one F.B.I. agent. It was further decided that the interviews would be conducted starting in the G.S.A. stores, one store at a time. In order to insure the integrity of the interview notes and the 302's, we required that at the end of each day's interviewing the agents would not proceed to further interviewing until their notes had been dictated into 302's.

The Bureau has a five-day policy, which requires interviews to be dictated within five days of the date of their occurrence. Because of the number of interviews that were going to occur and because we foresaw possibilities of challenges as a result of the volume of interviews, it seemed only logical not to have the interviewers conduct further interviews until they had finished the 302's on any date. A pattern of interviewing was thus created of one day of interviews, followed by a day of dictating, followed by a day of interviews, followed by a day of dictating. On several occasions the agents were unable to complete their 302's at the end of the intermediate day and they were not permitted to continue their scheduled interviews for the following day. In fact a trial of one of the defendants produced cross examination on the integrity of the interviewing and transcribing process. The answer provided by the agent of the procedures established so stunned the defense attorney that he commented in the presence of the jury the method established was extraordinary.

The interviews were conducted with an understanding that each individual interviewed would be advised of his or her constitutional rights and asked to sign a waiver of those rights. No one was to be promised any immunity and no one was to be lead to believe anything other than that they were and would remain targets of the investigation. In fact, the agents were instructed that at the end of each interview they were to advise the defendant to obtain an attorney familiar with the practice of criminal law in the federal courts and that that attorney was to call our office.

Informal immunity by letter was given to two Hilles employes who were secretaries and office managers. Those were given on the grounds that one of them no longer worked with Hilles and the other was crucial to making the cases against the Hilles president and vice-president. That pattern was followed in the other companies where the secretarial personnel were offered immunity in exchange for their full cooperation and testimony. As related to Hilles, once the two secretary/ office managers were in our camp, the president and vice-president had no choice but to cooperate and accept whatever plea offer was given.

The agents ran all of their interviews according to their time-honored system of interviewing.

They began

with the advice of rights, an explanation of what the subject matter of the interview was, and a request for volunteering of information by the target. In most cases the target had

nothing to say involving any culpability whereupon the agents produced the stack of documents showing false invoices to his store and the receipt by him of goods for his personal use. It is hard to estimate, but approximately 50% to 55% of the individuals when confronted with the documents confessed.

Over 150 interviews were conducted of both G.S.A. employees and employees for other agencies and other companies. The interviewing process went smoothly as did the procedure of using the F.B.I. as the only writers of the 302's. Part of that requirement, incidently, was based on our concern of using G.S.A. secretaries within the G.S.A. headquarters to type anything. We were all satisfied that we made the right decision on that issue.

It should be pointed out that through early May of 1978 the integrity of the investigation was absolute. NO one higher than Bill Clinkscales within the General Services Administration knew exactly what was going on. Once the interview process started, the rumors obviously became rampant, but the results of those interviews were never known. So well kept was our secret that in June of 1978, after at least two press reports and dozens of employer interviews, the Administrator of G.S.A. at the time was interviewed in Baltimore at the dedication of the Garmatz Building and told a reporter that the investigation only focused on Baltimore and that that was the only place that was having trouble in their G.S.A. store.

The interviews, obviously, fleshed out the facts learned through the documents and revealed a very simple scheme. James Hilles Associates and other companies through their salesmen went into the various G.S.A. stores to sell legitimate goods. In the course of those sales they learned that a well-established practice within the stores was to take the managers to lunch and to give them Christmas gifts of liquor and other items. Several companies pre-dated Hilles in their criminal activity. Atlantic Office Supply started in 1971 to give bribes to various store managers. The only difference between Atlantic and Hilles was that Atlantic gave bribes and from time to time filed false invoices to cover the expense of the bribes. Hilles had as a substantial part of its business the giving of bribes and the charging of the government for those bribes. If a store manager wanted an item, he would go to the retail store designated by the company from whom he was receiving the bribe. He would pick out whatever he wanted, be it a color television or clothing, and the item would then be charged to the account of the company giving the bribe. The company would receive the bill from the legitimate retailer and

would then prepare false invoices showing that goods had been delivered to the manager's store when in fact no such goods had been delivered. In the course of doing so, the bribing company would add a profit. In the case of Hilles that was their standard 30% markup on office supplies. Therefore, for example, if a pool table cost $1,000.00, Hilles would add 30% on and bill the store for $1,300.00 worth of goods that were never delivered. The store manager would then forward the false billing to G.S.A. for payment.

At this point a crucial question arises which goes back to the initial part of the investigation which was why the audit at store No. 17 in June of 1977 fail to show any irregularities. The reason was that the stores operated on a cash inventory basis and not on an item inventory basis. Simply stated, a manager was only responsible to show $20,000.00 in sales if he had made $20,000.00 in purchases on his BPA's or his goose contracts. The store managers devised two methods to cover passing false invoices through the system. One way was rather simple and related to the requirement that the G.S.A. stores were non-profit, that is, the cost of purchase to the store was supposed to be the same as the sales price to the purchasing government agencies. The managers, to cover their purchases of goods that had in fact never been delivered, would charge their customers an extra dollar or two per skid of xerox paper, resulting in a surplus in their inventory account. That surplus would then be balanced against the false invoices and mathematically you would have an even equation. Because of the cash method of inventory, it did not matter whether they bought $20,000.00 worth of pencils and sold $20,000.00 worth of xerox paper. There was no way of finding the

discrepancy absent an item inventory.

The second method the managers used to cover their inventory and accounting requirements was to enroll the purchasers of the goods in the bribery scheme. Frank Ellis at store No. 17 was particularly active in this regard and he enrolled half of the purchasers from Fort Meade. Those individuals would come in and would sign any invoice for goods placed before them by Frank Ellis and in return they would receive tires, or riding lawnmowers or color televisions. The employees of these various agencies did not have to account within their own agencies for what was returned as, at most of these agencies, the only check made was a dollar check. It would be difficult to say that the blame lies with these other agencies because it might simply be too much to have every delivery of yellow pads into the United States Attorney's Office accounted for or counted by some attorney to make sure that the administrative person is doing his or her job.

A crucial fact which has been little reported in the press and probably little understood about the entire

G.S.A. "scandal" was that this entire scheme never cost G.S.A. a penny. The money lost was not lost by G.S.A. but was rather lost by the agencies who were either overbilled or whose own employees became involved in the scheme and signed their own false invoices to keep the managers happy and the bribes flowing in their direction.

That

One of the failures of the G.S.A. system was that, even when they did an item inventory, it was their practice to give notice to the stores of approximately one week's duration. What the managers would do to cover their inventories, because so many were involved in the scheme, was simply to send out their trucks and shift inventories from store to store. We should note, that in the course of doing our initial interviews, the G.S.A. supervisor of federal supplies with Region 3 made a request that, because we were closing the stores in the process of interviewing the employees, they would like to know the night before so they could have a shift ready to go to keep the store open. particular individual, Jerald Sternberg, then did exactly what he did with all of the inventories and audits ever held: he gave the store notice before the agents ever arrived and the next morning, when the agents appeared, all the surprise was gone because Sternberg had called to tell them that the agents were on the way. During the intensive interviewing of all the store managers, there was one supervisor who told several of the people to keep their mouths shut and it would all go away. Unfortunately, a strong obstruction case could never be made against that individual, and, interestingly enough, he resigned within a month or two of the time he made those statements. Other than that there was little that hindered our investigation. On an amusing note the American Federation of Government Employees in the midst of these investigations, sometime in June or July of 1978, offered a seminar at the G.S.A. auditorium for G.S.A. employees on knowing their rights, when they should remain silent and when to seek an attorney. We gave some thought to investigating that but, quite frankly, did not have the time to spend.

As a result of the interviews, a icasonable number of the targets obtained counsel. Those who did not were sent target letters advising them that they should obtain counsel. What ensued was an intensive period of plea bargaining with all those who were involved. Because of the number of individuals being dealt with, it was the opinion of our office that a uniform policy on plea bargaining had to be established. The initial decision was an easy one which was that everyone would receive recommendations of incarceration. It was our view that government employees who steal from the government should be incarcerated and privately most defense counsel did not argue with that view. Ranges of recommended sentences were than established depending on the person's position within the investigation.

Store managers, who were our greatest volume, were to receive recommendations of between 6 and 18 months incarceration depending on the dollar amount of items received by them for their personal use. Assistant store managers who had received gifts at the grace of the store managers and government employees who worked for other agencies were to receive' recommendations of from 30 to 90 days incarceration depending upon their dollar figure as well as the factor of on how many separate occasions they took bribes. Finally, as to the corporate executives, it was determined that that would have to be on a company -by--company basis and, as to both Hilles executives, recommendations of two years incarceration were finally settled upon. Included within all these recommendations was the understanding that these individuals would cooperate fully, completely and honestly. As to several individuals, offers were made which did not include coop ration but which included substantially greater recommendations. Once the sentencing recommendation scheme was established it became a non debatable point during discussion with counsel other than as to which category their client would fall in. In some ways it made the plea bargain ng process casier and did offer certain assurances to the defendants that they were being treated in comparison to their co-defendants. An important fact to us was that, during the course of the fraud, each of the store managers knew who among them were the big takers and who were the small takers. It was imperative to us that there be an appearance of fairness even though with seven separate judys doing the sentencing we knew that there was no way to assure actual fairness.

Included within the process was a substantial amount of pre-indictment discovery. Normally, this would obviously not be a course to be followed. It was done in this case, however, because no attempts were made to convince counsel that the cases were good or that their clients should plead. They were presented with the facts and the opportunity to plead or not. Under the circumstances of this investigation that approach proved most effective as the first round of indictments included 18 individuals who had agreed to enter pleas and only one of those individuals failed to enter a plea at his arraignment.

In this particular investigation we did not have a lot of resistance by defense attorneys to giving up multiple clients when a conflict existed. One attorney chose to represent three defendants out of the five who came to see him because those three indicated an immediate desire to cooperate and plead guilty. When other attorneys called representing more than one individual, they were immediately informed of any possible conflicts and, in each case where there appeared to be a problem, they willingly gave up all but one of their potential clients. During the plea bar