Inspections of Firearms Dealers by the Bureau of Alcohol, Tobacco, Firearms and Explosives
Report Number I-2004-005
The ATF Does Not Conduct In-Person Application Inspections on All New FFLs to Verify Applicant Information and Ensure That They Understand Firearms Laws
According to ATF Field Operations Directorate officials, application inspections are critical for ensuring compliance with federal firearms laws.44 Application inspections not only enable the ATF to better ensure that the applicant is eligible for a federal firearms license, but also provide the new dealer an opportunity to ask ATF Inspectors questions and discuss with them issues related to firearms laws. Further, the ATF officials said that if an FFL violates federal firearms laws after having received an application inspection, it is easier for the ATF to meet the legal standard of demonstrating that the violations were "willful" in order to take adverse action.
According to the ATF's N-Spect database, which tracks inspections activity, the ATF conducted 8,123 application inspections in fiscal year (FY) 2002.45 According to the ATF's FLS database, which tracks all firearms licenses issued by the NLC, there were 15,359 newly licensed FFLs in FY 2002. Of that number, 7,977 were licenses issued for firearms businesses and 7,382 were issued to firearms collectors. By examining the "inspection date" field included in the FLS record for each FFL, we found that most (95 percent) of the FFLs that were inspected were businesses, not collectors.46 Therefore, the data available to us indicates that in FY 2002 the ATF inspected the preponderance of new firearms retailers.
However, our interviews and survey of ATF Headquarters and Field Division personnel found that many of those application inspections consisted only of a telephone call rather than an in-person visit. Ten of the 45 Area Supervisors we surveyed (22 percent) said that, because of limited Inspector resources, they directed Inspectors to perform at least some application inspections by telephone. One Area Supervisor stated that he assigned telephone application inspections because some applicants were located as far as 400 miles from his Area Office. He said that, in those cases, he tries to schedule an in-person compliance inspection for the following year, but cannot guarantee that it will occur.
ATF Headquarters officials, DIOs, Area Supervisors, and Inspectors told us that application inspections conducted by telephone are not as comprehensive as in-person inspections. For example, one important action Inspectors take during in-person inspections is to interview individuals who will work for the FFL. When application inspections are conducted by telephone, the Inspectors do not routinely interview store employees.
Subsequent to the initiation of this review, in an October 8, 2003, memorandum to ATF Field Divisions, the ATF Assistant Director for Firearms, Explosives, and Arson issued new guidance on application inspections. The new guidance stated that the ATF's goal is to inspect all new FFL applicants to ensure that they understand federal firearms laws before issuing them a federal firearms license. The memorandum established as an initial milestone that "Inspectors should ensure that all new applicants receive onsite inspections" at 14 select cities (emphasis in original).47
After receiving a draft of this report, the ATF attempted to more precisely determine the number of inspections conducted by telephone. Although in 2002 the N-Spect system did not capture the method by which an inspection was conducted, ATF's Field Management Staff attempted to estimate the number of application inspections conducted by telephone by querying the database to identify application inspections with words such as "telephone" or "phone" in a "Special Instructions" data field. The results of that query indicated that about 4 percent of application inspection records had such remarks. However, because there was no direction to the field to ensure consistent entry of inspection methods, and because words such as "telephone" could be entered for reasons other than to specify the inspection method, we found these results unreliable.
Although the new goal established in the October 8, 2003, memorandum is a step in the right direction, the ATF still does not comprehensively inspect all new applicants in person. Therefore, it cannot verify the accuracy of the information that each applicant provided to the NLC to become an FFL. Further, foregoing an in-person inspection means that the applicants' opportunity to ask questions of ATF Inspectors is limited. While FFLs should understand and follow the law, ATF personnel said that if an FFL fails to comply with federal firearms laws, the lack of an application inspection makes it harder for the ATF to meet the legal standard of proving that the violation was "willful" in order to take adverse action.
The ATF Does Not Regularly Conduct Compliance Inspections on Active FFLs, Including Large-Scale Retailers
According to the ATF Director, for the ATF to ensure compliance with federal firearms laws, FFLs should receive a compliance inspection at least once every three years. However, the ATF is currently unable to even begin to meet that goal. We found that most FFLs are inspected infrequently or not at all. ATF workload data show that the ATF conducted 4,581 FFL compliance inspections in FY 2002, or about 4.5 percent of the approximately 104,000 FFLs nationwide.48 At that rate, it would take the ATF more than 22 years to inspect all FFLs.
We reviewed inspection history files for a sample of 100 FFLs that had been in business for an average of 11.2 years, and verified that they did not receive regular compliance inspections (Figure 4).49 Specifically:
Source: ATF inspection history files
Large-scale retailers sell a higher volume of guns than small dealers. Despite the potential for large numbers of improper sales, we found that large-scale retailers also are not inspected on a routine basis. Our sample showed that the ATF conducted compliance inspections on large-scale national and regional chain retailers such as Wal-Mart, Sports Authority, or Big 5 (a sporting goods chain with 293 stores throughout 10 western states) with about the same infrequency as small dealers. The five Area Supervisors we interviewed told us that they avoid selecting large FFLs for compliance inspections because the large volume of records makes the inspections more difficult and time-consuming. The nine large-scale retailers in our sample were inspected about once every 9.9 years, versus once every 9.2 years for the overall sample.51
Further, we found that the 9 large-scale retailers included in our sample were only slightly more likely to have been subjected to a compliance inspection than the average FFL, with 33 percent (3 of 9) having received at least one compliance inspection, versus 29 percent of FFLs overall.52 Of the nine large-scale retailers we reviewed:
Because the ATF does not conduct regular inspections of FFLs, the ATF cannot effectively monitor the overall level of FFL compliance with federal firearms laws and regulations. In December 2003, the ATF initiated a program to conduct special Random Sample Compliance Inspections to develop a risk model for the FFL inspection program. Using data from those inspections, the ATF planned to "be able to project the overall level of compliance by" gun dealers, pawnbrokers, and collectors. 53 While the project to estimate the overall level of compliance with laws is needed to identify the challenges facing the ATF, it cannot take the place of regular compliance inspections for deterring and identifying noncompliance with gun laws.
The ATF Does Not Identify and Inspect All FFLs That Exhibited Indicators of Potential Violations or Gun Trafficking
The ATF has not identified all FFLs that exhibit characteristics of violations or firearms trafficking in order to properly manage its range of enforcement activities, as well as to inform the Attorney General, Congress, and the public of the scope of the potential trafficking problem at FFLs. Gun dealers are selected for compliance inspections either by ATF Headquarters (under the Focused Inspection, Demand Letter, or other programs) or by the 23 ATF Field Divisions. Under the Focused Inspection Program, the ATF Firearms Programs Division selects FFLs for mandatory compliance inspections by the Field Divisions. The Chief of the Firearms Program Division told us that the ATF and the NTC generally applied two principal criteria to select the FFLs: data on sales practices by FFLs, such as volume and multiple handgun sales; and time-to-crime for guns traced to that FFL.54 In FY 2002, ATF Headquarters assigned Focused Inspections on about 350 FFLs, or about 16 per Field Division.
Although the ATF's criteria for selecting FFLs for Focused Inspections targeted those FFLs that most significantly exhibited the established indicators of trafficking (i.e., multiple sales and short time-to-crime), the ATF did not identify or inspect all FFLs that exhibited indicators of trafficking. Instead, the NTC Director told us that the ATF established a numerical goal for Focused Inspections based on the resources available, and then changed the criteria to limit the number of FFLs identified to that number (350 in FY 2002). Therefore, the number of FFLs that exhibited indicators of trafficking and therefore should have been inspected was more than the 350 identified.
Gun Tracing Has Significant Shortcomings That Limit Its Use For Identifying FFLs That Should Be Inspected
Our review found that the crime gun trace data relied upon by the ATF to target inspections has significant limitations that reduce its effectiveness for identifying FFLs likely to be involved in firearms trafficking. Along with other information, firearms trace data can assist the ATF in identifying those FFLs that may be violating federal firearms laws and should be inspected. Although a trace does not in itself prove that an FFL is involved in gun trafficking, a high number of short time-to-crime sales and other patterns can indicate that an inspection is warranted. In FYs 2001, 2002, and 2003, LEAs submitted an average of a quarter million trace requests to the ATF each year. Table 1 shows the trace requests submitted by the 60 YCGII cities and other LEAs.
|Table 1: Trace Requests Submitted to the ATF|
|Fiscal Year||Total Firearms Trace Requests||Non-YCGII trace requests||YCGII trace requests|
|2001||232,272||133,962 (58%)||98,310 (42%)|
|2002||240,651||144,300 (60%)||96,351 (40%)|
|2003||280,947||(not available)||(not available)|
|Source: National Tracing Center|
As an indicator of whether the ATF overall used that data to direct its resources at FFLs that should be inspected, we examined whether the Field Divisions that had more gun traces in FY 2001 conducted more compliance inspections in FY 2002. The ATF was unable to provide data on traces to FFLs in each Field Division, but was able to provide us with the number of traces submitted by LEAs in each Field Division. ATF data shows that the preponderance of crime guns are recovered in the same geographic area in which they were originally sold by an FFL. Therefore, although we recognized limitations in the data, we expected that, if trace data were being used to target inspections at those FFLs that exhibited indicators of firearms trafficking, then higher numbers of trace requests would lead to more inspections in a Field Division. However, we found little correlation between the number of traces and the number of compliance inspections conducted the next year (Figure 5). This result indicated that the ATF overall did not focus its resources to conduct more inspections in those Field Divisions that had more crime guns traced.
|Figure 5: Comparison of Trace Requests to Inspections
[Not available electronically]
Although the ATF does not appear to be systematically conducting more inspections in Field Divisions with more gun traces, the Field Divisions may still use tracing data to select FFLs for those inspections that are not directed by Headquarters. For example, Area Supervisors, who determine FFL inspection assignments for their Area Offices, told us that they use trace data such as time-to-crime of two years or less, firearms with obliterated serial numbers, and multiple sales of handguns to target inspections.
There are limitations to more extensive use of trace data to target inspections. In response to our survey and in interviews, ATF Inspectors, Area Supervisors, and NTC management explained that tracing data is not fully useful for identifying potential problem FFLs. There are three limitations that reduce the utility of tracing data for targeting inspections. First, all LEAs do not trace all crime guns. Consequently, trace data are skewed toward dealers located in and around cities that participate in the YCGII program. As shown in Table 1 on page 24, about 40 percent of trace requests originate from the 60 YCGII cities, even though only about 15.5 percent of the U.S. population lives in those cities.55 Therefore, FFLs located in areas where the LEAs do not comprehensively trace crime guns are less likely to have guns traced to them. A Deputy Assistant Director told us that, because of this data limitation, at least three firearms "trafficking corridors" (i.e., routes along which guns from an area with lax firearms laws are illegally transported to an area with stringent firearms laws) are not picked up by tracing data.
Second, because trace data are not controlled for dealer sales volume, an elevated number of traces is not always evidence that an FFL is involved in firearms trafficking. Large-scale FFLs may have more guns traced to them simply because they sell more guns than smaller FFLs. As described to us by ATF Inspectors, a hypothetical FFL selling 40 guns a year of which 8 are subsequently traced as crime guns would be a much greater concern than a dealer selling 2,000 guns of which 15 are subsequently traced. However, the ATF's current database does not include sales volume, and therefore the ATF cannot easily use trace data to identify FFLs that have a high number of traces for their sales volume. Consequently, the smaller dealer in the example above would be less likely to be identified for inspection through tracing, despite the fact that 20 percent of the guns it sold became crime guns.
Third, tracing follows a particular firearm's trail from the manufacturer, through the FFL, to the initial purchaser of the firearm. As an Area Supervisor stated, a firearm may have been purchased later by another FFL (such as a pawnshop), "possibly two or three times, before it is actually involved in a crime and traced. The subsequent FFLs do not appear in the trace report."
Although correcting all of the above limitations would be difficult, more LEAs could be induced to comprehensively trace firearms if such tracing were made a condition of relevant Department of Justice grants. The ATF also could collect sales volume information (but not any personal information) during FFL inspections and enter that data into the NTC database.
A recent study indicated that some gun dealers are willing to sell guns to prohibited persons. The need to better identify the potential universe of dealers involved in gun trafficking and focus inspections on those dealers was highlighted in a 2003 study by the University of California, Los Angeles, School of Public Health. That independent study found that up to 20 percent of gun dealers exhibited a willingness to help likely prohibited persons obtain guns through straw purchases.56 The study used an independent telephone survey to test whether randomly selected FFLs were willing to sell a handgun, regardless of the intended recipient of the firearm. A total of 120 FFLs were surveyed. Of those, 87.5 percent said that they would sell a handgun to someone when the caller stated that the handgun was for his or herself, and 72.5 percent said that they would sell a handgun to someone when the caller stated that the handgun was intended to be a gift, both situations in which the sale would be legal. However, 52.5 percent also said that they would sell a handgun to the caller when told the gun was for a boyfriend or girlfriend "who needs it," a situation in which the FFL should have questioned the legality of the sale.
To address concerns that the FFLs may have been "playing along" with the interviewers, 20 additional calls were made after the initial study was completed. In these calls, the interviewer opened with "My girl/boyfriend needs me to buy her/him a handgun because s/he isn't allowed to." In 16 of the 20 calls, the dealers correctly informed the callers that they would not sell a firearm to them. However, four agreed to sell a handgun, even though they appeared to recognize that the sale would be illegal. Some even offered advice on avoiding the restrictions, stating for example:
In December 2003 the ATF's Firearms Programs Division randomly selected 760 FFLs for compliance inspections to identify FFL business characteristics that could be used to more effectively target FFLs for compliance inspections. By analyzing FFL business characteristics, such as number of employees and types of firearms sold, the ATF plans on developing a "risk model" based on the types and levels of firearms violations found at the selected FFLs.58 According to the ATF, the "risk model" will identify those FFLs most likely to be violating federal firearms laws. According to the Chief of the Firearms Programs Division, the ATF may, in the future, modify the FFL Renewal Application form to capture those factors identified by the risk model.
In sum, we disagree with the ATF's practice of limiting its identification of potential traffickers to the number that can be addressed within available resources. Unless the ATF identifies the full universe of FFLs exhibiting indications of firearms trafficking, it is unable to properly manage its range of enforcement activities, or inform the Attorney General, Congress, and the public of the scope of the potential trafficking problem at FFLs.
ATF's Field Divisions Implement FFL Inspections Inconsistently
During our review of FFL inspection files, we found that the process and amount of time spent conducting application and compliance inspections varied greatly among the ATF Field Divisions. For the 8,123 application inspections and 4,581 compliance inspections conducted by the ATF in FY 2002, the application inspections took an average of 11.8 hours and the compliance inspections took an average of 35.3 hours each. However, we found considerable variation in the average inspection time among the ATF's 23 Field Divisions. Time and workload data from the ATF's N-Spect database showed that the average time that Field Divisions spent conducting each application inspection ranged from 6.2 hours in the Kansas City Field Division to 25.5 hours in the Miami Field Division. For compliance inspections, the average inspection length ranged from 24.5 hours per inspection in the Nashville Field Division to 90.0 hours in the Washington Field Division. According to ATF Headquarters officials, the variance was due to Inspectors' discretion in determining the appropriate amount of time to spend examining FFL records. ATF officials also told us that inspection times were affected by the number of inexperienced Inspectors in the Field Divisions.
Our review of the ATF's Inspector Handbook confirmed that Inspectors have the latitude to abbreviate significant portions of a compliance inspection.59 According to the Inspector Handbook, Inspectors must complete 11 worksheets totaling 23 pages during an inspection. Although the Work Program and worksheets provide extensive direction for data collection, Inspectors are allowed to reduce specific inspection steps. For example:
Case Study: Gun Dealer Avoided Restrictions on Sales To Prohibited Persons
In June 2002, an ATF inspection was conducted at an FFL in the Seattle Field Division based on information from Special Agents at another Field Division. The inspection found that the FFL had been illegally conducting business at Nevada gun shows in violation of the Gun Control Act. The FFL admitted to inspectors that he had illegally transferred 52 firearms at the gun shows, and that he altered his A&D Book to hide the sales. Furthermore, the FFL admitted to having worked with two out-of-state gun show dealers to allow for firearms sales to be conducted without NICS checks. The FFL told inspectors this was done "because [the FFL] needed the money for...family." As a result of the inspection, the ATF revoked the FFL's license.
Source: ATF case files
Our discussions with Inspectors during our site visits also confirmed that Inspectors vary in the depth of their reviews. For example, all 18 Inspectors we interviewed said that they review Forms 4473 as part of the inspection process, but 14 said that they only examine the forms to see if they were filled out properly - not for indications that a purchaser may be part of a firearms trafficking ring or acting as a straw purchaser for someone else (e.g., purchasing patterns, similarities in purchasers' addresses). These 14 Inspectors said that they do not review Forms 4473 for firearms trafficking indicators because it was not a required part of the ATF's inspection process.
Although our interviews and review of the ATF's Inspector Handbook confirmed that Inspectors and their supervisors have the discretion to vary the extent of inspections, we also analyzed ATF performance data to identify any other factors that could be causing the disparity between Field Divisions. Our analysis of ATF data regarding application inspections found that the average time spent conducting application inspections was clearly correlated to the Field Divisions' staffing levels and the number of FFLs located in the Divisions. ATF data showed that Field Division staffing ranged from a high of 35 Inspectors in the San Francisco Field Division to a low of 9 Inspectors in the Baltimore Field Division. The number of FFLs in each Field Division ranged from 8,194 FFLs in the Kansas City Field Division to 1,172 FFLs in the Miami Field Division. However, the ATF had not distributed its Inspector resources among the Field Divisions to match the distribution of FFLs, resulting in significant workload imbalances among the Field Divisions. As shown in Figure 6, the Field Division that had higher numbers of FFLs to oversee with each Inspector spent less time conducting each application inspection.
|Figure 6: Comparison of Length of Application Inspections to FFLs per Inspector
[Not available electronically]
We also examined the ATF's explanation that compliance inspection times were affected by the number of inexperienced Inspectors assigned to each Field Division.61 As explained by ATF officials, inexperienced Inspectors receive on-the-job training from more experienced Inspectors. During that time, they observe or are observed by the experienced Inspector. Since inexperienced Inspectors are not fully productive, the average inspection time for Field Divisions with more inexperienced Inspectors could be higher. However, as shown in Figure 7, we found there was no correlation between the average time that a Field Division took to conduct compliance inspections and the percentage of inexperienced Inspectors among the Field Division's staff.
|Figure 7: Impact of Inexperienced Inspectors on Average Compliance Inspection Hours
[Not available electronically]
We also examined several performance indicators to see if the differing inspection times were related to outcomes, such as violations found and criminal referrals. As shown in Figure 8, we found that Field Divisions with longer average compliance inspection times found more instances of violations on each inspection. However, it was unclear whether the longer inspection times resulted from the need to document more violations, or if the greater number of violation instances found resulted from longer inspections.
|Figure 8: Comparison of Average Hours and Average Violation Instances for Compliance Inspections
[Not available electronically]
To examine whether the variation in average inspection times was due to some Field Divisions taking more time to document noncompliance by FFLs in order to pursue adverse actions, we analyzed whether Field Divisions that took longer to conduct compliance inspections in FY 2002 also took more adverse actions. We found that Field Divisions that had longer average inspection times took slightly fewer adverse actions than Field Divisions that took less time to conduct inspections. Figure 9 shows the average inspection time and number of adverse actions taken for each ATF Field Division.
|Figure 9: Comparison of Average Compliance Inspection Time to Adverse Actions, FY 2002
[Not available electronically]
Our analysis of the ATF Field Divisions' inspection efforts and outcomes also uncovered significant variances in inspection productivity. For example, our analysis of the ATF's FY 2002 workload and performance data found such variances as:
Inspection hours, violations, and adverse actions taken in FY 2002 are summarized by Field Division in Table 2, on the next page.
The variability in FFL inspections between ATF Field Divisions indicates that inspections are not being conducted according to standardized inspection procedures. Moreover, the excessive variability allowed inefficient and ineffective operations to persist. The ATF's failure to use the limited available Inspector resources efficiently also reduced its capability to carry out regular inspections of all FFLs, which, in turn, reduced the effectiveness of the ATF's FFL inspection program for ensuring FFLs comply with federal firearms laws. Further, the lack of standardization resulted in inconsistent treatment of FFLs in Field Divisions.
We believe that a standardized inspection approach is needed to ensure that FFLs in all Field Divisions are inspected using consistent inspection procedures and sampling criteria. Adopting a standardized inspection process designed to use the minimum review necessary to effectively gauge FFLs adherence to gun laws also would increase the efficiency of the FFL inspection program, and better enable the ATF to provide uniform, regular inspection coverage of the FFL population. An increase in inspection efficiency also would reduce the overall time that Inspectors spend at the FFLs' place of business.
|Table 2 - Inspector Hours, Productivity, Violation Instances Found, and Adverse Actions Taken, FY 2002|
|Field Division||Inspectors||FFLs per Inspector||FFLs Within Field Division||Total Inspector Hours (N-Spect Data)||Application Inspections||Compliance Inspections||Inspections per Inspector||Average Hours per Application Inspection|
|Field Division||Average Hours per Compliance Inspection||Inspections with Violations||% of Inspections Finding Violations||Violations Instances Found||Violation Instances per Inspection||Inspection Hours to Find Each Violation Instance||Warning Letters||Warning Conferences||Revocations|
Note: FFL and Inspector data are as of 08/01/03, other data are totals for FY 2002.
Suspected Criminal Violations Are Not Always Referred for Investigation
In addition to documenting regulatory violations, the FFL inspection program also identifies indications of potential criminal activity. Paradoxically, we found that Field Divisions that took longer to conduct inspections made slightly fewer referrals of suspected criminal activity (Figure 10). To analyze this result, we examined the total number of inspections completed by each Field Division and found that Field Divisions that took longer to conduct inspections completed fewer inspections. Further, the data showed that the number of inspections completed was correlated to the number of referrals made, and that taking longer to conduct an inspection made it no more likely that suspected criminal activity would be found and referred. Therefore, Field Divisions that took longer to conduct inspections completed fewer inspections and made fewer criminal referrals.
|Figure 10: Comparison of Average Compliance Inspection Time to Criminal Referrals, FY 2002
[Not available electronically]
In addition, we found evidence that the coordination between ATF Inspectors and ATF Special Agents could be improved. During our review, we determined that even when FFL compliance inspections identify significant violations of federal firearms laws by the FFLs or by gun purchasers, the violations are often not reported to ATF Special Agents for investigation. In our interviews, 12 of 18 Inspectors said that they rarely refer information gathered during FFL inspections to Special Agents because they did not believe that Special Agents would follow-up on the information. The other six Inspectors told us that they made one or two referrals per year to Special Agents.
We identified several cases where indications of potential criminal violations by FFLs, including gun trafficking, were identified but not referred for investigation. These FFLs were subsequently investigated after the illegal activity was discovered through other means. For example, Inspectors conducted compliance inspections in March 2000 and October 2002 on an FFL located in the southern United States. The 2000 inspection was based, in part, on information from a state LEA. During the 2002 compliance inspection, the Inspector found 40 firearms not entered into the FFL's A&D Book, several missing Forms 4473, and sales to out-of-state residents - all strong indicators of gun trafficking. Despite these findings, the FFL was not reported to Special Agents for investigation. Then, subsequent to the inspections, information from a confidential informant led to an investigation of this FFL. In December 2003, ATF Special Agents arrested the FFL for trafficking firearms.
Further, the latest data available indicate that investigations are not frequently initiated as a result of information provided by ATF Inspectors. In FY 2002 and FY 2003 respectively, Inspectors made 951 and 823 referrals of potential criminal activity identified during compliance inspections, an average of two per field-level Inspector. According to the ATF's Following the Gun report, published in 2000, of the 1,530 firearms trafficking investigations conducted from July 1996 to December 1998, just 43 - less than 3 percent - were initiated based on information found during inspections (see Table 3 on next page). Our survey of 45 Area Supervisors also found that few inspections result from information referred to Inspectors by ATF Special Agents. Most Area Supervisors (71.1 percent) said that they assigned five or fewer inspections each year based on information from Special Agents.
In December 2002, the ATF attempted to improve coordination between Inspectors and Special Agents by creating 23 Special Intelligence Inspector (SII) positions (one in each Field Division). The SIIs are responsible for collecting and disseminating information gathered by Inspectors and Special Agents between the two groups. As of November 2003, only 7 of the 23 authorized SII positions had been filled. Although ATF Headquarters officials said that they plan to fill the positions, as of March 2004 no deadline had been set.
The ATF Acts Infrequently to Revoke Federal Firearms Licenses, and the Process is Not Timely
We found that the ATF rarely revokes federal firearms licenses. In FY 2002, the 1,934 FFL inspections that found violations found an average of almost 70 violations each (for a total of 134,832 violations). In FY 2003, the 1,812 inspections that found violations found an average of over 80 violations each (for a total of 149,396 violations). However, in those years, the ATF issued Initial Notices of Revocation to only 30 and 54 FFLs, respectively.62 In addition to issuing a Notice of Revocation after a compliance inspection, the ATF also can effectively revoke an FFL's license by denying its request for license renewal. If an FFL's license expires during the course of revocation proceedings, the ATF's action is formally categorized as a denial of a renewal request - not as a revocation. In FY 2001, the ATF denied 28 requests for renewal.63
During our review, ATF officials told us that, before May 2003, the decision on whether to take adverse action (i.e., to revoke or deny renewal of an FFL's license) was left to the discretion of the Inspector, Area Supervisor, and the DIO of each of the 23 Field Divisions. However, in May 2003, subsequent to the initiation of our review, the ATF created guidelines for the Field Divisions to follow when determining whether or not to initiate an adverse action. When asked about the impact of these new guidelines, ATF personnel in the four Field Divisions we visited told us that they expected the guidelines to lead to an increase in the total number of license revocations. We found that after the ATF issued the adverse action guidelines, the number of FFL revocation hearings rose to 87 during FY 2003, and 59 Initial Notices of Revocation were issued and renewals denied during the first quarter of FY 2004 alone. As of March 2004, most of these cases have not been finalized.
The process for revoking or denying renewal of a federal firearms license is not timely. The ATF provided specific case tracking data for 50 closed denials and revocations completed in FY 2001 and FY 2002.64 The processing of these 50 cases averaged 379 days from the date that the Inspector recommended revocation or denial to the date that the case was closed by the NLC. We determined that the length of denial and revocation proceedings was due, in part, to the number of ATF officials involved (see Figure 11 on page 42). At least five ATF officials participate in these proceedings and each official reviews and approves the FFL inspection case file seriatim. A formal ATF Notice of Revocation is issued only after all of the officials have approved the action. Because the ATF has limited suspension and fining authority, FFLs remain in business during the adjudication of renewal denials and revocations.
During our interviews, DIOs and some Area Supervisors stated that most delays in the eight-step process occurred at the Division Counsel level as the ATF waited for Division Counsel to draft a Notice of Revocation for the FFL. We were unable to evaluate that perception from the case tracking data that ATF could provide, because it did not include the dates that the cases were processed by individual Field Division offices. However, Assistant Chief Counsels and Division Counsels we interviewed acknowledged delays in denial and revocation proceedings, which they stated were due, in part, to their heavy caseloads. They also stated that the quality of the initial compliance inspection report was a factor, as not all cases they received adequately detailed that the FFL "knowingly and willfully" violated federal firearms laws. This caused further delay while the Division Counsels obtained clarifying information from the Inspectors to ensure that the cases met legal standards.
We also noted that in some cases delays occurred due to a lack of legal staff within the Field Divisions. Although in 1999 the ATF established a standard Field Division structure that would include two staff attorneys, two of the Field Divisions we visited (Washington, D.C. Field Division, Seattle Field Division) had no attorneys on staff. When those Field Divisions needed legal assistance, they obtained it directly from their regional Assistant Chief Counsel's Office. For example, at the Washington, D.C., Field Division, the DIO told us that he sent revocation and denial cases to the Assistant Chief Counsel's Office located in Philadelphia.65 In one case, he said, it took four months to prepare an Initial Notice of Revocation for his signature.
Figure 11: FFL Denial or Revocation Process
We also found indications of limited communication between Field Division staff and Division Counsels. For example, we were told that Inspectors, Area Supervisors, and DIOs, usually do not seek advice from Division Counsels on inspections of FFLs found to have violated federal firearms laws until they request a Notice of Revocation or Denial. Furthermore, Division Counsels told us that they are not routinely notified when Warning Conferences are scheduled between FFLs and DIOs. One Assistant Chief Counsel told us that he believes that the ATF would benefit if Division Counsels had the opportunity to participate in such proceedings in case, later on, a Notice of Revocation is issued to the FFL.
New ATF Guidelines Begin to Address Inconsistent and Untimely Adverse Actions
In May 2003, the ATF took the initial steps toward standardizing adverse actions by issuing guidelines that were intended to ensure that Field Division personnel make more consistent and appropriate determinations on adverse actions (such as warning letters, warning conferences, and revocations) when FFLs are found to have violated federal firearms laws. The guidelines established standards for minimum adverse actions to be taken when violations are found. Although Field Divisions can deviate from the policy if they determine that the facts warrant another action, any deviation on inspections violations for which the standard action is revocation must be reviewed by Headquarters.
The guidelines specify that FFLs that commit minor non-repetitive, non willful violations that do not affect the lawfulness of a gun transfer, such as minor omissions or format errors, should receive warning letters or reports of violations. FFLs committing more serious violations that do not rise to the level of revocation, such as failing to record an acquisition within a specified time frame or failing to report multiple sales, should receive warning conferences. When FFLs commit the most serious violations, such as having more than a threshold number of guns missing from their inventory, the Field Division should begin the process of revoking or denying the renewal of the FFL's license.
The guidelines also address some of the problems we noted with the ATF's past failure to follow-up and take action when violations were found. For example, they address the frequent failure to routinely re-inspect FFLs found to have committed serious violations by directing that all FFLs that receive warning letters or warning conferences must be scheduled for a follow-up "recall" inspection in the following year. The guidelines also begin to address the lack of adverse actions taken against repeat violators by directing that adverse actions must escalate for repeat offenses. For example, an FFL that was issued a warning letter or directed to attend a warning conference for a violation cannot be given the same penalty if a subsequent inspection discovers further violations. Instead, the penalty must escalate to a warning conference (from a warning letter) or to revocation. The guidelines also establish a time frame for part of the adverse action process by directing DIOs to act on adverse action recommendations within 90 days after receiving the inspection report.
In addition to addressing the adverse action process, the guidelines also address part of the inspection process by requiring that follow-up inspections on FFLs that had unresolved inventory discrepancies include a full inventory, unless the DIO approves a statistical sample instead. Conducting a full inventory identifies all the guns that the FFL may have lost, and it brings the FFL's inventory records up-to-date. The FFL then formally reports any missing guns to the NTC as lost or stolen. The reporting of lost and stolen guns provides some benefit should one of those guns be recovered and traced, since it saves the NTC from contacting the manufacturer and dealer before the NTC knows that the gun was lost from a specific FFL's inventory. A full inventory may also identify more instances of violations to support potential adverse actions.
However, for the purpose of verifying that FFLs are complying with the requirement to maintain an accurate inventory, a policy of conducting full inventories in lieu of valid statistical samples may not be the most effective use of ATF resources. Conducting full inventories at larger gun dealers can be very time consuming. Moreover, once an inspection identifies discrepancies sufficient to document that an FFL's inventory system is deficient, completing a full inventory provides only incremental instances of missing guns. We would not disagree that there are cases, often related to an investigation, where a full inventory is desirable - cases like the Bull's Eye Shooter Supply.66 However, maintaining an accurate inventory remains the responsibility of the FFL. Given the limited resources available to the ATF to conduct gun shop inspections, restricting inventories to the minimum sample needed to provide a statistically valid check on the accuracy of the FFL's record-keeping system would reduce the length of inspections and enable the ATF to provide better coverage of FFLs.
By Streamlining and Standardizing Inspections, the ATF Could Dramatically Improve the FFL Inspection Program
We found that there is a critical need for the ATF to improve the efficiency and effectiveness of the FFL inspection program and ensure that FFLs are inspected using consistent inspection procedures and sampling criteria regardless of their geographic location. With the May 2003 guidelines, the ATF began to improve the consistency and timeliness of adverse actions, as well as to address the critical need to re-inspect FFLs that committed violations. However, the current variability in the Field Divisions' inspection procedures must be addressed to ensure that adverse actions taken under the May 2003 guidelines treat FFLs consistently. Requiring defined adverse actions for specific numbers of violation instances in the absence of standardized inspection procedures and sampling criteria will result in dissimilar treatment of FFLs in different Field Divisions. As discussed previously, the ATF's guidance on conducting inspections does not ensure consistent examinations of FFLs' compliance with gun laws.
We observed several areas in which the inspection process could be improved through standardization, such as:
Improving the efficiency of the inspection process would also reduce the need for additional staff. Adopting a standardized inspection process designed to use the minimum review necessary to effectively gauge FFLs' adherence to gun laws will increase the efficiency of the FFL inspection program. This will reduce both the need for additional Inspector staffing and the burden that inspections place on FFLs. In an April 2003 report to Congress, the ATF Director stated that, to fully implement the ATF's mission to regulate and enforce federal firearms and explosives laws, the ATF would need 1,775 Inspectors, an increase of 1,277 Inspectors from current staffing levels.67 Of the 1,775 Inspectors, 1,235 would be dedicated to conducting FFL compliance inspections in order to inspect all FFLs on a triennial basis. The Director based this figure, in part, on projections of FFL application and compliance inspections, as well as inspections initiated to support criminal investigations. Our review of FY 2002 ATF work hour data shows that it dedicated 628,117 staff hours (the equivalent of 302 staff years at 2,080 hours per year) to FFL inspections. The requested increase to 1,235 Inspectors would require 933 new Inspectors, and would more than quadruple the Inspector workforce dedicated to inspecting FFLs.68 In the Department's FY 2005 Budget Request, it recommended funding 126 new Inspector positions for the ATF.
We examined the ATF's calculations and question whether it needs as many Inspectors as stated. The ATF projection was calculated using an overall average inspection time of 62.7 hours per inspection (Table 4, on the next page). However, according to the ATF's FY 2002 data, the overall average inspection time was only 49.4 hours (628,117 inspection hours divided by 12,704 total inspections). The ATF request therefore appears to include an assumed 27 percent increase in the average length of inspections.
Calculating the staffing requirement using the ATF's actual historical inspection average of about 50 hours per inspection indicates that the ATF should need a total of only 984 Inspectors (682 new Inspectors) to accomplish inspections on a triennial basis. Moreover, as previously discussed, our analysis found wide variations in inspection implementation and productivity among the ATF Field Divisions. By standardizing and streamlining its inspection process the ATF could reduce the average inspection time from the current 49.4 hours, which would further reduce the number of additional Inspectors that needed to accomplish FFL compliance inspections on a triennial basis.69 For example, by reducing the overall average inspection time to 40 hours per inspection, the ATF should be able to implement a triennial FFL compliance inspection program with 788 Inspectors (Table 4).
|Table 4: ATF STAFFING REQUEST CALCULATIONS|
|Activity||Number of |
|Hours per |
|Total Hours||Inspector FTE |
@ 2080 Hours
|Non-YCGII FFL Compliance Inspections||22,889||66||1,510,674||726.3|
|YCGII FFL Compliance Inspections||11,444||80||915,520||440.2|
|Support Criminal Investigations||631||73||46,063||22.1|
|OIG RECALCULATION: ATF STAFFING AT 50 HOURS PER INSPECTION|
|Non-YCGII FFL Compliance Inspections||22,889||51.5||1,178,784||566.7|
|YCGII FFL Compliance Inspections||11,444||63.5||726,752||349.4|
|Support Criminal Investigations||631||73||46,063||22.1|
|OIG RECALCULATION: ATF STAFFING AT 40 HOURS PER INSPECTION|
|Non-YCGII FFL Compliance Inspections||22,889||40||915,560||440.2|
|YCGII FFL Compliance Inspections||11,444||50.8||581,360||279.5|
|Support Criminal Investigations||631||73||46,063||22.1|
Increasing the efficiency of the inspection process also is needed because the demands on ATF Inspectors to perform duties related to explosives licensees are increasing. Immediately after the terrorist attacks of September 11, 2001, the ATF initiated a policy of investigating all incidents of theft or loss of explosives materials, and conducting compliance inspections on all explosives license holders within 50 miles of major metropolitan areas. This meant that the ATF had to inspect 7,459 of 9,400 explosives license holders. In November 2002, the Safe Explosives Act (SEA) imposed new licensing requirements that increased the number of explosives licensees, and mandated that the ATF conduct on-site inspections of explosives licensees and permit holders at least once every three years.
The additional explosives work is already reducing ATF's ability to inspect FFLs. The Chief of Staff of the ATF's Firearms, Explosives and Arson Directorate confirmed that Inspector resources have been diverted to explosives work by Area Supervisors to meet the SEA's inspection requirements. As a result, she said, the ATF plans to re-examine Inspectors' firearms work because ATF Headquarters officials are "worried" that FFLs are not being inspected. Our review of preliminary data for FY 2004 inspections work confirmed these concerns. Preliminary data for early FY 2004 indicated that there has been a precipitous decrease in the number of FFL inspections. Through the first five months of FY 2004, the ATF completed 1,113 FFL compliance inspections. At that pace, the agency will complete less than 2,700 FFL compliance inspections during FY 2004. That is less than half the number that the agency reported that it completed in FY 2003, and less than 2.6 percent of the FFL population. For example, the Kansas City Field Division, which oversees the most FFLs of any Field Division (with 8,194 FFLs) completed only 21 FFL compliance inspections in the first five months of FY 2004.
Reducing the time spent at FFLs' places of business. The time that ATF Inspectors spend on-site at FFLs cannot be calculated definitively. Although the ATF's inspection hour tracking system shows that the ATF spent a total of 257,723 hours conducting FFL inspections in FY 2002, it does not allow the time spent on travel and other inspection related actions performed away from FFL locations to be segregated from the total inspection hours. Nonetheless, any increase in inspection efficiency would reduce the overall time spent conducting reviews on-site at the FFLs, and minimize any potential interruption of the FFLs' business operations.
ATF Does Not Consistently Report Inspection Performance
In response to our requests for inspections and workload data (such as the number of compliance and application inspections conducted and the Inspector work hours associated with completing these inspections during FY 2002), the ATF queried its N-Spect, FLS, and Standard Time and Attendance System (STATS) electronic databases. N Spect tracks direct time related to FFL inspections, FLS tracks information related to FFL licensees, and the STATS timekeeping system tracks direct and indirect hours for payroll purposes. During our examination of the performance and productivity of the ATF's FFL inspections program, we identified significant discrepancies between the systems with regard to the number and type of inspections conducted and the hours spent conducting the inspections. In addition, the data in the systems regarding the number of inspections done differed significantly from what the ATF included in published reports. Finally, data contained in the N Spect database contained significant errors: after extracting data to respond to our requests, ATF officials determined that several hundred inspections entered as compliance inspections were actually application inspections.
Table 5 contains examples of inconsistent data related to the ATF's FY 2002 ATF's FFL inspections program that the ATF provided to OIG and reported publicly.
|Table 5: Inconsistent N-Spect Totals of FY 2002 FFL Inspection Activities|
|Data Source||FY 2002 Totals||Inspection Type|
|N-Spect data provided to the OIG in May 2003||12,522||Total Inspections (the ATF claimed data could not be broken down by inspection type)|
|Revised N-Spect data provided to the OIG in February 2004||7,665
|Revised N-Spect data adjusted to correct miscoded inspections provided to the OIG Team on May 20, 2004||8,123
|FLS Licensee Inspection Data||4,722||FY 2002 new licensees with "inspection date" indicating that an application inspection was conducted|
|ATF Snapshot 2003||"Approximately 6,000"||Compliance Inspections|
|ATF Performance and Accountability Report 2002||"Eleven percent" of all FFLs were inspected.
11% of 104,300 FFLs is equivalent to 11,473 inspections
In addition to FY 2002 data discrepancies, current data from ATF's N-Spect database shows that the ATF conducted 5,729 application inspections and 4,035 compliance inspections in FY 2001, as well as 8,043 application inspections and 5,887 compliance inspections in FY 2003. Previously, the ATF had published reports and provided draft reports to the OIG indicating that it conducted more inspections: 5,497 application inspections and 5,016 compliance inspections in FY 2001 and 8,422 application inspections and 6,481 compliance inspections in FY 2003.
We discussed these substantial inconsistencies with ATF Headquarters officials. They explained that the inconsistencies were due to differences in how the queries of the N-Spect electronic database were constructed by different ATF analysts. For instance, according to the ATF Deputy Assistant Director for Field Operations, the ATF report on FY 2002 activities (ATF Snapshot 2003) reported a higher number of inspections completed because it aggregated all Inspector activities to determine the number of compliance inspections conducted by ATF Inspectors, and included limited purpose inspections as well as activities unrelated to FFL compliance inspections. Another reason for the discrepancies related to the project codes used to identify types of inspections. According to ATF Headquarters officials, Area Supervisors do not always use the correct project codes when assigning inspections because the codes are "confusing." For our review, ATF Headquarters officials had to query an N-Spect "Special Instructions" data field for "firearms application inspection" and "firearms compliance inspection" rather than the appropriate project code in order to more accurately determine inspection totals.
The ATF has taken steps to ensure that inspection totals are accurately measured. In October 2003, the ATF released a new version of N-Spect that requires Area Supervisors to use pull-down menus that are inspection-specific (e.g., "Application Inspection"). Moreover, although the enhancements to the N Spect database will increase the reliability of the data in the system, the ATF must adopt a standard approach for querying the N-Spect electronic database and ensure that queries are stored for future use so that subsequent requests for the same data will elicit comparable results. If the ATF does not adopt a standard method for querying and extracting historical data, it cannot consistently report accurate performance data.
New Restriction on Retention of Gun Purchaser Data will Hinder the ATF's Ability to Detect Fraudulent Background Checks
As discussed in the Background section, prior to transferring a firearm to an unlicensed individual, an FFL must complete a check of the National Instant Criminal Background Check System (NICS) to determine if the potential purchaser is prohibited from owning a gun. For each query, the NICS currently collects information on whether or not the purchase was allowed to proceed, and retains information on the potential purchaser for 90 days. The FBI assigns each query a unique NICS Transaction Number (NTN), which FFLs are required to enter onto the corresponding Form 4473.
For the majority of dealers, NICS is a valuable tool that enables them to quickly determine whether a potential customer is prohibited from purchasing a firearm. However, some gun dealers could attempt to hide transfers to prohibited persons by falsifying NICS information. To deter fraud and detect FFLs that may be providing false information to pass a NICS check in order to facilitate a sale to a prohibited person, during FFL compliance inspections ATF Inspectors verify that the information submitted to NICS matches the information on the Form 4473. The Inspectors copy information on selected Forms 4473 from the past 90 days and check with the FBI to ensure that the information on the Form 4473 matches the information that the FFL provided at the time of the sale. If discrepancies are found, it may indicate than an FFL submitted false information to NICS in order to receive an NTN associated with a background check for a non-prohibited person. The ATF reported that it has not found any NICS violations involving the falsification of purchaser information through the Forms 4473 review.70
However, the ability of ATF Inspectors to conduct this Form 4473 review was affected when the FY 2004 appropriations act reduced the time that the FBI can retain information submitted by FFLs during the NICS check.71 Beginning in July 2004, all purchaser information (e.g., name, address, date of birth) on NICS queries for which firearms sales are approved will no longer be kept for 90 days; it must be destroyed within 24 hours of the official NICS response to the FFL.72 For approved sales, the FBI can retain for 90 days the NTN, the license number of the FFL that contacted NICS, and the date that the NICS query was made. After 90 days, the FBI may retain only the NTN and the date that the number was issued.
In 2002, ATF Headquarters officials suggested that the ATF's FFL compliance inspections program would not be affected by a "Next Day Destruction" provision.73 Instead of submitting information copied from Forms 4473 to NICS, ATF Headquarters officials said that Inspectors could "recheck" the eligibility of purchasers by requesting that the FBI rerun selected NICS checks taken from FFL records during compliance inspections. However, while resubmitting Form 4473 information to NICS will determine whether the purchaser would be approved on the date of the recheck, it does not enable the ATF to effectively detect that the FFL supplied inaccurate information to NICS. There are reasons other than FFL fraud for a prior approved purchase to fail during a recheck. For instance, the purchaser may have become ineligible only since the sale, or the FBI NICS operator on the original query may have transposed letters or numbers resulting in an erroneous approval. Given these and other possible explanations, the lack of information in NICS will make it much more difficult for the ATF to prove that FFLs supplied false information initially.
Moreover, the shortened retention time will make it much easier for corrupt FFLs to avoid detection. We identified at least two potential ways that the new restriction would make it easier for corrupt FFLs to falsify the NICS check to hide a knowing transfer of a gun to a prohibited person.
Given the new restriction on retaining NICS data, after July 2004 an ATF Inspector arriving on-site could only check the information on Forms 4473 filled out by the FFL that day. Therefore, the likelihood of encountering a falsified NICS check would be remote.
We concluded that the ATF's FFL inspection program did not consistently ensure that FFLs comply with federal firearms laws. The lack of standardized inspection procedures resulted in inconsistent inspections of FFLs and significant variation in the implementation of the inspection program by Field Divisions. Our review of N-Spect performance data found that ATF's Field Divisions took an average of 35.3 hours to conduct FFL inspections to detect noncompliance with federal firearms laws, and one Division took only 24.5 hours on average to conduct its compliance inspections. We found no operational reasons why some Field Divisions averaged much longer, up to 90 hours, to conduct compliance inspections. In fact, we found little or no correlation between inspection times and enforcement activities, such as referrals of suspected criminal activity and adverse actions taken. Further, our finding that the Field Divisions varied significantly in such productivity measures as number of inspections finding violations and number of inspections done by each Inspector argues strongly for implementation of a more standardized and efficient inspection regimen.
Because the ATF does not conduct regular inspections of FFLs and lacks adequate resources to meet agency goals, it cannot effectively monitor the overall level of FFL compliance with federal firearms laws. In December 2003, the ATF initiated a program to conduct special Random Sample Compliance Inspections to develop a risk model for the FFL inspection program. Using data from those inspections, the ATF planned to "be able to project the overall level of compliance by" gun dealers, pawnbrokers, and collectors.75 While the project to estimate the overall level of compliance with laws is needed to estimate the challenges facing the ATF, it does not take the place of regular compliance inspections for deterring and identifying noncompliance with gun laws.
To ensure that all FFLs are treated consistently, and that the FFL inspection program is as efficient as possible, the ATF needs to implement a policy to ensure that inspections are conducted in a uniform manner, that inspections procedures are limited to the minimum steps needed to accomplish a valid review, and that violations are processed in a uniform and appropriate manner. A consistent and timely inspection process is essential for identifying and addressing scofflaw dealers and reducing the availability of illegal firearms to criminals.
We recommend that the ATF: