DEPARTMENT OF THE TREASURY BUREAU OF ALCOHOL, TOBACCO AND FIREARMS Washington, DC 20226 AUG 20 1979 MEMORANDUM TO: Director FROM: Assistant Director (Technical and Scientific Services) SUBJECT: Comments on the National Firearms Registration and Transfer Record Study Group Report of July 1, 1975 You have requested comments on the major problems cited by the study group, as well as the corrective action taken since the report was released. The major problems, more properly described as conditions, will be listed in the same order as shown on pages 3 and 4 of the report, followed by my comments. 1. Index card file contains numerous "charge-out cards" (estimates very near 1,000), some of which are dated several years back (1971, 1968, 1957, etc.)) and contain minimal information (weapon identification only). This indicates that the index cards (particularly those with prior year dates) are probably lost. If the registrant's name is not on the charge-out card, it would virtually be impossible to locate a transaction record to recreate a complete index card. The index card file maintained in the National Firearms Act (NFA) Branch has always been subject to very strict - 2 - Director controls. Proof of this is the existence of requirements that anyone who removes an index card from the file must replace it with a charge-out card. This charge-out card contains the serial number of the firearm, the model and type of firearm, and the name of the registrant. These controls have existed, even though the index card file is not the primary record known as the National Firearms Registration and Transfer Record (NFRTR). The NFRTR is, in fact, the alphabetical file of the documents required by law to be submitted by manufacturers, importers, makers, transferors, and, before the NFA was amended in 1968, certain possessors. The index cards referred to in the report are commonly called serial number cards in the NFA Branch since they are filed by the serial number of the firearm. While we will continue to use the term index card in this memorandum, a word of explanation is in order. The term "index card" was used in the report apparently because the file consists of cards which approximate the size and thickness of 3 x 5 inch index cards. The index cards represent a cross- reference file to the NFRTR that have only the serial number and description of the firearm. It also serves as a historical record of all registered owners of a firearm since all transactions involving the firearm, including transfers to subsequent owners, are posted to the index card. It is a "back-up file" created for our convenience. While there may have been 1,000 charge-out cards in the index file at the time of the report, this never constituted a serious problem relative to the integrity of the NFRTR. On May 30, 1979, one of our supervisors in the NFA Branch spot-checked a number of charge- out cards and found that, in all cases except one, the charge-out card contained all the essential information from the index card, such as the serial number, the type of firearm, and the name of the present registered owner. It was estimated that the present number of charge-out cards in the file is between 300 and 400. - 3 - Director It was found that the predominant cause of the charge-out cards remaining in the file was due to the procedure followed when firearms were exported. When firearms were exported, the index cards from the file were stapled to the Form 9, Application and Permit for the Permanent Exportation of Firearms, and filed in the "pending exports" file, pending proof of exportation by the exporter. When proof of exportation is received, the form 9 is removed from the pending file and, at this time, the index cards are returned to the file. Since the NFRTR is referred to in the law as "a central registry of all firearms in the United States which are not in the possession or under the control of the United States," the records pertaining to exported firearms are not, technically speaking, a part of the NFRTR. We have been advised by the Chief Counsel that we are required to maintain, in the NFRTR, only documents relating to the current registrant. Thus, the history of the firearm contained on the index cards is not information that is required to be maintained. In view of this, the filing of the index cards back into the index file for those firearms that are exported has always taken a low priority. The Branch has had summer employees working on the task in the past. At one time, these cards were destroyed since they no longer referred to the firearms requiring registration in the NFRTR. If charge-out cards had been made out, then these are still in the file. Another reason charge-out cards can be in the file from several years back is that when firearms were destroyed or seized, the cards were removed from the file and attached to the documents relating to the seizure or destruction, with the result that the charge-out cards placed in the file at that time remained int eh file. There has always been a question as to whether we should retain any records relating to firearms that have been exported, seized and destroyed by a governmental agency, or have been transferred to federal agencies, including the armed services. Generally, it has been out policy to keep all records. Since the date of the report, a xerox machine has been installed in the same room as the index file, Our - 4 - Director present procedure is to make a photocopy of the index card instead of removing the card and placing a charge-out card in its place. The only exception to the policy of not removing cards is in the processing of applications transferring firearm from one special taxpayer to another special taxpayer. The Industry Section, which processes these form 3 applications, still pulls the cards out of the file and replaces them with charge-out cards. The index cards for these transactions will normally be out of the file for two to three days, at the end of which time they are filed back and the charge cards removed. The large volume of this type transaction makes this procedure necessary. On August 12, 1975, a new system was inaugurated whereby all incoming applications for the making, transfer, and registration of NFA firearm are entered in a log (or register) which contains the names of the parties involved, the serial number of the firearm, and other details relating to the form. Each entry is numbered and this log number is stamped on the index card for each firearm and is written on all charge-out cards along with the type of firearm, the serial number, and the name of the current possessor. This offers another safeguard so that the required information is available should an index card be lost of misfiled. For the reasons stated, the problem cited above did not adversely affect the ability of the Bureau to certify whether a specific person lawfully possessed a specific NFA firearm. 2. Index cards exist, but no related transaction record can be located. The term "transaction record," as used in this memorandum, refers to any of the documents or forms - 5 - Director that are required to be submitted to register, or transfer and register, a firearm in the NFRTR. Where a serial number card is in the file, but a document cannot be located, this could indicate that the firearm was transferred, and the card, for some reason, was not noted. During one period in the past when a firearm was transferred, the document relating to the prior owner was removed from the file and stapled to the bottom of the document belonging to the current registered owner. In this case, the document would be filed under the name of the current registered owner. When the NFRTR records were transferred from the Diebold files and the new filing system was set up, providing a separate file folder for each registrant, the procedure included the separation of documents that were stapled together in the manner described above. The new file was completed approximately November 1, 1976. Thus, under our present system, this type of problem would not occur since the previous registrant's document would still be in the file under his name as shown on the index card and would contain a reference to identify the subsequent registered owner. Since the previous owner no longer possessed the firearm, had they remained stapled together, our failure to locate that document would be of no consequence. What is important is that the current registered owner be shown. In a situation where an index card exists but the transaction record cannot be located, as we have evidence that the person shown as possessing the firearm on the index card is the current possessor, we would accept the index card as the best record available and consider the firearm registered as indicated on the index card. 3. Transaction records exist, but no related index cards can be located. Where a document exists, but the index card cannot be located, it has always been out practice to type up a - 6 - Director "dummy" card. (This has never created any problems as, when the serial number card shows up at a later date, the information from the "dummy" card is transferred to the original index number card.) The absence of an index card would hamper our search in determining who the registered owner of a firearm is in a situation where only the serial number is available. This could be significant when a law enforcement agency recovers a registered firearm which has been stolen. However, since the law requires the registered owner to report any theft at once to ATF, we would have the report of the theft noted on the owner's record and we would notice the absence of an index card at that time. This condition would not affect the ability of the NFA Branch to certify whether a firearm is or is not lawfully possessed. As a practical matter, we would always have a name to search in the alphabetical file to determine whether the possessor was in lawful possession. 4. Both index card and transaction record are missing. We do not understand how the study group would be able to determine that this condition existed. It could only have been determined by finding a particular weapon somewhere in the field, the possessor of which had evidence of registration, and ATF had no record of the registration. The study group describes no such weapon. The study group may have been more interested in analyzing the different combinations of situations that could occur, e.g., have document, no index card; have index card, no documents; have neither index card or document. The study group's main purpose in studying the procedures of the Branch may have been for the purpose recommending whether the NFRTR should be computerized. - 7 - Director If both the index card and transaction record are missing, it would represent a complete breakdown of the system. All our energies are constantly geared toward preventing this from ever happening. To our knowledge, this has never occurred. 5. Registration forms (nine different types) exist in several revision formats. As a result, the format of information available differs which adds to research time. (Some registrations are based on "letter" applications.) The reason this condition exists is that the law requires different forms to be used under certain specified conditions. Some of the variables to consider are: one or both of the parties are licensees or individuals; is it a taxable or tax-exempt registration; is it a registration only; a transfer and registration; a making of a firearm; etc. The reason why the study group chose to make this comment may have been merely to point out the fact that, due to the variety of forms in the NFRTR, transcribing the information to a form for entry into a computer would be a greater burden due to the many revisions of the same form during the years since 1934. The "letter" applications referred to were common during the period after the Gun Control Act (GCA) of 1968 when forms were not yet printed for the new Act. It is true that the cited condition adds to the research time in going through the documents in a registrant's file since the vital information is listed in different locations on the same type form depending on which of the many different revisions of the form was used when the firearm was registered. The fact that firearms can be registered on different forms compounds this condition. - 8 - Director 6. In prior years, applications to register weapons with no weapon serial number and/or vague descriptions were accepted. As a result, definite registration status is difficult or impossible to establish in many instances. This condition exists because, during the 1968 amnesty period, registrations were accepted on a "no questions asked" basis. If a person submitted the proper form, the Bureau was obligated to accept it. In many cases, the description of the firearm was not complete, making it difficult to classify the firearm as to the type of firearm under the law. Many people attempted to register handguns and sporting-type rifles and shotguns which did not require registration. For many years, certain rifles and shotguns could be manufactured without serial numbers. This explains why many short-barreled shotguns and rifles were registered without serial numbers. All persons submitting amnesty registration forms for firearms without serial numbers were contacted by mail and instructed to have an IRS or ATF serial number assigned by one of our offices and stamped ont eh weapon. Many firearms were serialized in this manner, but, in spite of our efforts to contact these people, 1,234 firearms remain registered without serial numbers. If a person were apprehended with an NFA firearm which did not have a serial number, we could check the NFRTR for his name. In addition, we also maintain a current drawer of index cards for all firearms that were registered without serial numbers. Our records contain other descriptive information, such as the manufacturer's name, the model, the type of firearm, the length of the barrel and the overall length of the firearm. This information would be sufficient to conclude whether the firearm was registered to the individual. - 9 - Director Most registrations prior to the 1968 GCA and all registrations since the 1968 amnesty period (except form 2 registrations by manufacturers and importers) are accomplished by an application. Registration is accomplished when the Bureau approves the form. Thus, this situation does not exist now because the Bureau will not approve any registration form if the description of the weapon is not adequate or if an accurate serial number is not furnished. 7. Internal procedures for filing and updating records have been periodically changed and modified over the years as a result of personnel changes, etc. The result is a variety of procedures and systems which may or may not be consistent and compatible. Procedures have changed since 1934. With the increase in registration and transfers, we are obliged to improve our procedures. However, the basic system has remained the same since 1934. The NFRTR consists of an alphabetical file of persons possessing firearms that are required, by law, to be registered with the Federal Government. The same cross-reference file (index cards filed in serial number order) that was originated in 1934 is still being used. 8. Weapon history record maintained on reverse of index card, which requires the card be repeatedly removed from the file for updating, increasing the possibility of loss or misfile. This condition does not affect the Bureau's ability to accurately certify to the registration status of an NFA firearm. - 10 - 9. Research activities are restricted to "one-at-a-time," due to Diebold power file unit limitations. This condition existed at the time of the study. This condition was eliminated by placing all records relating to the NFRTR in conventional type filing cabinets. The Diebold power units were transferred to another division of the Bureau. 10. Records exist for weapons that are not required to be registered. This is a current condition. A number of firearms have been removed from the purview of the NFA and, unless the registrant writes us for removal, we have no way of removing the records (without going though the entire file). This in no way affects the integrity of the NFRTR. 11. Dual (or more) registrations for one weapon. Dual registrations generally came into being during the amnesty period. Under the law, we were required to accept the amnesty registrations on a "no questions asked basis." If the registrant had acquired the firearm unlawfully from another registrant, we could not require him to tell us where he obtained the firearm. As these dual registrations are encountered during our daily operations, the NFA Branch attempts to resolve the dual registration by writing the other registered owner to inquire if he or she still possesses the firearm. Not all of these situations are true dual registrations. In most situations the "dual" serves to show the history of the firearm, i.e., registered to one person at one - 11 - Director point in time, but the same firearm becomes registered to another person at a later point in time. This certainly in no way impugns the accuracy of the NFRTR. We have found, through inquiry, that two persons may be in possession of firearms bearing the same description and serial number. This is due to the manufacturer's error in duplicating serial numbers on two different firearms. Should a search of our files be made for either person, the NFRTR would accurately show that the person possessed a registered firearm. 12. Misfiled records - index cards and transaction records, serial number and/or alpha filed out of sequence. During the period from approximately June though November, 1976, the NFRTR was transferred from the Diebold system to individual filing cabinets. At this time, the special taxpayer files, the governmental entity files (sheriff's offices and police departments), and correspondence were combined into one file in alphabetical order. While the documents were previously filed in trays, san individual file folder was created in the filing cabinets for each registrant. During this process, the proper alphabetical sequence was verified when the folders were placed in the new filing system. Periodically, the serial number card file has been worked on to detect any misfilings. During a year period ending in July, 1978, the entire index card file was examined and the numerical sequence accuracy of each card was verified. 13. Large numbers of NFA weapons manufactured and reportedly exported have never been entered into the file. Firearms are reported manufactured and registered on a form 2. Export orders typically cover hundreds or - 12 - Director thousands of firearms. When we are advised by the manufacturer that the firearms will be exported, index cards are not prepared for these firearms. The form 2's are held in a "suspense" file pending the filing of the form 9 to export the firearm and proof that the firearms were exported. When the proof of export has been received, the form 9 is placed in a "closed export" file. The above procedure has been followed for the past six to seven years. Prior to this period, the documents in the NFRTR were not always noted to reflect the fact that the weapons were exported. Sometimes the form 9 for the export of the firearms was merely stapled to the form 2 registering the firearms. The NFA Branch is presently working to resolve this condition by verifying its records with the records maintained by the exporters. The law creating the NFRTR specifically refers to firearms "in the United States which are not in the possession or under the control of the United States." Thus, records pertaining to exported firearms are no longer required to be a part of the NFRTR. In summary, we disagree with the statement that large numbers of NFA weapons manufactured and reportedly exported have never been entered into the NFRTR. However, once we receive proof that the firearms have been exported, there is no requirement that the Bureau maintain any records, pertaining to these firearms, in the NFRTR. 14. No complete "back up" files to recreate NFA registration and transfer record in the event existing files are destroyed in whole or in part. This is still true. The only "back up" file is the index card file. However, since this file is located in the same room as the NFRTR file, in case of a fire or other disaster, it would not serve the purpose of a "back up" - 13 - Director file. When the NFRTR is computerized, a back-up system will exist. 15. No ability to readily assimilate any reliable statistical data. This is still true. 16. No ability to validate (in any volume) the current status of specific weapons registered since enactment of the law. If the study group is referring to our inability to arrive at the total number of each type of firearm that may be registered at any given time, then this situation still exists. With respect to specific firearms, we are certainly able to determine the registration status of any particular firearm. 17. Multiple weapons registered on one registration record. Except for form 2 and form 9, this causes a problem in determining the current status of a specific weapon. For example, one individual in Utah registered approximately 400 "destructive cannons" (projectiles) on one form 4467 in 1968. This is not a significant problem. Except for forms 2 and 9, only one firearm may be registered on as single form. Regarding the example cited, if 400 projectiles were registered on a single form, a list would, of necessity, have been attached to the form. Serial numbers would be circled and noted if individual projectiles were transferred. - 14 - Director Study Group Conclusion (The above problems clearly indicate the possibility of embarrassment to the Bureau. If we certify, for a court proceeding, that, based upon the research of our file, a specific NFA weapon has never been registered or that a specific individual (defendant) did not register a weapon, and an attorney produces a valid registration, I cannot assess the potential legal ramification on past or future actions relative to NFA weapons.) To the best of our knowledge, no one has ever produced a valid registration (transaction document) after the Bureau has certified that the weapon is not registered to that individual. The NFA Branch is acutely aware of the potential problem and strives to continually improve the recordkeeping system. In responding to the "problems" cited by the study group, we have demonstrated that none of the problems cited has prevented the Bureau from accurately certifying to the registration status of any NFA firearm. Further, our responses have discussed a number of improvements which have reflected the NFA Branch's continuing concern over maintaining the integrity of the NFRTR. C. Michael Hoffman DADobbs:dlj/dkb 8/14/79 [new memo] Dobbs 2-11-80 The Integrity of the National Firearms Registration and Transfer Record (NFRTR) The National firearms Act (NFA) Branch of the Technical Services Division has taken a number of actions since 1975 to insure the continued integrity of the National Firearms Registration and Transfer Record (NFRTR). A summary of these actions follows. Installation of a photocopier During the latter part of 1975, a photocopying machine was installed in the immediate area where the NFRTR is located. This eliminates the need to take documents out of the file since a photocopy of the document can be made and the document returned to the file immediately. This has diminished the need to make up charge-out cards for placement in the file while original documents are out of the file. Branch policy forbids the removal of documents from the immediate file area. Serial number cards (index cards) represent a cross-reference file to the original documents in the NFRTR. They are filed by serial number of the firearm and contain the entire history of each registered firearm. They are updated each time a transfer takes place. With the photocopying machine, the need to remove the cards from the file is greatly reduced. Whenever a card is removed (except for a removal where the card is updated and replaced immediately) a charge-out card containing all the essential information is placed in the file. Transaction Numbering System On August 12, 1975 a new system of logging in and numbering all incoming applications and correspondence was instituted. The transaction number is also stamped on the serial number card when the incoming document refers to a registered firearm. The log book contains all the essential information including a description of the firearm and the names of the.transferor and transferee. The -2- logs are maintained as permanent records and the transaction numbers are used for reference purposes when noting various records in the Branch. For instance, when a registrant transfers a firearm to another person, the original registrant's document is noted to reflect the fact that it was transferred to another person, and the transaction number is entered on the document. This system provides an additional source of the same information which appears on the serial number card for each registered firearm. Diebold Filing Cabinets Removed During the period from approximately June 1, 1976 to November 1976, the records were removed from the two Diebold electric filing cabinets containing the registration documents and a third Diebold cabinet containing the serial number cards. All records were placed in conventional filing cabinets. This eliminated the problems of limited access to the files due to the nature of the previous equipment. Consolidation of NFRTR into One Section and Creation of Separate File Folders for Each Registrant When the registration documents were filed in the two Diebold cabinets, separate sections existed for individuals registration and special taxpayers registrations and registration to government entities. Another section contained correspondence. The documents were filed in "tubs" and file folders were not used. During the process of removing all files from the two Diebold cabinets, all registration documents were consolidated into one file. A separate file folder was created for each registrant and correspondence was filed -3- together with the documents in the registrant's file folder. All registrations are now in one set of files, arranged alphabetically. During the process described above, extraneous correspondence and registrations to federal government entities who are not required to register firearms were removed from the file. Verification of File for Proper Alphabetical Sequence During the five month period ending November 1976, all files in the NFRTR were checked to determine that they were in proper alphabetical order. This was part of the procedure when the new registration file was created. Registration Documents Pertaining to Firearms that were transferred by the registrant were re-filed under the registrants name During one period of time extending up to approximately December 1971, the procedure followed, when a firearm was transferred, was to staple the registration document establishing registration to the transferor to the back of the registration document for the new registrant (transferee) and, filing both documents in the NFRTR under the transferee's name. This stapling procedure was followed only for individuals. Special taxpayer's inactive documents were always placed in a "dead" file under the special taxpayer's name. During the process of removing the records from the Diebold files, these documents were separated and each one filed under the transferee's name as shown on each individual document. This procedure will eliminate those situations where we are unable to -4- locate a particular document because it was stapled to the back of the present registered owner's registration form. Serial Number (Index) cards verified as being filed in the proper numerical sequence All serial numbers cards in our file were examined and any cards that were out of sequence were placed in the proper numerical sequence. This file maintenance process was completed in July 1978. Charge-out Card Project A project was initiated to determine why any charge-out cards dating prior to the current fiscal year remain in the serial number card file. A total of 271 charge-out cards were found in the file. In the process of listing the charge-out cards, 57 cards were removed from the file and not entered on the list because the original serial number cards were in the file next to the charge-out card. In other words, the cards were replaced in the file, but someone neglected to remove the charge-out card. This left 214 charge-out cards to be listed and resolved. To date, 44 of the charge-out cards have been removed from the list and pulled from the files since the original serial numbers cards were located. The serial number cards were located for 36 of the 44 charge-out cards as they had been filed. slightly out of order, or in some cases, the serial number had been corrected by the registrant and the card filed under the new number without removing the charge-out card from under the old number. -5- Replacement serial number cards were typed up for 6 firearms from the information on the charge-out card. This project has not been completed due to higher priority matters. However, the work completed so far has enabled us to accurately assess the problem. To date we have found only 11 charge-out cards with insufficient data to trace back and re- create a substitute card. This is not considered a problem since we are quite certain these old cards refer to firearms that have been exported and we are not required to keep records on them when exported. Additionally, since the serial number card is a back-up file, a means to cross-reference firearms registered in the NFRTR, if a firearm is registered, we will have the registration document on file under the registrant's name in the NFRTR. Project Alpha This project was started in November 1979. The alphabetically arranged files that make up the NFRTR were checked manually to verify that all folders were in proper alphabetical order. A second step in its process was to examine the documents in each folder to make certain that the file was maintained under the name of the present registrant. In other words, to verify that documents were not inadvertently filed under the previous owner instead of the present registered owner. At least one document in each file was checked to determine that the name was spelled properly on the tab oil the file folder. Miscellaneous files that contained correspondence to persons who were not NFA firearm registrants were removed and either destroyed or placed in the General File. Files-relating to special taxpayers who were no longer current special taxpayers and who had no registration documents of -6- any kind in their file (current or previous registrations) were removed and set up in a separate file. Many of these were Class 6 special taxpayers who paid the $10.00 special tax in error. Project Alpha was a file maintenance procedure designed to be certain that no files had been misfiled since November 1, 1976 when the present files were set up. As of February 11, 1980 project Alpha was over 80% completed. The remaining portion of this work should be completed by the end of February. -7- FUTURE PLANS AND SUMMARY Preliminary meetings have been held with our automated Data Processing Division and the NFA Branch plans to start entering the National Firearms Registration and Transfer Record (NFRTR) into a computer some time during 1980. Our plan is to compare each document to the serial number card prior to entry into the ADP system. This will complete the final phase of our purification project since we will be assured that all information relative to the registration status of a firearm will be reflected both on the documents and the serial number cards. When the entries are completed, an effective back-up system will exist, enabling the NFRTR to be reconstructed in the event the original documents were destroyed by fire,or other disaster since the computer data can be stored on tapes at another physical location. The present serial number card file would be replaced by the computerized file. At this point record searches could be performed automatically, since the computerized records could be accessed by both serial number and alphabetically. This would take much less time than the present method of manually searching two different record systems. Past performance has proven that the NFRTR is a reliable and accurate system of records. It is unfortunate that some internal memoranda that were written for the purpose of obtaining funds to assure the integrity of the NFRTR are being used to impugn the integrity of the NFRTR. Increased activity, reflected by an increased number of firearms registered, an increase in the number of weapons transferred, and an increase in the number of firearms investigations, had demonstrated the need for certain improvements in the system. The need for a back-up system was recognized earlier, but an intensive effort to acquire the resources needed for a back-up system by micro-filming the records started in 1971 and abandoned in 1974, made it obvious that the ultimate answer to the problem was entry of the records into a computer. Since -8- the money for a computer was not available, other methods had to be proposed. In the process, certain memorandums were written which emphasized and dramatized certain inadequacies in the records in an effort to obtain funds for additional personnel and equipment. An analysis of the problems cited in the memorandums, as well as an internal study of the NFRTR, fail to discredit the ability of ATF to accurately certify to the registration status of a particular firearm. [1981 memo] STATUS REPORT NATIONAL FIREARMS REGISTRATION AND TRANSFER RECORD (NFRTR) Deron A. Dobbs July 1, 1981 [newpage] TABLE OF CONTENTS Page Introduction 1 I. STATISTICS A. Number of Firearms Registered 1 B. Project Inventory 2 II. PAST CRITICISM OF THE NFRTR A. Alleged Inaccuracies in the NFRTR 4 B. Comments on Criticisms 4 C. Challenges in Court Regarding Accuracy of NFRTR 5 III. PAST ATTEMPTS TO CREATE A BACK-UP FILE TO THE NFRTR A. Microfilm Project 8 B. Proposed Xerox System 8 IV. ADDITIONAL ACTIONS TAKEN TO ENHANCE EFFECTIVENESS OF THE NFRTR A. Project Alpha 11 B. Purification Project 11 C. Verification of Special Taxpayers (STP) Inventories with our Records in the NFRTR 11 V. PRESENT SHORTCOMINGS ASSOCIATED WITH THE NFRTR A. Present Conditions 13 B. Comments on the Shortcomings 17 VI. FUTURE PROJECTS A. Retaining Inactive Records 19 B. Study of Special Taxpayers 19 C. Study of Out of Business Manufacturers as Large Dealers 20 VII. COMPUTERIZING THE NFRTR A. Benefits of a Computer 22 B. Disadvantages of a Computer 23 C. Anticipated operating Procedure for Entry of Original Data 24 D. Discussion of the Above Methods 25 [newpage] - 1 - INTRODUCTION The purpose of this report is to review past activities relating to the NFRTR, assess the present status of, and to discuss future activities relating to the NFRTR. I. STATISTICS A. Number of Firearms Registered Attached Exhibits A1 and A3 are estimates that have been used during the past 6 years. The estimate dated September 12, 1980 contains the most detailed information and is based upon a sampling of 3.5% of the records. The latest estimates can be summarized as follows: Type of Weapon Number Registered Percent of Total Machinegun 100,452 54 Rifles 14,036 7 Shotguns 19,734 11 Any Other Weapons 31,372 17 Silencers 7,766 4 Destructive Devices 12,156 7 186,516 100 - 2 - The above total is considerably less than our previous estimates. The reason for this is that the previous estimates were based upon the total number of serial number cards in our file and there was no attempt to allow for records in the file pertaining to firearms that are no longer active registrations. Inactive registration records pertain to firearms that have been exported, transferred to the Department of Defense or other Federal agencies, have been removed from the purview of the NFA, or have been abandoned, seized, or destroyed. We estimate that there are 33,132 firearms in this category. It should be noted that the number of cards in this group that relate to exported firearms represent only a minuscule amount when compared to the total firearms exported since the serial number cards have been destroyed when the firearms are exported or are never prepared when it is known that the firearms are manufactured for export. According to my estimates, we have 75,600 registered owners who possess one or more registered firearms. B. Project Inventory During the approximately eight years I have been assigned to the NFA Branch, we have received various requests relating to how many firearms are registered, how many of each type, and how many are registered in a certain state. These requests have come from State authorities, news media, congressional committees, and other government agencies. Our standard reply has been that we are unable to supply the requested information because our files are not automated and the information is just not available. In April of this year, I decided that to keep saying over a 46 year period that we do not have this information available could no longer be tolerated. As a result, I initiates Project Inventory to make a physical count, from our serial number registration cards, of the number of firearms registered, by type of firearm, condition of firearm as to whether serviceable or unserviceable, type of registrant (individual, special taxpayer or governmental entity), and by State where registered. While this project represents an enormous task, it will provide us with much needed information. In - 3 - addition to the actual tally of this information, certain file maintenance operations are being carried on at the same time. The correct numerical filing sequence is being verified, cross- reference cards (when firearms has more then one number on it that may cause confusion as to which one is the serial number) are checked out to determine which one is the master card and whether both are needed, and other items such as adding form numbers if missing on the card. This project should be completes during January 1982, based upon our present procedures which involves approximately 30 person/hours a week being spent on this project. I plan to reassess our progress and evaluate the program in view of other priorities. I still favor a total count, but will consider going for a large sample and then determine the make-up of the NFRTR from this sample. The NFA Branch has gone thru a 6 year period during which the prospects of computerizing the NFRTR continued to rise and fall. When Project Inventory was started, the Bureau was in the process of preparing retention registers prior to initiating RIF actions during the balance of FY 81 and as a result, the prospects for computerizing the NFRTR seemed remote. Events since that time, indicate that the ADP Division is now actively working on our program again. Should the program for entering the NFRTR data into our computer become operational, and should we receive assurances that we will be able to hire the extra personnel to process the data, I would consider an immediate cessation of Project Inventory Activities. - 4 - II PAST CRITICISM OF THE NFRTR A. Alleged Inaccuracies in the NFRTR Special Agent Philip Price, who was the NFA Branch Chief at the time, testified in the trial involving the National Rifle Associations action against the Government to recover 7 firearms. The trial was conducted without a jury on June 17, 1979 thru June 19, 1979. During the trial, the NRA challenged the accuracy of the records that make up the NFRTR. The source for their criticism of our records was the July 1, 1975 internal memorandum to the Director from the Chairman, NFA Registration File Study Group. This memo listed 17 items under the heading of Major Problems. Other internal memorandums also were acquired by the NRA which, when read by outsiders, and taken out of context, were very damaging and would lead readers to question the accuracy of our records. The memos which were acquired by the NRA are attached. B. Comments on the Criticisms A memorandum, dated August 20, 1979, from the Assistant Director, T&SS, to the Director responded to the Director's request for comments on the study group's report and included corrective action taken since the date of the report. On February 11, 1980 a summary was prepared, entitled, "The Integrity of the National Firearms Registration and Transfer Record." This paper summarized the action taken to date to insure the continued integrity of the NFRTR. In short, the 1975 memorandums were shown to have been written to emphasize and dramatize a need to obtain funds for additional personnel and equipment. Many people outside of government do not realize that the competition for funds in government agencies parallels the competition that exists in private industry. In a large corporation, for example, you have the Sales, Engineering, Production, Quality Control and other departments in a constant battle - 5 - for funds. This same competition exists in government. Mr. West, the NFA Branch Chief from 1974 thru March 1976, realized that the quality of the people in the Branch determines the quality of the work that the Branch performs. He was successful in bringing on board additional people as well as upgrading the positions in the Branch so that competent people could be recruited and retained. I have been assigned to the NFA Branch for nearly 8 years. During this period, it has been a standard procedure to cite the consequences of an error in conducting a records search. Our daily activities have always been guided by the constant awareness that an improper records search could cause an innocent person to be arrested. To my knowledge, this has never happened. A former Division Chief, Thruman W. Darr, testified in the NRA case that during the period he was assigned to the NFA Branch and when he served as Division Chief of the Technical Services Division, from 1958 until his retirement in January, 1976, that ATF had never prepared an erroneous certification. Since July, 1973, when I transferred to the NFA Branch, over 10,000 certifications have been prepared. To my knowledge, nobody has ever produced a valid registration document after we have certified that a firearm was not registered to that person. C. Challenges in Court Regarding Accuracy of NFRTR It is my opinion that the NRA became aware of the internal memorandums relating to the purification of the NFRTR thru their General Counsel, Mr. James Featherstone. Mr. Featherstone, before being employed by the NRA, was the Deputy Assistant Secretary of the Treasury (Enforcement) and in this capacity had access to these reports. After the NRA case, the NRA sent copies of the memorandums to various Public Defenders around the country. In addition, the memorandums were introduced in the oversight hearings on BATF, before the Subcommittee of the Committee on Appropriations, United States Senate, on July 11 and 12, 1979. - 6 - I first encountered the memorandums in court when I testified in October, 1979 (U.S. vs. Eisenstein) in New York City. The judge did not allow them to be introduced in this case and the defendant was found guilty on all counts. I testified in an FBI case in U.S. District Court, Seattle, Washington on December 4, 1979. This case involved charges against Judith Bissel relating to an attempted bombing of the Air Force ROTC Building at the University of Washington building by her and her husband. She and her husband had forfeited a $50,000 cash bond in 1970 when they disappeared after being arrested. Judith Bissel had been a fugitive for 10 years prior to the trial. Her husband was a fugitive at the tine of my testimony and to my knowledge remains so. The Public Defender in this case had been provided copies of the memorandums by the NRA. The defense attorney questioned me regarding these memorandums, in court, but not in the presence of the jury. The judge ruled in favor of the government and my certifications of non-registration were allowed to be introduced into evidence. Judith Bissel was found guilty of the charges. The accuracy of our records was brought up on September 10, 1979, in U.S. District Court, Orlando, during U.S. vs. John Mark Johnson and Charles Allen Stout. I was prepared to testify in this case but Judge John Reed that there was no evidence that we had ever certified improperly and that the memorandums were hearsay. Other challenges were made during a couple of cases in Ohio, that I am aware of as a result of receiving a subpena to testify, including one case that was handled by the Organized Crime Branch of the U.S. Attorney's Office in Cleveland during March 1981. An interesting memo of February 14, 1980, written by one of the NRA's attorney's, David T. Hardy, showed up in pre-trial proceedings of the Cleveland trial. [redacted material] - 7 - [redacted material] I was on call to testify in other cases in Philadelphia and Baltimore, but did not receive subpoenas in these cases as apparently the challenges were not allowed. On June 24, 1981, we received a call from the case agent in a trial that was to begin the next morning in Hartford, Connecticut. The Public Defender in this case had, at the last moment, produced copies of the July 1, 1975 study and accompanying memos and was challenging the accuracy of the records in the NFRTR. I discussed the memos with the case agent, Rick Smith, including past actions by the other Judges. Special Agent Smith informed me later that the Judge in Hartford also ruled the memos were hearsay. Most of the memorandums are now over 6 years old and they have not been brought up in any of the cases I have testified or been involved in since the September 1980 trial in Orlando, Florida. - 8 - III. PAST ATTEMPTS TO CREATE A BACK-UP FILE TO THE NFRTR A. Microfilm Project In February of 1971, a program was initiated to create a back-up system to protect the records in the event they were destroyed by fire or other disaster. The program involved a 3 step process: 1. The registrant's name, description of firearm and serial number were keypunched onto a computer tape. (This tape could only be read by taking it over to the National Bank of Maryland and having their computer print out a hard copy.) 2. The computer tape was used to generate a microfilm copy of the information. 3. A camera was used to create microfilm copies of the documents. Only a limited amount of work was performed on the program during the last half of 1973 and the first half of 1974. The program was abandoned completely in August 1974. The records pertaining to special taxpayers and law enforcement agencies never were entered into the system. As a result, only 50% of the registration forms were ever microfilmed. The system proved to be unsuccessful because the proposed method of updating the microfilm records with a delicate stylus did not work out. Since there are frequent and rapid turnovers in the registered owners of a firearm, a system that involved taking a picture of the documents, having the film developed, and splicing rolls of film was not a workable solution. The problems with this system were discussed in the 1975 memorandum. B. Proposed Xerox System The NFA branch Chief from 1974 through March 1976, the late Mr. Reginold T. West, was convinced that the NFRTR system should be converted from a manual system to a modern system employing computer technology. When promised funds and equipment were not made available, he was forced to seek other alternatives. His - 9 - suggestion was to duplicate all documents in the alphabetical files by xeroxing them, and then setting them up in a file located in another area of the building. I don't think this systems would have been practicable either since it would have merely created a second file to maintain. While it would have created a back-up file in the event of a disaster the requirement for maintaining 2 alphabetical files plus the serial number file would not have been practicable, and the predictable result would have been a decline in the maintenance of at least one of the files and the duplicity of effort could not be condoned. - 10 - IV. ADDITIONAL ACTIONS TAKEN TO ENHANCE THE EFFECTIVENESS OF THE NFRTR In addition to those actions taken that were discussed in the memorandums dated August 20, 1979 and February 11, 1980, a number of other actions have been initiated. A. Project Alpha Project Alpha was completed on March 12, 1980. This project was initiated because of our continuing responsibility to do everything possible to assure the accuracy of the NFRTR. The project was completed over a 3 1/2 month period. Most of the work was performed by Branch personnel during regular working hours. However, in order to expedite the project approximately $7,000 in overtime was authorized and expended. The purpose of the project was to go through each of the 39 five drawer filing cabinets that constitute the alphabetical section of the NFRTR to determine that each of the estimated 75,600 file folders were filed in proper alphabetical sequence. Additionally, at least one document in each folder was checked to verify that the tab on each folder reflected the name of the current registrant (and not the previous registrant's name) and that the name was spelled correctly. In addition to initiating and supervising this project, I personally checked 2 filing cabinets containing approximately 4,000 folders. I found a total of 22 discrepancies, but only one was an error of a serious type. A registrant whose last name started with the letter "L" was filed under the "C" section of the file. The registration form was an amnesty registration form which the Bureau accepted in 1968 on a "no questions asked" basis and without any approval action our part. The form was filled out in longhand. The first letter of his name as printed ont he form and as written in his signature appeared to be the same and was interpreted as a "C" instead of an "L". The serial number card for this firearm reflected the proper spelling of the man's - 11 - name. Thus, had we searched the serial number file, we would have determined that the firearm was properly registered to the registrant and this error would not have caused us to improperly certify as to the registration status of this firearm. The remaining discrepancies that I found were all of a minor nature such as a first name shown as Ray instead of Raymond, the "Jr." left off a name, or "& Sons" left off a firm's name, or a folder being filed out of sequence of order by 2 or 3 folders. B. Purification Project This project was initiated in July 1977. The purpose of the project was to verify that each firearm shown in the NFRTR had a corresponding serial number (SN) card in the SN card file. As each verification was made a symbol (a zero with a vertical slash made through it) was placed on the upper right hand corner of the document as well as on the SN card next to the annotations on the card pertaining to that document. As a starting point, a list of out-of-business special taxpayers was selected. When the project was started, we planned to incorporate these procedures into our computerization of the NFRTR. This is a continuing project. As time is available from required daily duties, any available person/hours is assigned to this project. Because of increased workloads, other projects, and a shortage of personnel due to hiring freezes, the work on this project has been limited, at least during the past 18 months, to our special taxpayers. C. Verification of Special Taxpayers (STP) Inventories with our Records in the NFRTR. Regulatory Enforcement has generally considered compliance inspections of NFA special taxpayers (manufacturers, importers and dealers) to take priority over inspections of licensees who are not special taxpayers dealing in NFA firearms. Inspections have been made on an annual basis. In discussing a particular STP with my NFA Specialist in the Evidentiary Documents section, it was decided to - 12 - resolve our problem by having the Inspector get the answer for us when he did the next compliance inspection of the dealer's premises. Out of this discussion, we realized that we could verify our records with the STP's records simply by providing the Inspector with a list of firearms that our records show to be in his possession. By providing the inventory prior to his making his inspection, he could verify our listing during the inspection and resolve any discrepancies. Regulatory Enforcement was very receptive to this program and on May 6, 1980, a TECS message went out instituting this procedure. This program has been in effect for over a year now and has been very helpful. It has uncovered a number of instances where a transfer was cancelled, but the transferor failed to send in the original copy of the transfer application and request that it be voided. This procedure is necessary to restore the original transferor as the lawful registrant in the NFRTR. It has also been brought to our attention the fact that thefts of NFA weapons are not always reported to us as required. We in turn notify Criminal Enforcement so that they can enter the stolen data into TECS and NCIC. Another benefit of the program is that we have been able to clear from our active records many records pertaining to firearms that large manufacturers either exported or transferred to the defense department. - 13 - V. PRESENT SHORTCOMINGS ASSOCIATED WITH THE NFRTR A. Present Conditions There are a number of conditions which still exist which have been described in the past under the headings "Problems and Discrepancies." These conditions are primarily the result of the structure of the law and are not caused by any action or inaction on the part of the NFA Branch. 1. Dual registrations. During the November 1968 Amnesty Period, approximately 65,000 firearms were registered on a "no questions asked" basis. As a result, people registered firearms that our records clearly indicated were in fact already registered to someone else. The form did not ask who the firearm was acquired from, nor could we ask, and the result was that we now have approximately 3,000 dual registrations in the NFRTR. Efforts were made to resolve these dual registrations, but the Bureau has always had higher priority work, and the result was that we only addressed those duals that have come to our attention when we receive an application pertaining to a firearm that is dually registered. Even then, we have not been very successful since we normally write letters and often we do not have current addresses. Criminal Enforcement informed us that we should no longer refer requests for dual registration investigations to them. They asked that future requests be limited to situations involving unusual circumstances. While dual registrations can be described as discrepancies in our records, I do not consider it a serious problem in that if one of these firearms or the person's name is the subject of a record search, the search would reveal that the firearm is registered to one of two persons, or possibly to both persons since it is possible for a manufacturer's error to result in 2 firearms being produced with the same serial number. When we have records showing a firearm registered in some year between 1934 and 1967, and then have - 14 - similarly described firearm registered in 1968, it might be logical to purge the earliest registration from the NFRTR, but without actually contacting the original registrant to find out whether he still has his firearm, we have always felt it was better to err on the side of having some extra paperwork in our files than to take a chance on purging a registration that may be a legitimate current registration. 2. Firearms registered without serial numbers or registered under a parts number. High powered rifles and handguns were not required to be serialized until July 1, 1958. Shotguns were not required to have serial numbers until the GCA of 1968. As a result, many registration documents were received during the 1930's and 1940's without serial numbers. These people were contacted by letter during 1970 and 1971 but many did not respond. Again, the government's position was to consider the firearms registered rather than to remove the registrations from the file when people failed to respond to our correspondence. This is another situation where, even though the lack of a serial number could be termed a discrepancy in our records, we could still certify properly since we would have the registrants name in our alphabetical file. 3. No requirement to update registration. Most record systems contain a provision whereby the record, whether it be a registration or license, must be renewed on an annual basis, usually accompanied by a renewal fee. While there is a making tax and a transfer tax, I would estimate that over 85% of our registrations involve registrations under conditions under which the registrant paid no tax or registration fee of any kind. We live in a highly mobile society. I would estimate that 5 out of 8 persons move within a 6 or 8 year period. The law requires persons to obtain authorization from the Director prior to any movement of most NFA firearms in interstate commerce. Also, we have the condition where people who registered firearms under the original - 15 - National Firearms Act at age 65 would now be 112 years old. We know that these people are dead and their heirs have not taken the necessary steps to contact us so that the involuntary transfer created by the registrant's death can be formalized in the NFRTR. 4. Some forms have not been noted to reflect that the firearm was transferred to someone else. At one time (prior to 1973), when a firearm was transferred, the document registering the firearm to the seller was stapled to the back of the new owner's registration document. Later these documents were separated and filed under each transferee's name. We have no guarantee that all of these documents were noted to indicate that the firearm was transferred. Part of our file purification project involves comparing our serial number card record with the registration documents. This process will result in the document being properly noted when the firearm has been transferred to someone else. 5. Peace officers could acquire firearms tax exempt under the original NFA. Police Departments and Sheriff's offices also could acquire firearms tax exempt using the same application form. In some instances, it is impossible to determine whether the firearm is registered to an individual officer or the the police department of sheriff's office. As we became aware of this type of condition, our practice is to place a cross reference card in the NFRTR so that a search under either name will recall the document. 6. Situation where transferor, after receiving an approved application to transfer a firearm, does not carry out the approved transfer and fails to cancel the application. This is not a serious problem as it relates to certifications since firearms cane be lawfully possessed by only the 2 parties to the transfer. Since our approval of a transfer application automatically registers the firearm in the name of the transferee, there is always a period of time when we show the firearm registered to the transferee - 16 - but the transferor still possesses the firearm pending shipping arrangements, and in many cases, payment in full of the purchase. There is a real need to correct this situation by providing for a definite period such as 60 or 90 days when the transfer must be physically completed or the applicant should be required to void the approved application. Another alternative solution to this problem would be to add a tear-off section to the transfer application and require the transferee to sign and return the form to ATF indicating that he received the firearm. I have never been in favor of this procedure since it would add an additional burden on the taxpayer and generate extra paperwork and procedures in our Branch. It has been suggested that we could require a third copy of all transfer applications. This third copy of the application could then be mailed to the transferee at the same time we mail the original approved application back to the applicant transferor. 7. We have firearms that we know have been transferred to the military. At one period of time, the serial number cards were destroyed, and if the registration documents were not noted in each case, we will have registration documents showing active registrations when, in fact, the records should have been noted to show that registration is no longer required since the firearms are under the control of the U.S. This is another example of excess records in the file. 8. Under the original National Firearms Act, unserviceable firearms could be transferred and then a "Notice of tax exempt Domestic Transfer of Firearms (form 5)" was sent to ATF. Obviously, if the form was lost in the mail, or the transferor forgot or neglected to send us the copy, we would have no knowledge of the transfer and we would show the firearms till possessed by the original registrant. The purchaser would have his copy of the form, however, and if he were to attempt to sell the firearm, we would - 17 - become aware of the variance in our records and could request a copy of the missing transfer application. 9. We have no ability to provide accurate up-to-date statistics on how many firearms are registered, the number of various type firearms that are registered, wheat States the registrants live in, or similar data. 10. Perhaps the biggest shortcoming to our present system of records that make up the NFRTR is that we have no back-up system in the even the records should be destroyed by fire or other disaster. B. Comments on the Shortcomings Many of the above items were cited as problems in the July 1, 1975 report of the NFA File Study Group. However, as I have pointed out previously, none of these conditions prevent the NFA Branch from accurately certifying whether a specific firearm is registered to a specific person. Despite certain shortcomings that are inherent in the system and which are beyond our control, we do have an effective system. The NFRTR is accessible by either name or serial number. Since we have 2 separate sets of files for each registered firearm, one system acts as a back-up system for the other system. If a document is mis-filed, we would still be have a record in the serial number file system which contains the name of the registered owner of the firearm. If a serial number card is mis-filed, we would still have a record of the registration in the Document Section of the alphabetical file. Each system is cross-referenced to the other system. In addition to our maintaining two separate sets of registration files, the law requires each registered owner to maintain proof of registration which shall be available upon request. We have capable people maintaining the system. The NFA Specialist in the Evidentiary Documents Section has been assigned to that position for the past 2 1/2 years. Prior to that she had 6 years experience tracing firearms in ATF's National Firearms Tracing - 18 - Center. Prior to that she worked for 2 years in our Regional Office in Cincinnati. The two applications examiner in the section are very well qualified to perform their duties. One has been assigned to the Branch for 5 1/2 years, most of which time has been in Evidentiary Documents Section. She has, in addition, 2 1/2 years experience in another office of the Bureau. The other examiner is also very well qualified. Before coming to Atf, as an Army officer, he performed various administrative duties for 13 years. He has been assigned to the Evidentiary Documents for over 4 years. Other application examiners and other personnel in the Branch who have access to the NFRTR are also well qualified. The newest examiner in the Branch has been with us 1 1/2 years, but she was an applications examiner with Customs prior to joining us. The remaining personnel have experience ranging from over 6 years in the Branch to over 20 years. It is for the above reasons that I must conclude that the attacks on the integrity of the record in the NFRTR cannot be justified. While the system may be an "old fashioned manual system," it has proven to be a reliable system. - 19 - VI. FUTURE PROJECTS A. Retaining Inactive Records We have debated whether to continue to retain records in our file that pertain to such inactive registrations. We have no obligation under the National Firearms Act to retain these records. However, we have felt that the records have value for tracing purposes such as when an investigation reveals the existence of a firearm that our records show was destroyed or exported. An alternative would be to place these records in a "dead" file, but this action would create a second file that would need to be searched when performing certain record searches. We have been retaining all copies of registration documents even though the registrant has transferred the firearm to another party. We need to review this procedure since it results in a continued expansion of the number of documents in our system, even if the number of registered firearms were to remain the same. We cannot automatically destroy the previous document when we approve a transfer since the buyer may not be able to raise the purchase price and the seller (transferor) could write to void the original transfer, and in this case we would need the previous registration form. We should be able to store some inactive records, especially those pertaining to some large manufacturers or dealers, in a records center. A program for removing documents from our file that do not pertain to current registrations should be started at some future date when the workload permits. I don't see this as a priority item at this point. As we get into our computerization of the files, we may find it convenient to work this project in at that time. I am convinced, however, that we cannot justify our present policy of retaining all records indefinitely. B. Study of Special Taxpayers There has been a loophole in the National Firearms Act that has caused the number of STP's to increase from 179 in 1964 to approximately 300 STP's in 1973 and to over 1200 STP's at the present time. Briefly, - 20 - persons who are not legitimate dealers acquire a STP status by purchasing the $30.00 Federal Firearms License and pay the $200.00 per year special tax. This action enables them to acquire the firearms they want from other "dealers" without paying the transfer tax. Furthermore, they can acquire the firearms from any other STP in any of the 50 States, and regardless of how many firearms they acquire, they never have to be fingerprinted or attach a photograph to any of their applications. A number of large STP's have actively publicized this loophole in order to facilitate the sale of their firearms. The practice has continued in part due to the fact that ATF has never satisfactorily defined the term "engaging in the business." In 1976, ATF Ruling 76-22 was issued to deter this practice. The ruling was so ineffective that, despite our attempts to perfect cases against STP's who obviously acquired firearms in direct violation of the ruling, I am not aware of a single firearm being seized or a single tax assessment made for a violation. This project would involve analyzing the records in the NFRTR pertaining to STP's to isolate those dealers who are not legitimate dealers and determine the amount of the transfer taxes that have been avoided. Another facet of this problem concerns the fact that STP's are acquiring firearms that they could not otherwise possess due to State laws. Further, they are importing firearms or acquiring imported firearms when they would not otherwise be authorized to possess were it not for their STP status. All of these actions are taken in direct violation of the intent of the National Firearms Act. C. Study of Out of Business Manufacturers and Large Dealers We have a large number of manufacturers and dealers who are no longer in business, but show substantial inventories of firearms. We need to make contact with these dealers either through Regulatory or Criminal Enforcement in order to verify their inventories. We know in most cases the manufacturers have disposed of their inventory to the military or by exportation without the proper forms - 21 - being submitted to this office. In other cases, we know that dealers sold their weapons legally under the DEWAT program and there seems to have been a lack of a requirement to notify us in these circumstances since the firearms, when welded up according to the DEWAT procedures, were not considered to be firearm under the NFA and no longer required registration, until the amended NFA in 1968 defined the receiver of a machinegun by itself to be a machine gun requiring registration to be lawfully possessed. Action on this project will further reduce "excess paper" in the NFRTR. - 22 - VII. COMPUTERIZING THE NFRTR A. Benefits of a computer 1. The primary benefit of a computer will be to provide a back-up system for the NFRTR so that it can be re-created in the event of a disaster. After each day's operations, everything in the system is stored on a tape. These tapes can be stored at another location and also serve as a back-up in the even something happens to the system during the say to cause us to "lose" our records. 2. The system will provide us with up to date statistics on the number and type of weapons registered as well as other data. 3. We anticipate that a modern computer system will enable us to process our applications faster and more economically than our present manual method. The primary factors in approving transfer applications of firearms involve: a. Verifying that the firearm is registered and that the applicant is the registered owner of the firearm. b. Determining that the transferee can possess the firearm without violating State laws. c. In the case of Form 3 transfers between Special taxpayers, the current Special Tax Status of both parties to the transfer must be verified. It may be possible to have sufficient information in the computer to enable us to process 75% of our applications by using the computer terminal. The only manual steps aside from signing the applications would be the noting of the previous document and filing our copy of the approved document in the alpha file. Certain transfers involving additional procedures which must be completed before the final operations can be conducted by using the computer. 4. The new record system should reduce the possibility of human error in making a records check. - 23 - 5. making contact with registrants for informational purposes as well as for the purpose of updating our records will be possible due to the computer's ability to print mailing labels. 6. While we will always be required to keep the registration documents in alphabetical file for evidentiary purposes, the combination of an accurate manual system and computerized system will provide us the best of both worlds. I think both are important since the people who serve on juries, the average citizen, have heard too many horror stories to place blind faith in them as a sole source of information relating to such an important piece of information as whether a firearm is registered to the defendant. 7. The new computerized records system may provide the nucleus for any new registration system of other type firearms, such as handguns, that may be mandated by the American people at some time in the future. With the system in place for our NFA records, if there is a sudden requirement for federal registration of handguns, the same system would handle other type registrations. In fact, in the unlikely event that handguns were required to be registered with the federal government, a computerized system would be an absolute necessity. B. Disadvantages of a computer 1. It is a well known fact that any computerized system of records will only be as accurate as the information that is entered into the system. We will still have our dual registration, firearms registered under parts numbers and firearms registered to 112 year old persons. Also, it is possible for errors to be made in entering the data in the system. 2. The initial entry of data into the system will be a Gargantuan task. It will require a lot of time as well as additional personnel in the Branch. These expenditures may be offset by lower costs in the future, but it is too early to - 24 - come up with any specific estimates at this time. The system could prove to be much more expensive than the present manual system. C. Anticipated Operating Procedure for Entry of Original Data Two methods are possible for original entry of the data into the system. 1. All data from the serial number card files could be entered. Perhaps an outside contractor could be hired who had 30 or 40 or more persons trained to do this work on a "crash program" involving a very short time period. The serial number file is not that bulky and it could be transported easily to the contractor's office, assuming they had the necessary security clearances. The entries in this case would likely be made direct from the serial number cards, presumably with some kind of verification system to assure accuracy. The above having been accomplished, all current transactions could be posted to the new system. The final step would be to start with the first folder in the "A" section of the alphabetical file, and on a document by document basis, verify that every registered firearm in the alphabetical system appear accurately in the computerized file system. Any error or discrepancies are resolved. 2. This approach or method involves starting entry of data by going to the first folder in the "A" section of the alphabetical file. The data in the alpha section would be compared and verified with the data in the serial number card file. Any errors or discrepancies would be resolved at this point and a "data transcription" form of some kind would be prepared. This operation would be performed by NFA Branch personnel who possess a good knowledge of our operation. - 25 - The entry into the new system would then be performed by lower grade personnel trained for this purpose. At some point, all of the "A" registrants will have been entered into the system, and at this time, we would then know that we could conduct all name searches for "A" persons by using the ADP system. D. Discussion of the Above Methods The first method described above would result in the system being operational in the shortest possible period of time. In essence, the present serial number card file would be recreated in our computer system. At this point the serial number cards would become obsolete and would be retained for reference purposes only pending complete verification of the new file with the alphabetical file. Funds would need to be made available for this project and a contractor that could do the job in an acceptable period of time would need to be found. My present thinking favors the second method outlined above. This method would be slower, but would have a number of advantages. 1. All entries would be verified (data in alpha file compared and in agreement with data in serial number file) before entry into the system. 2. All work would be under our immediate control so that we could have better assurance of accuracy. 3. Would be less interference with current workload since normal operations could be carried on at all times. We would not require a period where our operations might need to be shut down or, if we approve application's during this phase, a backlog of transactions to be updated in the computerized section would occur. 4. Trained NFA Branch personnel would do all of the verification work and a transcription form of - 26 - some kind would be used. Entry into system would be by clerical personnel and in such a manner that each entry is checked for accuracy by a second person. As of this writing, our ADP personnel are in the preliminary stages of developing the program for our ADP system. As we become more familiar with the system as it is developed, we may have different ideas on implementing the system. ADP personnel may also be able to assist us in this regard. Priorities and budget considerations will also influence our choices. Regardless of how we implement the new system, we should discontinue the serial number card file as soon as all the data has been entered into the computer. These files could be placed in storage after a period of time since our computerized file will then constitute our new serial number file. The new file will be accessible both by serial number as well as by name. There is always a possibility that we may not have the money to implement the system within fiscal year 1982 or 1983. In any event, I think it is important that the ADP section follows thru and completes the programming phase of this project. We will then be in a position to proceed once the money and personnel are made available. I can see a possibility of the National Firearms Act being amended in some manner that would create a huge workload for the Branch. This could be in the form of a new amnesty period, whether it be a general amnesty or one limited to certain firearms under certain conditions, or perhaps a requirement that all NFA firearms be re- registered. At any rate, I think it is our responsibility to be prepared to implement any new programs that may be assigned to our Branch.