Review of the Bureau of Alcohol, Tobacco, Firearms and Explosives’ Disciplinary System
Evaluation and Inspections Report I-2005-009
Office of the Inspector General
Employees did not consistently comply with the ATF’s misconduct reporting requirements.
ATF Order 2130.1 requires that “any allegation or information that the standards or rules of conduct have been violated must be promptly reported by the employee having the information” to the ATF Investigations Division.25 As an alternative, ATF Order 2130.1 states that employees can report misconduct directly to the OIG. Despite this reporting requirement, we found that ATF employees did not consistently report all allegations of misconduct to the Office of Professional Responsibility and Security Operations’ Investigations Division or to the OIG. Interviews with local management, ELRT specialists, and Investigations Division officials revealed that ATF employees often reported allegations of misconduct to their supervisors rather than directly to the Investigations Division and their supervisors did not consistently forward the complaints to the Investigations Division.
Specifically, the ATF could not document that 59 of the 230 employee misconduct cases that we reviewed were ever reported to the Investigations Division, as required. All 59 of these cases were handled by local management using the decentralized process. The 59 cases included such misconduct charges as insubordination, absence without leave, misuse of credit card, and misuse of position. The Investigations Division had no record of an incident report corresponding to any of these cases, nor did the disciplinary case files maintained by the ELRT document a report of the initial misconduct allegation to the Investigations Division.
Additionally, our employee survey confirmed that employees did not always report misconduct directly to the Investigations Division. Of the 421 employees who responded to our survey, 69 said that they were unaware of their obligation to report misconduct. Out of 345 employees who were aware of the obligation to report misconduct, only 110 stated that they had reported misconduct that they had witnessed. Some of these employees did not report it directly to the Investigations Division, as required. Only 35 of the 110 survey respondents who reported misconduct allegations (32 percent) properly reported the misconduct to the Investigations Division or the OIG, in accordance with ATF Order 2130.1. Chart 1 indicates where our survey respondents reported misconduct.
Chart 1: Survey Respondents’ Reporting of Misconduct Allegations
Although 68 of our survey respondents (62 percent) stated that they reported misconduct allegations directly to local management rather than the Investigations Division, local managers did not always report the allegations to the Investigations Division, as required. We were told in interviews and briefings that local management sometimes contacted the ELRT to discuss whether an allegation of misconduct could be handled through the decentralized process, without reporting the allegation to the Investigations Division. When this happened, ELRT specialists determined whether the misconduct allegation constituted more serious misconduct. If it did, the ELRT advised the manager to report the allegation to the Investigations Division so it could make a determination whether to conduct an investigation or refer it back to local management for action. If the ELRT did not believe the allegation met the criteria for more serious conduct, it would begin to provide advice to local management on how to proceed. The ELRT specialists did not report misconduct allegations they received from local management to the Investigations Division, nor did they require documentation verifying that the manager had reported the misconduct allegation to the Investigations Division.
The ATF did not comply with the OIG’s misconduct reporting requirements.
Because the ATF was not in compliance with its own reporting requirements, it also was not in compliance with the requirement to report misconduct to the OIG. Of the 76 decentralized process case files we reviewed, 58 misconduct cases occurred after the ATF transferred to the Department in January 2003 and, thus, were subject to the Department’s OIG requirement to report allegations of employee misconduct to the OIG. Our review of case files, incident reports, and OIG records showed that 47 of these 58 cases were not reported to the OIG in accordance with OIG policy.
An OIG memorandum dated March 26, 2003, OIG Investigative Procedures Relating to the ATF, requires that the ATF report allegations of employee misconduct to the OIG. Our review of the Investigations Division’s incident reports and OIG records showed that, for the time period reviewed, the Investigations Division reported all allegations that it received to the OIG. However, because local management sometimes reported misconduct allegations to the ELRT and not the Investigations Division as required, and the ELRT did not ensure that these allegations were reported to the Investigations Division or the OIG, the ATF could not ensure that it fully complied with the Department’s reporting requirement.
The ATF’s noncompliance with internal and OIG reporting requirements altered the way in which the ATF’s disciplinary system operated. The ATF’s failure to ensure that all allegations of misconduct were properly reported enabled local managers to determine whether the allegations were serious or minor, in contradiction of ATF Order 8610.1A that delegates this authority to the Investigations Division. This, in turn, caused six misconduct allegations in our case file review that would normally be investigated or monitored by the Investigations Division under OIG oversight to instead be handled through the decentralized process without any report to the Investigations Division or the OIG. The ATF’s noncompliance with reporting policies increases the likelihood that similar types of misconduct may be handled inconsistently through both disciplinary processes and, thus, may be subjected to different investigative and adjudicative standards. Additionally, the ATF’s failure to ensure proper reporting of misconduct allegations inhibited both the ATF’s and the OIG’s ability to compile, monitor, and report accurate statistics on ATF employee misconduct.
Figure 2, on the next page, diagrams how the ATF’s discipline system actually works, in comparison to Figure 1 on page 3, which diagrams how the system is supposed to work according to ATF regulations and OIG policy.
Investigations were not always thorough.
While we found that investigations conducted under the centralized process were generally thorough, investigations conducted under the decentralized process did not always exhibit sufficient evidence of a thorough investigation. The 76 decentralized process case files we reviewed did not contain sufficient evidence to show that the misconduct allegations were adequately investigated. In fact, 16 of the 76 case files that we reviewed contained no documentation at all to show that there had been any investigation into the alleged misconduct, even though discipline had been imposed. The remaining 60 decentralized process case files contained documentation of some informal evidence collection. The types of evidence maintained in these 60 case files varied widely. Some case files contained activity logs, time reports, or written guidance pertaining to the alleged misconduct. Others contained written statements from the employee being investigated that described the circumstances surrounding the alleged misconduct. One case file contained only a memorandum from the local manager who processed the case, describing a statement that he had obtained from the employee; the employee’s original statement was not included in the file. None of the decentralized process case files contained a report of investigation. Without a thorough investigation, the ATF may not have sufficient information to accurately determine whether misconduct actually occurred.
ELRT specialists routinely advise local managers on what types of evidence must be collected to support an allegation of misconduct. However, the ELRT specialist we interviewed stated that the ATF has no written standards or guidance for the decentralized investigative process. In fact, ATF Order 2750.1 states only that the Chief of the ELRT must review the proposal and “copies of all materials upon which the proposal is based” before discipline greater than 14 days’ suspension can be proposed. The Order also states that the ELRT must maintain in the disciplinary file a record of “any order effecting the action, together with any supporting material.” The Order does not, however, provide any guidance on how supporting materials should be collected or what types of materials are appropriate. Instead, each ELRT specialist determines the appropriate level of evidence individually. Based on the ELRT’s advice, local managers collect evidence and send it to the ELRT via facsimile or e-mail for inclusion in the disciplinary file. The ELRT does not track what types of evidence have been recommended or when evidence is received from local management.
In contrast to the investigations conducted by local managers, the OIG’s Investigations Division reviewed the investigations conducted by the ATF Investigations Division and concluded that they were thorough, objective, and well documented in reports of investigations. An OIG Special Agent reviewed 17 reports of investigations and found that the investigations conducted by the ATF Investigations Division were thorough. The OIG Special Agent concluded that the ATF investigators interviewed relevant witnesses and examined necessary documents, and that the investigative reports contained the information necessary to understand the actions taken during the investigations. The OIG Special Agent rated all 17 investigations reviewed as either very good or good.
Inadequate separation of responsibilities.
In the decentralized process, the same official was permitted to propose and decide discipline for misconduct cases that resulted in a penalty of a 14-day suspension or less. Specifically, ATF Order 1150.4 grants the authority to ATF officials to act as the proposing and deciding official for this category of cases. By contrast, the centralized process requires the Investigations Division to investigate misconduct allegations, the Professional Review Board to propose discipline, and the Bureau Deciding Official (or, previously, local managers) to decide discipline. Although ATF Order 1150.4 does not require local managers in decentralized cases to serve as both the proposing and deciding official on an individual misconduct case, the ELRT specialist we interviewed stated that, in her experience, local managers generally served in both capacities unless they raised a specific objection to the arrangement.
Of the 76 decentralized process case files we reviewed, 19 cases resulted in proposed suspensions of 14 or fewer days. In 13 of these 19 misconduct cases, the penalties were proposed and decided by the same individual, and grievances were filed in 4 of the 13 cases. The remaining 63 decentralized process misconduct cases, which did not involve a single individual serving as both the proposing and deciding official, resulted in only 3 grievances.
One of the grievances in a case in which discipline was proposed and decided by the same individual specifically alleged:
“My immediate supervisor began a campaign of retribution against me…. [He] tried to dissuade me from filing the Grievance [sic]…. I request[ed] that the [supervisor] be instructed to cease his personal vendetta against me.”
Another of these grievances expressed the employee’s concerns that the supervisor’s actions related to the misconduct case were “personally-motivated [sic] and vindictive.”
Our previous reviews of three other Department components’ disciplinary systems revealed that those components do not use this method of adjudicating misconduct.26 In the other components we reviewed, the proposing and deciding official responsibilities are assigned to separate individuals or entities regardless of the severity of the discipline. The practice of combining the responsibilities of collecting evidence, proposing, and deciding discipline removes the checks and balances and objectivity normally in place in a disciplinary process.
The ATF does not employ a consistent standard for considering prior misconduct when adjudicating misconduct cases.
We also found indications that the ATF did not employ a consistent, standard methodology for considering past misconduct when determining the appropriate application of more severe discipline for repeat offenses. For instance, we found one case in which the proposed penalty was based, in part, on misconduct that resulted in discipline 6 years prior to the current misconduct. Conversely, we found a case in which prior misconduct was not considered, although the employee had received discipline less than 2 years prior to the current misconduct. Our interviews with ATF officials confirmed that the determination to consider prior misconduct varied on a case-by-case basis, with no written standard to guide the decision. In contrast, our reviews of three other Department disciplinary systems revealed that prior misconduct cannot be considered if it occurred over 2 years ago.27
The ATF did not accurately record misconduct cases in its disciplinary database.
The ATF relies on its disciplinary database as its primary source of information for determining the consistency of penalties for similar misconduct.28 However, the ATF did not have sufficient internal controls to ensure that all information was entered or that the data was accurate.
We examined summary sheets from the disciplinary database that were included in the 230 case files we reviewed and found that incorrect or incomplete data had been entered in the database. Of the 230 case files, 134 files contained summary sheets from the disciplinary database. Of these 134 database summary sheets, 38 incorrectly identified whether misconduct had occurred. Additionally, we found that in 52 of the summary sheets, the charge or discipline recorded in the case file (specifically in the proposal and the decision letters) did not match the charge or discipline recorded in the disciplinary database. These inaccuracies prevent the ATF from successfully comparing similar misconduct cases to determine the consistency of proposed charges and penalties.
The ATF did not always adjudicate its misconduct cases consistently.
The charges that the ATF used to describe different types of misconduct were not always consistently applied. For example, 31 of the 230 case files that we reviewed included the charge of “poor judgment.” These 31 poor judgment charges encompassed a total of 22 different types of misconduct, as shown in Table 1.
Table 1: Poor Judgment Cases
Some of the types of misconduct cited as poor judgment in these cases, such as abuse of the Public Transportation Incentive Program, misuse of a government-owned vehicle, and failure to apply reimbursements to travel card expenditures, were also sometimes categorized as other, more specific charges. We found that the cases involving poor judgment charges sometimes appeared to result in more lenient discipline than the cases with more specific charges. For example, federal regulations require that government employees who willfully misuse a government-owned vehicle are subject to a mandatory 30-day suspension.29 The ATF has applied this mandatory penalty in cases involving the specific charge of “willful misuse of a government-owned vehicle.” Nevertheless, two “poor judgment” cases we reviewed involving misuse of a government-owned vehicle each resulted in only a 2-day suspension. Another similar vehicle misuse case classified as poor judgment resulted in a 4-day suspension.
Mitigation of penalties is not adequately documented.
During our review of misconduct case files, we found 38 misconduct cases in which the punishment was mitigated. Deciding officials may mitigate proposed penalties based on consideration of the Douglas Factors and any response provided by the employee. ATF Order 2750.1 describes what is to be documented in a decision letter, including a discussion of any mitigating factors.
In 21 of the 38 cases that involved mitigation of the proposed penalty, the reasons for the mitigation were not adequately documented in the decision letter. The mitigations in these cases ranged from the reduction of a proposed 1-day suspension to a letter of reprimand for a poor judgment charge, to the reduction of a proposed removal to a 7-day suspension for charges of submitting fraudulent travel vouchers, accepting per diem from both the ATF and another government agency, and engaging in unauthorized outside employment. Chart 2 below shows how much mitigation was applied in the cases that we found were not adequately explained.
Chart 2: Mitigations with Insufficient Explanation
The decision letters in these cases did not contain adequate explanations of why the punishment recommended by the proposing officials was mitigated. For example, some of the decision letters stated only that the Douglas Factors had been considered without giving explanation of which factors contributed to the mitigation and why. Other decision letters stated that the deciding official had considered the employee’s written or oral reply without citing specific reasons why the mitigation was appropriate. This occurred under both the centralized process (18 cases) and the decentralized process (3 cases).
We attempted to review all SF-50s documenting that the discipline in the 120 cases in our sample in which the employees were suspended, reduced in pay or grade, or removed was actually imposed.30 In 14 out of the 120 cases (12 percent), the ATF could not provide an SF-50 to verify that the punishment was implemented. The 14 cases for which the ATF could not provide SF-50s comprised 12 suspensions and 2 removals. The ATF could not adequately explain why no SF-50 was available for 13 of these cases. An ELRT specialist we interviewed confirmed that, at the time of the interview, no SF-50 had yet been initiated for 1 of these 14 cases. The decision letter for this case, imposing a 2-day suspension, was issued over 7 months before our interview with the ELRT specialist. According to the specialist, the local manager who decided the case through the decentralized process did not initiate an SF-50 to implement the suspension before leaving that field office. The ELRT specialist emphasized that, while the ELRT encourages managers to implement discipline in a timely manner, local management is ultimately responsible for initiating the SF-50s.
Because the ATF could not provide SF-50s in these cases, the ATF could not demonstrate that the employees’ pay had been withheld for the 12 cases involving suspensions. Although the ATF could not provide SF-50s for the two cases involving removals, we subsequently verified with ATF officials that the employees had in fact been removed from service.
The ATF has not implemented timeliness standards to measure its performance.
The ATF has not implemented any timeliness standards to measure the performance of its overall disciplinary system. Timeliness data for the centralized process is collected and recorded in a monthly status report generated by the Investigations Division.
However, timeliness data for the decentralized process is recorded in the disciplinary database only from the start of the adjudication and only for those cases in which discipline is proposed or imposed.31 We were unable to determine the timeliness of the investigations for this process because the ATF lacks any requirement that this type of information be uniformly collected and recorded by either field personnel or ELRT specialists who assist in decentralized process cases. As a result, the ATF cannot measure the timeliness or performance of its entire discipline system until it establishes formal data collection procedures and ensures that the procedures are followed.
In our three prior reviews of Department components’ discipline systems, we recommended that the components institute timeliness standards to measure system performance. All three components concurred with the recommendations. As a result, the components have either established timeliness standards or are in the process of testing, analyzing, and determining what those standards should be prior to formal implementation. The United States Marshals Service’s Office of Internal Affairs has established a standard of 90 days and the Drug Enforcement Administration’s Office of Professional Responsibility has established a standard of 180 days for completing investigations of employee misconduct. The Bureau of Prisons is in the process of formalizing investigations timeframes. The Bureau of Prisons and the Drug Enforcement Administration are also working on developing formal timeframes for adjudicating misconduct cases. The United States Marshals Service has formalized its adjudication timeframes.
ATF investigative and adjudicative efforts were within the range of other three Department components that we have reviewed.
Although the ATF does not require or track the timeliness of the discipline process, we calculated the timeliness of its disciplinary system based on our analysis of the 154 centralized process case files reviewed. On the average, it took the ATF 277 days to process a misconduct allegation – 112 days to investigate the allegation and 165 days to adjudicate the case. We were unable to incorporate the decentralized process cases into our calculations because the dates tracked in the decentralized process’s files were not comparable to the dates tracked by the centralized process.
The Investigations Division’s investigation times, based on 133 reports of investigation that we reviewed, are shown in Chart 3, below.
Chart 3: Processing Times for the Investigations Division’s Investigations
Some of the variation in the duration of the Investigations Division investigations may be attributed to the varying complexity of the misconduct investigations. Also, the time expended on each misconduct investigation may be affected by the Investigations Division’s prioritization of cases based on the severity of the offense and staff availability.
Our case file review indicated that the adjudication phase lasts, on average, 165 days. Under the centralized process, the proposal portion of the adjudication phase averaged 111 days from the completion of a report of investigation to the issuance of a proposal letter. The time expended to produce a proposal letter ranged from 32 days to 422 days. The average time expended to produce a proposal letter, based on our review of 99 proposal letters, is shown in Chart 4.
Chart 4: Processing Times for Centralized Process Proposals
Disciplinary decisions in the adjudication phase of the centralized process averaged 90 days to complete, following the issuance of a proposal letter. The ATF’s decision letter processing time ranged from 19 days to 394 days. Chart 5 shows the amount of time expended on 93 disciplinary decision letters we reviewed.
Chart 5: Processing Times for Centralized Process Decisions
Under the centralized process, letters of caution, admonishment, and reprimand, which are generally issued to employees without a proposal letter, averaged 119 days to produce, ranging from 56 days to 267 days. Additionally, our case file review showed that the ATF’s disciplinary system averaged 66 days to issue letters of clearance after an investigation is complete, ranging from 16 days to 145 days.
While neither the ATF nor the Department requires specific time standards, the Department encourages managers “to act in a timely manner” on misconduct cases.32 To determine whether the time taken by the ATF to process misconduct cases was within the range of the other three Department components we have reviewed, we compared the average amount of time the ATF took to investigate and adjudicate misconduct cases to the other three Department components. We found that the average amount of time that the ATF expended to investigate and adjudicate misconduct cases was within the range of the other three Department components (Table 2).
Table 2: Component’s Discipline Case Processing Times
Prior to June 2003, the centralized process relied on local managers to serve as deciding officials for employee misconduct cases. In June 2003, the ATF implemented a pilot project establishing a single Bureau Deciding Official to adjudicate all cases under the centralized process. A June 2, 2003, memorandum from the Director of the ATF to all ATF employees stated that:
In addition to providing a mechanism for ensuring consistent final decisions, the establishment of the Bureau Deciding Official should serve to enhance timeliness of the overall deciding official process. While managers currently serving as deciding officials must handle that responsibility in conjunction with their other wide-ranging daily responsibilities, the [Bureau Deciding Official] will have a singular focus on the prompt and fair adjudication of proposals received.
To evaluate whether the single Bureau Deciding Official was an improvement, we performed a comparative analysis of disciplinary decisions in the centralized process before and after the creation of the Bureau Deciding Official. We found that the Bureau Deciding Official’s disciplinary decisions were timelier than the decisions made prior to the June 2003 Bureau Deciding Official pilot project. For the time period reviewed, the Bureau Deciding Official pilot project improved the average disciplinary decision processing time by 43 percent. Based on 56 cases involving a decision prior to June 2003, the average time for local deciding officials to process a disciplinary decision was 111 days, ranging from 19 days to 394 days. Based on 43 such cases that were decided by the Bureau Deciding Official after June 2003, the average time required for the Bureau Deciding Official to decide cases was only 63 days, ranging from 26 days to 249 days. Chart 6 shows how long the centralized process took to issue decision letters, before and after the initiation of the Bureau Deciding Official pilot project.
Chart 6: Centralized Process Decision Letter Processing Times
The Bureau Deciding Official process also provides for more consistent decisions because the responsibility for deciding misconduct cases has been consolidated. According to the Merit Systems Protection Board, which hears employee appeals for misconduct cases involving suspensions of more than 14 days, reductions in pay or grade, or removals, deciding officials must issue disciplinary decisions consistent with their own prior decisions in their current duty station.33 Prior to June 2003, therefore, individual deciding officials were only required to ensure consistency with their prior decisions in similar cases and were not required to consider the decisions of other deciding officials or conform to an ATF wide discipline standard. Because deciding officials must issue disciplinary decisions consistent with their own prior decisions, the Bureau Deciding Official is required to ensure consistency with his own prior decisions in similar cases. Moreover, because the Bureau Deciding Official is now responsible for all centralized process disciplinary decisions ATF wide, the Bureau Deciding Official is better able to ensure consistent disciplinary outcomes throughout the ATF for the cases he adjudicates.34
In addition, the policies and procedures guiding the Bureau Deciding Official process promote a more consistent method for handling misconduct cases than the guidance previously in place. For example, the Bureau Deciding Official is required to hear, via telephone, all employees’ oral replies to proposed discipline. All employees receive the same opportunity to reply to proposed discipline in the same format. In contrast, previously deciding officials could hear employees’ oral replies via telephone or in person. Deciding officials were able to delegate the responsibility for hearing an employee’s oral reply to another individual. That individual would then summarize the oral reply for the deciding official’s consideration. This process increased the likelihood that misconduct cases or employees could receive inconsistent consideration during the adjudication phase.
We also concluded that the Bureau Deciding Official’s decisions were more reasonable than the decisions previously made by local deciding officials. Prior to the creation of the Bureau Deciding Official, local managers mitigated the proposed punishment in 26 of the 77 disciplinary decisions we reviewed. Fifteen of the 26 cases that were mitigated did not contain adequate explanation of the mitigation in the decision letters. As discussed previously, some decision letters included only a broad statement that the deciding official had “considered the Douglas Factors” or the employee’s reply. Other decision letters stated that the deciding official felt that the mitigated punishment appropriately addressed the misconduct, but did not give any reason why the discipline was mitigated.
In contrast, the Bureau Deciding Official mitigated the discipline proposed by the Professional Review Board in only 7 of the 77 cases we reviewed. Of those seven mitigations, three were not adequately explained in the decision letters, as required. These three inadequately explained mitigations were relatively minor, ranging from a 1-day mitigation to a 4-day mitigation of proposed suspensions. These results show that the introduction of the Bureau Deciding Official has reduced the incidence of undocumented mitigations in the centralized process.