receive less scrutiny during the inspection process if they applied under the VWPP. Several of these terrorists and criminals had records that would have prevented them from obtaining a visa, so the VWPP provided an alternative avenue of entry into the United States. Consider the following examples:
One of the conspirators in the World Trade Center bombing entered the country on a photo-substituted Swedish passport in September 1992 (Sweden is a VWPP country). When the terrorist arrived at John F. Kennedy International Airport in New York City, an INS inspector suspected that his passport had been altered. A search of his luggage revealed instructional materials for making bombs; the subject was detained and sentenced to six months' imprisonment for passport fraud. In March 1994 he was convicted for his role in the World Trade Center bombing and sentenced to 240 years in prison and fined $500,000.
Two Irish VWPP applicants attempted to pass through an INS pre-inspection facility in August 1998. INS inspectors questioned both applicants, one of whom admitted that he had served jail time for possession of explosives and had been a member of the Irish Republican Army (IRA). Irish immigration authorities informed the INS that this applicant was a current member of the Real IRA - a terrorist group that has broken away from the original IRA. The INS inspectors denied admission to both applicants.
U.K. citizens traveling under the VWPP are smuggling khat, a leafy narcotic from Africa, into the United States. Drug smugglers solicit help in local pubs, finding unemployed people who need to make extra money. The "mules" are told that they are transporting a plant or vegetable rather than an illegal substance. When Customs officers discover the drugs hidden in luggage, they refer the passengers back to INS for removal. INS inspectors with whom we spoke identified khat smuggling as a problem related to the VWPP.
It is highly improbable that the individuals in these illustrations could have obtained a visa at a Department of State consulate.
In addition, our inspection team observed the following case during our July 1998 field visit to Honolulu International Airport. The INS intercepted a Chinese citizen seeking entry under the VWPP by posing as a Japanese citizen and presenting a Japanese passport (Japan is a VWPP; China is not). The INS inspector noted that the Japanese passport contained characteristics consistent with recent trends in altered Japanese passports. The inspector also reported that the Chinese citizen appeared nervous. A forensic check revealed that the passport had a counterfeit biographical data page and a counterfeit back cover page.
During questioning, the Chinese citizen admitted that he was attempting to enter the United States illegally. He explained that a Japanese citizen on the flight was his escort and that a third man back in China had provided the escort and arranged his travel. The man in China had instructed the Chinese citizen to pay the escort $1,000 when they arrived in Honolulu; the Chinese citizen's family would then pay the man in China $10,000 when he successfully entered the United States.
B. Lookout System Not Queried for all VWPP Applicants
The IBIS lookout system is the INS's principal tool for helping inspectors identify aliens attempting to enter the United States illegally. As I mentioned previously, if VWPP applicants present passports that are not machine-readable, INS inspectors are required to manually enter applicants' names and dates of birth - but not passport numbers - and check them against the lookout system.
According to INS officials, manually entering passport numbers for all passports that are not machine-readable would hamper INS's ability to comply with the congressionally mandated requirement to inspect all passengers within 45 minutes of a flight's arrival. Depending on the volume of applicants at a port of entry, INS inspectors may have to perform quick, rather than thorough, inspections. INS inspectors told the OIG that they manually enter passport numbers only if they are suspicious of an applicant. Therefore, the full benefit of the lookout system is lost for many VWPP applicants whose passports are not machine-readable.
As part of our inspection, we selected a sample of stolen blank passport numbers from 10 VWPP countries and queried them against NIIS records to determine whether aliens had used these passports to enter the United States illegally. While fraudulent entry can be achieved using a variety of methods, we limited our test to stolen blank passports because it was possible to compile a list from which to draw a statistically valid sample.
From our sample of 1,067 blank passports reported stolen from VWPP countries, 40 passports (3.7 percent) had exact matches to NIIS records that showed the passports had been presented for successful entry into the United States. These 40 successful entries project to 2,350 successful entries based on our universe of 61,836 stolen blank passports.
An additional 61 passports (5.7 percent) had possible matches to NIIS records. For these 61 passports, the NIIS record listed a passport number that was slightly different than the number reported as stolen (e.g., the passport numbers in the NIIS record were missing one or more alphabetic characters). Despite the absence of an exact match, we conclude that it is likely that these cases also reflect the use of stolen passports to enter the United States under the VWPP.
C. A Case Study: Stolen Icelandic Passports
We queried all 138 stolen Icelandic passports from our sample universe because INS intelligence reports indicated that a significant number of these passports were being used by organized rings to smuggle children into the United States. We found that VWPP applicants had used 46 of these Icelandic passports to enter the United States successfully. Twenty-seven of the passports were used by more than one person (i.e., an adult traveling with one to three children). Some records indicate that the adults left a short time later but the children did not. The adults returned to the United States (sometimes at different ports of entry) with one to three additional children.
For example, NIIS records indicate that a woman accompanied by two children entered the United States via JFK Airport on July 14, 1996. NIIS records indicate she left the United States via Hartsfield International Airport in Atlanta on July 15, 1996, but the children did not. She again entered the United States through Newark International Airport on October 13, 1996, accompanied by two children. She left through JFK on October 15, 1996, but NIIS records indicate the children did not depart. In total, 116 VWPP applicants (adults and children) entered the United States using the 46 stolen passports.
The passports themselves were stolen from the Icelandic Embassy in Paris, France, in March 1996. The U.S. Department of State issued a cable on May 28, 1996, reporting the theft and the INS created a lookout record for the entire block of stolen blank passports on May 30, 1996. However, the INS's efforts to prevent illegal entry using these passports proved unsuccessful. Even after creation of lookout records and issuance of an alert to INS field personnel, VWPP applicants continued to enter the United States using stolen Icelandic passports - applicants successfully used 22 additional stolen Icelandic passports to enter the United States.
The fact that Icelandic passports are not machine-readable contributes to the success of applicants presenting stolen Icelandic passports. As with all passports that are not machine-readable, it is reasonable to assume that INS inspectors will not routinely check Icelandic passport numbers against the lookout system. Because the lookout record for these (and other) stolen blank passports is based on the passport numbers, INS inspectors will not get a match to a stolen passport record when they enter only applicants' names and dates of birth.
D. Problems with Entry of Data on Stolen Passports
Even if INS checked all passport numbers against the lookout system, the effectiveness of the checks is reduced by problems we identified in the INS's entry of stolen VWPP passport information into the IBIS database. We found that VWPP applicants using stolen passports have entered the United States without being intercepted because INS did not systematically collect information on all stolen passports and did not create lookout records based on the information that it did receive. We also found problems with the timeliness and accuracy of the lookout records that the INS did create.
We found that no single entity within the INS is responsible for the collection of information on stolen passports. While the INS Forensic Document Lab (FDL) receives cables and faxes from foreign governments and the Department of State about stolen passports, the INS LookOut Unit and various ports of entry also receive reports from the Department of State and INS headquarters. We could find no policies instructing these INS components to systematically obtain information on stolen passports from outside sources. A limited review of reports of stolen passports received over a nine-month period at the LookOut Unit and the FDL showed only one block of stolen passports that was reported to both offices. This suggests poor coordination among the various INS entities receiving stolen passport information.
Similarly, we found that no single office within INS was responsible for systematically entering stolen passport numbers into the lookout system. Personnel at the LookOut Unit, the FDL, and ports of entry were all creating lookout records. The result of so many different people entering lookout records is predictable - of the 500 stolen blank passports from our case sample of 1,067 that had lookout records, 202 passports (40.4 percent) had more than one lookout record. Without delegating the responsibility for creating lookout records to a single entity, duplication of effort among the various offices is inevitable.
We found that the remaining 567 stolen passports in our sample (53.1 percent) had no lookout record. In one case, a Dutch passport was stolen in August 1990. The FDL received information about the theft in December 1990. No lookout record was created at the time, and a Chinese alien used this passport to enter the United States in September 1993. Using the same stolen passport, the alien attempted another entry at a different port in March 1996. Although the alien was refused entry on this second attempt, his prior entry could have been prevented had INS created a lookout record. The stolen passport was finally entered into the lookout system by the INS more than five years after INS was alerted to the theft.
The INS inspectors we interviewed did not know who was responsible for entering stolen passports into the lookout system. Some suggested that the FDL should be responsible for entering this information because it already has contacts with foreign agencies and already receives much of the information on stolen passports. Others suggested that the LookOut Unit should be responsible. While there was no consensus for who should have this responsibility, there was support from the INS inspectors we interviewed that the process should be centralized at INS Headquarters.
Finally, passport numbers may contain both numeric and alphabetic characters; alphabetic characters may occur at the beginning, the end, or in the middle of a passport number. Because the lookout system searches for exact matches, the correct placement of these alpha characters is crucial to searching for stolen passport numbers. Of the 500 stolen passports in our sample that had lookout records, 112 (22.4 percent) had discrepancies between the placement of the alpha characters in the actual passport number and the way it appeared in the lookout record. Consequently, if an INS inspector queried the lookout database using accurate passport numbers, there would be no match and it is likely that INS inspectors would admit the individual.
E. Derivative Citizenship
In addition to fraudulent use of VWPP passports, INS Inspectors we interviewed described "derivative citizenship" as a potential source of abuse in the VWPP. Some countries automatically bestow derivative citizenship on children and grandchildren of citizens irrespective of where the children or grandchildren were born or reside. The INS officials specifically noted that Brazilians who have Italian parents or grandparents often obtain Italian passports. Therefore, these citizens and residents of Brazil (not a VWPP country) can enter the United States using passports from Italy (a VWPP country).
Derivative citizenship was described by some INS inspectors as a loophole and an unintended consequence of the VWPP. Using a VWPP passport obtained through derivative citizenship - rather than a passport from their country of residence - allows aliens who are not eligible for a U.S. visa because of economic reasons or a criminal history to circumvent the visa process and enter the United States under the VWPP. INS regulations require that a VWPP applicant must be a citizen of a VWPP country; however, the regulations do not include residency requirements.
III. Conclusions and Recommendations
The VWPP clearly expedites the entry of legal visitors to the United States, as intended. However, we found that the VWPP provides an avenue for terrorists, criminals, and other inadmissible applicants to enter the United States by removing the requirement that they first obtain visas.
We realize that some issues related to the VWPP are beyond the INS's control, such as the Department of State's ability to nominate countries for the VWPP, participating countries' passport issuing processes, and participating countries' reporting of stolen passports. In particular, we note that the law, as revised, states that "[t]he government of the [VWPP] country certifies that it has or is in the process of developing a program to issue machine-readable passports to its citizens." No time frame is specified for actually developing a machine-readable passport. At the time of our review, six countries had been in the process of developing a machine-readable passport for more than seven years.
Nonetheless, in the context of its role of administering the VWPP, the INS has clear responsibilities that it must meet. Our inspection found several areas in which INS's efforts to address the risks posed by abuse of the VWPP were insufficient. Our March 1999 inspection report made three recommendations:
The INS should modify its primary inspection policy to ensure that the passport numbers of all VWPP applicants are queried against the lookout system;
the INS should designate a unit to systematically collect information on stolen blank VWPP passports and to ensure timely entry of stolen passport numbers into the lookout system; and
the INS should develop guidelines for the entry of passport numbers when creating lookout records.
The INS concurred or concurred in part with all three of our recommendations and proposed corrective actions for each.
With respect to our first recommendation, the INS said that it planned to send guidance to the field by the end of the Fiscal Year 1999 that will direct INS inspectors to search all VWPP applicants' passport numbers in the lookout system, even those that are not machine-readable. As of the date of this hearing, we have not received formal notification that such guidance has been disseminated, although INS officials informed us that a memorandum to that effect was distributed to the field on October 4, 1999. We received this week an unofficial, draft copy of the INS's follow-up response to our inspection report together with a copy of the memorandum sent to the field. We are in the process of determining how the INS' actions respond to our three recommendations.
The INS agreed with our second recommendation to designate a unit to systematically collect information on stolen blank VWPP passports and ensure timely and accurate entry of stolen passport numbers into the lookout system. INS Headquarters officials planned to meet with staff from its Inspections Branch, Office of Intelligence, and LookOut Unit to determine the most effective method to complete this task, which it expected to implement by the end of FY 1999. To date, we have not received confirmation of INS action in this regard.
Finally, the INS agreed with our third recommendation to develop clear guidelines for the entry of passport numbers when creating lookout records. The INS said that such guidelines would be developed and implemented in conjunction with its designation of a unit to systematically collect and enter information on stolen, blank passports into the lookout system. Similarly, we have not received official confirmation that such guidelines have been developed or implemented.
This concludes my prepared remarks, Mr. Chairman. I would be pleased to answer any questions you or members of the subcommittee might have.