19196 1 BEFORE THE 2 POSTAL RATE COMMISSION 3 - - - - - - - - - - - - - - - X 4 In the Matter of: : 5 POSTAL RATE AND FEE CHANGES : Docket No. R97-1 6 - - - - - - - - - - - - - - - X 7 8 Third Floor Hearing Room 9 Postal Rate Commission 10 1333 H Street, N.W. 11 Washington, D.C. 20268 12 13 Volume 36 14 Friday, March 20, 1998 15 16 The above-entitled matter came on for hearing, 17 pursuant to notice, at 9:30 a.m. 18 19 BEFORE: 20 HON. EDWARD J. GLEIMAN, CHAIRMAN 21 HON. W. H. "TREY" LeBLANC, III, COMMISSIONER 22 HON. GEORGE W. HALEY, COMMISSIONER 23 HON. GEORGE A. OMAS, COMMISSIONER 24 25 19197 1 APPEARANCES: 2 On behalf of the United States Postal Service: 3 SUSAN DUCHEK, ESQUIRE 4 ERIC KOETTING, ESQUIRE 5 RICHARD COOPER, ESQUIRE 6 MICHAEL TIDWELL, ESQUIRE 7 ANNE REYNOLDS, ESQUIRE 8 DAVID RUBIN, ESQUIRE 9 KENNETH N. HOLLIES, ESQUIRE 10 SCOTT L. REITER, ESQUIRE 11 ANTHONY ALVERNO, ESQUIRE 12 United States Postal Service 13 475 L'Enfant Plaza West, SW 14 Washington, D.C. 20260 15 16 On behalf of American Business Press: 17 DAVID STRAUS, ESQUIRE 18 Thompson Coburn 19 700 14th Street, NW, Suite 900 20 Washington, D.C. 20005 21 (202) 508-1013 22 fax (202) 508-1010 23 24 25 19198 1 APPEARANCES: [continued] 2 On behalf of the Association of Alternate Postal Systems: 3 BONNIE S. BLAIR, ESQUIRE 4 Thompson Coburn 5 700 14th Street, NW, Suite 900 6 Washington, D.C. 20005 7 (202) 508-1003 8 fax (202) 508-1010 9 10 On behalf of Nashua Photo, Inc.; District Photo, Inc.; 11 Mystic Color Lab; Seattle FilmWorks, Inc.; ValPak Direct 12 Marketing Systems, Inc.; ValPak Dealers' Association; Carol 13 Wright Promotions: 14 WILLIAM J. OLSON, ESQUIRE 15 ALAN WOLL, ESQUIRE 16 JOHN S. MILES, ESQUIRE 17 JOHN F. CALLENDER, JR., ESQUIRE 18 William J. Olson, P.C. 19 8180 Greensboro Drive, Suite 1070 20 McLean, VA 22102-3823 21 (703) 356-5070 22 fax (703) 356-5085 23 24 25 19199 1 APPEARANCES: [continued] 2 On behalf of Readers Digest Association, Parcel Shippers 3 Association: 4 TIMOTHY J. MAY, ESQUIRE 5 Patton Boggs, LLP 6 2550 M Street, NW 7 Washington, D.C. 20037 8 (202) 457-6050 9 10 On behalf of Advertising Mail Marketing Association: 11 IAN D. VOLNER, ESQUIRE 12 Venable, Baetjer, Howard & Civilletti 13 1201 New York Avenue, NW 14 Washington, D.C. 20005 15 (202) 962-4814 16 fax (202) 962-8300 17 18 On behalf of the Dow Jones & Company, Inc.: 19 SAM BEHRENDS, ESQUIRE 20 MICHAEL F. McBRIDE, ESQUIRE 21 LeBoeuf, Lamb, Greene & Macrae 22 1875 Connecticut Avenue, NW 23 Washington, D.C. 20009 24 (202) 986-8018 25 fax (202) 986-8102 19200 1 APPEARANCES: [continued] 2 On behalf of the Major Mailers Association: 3 RICHARD LITTELL, ESQUIRE 4 1220 19th Street, NW, Suite 400 5 Washington, D.C. 20036 6 (202) 466-8260 7 8 On behalf of the Office of Consumer Advocate: 9 SHELLEY S. DREIFUSS, ESQUIRE 10 KENNETH E. RICHARDSON, ESQUIRE 11 Office of the Consumer Advocate 12 Postal Rate Commission 13 1333 H Street, NW, Suite 300 14 Washington, D.C. 20268 15 16 On behalf of the United Parcel Service: 17 JOHN E. McKEEVER, ESQUIRE 18 Piper & Marbury 19 3400 Two Logan Square 20 18th and Arch Streets 21 Philadelphia, PA 19103 22 (215) 656-3310 23 fax (215) 656-3301 24 25 19201 1 APPEARANCES: [continued] 2 On behalf of Hallmark Cards, Incorporated: 3 DAVID F. STOVER, ESQUIRE 4 2070 S. Columbus Street, Suite 1B 5 Arlington, VA 22206 6 (703) 998-2568 7 fax (703) 998-2987 8 9 On behalf of ADVO, Inc.: 10 JOHN M. BURZIO, ESQUIRE 11 THOMAS W. McLAUGHLIN, ESQUIRE 12 Burzio & McLauglin 13 1054 31st Street, NW, Suite 540 14 Washington, D.C. 20007 15 (202) 965-4555 16 fax (202) 965-4432 17 18 On behalf of Time Warner, Inc.: 19 JOHN M. BURZIO, ESQUIRE 20 TIMOTHY L. KEEGAN, ESQUIRE 21 1054 31st Street, NW, Suite 540 22 Washington, D.C. 20007 23 (202) 965-4555 24 fax (202) 965-4432 25 19202 1 APPEARANCES: [continued] 2 On behalf of the Direct Marketers Association: 3 DANA T. ACKERLY, II, ESQUIRE 4 MICHAEL D. BERGMAN, ESQUIRE 5 Covington & Burling 6 1201 Pennsylvania Avenue, NW 7 Washington, D.C. 20016 8 (202) 662-5296 9 fax (202) 778-5296 10 11 On behalf of the Newspaper Association of America: 12 WILLIAM B. BAKER, ESQUIRE 13 ALAN R. JENKINS, ESQUIRE 14 MICHAEL YOURSHAW, ESQUIRE 15 Wiley, Rein & Fielding 16 1776 K Street, NW 17 Washington, D.C. 20006 18 (202) 429-7255 19 fax (202) 429-7049 20 21 ROBERT J. BRINKMANN 22 Newspaper Association of America 23 529 14th Street, NW, Suite 440 24 Washington, D.C. 20045-1402 25 19203 1 APPEARANCES: [continued] 2 On behalf of the McGraw-Hill Companies, Inc.: 3 TIMOTHY W. BERGIN, ESQUIRE 4 Squire, Sanders & Dempsey 5 1201 Pennsylvania Avenue, NW, Suite 500 6 P.O. Box 407 7 Washington, D.C. 20044 8 (202) 626-6608 9 fax (202) 626-6780 10 11 On behalf of the Mail Order Association of America: 12 DAVID C. TODD, ESQUIRE 13 Patton Boggs, LLP 14 2550 M Street, NW 15 Washington, D.C. 20037 16 (202) 457-6410 17 fax (202) 457-6513 18 19 On behalf of David B. Popkin: 20 DAVID B. POPKIN 21 P.O. Box 528 22 Englewood, NJ 07631-0528 23 (201) 569-2212 24 fax (201) 569-2864 25 19204 1 APPEARANCES: [continued] 2 On behalf of the Magazine Publishers of America: 3 JAMES R. CREGAN, ESQUIRE 4 Magazine Publishers of America 5 1211 Connecticut Avenue, NW, Suite 610 6 Washington, D.C. 20036 7 (202) 296-7277 8 fax (202) 296-0343 9 10 On behalf of the Alliance of Nonprofit Mailers: 11 JOEL T. THOMAS, ESQUIRE 12 11326 Dockside Circle 13 Reston, VA 20191 14 (703) 476-4646 15 fax (703) 620-2338 16 17 On behalf of the National Newspaper Association: 18 TONDA F. RUSH, ESQUIRE 19 King & Ballon 20 P.O. Box 50301 21 Arlington, VA 22205 22 (703) 534-5750 23 fax (703) 534-5751 24 25 19205 1 APPEARANCES: [continued] 2 On behalf of the National Newspaper Association: 3 [continued] 4 SENNY BOONE 5 National Newspaper Association 6 1525 Wilson Boulevard, Suite 550 7 Arlington, VA 22209 8 (703) 907-7900 9 10 On behalf of the National Federation of Nonprofits: 11 CAROLYN EMIGH, ESQUIRE 12 Nonprofit Service Group 13 815 15th Street, NW, Suite 822 14 Washington, D.C. 20005 15 (202) 628-4380 16 17 On behalf of the Florida Gift Fruit Shippers Association: 18 M.W. WELLS, JR., ESQUIRE 19 Maxwell W. Wells, Jr., P.A. 20 105 E. Robinson Street, Suite 201 21 Orlando, FL 32801 22 (407) 422-8250 23 fax (407) 422-8262 24 25 19206 1 APPEARANCES: [continued] 2 On behalf of the Recording Industry Association of America, 3 and Advertising Mail Marketing Association: 4 N. FRANK WIGGINS, ESQUIRE 5 Venable, Baetjer, Howard & Civiletti, L.L.P. 6 1201 New York Avenue, NW 7 Washington, D.C. 8 (202) 962-4957 9 10 On behalf of Edison Electric Institute: 11 R. BRIAN CORCORAN, ESQUIRE 12 Oliver & Oliver, P.C. 13 1090 Vermont Avenue, NW, Suite 800 14 Washington, D.C. 20005 15 (202) 371-5656 16 fax (202) 289-8113 17 18 On behalf of American Business Press: 19 STEPHEN FELDMAN, ESQUIRE 20 Ramsey, Cook, Looper & Kurlander 21 c/o Thompson Coburn 22 700 14th Street, NW, Suite 900 23 Washington, D.C. 20005 24 (202) 508-1022 25 fax (202) 508-1010 19207 1 APPEARANCES: [continued] 2 On behalf of Douglas F. Carlson: 3 DOUGLAS F. CARLSON 4 P.O. Box 12574 5 Berkeley, CA 94712-3574 6 (510) 597-9995 7 8 On behalf of the Alliance of Non Profit Mailers: 9 DAVID M. LEVY, ESQUIRE 10 Sidley & Austin 11 1722 I Street, NW 12 Washington, D.C. 20006-3704 13 (202) 736-8214 14 15 On behalf of the National Association of Presort Mailers: 16 HENRY HART, ESQUIRE 17 Hazel & Thomas 18 P.O. Box 820 19 Alexandria, VA 22313 20 (703) 838-5153 21 fax (703) 836-8062 22 23 24 25 19208 1 APPEARANCES: [continued] 2 On behalf of Brooklyn Union Gas Company: 3 MICHAEL HALL, ESQUIRE 4 Cullen & Dykman 5 1225 19th Street, NW 6 Washington, D.C. 20036 7 (202) 223-8890 8 9 On behalf of Niagara Telephone Company: 10 TIMOTHY E. WELCH, ESQUIRE 11 Hill & Welch 12 1330 New Hampshire Avenue, NW, Suite 113 13 Washington, D.C. 20036 14 (202) 775-0070 15 fax (202) 775-9026 16 17 On behalf of the Coalition of Religious Press Associations: 18 JOHN STAPERT 19 Associated Church Press 20 18653 N. 41st Place 21 Phoenix, AZ 85024-3759 22 (602) 569-6371 23 fax (602) 569-6180 24 25 19209 1 APPEARANCES: [continued] 2 On behalf of the Greeting Card Association: 3 ALAN R. SWENDIMAN, ESQUIRE 4 Jackson & Campbell, P.C. 5 1120 20th Street, NW, Suite 300 South 6 Washington, D.C. 20036-3437 7 (202) 457-1645 8 fax (202) 457-1617 9 10 On behalf of LabOne, Inc., Osborn Laboratories, Inc., and 11 Clinical Reference Laboratory, Inc.: 12 JOSEPH C. BENAGE, ESQUIRE 13 Hillix, Brewer, Hoffhaus, Whittaker & Wright 14 2420 Pershing Road 15 Kansas City, MO 64108-2574 16 (816) 221-0355 17 fax (816) 421-2896 18 19 20 21 22 23 24 25 i 1 C O N T E N T S 2 WITNESS DIRECT CROSS REDIRECT RECROSS 3 RITA D. COHEN 4 BY MR. CREGAN 19212 5 BY MR. McKEEVER 19235 6 BY MR. KOETTING 19250/19272 7 HALSTEIN STRALBERG 8 BY MR. BURZIO 19276 9 BY MR. KOETTING 19295 10 BY MR. BURZIO 19308 11 CARL G. DEGEN 12 BY MR. KOETTING 19310 13 BY MR. WIGGINS 19367 14 BY MR. STRAUS 19373 15 BY MR. KEEGAN 19455 16 BY MR. KOETTING 19467 17 BY MR. KEEGAN 19469 18 STEPHEN E. SELLICK 19 BY MR. McKEEVER 19472 20 BY MR. GOLD 19491 21 ROGER C. PRESCOTT 22 BY MR. TODD 19503 23 BY MR. OLSON 19547 24 BY MR. TODD 19580 25 BY MR. OLSON 19581 ii 1 C O N T E N T S [continued] 2 WITNESS DIRECT CROSS REDIRECT RECROSS 3 LESLIE M. SCHENK 4 BY MR. REYNOLDS 19583 5 BY MR. LEVY 19649 6 GARY M. ANDREW 7 BY MR. TODD 19661 8 BY MR. CORCORAN 19730 9 BY MR. LITTELL 19772 10 BY MR. BAKER 19778 11 BY MR. COOPER 19806 12 BY MR. BAKER 19808 13 JOHN L. CLARK 14 BY MR. OLSON 19810 15 BY MR. McKEEVER 19823 16 BY MR. OLSON 19849 17 BY MR. McKEEVER 19852 18 BY MR. OLSON 19854 19 BY MR. McKEEVER 19854 20 21 DOCUMENTS TRANSCRIBED INTO THE RECORD: PAGE 22 Rebuttal Testimony and Exhibits of Rita 23 D. Cohen, MPA-RT-1 19215 24 Cross-Examination Exhibit No. USPS/MPA-RT-XE-1 19260 25 iii 1 DOCUMENTS TRANSCRIBED INTO THE RECORD: [continued] PAGE 2 Rebuttal Testimony and Exhibits of Halstein 3 Stralberg, TW-RT-1 19278 4 Rebuttal Testimony and Exhibits of Carl G. 5 Degen, USPS-RT-6 19312 6 Cross-Examination Exhibit No. ABP-XE-1 19422 7 Cross-Examination Exhibit No. ABP-XE-2 19427 8 Cross-Examination Exhibit No. ABP-XE-3 19437 9 Cross-Examination Exhibit ABP-XE-4 19451 10 Rebuttal Testimony and Exhibits of 11 Stephen E. Sellick, UPS-RT-1 19474 12 Revised Response of UPS Witness Sellick to 13 POIR-16 19487 14 Rebuttal Testimony and Exhibits of Roger 15 C. Prescott, MOAA-RT-1 19505 16 Rebuttal Testimony and Exhibits of Leslie 17 M. Schenk, USPS-RT-22 19587 18 Rebuttal Testimony and Exhibits of Gary M. 19 Andrew, MOAA, et al.-RT-1 19663 20 Cross-Examination Exhibit No. ABA/EEI/NAPM-XE-1 19769 21 Cross-Examination Exhibit No. ABA/EEI/NAPM-XE-2 19770 22 Cross-Examination Exhibit No. ABA/EEI/NAPM-XE-3 19771 23 Cross-Examination Exhibit No. 24 NAA/MOAA et al.-RT-1-XE-1 19797 25 iv 1 DOCUMENTS TRANSCRIBED INTO THE RECORD: [continued] PAGE 2 Cross-Examination Exhibit No. 3 NAA/MOAA et al.-RT-1-XE-2 19798 4 Rebuttal Testimony and Exhibits of John L. 5 Clark, CTC-RT-1 19813 6 Excerpt from the United States Postal 7 Service Household Diary Study, Fiscal 8 Year 1996, Executive Summary 19845 9 10 E X H I B I T S 11 EXHIBITS AND/OR TESTIMONY IDENTIFIED RECEIVED 12 Rebuttal Testimony and Exhibits of 13 Rita D. Cohen, MPA-RT-1 19214 19214 14 Cross-Examination Exhibit No. 15 USPS/MPA-RT-XE-1 19258 19258 16 Rebuttal Testimony and Exhibits of 17 Halstein Stralberg, TW-RT-1 19277 19277 18 Rebuttal Testimony and Exhibits of 19 Carl G. Degen, USPS-RT-6 19311 19311 20 Cross-Examination Exhibit No. ABP-XE-1 19420 19421 21 Cross-Examination Exhibit No. ABP-XE-2 19426 19426 22 Cross-Examination Exhibit No. ABP-XE-3 19436 19436 23 Cross-Examination Exhibit ABP-XE-4 19450 19450 24 Rebuttal Testimony and Exhibits of 25 Stephen E. Sellick, UPS-RT-1 19472 19472 v 1 E X H I B I T S [continued] 2 EXHIBITS AND/OR TESTIMONY IDENTIFIED RECEIVED 3 Revised Response of UPS Witness 4 Sellick to POIR-16 19486 19486 5 Rebuttal Testimony and Exhibits of 6 Roger C. Prescott, MOAA-RT-1 19504 19504 7 Rebuttal Testimony and Exhibits of 8 Leslie M. Schenk, USPS-RT-22 19585 19585 9 Rebuttal Testimony and Exhibits of 10 Gary M. Andrew, MOAA, et al.-RT-1 19662 19662 11 Cross-Examination Exhibit No. 12 ABA/EEI/NAPM-XE-1 19739 13 Cross-Examination Exhibit No. 14 ABA/EEI/NAPM-XE-2 19742 15 Cross-Examination Exhibit No. 16 ABA/EEI/NAPM-XE-3 19750 17 Cross-Examination Exhibit No. 18 NAA/MOAA et al.-RT-1-XE-1 19785 19795 19 Cross-Examination Exhibit No. 20 NAA/MOAA et al.-RT-1-XE-2 19788 19795 21 Rebuttal Testimony and Exhibits of 22 John L. Clark, CTC-RT-1 19811 19811 23 24 25 vi 1 E X H I B I T S [continued] 2 EXHIBITS AND/OR TESTIMONY IDENTIFIED RECEIVED 3 Excerpt from the United States 4 Postal Service Household Diary 5 Study, Fiscal Year 1996, 6 Executive Summary 19844 19844 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19210 1 P R O C E E D I N G S 2 [9:30 a.m.] 3 CHAIRMAN GLEIMAN: Good morning. 4 Today should be our final day of hearings on this 5 case, unless, of course, we go beyond midnight, in which 6 case tomorrow will be our final day of hearings. 7 Unfortunately, I'm afraid it may come to that, but hopefully 8 not. 9 We're scheduled to receive testimony in rebuttal 10 to the direct cases of participants other than the Postal 11 Service from Magazine Publishers of America Witness Cohen, 12 Time-Warner Witness Stralberg, United States Postal Service 13 Witnesses Degen and Schenk, United Parcel Service Witness 14 Sellick, Mail Order Association of America Witness Prescott, 15 and Mail Order Association of America, et al., Witness 16 Andrew, and then, finally, CTC Distribution Services 17 Witnesses Clark. 18 I have several procedural matters to mention 19 before we begin this morning. 20 There is one outstanding procedural issue still to 21 be resolved. 22 On March the 17th, the Alliance of Non-Profit 23 Mailers filed a motion to compel production of mail 24 acceptance logs underlying USPS-RT-22 or, in the 25 alternative, to strike portions of that testimony. 19211 1 The Postal Service filed a response in opposition 2 at noon yesterday, and last night, during breaks of the 3 basketball game, I reviewed that response and also had an 4 opportunity to leaf through Library Reference H-353, the 5 survey forms that are at issue, used by Witness Schenk in 6 preparing her testimony. 7 I'm going to have some questions to ask Ms. Schenk 8 before I rule on the ANM motion, but I am not going to do 9 that until all the appropriate counsel are in the room, 10 perhaps when Ms. Schenk comes to the stand later today. 11 Mr. Koetting, would you inform Postal Service 12 counsel who is representing Ms. Schenk that I will be asking 13 her questions concerning the use of actual copies of Form 14 8075, the so-called disqualification logs referred to in her 15 testimony that have now become the subject of the ANM motion 16 to compel production? 17 MR. KOETTING: I'd be happy to do that, Mr. 18 Chairman. 19 CHAIRMAN GLEIMAN: During Tuesday's hearing, I 20 addressed the issue of incorporating additional materials 21 into the evidentiary record and set March 27th as the 22 deadline for filing such motions. 23 If you have additional materials for inclusion in 24 the record, please file the appropriate motion, accompanied 25 by two copies of the designations by close of business 19212 1 Friday, March the 27th. 2 March the 27th is also the date for filing 3 transcript corrections for this round of hearings. 4 As I mentioned earlier, if a transcript correction 5 is related to the final round of hearings and is central to 6 an argument in an initial brief, please identify that 7 situation in the text or in footnote to the brief. 8 Does any participant have a procedural matter that 9 they would like to raise at this point in time? 10 [No response.] 11 CHAIRMAN GLEIMAN: If not, then we'll move on to 12 our first witness. 13 Our first witness today is appearing on behalf of 14 the Magazine Publishers of America, and Ms. Cohen is already 15 under oath in this proceeding. 16 Mr. Cregan, if you would introduce your witness 17 and enter her rebuttal testimony into the record. 18 MR. CREGAN: Thank you, Mr. Chairman. 19 Whereupon, 20 RITA D. COHEN, 21 a rebuttal witness, was called for examination by counsel 22 for the Magazine Publishers of America and, having been 23 previously duly sworn, was further examined and continued to 24 testify as follows: 25 DIRECT EXAMINATION 19213 1 BY MR. CREGAN: 2 Q Ms. Cohen, I'm handing you two copies of a 3 document designated MPA-RT-1, Rebuttal Testimony of Rita D. 4 Cohen on Behalf of Alliance of Non-Profit Mailers, American 5 Business Press, Coalition of Religious Press Associations, 6 Doe-Jones and Company, Inc., Magazine Publishers of America, 7 the McGraw-Hill Companies, Inc., National Newspaper 8 Association, and Time-Warner, Inc. 9 Are you familiar with this document? 10 A Yes, I am. 11 Q Was it prepared by you or under your supervision? 12 A Yes, it was. 13 Q Do you have any corrections or revisions today? 14 A No, I do not. 15 Q If you were to testify today orally, would your 16 testimony be the same? 17 A Yes. 18 MR. CREGAN: Mr. Chairman, I will hand two copies 19 of Ms. Cohen's testimony, designated MPA-RT-1, to the 20 reporter and ask its admission into evidence. 21 CHAIRMAN GLEIMAN: Are there any objections? 22 [No response.] 23 CHAIRMAN GLEIMAN: Hearing none, Ms. Cohen's 24 testimony and exhibits are received into evidence, and I 25 direct that they be transcribed into the record at this 19214 1 point. 2 [Rebuttal Testimony and Exhibits of 3 Rita D. Cohen, MPA-RT-1, was 4 received into evidence and 5 transcribed into the record.] 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19235 1 CHAIRMAN GLEIMAN: Two participants have requested 2 oral cross examination of Witness Cohen, United Parcel 3 Service and United States Postal Service. 4 Does any other participant wish to cross examine 5 the witness? 6 [No response.] 7 CHAIRMAN GLEIMAN: If not, then Mr. McKeever, when 8 you're ready, you can begin. 9 MR. McKEEVER: Thank you, Mr. Chairman. 10 CROSS EXAMINATION 11 BY MR. McKEEVER: 12 Q Good morning, Ms. Cohen. 13 A Good morning. 14 Q Ms. Cohen, could you turn to page 1 of your 15 testimony, please? On that page, at lines 20 to 21, you 16 state that Mr. Sellick's proposed distribution method cannot 17 be used without also using Witness Bradley's results. Do 18 you see that? 19 A Yes, I do. 20 Q Is it also true that it doesn't make sense to use 21 Dr. Bradley's variability results without, at the same time, 22 using Mr. Degen's proposed distribution methods? 23 A Well, I think that the attribution and 24 distribution methodologies used must be consistent. I think 25 Dr. Christensen put it pretty well in his testimony on 19236 1 rebuttal. I would say that you have to use a similar 2 framework. 3 I don't think -- as you know, in my direct 4 testimony, I have suggested revisions to Witness Degen, but 5 -- so, I don't think it had to be done exactly as Witness 6 Degen had proposed it, but with the changes that I 7 recommend, I do believe that those two would be acceptable 8 together. 9 Q It is your testimony that certain LIOCATT 10 distribution methods can be used while also using Witness 11 Bradley's variability results. Is that right? 12 A I did not advocate going back to the LIOCATT 13 method. We advocated using Witness Degen as a starting 14 point, making the necessary modifications, and then we had a 15 consistent set of attribution and distribution. 16 Q Don't you use some of the LIOCATT methods to 17 distribute at least some costs? 18 A Well, for the mixed mail, where you really have to 19 make a determination of the contents, we were unhappy with 20 the item distribution, and we really substituted what was 21 the basic function and CAG approach from the LIOCATT system, 22 but within the same framework that Witness Degen had 23 proposed. 24 Q What do you mean by "within the same framework"? 25 A Well, we didn't ignore the cost pools -- I mean, 19237 1 in terms of setting up the variabilities, the cost pools, 2 the consistency with cost drivers, all that would stay the 3 same, but there are obviously, as you go forward in your 4 distribution process, points where you have to make 5 assumptions in terms to get to sub-class, and we felt using 6 the basic function and CAG was a better approach at the 7 current time. 8 Q Okay. Thank you. 9 Could you turn to page 4, please? 10 A Yes. 11 Q There, in footnote 2, you state that you cannot 12 quantify -- this is at the end of the footnote -- that you 13 cannot quantify how fast it has been increasing. What is 14 the "it" that you are referring to there? 15 A Well, the particular grouping, they made some 16 changes this time in terms of some things moving from not 17 handling into mixed and some things moving from mixed to not 18 handling. So, it was slightly different. 19 I did notice that Witness Degen, in his rebuttal, 20 did examine another subset other than the one I had, which 21 was overhead, and he looked at the set of the shape-specific 22 and mixed shapes not handling category and showed the growth 23 in that, as well. 24 So, I think we have that information on the 25 record, also. 19238 1 But in terms of the particular movement, where 2 things that involved an item in container moved from not 3 handling to mixed and something else moved back, which I 4 will think of in a minute, but it's slightly different. 5 Q I apologize, Ms. Cohen, but I lost in there what 6 the "it" is. What is it that you could quantify that you're 7 referring to in that sentence? 8 A That the not handling category, as we're using it 9 in this case, is somewhat different than the not handling 10 category that we used in the preceding case in terms of what 11 codes are in there and what parts of codes. 12 Q Okay. 13 So, you couldn't identify the difference between 14 not handling in the last case and not handling in the case? 15 Do I understand you correctly? 16 A You can identify comparable pieces, okay? So, 17 like what Witness Degen did is he looked at the 5610 to 5750 18 series. So, you can look at that. 19 And I had shown what happened with what was 20 traditionally called overhead, which included the breaks and 21 the clocking in and out and the moving empty equipment, but 22 there are changes that -- with regard to empty equipment and 23 the not handling that make this category this time slightly 24 different, and my direct testimony -- I know it's confusing, 25 and it was confusing -- tried to make it as clear as 19239 1 possible -- I explained what, this time, the category had, 2 but it's somewhat different. 3 I don't know how better to do it. 4 Q Okay. I'll just ask you one more question. You 5 said the category, you tried to identify what the category 6 included. What is the category? 7 A The not handling as defined by Witness Degen in 8 this proceeding. 9 Q Okay. Thank you. 10 Still on page 4, at lines 11 to 15, you mention, 11 among other things, sorting outside sacks mechanically, that 12 activity. 13 A Actually, that should say outsides and sacks, but 14 -- sorry. 15 Q Outside? 16 A Outsides and sacks. 17 Q Okay. That's helpful. Could you tell me -- could 18 you describe for me that activity? 19 A Well, it is -- well, actually, it depends on where 20 it is. Witness Bradley, actually, for this, used a proxy 21 from the BMCs, but I think that, at MODS facilities, there 22 are not that many, actually, that have this operation, but 23 some do have sack sorting and outside sorting. 24 It is -- it can have a few different 25 configurations, but it can be mostly a conveyor system to 19240 1 get the sacks and outsides to a few different locations in 2 the facility or more traditional sack sort of the type at a 3 BMC. 4 Q Now, you note in lines 11 to 15 that not handling 5 costs are 50-percent higher in the mechanical sack sorting 6 activity than they are in the mechanical parcel sorting 7 activity. Is that right? 8 A Yes. 9 Q And then, in the BMC operations that you discuss 10 at lines 20 to 21, it's again the case that not handling 11 costs are 50-percent higher in the case of the sack sorting 12 machine than in the case of the parcel sorting machine. Is 13 that right? 14 A Yes. I did notice that those are actually in 15 comparable directions. I'm not sure that that makes me feel 16 anymore comfortable about it. 17 I think we found -- I tried to illustrate a few 18 cases. There are certainly others. In your own 19 interrogatory to me, we compared LSMs and small parcel and 20 bundle sorters, which had very different not handling 21 percents, as well. 22 These were some examples, but there are others. 23 Q And they move in the same direction at about the 24 same -- with about the same difference. 25 A No, not necessarily. 19241 1 Q In the examples you cite. 2 A In the two -- the BMC and the MODS, which have 3 similar kinds of mechanical sorting here, yes, it is in the 4 same direction. 5 Q Okay. 6 Ms. Cohen, could you turn to page 6 of your 7 testimony, please? 8 A Okay. 9 Q And in particular, lines 19 to 23. 10 A Okay. 11 Q There you criticize Mr. Sellick for not mentioning 12 in his written testimony the conclusion of the 1992 Foster 13 Associates report that, quote, "additional field operating 14 data are necessary to determine the proper causative 15 attribution of the break and sub-class costs in question and 16 those other costs which are presently attributed as mixed 17 mail or overhead activities," and you cite pages 14256 to 18 14258 of Mr. Sellick's oral cross examination there. Is 19 that correct? 20 A Yes. 21 Q Isn't it a fact that, despite your counsel's 22 attempt to cut Mr. Sellick off in this middle of his answer 23 to your counsel's question on that point, Mr. Sellick later 24 went on to state, in the next page, after the ones you cite, 25 that IOCS was modified to gather more information about 19242 1 exactly the question raised in that part of the 1992 study? 2 I do have the transcript here if you would like to 3 refer to it. 4 A That doesn't affect the field operating data. 5 Q Modifications to IOCS to collect data? 6 A Oh, that was clearly not what was intended here. 7 Q What was not intended where? 8 A Well, modifications to IOCS is not what this talks 9 about. It talks about getting into the field and looking at 10 operating data to determine what's actually going on in 11 terms of staffing, operations. You would look at the IOCS 12 as part of that. It's actually pretty similar to what I 13 think we're going to start, hopefully, soon, to look at for 14 periodicals. 15 Q Isn't it, in fact, the case that IOCS was modified 16 to ask some additional questions to obtain data associating 17 breaks with the activity that the employee was on break 18 from? 19 A Yes. 20 Q And that was done after the 1992 report. Is that 21 right? 22 A I don't remember exactly when that was -- when 23 that went in. 24 Q Well, there is a variable F94-19. Are you 25 familiar with that at all? 19243 1 A Okay. I've got it in my book. Do you want me to 2 look at it? 3 Q Yes, please. 4 A 92-19? 5 Q 94-19. 6 A Okay. Which question is it for IOCS? 7 Q It's in 18, I believe, but I'll have that for you 8 in just one moment. Actually, this is -- you're looking at 9 the IOCS booklet? What library reference do you have? 10 A Twenty-three. 11 Q Twenty-three. Okay. 12 A IOCS dictionary. 13 Q If you look at a page -- it's actually page 38 of 14 that document, but it's labeled page 20 of 36 on the bottom 15 of the page. That refers to F94-19. 16 A Okay. 17 Q And that says title of item, type distribution on 18 break from? Is that correct? 19 A Yes. 20 Q Okay. 21 Now, do you know what the structure of that SAS 22 item number means? Does the 94 symbolize anything? 23 A I am not a SAS programmer. I'm sorry. 24 Q Doesn't that -- let me see if I can help you a 25 little bit, and if you don't know, you don't know. Doesn't 19244 1 the 94 indicate when that question or when that particular 2 item was put into the program, the year, 1994? 3 A I'll accept it subject to check. 4 Q Okay. You don't know one way or the other. 5 A No. I didn't use those codes. I mean they're 6 used by my people, but I don't know. 7 Q Okay. 8 Ms. Cohen, is it your testimony that Mr. Degen's 9 distribution keys are piece handlings for the various 10 sub-classes in the various cost pools he deals with? 11 A That his distribution keys are? 12 Q Yes. 13 A No. 14 Q Is that your testimony? 15 A No. 16 Q Pardon me? 17 A No. 18 Q Okay. 19 Now, allied operations handle pre-sorted mail, do 20 they not? Or I should say pre-sort mail is handled in 21 allied operations. Is that correct? 22 A It can be. 23 Q Well, you say it can be. It is on occasion, at 24 least? 25 A Yes. 19245 1 Q Okay. 2 So, it's more than it theoretically can be. It 3 actually is at times. Is that what we're saying? I want to 4 make sure we're understanding each other. 5 A Well, I mean mail that is pre-sorted will 6 sometimes be handled in various ways at allied operations. 7 Q Okay. 8 Piece sortation operations do not generally handle 9 pre-sorted mail. Is that correct? 10 A Well, frequently you'll have an incoming 11 secondary. It would mostly bypass outgoing operations. 12 Some to carrier route would bypass the incoming secondary at 13 all. It really depends on the pre-sort. 14 Q Okay. It is handled in piece sortation operations 15 to a lesser extent than non-pre-sorted mail. Isn't that 16 true? 17 A Yes. 18 Q Now, if an IOCS data collector at an allied 19 operation observes mixed pre-sorted mail, do you know what 20 activity code would be assigned to that observation in IOCS? 21 A I'm sorry. Tell me again? 22 Q Sure. If an IOCS data collector takes a tally at 23 an allied operation and he observes mixed pre-sorted mail, 24 do you know what activity code would be assigned to that 25 observation? 19246 1 A Well, do you mean that the employee is clocked 2 into the allied operations or he's actually at the allied 3 operations? 4 Q Let's assume they're the same for purposes of this 5 question, that he's clocked in to that operation and that's 6 what he's actually doing. 7 A So, you are assuming he is clocked in and he is 8 there. 9 Q Yes. 10 A Okay. They -- well, it would depend on what the 11 mixed mail was. It probably could get a mixed all shapes of 12 5750. 13 Q 5750. Could it get a 5610? 14 A Not if he is at the allied operations. That's 15 only used if there is a predominant shape association. You 16 would get that if the employee was at the distribution 17 operations. 18 Q Only if he was at the distribution operations? 19 A Yes. 20 Q Am I correct that Mr. Degen's distribution method 21 distinguishes large CAG offices which are in MODS from 22 smaller CAG offices which are not MODS offices? 23 A He separates the MODS and non-MODS offices. 24 Q And aren't large CAG offices MODS offices and 25 smaller CAG offices non-MODS offices? 19247 1 A Well, there's some overlap, but all the CAG A 2 offices are in his MODS category. 3 Q Do you know what proportion of the MODS pool of 4 the MODS -- let me start over. 5 Do you know what proportion of the MODS pool costs 6 are in CAG A? 7 A A substantial portion; I don't know the exact 8 number right now. 9 Q Could it be 90 percent or more? Do you know? Is 10 that the right order of magnitude? 11 A I think we were talking about 85 percent for MODS 12 offices in general. I don't know. I'd say it's a lot. I 13 don't know the number. 14 Q Okay. Mr. Degen treats BMC costs separately, 15 doesn't he? 16 A Yes, he does. 17 Q And in LIOCATT, BMCs are their own CAG or cost 18 ascertainment grouping, aren't they? 19 A Yes. 20 Q Now I'd like to ask you a question about non-MODS 21 offices. Do you know what percentage of total cost segment 22 3 costs non-MODS offices account for? 23 A I think if I'm remembering the right number that 24 that was roughly the 85-15 split, but I'd have to check 25 that. 19248 1 Q So you think to the best of your recollection now 2 it's about 15 percent? 3 A I think so. 4 Q Okay. Do you know what activity code is assigned 5 by the IOCS system to an employee who is handling mixed 6 mail? 7 A Is this a general question? I mean -- 8 Q Yes. 9 A Let me ask the context. 10 Any IOCS data -- 11 Q How about -- well, let me ask you this one. Yes, 12 any IOCS data collector I assume -- 13 A Okay. 14 Q That -- I'm looking for the IOCS system, okay? 15 A Okay. 16 Q Do you know what activity code is assigned under 17 the IOCS system to not handling mail at an OCR machine? If 18 an employee is at an OCR machine but not handling mail, do 19 you know that activity code that would be assigned there? 20 Is that possible to say? 21 A 5610, which is the shape-specific letters. 22 Q Okay. And what shapes of mail -- maybe you've 23 just answered it, but let me make it clear. What shapes of 24 mail are the basis of LIOCATT's distribution key for mixed 25 mail activity code 5610? Do you know that? 19249 1 A The 5610 is letter-shape; 5620 is flat; 5700 is 2 parcels. 3 Q So for 5610 it would be all letter activity codes 4 would be the basis of the LIOCATT distribution key? 5 A Right, all letter shapes. 6 Q Okay. Do you know if that includes Priority Mail 7 letters? 8 A To the extent that there are letters for Priority 9 in the direct keys, yes. 10 Q Do you know if in fact that happens? 11 A I'd have to look at my numbers, but -- I could 12 check for you. It would take a few minutes. 13 Q Could you please? 14 A Actually, I don't have it by shape here so I would 15 have to -- well, no, I don't. I would have to go back and 16 check. Sorry. 17 Q Okay. Are you able to check with respect to 18 Express Mail letters? 19 A Yes, I can check for you. 20 Q I mean you can't do it now though, in other words? 21 A No. 22 Q Okay -- and I take it you are not in a position to 23 check now for presorted letters? 24 A Well, we know that there are letters. I mean I 25 think all these categories, to the extent there is letter 19250 1 mail in the direct tallies there will be an assignment of 2 the letter 5610. 3 MR. McKEEVER: That's all I have, Mr. Chairman. 4 CHAIRMAN GLEIMAN: Mr. Koetting. 5 MR. KOETTING: Thank you, Mr. Chairman. 6 CROSS EXAMINATION 7 BY MR. KOETTING: 8 Q Good morning, Ms. Cohen. 9 A Good morning. 10 Q I would like to start looking at Footnote 25 on 11 page 13 of your testimony and if you could look at that in 12 your -- 13 A My cryptic footnote? Okay. 14 Q Right, and I don't want to belabor this but the 15 footnote cites to LR-146 and earlier this week you filed a 16 Work Paper 1. 17 A Yes. 18 Q Would you agree that your work paper does not 19 contain any citations to data that are presented in H-146? 20 A The data itself is in H-23. That is where the raw 21 data is and the programs are in 218 which you need to read 22 146 to work with 218. 23 It's kind of not fun for people who don't do SAS, 24 but -- 25 Q I assume we can agree that actually now that 19251 1 PMA-RT-1 Work Paper 1 is available that the footnote would 2 now probably be more accurate and informative if it referred 3 to that work paper, correct? 4 A Sure. 5 Q And would you agree that even before that work 6 paper was available that a reference MPLR-1 might have been 7 a little more helpful than 146? 8 A Well, I mean LR-1 pulls the data from 23, using 9 146. I mean I'm sorry because there's a lot of Library 10 References and it has the same data and that's why on the 11 work paper I explained it could be gotten from either place. 12 Q Let's turn to the substance. I am still on page 13 13 of your testimony, lines 5 through 9. 14 You are discussing Witness Sellick's proposals and 15 presentations. Would it be a fair paraphrase of what you 16 are saying on lines 5 through 9 there that you find Witness 17 Sellick's distribution method significantly overstates the 18 volume variable cost of Priority Mail, periodicals and 19 Standard B relative to what you would consider a correct 20 distribution method? 21 A Yes. 22 Q And are you drawing that conclusion from your 23 analysis presented at Table 1 there on the bottom of the 24 page? 25 A That's one of the reasons I say that. 19252 1 Q Well, let's look at that table for a moment. 2 Does this analysis hold the variabilities constant 3 between the two allied cost distributions that are 4 presented? 5 A Yes. I mean in trying to do this illustration I 6 used Witness Sellick's assumption that all costs were 7 variable, and so I used the tally costs but reweighted them, 8 but didn't do the variability adjustment. 9 Q Okay. The first column with numbers in it is 10 labelled as the Allied Distribution on Allied Cost Pools, 11 correct? 12 A Well, actually the first column is -- oh, I'm 13 sorry. Are we looking at the work paper or the table? 14 Q I am sorry. I am looking at the table, Table 1. 15 A Right, okay. 16 Q Is the methodology, the allied distribution on 17 allied cost pools methodology that is reflected in that 18 first column, is that the same methodology as Witness 19 Sellick's proposed method? 20 A No. This is a simplification. I mean I did not 21 have Witness Sellick's work papers fully loaded. I really 22 couldn't go through and do an exact calculation but I 23 thought it was an important illustration to make, so I made 24 some simplifying assumptions and I used the term 25 "rudimentary" to describe what I had done. 19253 1 I simply said if you looked at allied and only 2 distributed on itself compared to if you distributed it on 3 everything. In fact, he did it within cost pools for most 4 of that. There were a couple of exceptions across pool but 5 he would have done it within allied and he also would have 6 used the items which I didn't play with here. 7 Q I think that's the basic point we're trying to get 8 at, but just to make sure, the second column with numbers in 9 it, the allied distribution across all costs, would you 10 agree that that isn't actually being -- doesn't actually 11 reflect your proposal or Witness Stralberg's either? 12 A Oh, I do propose that allied distribution should 13 be done across all cost pools. Yes, that is exactly what I 14 proposed. 15 Q But is this derived from your actual distribution 16 keys for mixed mail and not handling mail? 17 A No, this is derived doing a comparison for Witness 18 Sellick. I think that the points are -- can be generalized. 19 I think the conclusions would hold no matter if you used 20 mine. 21 Q If you could refer to your testimony at page 16. 22 We're looking at lines 4 through 6, please, specifically 23 where you state no severe distortions result from 24 distributing costs within CAG and basic function because 25 CAGs and basic function are cleaner separations. 19254 1 Individuals do not often move between CAGs or between basic 2 functions during a work shift. 3 I think we can all pretty readily agree that 4 individuals don't often move between CAGs during work 5 shifts, so if we can put that side of it aside for a moment, 6 would you agree that basic function is an IOCS concept? 7 A Well, not really. I mean, I know Witness Degen 8 said that in his rebuttal, but the concept of incoming and 9 outgoing is certainly integrated in MODS as well. I think 10 it's a well-known concept throughout the postal system. So 11 perhaps the measurement, you know, as we define it there in 12 the two pages he talks about, but I don't think it's an 13 alien concept just used for IOCS. 14 Q Given IOCS sampling procedures, is it possible for 15 an employee to be tallied twice during the same work shift? 16 A I think it is possible, but I don't think it 17 happens. 18 Q Okay. So whether it's theoretically possible or 19 not probably isn't a relevant question, but it doesn't 20 happen. 21 A Right. 22 Q And doesn't that imply that there's no IOCS data 23 with which your hypothesis -- your hypothesis being that 24 employees do not often move between basic functions during a 25 work shift -- there's no IOCS data that you could test your 19255 1 hypothesis with. Is that correct? 2 A Well, it's not really as if that's an important 3 idea. I think we were just trying to do a distinguishing. 4 The fact that in the cost pools we know that people move 5 pretty readily, whereas in basic function certainly at a 6 certain point in the shift you may switch, you go from doing 7 outgoing mail to incoming mail, but it's not the same 8 fluidity that I think you might find in MODS, and that's 9 what we were trying to distinguish. So I'm not saying I 10 would test it or have a hypothesis that one could not find 11 someone doing both on the same shift. 12 Q Well, let's explore that a little bit. In support 13 of your statement down in the footnote you cite transcript 14 page 26, 13826, which is page 12 of Witness Stralberg's 15 direct testimony. Do you happen to have that with you? 16 A Yes, I do. 17 Q If you would look, please, at lines 6 through 8 on 18 that page and footnote 5. 19 A Okay. 20 Q Would you agree that Witness Stralberg's argument 21 is based on the postulation that there is a limited 22 concurrency of the basic functions? 23 A I think he's talking about the general trends and 24 the fact you know the cycle of the day, you know when the 25 outgoing mail is, the period of time, then the switch to 19256 1 incoming, you know, you see more transit during the day. I 2 think that concept is what he's going for. 3 Q As you were preparing your testimony or any of 4 your -- you know, I don't want to focus exclusively on your 5 rebuttal testimony. I know you've been working on direct 6 and everything. Did you ever attempt to verify this point 7 using IOCS data? 8 A To look at whether you have different basic 9 functions on the same tour? 10 Q Correct. 11 A I didn't look at it to see. 12 Q Okay. And we sent you some cross-examination 13 exhibits -- 14 A Yes, you did. 15 Q That I believe you received. And let me 16 distribute those now. 17 A Okay. 18 Q Obviously I could well understand if you haven't 19 had any opportunity to try to replicate these exhibits, but 20 I take it you understand the basic source data that we 21 worked with to come up with the exercise that's -- 22 A Yes, you also used the same sources I did. 23 Q So would you accept for purposes of these 24 discussions that these tables are an accurate representation 25 of IOCS tally costs by basic function and time of the 19257 1 reading? 2 A I'll accept that subject to check. 3 Q Now the statement from Witness Stralberg's 4 testimony that you cited to that appears on that transcript 5 page 13826 is -- there obviously is overlap. He says 6 outgoing and incoming operations in postal facilities are 7 mostly done on separate shifts. Would you agree that while 8 his description may be technically accurate depending on how 9 you interpret words like "mostly," that there is in fact a 10 significant amount of concurrency of the basic functions 11 except at the non-MODS offices? 12 A Well, if I look at the chart that's all office 13 types, I would say I see a significant difference between 14 them in terms of the outgoing being largest on tour 3, 15 incoming predominant in tour 1, and a combination of 16 incoming and transit on tour 2, which is exactly what he 17 suggested. 18 Q Nevertheless you would agree that there are 19 also -- is a significant amount of overlap. 20 A There is overlap. 21 Q And would you agree that at BMCs, for example, 22 which is chart B, that there's nearly as much incoming as 23 outgoing processing taking place at BMCs at nearly all times 24 of the day? 25 A Well, I would like to back up one minute and say 19258 1 one other thing also. It is a little hard to draw 2 conclusions from this because this is really the aggregate 3 data, and different facilities could have different cutoffs 4 at times when they switch from one set of processing to 5 another. So it's a little hard to tell when you look at the 6 aggregate. I mean, you could have a cleaner break at 7 particular facilities. But that said, looking at the BMCs, 8 I would say that not surprisingly they have a more 9 consistent mix. 10 MR. KOETTING: Mr. Chairman, at this point I would 11 like to move that these Charts A through D be designated as 12 USPS/MPA-RT-1 Cross-Examination Exhibit Number 1, and I 13 would request that they be transcribed and I believe it 14 would also be appropriate for them to be moved into 15 evidence, since they do contain summations of information 16 that is already on the record in a fairly well accepted 17 format. 18 MR. CREGAN: Mr. Chairman, we would note that the 19 sources are rather cryptic but we, subject to Ms. Cohen's 20 acceptance of these charts subject to check, we have had no 21 opportunity to analyze them obviously, we have no objection. 22 CHAIRMAN GLEIMAN: And I will direct that 23 Cross-Examination Exhibit Number 1 be transcribed into the 24 record and admitted into evidence. 25 [Cross-Examination Exhibit No. 19259 1 USPS/MPA-RT-XE-1 was marked for 2 identification, received into 3 evidence and transcribed into the 4 record.] 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19264 1 BY MR. KOETTING: 2 Q And while we are on this same general topic, I 3 would like to ask is it possible for someone to move to 4 another basic function without moving between cost pools as 5 cost pools are developed and utilized by Witness Degen? 6 A Well, the cost pools are an aggregation of 7 individual three-digit codes, which include some outgoing 8 and some incoming, and so you do have a combination of 9 outgoing and incoming mail in each of the cost pools, and I 10 think one of the nice things about using the basic function 11 information is it allows you to distinguish that. 12 Q For example though, an employee working at a BCS 13 for the duration of the shift could be processing on the 14 outgoing scheme for part of the shift and processing on the 15 incoming scheme for the remainder of the shift, is that -- 16 A I think so. 17 Q -- a not infrequent occurrence? 18 A Could be. That could happen. I don't know the 19 frequency. 20 Q Under the circumstances they would be staying 21 within the same cost pool but they would be changing basic 22 function? 23 A Right, and what I am saying is if you use basic 24 function as part of the distribution you can capture that. 25 Q Well, let's go back to the statement in your 19265 1 testimony on page 16 that we started with. 2 You say that basic functions, CAGs and basic 3 functions are cleaner separations, and I guess if you could 4 clarify cleaner separations than what? 5 A Than the MODS cost pools in terms of the ability 6 of people or the frequency with which people can be moving 7 between MODS cost pools. 8 Q Are you suggesting that there is no ambiguity in 9 basic function when that is recorded by the IOCS data 10 collector? 11 A They have been using it for 25 years. I think 12 it's pretty clean. 13 Q Let's move to a topic that I think was covered 14 briefly by Mr. McKeever but try to avoid rehashing exactly 15 the same points. 16 Is the MODS CAT a stratum employed by you and 17 Witness Stralberg an aggregate of multiple IOCS sampling 18 strata? 19 A The MODS A, the CAG A facilities you're saying, 20 including BMCs I guess, is that what you are asking me 21 about? 22 Q Not including BMCs. 23 A Okay. Well, then what sense do you mean strata? 24 Q Are you aware for example that there are separate 25 strata for the large mail processing plants and the customer 19266 1 service facilities like stations and branches? 2 A Well, the Function 4 are the station and branches 3 which are part of the MODS, yes. 4 Q Right, and the labor at mail processing plants, 5 what is referred to in MODS and the NWRS is Function 1 6 labor, correct? 7 A Right. 8 Q And the stations and branches is Function 4 labor? 9 A Right. 10 Q And is it your understanding that Witness Degen 11 has separate cost pools for Function 1 and Function 4 labor 12 in the MODS office groups? 13 A He separates those into different cost pools, yes. 14 Q Would you agree, on the other hand, that you and 15 Witness Stralberg employ no directly comparable separation 16 because despite the fact -- let me start over again. 17 Would you agree that you and Witness Stralberg 18 employ no directly comparable separation despite the fact 19 that there is a separate IOCS CAG A stratum for the large 20 customer service operations? 21 A Well, you know, when Witness Bradley does his 22 analysis he doesn't have separate variabilities for those 23 either, so I think, yes, I would say that we treat those 24 together but it is consistent with what Witness Bradley did. 25 Q For distribution purposes, does this make your 19267 1 MODS CAG A group a less clean separation than it could be? 2 A Well, no, because people are not moving between 3 them. I think that's what we were talking about, the fact 4 that people could be clocked into one MODS operation and 5 working in another or clocked into a MODS operation but 6 functioning in some other place, but they are not going to 7 be in a MODS facility and at the station and branch. 8 Q But doesn't this mean that you are basically 9 pooling the handling categories at plants and at stations 10 and branches? 11 A I am treating them together when I do the 12 distribution with the IOCS codes. 13 Q So you are not keeping them within the same 14 facility necessarily? 15 A Right. The tallies are distributed. Their 16 tallies would go into the general pool that are distributed. 17 Q Still on page 16, lines 8 through 10 -- 18 A Okay. 19 Q -- you note that there are variations in the 20 percentages of mixed mail and not handling mail tally 21 observations between the largest and the smallest CAGs. 22 By the smallest CAG do you mean CAG H? 23 A Yes. 24 Q Is it your understanding that the MODS, BMC, 25 non-MODS distinction used by Witness Degen separates the 19268 1 analysis of cost for the small CAGs from that of the largest 2 CAGs? 3 A Yes. 4 Q If we could turn now to page 3 of your testimony, 5 lines 27 through 28. 6 And I'm referring to the statement that reads: In 7 spite of the significance and magnitude of not handling 8 costs, the record in this case contains no evidence 9 pertaining to the causality of these costs. 10 Is it your position that investigating the causal 11 relationship between it's cost driver TPH and cost in the 12 cost pool that Dr. Bradley somehow omitted what you call not 13 handling cost from his cost pool? 14 A No, that's not what I said. 15 Don't you want me to elaborate? 16 Q No. 17 A I didn't think so. I have a good answer. 18 Q Well, I'm sure your counsel can handle that on 19 redirect. 20 I think this is my last line, hopefully. On page 21 11, lines 6 through 15, again I think there's a little bit 22 of overlap here through -- with what Mr. McKeever went 23 through, but maybe we can move through it quickly. 24 Basically in that paragraph on page 11, you're emphasizing 25 how the allied operations support direct distributions; 19269 1 correct? 2 A Yes. 3 Q And isn't it true that there is a component of the 4 allied operation workload that relates to mail that does not 5 require piece distribution in that particular facility? 6 A Yes. 7 Q And would that mail specifically tend to be 8 presorted mail? 9 A Yes. The reason that I proposed doing it over all 10 distributions and not just -- over all operations and not 11 just the distribution is recognition that there is in fact 12 workload. Witness Christensen talked about that as well. 13 He said you'd have to include the workload at the allied as 14 well as the distribution. And that I propose to do. 15 Q On line 15 there on page 11 you're stating that 16 that should be distributed across distribution operations. 17 A I mean to include across all operations. I'm 18 sorry. That wasn't clear. 19 MR. KOETTING: That's all we have, Mr. Chairman. 20 Thank you, Ms. Cohen. 21 CHAIRMAN GLEIMAN: Is there followup? 22 Questions from the bench? 23 Ms. Cohen, I have a couple of questions I need to 24 ask you. When I go through these proceedings I kind of move 25 back and forth between looking at trees and trying to see 19270 1 the forest, and I'm at a point where I've moved away from 2 some trees and I need to look at the forest a little bit. 3 Perhaps you can help me understand. 4 Earlier on Mr. McKeever asked you some questions, 5 and your response made reference to Witness Christensen and 6 his endorsement of the immutable link as it appears to be 7 between Witness Bradley and Witness Degen. You endorsed 8 Bradley's attributions flat out and Degen's distributions, 9 but only if they're modified. That is, in my mind you found 10 some fault with Witness Degen, but not with Witness Bradley. 11 Is that correct? 12 THE WITNESS: Yes. 13 CHAIRMAN GLEIMAN: Okay. Now again, having moved 14 back from the -- maybe the buds on the trees to the forest, 15 as I understand Bradley now, after all the discussion we've 16 had in the hearing room, his attributions basically lower 17 the level of the sea for everybody. Is that pretty much the 18 case? 19 THE WITNESS: He shows lower attribution for mail 20 processing costs. Yes. But it differs at different 21 operations. 22 CHAIRMAN GLEIMAN: But everybody's attribution is 23 basically good. 24 THE WITNESS: Well, some are close to 100 percent, 25 but most of them are lower. 19271 1 CHAIRMAN GLEIMAN: When one does this, that is, 2 lowers the level of the sea, those closer to the margin, 3 that is, those with the lowest markups, tend to benefit the 4 most, do they not? 5 THE WITNESS: Well, if you continue the same 6 markup, I mean I guess you have to accept that everything 7 else stayed the same. 8 CHAIRMAN GLEIMAN: If you continued the same rate 9 and you lowered the level of the sea. 10 THE WITNESS: Right. Those people who had less 11 of -- received less of the institutional costs would 12 benefit. 13 CHAIRMAN GLEIMAN: Would have a higher markup, and 14 if someone then argued -- I don't mean to confuse you. 15 THE WITNESS: Okay. 16 CHAIRMAN GLEIMAN: If you lowered the attribution 17 level on mail processing -- 18 THE WITNESS: Right. 19 CHAIRMAN GLEIMAN: Those parties who are closest 20 to the margin, if you kept the same markup would wind up 21 with a lower rate. 22 THE WITNESS: Well -- 23 CHAIRMAN GLEIMAN: No? 24 THE WITNESS: You have to adjust the markup if 25 you're going to recover all the costs. I mean, you would 19272 1 have a higher markup across the board. 2 CHAIRMAN GLEIMAN: But the parties, if you -- 3 let's take publications, for example. If the markup on 4 publications was five percent, and you decided you wanted to 5 continue a five-percent markup on publications, and you 6 adopted Bradley's attribution levels -- 7 THE WITNESS: Then rates would go down. 8 CHAIRMAN GLEIMAN: So people who are closer to the 9 margin would benefit the most under that example. 10 THE WITNESS: Well, if you chose to keep the cost 11 coverage the same and you lowered the attributable cost, 12 everybody would have that same situation. 13 CHAIRMAN GLEIMAN: Let me ask you now a 14 hypothetical. If the current level of attribution was .5 15 and Bradley's econometric model raised the level of the sea 16 to .9, would you still endorse the use of Bradley flat out? 17 THE WITNESS: Well, if Witness Bradley did a 18 credible, good job on the variability and I accepted that he 19 represented what the variability was, I would accept it. 20 CHAIRMAN GLEIMAN: Okay. 21 Thank you. That's what I needed to know. I have 22 no further questions. Do you? 23 FURTHER CROSS EXAMINATION 24 BY MR. KOETTING: 25 Q Just to follow up on the Chairman's question, Ms. 19273 1 Cohen, he was talking about changes in the overall level of 2 variability and then he was talking about the results for 3 specific cost coverages. 4 Somehow implicit in that question must be 5 something about distributions if we are going to talk about 6 specific subclasses, isn't there? 7 A Distribution of the attributable costs or the 8 institutional? 9 Q Correct. 10 [Laughter.] 11 BY MR. KOETTING: 12 Q Distribution of the attributable costs, I'm sorry. 13 A Right. Somehow when you take a new attribution 14 level you will have presumably a new distribution of cost 15 that goes with it. 16 Q The same distribution -- but I mean if you are 17 going to start talking about subclasses my point is you 18 can't just talk about overall changes in variability without 19 either implicitly or explicitly including the distribution 20 methodology in there, correct? 21 A That's correct. 22 MR. KOETTING: Thank you. 23 CHAIRMAN GLEIMAN: By the way, Ms. Cohen, did you 24 do as detailed an examination of Witness Bradley's 25 attributions as you did Witness Degen's distributions? 19274 1 THE WITNESS: No. I mostly worked on Witness 2 Degen. Witness Higgins, who testified on our behalf, picked 3 up the ball for me and did a more credible job or an 4 extensive job on Witness Bradley. 5 CHAIRMAN GLEIMAN: Extensive. 6 THE WITNESS: Extensive, right. I was correct. 7 CHAIRMAN GLEIMAN: Okay. That brings us to 8 redirect, counsel. 9 MR. CREGAN: Can we have a few minutes? 10 CHAIRMAN GLEIMAN: Certainly. Let's take our 11 10-minute break now and then] we'll pick up from there. 12 [Recess.] 13 CHAIRMAN GLEIMAN: Mr. Cregan, do you have any 14 redirect? 15 MR. CREGAN: No redirect, Mr. Chairman. 16 CHAIRMAN GLEIMAN: No redirect. Well, Ms. Cohen, 17 if that is the case then that ends your testimony before us, 18 hopefully for this case. We appreciate your appearance here 19 today and your contributions to the record. 20 I would like to note for the record that Diogenes 21 is probably spinning in his grave. We found an honest 22 person here -- not that everyone else isn't honest, but you 23 are a seeker of the truth, it was clear from your answers to 24 my last question there. 25 THE WITNESS: Thank you. 19275 1 CHAIRMAN GLEIMAN: And if there is nothing 2 further, you are excused. 3 THE WITNESS: Thank you. 4 [Witness excused.] 5 CHAIRMAN GLEIMAN: Our next witness is appearing 6 on behalf of Time Warner. Mr. Stralberg is already under 7 oath in the proceeding and I have no doubt that Diogenes 8 would be happy to have Mr. Stralberg on the stand also -- 9 lest we show prejudice towards one witness or another. 10 Counsel, as soon as you are ready, you can proceed 11 to introduce Mr. Stralberg's testimony. 12 MR. BURZIO: Good morning, Mr. Chairman. I am 13 John Burzio, appearing on behalf of Time Warner. Appearing 14 with me at counsel table is Timothy Keegan, and Time Warner 15 calls Halstein Stralberg. 16 Whereupon, 17 HALSTEIN STRALBERG, 18 a rebuttal witness, was called for examination by counsel 19 for Time Warner, Inc., also on behalf of Alliance of 20 Nonprofit Mailers, American Business Press, Coalition of 21 Religious Press Associations, Dow Jones $ Company, Inc., 22 Magazine Publishers of America, the McGraw-Hill Companies 23 Inc., and National Newspaper association and, having been 24 previously duly sworn, was further examined and testified as 25 follows: 19276 1 DIRECT EXAMINATION 2 BY MR. BURZIO: 3 Q Would you please identify yourself for the record? 4 A My name is Halstein Stralberg. 5 Q Do you have with you at the table, Mr. Stralberg, 6 a document that has been marked for identification as 7 TW-RT-1 and entitled, "Rebuttal Testimony of Halstein 8 Stralberg on behalf of Alliance of Nonprofit Mailers, 9 American Business Press, Coalition of Religious Press 10 Associations, Dow Jones & Company, Inc., Magazine Publishers 11 of America, the McGraw Hill Companies, Inc., National 12 Newspaper Association, and Time Warner, Inc."? 13 A Yes, I do. 14 Q Did you prepare that testimony? 15 A Yes, I did. 16 Q If you were to deliver your testimony orally 17 today, would it be the same as contained in this document? 18 A Yes, it would. 19 MR. BURZIO: Mr. Chairman, I move that TW-RT-1 be 20 received in evidence and transcribed in the record, and let 21 the record show that I am handing two copies to the 22 reporter. 23 CHAIRMAN GLEIMAN: Are there any objections? 24 Hearing none, Mr. Stralberg's testimony and exhibits are 25 received into evidence and I direct that they be transcribed 19277 1 into the record at this point. 2 I apologize for misidentifying the witness. 3 Sometimes we get a little lost in all the papers and I did 4 not say Time Warner, et al. and I appreciate the correction 5 for the record, Mr. Burzio. 6 [Rebuttal Testimony and Exhibits of 7 Halstein Stralberg, TW-RT-1, was 8 received into evidence and 9 transcribed into the record.] 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19295 1 CHAIRMAN GLEIMAN: Two participants requested oral 2 cross examination of Witness Stralberg -- United Parcel 3 Service and the United States Postal Service. 4 Does any other participant wish to cross examine 5 the witness? If not, then Mr. McKeever, when you are ready 6 to begin. 7 MR. McKEEVER: We have no questions, Mr. Chairman. 8 CHAIRMAN GLEIMAN: Mr. Koetting? 9 MR. KOETTING: Thank you, Mr. Chairman. 10 CROSS EXAMINATION 11 BY MR. KOETTING: 12 Q Good morning, Mr. Stralberg. 13 A Good morning. 14 Q I would like to start on page 6 of your testimony, 15 lines 26 and 27. 16 A Okay. 17 Q You state that mail that bypasses piece 18 distribution produces mostly directly IOCS tallies, correct? 19 A Yes. 20 Q What is the basis for that statement? 21 A Generally what is called direct mail in the mail 22 processing facilities, mail that because of its presort can 23 bypass sorting in that particular facility. This generally 24 is mail that comes from the mailers and therefore it 25 contains identical pieces. 19296 1 Q So what you are stating at least as one of the 2 bases for your statement is that it is primarily identical 3 pieces? 4 A Yes, that is one way that you would produce direct 5 tallies. 6 Q Okay. Now for an identical mail item or 7 container, to turn into an identical mail tally doesn't the 8 data collector have to be able to identify the mail as 9 identical? 10 A When you say "identical mail tally" you mean for 11 him to identify it as being identical mail? 12 Q Correct -- yes. 13 A So by definition he has to identify it as 14 identical mail. 15 Q If the item is subject to the top piece rule, is 16 there any reason why a mailer-prepared item would be more 17 likely to receive a direct tally than a Postal Service 18 prepared item? 19 A Yes -- as far as letter in flat trays and bundles 20 are concerned, most of those are subject to the top piece 21 rule and so they would also produce direct tallies. 22 Q But if it is an item that is subject to the top 23 piece rule, is there any reason why a mailer-prepared item 24 would be more likely to receive a direct tally than a Postal 25 Service prepared item? 19297 1 A In the case of the top piece rule items you are 2 not talking about items in containers. You are talking 3 about individual items being handled, so in those, in the 4 case of individual items being handled that are subject to 5 the top piece rule the chances are about the same. 6 Q Let's go back to the question I had earlier about 7 the necessity to be able to identify identical mail as 8 identical in order for it to be tallied as such. 9 If the data collector is trying to determine 10 whether in fact the mail is identical, wouldn't that task be 11 harder, for example, if the item is shrink-wrapped? 12 A shrink-wrapped? Are you talking about the pallet 13 for example or a bundle or -- 14 Q Anything that would be shrink-wrapped -- pallets I 15 suppose would certainly be an example. 16 A Well, yes. If he cannot clearly see through the 17 shrink-wrap, if he sees a bundle of magazines inside the 18 shrink-wrap, he will generally know that it's one magazine. 19 Q If an item is being cross-docked, do you think it 20 is more or less likely that the shrink-wrap would be broken 21 relative to the case where the item is destined for a direct 22 operation in the plant? 23 A First of all, you are talking about an item in a 24 strict IOCS sense now, an individual sack being 25 cross-docked? 19298 1 Q I apologize. Let's say a pallet is being 2 cross-docked. 3 A A pallet is being cross-docked. And what was the 4 question about it? 5 Q Right -- let's assume we are talking about a 6 shrink-wrapped pallet. 7 A Yes. 8 Q And the question is is it more or less likely that 9 the shrink-wrap will be broken if it is being cross-docked 10 relative to a shrink-wrapped pallet where the item is 11 destined for direct operations in the plant? 12 A Generally if it is being cross-docked in most 13 cases that would be -- well, it could be shrink-wrapped 14 sacks also, which the IOCS clerks have no ways of 15 identifying, but if it is a -- so that would be a mixed 16 tally, but if it is let's say a pallet with Time Magazine it 17 would be pretty easy for the data collector to see what it 18 is through the shrink-wrap. 19 Q Since you mentioned Time magazine, let's imagine a 20 pallet with Time magazine and Sports Illustrated on the same 21 pallet. Would that properly be recorded as identical mail 22 tally? 23 A Well, I don't believe that ever occurs. If could 24 happen in the case of copalletization, which is practiced to 25 some extent by some periodicals. As far as Time and Sports 19299 1 Illustrated, first of all, they publish at different times 2 of the week, and they would in any case not appear together. 3 Q Okay. Well, let's leave my more colorful perhaps 4 but artificial hypothetical and move to a more general 5 hypothetical, as you mentioned, just two periodicals that 6 are being copalletized and appear on a shrink-wrapped 7 pallet. You I believe did just suggest that that in fact 8 can occur. 9 A It does occur, although it's a fairly limited 10 extent at this point in time. 11 Q And when that does occur, would that properly 12 generate an identical mail tally? 13 A It could if the data collector thinks it's only 14 one magazine, but if he sees there's two, then it should -- 15 or if he doubts if they're all the same he should generate a 16 mixed pallets tally. 17 Q If a sampled employee's driving a tow motor and 18 pulling one or more rolling containers across the platform 19 at the time the data collector is making the observation, do 20 you think it is more or less likely that the data collector 21 will be able to obtain the identical mail information 22 necessary to record it as an identical mail tally relative 23 to the case where the employee is handling a single rolling 24 container manually? 25 Do you understand my question? 19300 1 A Yes. I am not sure what the rule is in that case, 2 if he's pulling multiple containers. I think there is rules 3 about taking the first container in that case. In other 4 words, when he's handling more than one, looking at only a 5 single container. In any case, there are very few 6 containers with identical mail on them. At least they 7 represent a very small portion of the IOCS samples. 8 Q Would you say that the mail we're talking about 9 here that bypasses the piece distributions would usually be 10 either on a pallet or on a rolling container? 11 A That's most likely except in a -- if you have a 12 facility with a sack sorter machine. It might actually go 13 on a sack sorter and end up at some other part of the dock 14 or a different platform. 15 Q But it would arrive as either a pallet of sacks or 16 a rolling container of sacks probably? 17 A Well, generally in that case it comes out in a 18 sorters area, and they put it on the -- they usually do some 19 distribution of the sacks there. It comes out in what they 20 call a sorters area, which is an area where they do a 21 further distribution to individual trucks. 22 Q But I believe you did agree that it would more 23 frequently be either on pallets or in rolling containers. 24 A Most of the stuff that gets crossed up nowadays is 25 pallets or rolling containers. Yes. 19301 1 Q Let's talk about rolling containers first then. 2 In general, do you know how often observations of rolling 3 containers result in direct tallies? 4 A Well, as I mentioned earlier, it's fairly 5 infrequent. 6 Q Would you agree, however, that there is a fair 7 amount of presorted mail that arrives in rolling containers? 8 A Presorted letters do; yes. 9 Q I'm staying on this general topic, but I would 10 like to refer you to Exhibit 4 of your direct testimony. It 11 appeared at transcript 26 page 13863, and if you don't have 12 it or if it's inconvenient -- 13 A I do have my direct testimony. Give me a minute 14 to find it. Exhibit 4, you said? 15 Q That's correct, and I will distribute copies to 16 anybody else who might not have that in front of them. 17 A And that exhibit has four pages. Which page did 18 you say? 19 Q Table 4-1 on the first page. 20 A Okay. 21 Q Okay? 22 A Okay. I've got it. 23 Q On that table there are a number of columns, and 24 there's one column -- actually there are two columns that 25 have the phrase mixed over them, and underneath that there 19302 1 is a counted column and an uncounted column. And I would 2 like to direct your attention to the what I will refer to as 3 the mixed uncounted column, which would be the third column 4 of numbers on the table. 5 Are you with me? 6 A Yes. 7 Q Would it be correct to say that that mixed 8 uncounted column reflects nonempty item tallies which have 9 mixed mail activity codes? 10 A Yes, they have mixed mail activity codes. 11 Q And let's look at the pallet line on that table. 12 In that mixed, you know, counted column, do you see the 13 number of 4374? 14 A Four million; yes. 15 Q And in the same line, total nonempty pallet costs 16 are 15099 million; correct? 17 A Yes. 18 Q If we were to divide the first number, 4374, by 19 the second number, 15099, I get approximately 29 percent. 20 You can either, if you prefer, accept that subject 21 to check, or I have a calculator here if you'd like to do 22 the division, or -- 23 A You're dividing the uncounted by the counted, by 24 the total. 25 Q Exactly, the mixed uncounted of 4374 by the total 19303 1 nonempty of 15099. 2 A Yes, the mixed pallets that were not counted. 3 Q And so uncounted pallets then, if that number is 4 properly calculated at 29 percent, the uncounted mixed 5 pallets number is about 29 percent of the total nonempty 6 pallet costs. Correct? 7 A That's what it appears; yes. 8 Q Now let's look at the total items line, which is 9 the very last line on the bottom of the page. And I 10 basically would just like to do the same exercise in the 11 total uncounted column. See the number of 66,012? Correct? 12 A Yes. 13 Q And in the total nonempty column we see the number 14 of 1139820; correct? 15 A Yeah, that's when you include the top piece rule 16 items as well. 17 Q Correct. 18 A Okay. 19 Q And if we do the -- obtain the same percentage, we 20 divide the 66 number by the 1139 number and according to my 21 calculation that's about 5.8 or 6 percent. Does that seem 22 to be a -- 23 A Well, I'll accept that. 24 Q Based on your exhibit, doesn't this therefore 25 suggest that pallets are more likely to result in a mixed 19304 1 mail tally in the IOCS relative to all items? 2 A Well, I think this comparison really makes very 3 little sense. As I indicate in my direct testimony, pallets 4 would more probably be considered with containers. They -- 5 a pallet is a container that can contain sacks, trays, or 6 bundles. And for a shrink-wrapped pallet, for example, to 7 compare that with an individual letter tray really doesn't 8 make too much sense. But I'll agree with your arithmetic. 9 Q Isn't the calculation we have done simply an 10 exercise to show the proportion of times that pallets 11 receive mixed mail tallies relative to all items? 12 A Well, like I said, I agree with your arithmetic. 13 Q Say that pallets were treated as containers, would 14 they be more or less likely to receive mixed mail tallies 15 under the current IOCS rules? 16 A They would probably be more likely to receive 17 mixed mail tallies, like other containers who receive almost 18 all mixed mail tallies. 19 If you take a comparison of an APC with a letter 20 tray, although the individual -- the container as a whole 21 may be mixed, if you take an individual letter tray on it, 22 it would probably be a direct tally. 23 Q I would like to go to page 5 of your testimony. 24 Here I am back with your rebuttal testimony. 25 A Okay. 19305 1 Q And I am looking at the first paragraph, in 2 general, if you want to take a look at that. 3 A Okay. 4 Q On lines 5 through 9, you describe a scenario in 5 which the peak of activity in the platform operation in a 6 major mail processing facility would be in the late 7 afternoon and be relatively -- a relative level of 8 inactivity on the platform two or three hours later. Is 9 that a fair characterization of what you say on those lines? 10 A That is what I am saying, yes. I maybe shouldn't 11 have said major because it, in fact, occurs -- is even more 12 clear in smaller mail processing facilities. But that is 13 what I am saying here. 14 Q So you would say that this would hold true at 15 major and small facilities? 16 A Mail processing plants, yes. 17 Q Do you know if there are peaks on the platforms at 18 any other times of the day? 19 A Well, there tends to be peaks on the platform. 20 This, of course, depends on the facility. But there is also 21 generally a peak when they have to distribute the mail to 22 the stations and branches and associate offices later in the 23 morning, during Tour 1, there tends to be more activity. 24 Individual facilities may, of course, have minor peaks at 25 other times. But, really, the one associated with outgoing 19306 1 mail -- with originating mail, in my opinion, is the most 2 dominant. 3 Q You mentioned a peak in the early morning as mail 4 is dispatched to the stations and branches, correct? 5 A Yes. 6 Q What about a peak later in the night, that is 7 relative later, relative to when the collection mail that 8 you are talking about comes in? Might there might not be a 9 peak later in the night that is related to the incoming mail 10 processing windows? 11 A To the incoming mail processing windows. Incoming 12 mail, this all depends on truck schedules, of course. 13 Trucks arrive at facilities from many places and, for the 14 most part, they are spaced out. It depends on driving 15 distances and so on. A star route, for example, may have to 16 stop at several facilities. What is most concentrated is, 17 again, the peaks that I am referring to. 18 Q Did you attempt to use the IOCS tally information 19 that was available to plot out platform tallies by time of 20 day? 21 A No. I would have the same comment to that as on 22 the basic function chart you showed Ms. Cohen, that 23 individual facilities really have different schedules. So 24 if you try to test that kind of information with IOCS, which 25 really may only have a few samples from each facility, it is 19307 1 not clear that the information will be meaningful. It may 2 be very clear in an individual facility. 3 Q But it is your opinion that employees working 4 platform operations are the ones that handle collection mail 5 in most plants? 6 A In the processing plants, they generally -- the 7 platform people are the ones that pull the mail off the 8 trucks, and usually, there has to be a lot of staging at 9 that point, and they then bring it inside to the 10 cancellation area, where other employees will take over. 11 Q So, it's not -- in your understanding, the 12 collection mail is mostly unloaded by employees -- by either 13 the truck drivers or the employees working the cancellation 14 mail prep operation 010. 15 A It all depends on the facility. 16 Generally speaking, we have mail handlers on the 17 platforms, at least in the larger facilities, who will do 18 that. 19 In smaller facilities, they may not necessarily 20 have mail handlers; it may be the same people who will go 21 out and get the mail and also take it inside and start 22 preparing it. 23 Q Are there mail handlers assigned to the 24 cancellation mail prep operation for that purpose? Do you 25 know? 19308 1 A Yes, I believe there may be. 2 MR. KOETTING: That's all we have, Mr. Chairman. 3 Thank you, Mr. Stralberg. 4 CHAIRMAN GLEIMAN: Is there any follow-up? 5 [No response.] 6 CHAIRMAN GLEIMAN: Questions from the bench? 7 [No response.] 8 CHAIRMAN GLEIMAN: Would you like some time with 9 your witness before rebuttal? Five minutes? 10 MR. BURZIO: Yes. 11 [Recess.] 12 CHAIRMAN GLEIMAN: Mr. Burzio? 13 MR. BURZIO: Thank you, Mr. Chairman. 14 REDIRECT EXAMINATION 15 BY MR. BURZIO: 16 Q Mr. Stralberg, in your colloquy with Mr. Koetting 17 about your Exhibit 4 from your direct testimony, do you 18 recall he asked you some questions about the item under the 19 mixed, uncounted table where the number of 4374 appears for 20 pallets, and I believe that, in the course of your answers, 21 you indicated that pallets would be less likely to be 22 treated as a direct tally than other containers. Did you 23 mis-speak? 24 A Well, we were talking about the hypothetical case, 25 if pallets were treated like other containers, which I 19309 1 believe would be more appropriate, and I may have mis-spoken 2 and said they would be less likely to produce direct 3 tallies. Compared to other containers, they would be much 4 more likely to produce direct tallies. 5 MR. BURZIO: That's all we have, Mr. Chairman. 6 CHAIRMAN GLEIMAN: Thank you, Mr. Burzio. 7 Is there any recross? 8 [No response.] 9 CHAIRMAN GLEIMAN: If there's no recross, then I 10 want to thank you, Mr. Stralberg. We appreciate your 11 appearance here today and contributions to our record, and 12 if there's nothing further, you're excused. 13 THE WITNESS: Thank you. 14 [Witness excused.] 15 CHAIRMAN GLEIMAN: Appearing next is Postal 16 Service Witness Degen. Mr. Degen is already under oath in 17 these proceedings, and as soon as everybody gets settled in, 18 Counsel, you can proceed to introduce his rebuttal 19 testimony. 20 Whereupon, 21 CARL G. DEGEN, 22 a rebuttal witness, was called for examination by counsel 23 for the United States Postal Service and, having been 24 previously duly sworn, was examined and testified as 25 follows: 19310 1 DIRECT EXAMINATION 2 BY MR. KOETTING: 3 Q Mr. Degen, could you please identify yourself 4 formally for the record? 5 A Yes. My name is Carl Degen. 6 Q I'm handing you a copy of a document entitled 7 "Rebuttal Testimony of Carl G. Degen on Behalf of the United 8 States Postal Service," which has been designated as 9 USPS-RT-6. Are you familiar with that document? 10 A Yes, I am. 11 Q Was it prepared by you or under your supervision? 12 A Yes, it was. 13 Q Does the copy that I have handed you exclude the 14 materials on pages 31, 32, and 33 that were stricken by the 15 presiding officer's ruling earlier this week? 16 A Yes. As far as I understood the order, I think 17 that's all been excluded. 18 Q If you were to testify orally today, would this be 19 your testimony? 20 A Yes, it would. 21 MR. KOETTING: Mr. Chairman, I'm going to hand the 22 two copies of USPS-RT-6, Rebuttal Testimony of Carl G. Degen 23 on Behalf of the United States Postal Service, to the 24 reporter and request that they be accepted into evidence in 25 this proceeding. 19311 1 CHAIRMAN GLEIMAN: Are there any objections? 2 [No response.] 3 CHAIRMAN GLEIMAN: Hearing none, Mr. Degen's 4 testimony and exhibits are received into evidence, and I 5 direct that they be transcribed into the record at this 6 point. 7 [Rebuttal Testimony and Exhibits of 8 Carl G. Degen, USPS-RT-6, was 9 received into evidence and 10 transcribed into the record.] 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19367 1 CHAIRMAN GLEIMAN: Three participants requested 2 oral cross examination of the witness -- the Advertising 3 Mail Marketing Association, the Alliance of Non-Profit 4 Mailers, and the American Business Press. 5 Does any other party wish to cross examine the 6 witness? 7 MR. KEEGAN: Yes, Mr. Chairman. Timothy Keegan 8 representing Time-Warner, Inc. We did not file a notice of 9 intent to cross examine, but we would like to reserve the 10 right for follow-up cross examination. 11 CHAIRMAN GLEIMAN: Certainly. That's not a 12 problem. 13 There is no other party, as best I can tell. 14 So, in that case, Mr. Wiggins, if you're ready, 15 you can proceed. 16 MR. WIGGINS: Thank you, Mr. Chairman. 17 CROSS EXAMINATION 18 BY MR. WIGGINS: 19 Q Mr. Degen, Frank Wiggins for the Advertising Mail 20 Marketing Association. 21 You discuss in the section of your testimony -- 22 your capacious testimony -- marked as Roman numeral XXIII, 23 beginning in page 44, the testimony of Gary Andrew. Do you 24 have that handy? 25 A Yes, I do. 19368 1 Q And over on page 45, beginning on line 3, you say 2 Witness Andrew argues that one also needs to distribute 3 non-MODS costs by sub-groups. 4 You go on to say, down at the end of that 5 paragraph, Witness Andrew's criticism may have theoretical 6 validity, but in this instance, the empirical results show 7 that it is not a problem, and you then, in the course of 8 that, refer to Table 7, which shows up on page 46 of your 9 testimony. 10 A Yes. 11 Q I take it that it's your belief that Table 7 12 demonstrates why it is that Dr. Andrew's theoretically valid 13 objection to the distribution of mail processing costs from 14 non-MODS offices is not a problem. Is that right? 15 A Yes. 16 Q I'm having a little difficulty making the 17 connection, and maybe you can help me out. Witness Andrew 18 testified that, if you redistributed the mail processing 19 costs of the non-MODS offices, the thing that you said is 20 theoretically valid, it would result in a decrease in the 21 measured cost of parcels of 2.3 cents. Was that your 22 understanding of his testimony? 23 A I'm afraid I don't have the ability to recall it 24 at this point. I'd be happy to look at it. 25 Q Do you have his testimony handy? If not, I do. 19369 1 If this will be quicker, I'll just hand you Tables 4 and 5 2 of his testimony, which demonstrate -- 3 A That would be fine. 4 Q -- his development of that. 5 MR. WIGGINS: If I may approach the witness, Mr. 6 Chairman. 7 CHAIRMAN GLEIMAN: You certainly may, sir. 8 BY MR. WIGGINS: 9 Q Have you had a chance to look at Tables 4 and 5 of 10 Witness Andrew's testimony, which I believe demonstrates his 11 redistribution of the non-MODs office costs and, at Table 5, 12 demonstrates a restated parcel cost which is 2.33 cents less 13 than what resulted from the Postal Service analysis 14 A Yeah, I mean I haven't replicated his analysis or 15 gone into it in depth, but that appears to be what he is 16 doing here. 17 Q Do you have a problem with his calculation? In 18 your analysis, is the 2.33 cents of decrease in parcel costs 19 an accurate application of his theoretically valid objection 20 to what the Postal Service did? 21 A No, I am really not comfortable with drawing that 22 parallel. My impression from reading his testimony, and not 23 specifically these tables, was that he was theorizing that 24 the distribution of costs, if the distribution of costs was 25 done by subgroups within the non-MODs offices, we would get 19370 1 a very different distribution of mail. And my statement in 2 my testimony is that might be true, but I tried it and it 3 didn't change. 4 I really didn't mean to offer a specific opinion 5 to any of his tables here, and I am not comfortable with you 6 saying that what I said might have been theoretically 7 correct has anything to do with these tables. When I was 8 talking about his theory, I was talking specifically about 9 my interpretation of what he was saying, which is that the 10 distribution of costs within subgroup for non-MOD offices 11 would be different than it is in the aggregate, and I showed 12 that it wasn't, and that's -- that's as far as I went. I 13 can't endorse the theory behind these tables, I really 14 haven't studied that. 15 Q You can't endorse it, but you can't specifically 16 dispute it, is that correct? 17 A No, and I don't think I have attempted to in my 18 testimony. 19 Q What Dr. Andrew did here was a redistribution of 20 the non-MODs office cost by shape, correct? Distinguishing 21 between parcels and flats. 22 MR. KOETTING: I think this is getting beyond the 23 scope of Witness Degen's rebuttal testimony. He 24 specifically cites one page of Dr. Andrew's testimony, which 25 is in fact an Interrogatory response. And, again, as he 19371 1 just stated, his analysis was limited to that particular 2 statement as illustrated in that Interrogatory response by 3 Dr. Andrew, and he responded to that point and he didn't 4 address the substance of Dr. Andrew's direct testimony at 5 all. 6 MR. WIGGINS: Well, the testimony to which the 7 citation -- there are two citations, I believe. The first 8 of them on lines 16 through 18 of page 44 of Witness Degen's 9 testimony says, "The impact of interaction between 10 individual cost pool variabilities and distribution key can 11 distort the differences between shapes." And the witness 12 goes on to say, "The empirical results show that it is not a 13 problem." 14 I am citing him to the portions of Dr. Andrew's 15 testimony that show the results of Dr. Andrew's shape 16 redistribution, and I am just trying -- I don't see in Table 17 7 a shape redistribution or a contest with Dr. Andrew's 18 results. I am just trying to bridge that gap, Mr. Chairman. 19 CHAIRMAN GLEIMAN: I think you can continue with 20 your line of questioning. 21 BY MR. WIGGINS: 22 Q What Dr. Andrew does in the two tables that I have 23 put in front of you is to demonstrate the redistribution of 24 non-MODs mail processing costs by shape, is it not, Mr. 25 Degen? 19372 1 A Well, I think I have already said, and I just want 2 to say this very clearly, -- 3 Q Sure. 4 A -- I am not familiar with what he did with respect 5 to redistribution by shape at all. 6 Q Okay. And you don't offer any redistribution by 7 shape? 8 A Well, inasmuch as he alleged that the shape 9 effects were related to the failure or to not do sub-cost 10 pool distribution -- distribution of costs within subgroups 11 of non-MODs, I went through the exercise of doing that 12 distribution by subgroup and non-MODs and, from the point of 13 view of the overall subclass distribution, I don't see a 14 chance. To the extent his theory regarding shape relies on 15 the fact that that subclass distribution is different, that 16 is the point I have addressed. 17 I am not addressing his specific alternative shape 18 redistribution in my testimony and, to the extent it is 19 independent of the fact that I have shown that the subclass 20 redistribution doesn't change, then it would stand on its 21 own. To the extent it relies on that fact, then I guess I 22 am disagreeing with it. 23 Q Understood. Thank you. 24 MR. WIGGINS: Mr. Chairman, I have nothing 25 further. 19373 1 CHAIRMAN GLEIMAN: The Alliance of Nonprofit 2 Mailers. 3 [No response.] 4 CHAIRMAN GLEIMAN: That moves us to the American 5 Business Press. 6 CROSS EXAMINATION 7 BY MR. STRAUS: 8 Q Good morning, Mr. Degen, I am David Straus for 9 American Business Press. How many non-handling codes are 10 there? 11 A Could you be more specific? You mean how many 12 different activity codes can be assigned to a person who is 13 not observed handing a piece of mail or an item? 14 Q Yes. 15 A I don't know off the top of my head. There would 16 be several classes of such codes. 17 Q Are you familiar with the major codes? 18 A Perhaps you could direct me to what you consider 19 to be the major codes and I will tell you if I am familiar 20 with them or not. 21 Q How about 5610, 5620, 5670 and 5040? 22 A I need a little refreshing on what 5040 is. 23 Q But you know what 5610 and 5620 are? 24 A I believe those are letter and flat shape, 25 respectively, in the case of someone who is observed not 19374 1 handling mail but in a shape dominant operation. 2 Q And 5700? 3 A I understand that to be all shapes, all classes, 4 the most general mixed mail activity code. 5 Q Would you have any way to check what you 6 recollection of 5700 is, because I believe it is parcel 7 shape? 8 A Yes, I think you are right, I'm sorry. 9 Q And the mixed would be 5750? 10 A Yeah, I think I had those backgrounds. I'm sorry. 11 Q And 5620, did we do that one? 12 A Yeah. Wasn't that flats? 13 Q Yes. 14 A Yeah. 15 Q Well. Okay. So -- 16 A And there is one in there you asked me about that 17 we haven't repeated, 5740. 18 Q 5040. Well, let's just move on. 19 A Okay. All right. 20 Q If an employee is seen near a flat case, but is 21 not handling, will that result in a 5620 tally? 22 A I am not comfortable with the characterization of 23 seen near a flat case. I mean is he -- 24 Q If the IOCS observer observes him standing next to 25 a flat case but he is not handling mail, what kind of a 19375 1 tally would produce? 2 A Okay. Can I assume that the data collector has 3 determined that he is working there? Or is this somebody 4 walking by? 5 Q He's just there. 6 A I am not comfortable with that kind of question. 7 I mean -- 8 Q Well, why not? I mean is the data collector any 9 more comfortable with it? 10 A I believe so. I think the data collector has the 11 wherewithal to reasonably ascertain whether this person is 12 working there. For instance, if they are standing in front 13 of the case, it would make a difference than if they are 14 standing behind it. I mean I think data collectors do have 15 the sense to determine whether or not somebody appears to be 16 working in the flat operation. 17 Q If the data observer saw that he had just 18 delivered mail to that flat opening unit, and now is doing 19 nothing, having just delivered that mail, then how would 20 that observation be recorded? 21 A Is it the instant he is let go and he is about 22 ready to turn around and go back, or -- 23 Q Yes. 24 A I mean the data collector understands that this 25 mail has been delivered to the operation and that he had 19376 1 been functioning to deliver mail to that operation? 2 Q Yes. 3 I am trying to find the tally in that case. 4 A Yeah. I don't believe it would be 5620. Was this 5 a flat case? 6 Q Yes. 7 A Yeah. I don't believe it would be 5620. I don't 8 think he would be associated with the flat operation at that 9 point. 10 Q Even though the data collector knew he just 11 delivered flats to the flat case? 12 A Right. I mean if he knew his purpose there was to 13 bring them from the platform, I don't believe they would 14 associate that under the 5620 rule. 15 Q What would it be? 16 A I think the 5750. Well, no, I'm sorry, I think it 17 would be straight not handling. 18 Q With no shape reference at all? 19 A I don't think so. If he is on his way back from 20 having delivered mail, I don't think they would force that 21 shape connection. 22 Q What -- you said straight not handling. What 23 activity code do you have in mind? 24 A 5750, no shape connection. 25 Q Are any employees logged into opening units 19377 1 recorded as 5620 entries? 2 A I'm not sure. I would have to check that, and 3 even if they were that would not necessarily counter your 4 example inasmuch as the data collector may have mistakenly 5 determined that the employee was working in that operation 6 rather than just delivering mail to it? 7 Q Maybe you should explain to me why an employee who 8 the data collector knows has just been handling flats, 9 brought them to a flat case, and is about to return would 10 not be recorded with a flat-shaped indicia of 5620? 11 A My understanding of the 5620 activity code is that 12 it specifically handles cases where a person is working in a 13 shape-dominant operation but not specifically handling mail 14 at the time of sampling, but if the data collector doesn't 15 determine that this person is working in that operation, but 16 rather sees him in transit from that operation I don't know 17 that that is what was intended by that rule. 18 Q What functions are typically performed by opening 19 unit employees? 20 A Opening units generally have responsibility for 21 processing containers and items that are entering or leaving 22 the facility. In terms of an incoming opening unit, their 23 function would be to look at sacks and trays and determine 24 whether that needs to be sorted for direct outgoing dispatch 25 and determine which of those containers or items need to be 19378 1 directed to piece sortation operations or bundle handling 2 operations within the plant. 3 Q Would an opening unit employee be traying letter 4 mail and putting the trays on an APC? 5 A Certainly when letter mail was bundled, you saw 6 some of that. Now that, you know, bundles are essentially 7 gone, that is a relatively infrequent activity. 8 They would be handling whole trays of letter mail. 9 They certainly wouldn't be traying up any collection mail or 10 loose letter mail. I mean that is not the function of the 11 opening unit. 12 Q But they could be loading trays onto an APC? 13 A Yes. They could be loading whole trays onto an 14 APC. 15 Q And then pushing that APC to another station? 16 A Possibly yes. 17 Q And while pushing that container, while the 18 employee is pushing that container would that be a mixed 19 mail cost? 20 A If the -- well, it would fall into that area if 21 the contents of the container are counted. It would be a 22 counted item cost that would be distributed based on the 23 proportion, the item proportions inside of it. 24 Q But he is pushing an entire container. 25 A Yes, he is. 19379 1 Q And typically that would be -- you said if they 2 are counted, but typically that would be a mixed mail cost, 3 wouldn't it? 4 A Yes, mixed mail in the broadest sense, but it is 5 not the same as mixed -- we don't have any idea what is 6 going on here. It is mixed in the sense of we have a 7 container, we know what kind of items are inside of it, and 8 we'll distribute those items in proportion to the kind of 9 mail we find in those items. 10 Q So if you have two employees in an opening unit, 11 one has letters in trays, one has flats in trays, and they 12 both -- the letter employee puts the letter trays in an APC 13 and the flat guy puts the flat trays in an APC and they both 14 start pushing their APCs off to the next station, would you 15 distinguish those costs by container type in your approach? 16 A I would to the extent the data collector counts 17 the container in each case. We would distribute the 18 container containing the letter trays according to the kinds 19 of mail we find in letter trays, and we would distribute the 20 costs associated with the flat tubs, if you will, according 21 to the kind of mail we find in flat tubs. 22 Q So the answer is at times yes, you would 23 distinguish, so that the letter cost would be imposed upon 24 letter mail and the flat cost on flat mail? 25 A Well, I think more than "at times" -- to the 19380 1 extent the contents of those containers are counted it would 2 always happen. 3 Q Okay. Now let's assume that these two employees 4 reach their station as we discussed before and they let go 5 of the cart and they are standing there talking to the 6 person at the next station or just getting ready to move 7 back to the loading dock, or maybe they are waiting for an 8 empty cart to bring back to the loading dock. 9 Is this wasted time in your opinion or is it 10 associated with a function? 11 A Well, you slipped in a new wrinkle there in that 12 he is standing there chatting with the guy at the case. 13 Q Forget the chatting part. 14 A Okay. No, I don't view that as wasted time. I 15 view that as productive time. 16 Q And how again would that time be recorded? 17 A I believe 5750. 18 Q Do you know the extent of the 5610 costs in 19 opening units? 20 A No, I don't. Not off the top of my head. 21 Q What about in pref. opening units? 22 A No, I don't have those data in front of me. 23 Q What would account for the 5610 costs in a pref. 24 opening unit? What kinds of activities? 25 A I believe it would be the data collector's 19381 1 understanding that at the point such an employee was 2 observed that they were associated with a letter sortation 3 operation. 4 Q At an opening unit. 5 A Well, or if they're delivering mail and the data 6 collector misinterprets them as actually working in that 7 unit and gives them a 5610 code, I believe that's where that 8 would come from. 9 Q So if there are say $30 million of 5610 costs at 10 pref. opening units, 5610 not handling costs -- we're still 11 on not handling. 12 A Right. 13 Q Again, what kinds of activities would that be? 14 A If at the time that tally was taken the data 15 collector associated that employee with one of the 16 operations for which they're instructed to classify not 17 handling as 5610. 18 Q Right. But you're telling me what the observer 19 does. And what kinds of observations by -- of the employees 20 would lead the collector to record a 5610? 21 A I think that's what I just answered. 22 Q No, I think you gave me a general description, and 23 I'm asking for a specific activity. You said if he thought 24 it was associated with a letter-opening activity. I'm 25 saying what specifically -- what kinds of activity. 19382 1 A If it looked like the employee was working there. 2 Q Like what? Doing what? 3 A Sorting mail, moving mail from the container onto 4 ledges, for instance. In your example we're leaving the 5 container -- 6 Q Aren't those handling costs rather than not 7 handling costs? I'm looking for the 5610 not handling costs 8 at the pref. opening unit. 9 A Well, they're not all handling costs. If you're 10 unloading a container when you have the piece you're taking 11 out of the container and putting it to the ledge, you're 12 handling. If you're going back to container, you're not 13 handling. There's a lot of opportunity to observe someone 14 not handling, even though they're in an operation that we 15 would all generally associate with handling mail. 16 Q So you're saying then that if somebody is moving 17 back from the ledge to get more letter-shaped mail it would 18 be a 5610. But if someone -- 19 A No, I'm not saying that. I'm saying if the data 20 collector believes that employee is working in that 21 operation, they correctly associate the fact that that 22 person isn't handling mail with the 5610 activity code, 23 because they believe them to be associated with a 24 piece-handling operation. 25 Q But if that same employee is seen delivering that 19383 1 tray or having just delivered that mail to an opening 2 unit -- from an opening unit -- that would not be a 5610. 3 A No, if it's clear you're on the way back, I don't 4 believe so. 5 Q But if you are on the way back from the ledge to 6 the container it would be? 7 A Because in some cases the unloading of containers 8 would be performed by employees working in that operation, 9 and in other instances, time permitting, the opening unit 10 person may be helping out. 11 Q Okay. So, the employee who has just delivered 12 letters from the opening unit, his time would not be 13 recorded as a letter cost. 14 A Could you be more specific? 15 Q For example, if the man -- the employee -- excuse 16 me -- who delivered the cart, the APC, dropped it off and is 17 starting his return trip. 18 A And a portion of his time will be recorded as 19 letter cost to the distribution of mixed mail within the 20 opening unit operation. 21 Q But unrelated to his activity at the time, though. 22 It wouldn't matter what he was doing at that time, it still 23 would have been distributed the same way. 24 A No. We use the fact that he's clocked in to the 25 opening unit to distribute his time into proportion of the 19384 1 kinds of mail that are handled in the opening unit. So, his 2 activity is very much taken into account. 3 Q His activity in delivering letters to a letter 4 sorter is not taken into account, is it? 5 A Inasmuch as -- I would say that, to the extent 6 that that activity is a certain proportion of general 7 opening unit activities, it would be precisely taken into 8 account. 9 Q Yes, but you only take into account on the basis 10 in proportion to the direct tallies, don't you? 11 A Yes, specific to that opening unit. 12 Q So, the ratio of direct tallies would tell you how 13 you would distribute all of the costs of bringing mail from 14 the opening unit to processing stations. 15 A No. It would only tell you how to distribute the 16 costs associated with the return trip when the employee is 17 not observed with an item or container in his hand. 18 Q What about the trip going if it's a mixed mail 19 cost? 20 A Well, if it's a mixed mail cost, then I'll assume 21 for purposes of this discussion that it's a counted 22 container that he's using. Then we use the fact that that 23 -- are we talking a flat case here or a letter case -- 24 letter case. 25 Q Letter case. 19385 1 A If he's observed pushing a pie cart, if you will, 2 full of trays and we know that it's full of trays, those 3 costs are all basically going to go to the kind of mail 4 that's in trays, which is letters. 5 Q Please look at page 1 of your testimony. 6 A Yes, I have it. 7 Q On line 10 and 11, you use the phrase "economic 8 marginal cost." Could you define that, please? 9 A I understand marginal cost to be the cost 10 associated with producing an additional unit of volume. 11 Q I asked about the phrase "economic marginal cost." 12 A Perhaps "economic" is a little redundant in that 13 phrase. I'm not sure people outside of economics talk about 14 marginal cost. 15 Q You aren't using -- you didn't mean economic 16 marginal cost as opposed to some other kind of marginal 17 cost. 18 A No, I did not. 19 Q So that sentence means that you must distribute 20 processing costs the way you've proposed here in order to 21 determine the marginal cost? 22 A There's a couple elements to my distribution 23 methodology perhaps that maybe have not been adequately 24 distinguished to this point. 25 Q Is this in response to my question? 19386 1 A Yes, it is. 2 Q Okay. 3 One element of my distribution methodology is what 4 to do with mixed mail, and how you estimate, you know, what 5 the underlying sub-class detail of mixed mail is certainly 6 not dictated by the desire to get marginal cost. 7 When I'm referring to my distribution methodology, 8 I'm specifically referring to the fact that Dr. Bradley has 9 estimated for me the portion of each cost pool that is 10 volume variable, and I need to keep those costs associated 11 with the mail that's being handled in that particular cost 12 pool, and in that sense, my distribution methodology needs 13 to be done the way I do it independent of my mixed mail 14 distribution. 15 And that is the only way to determine marginal 16 costs? 17 A The sentence I have in lines 9 and 10 says, "I 18 will explain why the distribution of mail processing costs 19 must be done the way the Postal Service has done it in order 20 to provide economic marginal cost." 21 What I am really saying there is given the cost 22 pool framework and Dr. Bradley's variability estimates, my 23 distribution methodology is the piece that completes that 24 puzzle. 25 I do not mean to say that there is not some 19387 1 completely different approach that might get to you to 2 marginal cost. 3 Q Please turn to page 3 of your testimony, the 4 sentence beginning on line 5. You say, "Under the new 5 method not handling tallies are effectively ignored in most 6 cost pools." 7 What happens to those tallies? 8 A The tallies themselves are not used. The costs 9 associated with those tallies -- well, all the costs 10 associated with the cost pool are part of the pool of costs 11 to which I apply Dr. Bradley's variability. 12 Dr. Bradley's variability analysis for the pool as 13 a whole segregates nonvolume variable costs from volume 14 variable costs. Some of those nonvolume variable costs may 15 be not handling. Some of the nonvolume variable costs may 16 be direct. But he has done the causality test to determine 17 that some portion of that pool's costs are not volume 18 related and in terms of what happens to those, we do not 19 distribute them to classes of mail but my understanding is 20 they are collected in the form of markup on the final 21 attributed costs. 22 Q But you do distribute the nonhandling costs? 23 A No, that is not an accurate statement. 24 I distribute the portion of costs that Dr. Bradley 25 has associated with volume variable -- the portion of costs 19388 1 that Dr. Bradley has identified as volume variable. 2 Some of those may be not handling. Some of those 3 may be direct. 4 Q When you say they are effectively ignored, you 5 mean they are not part of the distribution key? 6 A No. I mean more than that. 7 Under LIOCATT the formation of cost pools uses the 8 not handling tallies to determine the size of the cost pool. 9 My cost pool costs are independently determined 10 through use of NWRS and MODS data so I don't need to rely on 11 the not handling costs to measure that portion of my cost 12 pool. 13 Q On that same page you discuss the growth in the 14 not handling and mixed mail costs and say that it is in part 15 due to a change in the IOCS Question 20 instruction in 1992. 16 A Yes. 17 Q Do you see that? And that instruction was that 18 collectors should not ask an employee to pick up mail? 19 A That's correct. 20 Q How much did that change affect the mixed mail 21 costs? 22 A I don't know off the top of my head. 23 Q You wouldn't expect it to be large, would you? 24 A I'd have to think about it to offer an opinion 25 either way. 19389 1 Q Well, it is true, isn't it, that prior to that 2 change instruction there was no contrary instruction? They 3 were not asking people to pick up a piece of mail. They 4 just more or less in 1992 clarified that they should not do 5 it. 6 MR. KOETTING: Could I get a clarification on what 7 I think is the question that is still under discussion here? 8 You asked about the effect of the change on not 9 handling costs or on mixed mail costs? 10 MR. STRAUS: I don't remember, but the witness 11 heard the question and gave me an answer so I don't -- 12 MR. KOETTING: I would like for the record to be 13 clear as to what the question was so that the witness can 14 make sure that he heard the question correctly. 15 MR. STRAUS: I am sure the record is clear. My 16 memory is not. 17 MR. KOETTING: Well, then I would ask that it be 18 read back -- if counsel doesn't remember what the question 19 was. 20 CHAIRMAN GLEIMAN: Can you read the question back? 21 [The reporter read the record as requested.] 22 MR. KOETTING: I think the question really was 23 phrased in terms of a mixed mail cost when the previous 24 conversation had been in terms of a not handling cost. 25 CHAIRMAN GLEIMAN: Well, I am glad that you now 19390 1 understand what the question was. You can move on now, Mr. 2 Straus. 3 BY MR. STRAUS: 4 Q Do you know how much of a change in not handling 5 costs resulted from the instruction in 1992? 6 A If you mean can I specifically quantify the change 7 in not handling costs associated specifically with that 8 change in instruction, the answer is no, I cannot. 9 Q Was that instruction then reversed in 1996? Your 10 footnote at page three. 11 A I would not characterize it as a reversal. I 12 believe additional instructions were given so that employees 13 could associate a piece of mail with a sampled employee in a 14 non-subjective manner. 15 Q Even with the 1992 change, didn't the not handling 16 costs increase from less than 20 percent in 1986 to about 28 17 or 29 percent in 1991, which is before the change in 18 instruction? I'm looking at your graph on page four. 19 A Yes, that's approximately correct. I don't have 20 the actual specific numbers associated with each bar, but 21 that's approximately correct. 22 Q That's about a 40 percent increase in not handling 23 costs? 24 A Well, to go from, what did you say, 20 to about 25 30, is about ten points. 19391 1 Q From 20 to about 28. 2 A So it'd be about eight points. 3 Q About 40 to 50 percent increase in not handling 4 costs prior to the change in instruction? I'm not saying 5 percentage point increase. I'm saying a percentage 6 increase. I understand it's -- 7 A You are looking at the percentage change in the 8 percentage? 9 Q Yes. If it goes from 20 percent to 30 percent, 10 that's a 50 percent increase in the percentage; isn't that 11 right? 12 A Yes, that's the percentage change in the 13 percentage; yes. 14 Q So we could then say that between 1986 and 1991, 15 as a percentage, the not handling costs increased in excess 16 of 40 percent? 17 A Well, the percentage went up about eight points 18 and the percentage increase in the percentages, 19 approximately 40 percent. 20 Q From 1986 to 1996, they about doubled; right? 21 A The percentage of not handling approximately 22 doubled; yes. 23 Q On page eight of your testimony, the sentence 24 beginning on line six. 25 A I have it. 19392 1 Q My high school English teacher would charge you 2 for dangling that participle there, and the reason is you 3 can't understand what it means. You say "instead of 4 rejecting the new method," who are you talking about there, 5 the Postal Rate Commission, the Postal Service? Who would 6 be doing the rejecting in that sentence? 7 A Anyone. I basically mean to say no one should 8 reject the new method because it makes different assumptions 9 in LIOCATT, that there should be different criteria used to 10 evaluate it. 11 Q Are you suggesting that the Rate Commission or the 12 Postal Service should test the assumption underlying your 13 method? 14 A No, I think I'm much more along the lines of Mr. 15 Stralberg when he invokes the application of common sense 16 and our operational knowledge to the evaluation of the 17 assumptions. 18 Q So you think that a comparison of assumptions is 19 just as good as a testing of assumptions? 20 A I think I've tried to make it clear in my 21 testimony that these aren't really the kinds of assumptions 22 you are testing. There doesn't seem to be a dispute in this 23 record that the sub-class distribution across cost pools 24 differs, and I find that's sufficient information to say 25 that costs should be distributed -- mixed mail costs should 19393 1 be distributed within the cost pool. 2 I've seen testimony from several Intervenors 3 making it very clear that when you find a tray in the 4 collection operation, it's probably going to have first 5 class single piece in it, and to me, that tells me that if I 6 take the mixed mail costs associated with collection mail 7 and distribute them on a broader key, that is not as good an 8 assumption as saying a tray in collection that we don't know 9 the contents of is more likely going to have the contents of 10 the other trays we do know the contents of within the 11 collection operation. 12 Q Again, I mean that's a nice answer and it's long, 13 but you are relying still on comparing the common sense of 14 assumptions rather than testing those assumptions, and you 15 say Mr. Stralberg did not test assumptions. He's not in a 16 position to conduct those tests, is he? 17 A I'm not sure really any of us are in a position to 18 conduct those tests, and your earlier characterization that 19 I rely only on common sense is not true. I thought my 20 answer made it clear that I'm also relying on what I believe 21 is the consensus in the record that the sub-class 22 distributions do differ by cost pool. I mean that's an 23 important point. If there were a dispute about that, we 24 might want to test it, but my understanding is there is 25 agreement about that, and given that, I don't think it's a 19394 1 big leap of faith to say that cost distribution should be 2 done within the cost pool. 3 Q You do testify on page 9 that the proportions and 4 composition of direct mixed and not handling tallies vary 5 across cost pools. Isn't that right? 6 A Yes. 7 Q Does the extent of clocking into one pool and 8 working in another pool also vary across cost pools? 9 A To the extent that I think that's a negligible 10 amount, I think it's a negligible amount in all cost pools, 11 and to that extent I wouldn't say it varies -- the variation 12 in it would matter. 13 Q Your conclusion that it's a negligible amount, is 14 that common sense or is that an assumption or is that 15 tested? 16 A It's tested to the extent that Dr. Bradley's 17 estimates are robust in terms of his ability to explain the 18 hours associated with a particular cost pool by the cost 19 drivers he uses for that pool. 20 Q Do you agree with Postal Service Witness Steele 21 that a good manager shifts employees around to maximize 22 productivity? 23 A Yes, I do. 24 Q And do you know whether those shifted employees 25 reclock every time they're shifted? 19395 1 A My understanding is they don't reclock every time 2 they're shifted, but to the extent we're talking about cost 3 pools, my understanding is that they do reclock when they're 4 shifted, so that you may move from one three-digit MODS 5 operation to another within a general group of operations. 6 Maybe, you know, you've switched from incoming to outgoing 7 within a manual distribution operation. 8 My understanding is that that reclocking is not 9 always done, but those employees are generally under the 10 supervision of a single supervisor. I think supervision is 11 closely aligned with the cost pool definitions we've 12 created, so that I would find it unusual for people to move 13 among supervisory domains without reclocking. 14 Q What are brown sacks supposed to be used for? 15 A In general they're used for periodicals. 16 Q And you understand that not everything in brown 17 sacks is periodicals? 18 A Yes, I do. 19 Q Do you understand that some periodicals are not in 20 brown sacks? 21 A Yes, I do. 22 Q Have you ever discussed sack shortages with 23 printers or Postal Service field personnel? 24 A I must have, because I'm certainly aware of the 25 problem. 19396 1 Q And to the extent there are sack shortages, you 2 might find stuff in the wrong-color sacks. 3 A That could be a possible reason; yes. 4 Q You say, beginning on line 8 of page 11, something 5 I think you expounded upon before in one of your answers, 6 which is that you say there's no evidence that mixed mail 7 costs would not have the same subclass distribution as 8 direct costs in a pool defined by operation group and item 9 or container type. Is there any evidence that they would? 10 A Yes. 11 Q And what is that evidence? 12 A Well, I'll go back to my earlier example about 13 collection mail, that when we observe a tray in a collection 14 operation and for whatever reason we don't invoke a top 15 piece rule or we associate a non-handling tally with it, the 16 evidence is clear that presort mail should not be passing 17 through cancellation. I mean, it may happen, but most of 18 the time, it doesn't, and so there is strong operational 19 evidence that the containers observed in that cost pool have 20 a very focused, a very specific distribution, if you will, 21 related to the kinds of mail that are handled in that cost 22 pool, just the same way that I would not expect, you know, 23 the time spent in letter sorting operations to be related to 24 flats or vice versa. 25 I mean, it may happen, but there's clearly a 19397 1 dominant subclass distribution in cost pools that differs 2 from other cost pools. 3 Q But the extent of that domination differs among 4 cost pools, and you have given some examples in which your 5 conclusion, you believe, is strongly reinforced. There are 6 others where the relationship is less dominant; isn't that 7 the case? Such as opening units. 8 A No, I -- I think I have given examples that are 9 very clean, you know, for illustrative purposes. That 10 doesn't mean to suggest that I don't believe that the 11 underlying subclass distribution in any cost pool isn't 12 strong enough to warrant distributing costs within it. 13 I mean, I gave an example where there would be a 14 pattern of a single underlying subclass, but if you had a 15 pattern of three or four, that would not weaken my assertion 16 at all. 17 Q Your assertion is based I think you said on your 18 belief. 19 A No. The operational realities of the work room 20 floor was my basis. 21 Q That could be tested, though, couldn't it, by 22 counting the mixed mail samples at the same cost pools where 23 the direct tallies are observed? 24 A I think we went through this when I testified on 25 direct, that, you know, once you count them, they're not 19398 1 mixed anymore. I mean, my understanding is, for whatever 2 reason the mixed aren't countable, to test that, you somehow 3 have to override those or predict, absent the presence of 4 your testing people, which ones, in fact, would have been 5 mixed. I think you have to make a lot of assumptions to try 6 and create such a test. 7 Q Well, you say uncountable. That's uncountable in 8 the terms of interfering with the ongoing mail processing, 9 but a special test to determine the validity of your thesis, 10 anything is countable, isn't it? 11 A I think that's technically true, but I have never 12 been aware of the Postal Service's willingness to tolerate 13 holding up the mail. I mean, my understanding is they've 14 never compromised that. 15 Q As the old joke goes, it happens all by itself? 16 A It's an old joke. 17 Q Getting back to common sense and assumptions, on 18 page 12 of your testimony, on line 11, you list the first of 19 what you call a straightforward assumption, which is that 20 the subclass distribution of uncounted items is the same as 21 the subclass distribution of counted items within the cost 22 pool. Are you recommending that the Commission reverse the 23 determination it made in R94-1? 24 A If it's contrary to that, I am. 25 Q And then after stating your second assumption, you 19399 1 say that common sense tells you that the assumptions are 2 true or more nearly true than the alternative. Are mail 3 processing costs 100 percent variable? 4 A I do not believe they are. 5 Q Didn't common sense tell us for about 25 years 6 that mail processing costs were 100 percent variable? 7 A I think that may have been more of convenience 8 than common sense. 9 Q And Dr. Bradley tested that conclusion and found 10 it to be wrong, didn't he? 11 A Dr. Bradley tested the volume variability of costs 12 within cost pool. The previous assumption of 100 percent 13 variability applied generally to mail processing costs but 14 did not apply specifically to the cost pools as I have 15 defined them. 16 Q Well, wait a minute. If it's 100 percent across 17 all cost pools, how can it be something other than 100 18 percent within each individual cost pool? 19 A I can explain that. 20 Q Okay. 21 A And the reason is because under LIOCATT, not 22 handling mail costs were not distributed within cost pools. 23 In essence, not handling mail costs were treated as variable 24 with respect to all mail processing, but not variable with 25 respect to a specific cost pool, and that's an important 19400 1 difference. 2 Q Nevertheless, Dr. Bradley's conclusion about less 3 than 100 percent variability would require, if adopted, the 4 Commission to cast aside what everyone assumed to be the 5 case for 25 years. 6 A Well, my recollection is that previous dockets are 7 filled with people questioning that assumption, so your 8 characterization that everyone assumed it to be true I think 9 is wrong. 10 Q Well, the Postal Service assumed it to be true. 11 A I am not even sure that's true. I mean they used 12 it because it was there. I mean it was the best assumption 13 they could make. 14 Q Even though they didn't -- and they didn't think 15 it was a valid assumption but they used it anyway? 16 A They didn't have any evidence to the contrary. 17 Q Did they try to develop evidence to the contrary? 18 A I am not in a position to address that other than 19 the work with which I am familiar that supports their 20 proposal in this docket. 21 Q Is Dr. Bradley's work dependent upon some new 22 invention or theory that was developed only in the past year 23 or two, or could that study have been done 10 or 15 years 24 ago? 25 A It would have been very difficult to do it 10 or 19401 1 15 years ago. 2 Q It was difficult to do it now, wasn't it? 3 A I don't think so, I think it was very 4 straightforward. 5 Q I would like to direct your attention to page 16 6 of your testimony. 7 A I have it. 8 Q You say that employees and the Postal Service have 9 strong incentives to have the clock ring data correct. Do 10 you see that? Line 9. 11 A Yes, I do. 12 Q What do you mean by employees there? 13 A Well, I mean that if employees want to get paid, 14 they need to be clocked into an operation. 15 Q But they don't need to be clocked into the right 16 operation, do they? 17 A Not in terms of getting paid. 18 Q So their real incentive is to clock in as quickly 19 as possible, not to clock in as accurately as possible, 20 isn't that right? 21 A I don't believe that is true. I think they are 22 penalized for clocking in too early or too late from a 23 designated start time. 24 Q Are they penalized for clocking into the wrong 25 operation? 19402 1 A Not in terms of their pay, but I would expect that 2 if they were found clocked into the wrong operation, it 3 would bring some kind of supervisory reprimand. 4 Q So if there's -- if the start time is exactly 8:00 5 o'clock a.m. and there's one place to clock in that's two 6 feet away, another that is 200 yards away, in neither case 7 would that employee be clocking in too early, -- never mind, 8 I'll withdraw that. 9 You call, on line 19, you call MODS an accounting 10 system. Could you tell me what you mean by that? 11 A Yes. When I refer to it as an accounting system, 12 I am really referring to the underlying time and attendance 13 system from which MODS hours are derived. 14 Q So you are saying the data are from an accounting 15 system? 16 A The hours data, yes. 17 Q Was the MODS system designed to relate costs of 18 operations to subclasses? 19 A Not directly. 20 Q I would like to direct you to page 21 of your 21 testimony, specifically, the statement beginning at line 13. 22 You saw that both Witnesses Stralberg and Cohen recommend 23 that all not handling costs be treated as institutional. Is 24 that a correct statement? Is your testimony correct? 25 A Given the way you are asking it, I am sure there 19403 1 is some detail I have overlooked, so -- 2 Q Well, let me ask you about your reference. What 3 did you read from which you concluded that? 4 A I don't have a specific reference and you will 5 probably make me wish I had put the word "essentially" in 6 there somewhere. 7 Q No, I wish you had put "not all" instead of "all". 8 A Yeah. Okay. 9 Q I am just wondering, you know, you reach this 10 conclusion that they recommend all not handling costs. Do 11 you have their testimony with you, that you are rebutting? 12 A Yeah, I think I have both of theirs here. 13 Q Could you make take a quick look at, say, Ms. 14 Cohen's testimony and see what you might have had in mind 15 with that statement? 16 A Well, I can explain what I had in mind with that 17 statement. 18 Q I know what you had in mind by explanation. I 19 want to know what in her testimony you relied on for your 20 conclusion. 21 A I relied on the recommendation that not handling 22 costs be removed from the cost pools. And I don't think we 23 need specific citations for that, I think that is pretty 24 clear in both of their testimonies, that we not distribute 25 not handling costs within a cost pool. 19404 1 Q Let me read you from Ms. Cohen's testimony, page 2 36, beginning at line 15. 3 A This is her rebuttal testimony or the direct 4 testimony? 5 Q No, the testimony that you rebutted from which you 6 drew this conclusion. 7 A What page were you on, sir? 8 Q Page 36. 9 A Okay. 10 Q She says, "I recommend that the Commission 11 similarly use its statutory discretion in this case to 12 refrain from attributing to classes and subclasses of mail 13 the portion of volume variable mixed mail and not handling 14 costs that is due to inefficient operations. Do you 15 translate the portion that is due to inefficient operations 16 to mean all? 17 A No, I do not. 18 Q Do you know what portion she was talking about 19 there? 20 A Well, I think there has been considerable 21 discussion in this docket that we -- that she -- I think, 22 and I can't quote the exact spot, but at one point she says 23 I don't know what the number for not handling is. So I 24 don't think I can say proportion she was referring to, and I 25 believe she has admitted in her testimony that she doesn't 19405 1 know what the number is. 2 Q Do you know what she actually used? 3 A No. But that does not change the conclusion of 4 the sentence to which you originally directed me. 5 Q Does Witness Stralberg distribute the same costs 6 as you do? 7 A Could you be more specific? 8 Q The same segment 3 costs as you do, the same 9 amount of costs. 10 A In total? 11 Q Yes. 12 A Yes. 13 Q So does he -- he distributes them, but he treats 14 them as institutional, is that what you are saying? 15 A Perhaps my phraseology there as institutional was 16 misleading. But the point I was trying to make there is the 17 point that I made earlier, and that is, under the LIOCATT 18 method, even though it is nominally an assumption of 100 19 percent variability, that is 100 percent variability with 20 respect to mail processing as a whole, and while he does 21 distribute all those costs to subclass, he does not do it 22 within cost pool. And it would be wrong to do it after we 23 have applied Dr. Bradley's variabilities which serve the 24 function of telling us what the correct number is in terms 25 of which costs should stay in the cost pool and which should 19406 1 not. 2 Q Let me direct you to Ms. Cohen's testimony on page 3 38. 4 A I have it. 5 Q Have you read that before? 6 A The page? 7 Q Yes. 8 A Yes, I have. 9 Q If you look at the top half of that page, does 10 that refresh your recollection on the amount or the 11 percentage of costs that Ms. Cohen recommends be treated as 12 institutional? 13 A Yes. 14 Q And that is about 20 percent? 15 A Yes. But I don't agree with it. 16 Q No, but it is certainly not -- 20 percent isn't 17 the same all. 18 A That is correct. 19 CHAIRMAN GLEIMAN: Mr. Straus, do you have a sense 20 of how much longer you are going to go? 21 MR. STRAUS: Half an hour. 22 CHAIRMAN GLEIMAN: I think we will take a 10 23 minute break now then. It is my intention to finish up with 24 this witness, and then when we finish up with this witness, 25 we will take a lunch break. So parties who are interested 19407 1 in witnesses who come later in the day can plan accordingly. 2 My guess is we are looking at lunch around 1:00 o'clock, 3 coming back around 2:00 and picking up with Witness Sellick 4 at that time, give or take a little bit. Either 20 percent 5 or all. 6 [Recess.] 7 CHAIRMAN GLEIMAN: Mr. Straus, are you ready to 8 pick back up? 9 MR. STRAUS: Yes. 10 BY MR. STRAUS: 11 Q Mr. Degen, we are now on page 22 of your 12 testimony. 13 A I have it. 14 Q On line 18 you use the phrase, to describe your 15 firsthand experience, "hundreds of work floor situations". 16 I am not exactly sure what you mean, so could you explain 17 what hundreds of work floor situations means? 18 A They would be the situations you observe in 19 spending hundreds of hours on work room floors looking at 20 what people are doing. 21 Q You are not contending, are you, that you have 22 more firsthand observation experience than Mr. Stralberg? 23 A I am not familiar with his, but I would give you 24 even money on it. 25 Q My question remains, are you contending that you 19408 1 do? 2 A No. 3 Q Are you contending that you have more experience 4 than Ms. Cohen? 5 A No. 6 Q Are you, in essence, a full-time Postal 7 employee/consultant? 8 A Not quite. 9 Q Over the past five years, how much of your time do 10 you suppose you have devoted to Postal matters? 11 A Do we just want to do five? 12 Q So far. 13 A Ninety percent. 14 Q When you say that nearly all non-handling costs 15 are associated with productive activities, are you including 16 or excluding break time? 17 A Excluding break time, given that it is a 18 contractual requirement. It's just the cost of doing 19 business. 20 Q How do you explain the increase in break time? 21 A I don't know, I haven't studied that. 22 Q Do your observations tell you anything? 23 A I don't have any theories developed to the point 24 that I would care to share them now. 25 Q Please turn now to page 26 of your testimony. 19409 1 There you have Table 4 which shows increases in the 2 percentage of not handling time for both allied and, 3 separately, non-allied operations. What is your most 4 important, or most significant explanation for the increase 5 in allied not handling time? 6 A I believe that the -- well, in general, the 7 increase in not handling time reflects a changing technology 8 of processing the mail, an increased move to 9 containerization, increased use of automation, floor 10 layouts, et cetera. I believe it is a manifestation of 11 changes in the underlying technology for both allied and 12 non-allied. 13 Q So you don't distinguish between the two, you 14 don't have any different explanations for allied growth or 15 non-allied growth? 16 A If I spent some time, I could probably come up 17 with, you know, with has affected each differently, but not 18 off the top of my head. 19 Q I'd like to turn to a different subject now, which 20 is the disagreement over the importance of and the results 21 of comparing the costs of handling periodicals with the 22 Postal Service's wage costs. 23 One of your conclusions at page 27, line 5, based 24 upon your analysis is that by 1996 periodicals' costs were 25 slightly higher in real terms than they had been in 1989, do 19410 1 you see that? 2 A That is not really a conclusion of mine. That is 3 a conclusion reached from the graphs that are being 4 discussed in the section above it. I don't -- I believe I 5 go on to explain why I think that is not quite true because 6 of the illegitimacy of the comparison. 7 Q By saying it was a conclusion of yours, I was 8 simply trying to explain that. I don't necessarily agree. 9 A Okay. 10 Q Have their been changes between 1989 and 1996 in 11 the amount of work-sharing done by periodicals mailers? 12 A Yes, I believe there have. 13 Q Isn't it true that there's been a significant 14 increase in the amount of barcoding? 15 A Yes, I believe there has. 16 Q In fact, isn't it true that between fiscal year 17 1993 and fiscal year 1996 the amount of Level A barcoding 18 has more than doubled and the amount of Level B barcoding 19 has nearly doubled? 20 A I am a little concerned that I have given an 21 impression here that I am saying that this barcoding 22 represents work-sharing -- I mean in the order you have 23 asked these questions. I wouldn't want that implication to 24 be made. 25 Q All right. Forgetting about the characterization, 19411 1 isn't it true that barcoding has about doubled between -- 2 just between fiscal year '93 and fiscal year '96? 3 A I am not comfortable saying that. I don't really 4 have those numbers at my fingertips. 5 Q Well, will you accept it subject to check? 6 A I will agree that it has increased. 7 Q Well, will you accept that it has nearly doubled 8 based upon the Postal Service's own mail characteristics 9 data? 10 A If I accept it, then it is my responsibility to go 11 check it? 12 Q Yes, it is. 13 A No. 14 Q To the extent that it has increased and you have 15 admitted it is increased, that would reduce mail processing 16 costs by the Postal Service, would it not? 17 A To some extent, yes. 18 Q How big is the barcode discount for Level A? 19 A I don't know off the top of my head. 20 Q Well, can you accept subject to check -- 21 A Can I correct my earlier answer? 22 We were talking specifically about barcoding and 23 its ability to reduce costs. 24 Q Yes. 25 A I think that is an open question -- you know? I 19412 1 know that there are machines out there deployed with barcode 2 readers but there are certainly contentions that at this 3 point there isn't full enough deployment to warrant 4 segregating barcoded mail and so I can't agree that that has 5 necessarily resulted in decreasing costs. 6 Q You are saying that the Postal Service does not 7 segregate prebarcoded periodicals from non-prebarcoded 8 periodicals? 9 A Not in every instance, and there is also the issue 10 of whether or not the barcode is on a piece that is from a 11 practical standpoint machinable. 12 My understanding is there are some issues out 13 there like that and I have not studied it so I am not 14 comfortable agreeing with you. 15 Q I'll accept that you haven't studied it. 16 Doesn't the Postal Service offer a discount in the 17 neighborhood of 3 cents apiece for prebarcoded periodicals? 18 A That sounds reasonable. 19 Q And don't you think they do that because they 20 think they save about 3 cents apiece in processing costs? 21 A I am sure that was the plan when the discount was 22 offered. 23 Q What is the average processing cost for a 24 periodical? 25 A I don't know off the top of my head. 19413 1 Q Give me an estimate. Is it 15 cents or a buck and 2 a half or how much do you think it is? 3 A Closer to 15 cents than a buck and a half. 4 Q Do you know what the average postage per 5 periodical is? 6 A Closer to 15 cents than a buck and a half. 7 Q Do you think it's maybe in the low 20s? 8 A Seems reasonable. 9 Q And so a three cent discount is a pretty hefty 10 percentage of the total postage and therefore an even bigger 11 percentage of the processing costs, isn't that right? 12 A That is not how I usually use the phrase "pretty 13 hefty." 14 Q Has there been an increase in the amount of 15 presortation, the depth of presortation between 1989 and 16 1996? 17 A Are you referring to presort level of bundles or 18 containers? 19 Q Pieces for periodicals. 20 A I believe there has been an increased depth of 21 sort. 22 Q Would you accept that the amount of Level C 23 presort has increased 50 percent in that period? 24 A I'll accept that it has increased. 25 Q But you won't -- I guess you won't accept subject 19414 1 to check because you don't want to check whether it is 50 2 percent? 3 A That's right. 4 Q Well, Mr. Degen, you have got testimony here which 5 tries to explain the trends between mail processing costs 6 and wage costs and you point to -- and we will get to this 7 in a minute -- palletizing practices, which you apparently 8 have gone into in some depth, and you even hypothesized 9 about truckers wanting to fill up trucks. 10 I am curious as to why you -- why neither you nor 11 anyone else has investigated these other cost driving 12 factors -- the amount of presort, the amount of barcoding, 13 to get a complete picture of where costs are going and 14 should have gone compared to wage rates. 15 Can you explain that to me? 16 A The focus of my analysis has been more recent than 17 what you have given me, and most of the examples you are 18 giving me are over a longer time period than I looked at. 19 We have looked at some of the billing determinants 20 in recent years and they have not changed to the extent you 21 are indicating they have changed since 1989, so my analysis 22 was focused on a more narrow period than most of the numbers 23 you have put out here today. 24 Q The barcoding numbers I gave you began in 1993, 25 didn't they? 19415 1 A I don't remember. 2 Q Well, they did. 3 A Okay. 4 Q When does your pallet analysis begin? 5 A In '93. 6 Q So you looked at the pallet changes since 1993, 7 what you believe them to be, but you didn't look at the 8 barcode changes since 1993? 9 A No, I accepted that the barcoding had increased 10 quite a bit, but I expressed some doubt as to whether those 11 costs had been fully realized over that period. 12 I thought I addressed that pretty squarely. 13 Q Does Postal Service management share your view 14 that there is significant doubt that the barcode discounts 15 make any sense? 16 A I didn't characterize them as not making any 17 sense. 18 Q Well, if there's no savings they make no sense, 19 right? 20 A The timing of realizing savings can be a difficult 21 thing. You have to generate mail flows of that type so that 22 you can deploy automation and operating procedures that 23 capture them, and all I am saying is that I am not sure that 24 the instant barcoded mail appeared that it achieved those 25 savings. 19416 1 Q There were a billion barcoded pieces, prebarcoded 2 periodicals, in 1993. Isn't that enough to start realizing 3 some cost savings? 4 A You have to know a lot more about it than that. 5 You have to know about the practical machinability 6 of the pieces and the way they are deployed. 7 To the extent some of those pieces are in delivery 8 units it wouldn't even matter. 9 Q Let me go to the pallets then for a minute. 10 Your hypothesis, and we will get to it again, is 11 that mail has shifted from five-digit pallets to three-digit 12 pallets and that has increased costs. 13 A Yes. 14 Q And am I correct that the increase in costs is 15 limited to one additional bundle handling? 16 A No. 17 Q Well, have you calculated what the additional cost 18 is of a three-digit pallet on a per piece -- that is, a per 19 magazine, basis? 20 A No, I have not. 21 Q Do you know whether it is less than or more than 3 22 cents apiece? 23 A No, I do not. 24 Q So you haven't quantified this pallet shift at 25 all? 19417 1 A No. 2 Q But you have hypothesized that this is a 3 significant contributing factor to the handling cost of 4 periodicals? 5 A Well, I think "hypothesized" doesn't do it 6 justice. 7 I have met with a number of field operating 8 personnel who when asked about why periodicals' costs appear 9 to be rising talk about the increased handling associated 10 with more aggregate pallets and the resulting piece 11 handlings from breakage of those bundles due to additional 12 handlings. 13 This analysis is not presented here to be a 14 complete explanation of why periodicals' costs have risen, 15 and I don't hold myself up as an expert in that. This was 16 done for rebuttal in a very limited timeframe, and what I am 17 saying is there is a makeup trend that does seem to coincide 18 closely with both the decline in periodicals' unit costs and 19 their subsequent rise in the 1990s. 20 Q Let's turn now to your analysis of the 21 relationship between periodical processing costs and wage 22 rates. 23 Now what you have done, as I understand it, is you 24 have used in Figures 2 and 3, you have compared regular rate 25 periodicals' mail processing costs with all clerk and mail 19418 1 handler wages and you have indexed them in Figure 2 to 1986 2 and in Figure 3 to 1989, is that right? 3 A That's correct. 4 Q Have you had a chance to examine the various 5 similar charts provided to you yesterday as a potential 6 cross examination exhibit? 7 A Yes, I have. 8 Q Do you have that with you? 9 A Yes, I do. 10 MR. STRAUS: Mr. Chairman, I'd like to distribute 11 copies of that. 12 To save time, I'm going to be distributing two 13 charts, one with the weighted and one with un-weighted clerk 14 and mail handler costs, and we'll do them separately. 15 CHAIRMAN GLEIMAN: Mr. Straus, you'd better 16 distribute faster. This is counting against your half-hour. 17 BY MR. STRAUS: 18 Q Mr. Degen, please look at the chart marked 19 "Comparison of Regular Periodicals Mail Processing Cost and 20 Clerk and Mail Handler Wages." Have you had a chance to 21 verify the accuracy of those graphs? 22 A I didn't know I was supposed to. 23 Q Does that mean you've had a chance but didn't do 24 it or didn't have a chance and didn't do it? 25 A My attorney gave them to me yesterday, and I 19419 1 looked them over, and they're reasonable, but I'm not 2 prepared to attest to their accuracy. 3 Q Well, when you compare starting points of '86 4 through '93, as these graphs do, with your two graphs which 5 are indexed at two of those years, '86 and '89, do you see 6 any differences? 7 A No. I think all the plots on this chart are 8 consistent with my verbal description of, year by year, 9 what's happened to periodical costs relative to inflation. 10 Q What's your explanation for the very rapid 11 increase shown on the bottom two charts, the one beginning 12 in '92 and the one beginning in '93? 13 A I can't say with certainty that I know why those 14 costs have changed, but I observed that, in general, there 15 has been a move toward less aggregate pallets during that 16 period. 17 In fact, I think if you look at the third plot 18 down in the left-hand column, that shows you a very good 19 picture of costs declining in '92, when I understand there 20 was a significant movement toward pallets but, you know, a 21 relatively large percentage of five-digit pallets, and my 22 understanding from discussions with field personnel and just 23 comparing mail characteristics data is that, since 1992, 24 there's been a movement away from those five-digit pallets 25 in favor of three-digit or SCF pallets. 19420 1 So, I offer that as an explanation. I can't say 2 for certainty that that's what's happened, but it's a more 3 plausible explanation than costs are just out of control. 4 MR. STRAUS: Mr. Chairman, I'd like to ask that 5 this document be copied into the -- identified as ABP -- 6 these have been done several different ways in this case. 7 The simplest way would be ABP-XE-1, but some parties have 8 identified them with the party. 9 CHAIRMAN GLEIMAN: Your marking is fine. Do you 10 want to mark both of these that way and have both of them 11 in? 12 MR. STRAUS: No, let's just do the first one 13 first. 14 CHAIRMAN GLEIMAN: I think the reporter can mark 15 the copies, if he's sure he knows which of the two you're 16 making reference to. 17 [Cross-Examination Exhibit No. 18 ABP-XE-1 was marked for 19 identification.] 20 MR. STRAUS: And I'd ask that it be admitted into 21 evidence. 22 MR. KOETTING: No objection. 23 CHAIRMAN GLEIMAN: Thank you, Mr. Koetting. I 24 appreciate that. 25 BY MR. STRAUS: 19421 1 Q Now, Mr. Degen, turning to your Figure 4 -- 2 CHAIRMAN GLEIMAN: Excuse me, Mr. Straus. I'll 3 direct that the cross examination exhibit designated as 4 ABP-XE-1 be transcribed into the record and entered as 5 evidence. 6 [Cross-Examination Exhibit No. 7 ABP-XE-1 was received into evidence 8 and transcribed into the record.] 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19423 1 BY MR. STRAUS: 2 Q Turning to your Figure 4, Mr. Degen, I'm confused, 3 because I guess I don't understand the terminology. You 4 describe it at page 27. You say that index is based to the 5 overall clerk and mail handler wage index in 1993. But the 6 index shown on Figure 4 is 1989. Is one of those numbers 7 wrong, or am I just not understanding the line in the 8 testimony? 9 A Could you direct me to the line on page 27 you 10 want me to look at? 11 Q It begins at the very bottom on page 20 and runs 12 over to the top of page 21 -- excuse me -- line 1 on page 13 28. I'm reading the testimony that it's a '93 index, but 14 I'm reading the chart that it's an '89 index, and I may just 15 not understand what you're saying in the testimony. 16 A Okay. What I mean to say there is that we only 17 did the LDC-based index from '93 forward. 18 So, the '93 index was based to the -- the LDC 19 index was scaled so that its '93 value was equal to the 20 value corresponding to an overall clerk and mail handler 21 index that's based to one in 1989. 22 If you will, we pasted on those last four years, 23 using the same number for 1993 and then growth rates based 24 on the LDC index going forward. 25 Q It's safe to say, isn't it, that in 1996, for 19424 1 example, you show that the two points are virtually on top 2 of each other, both for mail processing costs and for wages. 3 A Yes. 4 Q And have you had a chance to look at the cross 5 examination exhibit provided to you yesterday that was 6 titled "Comparison of Regular Periodicals Mail Processing 7 Costs and LDC-Weighted Clerk and Mail Handler Wages"? 8 A Yes, I have. 9 Q And are those charts, those graphs, the equivalent 10 of yours, but rather than a 1989 index, using various 11 indices other than 1989, in addition to 1989. 12 A That, along with changing scales to make the 13 differences more pronounced as you use shorter and shorter 14 time periods. 15 Q Do those -- by using a different index here from 16 1989, do those graphs typically show a different result? 17 A Well, it's not really a different result. What 18 you achieve in the bottom two graphs, for instance, is you 19 take out the big decline in periodicals' costs between 1991 20 and 1992, and that's really what's going on here, that cost 21 -- let's look at the left-hand column, the third one down I 22 think is a good illustration. 23 In 1991 and 1992, periodicals' unit costs actually 24 declined relative to inflation, and then since 1992, they 25 have risen relative to inflation. That basic story doesn't 19425 1 change no matter how you plot it, and that issue is 2 explicitly stated in my testimony. 3 Q What about from 1986 to 1989, how do those costs 4 do during that time period? 5 A Let me go back to my -- 6 Q Can't we just look at the upper left-hand chart? 7 A I'd prefer mine. They're a little more readable. 8 The other one was a fax. Between 1986 and 1989, mail 9 processing unit costs rose relative to inflation. Between 10 1989 and 1990, they were pretty constant, relative to 11 inflation; between 1990 and 1991, they declined slightly;by 12 1992, they declined considerably and then they rose relative 13 to inflation in the remaining years. 14 Q What was the percentage increase between 1986 and 15 1989? 16 A Percentage increase in what? 17 Q Mail processing unit costs. 18 A Mail processing unit costs? About 30 percent. 19 Q And in wages, LDC weighted wages? 20 A About 20 percent, maybe a little less. 21 Q You can't explain that difference with the pallet 22 changes you believe occurred in the 1990's; right? 23 A Well, unfortunately, what I found is even trying 24 to look back to 1993 is an extremely difficult proposition 25 and I didn't even attempt to go back to 1989, just because 19426 1 data on mail preparation are very hard to obtain. 2 MR. STRAUS: Mr. Gleiman, I'd like to have marked 3 as ABP-XE-2, this chart we were just discussing with various 4 alternative ways to graph the comparison of periodicals' 5 mail processing costs against LDC weighted clerk and mail 6 handler wages, and ask that it be copied into the record and 7 admitted into evidence. 8 [Cross-Examination Exhibit No. 9 ABP-XE-2 was marked for 10 identification.] 11 CHAIRMAN GLEIMAN: It is so directed. 12 [Cross-Examination Exhibit No. 13 ABP-XE-2 was received into evidence 14 and transcribed into the record.] 15 16 17 18 19 20 21 22 23 24 25 19428 1 BY MR. STRAUS: 2 Q Finally getting to pallets. What do you mean by 3 an aggregate pallet? 4 A By a more aggregate pallet. I specifically mean, 5 and I think I indicate somewhere in my testimony that by 6 more aggregate, I mean, for example, a three digit pallet 7 rather than a five digit pallet, something with a lower 8 level of presortation. 9 Q The theory that more aggregate pallets is a 10 contributor to the cost increases for periodicals, was that 11 a theory that you developed or that somebody at the Postal 12 Service suggested to you? 13 A I mean you could say I developed it but it was 14 based on discussions. Actually, it wasn't an issue so much 15 that I was raising as just feedback that I was getting from 16 people in the fields complaining about getting mail on more 17 aggregate pallets and how it was driving their costs up. 18 You know, subsequent to my suggesting that as a possible 19 explanation, we had more structured discussions with field 20 personnel. It seemed to be generally confirmed by people 21 who were even more familiar with operations than I am. 22 Q Which drop shippers did you -- with which drop 23 shippers did you discuss their wanting to improve the cube 24 utilization of trailers? 25 A Most of that is probably secondhand from DMU 19429 1 personnel. DMU is detached mail units. So, it would have 2 been obtained secondhand from talking to postal acceptance 3 clerks associated with printing plants. Some of that would 4 have been my own discussions. Some of that would have been 5 discussions of my associates with such people. 6 Q These are discussions with Postal people, not with 7 the truckers or the printers or publishers? 8 A I don't think I've personally had any discussions 9 with the printers themselves. Some of my associates may 10 have. 11 Q You blamed some of the -- "blamed" isn't the right 12 word. You attribute some of the increase in periodical 13 processing costs to what you claim is a trend of mail moving 14 from five digit pallets to three digits; is that right? 15 A That's correct. 16 Q And that leads to more bundle handlings as a 17 result of having to open those pallets at a three digit 18 unit? 19 A That's correct. 20 Q And the Postal Service performed three studies 21 that you have examined on the -- there was an 1989 study and 22 a 1993 study of pallet make up? 23 A I think I looked at 1993 and 1996. 24 Q Was there one for 1989? 25 A My understanding is that there was, and we 19430 1 obtained data from that docket but could not in the time 2 available feel comfortable using it, so we did not use it. 3 Q Now your claim is that the 1993 study shows 43 4 percent five-digit pallets; is that right? 5 A No, my claim was that our corrected numbers from 6 1993 show that. When we looked at the 1993 data, we 7 observed that there was an apparent inconsistency with 8 billing determinants from 1993 because the mix of sack mail 9 and pallet mail in that study seemed to be out of line to 10 us, and actually our understanding of how those data were 11 collected confirmed that it did not do a good job of 12 estimating the relative proportions of sack versus pallet 13 mail. 14 So in order to deem those -- in order to use those 15 data, we felt compelled to reweight the sack pallet volumes 16 such that they would accurately reflect the billing 17 determinants we had available, and the 43-percent number is 18 the result of that process. 19 Q Have you reviewed the potential cross-examination 20 exhibit provided to your counsel yesterday marked USPS 21 Periodical Pallet Volumes 1993? 22 A Yes, I have. 23 MR. STRAUS: Mr. Chairman, I'd like to distribute 24 copies to the relevant parties. 25 CHAIRMAN GLEIMAN: Certainly, Mr. Straus. 19431 1 BY MR. STRAUS: 2 Q Did you examine this? 3 A Yes, I did. 4 Q After examining it, have you concluded that the 5 43-percent total is correct? Remains correct? 6 A Yes. 7 Q Do you -- have you determined whether the data on 8 this chart are -- accurately portray the data that were on 9 the sources shown? 10 A No, I haven't really had time to do that. I mean, 11 we got this when we were out here, and our office is in 12 Madison, so we weren't in a position to replicate the 13 analysis. 14 Q Well, the data come from Postal Service library 15 references, do they not? 16 A Well, they do. 17 Q You must have access to those while at the Postal 18 Service. 19 A I do, but I don't have access to all of my staff 20 and the computer facilities to do it. 21 Q Well -- 22 A I mean, I just didn't have time to do it. 23 Q Let me give you my hypothesis. You can tell me 24 where I'm making a mistake. You see the presort code D? 25 A Yes. 19432 1 Q On the left. And that's marked three-digit 2 carrier route, and if you -- and that's 390 million pieces. 3 Right? 4 A That's correct. 5 Q Now do you consider the three-digit carrier route 6 pallet to be a five-digit pallet? 7 A I don't believe we did for purposes of calculating 8 the 43-percent number, but I'd have to check that. 9 Q Well, can you tell me how you got to the 10 43-percent number without it? 11 A I can't off the top of my head. I mean, I'm not 12 that familiar with the spreadsheet. 13 Q What's a carrier route pallet? That's presort 14 code B. 15 A I believe it's a pallet with carrier route bundles 16 on it. 17 Q Is there such a thing as a carrier route pallet? 18 Other than a five-digit carrier route or a three-digit 19 carrier route, which are separately identified? 20 A The study would seem to indicate there are, but 21 I'm not sure that that's a legitimate makeup. 22 Q This study you're referring to is what? 23 A The 1993 mail characteristics study. 24 Q Have you seen the instructions for completing 25 record of pallet contents, form C, from that pallet study? 19433 1 A Not recently. 2 Q Let me hand you a copy. 3 A Okay. 4 Q Now, would you confirm that the pre-sort codes 5 listed there, A through K, are consistent with those shown 6 on this cross examination exhibit? 7 A Yes. 8 Q And for pre-sort code A, firm, it says not 9 applicable, does it not? 10 A Yes. 11 Q And then this exhibit would show a zero. Is that 12 right? 13 A That's correct. 14 Q And what does it say for pre-sort code B, carrier 15 route? 16 A Not applicable. 17 Q But the study showed how much volume on pre-sort 18 code B pallets? 19 A Three hundred and seventy-six million. 20 Q Can you explain why, if it's not applicable, there 21 are 376 million pieces on such pallets? 22 A No, I cannot. 23 Q Can you explain why -- well, never mind. 24 MR. STRAUS: Mr. Chairman, we believe that the 25 witness has made an arithmetic mistake and that adding up 19434 1 the five-digit pallets from these numbers, from this study, 2 don't produce the 43 percent he said but produce 35 percent. 3 CHAIRMAN GLEIMAN: I'm not sure why you're 4 addressing me on this matter at this point in time, Mr. 5 Straus. 6 MR. STRAUS: All right. 7 CHAIRMAN GLEIMAN: I think you ought to continue 8 with your cross examination. 9 BY MR. STRAUS: 10 Q Mr. Degen, would you check your math and see 11 whether the 43 percent is -- whether the five-digit 12 percentage is 43 percent or some lower number and report 13 back? 14 A Do I have to? 15 Q You swore that your testimony was the truth, and 16 if you made a mistake, I think you'd want to correct it. 17 CHAIRMAN GLEIMAN: Postal Service does have an 18 obligation, if there are errors that they discover or that 19 are pointed out to them, to correct their testimony, and you 20 know, we have a continuing flow of corrections flowing in -- 21 coming into this place on almost a daily basis. 22 MR. KOETTING: I guess my objection, if it has to 23 be an objection, is asked and answered. It was asked if the 24 43 percent came directly from the study and, as I recall, 25 the witness said no, the data from the study were adjusted 19435 1 to reconcile them with billing determinate information. 2 MR. STRAUS: That's not the adjustment I'm talking 3 about. 4 MR. KOETTING: Okay. 5 CHAIRMAN GLEIMAN: If there is an error, I expect 6 the witness to report back on a piece of paper to the Postal 7 Rate Commission in response to this cross examination that 8 there is an error and to provide the corrected figures -- 9 figure or figures. 10 The Postal Service has an obligation to do that. 11 We have to understand what the numbers are. You're the only 12 people that have the numbers. 13 You can proceed with your cross examination. 14 MR. STRAUS: Thank you. 15 BY MR. STRAUS: 16 Q Mr. Degen, when you do that, I would like you to 17 examine both the three-digit carrier route pallet entry in 18 this study from which you derived your 43 percent and the 19 pre-sort code B carrier route pallet entry to determine 20 whether those are -- either of those or both of those are 21 appropriately included within five-digit pallets. 22 MR. STRAUS: At this point, I would like this 23 document entitled "USPS Periodical Pallet Volumes, 1993" to 24 be marked, copied into the record, and admitted into the 25 evidence marked as ABP-XE-3. 19436 1 CHAIRMAN GLEIMAN: It is so directed. 2 [Cross-Examination Exhibit No. 3 ABP-XE-3 was received into 4 evidence, and transcribed into the 5 record.] 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19438 1 BY MR. STRAUS: 2 Q Are you pretty confident in your number that the 3 number of five-digit pallets by 1996 was 11 percent of total 4 palletized mail? 5 A With respect to that, I'm confident that there has 6 been a significant decline in the percentage of palletized 7 mail on five-digit pallets, and that's the extent to which I 8 rely on those numbers. I don't mean to offer it as 9 necessarily accurate, you know, right to the last digit. 10 Q But you've forecast a 75 -- you've forecast a 11 decline from 43 percent to 11 percent in only three years. 12 That's a 75-percent decline, isn't it? 13 A In terms of the change in the percentages, yes, 14 but again, I didn't forecast that. I believe -- I mean 15 those are the numbers that I have obtained looking at this 16 study, and even -- and my point is the same. Even if it's 17 35 percent to 11 percent, I'm simply observing that there 18 has been what I believe to be a significant increase in 19 five-digit pallet use in '92 and '93 and then a decline 20 therefrom in subsequent years. 21 Q Is Time-Warner one of the biggest palletizers of 22 periodicals? 23 A In terms of today? 24 Q Yes. 25 A Yes. 19439 1 Q Did you look at the Time-Warner numbers 2 introduced, the palletization numbers for Time-Warner 3 introduced in the pallet case and introduced in this case to 4 see what kind of five-digit volumes they were producing and 5 what their trend was? 6 A I've seen them. You gave them to me as a cross 7 examination exhibit. I hadn't gone back and looked at them, 8 but I don't find them relative, because they're lacking they 9 key element, which is what was the pallet profile in 1992 10 and '93. 11 Q Did you ask for 1993 data? 12 A Last night I offered to personally buy dinner for 13 anybody on my staff who could find a Time Warner pallet 14 profile for '92 or '93. That's how bad I wanted one. 15 MR. KOETTING: I would add for the record, Mr. 16 Chairman, that the Postal Service has exactly the kind of 17 information the Postal Service sought to obtain through the 18 discovery process in the earlier phase of the case. So the 19 information has been requested. 20 MR. STRAUS: And Time Warner provided some 21 information, as did some of the other publishers, about 22 palletized data, but not everything that was requested. Is 23 that right? 24 MR. KOETTING: I would believe that's correct, and 25 I'm sure that the record will show whether or not it 19440 1 included any information from 1993 or not. 2 MR. STRAUS: Well, let me state that it did not, 3 but that when Mr. Degen's rebuttal testimony began to be 4 analyzed, 1993 data have now been dug out for some of the 5 periodicals, the major ones, and if the Postal Service would 6 like to see it as badly as Mr. Degen says, I'd be happy to 7 offer into evidence an affidavit by Mr. James O'Brien from 8 Time Warner. I believe Mr. O'Brien is in the room and could 9 personally swear to the veracity of this data. 10 MR. KOETTING: The Postal Service would most 11 vehemently object to that at this late date in this 12 proceeding. 13 MR. STRAUS: That's what I thought. 14 CHAIRMAN GLEIMAN: Do I take it then that you're 15 not trying to move that in, and that I don't have to rule on 16 the objection? 17 MR. STRAUS: I don't like to lose. 18 BY MR. STRAUS: 19 Q What's the biggest -- what's the 20 highest-circulation periodical in the country? 21 A I believe it's TV Guide. 22 MR. KOETTING: Was that question circulation by 23 mail or total circulation? 24 BY MR. STRAUS: 25 Q By mail. Your answer is still TV Guide, I assume. 19441 1 A Yes, that was the context with which I was 2 answering. 3 Q And that would be about 8 million a week or about 4 400 million a year? 5 A Oh, I don't have those numbers at my fingertips. 6 I just have a general sense of the ordering. 7 Q Well, you don't have them at your fingertips. Do 8 you have your stricken testimony? 9 A No, it was stricken. 10 Q And you threw it away? 11 Well, let's assume subject to -- will you assume 12 subject to check that it's about 8 million a week? That's 13 not a hard one to check. 14 A I'd just as soon not check it, but sounds 15 reasonable to me. 16 Q Well, I think I'm going to ask you again, or maybe 17 ask counsel, will you accept subject to check that TV Guide 18 has a circulation of 8 million a week? 19 CHAIRMAN GLEIMAN: I think you need to ask the 20 witness subject to check, not counsel subject to check. 21 MR. STRAUS: Well, maybe counsel's willing to 22 check it if the witness isn't. 23 CHAIRMAN GLEIMAN: Would you ask the question of 24 the witness, please, so that we can move on. 25 BY MR. STRAUS: 19442 1 Q Assume that it's eight million a week, okay? Is 2 TV Guide predominantly palletized on five-digit pallets? 3 A That's my understanding. I wouldn't say 4 predominantly, but it's -- it uses a higher proportion of 5 five-digit pallets than most other large publications I'm 6 familiar with. 7 Q You wouldn't say predominantly? 8 A I'm not -- I don't have the numbers at my 9 fingertips. I am comfortable at this point saying that they 10 are -- their proportion of five-digit pallets is higher than 11 most large publications. 12 Q Do you ever recall writing something that said 13 that said that TV Guide's five -- is predominantly a 14 five-digit palletizer? 15 A Yes. 16 MR. KOETTING: Mr. Chairman, at this point I'm not 17 sure exactly on what basis I want to interject here, but 18 there are a couple of things going on. Having moved to 19 strike portions of Mr. Degen's testimony, Mr. Straus now 20 seems intent on conducting oral cross-examination of it 21 nonetheless, and he obviously is venturing into territory of 22 mailer-specific information, which is exactly the area that 23 the Postal Service had concerns about, which led fairly 24 directly I believe to that testimony being stricken. 25 So therefore I'm -- I can't say that I have a 19443 1 specific objection or anything, but I do want to point out 2 that there are some problems with this line of inquiry on 3 that basis. 4 CHAIRMAN GLEIMAN: If you didn't make a specific 5 objection, I guess I don't have anything specific to say 6 about it. And my string of untarnished procedural rulings 7 will continue. 8 [Laughter.] 9 MR. KOETTING: I raise it, Mr. Chairman, merely to 10 set the background for whenever there is an objection, we 11 don't have to go into it again. 12 [Laughter.] 13 CHAIRMAN GLEIMAN: Thank you, sir. I appreciate 14 that help. 15 BY MR. STRAUS: 16 Q Mr. Degen, would you accept, subject to check, 17 that TV Guide is 6.72 percent of all regular-rate Second 18 Class pieces? 19 A No. I mean, I don't want -- I don't want 20 homework. 21 Q Did you check to see what percentage of the total 22 pallets in this -- in 1996, the total five-digit pallets are 23 TV Guide pallets? 24 A No, that's not a -- well, let me ask you to ask 25 that again. Maybe I'm missing -- 19444 1 Q Let me go through some arithmetic that you won't 2 accept subject to check. 3 A Okay. 4 Q Let's assume that TV Guide is 6.72 percent of all 5 regular-rate periodicals and that as you say 56 percent of 6 all periodicals were palletized -- no, not five digits -- 56 7 percent of all periodicals were palletized to any extent in 8 1996. Then TV Guide would be 11 to 12 percent of all 9 palletized mail. 10 A Where did I say the 56 percent? 11 Q Isn't that what you say for the total 12 palletization for 1996? 13 A Could you point me to that? 14 Q Library Reference 190 -- are you -- 15 A Library Reference what? 16 Q 190. 17 A What's the name of that one? 18 Q It's Christensen Associates study of 19 palletization. 20 A Okay. All right. Okay. 21 Q Mail Characteristics Study. 22 A Yes. 23 MR. KOETTING: I would point out that that wasn't 24 entered into evidence by this witness. 25 BY MR. STRAUS: 19445 1 Q Okay. But does it say 56 percent of the 2 periodicals are palletized? 3 A I'll accept that subject to check. 4 Q Is Meredith a big palletizer? 5 MR. KOETTING: Mr. Chairman, at this point, I am 6 going to inject an objection because Mr. Degen, as a 7 consultant for the Postal Service, has limitations imposed 8 on the use with which he can make of the information that he 9 examines from Postal Service records. It's not clear to me 10 that he can really proceed without having some problems in 11 that regard. 12 BY MR. STRAUS: 13 Q Well, let me ask you, Mr. Degen. Did you review 14 the information provided by Mr. Littell of Meredith to the 15 Postal Service in response to its discovery requests? 16 A Yes, I did. 17 Q Do you have any reason to believe that has to 18 remain confidential? 19 A No. 20 Q Is Meredith a heavy palletizer? 21 A I am not comfortable subjectively characterizing 22 people as heavy or big, and the truth is because this 23 portion of my testimony was stricken, I have not recently 24 familiarized myself with the statistics to which you are 25 trying to point me. I'm just not comfortable making the 19446 1 kinds of characterizations you are asking me to make. 2 Q The statement that only 11 percent of the 3 palletized mail is five digit was not stricken, was it? 4 A No, it was not. 5 Q Yet, we have data in the record from Time Warner 6 showing that 11 percent of their pallets, roughly 11/12 7 percent, for their major magazines, are five digit. We have 8 T.V. Guide, which is the biggest mailer in the country, 9 which you admit is predominately five digit. We have Mr. 10 Littell submitting discovery responses from Meredith saying 11 that 35 percent of Better Homes and Gardens is five digit. 12 I'm asking you whether in light of all those 13 facts, you still contend that only 11 percent of the pallets 14 are five digit pallets? 15 A Yes, there is a lot of other periodicals out 16 there. 17 Q Do you know what percentage of the total 18 periodicals consist of Time Warner periodicals -- total 19 number of pieces are published by Meredith, Time Warner and 20 T.V. Guide? 21 A Not off the top of my head. 22 Q Do you think it's half? 23 A I don't have an opinion. 24 Q What's the minimum pallet weight? 25 A I believe it's currently 250 pounds. 19447 1 Q How long has it been 250 pounds? 2 A I think somewhere in the early '90s. 3 Q Is it easier or harder to make up a 5-Digit 4 pallet -- well, let me back up. 5 Before it was 250 pounds, how many pounds was it? 6 A I believe before that it was 500 and 650 before 7 that. 8 Q So as the weight reduced, did it become easier or 9 more difficult to make up 5-Digit pallets? 10 A Easier in what sense? 11 Q Easier to accumulate enough weight to make up a 12 pallet to a 5-Digit zip code? 13 A Yes, one would need fewer pieces to make up a 250 14 pound pallet than they would for a 500 pound pallet. 15 Q What pallet options are available to a mailer with 16 at least 500 pounds to a 5-Digit zip? 17 A They should be making up a 500 pound pallet. 18 Q To a 5-Digit zip? 19 A To a 5-Digit zip, yes. 20 Q You say they should be. Are they required to? 21 A You mean in practice or in the regs? 22 Q The Domestic Mail Manual is the regulations. Does 23 the Domestic Mail Manual require you to prepare a 5-Digit 24 pallet when you have 500 pounds? 25 A I believe it does. 19448 1 Q You know how the average weight of periodicals in 2 1996 compares with 1993? 3 A No, not as I sit here right now. 4 Q Do mailers have incentives to make up 5-Digit 5 pallets? 6 A What kind of incentives did you have in mind? 7 Q I am asking you whether you know whether -- but I 8 had in mind for example service, better service. 9 A Service is certainly an incentive that would argue 10 for smaller pallets. 11 I think, you know, savings in make-up and 12 transportation might argue for larger pallets, so there 13 might be other incentives. 14 Q When you say smaller and larger, are you saying -- 15 A More aggregate. I am equating the notion of a 16 more aggregate pallet with a heavier and bigger pallet. 17 Q Isn't it true that effective January 1, 1997 the 18 Postal Service implemented rules that could reduce the use 19 of 5-Digit pallets? 20 A I think that is true but I am not real familiar 21 with those. 22 Q Why do they do that? 23 A I am not sure. 24 Q To increase costs, do you think? To decrease 25 costs? You have no idea? 19449 1 A I am not familiar with the reasoning for the 2 change in regs. 3 Q To the extent that mail has shifted from sacks to 4 pallets, would that increase or decrease processing costs? 5 A I think there's really not enough information in 6 that statement for me to offer an opinion. 7 If you mean to the extent that used to be in 8 5-Digit sacks is now in 5-Digit pallets, I believe that 9 should decrease costs, processing costs. 10 If mail that was in 5-Digit sacks is now on all 11 mixed pallets, I don't think it would. 12 Q Has the Postal Service encouraged mailers to move 13 out of sacks and onto pallets? 14 A I believe they have. 15 Q Do you know why? 16 A I believe they expect That to lower costs. 17 Q You don't share that expectation? 18 A I do. I don't think anywhere here I have said 19 that I think 5-Digit mail or a 5-Digit pallet is more costly 20 to handle than a 5-Digit sack. I believe my testimony and 21 my discussion here has all been with respect to the change 22 in the make-up of the container. 23 Q Please look at page 34. 24 A I have it. 25 MR. STRAUS: Well, before we do that, maybe we 19450 1 ought to put into the -- get marked an exhibit we discussed. 2 Q You had I guess on your own mentioned that we had 3 provided you with Time Warner palletization data from 1991 4 and 1997. You said that you wished you had had the 1993, 5 but the 1991 and 1997 data we did provide you a cross 6 examination exhibit, did we not? 7 A That's correct. 8 Q And the source of that data is contained in the 9 records and the Postal Rate Commission proceedings, isn't 10 that right? 11 A That is what the table indicates. 12 Q Do you have any reason to doubt the validity of 13 the data? 14 A No, I do not. 15 MR. STRAUS: Mr. Chairman, I would like to 16 distribute this document and have it marked as ABP-XE-4, 17 transcribed into the record and admitted into evidence.c 18 MR. KOETTING: No objection. 19 CHAIRMAN GLEIMAN: It is so ordered. 20 [Cross-Examination Exhibit ABP-XE-4 21 was received into evidence and 22 transcribed into the record.] 23 24 25 19452 1 CHAIRMAN GLEIMAN: Mr. Straus, what was the page 2 number you were referring to in the testimony? 3 MR. STRAUS: 34 is where we are going next. 4 CHAIRMAN GLEIMAN: Thank you. 5 BY MR. STRAUS: 6 Q There you are critical of MPA Witness Little for 7 comparing, for relying on a comparison of costs of classes 8 of mail, isn't that right? 9 A Yes, that's correct. 10 Q Isn't it true that Mr. Little not only compared 11 between classes but also compared costs with wage rates? 12 A I believe that's true. 13 Q Isn't it also true that in R94-1 Postal Service 14 Witness Barker did a cross-class comparison of processing 15 costs by class? 16 A I am not familiar with that. 17 Q Again revisiting a subject you had discussed 18 earlier in your testimony -- we did some cross examination 19 on it -- page 34, lines 16 through 18 -- where you say that 20 presortation, drop shipping and mail piece readability can 21 have a substantial impact on the observed trend and 22 aggregate costs. 23 Now you confirmed, did you not, that there is more 24 presortation of periodicals than there used to be? 25 A With respect to pieces in bundles, yes, I did. 19453 1 Q And there is more -- is there more drop shipping? 2 A I believe that's true. 3 Q And what did you mean by mail piece readability? 4 A I probably should have said machinability there is 5 a broader term, but what I meant was, you know, the ability 6 of machines to read the barcode. 7 Q And there is more of that as well? 8 A You are talking about periodicals? 9 Q Yes. 10 A Certainly it's the case that there are more 11 barcodes on mail. 12 Q And so in each of these three respects, these 13 factors should have led to a reduction in aggregate unit 14 costs, should they not, taken in isolation? 15 A Yes, taken in isolation I would expect that to 16 have been true. 17 Q And again, you did not attempt to quantify either 18 the degree of change or the cost effect of any of these 19 changes, did you? 20 A No, I did not. 21 Q On page 38, beginning on line 3, you discuss the 22 Postal Service Governors' approval of the addition of 23 barcode readers to the FSM-1000 sorters, flat sorters. This 24 is underneath a heading that says that the Postal Service 25 has initiatives underway that will improve service, control 19454 1 costs, and work with mailers for further improvements. 2 A Yes. 3 Q So the barcode readers on the FSM-1000s, will they 4 control costs? 5 In other words, will they reduce costs? 6 A I believe that is the expectation. 7 Q And when will the mailers see the benefits of 8 those reduced costs? 9 A My understanding is that the component of the 10 roll-forward in this case accounts for additional FSM 11 deployments. I don't have any first-hand knowledge as to 12 whether an element of that is the deployment of barcode 13 readers or not, so I am not really in a position to say. 14 Q Well, the deployment of barcode readers doesn't 15 affect the rates paid by mailers, does it? 16 A I think to an extent that an adjustment is made in 17 the roll-forward process to determine test year costs, I 18 believe it does have an impact. 19 Q Let me be more direct then. When will there be a 20 presort discount associated with these barcode readers on 21 the FSM-1000s? 22 MR. KOETTING: Mr. Chairman, that is clearly 23 beyond the scope of this witness's testimony, to predict 24 when there is going to be a presort discount in the future. 25 CHAIRMAN GLEIMAN: Sustained. 19455 1 MR. STRAUS: I'm finished. 2 CHAIRMAN GLEIMAN: Does that mean you don't have 3 any other questions? 4 [Laughter.] 5 MR. STRAUS: I've got Bonnie Blair coming to 6 finish for me. 7 CHAIRMAN GLEIMAN: Is there any follow-up? 8 MR. KEEGAN: Yes, I'm sorry to say there is, Mr. 9 Chairman. 10 CHAIRMAN GLEIMAN: It's okay. We spend more than 11 90 percent of our time on postal matters. 12 [Laughter.] 13 CROSS EXAMINATION 14 BY MR. KEEGAN: 15 Q Good afternoon, Mr. Degen. I am Timothy Keegan, 16 representing Time Warner. 17 A Good afternoon. 18 Q I just want to follow up on several of Mr. 19 Straus's lines of cross examination. I expect this will be 20 brief. I will try to make it so. 21 You mentioned that mail processing, periodicals 22 mail processing costs, had declined relative to postal wages 23 in 1992, is that correct? 24 A Yes. 25 Q Do you know whether that decline may be due in any 19456 1 part to changes in the IOCS in that year or do you have an 2 opinion on that? 3 A I don't have the exact timing but I am aware of 4 some changes to IOCS that have improved the identifications 5 of periodical mail that may be more accurately measuring 6 periodicals' costs, but I am not prepared to discuss the 7 details of that or the exact timing or its impact. 8 Q Would you refer to page 3 of your testimony at 9 line 8, where you are discussing a change in the IOCS and 10 you say beginning in FY 1992. 11 A Yes. 12 Q And then describe the change. Do you happen to 13 know whether the changes were made all at the same time, as 14 opposed to a series of changes at different times? 15 A I don't know the exact timing of the changes. I 16 don't think they were as one big change. It may have been 17 ongoing. 18 Q Okay. A different subject, and this is from very 19 early on in Mr. Straus's cross examination you were 20 discussing 5610 tallies, which are, as I understand it, 21 letter-specific mixed-mail tallies, and in particular you 22 were discussing -- I'm sorry letter-specific tallies, and in 23 particular, Mr. Straus was asking about 5610 not handling 24 tallies and the circumstances in which one might have such a 25 tally. Is that correct? 19457 1 A I believe that was the gist of the conversation. 2 Q Just to follow up on that a bit, I would like you 3 to assume that an employee is clocked into an opening unit 4 and that that employee is recorded by IOCS with a 5610 not 5 handling tally. 6 A Okay. 7 Q In that instance, would you characterize that as a 8 case of mis-clocking? 9 A No. 10 Q Would you say that the employee in that instance 11 -- let me go back one step. Let me stipulate -- I think I 12 did -- the employee is clocked into an opening unit 13 operation. 14 A That's correct. 15 Q In that case, would you say that the employee was 16 properly clocked into the operation but nevertheless, he was 17 recorded at a letter operation not handling mail? 18 A Well, it's your hypothetical, so I have to say, 19 you know, was he or was he not properly clocked in. Which 20 would you like me to assume? 21 Q This hypothetical involves about $100 million, and 22 which we have such tallies. Let me ask if it is one of the 23 possibilities that in that case, the employee is properly 24 clocked into an opening unit, but is in fact observed by the 25 IOCS data collector at a letter operation not handling mail? 19458 1 A What's the question? 2 Q The question is if you assume that the employee is 3 clocked into an opening unit and is recorded with a 5610 not 4 handling tally, is it possible that what that situation is 5 recording is an employee who is properly clocked into the 6 opening unit but who is observed by the IOCS data collector 7 at a letter opening operation not handling mail? 8 A Yes, that's a possibility under your hypothetical. 9 Q Well, can you tell me what other possibilities 10 there are? You have said mis-clocking is not -- you would 11 not conclude there is mis-clocking, so in addition to those 12 two -- 13 A I didn't say mis-clocking wasn't a possibility. I 14 said I wouldn't necessarily conclude that. 15 Q In addition to those two possibilities then, what 16 other possibilities do you think are covered by the 17 assumptions that I have given you? 18 A Well, that the data collector incorrectly applied 19 the 5610 rule in terms of incorrectly determining that the 20 person was really working in that operation, but in terms of 21 does that create an error, I don't think it presents a 22 problem in terms of my mixed mail distribution because in 23 fact, it has the effect of associating that employee with 24 the shape of mail with which he was working. 25 Q Would you turn to page nine of your testimony, and 19459 1 I'd like to ask you about lines five through seven, which 2 Mr. Straus also asked you about, and I just wanted to follow 3 up a bit. 4 You state starting at line five on page nine, "We 5 know that the proportions and composition of direct, mixed 6 and not handling tallies vary across the MODS based cost 7 pools. This fact alone would argue definitively for the use 8 of these cost pools." Is that a correct reading of your 9 testimony? 10 A Yes, it is. 11 Q Would you again hypothetically for me just assume 12 that you take all your MODS cost pools and put them into a 13 blender and turn on the blender and then turn it off and 14 come out of the blender with a set of random cost pools, 15 covering the same total costs, but randomly organized. In 16 that event, would you expect that the proportions and 17 composition of direct, mixed, and not handling tallies would 18 vary across the resulting cost pools? 19 A If I understand your hypothetical correctly, you 20 are just introducing random variance into these cost pools 21 which essentially destroys what we know about the cost 22 pools, as I use them from an operational basis. 23 Q That's correct. 24 A I think you are saying if you make these things 25 completely random, am I going to be surprised that they are 19460 1 completely random, and I think the answer is no. 2 Q I'm not asking you that. I'm asking you whether 3 you would expect that the proportions and composition of 4 direct, mixed and not handling tallies would vary across 5 those random cost pools. 6 A I think if your blender does a good job and the 7 sample is large enough, they wouldn't. 8 Q They would not? All right. 9 A If I understand your hypothetical properly, I 10 don't believe they would. If by your blender, you mean that 11 you distribute them so that the cost pools are homogeneous, 12 I'm expecting they will be homogeneous. 13 Q That's fair. I accept that. Let's change the 14 hypothetical and throw in only half the MODS cost pools into 15 the blender. Same question. Would you expect that the 16 resulting set of pools would show proportions -- for the 17 resulting set of pools, the proportions and composition of 18 direct, mixed and not handling tallies would vary across the 19 pools? 20 A I would expect there would be a variation that is 21 dampened by having homogenized half of the tallies in the 22 cost pools, but the underlying operation based variation, I 23 think, would still be present. 24 Q In that case, would you conclude that the mere 25 fact of that variation argued definitively for the use of 19461 1 those pools? 2 A Yes, even the fact that half of these things 3 hadn't been tossed into that blender means there is some 4 information there based on our operational breakdown, and I 5 think that argues definitively for their use. 6 Q The pools in which half are entirely random and 7 half for your original pools? 8 A Yes, because there is still information there. I 9 think you want me to say we should chuck it all back 10 together, but until you put the whole thing in the blender, 11 there's no point to that. 12 Q Is the implication of what you are saying that the 13 set of pools that displayed the greatest possible degree of 14 variation would be the set of pools to be preferred over all 15 others? 16 A No, I don't think that's what I'm saying at all. 17 If you mean that the pools with the largest single sub-class 18 association should be preferred over the others, no, I don't 19 think so. I mean that's just the operational reality of 20 that pool versus another that doesn't have such a strong 21 sub-class association, but I don't believe I've relied on 22 sub-class -- individual sub-class association as a criterion 23 in any of my discussion. 24 Q Nor did I ask you about it. I asked you simply 25 about variation in terms of the proportions and composition 19462 1 of direct, mixed and not handling, however you wish -- those 2 are your words, however you wish to interpret them. 3 My question is would variation of that sort, with 4 the greatest possible extent of the variation of that sort 5 mean you had the optimal cost pools? 6 A No, unless -- the important thing is that the 7 variation is not the result of a blender. The important 8 thing is that to the extent that we know there are strong 9 operational differences among cost pools that give us a 10 strong a priori reason to believe that the underlying 11 sub-class distribution in that pool is different from 12 another pool, then we should separate them. 13 Q One final line of questions, and this has to do 14 with the possibility of testing your assumptions, which as 15 you rightly pointed out, did come up on your cross 16 examination on your direct testimony. Correct me if I'm 17 wrong, I think you indicated that with respect to testing, 18 for example, assumptions about counted items and the 19 representativeness of counted items, of uncounted items, 20 that's really not a testable assumption, as you see it? 21 A I haven't put a lot of effort into studying that, 22 but I see a problem in trying to discern a priori which 23 would have been counted and which wouldn't have. You'd need 24 some kind of hypothesis to say, you know, but for us doing 25 this test, this one would have been counted and this one 19463 1 wouldn't have. I see that as very problematic. 2 Q Let me give you a scenario and tell me what's 3 wrong with it. Suppose that for the next three months, 4 random selection of mail processing facilities that you were 5 satisfied was representative, you had two sets of data 6 collectors and as soon as the first data collector decided 7 that we have here a mixed mail tally for a container, and at 8 that point, the second data collector would come along and 9 count it. At the end of those three months, you would 10 compare the actual counts with the proportion between the 11 counted and the uncounted mixed tallies and see whether they 12 were comparable. Could that be done? 13 A I think once you've done the first one, the word's 14 out that you're there, and the behavior of the data 15 collectors could be modified by the fact that the study's 16 going on. That's one of my concerns. 17 Another of my concerns is that there are reasons 18 why some items are not counted, and as I pointed out 19 earlier, an important one is the exigency of dispatch of the 20 mail, and clearly Mr. Straus didn't think that was that 21 important. But I think the Post Office takes that very 22 seriously, and if you're sampling an employee at a point in 23 time that he's wheeling the last container on a truck and 24 they want to close that door, I can tell you from firsthand 25 experience, it's very difficult from an operations 19464 1 standpoint to be able to count that mail. We've tried. 2 Q So that such a test in your view is simply an 3 impossible sort of dream and not worth investigating or -- 4 A Well, I think I've fully explained why I don't 5 think it's necessary, since we seem to have consensus that 6 the underlying subclass profiles differ. I don't see the 7 point to it, and I'm sure that if I came in here offering 8 such a test, these are the kind of criticisms I'd face for 9 it. So I'm not sure that we'd gain a whole lot doing it. 10 Q I misspoke when I asked you about counting 11 containers, did I not? Isn't it items that are counted? 12 A Well, we talk about counted containers in terms of 13 estimating the proportion of the container that is related 14 to each item type, so I was willing to let you go. 15 Q But in fact containers -- containers themselves 16 are never counted. 17 A Well -- 18 Q The pieces in containers are not counted when you 19 have a container tally? 20 A The items in containers are counted. The pieces 21 inside of the items in containers are not counted. 22 Q Thank you. 23 MR. KEEGAN: That's all, Mr. Chairman. Thank you. 24 CHAIRMAN GLEIMAN: There are no questions from the 25 bench. 19465 1 That brings us to redirect. Mr. Koetting, would 2 you like an opportunity to consult with your witness before 3 determining whether you want to do redirect? 4 MR. KOETTING: Yes, Mr. Chairman. 5 CHAIRMAN GLEIMAN: In that case you can have an 6 hour and 15 minutes to consult. And we'll come back -- 7 gosh, it's quarter to two. That'll make it three o'clock 8 when we get back here from lunch. You all should be 9 prepared for a long evening. 10 [Whereupon, at 1:45 p.m., the hearing was 11 recessed, to reconvene at 3:00 p.m., this same day.] 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19466 1 AFTERNOON SESSION 2 [3:00 p.m.] 3 Whereupon, 4 CARL G. DEGEN, 5 the witness on the stand at the time of the recess, having 6 been previously duly sworn, was further examined and 7 testified as follows: 8 CHAIRMAN GLEIMAN: Yes, sir, Mr. Koetting, do you 9 have some rebuttal? 10 MR. KOETTING: Redirect. 11 CHAIRMAN GLEIMAN: Redirect, I'm sorry. 12 MR. KOETTING: Mr. Chairman, I'm afraid that an 13 hour and 15 minutes was not sufficient time. With another 14 half an hour I think we would have been down to no redirect. 15 [Laughter.] 16 We got stuck in the middle with a few questions. 17 CHAIRMAN GLEIMAN: Well, let's do a cost-benefit 18 analysis here. How much redirect do you think you have? Is 19 it more than a half-hour? Because if it's more than a 20 half-hour, we'll let you have another half-hour to talk, and 21 we'll come out ahead on this, as I understand the way things 22 work. 23 MR. KOETTING: I agree with your analysis, but I 24 think I can assure you that I'm not going to be doing a half 25 an hour of redirect. 19467 1 CHAIRMAN GLEIMAN: I can only hope that you'll 2 agree with my analysis when we issue our decision. 3 MR. KOETTING: Hope springs eternal. 4 [Laughter.] 5 CHAIRMAN GLEIMAN: Doesn't it, though. For lots 6 of folks in the room. 7 [Laughter.] 8 Why don't you fire away, Mr. Koetting. 9 REDIRECT EXAMINATION 10 BY MR. KOETTING: 11 Q Mr. Degen, in your conversations with counsel for 12 ABP and Time we had a lot of talk about different types of 13 tallies being coded under a variety of very specific fact 14 situations. I'd just like for you to try to clarify for the 15 record, do the data collectors on the scene make the 16 assignment to the activity codes such as those 17 shape-specific not handling activity codes that were 18 discussed at some length this morning? 19 A No. In fact, none of the activity codes are 20 assigned by the data collector. They're assigned by 21 programs contained in Library Reference H-21. I think one 22 of them that does a significant number of assignments is ALB 23 40. But the data collectors report or answer the questions 24 such as 18 and 19 that ask questions about the activity, and 25 then the coding is done by the computer program. 19468 1 Q Could you turn to page 21 of your rebuttal 2 testimony, line 14. 3 A I have it. 4 Q Again, there was some conversation this morning 5 regarding your statement regarding the recommendation about 6 Witnesses Stralberg and Cohen treating not handling costs as 7 institutional. I would like -- do you have Dr. Stralberg's 8 direct testimony with you? 9 I would like you to look at the bottom of page 39 10 of that, starting around line 27. 11 A Yes, I have it. 12 Q Could you read the statements made by Mr. 13 Stralberg at that point in his testimony? 14 A Yes. It begins: In particular, little is known 15 about what really causes the $3,727 million accrued, and 16 then parenthetically, 2,733 million in volume variable, end 17 of parentheses, costs referred to above as general overhead 18 or not handling costs. All that can be said with certainty 19 about these costs is that they grew anomalously during the 20 past ten years, when the automation program was being 21 implemented. The Commission should seriously consider 22 treating these costs as institutional until the Postal 23 Service provides more reliable information about what causes 24 them. 25 Q Was that the type of statement you had in mind 19469 1 when you wrote your testimony on page 21 that we were just 2 discussing at line 14 and following? 3 A Yes. While I wasn't able to point to it 4 specifically, it certainly created the impression in my mind 5 that he was calling for institutionalization of all of these 6 costs. 7 Q Moving to another topic, does your methodology 8 assume that the tallies associated with the handling of 9 brown sacks will always be entirely associated with 10 periodicals? 11 A No. We do -- you know, brown sacks is one of the 12 item types that we define, but the actual distribution of 13 those costs is based on the -- what the exact class -- 14 subclass distribution that is observed in those brown sacks. 15 I think there's quite a bit of information in this record 16 that indicates that it is not 100 percent periodicals, but 17 it is a very high percentage, and again we use the strong 18 association as a reason for delineating that item type, but 19 the actual distribution reflects what's actually found in 20 those sacks. 21 MR. KOETTING: Thank you, Mr. Degen. That's all. 22 COMMISSIONER HALEY: Did the redirect generate any 23 further cross exam? 24 MR. KEEGAN: Yes, Commissioner Haley. Thank you. 25 RECROSS EXAMINATION 19470 1 BY MR. KEEGAN: 2 Q Mr. Degen, would you return to page 39 of Mr. 3 Stralberg's testimony, from which you read a sentence a 4 moment ago? 5 A I have it. 6 Q Could you read that sentence again, please? 7 A All of the sentences that I read? 8 Q The sentence beginning at line 27 with the words 9 "In particular." 10 A "In particular, little is known about what really 11 causes the 3727 million accrued, 2733 million volume 12 variable costs referred to above as general overhead not 13 handling costs." 14 Q Were you aware the first time you read that 15 sentence, you read it as general overhead or not handling 16 costs? 17 A No, I was not. I am sorry. 18 Q Is it your understanding that Mr. Stralberg uses 19 the term "general overhead not handling costs" synonymously 20 with the term "not handling costs?" 21 A No, I don't believe he does. 22 MR. KEEGAN: That's all. Thank you. 23 COMMISSIONER HALEY: Very good. Mr. Koetting? 24 MR. KOETTING: Nothing, Commissioner Haley. Thank 25 you. 19471 1 COMMISSIONER HALEY: All right. Well, then, Mr. 2 Degen, we certainly appreciate your appearance here today 3 and your contributions to the record. If there is nothing 4 further, then you are excused. 5 THE WITNESS: Thank you. 6 [Witness excused.] 7 COMMISSIONER HALEY: I'm sure you thank us. 8 Our next witness is appearing on behalf of the 9 United Parcel Service, Mr. Sellick. I believe he's already 10 been sworn in. 11 MR. McKEEVER: That's correct, Commissioner Haley. 12 COMMISSIONER HALEY: Mr. McKeever, will you 13 introduce your witness? 14 MR. McKEEVER: United Parcel Service calls to the 15 stand Stephen E. Sellick. 16 COMMISSIONER HALEY: Mr. Sellick, since you have 17 been sworn, we'll just get together today. 18 Whereupon, 19 STEPHEN E. SELLICK, 20 a rebuttal witness, was called for examination by counsel 21 for the United Parcel Service and, having been previously 22 duly sworn, was further examined and testified as follows: 23 COMMISSIONER HALEY: We will enter his rebuttal 24 testimony into evidence. 25 MR. McKEEVER: Thank you, Commissioner Haley. 19472 1 DIRECT EXAMINATION 2 BY MR. McKEEVER: 3 Q Mr. Sellick, I've just handed you a copy of a 4 document entitled Rebuttal Testimony of Stephen E. Sellick 5 on behalf of United Parcel Service, and marked as UPS-RT-1. 6 Was that document prepared by you or under your direction 7 and supervision? 8 A Yes, it was. 9 Q And if you were to testify orally here today, 10 would your testimony be as set forth in that document? 11 A Yes, it would be. 12 MR. McKEEVER: Mr. Chairman, I move that the 13 rebuttal testimony of Stephen E. Sellick on behalf of United 14 Parcel Service and marked UPS-RT-1 be admitted into evidence 15 and transcribed into the record. I do have two copies for 16 the Reporter. 17 CHAIRMAN GLEIMAN: Are there any objections? 18 [No response.] 19 CHAIRMAN GLEIMAN: Hearing none, Mr. Sellick's 20 testimony and exhibits are received into evidence and I 21 direct that they be transcribed into the record at this 22 point. 23 [Rebuttal Testimony and Exhibits of 24 Stephen E. Sellick, UPS-RT-1, was 25 received into evidence and 19473 1 transcribed into the record.] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19486 1 CHAIRMAN GLEIMAN: On March 17th, Witness Sellick 2 filed a revised response to Presiding Officer's Information 3 Request No. 16, and I'm going to hand two copies of the 4 revised response to the Reporter and ask that it be included 5 in the record at this point. 6 7 [Revised Response of UPS Witness 8 Sellick to POIR-16 was received 9 into evidence and transcribed into 10 the record.] 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19491 1 CHAIRMAN GLEIMAN: Two participants have requested 2 oral cross examination of Witness Sellick, Magazine 3 Publishers of America and United States Postal Service. 4 Does any other party wish to cross examine the witness? 5 [No response.] 6 CHAIRMAN GLEIMAN: If not, Mr. Gold, please 7 proceed when you are ready. 8 MR. GOLD: Thank you, Mr. Chairman. 9 CROSS EXAMINATION 10 BY MR. GOLD: 11 Q Good afternoon, Mr. Sellick. 12 A Good afternoon. 13 Q I'd like to begin with page four of your 14 testimony. 15 A I have that. 16 Q Specifically, table one. 17 A Yes. 18 Q In that table, you show certain sack types that 19 are associated with certain mail classes; is that correct? 20 A Yes, it is. 21 Q The source of this table is Ms. Cohen's table four 22 from MPA-T-2; is that correct? 23 A In part, yes. 24 Q You prepared this table, this is not her 25 testimony, this is your table; correct? 19492 1 A That's correct. 2 Q In fact, Ms. Cohen did not distribute the costs 3 this way, did she? 4 A No, she did not. 5 Q Basically, this is an implicit distribution that 6 you have derived from various sources? 7 MR. McKEEVER: Mr. Chairman, may I ask for 8 clarification? There are two columns of numbers there. May 9 I ask which column Mr. Gold is referring to? 10 MR. GOLD: Well, my questions went to the table, 11 but specifically to the last column where it says Cohen, 12 mixed mail distribution to associated classes. 13 MR. McKEEVER: Thank you, Mr. Chairman. 14 BY MR. GOLD: 15 Q Mr. Sellick, do you know the general magnitude of 16 not handling and mixed mail costs in this record? 17 A Not handling and mixed mail together? Something 18 on the order of $2 to $3 billion. 19 Q Isn't it more on the order of $7 billion? 20 A That's I believe if you count overhead as well, 21 but mixed mail and not handling separately I think are two 22 to three. 23 Q Of course, overhead is part of not handling; is it 24 not? 25 A Overhead is typically defined as the break -- the 19493 1 on break, the clocking in and out, and I believe moving 2 empty equipment time. In past cases, it's been treated a 3 bit separately and differently from mixed mail and not 4 handling mail costs. In this case, they are treated a bit 5 more together. 6 Q But what I'm trying to focus on, the pool of costs 7 that Ms. Cohen was focusing on included the large number, 8 did they not, including the overhead? 9 A Yes, all of those have to be distributed on some 10 basis to sub-classes of mail. 11 Q Do you know the general magnitude of the uncounted 12 sack costs represented by this table? 13 A Not off the top of my head, no, I don't. 14 Q Would you accept subject to check that it's 15 approximately $45 million? 16 A Sure. 17 Q Does that sound right? 18 A That sounds fine. 19 Q Therefore, the amount of cost discussed in this 20 section as a percentage of the mixed mail and not handling 21 costs is approximately .7 percent, if you accept my 22 arithmetic. Does it sound about right? 23 A .7 percent of the total amount of moneys 24 distributed like that? 25 Q My numerator's 45 million and my denominator is 7 19494 1 billion. 2 A I'm not sure that's an appropriate way of looking 3 at this, because the overhead costs are distributed on top 4 of direct and mixed and counted mixed and distributed mixed 5 costs, so it isn't necessarily that order. But this amounts 6 to $45 million. 7 Q Well, Mr. Sellick, I wasn't suggesting that this 8 table implies any distribution of those overhead costs. 9 What I'm trying to establish is the general magnitude of the 10 cost that you're focusing on with the pool of costs that Ms. 11 Cohen was focusing on, and in that light is my .7 percent 12 about right? 13 A I'll accept the calculation. 14 Q And if we looked just at mixed mail costs, the -- 15 would you accept that those are approximately $1-1/2 16 billion? Does that sound about right? 17 A For mixed mail? 18 Q For mixed mail. 19 A All mixed mail, not limited to items. 20 Q Correct. 21 A I'll accept that number. 22 Q And again doing the arithmetic, that's 23 approximately three point -- this table represents 24 approximately 3.1 percent of the total pool of costs that 25 she was discussing with respect to mixed mail. 19495 1 A Again, I'll accept the calculation, but again 2 point out that the discussion might be appropriately more 3 limited to item type distribution, which is the basis of 4 sacks. 5 Q Now this table was first derived and discussed in 6 UPS/MPA-2-T-1. Is that not correct? 7 A I believe that's correct. Yes. 8 Q Do you have a copy of that interrogatory? 9 A I don't believe I do with me; no. 10 MR. GOLD: Does counsel? 11 And for the record, that's at transcript 26, 12 14091. 13 MR. McKEEVER: Do you want me to give it to him? 14 MR. GOLD: Go ahead, unless you want me to. 15 THE WITNESS: I have that citation now. 16 BY MR. GOLD: 17 Q Okay. And would you look at question B under that 18 interrogatory? 19 A I have that. Question B or answer B? 20 Q Well, first question B. In that question UPS 21 asked her to confirm the distribution of the costs. 22 A Yes. 23 Q Is that not correct? 24 A Yes. 25 Q And her answer was I cannot confirm -- cannot 19496 1 confirm, because I do not know how you calculated the 2 percentages in the Cohen distribution to associated class 3 column. I did not perform such a calculation because I 4 distributed mixed mail costs by activity code, not by mail 5 type. 6 And then she goes ahead and gives a table 7 comparing the direct and mixed mail costs and shows that 8 they're closely associated, does she not? 9 A Yes, she does. 10 Q And looking at question C in that interrogatory -- 11 A Yes. 12 Q The question was please confirm that with the 13 exception of green sacks your distribution methodology would 14 result in a significantly reduced proportion of mixed mail 15 sacks, sack costs being distributed to their associated 16 classes relative to Witness Degen's distribution. 17 Is that a correct reading of that interrogatory? 18 A Yes, I believe it is. 19 Q And would you look at her answer to that? 20 A I have that. 21 Q And there she suggests that there is strong 22 evidence on the record that such associations would be far 23 weaker in mixed mail sack tallies than in direct sack 24 tallies, particularly identical sack tallies. 25 Do you see that in her answer? 19497 1 A I do see that. I recall her response and I see it 2 here. Yes. 3 Q Do you have any reason to disagree with that? 4 A No. Identical sack tallies would tend to, when 5 combined with accounted mixed, would tend to increase the 6 proportion, as I believe I point out in a footnote in my 7 rebuttal testimony, that the association on page 5, footnote 8 5, the association between sack type and mail class is 9 slightly less strong for counted sacks alone, as opposed to 10 counted and direct sacks together, but the conclusion 11 remains the same. 12 Q And getting to the broader question in this 13 regard, doesn't she say that the question seems to suggest 14 that there is a known association between classes and sack 15 type for mixed sack tallies, and she seems to believe that 16 there's not. 17 A She -- yes, she does believe that there is not. I 18 do not agree with that belief. 19 Q In essence your table is based on an average, is 20 it not, an aggregation across 50 cost pools? 21 A I'm not sure I understand your question. 22 Q Well, did the Postal Service distribute these 23 costs in this way? Didn't they do it within cost pools? 24 A They did it within cost pool by item type where 25 appropriate. Yes. 19498 1 Q Would you turn to your testimony that begins on 2 the bottom of page 7? 3 A I have that. 4 Q From page seven to page eight, you spend 5 approximately 21 lines on this area of testimony; is that 6 correct? 7 A That seems about the right number of lines. 8 Q Is this the summary of all your criticisms about 9 this area? 10 A It says what it says. It represents points I wish 11 to make on rebuttal testimony. 12 Q Specifically looking at lines eight to 14, page 13 eight, would you look at that? 14 A I have that. 15 Q That says whether one believes that differing 16 degrees of not handling mail costs by mail processing 17 operation, in this case, MODS cost pools, are the result of 18 shifting employee assignments or instead, a manifestation of 19 evolving mail processing environments, increased 20 containerization and other factors, the fact remains that 21 not handling mail costs are higher in certain proportions 22 than in others. That's your testimony, is it not? 23 A Actually, with one small correction, certain 24 operations than in others. 25 Q I'm sorry. 19499 1 A Yes. 2 Q Certain operations than in others. Is it your 3 position that the various sub-classes of mail that are 4 "handled in an operation" should bear those costs regardless 5 of whether they caused those costs? 6 A That is the effect of distributing those costs, 7 that they bear the overhead costs in those pools. 8 Q When you say they bear those costs, does that 9 necessarily mean they cause those costs? 10 A It is not necessarily the conclusion that they 11 cause those costs, no. 12 Q Would you turn to your section on data thinness? 13 A I have that. 14 Q Do you agree that there needs to be five 15 observations for each cell in a distribution? 16 A I'm not sure -- I have not put that out as a 17 standard. I am commenting on Ms. Cohen's previous standard 18 that she established in that regard. 19 Q Do you have an opinion in that regard? 20 A Actually, having reviewed Mr. Degen's rebuttal 21 testimony and looking at his calculations on the coefficient 22 of variation, I believe that's a good way of looking at 23 determining whether the distribution method results in 24 significantly more variance than the LIOCATT system. 25 Q Does this mean that if there are six classes of 19500 1 mail observed in an item type, there would need to be 30 2 observations for there to be adequate data to do the 3 distribution? 4 A I'm sorry. Could you repeat the question? 5 Q We established that you need five observations for 6 each cell; correct? 7 A Well -- 8 MR. McKEEVER: Objection, Mr. Chairman. Mr. 9 Sellick indicated that was Ms. Cohen's test. If Mr. Gold, I 10 believe, is now trying to ask him what is Ms. Cohen's test. 11 MR. GOLD: I withdraw the question. 12 BY MR. GOLD: 13 Q You do agree that coefficient of variation is the 14 proper way to address this issue, do you not? 15 A In reviewing Mr. Degen's rebuttal testimony, it 16 seems to me that it's a good way to look at it. 17 Q In fact, the record in this case reveals that for 18 coefficients that she examined, for 70 percent of the 19 numerators, the coefficient of variations were so large that 20 there is no basis to suggest that the numerators are not 21 zero? Do you recall that testimony? 22 A I do recall some testimony to that effect, but I 23 was specifically referring to Mr. Degen's point that the 24 best measure, as I recall his testimony, is the final cost 25 estimates and the coefficient of variation of those final 19501 1 cost estimates, which are ultimately the numbers that are 2 used. 3 Q Would you turn to page ten of your testimony? 4 A I have that. 5 Q And at the very end of your testimony you state: 6 I strongly urge the Commission to adopt the cost 7 distributions provided in my supplemental testimony which 8 result from Mr. Degen's approach and are based on returning 9 the variability assumptions to those previously determined 10 by the Commission. 11 Is that your testimony? 12 A Yes, it is. 13 Q Is it your position that you can implement the 14 Degen approach if all cost pools are assumed to have the 15 same variability? 16 A The -- my position is that you can use the 17 approach I have taken, and that is taking costs using Mr. 18 Degen's approach with the previous Commission assumptions of 19 volume variability. I only state that -- restate that 20 slightly, in that although the cost pools for purposes of 21 running the SAS programs are returned to 100-percent 22 variability, implicitly some of them are not, because they 23 are treated as fixed mail processing and so on as discussed 24 in some of my work papers. But yes, I do believe that Mr. 25 Degen's approach can be used as I have used it. 19502 1 Q Do you have an opinion about whether this distorts 2 the cost distributions to allied activities? 3 A Distorts in what regard? 4 Q The fact that the -- all the cost pools are 5 assumed to have the same variability. 6 A I don't -- I'm sorry. 7 Q Do you have any opinion about the effect of that 8 on allied activities? 9 MR. McKEEVER: Objection, Mr. Chairman. I believe 10 that's beyond the scope of the rebuttal testimony, which 11 does not deal with allied operations at all, it deals with 12 two very narrowly defined subjects. 13 CHAIRMAN GLEIMAN: I believe Mr. Gold just nodded 14 his head in agreement. 15 MR. GOLD: I withdraw the question. 16 That's all I have, Your Honor -- Mr. Chairman. 17 CHAIRMAN GLEIMAN: The Postal Service? 18 MR. KOETTING: We have no questions, Mr. Chairman. 19 CHAIRMAN GLEIMAN: And? Is there any followup to 20 Mr. Gold's cross-examination? 21 No. 22 And there are no questions from the bench that I'm 23 aware of. 24 That brings us to redirect. Would you like some 25 time with your witness? 19503 1 MR. McKEEVER: Yes, Mr. Chairman, I'd appreciate 2 just a few minutes. 3 CHAIRMAN GLEIMAN: Okay. Why don't you take five 4 minutes? 5 MR. McKEEVER: Thank you. 6 [Recess.] 7 CHAIRMAN GLEIMAN: Mr. McKeever. 8 MR. McKEEVER: We have no questions, Mr. Chairman. 9 CHAIRMAN GLEIMAN: Inasmuch as there's no 10 redirect, I want to thank you, Mr. Sellick. We appreciate 11 your appearance here today and your contributions to the 12 record. And if there's nothing further, you're excused. 13 [Witness excused.] 14 CHAIRMAN GLEIMAN: Mr. Todd, I believe -- I 15 thought I saw you over there somewhere. I believe your 16 witness is the next witness. If you could identify him, so 17 that I can swear him in. 18 MR. TODD: The Mail Order Association of America 19 would like to present as its witness Mr. Roger C. Prescott. 20 Whereupon, 21 ROGER C. PRESCOTT, 22 a witness, was called for examination by counsel for the 23 Mail Order Association of America and, having been first 24 duly sworn, was examined and testified as follows: 25 DIRECT EXAMINATION 19504 1 BY MR. TODD: 2 Q Mr. Prescott, do you have in front of you a 3 document entitled Rebuttal Testimony of Roger C. Prescott, 4 which has been marked as MOAA-RT-1? 5 A Yes, I do. 6 Q And was this testimony prepared by you or under 7 your direction? 8 A Yes, it was. 9 Q And do you adopt it today as your testimony in 10 this case? 11 A Yes, I do. 12 MR. TODD: Mr. Chairman, I am handing two copies 13 of the identified testimony to the reporter with a request 14 that it be admitted into evidence and transcribed into the 15 record at this time. 16 CHAIRMAN GLEIMAN: Are there any objections? 17 Hearing none, Mr. Prescott's testimony and 18 exhibits are received into evidence, and I direct that they 19 be transcribed into the record at this point. 20 [Rebuttal Testimony and Exhibits of 21 Roger C. Prescott, MOAA-RT-1, was 22 received into evidence and 23 transcribed into the record.] 24 25 19547 1 CHAIRMAN GLEIMAN: Only one participant, Val-Pak 2 Marketing -- excuse me, Val-Pak Direct Marketing Systems, 3 Inc., Val-Pak Dealers Association, Inc., and Carol Wright 4 Promotions, Inc., filed a request for oral cross-examination 5 of the witness. 6 Does anyone else wish to cross-examine this 7 witness? 8 If not, Mr. Olson, you can begin when you're 9 ready. 10 CROSS-EXAMINATION 11 BY MR. OLSON: 12 Q Mr. Prescott, William Olson, representing Val-Pak 13 and Carol Wright. 14 And I want to ask you to begin by turning to 15 Appendix B of your testimony. 16 A Yes, I have it. 17 Q Okay. And the purpose of your testimony is to 18 rebut the testimony of Dr. Haldi on behalf of Val-Pak, Carol 19 Wright, VPCW-T-1; correct? 20 A Yes, it is. 21 Q And this is where you set out the errors and 22 assumptions in Witness Haldi's procedures in his what -- in 23 his presentation of ECR rates; correct? 24 A In this appendix and in my text; yes. 25 Q Okay. Well, what does the appendix -- the 19548 1 appendix says errors and assumptions, and then it says it 2 has theoretical errors, mathematical errors, faulty 3 assumptions -- aren't most of those in the appendix, or do 4 you have some division between what's in the text and what's 5 in the appendix? 6 A Well, I would describe the text as being an 7 overview of the theory and my explanation of the steps 8 followed by Dr. Haldi, and Appendix B is intended to 9 highlight some of the specific problems that occur in his 10 procedures. 11 Q The mathematical errors and the faulty 12 assumptions. 13 A Yes. 14 Q Okay. And in your testimony on page 1 you start 15 by saying at line 11: While the correction of mathematical 16 errors apparently would have little impact on the rates 17 proposed by Witness Haldi, I have identified all errors that 18 I have found in order to provide as complete a record as 19 possible. Correct? 20 A Yes. 21 Q Do you have an idea as to what the impact would be 22 of the correction of all of the errors that you have found 23 in terms of an order of magnitude of the effect on rates? 24 A The mathematical errors referenced in that 25 sentence? 19549 1 Q Yes. 2 A There were three specific mathematical errors that 3 I identified, and those were in table A-8, A-15, and table 4 C-2. And if you were to make those corrections only, it 5 would have the impact of approximately a tenth of a cent on 6 the rate structure. 7 Q Okay. And just for clarity, where do you discuss 8 the errors in A-8 in your appendix? 9 A The errors in Table A-8 would be discussed in 10 pages 4 and 5 of my Exhibit 1-B or the Appendix B. 11 Q And then E-15 is page 9? 12 A That's correct. 13 Q And the C-2 problems are on page 12? 14 A The C-2 mathematical error is discussed on pages 15 12 and 13. 16 Q Okay. Those are mathematical errors. 17 Now you also say there are faulty assumptions and 18 theoretical errors. Have you attempted to quantify the 19 degree of error from the theoretical errors and the faulty 20 assumptions? 21 A I have not restated the faulty assumptions or 22 provided my own assumptions, no. 23 Q So you don't know whether that would have little 24 impact on the rates proposed by Dr. Haldi or great impact on 25 the rates proposed by Dr. Haldi? 19550 1 A Well, it would depend on how you would change 2 those assumptions. For example, his assumption in Table 3 C-2, which has a weighting of 90 percent and 10 percent, if 4 you changed that weighting to 50/50, for example, assumed 5 that instead, you would get quite different results. 6 Q Actually you would get very much higher basic 7 rates, would you not, and much lower saturation rates? 8 A I believe that to be correct, yes. 9 Q Okay. Let me say then as you discuss these errors 10 and assumptions, mathematical errors, faulty assumptions, 11 that you discuss them table by table and then you, wherever 12 a number is rolled from one table into the next table, you 13 then say it's an error in that table because of the prior 14 analysis, correct? 15 That's the approach you took? 16 A As a general statement I would say yes. 17 Q Okay. So it is not just identified in the table 18 where the error is made according to your testimony, but 19 then wherever it rolls forward into a subsequent table you 20 have tried to point that out also, correct? 21 A That's correct. Dr. Haldi's structure to his 22 calculations are each table will build on a prior table. 23 Q Right. 24 A Right. 25 Q Let me ask you to take a look at your page 2 and 19551 1 your analysis of Table A-1, and there your criticism as I 2 understand it is that Dr. Haldi in Table A-1 relies on 3 overall average volume for letters and flats combined, for 4 transportation, and "other" -- correct? 5 A What line would you be referring to? 6 Q From your testimony, you mean? 7 A Yes. 8 Q 4 and 5, page 2.t 9 MR. TODD: Page 2 of the Appendix, counsel? 10 MR. OLSON: Yes. Everything will deal with the 11 appendix. 12 THE WITNESS: Yes, I see that. Yes. 13 BY MR. OLSON: 14 Q Okay, and so do you have Dr. Haldi's Table A-8 15 there -- A-1 on page A-8? Do you happen to have his 16 testimony? 17 A Yes, I have that. 18 Q Okay. So what you are criticizing, are you not, 19 is that in the transportation column for example he uses 20 .1877 for automation, basic, high density and saturation, 21 correct? 22 A For both letters and nonletters, yes. 23 Q Exactly -- irrespective of whether it is a letter 24 or nonletter and irrespective of condition of sortation, 25 correct? 19552 1 A That's correct. 2 Q Do you understand what the purpose of Table A-1 in 3 Dr. Haldi's analysis was? 4 Do you understand Table A-1 and A-2 to be the 5 development of a benchmark from which later analysis would 6 be derived, but that was not the totality of the analysis? 7 A That is not the totality of the analysis. 8 Q Okay, so for example, what Table A-1 and A-2 does, 9 correct me if I'm wrong, is attempt to extract from the 10 Postal Service's case unit costs for mail processing, 11 delivery, transportation and other and then multiply it by 12 test year after rates volumes, correct? 13 A Yes. 14 Q Okay, and then he comes up with test year -- 15 excuse me, Dr. Haldi compares that to CRA costs derived from 16 Witness Patelunas at the end of Table A-2, correct? 17 A He compares it to the CRA costs in total. In 18 Table A-2 he also develops the costs separately between 19 letters and nonletters. 20 Q Okay. 21 A And that is the point at which the costs are no 22 longer valid because of the assumptions that he has made in 23 Table A-1. 24 Q Okay, but when he goes -- when he runs his numbers 25 and compares it to CRA costs, do you have a problem with his 19553 1 analysis and how far it is away from CRA costs? 2 Have you looked to see how close this benchmark 3 got him? 4 A In Table A-2, the CRA costs are quite close to the 5 costs developed by Dr. Haldi. The problem is that in Table 6 A-2 he is separating the total costs between letters and 7 nonletters and the problem is in that separation, and those 8 numbers, those aggregate costs for letters and nonletters 9 are used later on in his analyses to develop his rates. 10 Q Okay, but when you criticize him for using the 11 same transportation number, an average transportation number 12 in Table A-1, you are not saying, are you, that that is 13 where his analysis ends, but rather does he not develop 14 shipping, unit shipping costs at a later point in his 15 analysis? 16 A Yes, he does. 17 Q Okay. So the Postal Service did not provide -- 18 isn't it true the Postal Service did not provide information 19 from which to be able to put in -- information from which 20 Dr. Haldi could have drawn -- unit costs for transportation 21 and others -- to put different numbers into this first 22 table? 23 A The Postal Service did not provide numbers 24 separated between letters and nonletters, that's correct. 25 Q Okay. Let me ask you to turn to the next page, to 19554 1 page 3 and specifically Section D, where you discuss Table 2 A-6. 3 A Okay. 4 Q You say that Witness Haldi, and I'm reading from 5 lines five through seven here, Witness Haldi's separation of 6 total pieces for ECR pound rated mail by destination entry 7 profile in Table A-6 is based on the UPS' separation of the 8 pounds by destination entry profile; correct? 9 A Yes. 10 Q There, we are talking about Witness Moeller's 11 workpaper one, page 20; correct? 12 A Yes. 13 Q Witness Moeller in that workpaper or anywhere else 14 in this case, I think you will confirm, did not provide 15 piece break outs for pound rated pieces, whether they be 16 flats -- whether they be in the standard A regular or ECR 17 sub-class, he didn't provide a piece break out for pound 18 rated pieces, correct, by point of entry, BMC, SCF or DDU? 19 A Witness Moeller did not; that's correct. 20 Q Therefore, those numbers had to developed; 21 correct? 22 A Well, I don't know if they had to be developed. 23 Dr. Haldi developed them. 24 Q The way that he developed them was using the 25 percentages that appear in columns three, four and five of 19555 1 that workpaper; correct? 2 A Yes. 3 Q He multiplied the percentage that was drop shipped 4 to BMC's, SCF's and DDU's as against total pieces and put 5 those numbers into columns six, seven and eight, in essence, 6 where there are blanks; correct? 7 A Yes. 8 Q And the problem you have with that I take it is 9 that -- could you state the problem you have with that? 10 A The problem with distributing pieces for pound 11 rated non-letters based on the distribution of the pounds is 12 that it assumes that all pieces have the same weight, the 13 same average weight. 14 Q Same average weight; correct. 15 A The data available shows that's not true. 16 Q Is there any data available for test year? There 17 obviously isn't. You are talking about base year 1996 data; 18 correct? 19 A Yes. That would be the same data that Dr. Haldi 20 relied on in developing his weight for letters and piece 21 rated non-letters. 22 Q Let's get to that in a second. Let's just deal 23 with this. Do you know of another way to develop the 24 numbers of pieces for pound rated pieces, other than to use 25 the approach Dr. Haldi used, which does make the assumption 19556 1 that you stated, that the pieces have an average -- the same 2 average weight? 3 A Yes. 4 Q What's the other way? 5 A There are a number of other ways. An easy way, I 6 think, would be to assume the same type of distribution that 7 occurred in base year 1996. 8 Q If you use the distribution in base year 1996 as 9 between BMC entry, SCF entry and DDU entry, have you ever 10 tried to do that, do you know any way to be able to 11 aggregate to the numbers that appear in Witness Moeller's 12 workpaper? 13 A You could aggregate to those numbers, yes. 14 Q Have you ever tried to do that? 15 A No, I have not done it. 16 Q Do you realize that implicit in -- strike that. 17 Now this is one of your criticisms of Dr. Haldi, I 18 take it, about a theoretical assumption I guess you could 19 call it. 20 Would you call this an error about a theoretical 21 assumption? 22 A Yes. 23 Q Where does Dr. Haldi use these numbers in 24 developing rates? 25 A Are we referring to Table A-6 now? 19557 1 Q We are referring to Table A-6 and your analysis, 2 your criticism of it. 3 A Do you want me to go through and list the tables 4 that rely on the values that are in A-6? I am not sure I 5 understand the question. 6 Q Well, I am asking you is it your position that Dr. 7 Haldi uses these numbers in developing his rates for 8 pound-rated pieces? 9 A Not directly, no. 10 Q Okay, so he doesn't even use these numbers to 11 develop rates, correct? 12 A No, not directly. That's correct. 13 Q In lines 14 through 16, you say, "Comparison of 14 the implicit average weights used by Witness Haldi in Table 15 A-6 with the average weights using actual 1996 billing 16 determinants, Table A-5, indicates large disparities." 17 I would ask you if you could turn to that Table 18 A-5? 19 A Yes. 20 Q The large disparity that you point out has to do 21 with high density where there is -- you say which has an 22 average weight of .21 pounds. 23 Where do you get that number from? 24 A The .21 on line 17 should be .24. 25 Q In fact, shouldn't it be .2417? 19558 1 A I was rounding to the nearest hundredths. 2 Q Okay, so that is an error? 3 A That is a typographical error, yes. 4 Q Now you are saying that the 1996 billing 5 determinants should have been used to develop these 6 separations, correct? 7 A I am saying that that is a way of doing it. 8 Q Okay. Take a look at pound-rated nonletters for a 9 second on Table A-5 in Dr. Haldi's testimony and tell me for 10 saturation mail, for example, what systematic variation do 11 you see of average weight depending on point of entry? 12 A I don't see one. 13 Q And in fact those numbers are fairly close, aren't 14 they, for saturation -- .28, .29, .31 and .28. 15 A Excuse me, I was comparing saturation to high 16 density. Could we go back? 17 Q Sure. Let me ask you if you see any kind of 18 systematic variation among the average weight of the 19 pound-rated nonletters for the saturation tier. 20 A The range is from .28 pounds to .32 pounds. 21 Q Okay. Any systematic variation? 22 A Maybe I am having a problem with the word 23 systematic. 24 Q Well, as you get -- as you drop into the system 25 closer to the customer, does it become heavier, typically a 19559 1 heavier weight piece or a lighter weight piece? 2 A No. 3 Q It just varies randomly, correct? 4 A I don't have a basis to say that that's a random 5 variation. 6 Q You don't have an explanation for the variation 7 though, correct? 8 A Well, yes, I have an explanation. It's based on 9 the Postal Service's pounds and pieces and you divide one by 10 the other and you get -- 11 Q Sure. It's what happened -- it's what happened to 12 have happened in Fiscal '96, correct? 13 A It's actual data. 14 Q Actual data as to what happened to have happened 15 that year? 16 A Yes. 17 Q And there happens to be more density -- excuse me 18 -- more variation, I guess, in the high density, 19 specifically the number you pulled out to use, which you now 20 corrected to be .24, there seems to be more variation in 21 high density, correct, than there is in basic or saturation? 22 A The variation in high density is between .24 and 23 .33. 24 Q Correct. That's much greater than it is for basic 25 or saturation; correct? 19560 1 A I don't know that it's much greater. It's 2 different. 3 Q It's greater, right? 4 A It's greater, yes. 5 Q Do you know how much of total volume is high 6 density within ECR? 7 A The 1996 billing determinants would show that high 8 density is 287 million pieces out of 6.6 billion pieces. 9 Q A fairly small percentage compared to certainly 10 saturation and basic; correct? 11 A It's smaller; yes. 12 Q You said before that there might be other ways to 13 make the allocation that you criticized Dr. Haldi for using, 14 the percentage of pounds and applying those to volume, you 15 said there might be other ways, one you said was to use the 16 fiscal 1996 billing determinants. Anything else you can 17 think of? 18 A You could do a time series. You could test the 19 values over time. You could do a regression. You could 20 take the test year pieces and adjust them based on the test 21 year pounds but recognizing variation in weight. 22 Q Do you know if you can do any of those with the 23 information that's been provided so far in this case by the 24 Postal Service? 25 A I don't know. 19561 1 Q Have you tried to do any of those? 2 A No. 3 Q Let me ask you to look at your testimony at page 4 four, your discussion of table A-7. There you criticize Dr. 5 Haldi, I think, for -- I'll start at line nine. You say 6 "However, in developing the pounds for automation letters, 7 Witness Haldi assumed that these automation letters have the 8 same average weight as basic presort letters." Then you say 9 that assumption is false; correct? 10 A Yes. 11 Q Would you take a look at his table A-3, which is 12 1996 billing determinant data, and for automation, does the 13 Postal Service provide a break out of automation letters by 14 BMC, SCF or DDU? 15 A There is no break out in table A-3. 16 Q In fact, there is no break out as part of the 17 record in this case; correct? 18 A I'm not sure. 19 Q Do you know why there was no break out provided? 20 I'll ask you to assume there was none for the purpose of the 21 question. Do you know why automation might not have a break 22 out where the others do? 23 A No. 24 Q Are you familiar with the implementation date of 25 the rates in Docket No. MC95-1, the rate class case? 19562 1 A Yes. 2 Q And the fact that these automation ECR rates went 3 into effect during the course of fiscal 1996? 4 A Yes. 5 Q In fact, we have probably a quarter's worth of 6 data for those automation letters? 7 A I don't know. 8 Q Otherwise, those letters -- where do those letters 9 appear for the other three-quarters of the year in that 10 chart? 11 A I'm not sure. 12 Q If I were to suggest they are in the basic 13 category and in fact, I believe that is a response to an 14 interrogatory, which I'll reference for the record, which I 15 think is USPS-ValPak-Carol Wright-T1-18 and 19, and I don't 16 have a page reference. I can correct that if necessary. 17 That's what my notes indicate here. 18 Automation was a subset of basic letters for most 19 of the year in the 1996 billing determinants. Would you 20 accept that subject to check? 21 A Yes. 22 Q If that's so, if automation was in fact a subset 23 of basic, is it irrational to assume that they might have 24 the same average weight, that might be the best proxy that's 25 available to determine the average weight of automation 19563 1 pieces? 2 A It still wouldn't explain the variation in the 3 average weight for the automation letters that are shown. 4 Q Do you think it's a good proxy or a reasonable 5 proxy? 6 A No. 7 Q It's because of your numbers in table A-5? 8 A Yes. 9 Q There, the numbers are for the pieces that have no 10 destination entry; correct? Or they are put in that column 11 anyway for no destination entry? Do you see that? 12 A In table A-5? 13 Q Yes. 14 A Yes. 15 Q Do you see that the Postal Service did not provide 16 average weight in the billing determinants for letters, 17 based on point of entry? I guess they are all just under no 18 destination entry. We just talked about how in table A-3, 19 they did not provide BMC entry, SCF entry or DDU entry break 20 out; correct? 21 A Yes. 22 Q You can't tell what the -- they are all lumped 23 under no destination entry, irrespective of point of entry; 24 correct? 25 A The only place that automation letters appear is 19564 1 under no destination entry; that's correct. 2 Q Let me ask you to look at the bottom of page 4 3 where you talk about table A-8. 4 A Yes. 5 Q And there you say the -- line 17, and going to the 6 next page -- the transportation cost for SCF shown by 7 Witness Haldi equals 0.72 cents per pound and is not 8 correct. The actual value from USPS Library Reference 9 LRH-111 equals 0.73 cents per pound. Correct? 10 A Yes. 11 Q Okay. Where did you get that number from the 12 library reference? 13 A Appendix C, table 1, revised November 20, 1997. 14 Q Okay. Let me ask you if you can look at page 2 of 15 the library reference for a moment. 16 A This would be Library Reference 111? 17 Q Yes. 18 A I don't have the entirety of that library 19 reference with me. 20 Q Okay. Well, let me just read you one sentence for 21 the library reference. Actually, let me read from page 6. 22 The nontransportation cost avoidances is what we're talking 23 about. It says the cost avoidances are 2.71 cents for DDU, 24 1.99 cents for DSCF, and 1.35 cents for DBMC. 25 Based on those numbers, what would be the 19565 1 nontransportation costs for SCF? Would it not be the DSCF 2 2.71 minus the DDU 1.99? 3 A The 1.99 was revised. 4 Q To 2.0 or to 1.98? 5 A Yes. 6 Q Okay. Do you have the date of that? 7 A Appendix C, table 1, revised November 20, 1997. 8 Q Okay. That's what your criticism is then, the 9 failure to recognize the revision? 10 A Yes. 11 Q By the penny. 12 A Yes. 13 Q And you do realize that the number you cite, you 14 do have that page from Appendix C that you cited; correct? 15 A Yes. 16 Q The bottom of table 7? 17 A The page that I have doesn't have a table 7 on it. 18 I'm looking at Appendix C, table 1. 19 Q Right. Appendix C, table 7 has a number carried 20 out to more decimal points. It's .007264. Dr. Haldi used 21 .72. You used .73. Correct? 22 A Yes. 23 Q Let me ask you to turn to your analysis there of 24 table A-10. 25 You say at the beginning of line 11: In addition 19566 1 Witness Haldi's analysis assumes that shipping costs for 2 piece-rated mail, i.e., below 3.3 ounces, varies in direct 3 proportion to weight. 4 Are you familiar with the way in which Library 5 Reference 111 determines drop ship savings for Standard A 6 mail? 7 A Generally, yes. 8 Q Okay. Is it on a piece -- per-piece or a 9 per-pound basis? 10 A It's on a per-pound basis. 11 Q Okay. Every single drop ship savings is on a 12 per-pound basis; correct? In that library reference? 13 A I don't know if every single one is. Generally -- 14 the answers are expressed on a per-pound basis. The Postal 15 Service in developing their discounts for piece-rated mail 16 develops the discount based on 3.3 ounces. 17 Q And you say that the assumption has not been shown 18 to be valid. Can you explain that to me? 19 A Dr. Haldi has assumed that the destination entry 20 cost savings are linear with changes in weight, when in fact 21 it's not been shown that those changes are linear, and the 22 Postal Service doesn't apply it that way. 23 Q Do you have a problem with Library Reference 111 24 in the way they develop their drop ship savings? 25 A No. 19567 1 Q But those are purely on a pound basis, are they 2 not? 3 A Yes. 4 Q Okay. So they're directly -- in direct proportion 5 to weight, are they not? 6 A There's no savings developed on a per piece basis, 7 so there's no -- the issue of the cost savings being 8 proportional to changes in weight is not addressed in that 9 library reference. 10 Q You say that this is refuted by the data shown in 11 Witness Haldi's Appendix D. Could you turn to his Appendix 12 D? 13 A Yes. 14 Q Can you tell me what you're referencing? I guess 15 it's the footnote 4 on that page, where you say that a -- 16 Dr. Haldi's Appendix D -- that says a carrier route letter 17 weighing one ounce costs more than letters weighing two to 18 four ounces. Is that what you're referring to. 19 A Yes. 20 Q That would be in Table D3? 21 A Yes. 22 Q Okay. And so, you say that this assumption about 23 shipping costs varying in direct proportion to weight is 24 refuted by this table, correct, in Dr. Haldi's own 25 testimony. 19568 1 A Yes. 2 Q Okay. Did you read Dr. Haldi's description of 3 these tables, D1 through 3? 4 A Yes, I did. 5 Q Okay. And on page D9, do you see that Dr. Haldi 6 discusses Table 3, which is -- Table 1 and 2 were provided, 7 Table 3 he calculated, and he says inspection of Table 3 8 shows that, for carrier route mail, the results are 9 obviously absurd? Do you see that? 10 A Yes. 11 Q And so you're saying that he's inconsistent and 12 that his assumption has been proved to be false by data that 13 he presents in his testimony which he describes as obviously 14 absurd? 15 A Well, the statement's equally valid to Tables D1 16 and D2 in his appendix. 17 Q But D1 and D2 are provided by the Postal Service, 18 and it's D3 that shows that all the data are obviously 19 absurd. Isn't that correct? 20 A I'm not sure that Dr. Haldi was claiming that all 21 the data in Table D3 is absurd. If you look at the data in 22 those three tables, they are uniform in that, for carrier 23 route mail, the cost for a one-ounce piece is greater than a 24 cost for a two-ounce, three-ounce, or four-ounce piece. 25 Q Okay. Well, take a look at his analysis at the 19569 1 end of page D9. He says serious weight cannot be given to 2 data for flats when the residual produces results such as 3 these. The data for flats are in Table D2, correct? 4 A Yes. 5 Q Okay. So, certainly, at least Table D2 and D3 he 6 said were absurd and could not be given serious weight, 7 correct? 8 A Yes. 9 Q Let me ask you to turn to page 8, where you 10 discuss Table A14, and there, beginning on line 14, you say 11 Witness Haldi assumes that 2.33 cents per piece should be 12 considered weight-related for all non-letters. Is that an 13 accurate statement of what you believe Dr. Haldi does? 14 A Yes. 15 Q So, you think he is of the view that 2.33 cents 16 per piece should be considered weight-related. 17 A In purposes -- for purposes of the calculations 18 that he's made in his appendices, yes. 19 Q Okay. And what were the purposes of those 20 calculations that he made in his appendices? 21 A Well, this gets into the inputs into the rate 22 design versus the calculations that he's made. 23 When he developed his aggregate cost true-ups and 24 his tests of the distribution of the costs, he assumed a 25 weight piece of 2.33 cents. 19570 1 In his actual rate design for non-letters, it 2 never appears. 3 Q Absolutely. He did not use 2.33, correct? But 4 rather -- 5 A But he has. He uses that number in many of his 6 tables, and because my exercise here in this appendix was to 7 critique the issues and the problems with each of the 8 tables, I needed to address the 2.33 cents. 9 Q Okay. Well, take a look at Table A14 very 10 quickly, if you could, in his testimony. Do you see that he 11 describes the 2.33 cents as case one? 12 A Yes. 13 Q And case one finishes with table A-19 and then it 14 goes to case two, and at case two, a very different 15 assumption is made, of .5825 cents per piece treated as 16 weight related costs? 17 A The case two analysis is never brought forward to 18 his Appendix C, though, just the case one analysis. 19 Q You criticize him on page 12 for having assumed, I 20 guess, or you say for conducting a mathematical exercise 21 regarding this .5825 cents, and it was just a number, and 22 you said it's arbitrary and not supported by the workpapers. 23 I would just ask you, if you didn't get the drift 24 of what Dr. Haldi was doing, saying that since no one knew 25 what the weight cost relationship was, he would take two 19571 1 extreme cases and assume that it was somewhere in between, 2 and that one of the cases was 2.33 cents per piece, that's 3 case one, and case two was 0.5825 per piece and that was 4 case two, and then he compared the two, and isn't that what 5 he was doing with those two numbers rather than vouching for 6 those numbers? 7 A Well, when he went to this Appendix C to develop 8 or finish his analysis, for example, table C-8 is based on 9 the 2.33 cents per piece. There is no comparable table to 10 C-8 based on the .5825. I would assume that he's giving 11 more credibility to the 2.33, since he's using that in the 12 later steps in his analysis. 13 Q In your own testimony, on the last line of this 14 page, you say Witness Haldi's analysis of the costs 15 associated with rate as utilized in Table A-14 or subsequent 16 tables, have no bearing on the ultimate rate design for 17 pound rate mail; correct? 18 A They have no bearing on the rates that he shows in 19 his table C-10. It has a bearing on the costs that he shows 20 in his table C-9, his costs in his table C-8, those two 21 tables. 22 Q But it has no bearing on his proposed rates; 23 correct? 24 A Yes. 25 Q And that he accepted Witness Moeller's .53 pounds 19572 1 -- excuse me -- .53 cents per pound -- 53 cents a pound as 2 the pound rate; correct? 3 A Yes. 4 Q Do you have any difference of opinion with the 5 concept that there are -- I don't mean to have that long a 6 pause in the question. Just strike what I've said and let 7 me start over again. 8 Do you have a problem with Dr. Haldi's analysis 9 that we need to know more about the effect of weight on cost 10 with respect to non-letters? 11 A I've not addressed that in my testimony. 12 Q Do you believe there is a cost component of weight 13 for non-letters, that cost varies with weight? 14 A In some regard, yes. 15 Q Do you have an opinion as to whether the proper 16 weight cost relationship lies within or without the 17 parameters of his case one and case two? The 58 cents and 18 the -- 19 A I don't have a basis to say that it's within those 20 parameters or outside of those parameters. 21 Q Take a look, please, at your testimony on page 22 nine regarding A-16, table A-16. There you say at line 16, 23 for the pound portion of pound rated mail, Witness Haldi 24 assumes that costs do not vary with sortation or destination 25 entry; is that correct? 19573 1 A Yes. 2 Q Do you have A16 before you? 3 A Yes. 4 Q Okay. The e.g. you give for that proposition is 5 that the pound rated portion -- excuse me, the pound portion 6 of pound rated mail for basic sortation without drop 7 shipping is the same as another rate cell, correct? What is 8 the pound portion of pound rated mail for basic sortation 9 without drop shipping, according to the chart? 10 A For basic mail it is .257. 11 Q Then you compare that to the pound portion for 12 saturation mail drop shipped to the DDU. And what is that 13 number, from Table A16? 14 A The DDU number for saturation is .1191. 15 Q Okay. So the first number that you use there, the 16 pound portion of the pound rated mail for basic sortation 17 without drop shipping is 25.7 cents, and you say that is the 18 same as the pound portion for saturation mail drop shipped 19 to the DDU, which you said is 11.91 cents. 20 A I think the statement -- my statement in the text 21 is not very clear. The point that I was trying to make in 22 that statement was if you are in Table A16, you see two 23 things. First, that for pound rated mail, the per pound 24 amount is the same regardless of sortation. In other words, 25 basic equals high density equals saturation. 19574 1 The second thought that I was trying to get across 2 here was that for the piece portion of pound rated mail, the 3 values are the same across all -- regardless of destination 4 entry. So that if you are, and this is one section up on 5 Table A16, if you are basic mail, the assumption is that the 6 cost is 5.90 cents for no destination entry but it also 5.90 7 cents for destination entry at the DDU. 8 Q And isn't it correct that Dr. Haldi has been 9 critical of the Postal Service for failure to develop these 10 weight cost relationships any quicker than they have -- than 11 they have done so, and that he is attempting to further the 12 analysis? 13 A Well, Dr. Haldi stated the purpose of his 14 testimony was to develop bottom up costs. 15 Q And with respect to weight, he notes serious 16 deficiencies in the available data, does he not? Isn't that 17 what Appendix D is about? 18 A His Appendix D discusses deficiencies in the data 19 related to weight and cost. 20 Q The entire topic of the section, isn't it? 21 A Yes, it is. 22 Q Okay. Well, so just to clarify, let me just make 23 sure, you are not contending, are you, that the first 24 number, 25.7 cents, is the same as 11.91 cents? 25 A No, I not. 19575 1 Q So this is an error? 2 A Well, the statement is not very articulate. 3 Q Well, isn't it just in error? Just wrong? 4 A Well, yes. 5 Q Let me ask you to turn to page 12, and there you 6 talk about the 90 percent margin, 10 percent markup that Dr. 7 Haldi uses in his recommended rates, do you recall that? 8 A Yes. 9 Q Okay. You first say on line 16 that Witness Haldi 10 only relies on the results of BMC mail in his rate proposal. 11 In other words, you are being -- you are criticizing him, I 12 take it, that he didn't develop a separate analysis for SCF 13 and DDU and no destination entry, correct? 14 A Separate analysis meaning rates, yes. His rates 15 for the other destination entry are based on the Postal 16 Service's discounts. 17 Q Right. Have you attempted to do the analysis for 18 SCF, DDU, no destination entry and compare it to the way 19 that it comes out with Dr. Haldi's analysis using BMC? 20 A No. 21 Q Okay. Secondly, you say the 90-10 allocation is 22 arbitrary and not supported. Does that mean that you don't 23 believe it is supported in his testimony? 24 A It appears in his testimony. There's no rationale 25 for the 90 percent/10 percent split versus any other split. 19576 1 Q Okay. Let me ask you to turn to page 40 of Dr. 2 Haldi's testimony. 3 A Okay. 4 Q On page 40, Dr. Haldi starts an analysis of target 5 margins versus target mark-ups. You have read this, 6 correct? 7 A Yes. 8 Q Okay. And he first says, and I am reading from 9 line 13 here, "The extent to which either of these two 10 approaches, margins or mark-ups, is more appropriate for any 11 given class or subclass depends on the competitive 12 environment for postal services as explained below." 13 Do you have any problem with that statement? 14 A Yes. 15 Q You have a problem with it? 16 A Yes. I don't believe that either of these two 17 approaches is appropriate from the standpoint of his 18 analysis. 19 Q Okay. Well, first, let's go through his analysis 20 to make sure we are clear on that and then I'll ask you 21 that. 22 First he discussed target margin at the bottom of 23 page 40, going on to 41, and he talks about assumptions 24 which underlie a pricing strategy of relying on margin are 25 when the Postal Service faces perfect or near-perfect 19577 1 competition for sortation and shipping but has a perfect or 2 near-perfect monopoly of the delivery function. 3 Do you see that? 4 A No. 5 Q Lines 9 through 14 on page 41. 6 A Yes. 7 Q Okay, and then when he discusses mark-up at the 8 bottom of 41 and on to 42, he talks about how that would -- 9 and I am referring now to line 8 on page 42 -- how that 10 would implicitly treat all postal services being subject to 11 fairer competition. 12 Do you see that? 13 A I see the statement, yes. 14 Q Okay. Do you have any problem with his analysis 15 there? 16 A Yes. 17 Q Okay, well, we will get to that. Then on page 43 18 he does an assessment of the competitive environment, which 19 according to his way of looking at it is very significant in 20 determining whether these are mark-ups or margins, and he 21 does an assessment or the competitive environment for 22 delivery -- at the bottom of page 43, correct? 23 A Yes. 24 Q It goes on to 44 and then he goes into the 25 analysis of the competitive environment for sorting and 19578 1 shipping, correct? 2 A Yes. 3 Q Okay, and then he goes on to 46 and he leaves the 4 issue and then he goes to page 49 where he says -- he picks 5 up this discussion and he says, line 15, "First initial 6 target rates were derived by adding to unit costs a constant 7 mark-up of 7.379 plus a mark-up of 10 percent. The 10 8 percent mark-up is a conservative recognition of the fact 9 that the Postal Service faces competition from alternative 10 delivery in a number of markets." 11 Is it fair to say that Dr. Haldi did not provide 12 support for the 90-10 allocation, as you have just said? 13 A It appears in his text but there is no 14 quantitative analysis here that says when you look at 15 competitive positions for advertising mail that you should 16 use 90 percent on a fixed margin and 10 percent on a 17 mark-up. 18 His analysis that you have pointed to in this 19 testimony doesn't have any quantitative support for a 90-10. 20 Q If the number were lower, if it were 80-20, it 21 would result in higher basic rates, correct? 22 Anything lower than 90 would result in higher 23 basic rates of the sort used by catalogs? 24 A Do you mean basic rates or initial target rates? 25 Q Basic rates. 19579 1 A I don't know because of the impact of the true-up 2 that he uses in Table C-3. 3 Q Okay, but if -- I guess I will do the rest of this 4 in brief, but if -- the only question I want to ask you is 5 isn't it true that when you move down from a 100 percent to 6 90 percent you put at least some of the mark-up on 7 processing and transportation and not all of it on delivery? 8 Is that one way to look at it? 9 A I don't think so. 10 Q Is the -- when you move from 100 to 90 is it true 11 that you are favoring the classes of mail that have higher 12 unit cost by giving them less of a mark-up? 13 A Well, let me answer that by using his table. 14 In his Table C-2, his part A of C-2 reflects a 15 constant mark-up of 8.19 cents. Part B of his Table C-2 16 uses a constant percentage mark-up of -- a ratio of 2.44. 17 He then weights Part A with Part B to get his 18 rates. 19 Now, within the framework of those values, if you 20 assume that the values he's got in that table are correct, 21 as you decrease the constant piece and increase the ratio 22 piece, Part C of his Table C-2, the basic rate, would 23 increase. 24 MR. OLSON: Mr. Chairman, that's all we have. 25 Thank you. 19580 1 CHAIRMAN GLEIMAN: Is there any followup? 2 Questions from the bench? 3 If not, that brings us to redirect. 4 Mr. Todd, would you like some time with your 5 witness? 6 MR. TODD: Mr. Chairman, please, yes. 7 CHAIRMAN GLEIMAN: I think we'll take ten then, 8 and make it a break. 9 [Recess.] 10 CHAIRMAN GLEIMAN: Yes, sir. Mr. Todd, did you 11 have any redirect? 12 REDIRECT EXAMINATION 13 BY MR. TODD: 14 Q Mr. Prescott, are you ready? 15 A Yes. 16 Q Could you state what your understanding is of what 17 Dr. Haldi was attempting to do in the testimony which is the 18 subject of your rebuttal testimony? 19 A The purpose of Dr. Haldi's testimony was to 20 develop what he called bottom-up costs, to look at the 21 Postal Service's rate design and to propose a rate design 22 which in his opinion followed the logic of bottom-up costs. 23 Q Do you believe that he was able to develop rates 24 based upon bottom-up costing in his testimony, valid 25 bottom-up costing in his testimony? 19581 1 A No. 2 Q And why do you think that he failed in his 3 attempt? 4 A The data isn't available for him to do the 5 bottom-up cost analysis that he wanted to do and from there 6 to develop rates based on bottom-up costs. And because of 7 that he had to make a series of assumptions, and ultimately 8 when he went to his rate design he utilized the Postal 9 Service's rate design except for a modification to the 10 sortation discounts and the discount for the base rate -- 11 excuse me, and the rates for base rates. 12 Q Thank you, Mr. Prescott. 13 MR. TODD: That's all I have, Mr. Chairman. 14 CHAIRMAN GLEIMAN: Is there any recross? 15 Mr. Olson? 16 RECROSS-EXAMINATION 17 BY MR. OLSON: 18 Q Mr. Prescott, you say that the data isn't 19 available to do the kind of bottom-up analysis Dr. Haldi set 20 out to do; correct? 21 A Yes. 22 Q Are all of the data insufficiencies that you know 23 of set out in your testimony? In other words, you're not 24 referring to something new now that isn't already in your 25 rebuttal testimony; correct? 19582 1 A That's correct. 2 Q All data insufficiencies of which you are aware 3 are in your rebuttal testimony; correct? 4 A Yes. 5 MR. OLSON: That's all I have, Mr. Chairman. 6 CHAIRMAN GLEIMAN: Any further redirect? 7 MR. TODD: No. 8 CHAIRMAN GLEIMAN: If that is the case, we want to 9 thank you, Mr. Prescott. We appreciate your appearance here 10 today and your contributions to our record, and if there's 11 nothing further, you're excused, but your counsel isn't. He 12 has to stay around for a while. 13 [Witness excused.] 14 CHAIRMAN GLEIMAN: I take it you do have another 15 witness. Right? 16 MR. TODD: Yes, and it would be helpful, Mr. 17 Chairman, if there could be some indication of how long the 18 next witness is likely to take. I'm not sure who that is 19 offhand, but -- 20 CHAIRMAN GLEIMAN: Well, I'm about to -- 21 MR. TODD: You're going to do that. 22 CHAIRMAN GLEIMAN: Apprise you, and then we'll all 23 find out. 24 Our next witness is appearing on behalf of the 25 United States Postal Service. Ms. Schenk is already under 19583 1 oath in the proceeding. Ms. Reynolds is counsel. 2 MS. REYNOLDS: Yes, the Postal Service is calling 3 its next witness, Leslie M. Schenk. 4 Whereupon, 5 LESLIE M. SCHENK, 6 a witness, was called for examination by counsel for the 7 United States Postal Service and, having been previously 8 duly sworn, was examined and testified as follows: 9 CHAIRMAN GLEIMAN: If you could proceed to enter 10 Witness Schenk's rebuttal testimony, and while you're on 11 your way over there -- well, let's wait a moment before we 12 find out from Mr. Levy. Why don't you proceed, Ms. 13 Reynolds. 14 DIRECT EXAMINATION 15 BY MS. REYNOLDS: 16 Q All right. Dr. Schenk, I've handed you two copies 17 of a document entitled the Rebuttal Testimony of Leslie M. 18 Schenk on Behalf of the United States Postal Service, and 19 designated USPS-RT-22. 20 Are you familiar with these documents? 21 A Yes, I am. 22 Q Were they prepared by you or under your direction? 23 A Yes. 24 Q And if you were to testify orally here today, 25 would this be your testimony? 19584 1 A Yes, it would. 2 Q Do you have any changes you wish to make to these 3 documents? 4 A Yes. There are a couple of errata to announce 5 from the filed version. On page -- and these corrections 6 are made in the version that we're presenting today. 7 On page 3, line 3, it should read 0.228 billion 8 for FY '96, not million. 9 On page 12, line 2, the phrase in parentheses 10 should be deleted. 11 And then in Appendix B, Exhibit 2, on page 29, we 12 are filing the testimony with a revised version of that 13 exhibit. The difference between the old version and the new 14 version is that we received three additional responses after 15 the filing of the testimony. Those responses are included 16 in the new version of the exhibit, but the inclusion of 17 those responses does not change the magnitude of the 18 estimate derived substantially, so -- and there was also one 19 correction to one of the formulas. It was just a 20 typographical error. So those are included in this revised 21 version of the exhibit. 22 MR. LEVY: May I inquire, Mr. Chairman, of the 23 witness which lines of the table are changed? 24 THE WITNESS: The lines would be lines 15, line 25 19, and -- I can't find the other one here right away. 19585 1 MR. LEVY: 244 and 249, perhaps? 2 THE WITNESS: Oh -- and 20. Oh -- and the 3 typographical error in the formulas would be the formulas 4 for lines 5 and 6. 5 BY MS. REYNOLDS: 6 Q That's inflation factor strata 1 and inflation 7 factor strata 2. 8 A Yes. 9 MS. REYNOLDS: At this time, Mr. Chairman, I've 10 got two copies of the testimony for the reporter, and we ask 11 that they be moved into evidence. The revisions are marked 12 in the documents that the reporter is receiving. 13 CHAIRMAN GLEIMAN: Almost afraid to ask. Are 14 there any objections? 15 MR. LEVY: Other than the pending motion, no, Mr. 16 Chairman. 17 CHAIRMAN GLEIMAN: Okay. Recognizing that there 18 is a pending motion and it does raise an objection, I'm 19 going to direct, for the moment, that the testimony and 20 exhibits of Witness Schenk be received into evidence, and I 21 direct that they be transcribed into the record at this 22 point. 23 [Rebuttal Testimony and Exhibits of 24 Leslie M. Schenk, USPS-RT-22, was 25 received into evidence and 19586 1 transcribed into the record.] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19636 1 CHAIRMAN GLEIMAN: I have some questions I need to 2 ask Witness Schenk before we proceed. 3 Witness Schenk, on page 25, at lines 4 and 5 of 4 your testimony, you describe disqualification logs 5 maintained by acceptance units. You then describe a survey 6 undertaken by LRCA which you present in your testimony. 7 THE WITNESS: Yes. 8 CHAIRMAN GLEIMAN: I've looked at the survey 9 forms. I have the Library Reference H353 here. These forms 10 reflect information provided by Postal Service employees at 11 the sample -- the 30 samples offices. Is that correct? 12 THE WITNESS: That's correct, the sites that 13 responded to our survey. 14 CHAIRMAN GLEIMAN: Is that number now 33? 15 THE WITNESS: That number is now 27 of the 16 original 30. Exhibit 2 in my appendix clearly indicates 17 that there are some sites that we have not received 18 responses from yet. 19 CHAIRMAN GLEIMAN: You'll have to bear with me. 20 THE WITNESS: Uh-huh. 21 CHAIRMAN GLEIMAN: Now, I couldn't determine very 22 easily from the library reference whether Postal Employees 23 who were providing information on the possibility that mail 24 with non-profit standard indicia paid standard A regular 25 rates were actually taking information from Form 8075. Did 19637 1 you ask the employees at each office whether they were using 2 actual Form 8075 data for 1996? 3 THE WITNESS: They indicated whether those forms 4 were available, whether they had discarded them yet, or 5 whether they were able to obtain them from storage if they 6 had not been discarded yet. They indicated whether they had 7 them or not and were able to use them to give us -- to 8 provide us that information. 9 CHAIRMAN GLEIMAN: I understand that you ask all 10 those but questions, but my questions is, did you ask the 11 employees at each office whether they were using actual Form 12 8075 data for 1996? 13 THE WITNESS: Yes. 14 CHAIRMAN GLEIMAN: Did you note on the survey 15 forms contained in the Library Reference H-35 -- did you 16 note that on the survey forms contained in the Library 17 Reference? And, if so, could you show me where the 18 notations of that fact are made? 19 THE WITNESS: I don't recall if we noted that 20 explicitly, but the information on the survey forms would 21 indicate that. If you would bear with me for a minute, I'll 22 find those responses. I believe there were only two sites 23 that were able to obtain the FY '96 logs and use them to 24 provide the information. 25 The first site would the Survey Numbered 1 in the 19638 1 Library Reference, and what indicates that they actually did 2 look at the disqualification -- or the acceptance logs, was 3 that we received a list of the mailings that were sent 4 regular rate but with nonprofit indicia, we received a list 5 of those from the survey -- from the site themselves and 6 that list is included in the Library Reference at the end of 7 that Survey No. 1. 8 CHAIRMAN GLEIMAN: Do I understand you correctly 9 that two sites actually used the data from Forms 8075 that 10 they had in hand? 11 THE WITNESS: Yes. 12 CHAIRMAN GLEIMAN: And the others did not? 13 THE WITNESS: Yes. 14 CHAIRMAN GLEIMAN: Okay. 15 THE WITNESS: The other one would have been site 16 No. 4. 17 CHAIRMAN GLEIMAN: Did you inquire whether Form 18 8075s were still in existence at each of the offices, the 19 other offices? 20 THE WITNESS: Yes. And their response would have 21 been indicated on question No. 3 of the survey form. 22 CHAIRMAN GLEIMAN: And can you tell me how many of 23 the offices who did not provide information from 8075s told 24 you those forms were still in existence? 25 THE WITNESS: I don't recall offhand, I did not 19639 1 quantify that. But it would be in question No. 3. 2 CHAIRMAN GLEIMAN: Okay. Did you ever ask Postal 3 Service employees to go get Form 8075 before responding to 4 the survey questions? 5 THE WITNESS: We did attempt to -- we did ask them 6 that if they knew that they existed, to try to find them and 7 to get the information from them. But since they were from 8 FY '96, this is quite a number of years ago, the sites 9 indicated that these were in storage and they -- it would 10 take some time to find them. In fact, one of the sites that 11 did -- both sites that found them, in took several days in 12 order to find the boxes that had the forms in them. 13 The other sites could not spare the personnel to 14 try to find them in the time frame that we needed to provide 15 them for -- or get that information for rebuttal testimony. 16 CHAIRMAN GLEIMAN: Page 29 of your testimony, you 17 list in the table 30 offices in your survey. It's the table 18 that you have just offered us in revised form. 19 THE WITNESS: Yes. 20 CHAIRMAN GLEIMAN: Would you please identify which 21 of those entries reflect the actual? As I understand it, 22 it's No. 1 and No. 4? 23 THE WITNESS: It was No. 1 and No. 4 in the -- the 24 way they are listed in the table, it would have been No. 10 25 in strata 1 and No. 52 in strata 2. 19640 1 CHAIRMAN GLEIMAN: Okay. Did you give 2 instructions to any Postal employees who provided 3 information as to how they should treat any of the notes or 4 forms they may have used to develop the information for you? 5 Or to put it another way, did you ask those people who had 6 access to Form 8075 to keep those forms available? 7 THE WITNESS: The only sites that were able to 8 obtain them in time to provide the information for rebuttal 9 testimony were the two sites that I have already listed. I 10 did not ask them to retain those forms nearby. 11 CHAIRMAN GLEIMAN: I have got to go back to one of 12 my earlier questions now, and make sure I understand. I 13 asked you earlier about whether you made an inquiry about 14 Form 8075s and whether they were still in existence at each 15 office. 16 THE WITNESS: Yes. 17 CHAIRMAN GLEIMAN: And you told me that that is in 18 question -- I could find that information in question No. 3. 19 THE WITNESS: Yes. 20 CHAIRMAN GLEIMAN: We are going to take a break in 21 a moment, and I am going to ask you to go through the 22 Library Reference and tell me how many offices indicated 23 that the 8075s still existed, separate and apart from the 24 two who we have established actually used that form to 25 provide data for you. Okay. 19641 1 Before we take the break, though, in the interest 2 of accommodating others, Mr. Levy, assuming for the sake of 3 discussion that we proceed on the basis of the testimony 4 that is in the record now, can you give me a worst case 5 scenario on how long you might cross-examine? 6 All right. No one else has been -- 7 MR. LEVY: Two hours. 8 CHAIRMAN GLEIMAN: I beg your pardon? 9 MR. LEVY: Two hours. 10 CHAIRMAN GLEIMAN: Two hours. Okay. That is an 11 outside worst case scenario, but it could be shorter. 12 MR. LEVY: It could be shorter. It is a lawyer's 13 worst case scenario. 14 CHAIRMAN GLEIMAN: All right. At this point we 15 are going to take a 10 minute break and Witness Schenk, 16 during the break, I would like you to go over the Library 17 Reference and tell me how many question No. 3 answers 18 indicate the continued existence of Form 8075. 19 [Recess.] 20 CHAIRMAN GLEIMAN: When last we met and before we 21 went into shock at the prospect of two hours of cross 22 examination by Mr. Levy and crew, I had asked you to count 23 up question number three responses from your library 24 reference to let me know how many other of the offices you 25 surveyed said they actually still had 8075's. 19642 1 THE WITNESS: Of the 27 responses we have received 2 so far, there were 11 responses that had a "yes" to the 3 question as to whether the FY'96 logs were available, but I 4 want to clarify what that means. 5 This means that they have not destroyed those 6 logs. It does not mean that they had them available on 7 hand. These often were -- I noticed as I was going through 8 it, we have written it clearly in some of the responses that 9 these are logs that are kept in boxes in storage, and it may 10 be difficult to find. 11 Like I said for one of the sites where they were 12 able to find those logs, it took several days and many hours 13 of a supervisor's time in finding those logs. 14 A "yes" response to that question means they 15 haven't been destroyed. It doesn't necessarily mean they 16 are readily available for us to look at and may take days to 17 look -- 18 CHAIRMAN GLEIMAN: I understand you. We are not 19 dealing with -- you surveyed 30 sites. 20 THE WITNESS: Right. 21 CHAIRMAN GLEIMAN: You had 27 responses? 22 THE WITNESS: Uh-huh. 23 CHAIRMAN GLEIMAN: We have two who for sure have 24 8075's in hand because they use them? 25 THE WITNESS: Yes. 19643 1 CHAIRMAN GLEIMAN: We have 11 who think they have 2 them in a box somewhere? 3 THE WITNESS: Right, and there were an additional 4 four that said maybe. The other thing I would like to 5 clarify is you characterized this information as data. I 6 wanted to clarify that on the acceptance logs, they do not 7 necessarily have listed what the volumes or even the revenue 8 deficiencies related to these entries are. That information 9 is kept in files for particular mailers, and getting that 10 information would take even longer, a much longer time to 11 get. 12 CHAIRMAN GLEIMAN: Well, I have three choices. I 13 can rule to compel disclosure of what we know exists for 14 sure and ask that you, "you" the Postal Service, make a 15 concerted effort on the probables and the maybes, 11 and 4, 16 and in the alternative, we can strike -- the problem I'm 17 faced with is I never like to strike anything. I think it's 18 always better to have a fuller record than not. 19 We are starring at a mid-May delivery date for a 20 recommended decision. I am going to leave it up to the 21 Postal Service. Counsel, you have a choice. You, the 22 Postal Service, have a choice. 23 Since only a limited number of forms were actually 24 used and we know we can get two of them, and 11 more are 25 maybe in boxes and four, we are not even sure they are in 19644 1 boxes, you can have a week to get the forms. That is I'll 2 compel disclosure with respect to that 13, 17 forms, with 3 the understanding that four of those may not exist, and 4 direct that a concerted effort be made and that a response 5 to my ruling be provided by close of business next Thursday, 6 or in the alternative, we will strike Witness Schenk's 7 survey. 8 MS. REYNOLDS: Could I clarify? 9 CHAIRMAN GLEIMAN: You bet. 10 MS. REYNOLDS: If it turns out that one of these 11 sets of forms, which may be in a box in deep storage 12 somewhere, turn out to be indeed irretrievable, how would 13 the Chairman like to handle that? 14 CHAIRMAN GLEIMAN: You find what you can find 15 between now and next Thursday. We know that you should be 16 able to find two of them. At that point, we will have what 17 I hope would be a relatively short hearing, at which Ms. 18 Schenk would reappear and ANM would have an opportunity to 19 cross examine on the materials that were uncovered pursuant 20 to the ruling in favor of the motion to compel, and if there 21 are only two forms, then the Commission will give 22 appropriate weight to Ms. Schenk's survey, based on where we 23 know the data came from for sure versus where it may have 24 come from, may it have been hidden in a box. 25 That's the best I can do. I'll give you about 19645 1 five minutes to talk it over with your team, and the choices 2 are come up with the forms that you can, two of them for 3 sure, and whatever else out of that 11 plus 4 by next 4 Thursday, close of business, or we will strike. 5 Five minutes. 6 MS. REYNOLDS: Mr. Chairman, another quick 7 clarification? 8 CHAIRMAN GLEIMAN: You bet. 9 MS. REYNOLDS: You prefaced your comments by 10 saying we had three options. So far, I only heard to. 11 CHAIRMAN GLEIMAN: No, I said we had three 12 options. My option is if we have a motion to compel and you 13 don't respond, then you'd be in violation of the lawful 14 order and I'd ask my fellow Commissioners in joining me in 15 issuing a C(2) order. 16 MS. REYNOLDS: Good enough. 17 [Recess.] 18 CHAIRMAN GLEIMAN: Just let me say that I didn't 19 mean to be cavalier throwing around reference to a C(2) 20 order, because none of us want anything like that to come to 21 pass. 22 We know from yesterday's hearing that the Postal 23 Service is in dire financial straits, and it's getting worse 24 every minute, and we wouldn't want to do anything to 25 endanger their situation, but I just wanted to clarify for 19646 1 the record that that's, you know, a last resort, far, far 2 out, and hopefully on a horizon that none of us ever have to 3 come to. 4 Ms. Reynolds, you have a decision to tell me 5 about. 6 MS. REYNOLDS: Yes, we do. We are going to take 7 your option whereby we will make every effort to obtain the 8 information that you're looking for by next week and recall 9 Dr. Schenk to respond to questions on it at that time. 10 CHAIRMAN GLEIMAN: Okay. Let's just make sure we 11 understand one another. 12 We know we're going to get two forms, or at least 13 we think we know we're going to get two forms, and we may 14 get a number more, and we're -- they're going to be 15 submitted by close of business next Thursday, and we will be 16 in touch with both the Postal Service and counsel for ANM 17 about scheduling what hopefully will be a, relatively 18 speaking, short hearing to allow oral cross examination on 19 the material that is provided. 20 MS. REYNOLDS: I have a procedural question. 21 Regarding the filing of these documents, they are 22 likely to be considerably voluminous, and I am wondering if 23 we might waive the Commission's filing requirement whereby 24 we are required to file 30 copies of them? 25 CHAIRMAN GLEIMAN: I believe we can do that. 19647 1 MS. REYNOLDS: What would be an appropriate number 2 of copies? 3 CHAIRMAN GLEIMAN: Well, I think that, if you can 4 provide a copy to ANM, which has a seemingly deep-seated 5 interest in these documents, and two additional copies, that 6 would probably suffice for all of our purposes, and I 7 suspect that the volume of the documents will depend upon 8 the success of the search through all those boxes out there. 9 So, we'll see what we get. Okay. 10 Now, let me ask you a question, Mr. Levy. Would 11 you like to reassess the probable length of your cross 12 examination today? 13 MR. LEVY: Yes, Mr. Chairman. 14 CHAIRMAN GLEIMAN: Again, this is not to put 15 pressure on you. It's to help some others who have been 16 sitting around and are going to have to be here late 17 tonight. 18 MR. LEVY: My answer will depend on a question I 19 would like to pose to you and you may not be in a position 20 to answer, which is, for the sites in the list of 27 that 21 don't have these disqualification logs, is the chair -- is 22 it the chair's intention to disregard those sites? I assume 23 the Commission probably doesn't know at this point. 24 CHAIRMAN GLEIMAN: I will tell you what I think I 25 said earlier, when I was giving Ms. Reynolds the options and 19648 1 some clarifications, and that is, if she chose the option 2 that she did choose on behalf of the Postal Service, that we 3 would await the material and we would determine the 4 appropriate weight to give it based on what came in. 5 MR. LEVY: In that event, then I'm afraid the 6 ruling doesn't reduce my estimate very much, because I think 7 I need to protect my position, to ask about sites where it 8 appears that the witness may argue that she is relying on 9 independent judgement of local field personnel. 10 CHAIRMAN GLEIMAN: Well, it will be difficult, I 11 think, to parse out at this point -- I'm sure that it's 12 possible, but it might be difficult to parse out at this 13 point which parts of the survey we're going to get a 14 response to and which parts of the survey we're not going to 15 get a response to, and I would respectfully suggest that we 16 could reserve your rights and you could question on the 17 survey in toto when we receive a response to the order to -- 18 in response to your motion to compel. 19 MR. LEVY: If I can defer questions about the 20 survey methodology, that would cut it way down. I could 21 probably do about half-an-hour. 22 CHAIRMAN GLEIMAN: It's not that we want to get 23 out of here early tonight, but I think that that's a prudent 24 course. The survey is a package, and conceivably, questions 25 could occur to you or responses may become -- information 19649 1 become -- may become evident to you once you see the 2 material that's produced or not produced. 3 So, do we understand and agree that the survey 4 --all aspects of questioning on the survey, in essence, will 5 be reserved for that hearing that we're going to have 6 sometime shortly after next Thursday? 7 My guess is we're looking probably at March the 8 30th, would be a good guess, because that would give you an 9 opportunity to review the material that came back in. I 10 believe that's a Monday. 11 But we're not firm on that. We'll talk with the 12 parties to make sure that everyone can -- 13 MR. LEVY: Then, if the questions about the survey 14 can be deferred until then, then my time estimate goes down 15 to 15 to 30 minutes, because I will be asking only about the 16 witness' criticism of Dr. Haldi's survey. 17 CHAIRMAN GLEIMAN: Well, we'll proceed on that 18 basis, and you can begin your cross examination. 19 MR. LEVY: If I may have a moment to reorganize my 20 notes. 21 CHAIRMAN GLEIMAN: Whenever you're ready. 22 CROSS-EXAMINATION 23 BY MR. LEVY: 24 Q Good evening, Dr. Schenk. As you know, I'm David 25 Levy for the Alliance of Non-Profit Mailers. 19650 1 Would you turn to page 5 of your testimony? 2 A Yes. 3 Q Now, on lines 10 to 11, you have the following 4 statement. Quote, "In addition, the ANM survey responses do 5 not indicate how the volumes for these mailings were 6 recorded in Postal Service's databases." Do you see that? 7 A Yes. 8 Q How would the ANM survey respondents be in a 9 position to know the answer to such a question? 10 A I did not imply that the ANM survey respondents 11 should know, and that is exactly my point is that that is 12 the inference that Dr. Haldi was making in his analysis was 13 that the volumes and cost data were not consistent with one 14 another but the survey responses do not in any way show how 15 the volumes were recorded. 16 Q Well, how is ANM at the time we filed Dr. Haldi's 17 testimony to indicate how the volumes for these mailings 18 were recorded in Postal Service databases? 19 A I'm sorry, I didn't catch your question. 20 Q How was the Alliance of Nonprofit Mailers at the 21 time we filed Dr. Haldi's testimony supposed to find out how 22 the volumes for these mailings were recorded in Postal 23 Service's databases? 24 A I don't know, but Dr. Haldi makes inferences about 25 how those volumes are recorded in the Postal Service 19651 1 databases without that information, and that is one of my 2 criticisms of his analysis. 3 Q So he shouldn't have filed his study at that 4 point? 5 A I believe he does not have the information 6 available to make his conclusions. 7 Q Are you aware that at the time we filed his 8 testimony we had an outstanding request to the Postal 9 Service for that information? 10 A Yes, I am aware of that request. 11 Q And the Postal Service's response at that time was 12 they didn't know? 13 A I don't believe that that was -- my understanding 14 of the Postal Service response was that the information 15 would take a lot of time to get if it were available at all. 16 I am not sure I am characterizing that correctly, 17 but that was my understanding of the Postal Service 18 response, and also I believe part of that response had to do 19 with the timing of the filing for that request. 20 Q In your testimony you attach a document that was 21 previously filed as an interrogatory answer? Is that 22 correct? 23 A Yes, that's correct. 24 Q And that appears starting on page 30 of your 25 rebuttal testimony, is that correct? 19652 1 A Yes. 2 Q That is the Postal Service's response to 3 ANM/USPS-28, is that correct? 4 A Yes. 5 Q And that was prepared by you? 6 A Yes, it was prepared by me and my colleagues. 7 Q And that was filed after the filing date for the 8 Intervenors' testimony, wasn't it? 9 A Yes. 10 Q Would you turn to page 8 of your rebuttal 11 testimony? 12 A I'm there. 13 Q Starting on line 15, you state, "The survey 14 responses provided in ANM-LR-1 do indicate that at least 15 one-third of survey responses were received from members of 16 the American Association of Museums. It is highly unlikely 17 that one-third of all nonprofit Standard A volumes are 18 associated with this group." 19 Have you seen any information in the course of 20 preparing your testimony that museums have a higher 21 incidence of nonprofit mail rejection than other nonprofit 22 mailers on average? 23 A No, I have not seen any information of that sort. 24 Q Would you turn to page 9 -- 25 A And I do want to clarify that my sentence in that 19653 1 section was referring to the way that the sample was drawn 2 and whether the sample was representative of nonprofit 3 mailers in general, and that is what my statement was 4 referring to. 5 Q I understand that, but an unrepresentative sample 6 can give correct results if the different populations being 7 surveyed -- if the sample population doesn't have traits 8 that are significantly different from the rest of the 9 universe. Isn't that correct? 10 A But there was no analysis in either the ANM survey 11 responses or in Dr. Haldi's analysis that showed whether the 12 respondents were representative of the universe or whether 13 the nonrespondents were -- had similar or different 14 characteristics to the respondents. And that is generally 15 accepted methodology in survey methodologies to show that, 16 especially with such a high nonresponse rate. 17 Q Dr. Schenk, you answered my question with the word 18 "but." Could you first answer my question, which was, if 19 there is not a significant difference between the sample 20 population in its traits and the traits of the universe as a 21 whole, then even an unrepresentative sample can produce 22 accurate results. Isn't that correct? If. 23 A Could you restate that again? I just want to make 24 sure that I'm understanding your question. 25 Q Yes. If hypothetically American museums 19654 1 experience disqualification of mail tendered at nonprofit 2 rates at the same frequency as the rest of the universe of 3 nonprofit mail, then overweighting of museums in the sample 4 shouldn't distort the results. Isn't that correct? 5 I'm not asking you -- 6 A I believe if -- I believe given your hypothetical 7 that that would be correct. 8 Q And then you added: but we have offered no 9 evidence that the hypothetical is correct. 10 And my question to you is: Have you offered any 11 evidence that the hypothetical -- that the population -- 12 that American museums experience mail rejection at a greater 13 rate than the average nonprofit mailer? You being 14 Christensen Associates or the Postal Service. 15 A I have no evidence to that, but also there has 16 been presented no evidence that they are representative, and 17 that is standard procedure in a survey like this to show 18 that the respondents are representative when there's a high 19 degree of nonresponse. 20 Q Representative of the universe. 21 A Of the universe. 22 Q Do you know whether any other organization or 23 entity in the world besides the Postal Service knows who the 24 whole universe of its nonprofit mailers are? 25 A There are a number of people who study nonprofit 19655 1 organizations, and in fact I spent a year as I was a 2 graduate student as a research assistant studying nonprofit 3 organizations under a professor who studies these. So yes, 4 there are people who know about the population of nonprofit 5 organizations. 6 Q And those people know how much mail each nonprofit 7 organization enters at nonprofit rates in fiscal year 1996? 8 A I do not know if anyone has studied that issue. 9 Q In fact, isn't it illegal for the Postal Service 10 to disclose the volumes of individual mailers to the public? 11 A I'm not sure about the legality. Generally we do 12 not provide that information in our studies to maintain 13 confidentiality. 14 Q Would you turn to page 10 of your testimony, 15 starting at the very first line? 16 There you state that for 26 of the survey 17 responses the data recorded in the exhibit ANM-T1-1 do not 18 match the answers provided in the survey forms provided in 19 ANM-LR-1. Do you see that? 20 A Yes, I do. 21 Q Now do you know whether the data recorded in the 22 Exhibit ANM-T1-1 includes information that was recorded 23 directly into a computer data base or spreadsheet rather 24 than on an intermediate basis onto the survey forms? 25 A What I'm referring to in that paragraph is the 19656 1 fact that some of the responses were misinterpreted, not 2 that they were recorded incorrectly in terms of data entry, 3 but that they were misinterpreted. And that, as I explain 4 later in that paragraph, that it was interpreted that if the 5 mail was sent with a nonprofit permit, question 2(b), then 6 it was assumed that it had nonprofit indicia. But in fact 7 in question 2(c) they asked specifically for what indicia 8 the pieces were sent at. And for these 26 survey responses, 9 I noted that those volumes were indeed according to question 10 2(c) sent with regular rate indicia. So there would be no 11 discrepancy between the volumes recorded and the indicia on 12 the piece. 13 Q I'm sorry, could you repeat the last sentence of 14 your answer. I didn't follow that. 15 A According -- for these 26 survey responses, the 16 volumes that were noted in the original exhibit, ANM-T1-1, 17 indicated that these pieces were sent at regular rate with 18 nonprofit indicia, but in fact question 2(c) indicated that 19 they had nonprofit indicia -- or, I'm sorry, regular rate 20 indicia on them, and therefore those pieces would have been 21 recorded as regular rate volumes, and also if they were 22 sampled in IOCS would have been recorded with regular rate 23 indicia. 24 Q Now -- 25 A But they were marked incorrectly in the exhibit, 19657 1 at least I assume because question 2(b) said they were sent 2 with a non-profit permit, which doesn't really indicate what 3 the indicia on the piece was. 4 Q Did it occur to you that if there was an ambiguity 5 of that sort, that Dr. Haldi or someone under his direction 6 might have done a follow up telephone call to the people 7 that answered the surveys? 8 A But there is nothing indicated on those survey 9 forms that I could tell that indicated that additional 10 information. 11 Q One wouldn't see information, additional 12 information entered on the survey forms if the information 13 was entered directly in the spreadsheet, would you? 14 A But the library reference was provided as evidence 15 and I assume if you are following the evidence rules -- I 16 could recreate the numbers in the exhibit from this library 17 reference, and that's what I was going on. 18 Q Just as we could recreate your numbers from your 19 library references? 20 A Yes. 21 Q Did it occur to you to see if somebody could ask 22 Dr. Haldi through his counsel to explain the discrepancy 23 before you filed your rebuttal testimony? 24 A I was going by -- this was provided. It was 25 provided -- I can't remember the exact date, but it was 19658 1 provided at the end of February. We didn't have very much 2 time to follow up. You know, we had to spend some time 3 getting survey responses that were missing out of it, and 4 trying to get the information together, but since this was 5 filed as the supporting evidence to his testimony, I assume 6 the numbers here could be used to recreate the numbers in 7 his exhibit. 8 Q Did you suggest to anyone on your team that Dr. 9 Haldi be questioned about this on cross examination? 10 A I don't recall. 11 Q Line 11 on the same page, you say 22 of the 108 12 responses are marked with two numbers on a single response 13 form? 14 A Yes, I see that. 15 Q It goes onto say nothing in ANM-LR-1 indicates any 16 reason for this, the double numbering does not appear to 17 correspond to mail, who mailed it, both commercial and 18 non-profit rates. Do you see that? 19 A Yes, I see that. 20 Q Again, did it occur to you to ask anyone to 21 explain it? 22 A Actually, the wording of that second sentence was 23 in response to some information that Ann Reynolds had gotten 24 for me from your office. I was trying to determine why 25 there were 108 responses listed, but it didn't appear to me 19659 1 there were 108 separate survey forms. I was trying to 2 determine that, and that was the only information that we 3 had received from your office, and it didn't appear to me 4 that was the case because there were other survey forms with 5 only one number on them that had both mailed at commercial 6 rates and non-profit rates. That didn't seem to answer the 7 question, but as I said before, we did not receive the 8 library reference until the end of February and there was 9 not much time available to follow up with further questions. 10 Q Would it surprise you if the double numbering 11 refers to mail entered at both commercial and non-profit 12 indicia? 13 A It would surprise me, yes, because there are forms 14 in there, as I recall, that have one number on them, that 15 have mailed with regular rate and non-profit indicia. 16 MR. LEVY: That's all I have, Mr. Chairman, at 17 this time. 18 Thank you, Dr. Schenk. 19 CHAIRMAN GLEIMAN: Is there any follow-up? 20 [No response.] 21 CHAIRMAN GLEIMAN: There's no follow-up. That 22 brings us to redirect. Would you like some time with your 23 witness? 24 MS. REYNOLDS: Just a few minutes, please. 25 CHAIRMAN GLEIMAN: Certainly. 19660 1 [Recess.] 2 CHAIRMAN GLEIMAN: Ms. Reynolds. 3 MS. REYNOLDS: The Postal Service does not have 4 redirect. However, given that Mr. Levy's intended 5 cross-examination was going to be considerably longer, the 6 Postal Service thinks that our future hearing would probably 7 be expedited if Mr. Levy could share with us any 8 cross-examination exhibits that he had prepared regarding 9 Dr. Schenk's survey. 10 CHAIRMAN GLEIMAN: Mr. Levy? 11 MR. LEVY: I can tell counsel exactly what they 12 are. The survey forms and Ms. Reynolds' letter and 13 attachments to me dated March 13th and March 16th of this 14 year. I believe she has possession of all of those, but I 15 can make duplicate copies if that is desired. 16 MS. REYNOLDS: No, I think I can dredge those up. 17 CHAIRMAN GLEIMAN: They are not hidden away in 18 boxes somewhere in some Postal facility. 19 MS. REYNOLDS: I don't assume. 20 CHAIRMAN GLEIMAN: Well, if there is no redirect, 21 then, Ms. Schenk, we want to thank you. We appreciate your 22 appearance today and your contributions to the record. I am 23 sorry that we are going to have to have you come back, but 24 -- 25 THE WITNESS: I will enjoy another visit to D.C. 19661 1 CHAIRMAN GLEIMAN: We sure hope it will be 2 enjoyable. I really do want to thank you and also your 3 counsel for cooperating in this exercise today. It is, you 4 know, in the furtherance of having as complete a record as 5 we possibly can to make a decision on. And I think all of 6 us do appreciate that you are being cooperative and helping 7 us achieve that. So, thank you, if you have nothing 8 further, you are excused. 9 [Witness excused.] 10 CHAIRMAN GLEIMAN: Our next witness is appearing 11 on behalf of the Mail Order Association of America et al. 12 Mr. Andrew is already under oath in this proceeding. 13 Whereupon, 14 GARY M. ANDREW, 15 a witness, was called for examination by counsel for the 16 Mail Order Association of America, and also on behalf of the 17 Advertising Mail marketing Association, and The Direct Mail 18 Marketing Association, Inc. and, having been first duly 19 sworn, was examined and testified as follows: 20 CHAIRMAN GLEIMAN: So, Mr. Todd, if you would 21 introduce your witness and enter his rebuttal testimony into 22 the record, we can, hopefully, get out of here at a 23 reasonable hour tonight. 24 DIRECT EXAMINATION 25 BY MR. TODD: 19662 1 Q Mr. Andrew, do you have before you a document 2 entitled "Rebuttal Testimony of Gary M. Andrew" which has 3 been presented on behalf of the Mail Order Association of 4 America, the Advertising Mail Marketing Association and the 5 Direct Mail Marketing Association, Inc., which has been 6 identified as MOAA, et al.-RT-1? 7 A Yes, I do. 8 Q And do you adopt this testimony today as your 9 testimony as having been prepared by you or under your 10 direction and control? 11 A Yes. 12 MR. TODD: Mr. Chairman, I move that this 13 testimony be admitted into evidence and transcribed into the 14 record at this point. 15 CHAIRMAN GLEIMAN: Are there any objections? 16 [No response.] 17 CHAIRMAN GLEIMAN: Hearing none, Mr. Andrew's 18 testimony and exhibits are received into evidence and I 19 direct that they be transcribed into the record at this 20 point. 21 [Rebuttal Testimony and Exhibits of 22 Gary M. Andrew, MOAA, et al.-RT-1, 23 was received into evidence and 24 transcribed into the record.] 25 19730 1 CHAIRMAN GLEIMAN: Three participants have 2 requested oral cross-examination and we have not been able 3 to wear them out today, so here they are, the American 4 Bankers Association, Edison Electric Institute, and the 5 National Association of Presort Mailers, Major Mailers 6 Association and the Newspaper Association of America. 7 Does any party -- is any other party still 8 standing? Does any party wish to cross-examine? 9 [No response.] 10 CHAIRMAN GLEIMAN: If not, then counsel for the 11 Bankers Association, et al. can proceed when ready. 12 MR. CORCORAN: Thank you, Mr. Chairman. 13 CROSS-EXAMINATION 14 BY MR. CORCORAN: 15 Q Dr. Andrew, I am Brian Corcoran. I represent EEI 16 in this proceeding. 17 A Good evening. 18 Q I would like to begin with a few, I guess what I 19 would call matters of language with your testimony. Could 20 you turn to page 25, please? 21 A Yes. 22 Q At lines 22 to 23 you indicate that Dr. Clifton 23 proposes a reduction for presort business cards between 1 24 and 1.6 cents. Do you see that? 25 A Yes. 19731 1 Q Those lines? 2 A Yes, sir. 3 Q And as support for that, you cite Dr. Clifton's 4 answer to a USPS Interrogatory 20. Did you review that 5 Interrogatory? 6 A Yes. 7 Q Is it true that, in response to that 8 Interrogatory, Dr. Clifton specifically states, quote, "The 9 rates for First Class work-shared cards have been kept at 10 their current rates instead of being set at the rates in the 11 USPS proposal. I am not adopting this as part of my formal 12 proposal." Wasn't that his testimony? 13 A That was his testimony, but in his final 14 statements of accounts in Technical Appendix D, he included 15 these reductions, implicitly? 16 Q It wasn't abundantly clear to you that, when he 17 said I am not adopting this as part of my formal testimony, 18 that he was using it simply to complete his technical 19 appendices, and they were for illustrative purposes only? 20 A Well, that's how he got his whole system to 21 balance from terms of the revenue -- meet the revenue 22 requirements, so I assume they are in there implicitly. 23 They being the reduction from the USPS proposed rates, or 24 stating another way, keeping the -- maintaining the same 25 rates as they are today. 19732 1 Q Well, that's fine. 2 Let me move to one other area briefly. Is it 3 correct that the current bench-mark used to set discounts 4 for first class work-shared mail is the first class single 5 piece rate? 6 A Today? 7 Q The current bench-mark, correct. 8 A The bench-mark that's used in this case or the 9 bench-mark that was used in '95? 10 Q I'll repeat the question. Is it correct that the 11 current bench-mark used to set discounts for first class 12 work-shared mail is the first class single piece rate? 13 A I don't know what you mean by "current." 14 Q It's the one that the discount -- the current 15 discounts are based upon. 16 A Okay. That's all I needed to know. The one that, 17 today's date, is based on the single piece, but it was used 18 as a proxy because the bench-mark or the data necessary to 19 do the bulk rate metered mail that was proposed was not 20 available in '95, but it is available today. 21 Q Excuse me. Was the answer to your question -- and 22 your explanation is fine -- was the answer to the question, 23 yes, that's the current bench-mark? 24 A That's correct, with modifications. 25 Q And is it correct that Dr. Clifton uses the 19733 1 current bench-mark for purposes of setting discounts for 2 first-class work-shared mail in his testimony? 3 A It is true. 4 Q At page 24, line 7 of your testimony, you've 5 chosen to characterize Dr. Clifton's use of the current 6 bench-mark as representing an increase in the bench-mark 7 used to determine cost savings for work-shared discounts. 8 Do you see that language? 9 A State the line, please? 10 Q I believe it's line 7 on page 24. 11 A Yes. 12 Q It's correct, therefore, to state, is it not, that 13 Dr. Clifton has not proposed to increase the bench-mark. 14 A Over the existing or over what the USPS has 15 proposed? 16 Q He's using the current bench-mark as you 17 previously testified. Isn't that right? 18 A That's correct. 19 Q And it's correct to state, therefore, that Dr. 20 Clifton has not proposed to increase the bench-mark. Isn't 21 that correct? 22 A That's correct. 23 Q Are you an economist? 24 A I consider myself an economist. I don't have a 25 Ph.D. in economics, but I've taken all the courses 19734 1 necessary. My Ph.D. is in operations research. 2 Q I see. At page 24 of your testimony, you 3 criticize Dr. Clifton's reduction in cost coverage for first 4 class work-shared mail as being based upon considerations of 5 efficiency and equity. Do you see that? 6 A On 24? 7 Q Correct. Line 11, specifically. 8 A Got it. 9 Q Is it correct that a Ramsey price is more 10 economically efficient that a price set in excess of a 11 Ramsey price? 12 A Yes. 13 Q Is it correct that an improvement in efficiency 14 occurs when the departure from Ramsey price is lessened? 15 A Yes. 16 Q Is it correct in this case the Postal Service has 17 proposed rates for work-shared first class yielding cost 18 coverage of 283 percent? 19 A Say again? 20 Q Is it correct that, in this case, the Postal 21 Service's proposed rates for work-shared first class mail 22 yield a cost coverage of 283 percent? 23 A That is correct. 24 Q Is it correct that had the Postal Service set 25 rates for work-shared first class mail based upon Ramsey 19735 1 pricing, the cost coverage would have been 245 percent? 2 A I'd have to look that up. I don't know -- don't 3 have it at my fingertips. 4 Q Did you read Witness Bernstein's testimony? 5 A A long time ago, yes. 6 Q Isn't it true that his testimony would result in 7 that 245 percent or don't you recall? 8 A I said I'd have to check. 9 Q Okay. Well, let's assume for purposes of my next 10 question that's the case. 11 A Subject to check. 12 Q Thank you. 13 Therefore, since Dr. Clifton has proposed to 14 reduce the cost coverage for first class work-shared mail, 15 the result must be, under Ramsey pricing, more efficient 16 than the Postal Service's, correct? 17 A With respect to Ramsey pricing, yes. 18 Q Is it also correct that economists deem Ramsey 19 pricing criteria to be objective and not subjective? 20 A Yes. 21 [Whereupon, at 6:00 p.m., the hearing continued in 22 evening session.] 23 24 25 19736 1 EVENING SESSION 2 [6:00 p.m.] 3 BY MR. CORCORAN: 4 Q So much for language. Let's move to your mail 5 mix. This is Section A of your testimony. That begins on 6 page 27, I believe. Is it your testimony that changes in 7 mail mix are the principal reason for decline in unit costs 8 for first class work shared mail? 9 A One of the principal reasons, yes. 10 Q Doesn't your testimony at 27, the heading, say the 11 primary reason? 12 A Yes. 13 Q Now, in effort to demonstrate your conclusion that 14 mail mix is the primary reason for decline in unit costs, 15 you include several tables in your testimony, including 16 table 10, which is on page 29 of your testimony, and which 17 shows the percentage shares of non-automation and automation 18 work shared first class mail for the years 1994 to 1996; 19 correct? 20 A Yes. 21 Q It's your testimony, I take it, that a shift of 22 this magnitude, the 12.7 percent, would cause a reduction in 23 overall unit costs in the CRA; is that correct? 24 A It can; yes. 25 Q Excuse me. Look at your testimony on lines 13 to 19737 1 15. I've paraphrased it, but that's specifically what you 2 say there, isn't it, a shift in volume within work shared 3 mail of this magnitude, i.e., 12.7 percentage points, from a 4 higher cost rate category to a lower cost category would 5 cause a reduction in overall unit costs? That's your 6 testimony; correct? 7 A Yes. 8 Q To demonstrate your contention, if you would turn 9 to page 30, please, of your testimony, to demonstrate your 10 contention, you developed table 11, in which you show a 11 hypothetical example where a decrease in average costs for 12 presort first class mail is consistent with increases in 13 unit costs for each rate category, due to volume mix shifts 14 from higher cost non-automation to lower cost automation; 15 correct? 16 A Yes. 17 Q As a somewhat preliminary matter, does your 18 testimony provide data showing the unit costs for rate 19 categories within work shared first class mail for the 20 period 1994 through 1996? 21 A Not in this table. 22 Q As to the -- you don't have any data in here, do 23 you, specifically with respect to unit costs for rate 24 categories within first class mail for the period 1994 to 25 1996, do you? 19738 1 A No, I do not. 2 Q Now, as to the hypothetical, and you do label this 3 as a hypothetical example, the unit costs for non-automation 4 and automation work shared first class mail are made up, 5 aren't they? 6 A Yes, the costs. 7 Q Yes, exactly. In addition, the relationship 8 between the unit costs of non-automation to automation, 9 i.e., with non-automation being 100 percent higher, is made 10 up; correct? 11 A Yes. 12 Q In fact, they bear no relationship to the historic 13 cost relationship between the two types of mail; is that 14 correct? 15 A That's correct. 16 Q Similarly, the percentage shares that you show -- 17 excuse me. I've lost myself. Similarly, the percentage 18 share of the mail stream represented by non-automation and 19 automation is made up; correct? 20 A Yes. 21 Q Further, as you demonstrated on table 10 where you 22 did use the actual percentage shares, the percentage of 23 shares you employ in table 11 represent a complete departure 24 from the actual results for the time period; correct? 25 A Yes. 19739 1 MR. CORCORAN: I have a cross examination exhibit 2 I would like to show you. 3 [Cross-Examination Exhibit No. 4 ABA/EEI/NAPM-XE-1 was marked for 5 identification.] 6 BY MR. CORCORAN: 7 Q Do you have what I have labelled ABA/EEI/NAPM 8 Cross-Examination Exhibit 1 before you? 9 A Yes. 10 Q Let me just identify the source of the data. For 11 1994, do you see that we have -- it's an attempt to 12 replicate your Table 11. That cost of nonautomation and 13 automation, do you see that? -- 4.8 cents for automation? 14 A Yes, sir. 15 Q Do you see that? That comes from Witness Smith's 16 testimony in MC-95-1 and represents the cost of automation 17 for 3-Digit mail and we have rounded for the 4.7 cents to 18 4.8 and the nonautomation share represents the difference, 19 as you can see in the footnote, between a 3-Digit automation 20 share and nonautomation mail, as provided by Witness Smith 21 in that case and that difference is calculated to be 47 22 percent. 23 So this -- 24 A Say again? 25 Q The difference between automation and 19740 1 nonautomation is 47 percent. That is how I derived the 7.1 2 cents. Okay? 3 A All right. 4 Q Are you with me? All right, and the share, as 5 noted in the Footnote 2, is the share of nonautomation and 6 automation mail in 1994 -- 42 percent for nonautomation and 7 58 percent for automation as compared to what you have used 8 in Table 11, 75 percent and 25 percent, respectively. Okay? 9 A Yes. 10 Q Now is it correct that under this example, given 11 more realistic unit cost figures, and a more realistic 12 spread between the two categories, and using actual mail mix 13 changes that the mail mix change, assuming no change in 14 costs, causes the average unit cost to decrease by 5.2 15 percent? 16 A That is exactly what we were trying to show in 17 Table 11 but we wanted to stay away from any real numbers so 18 we wouldn't have any arguments about details. 19 Q Wait a minute. Table 11 shows an increase, a 10 20 percent increase in costs, does it not? That is not what 21 my -- 22 A Say again? 23 Q Your Table 11 -- does that have a -- 24 A Sure -- 25 Q -- a 10 percent increase in costs? 19741 1 A It has a 10 percent increase in costs -- 2 Q Yes, sir. 3 A -- and it has an 11.4 decrease in cost on a 4 average basis. 5 Q Right. 6 A The actual costs of each individual rate category, 7 those costs were going up by 10 percent, but because of the 8 mix change the overall average made a decrease and that is 9 what we were trying to show. 10 Q That's fine, but in -- we will get to that 11 assumption about the increase momentarily, but bear with me 12 on Cross-Examination Exhibit 1. 13 That table, given more realistic cost figures, a 14 more realistic spread between the two rate categories, and 15 assuming -- and actual mail mix, and assuming no change in 16 costs causes the average unit cost to decrease by 5.2 17 percent, isn't that what this table shows? 18 A Yes. 19 Q And isn't it correct that Dr. Clifton's analyses 20 demonstrate mail processing costs for work-shared First 21 Class mail declined in excess of 5.2 percent over the period 22 1994 to 1996? Correct? 23 A Yes. 24 MR. CORCORAN: We have another cross examination 25 exhibit for you. 19742 1 [Cross-Examination Exhibit No. 2 ABA/EEI/NAPM-XE-2 was marked for 3 identification.] 4 BY MR. CORCORAN: 5 Q Do you have it in front of you, Dr. Andrew? 6 A Exhibit 2. 7 Q Yes, sir, marked ABA/EEI/NAPM Cross-Examination 8 Exhibit 2, and it's the same table, essentially, as number 9 1, except that we've increased the costs by 10 percent in 10 1996. Do you see that? 11 A Yes. 12 Q Now, under these conditions and contrary to the 13 conclusion you draw from your Table 11 that volume exchanges 14 would cause average unit cost to decrease, here we see 15 average unit costs actually increase by 4.3 percent, 16 correct? 17 A Given the conditions you have here, that is 18 correct. 19 Q Right. And in his testimony -- well, you've 20 already answered that. 21 Now, given our Cross-Examination Exhibit 1, which 22 shows that actual volume mix shift accounts for a 23 5.2-percent decline in average costs, what, in your view, 24 accounts for the balance of the decline in unit costs that 25 are included in Dr. Clifton's analyses -- i.e., the 19743 1 difference between the 5.2 percent and his decline of 2 approximately 13 percent? 3 A It could be a number of factors. 4 One, it could be a difference in the change of the 5 costs by rate category going from '94 to '96. It could be a 6 -- that's a relative change between the two. 7 Q Isn't that your mail mix argument? 8 A No. I'm talking about the cost -- the unit cost 9 of the category, not the mixes of the category, but the unit 10 costs of the category. So, your 7.1 and 4.8 -- those may 11 not change uniformly through time. 12 So, there could be a change in the level of those 13 costs relative to one another, and there could be a change 14 in the base -- in other words, they move up together or they 15 -- so, there's two things operating -- the relative 16 difference between the actual costs of the non-automated 17 versus the automated and the relative level or base of the 18 two, and given whatever happens in there, you can get all 19 sorts of numbers down on the right-hand side at the bottom. 20 I also want to know on the 7.1 -- if we're going 21 to -- are you going to ask anymore questions on this, sir? 22 Q Perhaps. 23 A I need to know the source of your 10-percent 24 increase. Since we're using actuals -- 25 Q That's easy. I just attempted to have replicated 19744 1 your Table 11, and when you use more realistic numbers, when 2 you use actual volume shifts, when you use costs that are -- 3 where the spread is more comparable to the actuals, you get 4 a result that is completely at odds with your Table 11. 5 A It's not at odds at all, because in the first 6 case, where there's no change, you're having a decrease 7 that's strictly a function of mix, and that's what the whole 8 exercise in 11 was to show. 9 Q Right. You claim that it's solely volume mix that 10 causes the costs to go down. 11 A No, I didn't say that. 12 Q That's your -- that's what you started out in your 13 heading to this testimony, that changes in mail mix 14 categories are the primary reason for declining unit costs. 15 A Primary, but they're -- 16 Q Nowhere in your -- excuse me -- nowhere in your 17 testimony do you say that there may be some other factors 18 such as the items you just were elaborating on. 19 A Okay. 20 Q True? 21 A I haven't read my testimony today. I thought I 22 had, but if I didn't, I didn't. 23 Q And isn't it possible that -- besides the mail 24 mix, isn't it possible that attributable costs were actually 25 dropping and that's the difference between the 5.2 that we 19745 1 show in Exhibit 1 and Dr. Clifton's results, which show 2 minus -- negative 13 percent? It's because the costs are 3 dropping. It's not mail mix. The mail mix is 5.2 percent, 4 correct? 5 A The mail mix, with your numbers, give a 6 5.2-percent decrease. 7 Q And if the results that Dr. Clifton shows is, for 8 example, 13 percent, the difference would have to be, at 9 least in part, dropping attributable costs, correct? 10 A Not necessarily. 11 Q In part, they wouldn't have to be, even? 12 A No. I can construct you an example, if you would 13 like me to -- I won't do it on the stand -- 14 Q Well, thank you. 15 A -- but I'll give it to you. 16 Q And I would note you didn't do it in your 17 testimony. You just said mail mix was the primary reason. 18 MR. TODD: I would like an opportunity for the 19 witness to complete his answer, Mr. Chairman. 20 THE WITNESS: I would be glad to make a 21 counter-example or an example that would show the situation 22 where this will go farther negative and the costs will at 23 least stay the same or they certainly won't decrease. 24 BY MR. CORCORAN: 25 Q Well, let's move on. 19746 1 A I take it that's a negative on the offer. 2 Q Even if it wasn't 6:30, it would be a negative. 3 Now, you also criticize -- and I'll find it here 4 -- at page 31 -- and I believe it's lines 4 and 5. In this 5 section A of your testimony, you characterize Dr. Clifton's 6 use of only two years of change in historical -- i.e., CRA 7 data -- from '94 to '96 as suspect. Do you see that on 8 lines 4 and 5? 9 A Yes. 10 Q And by this, is it your testimony that Dr. Clifton 11 should have considered a longer historical period? 12 A He definitely should have done that and/or modeled 13 -- done a model of this system he was working with. 14 Q Now, given your criticism of the historical time 15 period used by -- excuse me, I'll start again. Isn't your 16 criticism of the historical time period used by Dr. Clifton 17 irrelevant given your contention that the change in mail mix 18 is the principal reason for the declining unit costs for 19 First Class work-shared letter mail? 20 A Not at all, because the change in mail mix may not 21 continue, and that is another reason for the criticism of 22 using only the two year without a modeling of the migration 23 from higher cost to lower cost categories. 24 Q Well, I was going to get to this later, but to 25 keep it in one spot in the transcript, because it will be 19747 1 riveting reading, I know, isn't it the case that -- and I'll 2 get to it momentarily -- isn't it the case that Witness 3 Tolley projects a decline of 13.1 percent of non-automation 4 share before rates from a 28.76 percent in the base year? 5 A I don't remember those numbers. If you will give 6 me the citation, I have -- 7 Q That's fine. If you don't remember, that's fine. 8 A Okay. 9 Q His testimony will speak for itself. And you 10 also, is it correct that Witness Thress -- Thress -- 11 whatever -- projects an interim year, i.e., '97, decline in 12 non-automation share before rates of 14.9 percent? 13 A I will have to take that subject to check. 14 Q See, what I don't understand then, Doctor, you 15 just said -- don't those -- if those two are accurate, 16 that's projecting a decline into the interim year and into 17 the test year, correct? So the decline is, by the Postal 18 Service, is projected to continue, correct? 19 A Not at the level in the test year that has been in 20 the past. 21 Q You mean the 12.7? It declined 12.7 percentage 22 points. 23 A Would you cite that for me in -- 24 Q Well, that was your testimony on page 10, your 25 Table 10. 19748 1 A Say again? 2 Q You show on Table 10 a decline, those are actual 3 mail mix shifts. It's a percentage change, percentage point 4 change of 12.7 percent, correct? 5 A That's correct. But you are talking about going 6 into the test year after rates. 7 Q Right. And just bear with me, if the numbers I 8 quoted were correct, and Witness Tolley projects a decline 9 to 13.10 percent from the base year of 28.7 percent, that's 10 a larger decline that you show in Table 10, correct, in 11 terms of percentage points? 12 A In the interim year. 13 Q No, sir. Assume with me that I have read Tolley 14 correctly. Then he projects a decline to 13.1 percent from 15 the base year, which you have got right here, of 28.7. 16 That's a decline in percentage point terms greater than what 17 you show in Table 10, correct? 18 A Would you please give me the citation from which 19 you are reading? And then I can follow it with you, sir, 20 because I have the documents here. 21 Q You'll have to take it subject to check, I don't 22 have a page number for Tolley. So assuming that I am right, 23 the projection -- 24 A Well, I have it, it's in -- it's reproduced in 25 Table 8 of Witness Clifton's, on page 10 -- I'm sorry, page 19749 1 19. 2 Q Nineteen. 3 A Now if you care to go ahead, I'll follow with you. 4 Q My question was, and I thought this would be easy 5 to keep it all in one spot, but the question is, if Witness 6 Tolley projects a decline in the nonautomation share for the 7 test year to 13.1 percent, that's a decline that's greater 8 than in terms of percentage points, given that the base year 9 is 28.7, it's a decline that's greater than the 12.7 percent 10 you show in your table 10. Correct? 11 A That is correct, but you've got to look at the 12 progression that's taking place. In '97, the interim year, 13 the total is 14.9 percent, so that's a reduction of 13.8. 14 And then from 14.9 in '97 it goes to 13 even in '98, or a 15 1.9-percent decrease. So we've got a definite leveling off 16 which one would expect because there's only so much shift 17 that can take place in these migrations. We've experienced 18 that in Standard A mail over the years. And you've 19 experienced it in your own -- 20 Q Sure. 21 A Systems. 22 Q But your table 11 -- excuse me, your table 10 -- 23 shows the decline for two years. 24 A That's correct. 25 Q And from the base year to the test year is a 19750 1 decline greater than you show in table 10. I think you've 2 agreed. Is my math correct? 3 A Yes. Yes. 4 Q Okay. Let's go back. I didn't mean to -- 5 A But the dynamic is the importance here, because 6 the dynamic changes through time. 7 Q Let's go back to the -- where I left off, which 8 was your criticism of Dr. Clifton concerning the time period 9 he used, and your testimony was that your criticism is not 10 irrelevant -- notwithstanding your reliance that the 11 principal cause for the decline in unit costs is first -- is 12 the mail mix shift. Just setting the table here just to set 13 us back. 14 MR. CORCORAN: Let me show you or provide to you 15 another exhibit. 16 [Cross-Examination Exhibit No. 17 ABA/EEI/NAPM-XE-3 was marked for 18 identification.] 19 BY MR. CORCORAN: 20 Q Preliminarily, Dr. Andrew, if you would turn to 21 page 32. 22 A Yes. 23 Q Is the purpose of this table -- and I guess I said 24 Table 32, I apologize -- page 32, Table 12 -- is the purpose 25 of this table to demonstrate that Dr. Clifton's use of the 19751 1 CRA data from the period '94 to '96 was too short a period 2 to be representative? 3 A That and the fact that it's the most negative -- 4 largest decrease of the whole period. In fact, it's the 5 only decrease in the whole period. 6 Q And -- well, that's not quite true, is it? 7 Doesn't '93 go down slightly for a one-year period? 8 A Yes, on the one-year period, you're right. 9 Q Now, your Table 12 shows the cost per piece and 10 percentage change for one year and two years for work-shared 11 first class mail, correct? 12 A Yes. 13 Q And now we can go to what I've labeled ABA/EEI 14 Cross-Examination Exhibit 3, and it was handy that you 15 pointed to Dr. Clifton's Table 8, because as you note, we've 16 attempted to replicate your Table 12 for the years 1988 17 through '96. 18 The non-automation share, column two, comes from 19 Dr. Clifton's Table 8. The percentage change is just the 20 calculation of the difference between the years, and the 21 automation share also comes from Dr. Clifton's Table 8, and 22 again, the percentage point change is simply the difference 23 in the year, and the costs in item -- excuse me -- row 6 are 24 yours, from your Table 12, as is the percent change. Do you 25 see that? 19752 1 A Yes. 2 Q Now, given your testimony about the -- that mail 3 mix would cause -- caused the decrease, let me direct your 4 attention, for example, to 1992. You see the automation 5 share from the period from '91 to '92, okay? You see that 6 the automation share in '92 is 62.3 percent, correct? 7 A Yes. 8 Q And that's a drop from '92 of 16.2 percentage 9 points, correct? 10 A No. 11 Q What is it? 12 A You said it was a drop from '92? Drop from '91. 13 Q Thank you. I mis-spoke, and I apologize. 14 A Yes, from '91 to '92, the change is 16.2 percent. 15 Q And the automation share went up by an equal 16 amount. 17 A Correct. 18 Q And the unit cost you show went up by 3.6 percent, 19 correct? 20 A That's correct. 21 Q And in fact, Dr. Clifton -- excuse me -- Dr. 22 Andrew, in year 1989, there's a volume shift; in 1990, 23 there's a volume shift; in 1991, there's a volume shift; and 24 in each of those years, the costs go up, correct? 25 A That's correct, and in each of those years we had 19753 1 higher rates of inflation that we do currently. So, your 2 base -- when I talked about what could change -- what can 3 change the average cost of the piece within that work-shared 4 mail, it can be the general level of cost, as well. 5 Q When you refer to inflation, are you referring to 6 postal costs specifically or just your general sense of how 7 inflation was at the time? 8 A Well, the one I looked at was transportation, I 9 think. I just looked at it to see what was happening back 10 there, and most costs in that period were more than they are 11 today, in terms of the indices. 12 Q I see. And did you look at that after I sent you 13 this cross-examination exhibit yesterday? 14 A Yes, but I looked at it specifically again just to 15 make sure. 16 Q Sure. And if -- could -- 17 A By the way, I didn't get this till this morning, 18 but that's okay. 19 Q Well, I knew we wouldn't be up till -- 20 Is it possible that the decline that's shown, say, 21 in '96, the decline in cost there, due to inflation being 22 reduced? 23 A It could. 24 Q And also, you notice those costs go down in '95 25 and '96. Classification reform was implemented in '95, 19754 1 perhaps? Do you know? 2 A Ninety-six. 3 Q I see. Were mailers taking steps in '95 in 4 anticipation of the classification reform? 5 A I do not know. 6 Q If they were, would that have influenced costs 7 during that period? 8 A I do not know. It could have, I suppose. 9 Q Okay. Classification reform was in effect in '96. 10 Could that cause these costs to be reduced? 11 A You just asked that question, didn't you? 12 Q In '96, not '95. You said -- I thought you told 13 me that classification reform became effective in '96. 14 A Yes. 15 Q Could that be a contributing factor to the reduced 16 costs shown in the last column here? 17 A I do not know. I haven't analyzed those changes. 18 Q Okay, that's fine. 19 Turning to a different topic, do you understand 20 that Dr. Clifton's negative 3.6 roll-forward factor applies 21 to mail processing costs for First Class work-shared mail? 22 A Yes. 23 Q Do you agree that Dr. Clifton's volume forecasts 24 reflect the regression methodology and equations developed 25 by Witnesses Thress and Tolley? 19755 1 A Would you repeat the question, please? 2 Q Do you agree that Dr. Clifton's volume forecasts 3 reflect the regression methodology and equations developed 4 by USPS Witnesses Thress and Tolley? 5 A Yes. They purport to do that. 6 I have not checked them in great detail, however, 7 because we were never able to obtain a machine-readable form 8 other than the hard-wired copy with the numbers in it of Dr. 9 Clifton's workpapers. 10 Q Did you request from counsel any assistance? 11 A Yes. 12 Q Which -- I don't mean yours. I mean me or Mr. 13 Warden, who is at my right. 14 A It was my understanding that there had been two or 15 three iterations of it and we finally gave up and did it by 16 hand, what we could do. 17 Q Well, I will just say that it wasn't passed on to 18 me. 19 Your Table 9, which is on page 26, shows the 20 combined effects of Dr. Clifton's proposal concerning First 21 Class mail, correct? 22 A Yes. All of his proposed changes that are 23 reflected in the combination, namely Technical Appendix D as 24 in "dog." 25 Q Right, and his Technical Appendices break out his 19756 1 proposals separately, correct? 2 A Yes. 3 Q So for example with respect to what he has called 4 his discount proposal, that is in Technical Appendices C-1? 5 A I believe that is correct. 6 Q Okay. Did you examine -- well, let me phrase it 7 this way. 8 Under his discount proposal, isn't it correct that 9 revenues for First Class work-shared mail increased by $205 10 million? 11 A That is what your exhibit -- potential 12 Cross-Examination Exhibit page 2 says and we checked it and 13 the answer is yes. 14 Q And isn't it also correct that there is no impact 15 on Standard A commercial mail? 16 A That's right. 17 It is also true that the revenue is in imbalance 18 inn First Class and you are -- you have used up $152 million 19 that you don't have. 20 Q Well, let me see if I understand your position. 21 Your testimony addresses his combined proposals. Is it also 22 your testimony or the position of your clients that you 23 object to Dr. Clifton's discount proposal, which has no 24 impact on Standard A mail? 25 A I can't speak for counsel. 19757 1 Q Okay -- never mind your clients. Let me limit 2 it -- let me limit it to you and I will rephrase it. 3 Your testimony addresses the combined effects. As 4 we just discussed, his discount proposal has no impact, no 5 revenue impact, on Standard A mail. 6 Do you or is it your testimony that you object to 7 this portion of his analyses or is it only the combined 8 presentation? 9 MR. TODD: May I -- Mr. Chairman, even there I 10 suppose the witness can speak for himself. It seems to me 11 that the question of what this testimony does is contained 12 in the testimony. I don't believe that asking a witness in 13 the abstract whether he objects to a certain result within a 14 given class of mail is a proper question. 15 I think the question should be focused on the 16 analysis which has been done by the witness and whether that 17 analysis -- whatever questions he may want to ask about the 18 analysis, but getting back to the question of what is the 19 ultimate result of this and whether he thinks it's a good 20 idea or a bad idea I think is well beyond the scope of his 21 testimony. 22 CHAIRMAN GLEIMAN: Mr. Corcoran, do you want to 23 comment before I rule? 24 MR. CORCORAN: Okay. His testimony is -- he had 25 Dr. Clifton's testimony in front of him. Dr. Clifton's 19758 1 testimony consists of three proposals. He has combined all 2 of them, I am attempting to examine one portion of that 3 testimony. I think it is proper and the witness should 4 answer. 5 CHAIRMAN GLEIMAN: I'm not sure that I follow. 6 MR. CORCORAN: I am examining -- he has combined, 7 in his Table 9, and throughout his analyses, Dr. Clifton's 8 various proposals with respect to First Class mail. 9 My client is only interested in the discounts 10 testimony and that is what I am focusing on at the moment. 11 The discount proposal has no impact in terms of increased 12 revenues on standard mail. So the question to the witness 13 was simply, do you object to that proposal? 14 MR. TODD: If I may just perhaps clarify things, 15 or perhaps not. But, as is stated on page 1, a note to 16 footnote 1 of the testimony of Mr. Andrew, the reason he 17 combined this is because Dr. Clifton chose to combine it 18 and, therefore, there wasn't any other intelligent way of 19 putting the matter. 20 Now, if Mr. Corcoran would like to know whether 21 the Mail Order Association of America or the other parties 22 submitting this testimony are particularly concerned, at 23 least insofar as this analysis goes, as to what happens 24 within the First Class mailstream so long as it doesn't 25 negatively affect Standard A, the answer is we really don't, 19759 1 as a general matter. It doesn't mean that in certain areas 2 we may not take positions pro or con particular issues. 3 But, again, Mr. Andrew's testimony is an analysis, 4 in part, of Dr. Clifton's testimony, and it seems to me that 5 any cross-examination concerning that analysis is proper. 6 But going beyond to say whether he likes the end rate 7 results or not is not proper. 8 CHAIRMAN GLEIMAN: I am afraid I am going to have 9 to ask you to reconsider your question or rephrase your 10 question, or withdraw your question, Mr. Corcoran. 11 MR. CORCORAN: Okay. I will deem it withdraw. 12 BY MR. CORCORAN: 13 Q Switching topics, is it your testimony that the 14 cost characteristics of automation work-shared First Class 15 mail and Standard A mail are not similar? 16 A The cost characteristics? 17 Q Yes. 18 A Yes, it is my testimony that they are not similar. 19 Q That they are not similar. I see. Did you bring 20 USPS Exhibit 29(c) with you? 21 A Yes. 22 Q Could you take that out, please? 23 A I have it, sir. 24 CHAIRMAN GLEIMAN: Actually, Mr. Corcoran, I think 25 it's best that you identify what they are on the record, so 19760 1 that the record shows what it is you are talking about. 2 MR. CORCORAN: Sure. Thank you, Mr. Chairman. 3 BY MR. CORCORAN: 4 Q Dr. Andrew, do you have before you USPS Exhibit 5 29(c), page one, which is first class unit cost estimates? 6 A That's page one of six? 7 Q Yes, sir. 8 A Yes, sir. 9 Q And page two, which is standard regular unit cost 10 estimates? 11 A Yes, sir. 12 Q The columns there show mail processing and 13 delivery costs for various types of first class mail and 14 standard A mail; is that correct? 15 A Yes. 16 Q Would you agree that the mail processing and 17 delivery costs, for example, for first class automation 18 three digit and standard A automation three digit, are 19 similar? 20 A They are within two-tenths of a cent, yes, on a 21 base of eight. 22 Q I'm sorry. Your answer was yes, they are similar? 23 A Yes. 24 Q Are the mail prep requirements for these two types 25 of mail similar, that is automation presort, first class 19761 1 presort and standard A automation? 2 A I am not familiar with standard A. 3 Q Are they processed on the same machines by the 4 Postal Service? 5 A Some of the materials are, yes. 6 Q Letter shaped mail? 7 A Yes. That's my understanding. 8 Q Except for perhaps certain air transportation, are 9 they transported in the same vehicles? 10 A Oh, definitely not, because standard A mail is 11 heavily drop shipped and first class mail does not drop 12 ship. 13 Q Once they get to the destination facility, whether 14 it's a SCF or DDU, whatever it happens to be, aren't they 15 commingled at that point and transported together? 16 A But you are trying to make an identification that 17 the costs are so similar between these two and they 18 definitely are not when it comes to transportation. 19 Q But overall, the costs, as you just noted, between 20 automation, work shared, first class, three digit and 21 standard automation are remarkably similar; correct? 22 A For those two functions, yes, for mail processing 23 and delivery. 24 Q Is it correct that in the Postal Service's 25 analyses, Witness Daniel and Witness Hatfield used the same 19762 1 productivities for standard A mail and work shared first 2 class mail? 3 A I did not check standard -- first class. 4 Q Is it correct that standard A letter mail weighs 5 approximately 50 percent more than first class automation 6 work shared letters? 7 A Would you repeat that? 8 Q Is it correct that standard A letter mail weighs 9 approximately 50 percent more than first class automation 10 work shared letters? 11 A That's what the exhibit showed that you gave me 12 and we checked it and the answer is yes, according to the 13 Postal Service's information, it's actually 63 percent, but 14 it turns out -- 15 Q That's what I had, too. 16 A But it turns out that the cost is a J shaped curve 17 in that area of weight and as you get lighter and lighter 18 pieces, the costs have been shown to go up. 19 Q Pardon me? Could you repeat that, please? 20 A Yes. If you draw a graph and on the horizontal 21 axis, show weight in ounces, and on the vertical axis, you 22 show costs, unit costs of mail, as the piece gets lighter in 23 the area of two to one to zero or 6.66, which I think is 24 your average, the costs go up, not down. 25 Q What costs are they? What costs go up? 19763 1 A The processing costs. 2 Q Mail processing costs? 3 A Yes. 4 Q That's not born out by the USPS Exhibit 29(c), 5 pages one and two. 6 A You don't have weights in here. 7 Q You have the average weights that I forwarded to 8 you. Aren't those reflected in this data by the Postal 9 Service in 29(c)? 10 A Yes, but you have different handling 11 characteristics of each one of them. Take and compare one 12 to one. 13 May I confer with counsel a moment? 14 Q This could be done on redirect, perhaps. I'm 15 willing to move on, given the hour, and if they want to do 16 it on redirect, that's fine. 17 A I have references to Library Reference 182 on this 18 matter. 19 CHAIRMAN GLEIMAN: Well, if Mr. Corcoran is 20 willing to withdraw whatever question it was he was asking 21 you -- 22 MR. CORCORAN: Yes. 23 CHAIRMAN GLEIMAN: -- then we don't have to worry 24 about that right now. 25 THE WITNESS: Thank you. 19764 1 BY MR. CORCORAN: 2 Q You talk about, at page 36 of your testimony -- 3 and this follows up on the cost characteristics issues, and 4 you talk about first class letters have a higher value and 5 that's demonstrated by certain specific characteristics 6 which you detail on lines 6 through 11 on that page, 7 correct? 8 A Yes. 9 Q Now, in Witness Hatfield's analysis, isn't it 10 correct that he employed a paid premium adjustment to 11 reflect the fact that first class mail is being accorded 12 expeditious handling and a higher delivery priority? 13 A Would you repeat that and give me a reference in 14 -- 15 Q Page 13 -- Appendix 1, page 13 of USPS-T-26. 16 Isn't it correct that, in his analysis, he employs a paid 17 premium adjustment that reflects first class mail being 18 accorded expeditious handling and delivery priority? 19 A Give me your cite again, please. 20 Q It's page 13 of Appendix 1. 21 MR. CORCORAN: May I approach the witness just to 22 show him the page? 23 THE WITNESS: I've got it. 24 MR. CORCORAN: And just for the record, we 25 indicated in correspondence to Mr. Todd that we would -- we 19765 1 may use this material, that's all. 2 BY MR. CORCORAN: 3 Q Do you have it? 4 A I have it. 5 Q Do I need to repeat the question? 6 A No. Just give me your reference on the page. 7 There's a lot of numbers on the page. 8 Q It's column six, premium pay adjustment. 9 A Yes. 10 Q All right. 11 Now, isn't it the case that Witness Hatfield 12 increased unit costs by 1.1 percent to reflect that or do 13 you know? 14 A He changed it with the premium pay, but the 15 premium pay does not reflect all of the costs associated 16 with that. That's a proxy. 17 Q It's a proxy for what? 18 A For the correction that your question asked. 19 Q Doesn't it relate -- the premium pay relates to 20 expeditious handling and delivery, does it not? 21 A Yes. 22 Q Now, isn't it also true that Witness Daniel also 23 reflected a negative premium pay adjustment with respect to 24 standard A mail? 25 A Yes, I recall that. 19766 1 Q And these are both cost items. That's your item 2 -- line 7, first class mail is accorded expeditious handling 3 and delivery? 4 A Yes. 5 Q Isn't it true, therefore, that those costs are 6 already reflected in the attributable costs for this mail? 7 A Only part of them. 8 Q Isn't it also true that with respect to your line 9 9, item D, First Class mail benefits from free forward and 10 return, that that too is a cost item reflected in the 11 attributable costs of First Class mail? 12 A Yes, but it's not in the model. I do not believe 13 it's in the model used by Witness Daniel for First Class. 14 Q Witness Daniel didn't do the First Class, did she? 15 A No, she used somebody else's. But she made no 16 correction in there. 17 Q Isn't it also true that the attributable costs for 18 First Class work-shared mail reflect the benefits of 19 dead-letter operations which you cite on line 10 of your 20 testimony? 21 A Yes, I believe it does. 22 Q And isn't it also true that the attributable costs 23 for First Class mail reflect what you cite at line 6, the 24 long-distance mailings are transported by air? 25 A No. 19767 1 Q That's not an attributable cost item? 2 A It is an attributable cost item, but if you look 3 at how the -- I take that back. Yes, it is in the total 4 cost. 5 Q Thank you. So isn't it true of the five items you 6 identify here, only the sealed against inspection is a value 7 item? 8 A No, they're all value items. Some of them have 9 had some accounting for the additional costs, but not all. 10 Q Some have attributable costs associated with them, 11 and some don't. Is that your testimony? 12 A And the ones that do, we have no guarantee that 13 all the attributable costs are accounted for. 14 Q And would that be -- if they're counted as 15 attributable costs and then they're counted as a value item, 16 would that be in the parlance of regulatory terms a double 17 counting of the same factors? 18 A Yes. If they both had the total. 19 Q Thank you, Dr. Andrew. 20 MR. CORCORAN: Mr. Chairman, thank you very much. 21 Can I move the transcription of the 22 Cross-Examination Exhibits 1 through 3, please? 23 CHAIRMAN GLEIMAN: I'll direct that ABA et al. 24 Cross-Examination Exhibits 1, 2, and 3 be transcribed into 25 the record at this point. 19768 1 [Cross-Examination Exhibit Nos. 2 ABA/EEI/NAPM-1 through 3 ABA/EEI/NAPM-3 were transcribed 4 into the record.] 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19772 1 CHAIRMAN GLEIMAN: I think I'd like to take a 2 ten-minute break right now. 3 [Recess.] 4 CHAIRMAN GLEIMAN: Mr. Littell? 5 CROSS-EXAMINATION 6 BY MR. LITTELL: 7 Q Mr. Andrew, are you ready? 8 A Yes, sir. 9 Q In my notice of intent to cross examine you, I 10 asked that you bring with you all your work papers and 11 computer runs relating to Dr. Clifton and MMA Witness 12 Bentley's proposals on automation discounts. Do you recall 13 that? 14 A Yes, sir. 15 Q In response to my request, didn't I meet this 16 afternoon with you and your attorneys to look through all 17 those back-up materials? 18 A Yes, sir, you did. 19 Q Isn't it true that those back-up materials do not 20 include any computer runs or computations concerning Mr. 21 Bentley's testimony or exhibits? 22 A No computer runs, and the only computations were 23 verifications in the margins of his values, that is true. 24 Q Now, your testimony includes, in addition to the 25 written text, three exhibits denominated as MOA -- 19773 1 MOAA-RT-1A through RT-1C. Is that right? 2 A That is correct. 3 Q Exhibit MOAA-RT-1A discusses the testimony of 4 Witness Chown and not Mr. Bentley. 5 A That is correct. 6 Q And Exhibit MOAA-RT-1B also discusses the 7 testimony of Witness Chown and not Mr. Bentley. 8 A That is true. 9 Q And the third and last exhibit, MOAA-RT-1C, 10 compares proposals of Mr. Clifton and Witness Hatfield and 11 not Mr. Bentley. 12 A That's true, but what Mr. Bentley was working on 13 was the same thing as in -- or parts of the same material 14 that this flow chart shows. 15 Q Thank you for that qualification. 16 Now, let's turn to your testimony rebutting MMA 17 Witness Bentley. That testimony begins on page 44 of your 18 rebuttal testimony and continues for three lines on page 45, 19 correct? 20 A That is correct. 21 Q And your entire rebuttal to Mr. Bentley is four 22 paragraphs long. That's correct, too, isn't it? 23 A That is correct. 24 Q Now, isn't the first of those four paragraphs just 25 an introductory description of Mr. Bentley's testimony 19774 1 without any specific criticism? 2 A Yes. 3 Q Now, please look at the second and third 4 paragraphs of your testimony. 5 In those paragraphs, isn't your only criticism of 6 Mr. Bentley that he used the Commission's traditional 7 methodology that classifies mail processing labor costs as 8 100-percent variable instead of the Service's proposed 9 methodology? 10 A That's what the second paragraph says, yes. 11 Q And the third does, too, including your 12 affirmation that you prefer the Service's methodology. 13 A That's correct. 14 Q All right. 15 Now, please turn to page 24 of your testimony. 16 A Yes. 17 Q There, didn't you state four criticisms of Dr. 18 Clifton's adjustments to automation discounts with which you 19 disagree? 20 A Yes. 21 Q Now, looking at lines 5 and 6 on that page, isn't 22 one of your criticisms of Dr. Clifton directed at Dr. 23 Clifton's reduction in first class work-share letter mail 24 processing costs that manifests itself as a reduced 25 roll-forward factor in USPS Witness Hatfield's model? 19775 1 A That's correct. 2 Q Incidentally, did Mr. Bentley recalculate the 3 Service's roll-forward factor? 4 A I do not believe he did. 5 Q Looking at lines 6 and 7 of that page, isn't your 6 second criticism of Dr. Clifton that he proposes a test year 7 reduction in First Class work-sharing letter delivery costs? 8 A Yes. 9 Q Did Mr. Bentley propose any reduction in the 10 Service's deliver costs? 11 A I do not recall precisely. 12 MR. LITTELL: Mr. Chairman, with your permission, 13 I would like to approach the witness and show him some pages 14 from a transcript. And his counsel already has a copy of 15 that. 16 CHAIRMAN GLEIMAN: Please proceed. 17 THE WITNESS: Okay. 18 BY MR. LITTELL: 19 Q Why don't we switch? 20 A Okay. Good. Thank you. 21 Q Looking at transcript page 11236 from Volume 21, 22 didn't Mr. Bentley respond to a Postal Service Interrogatory 23 by conceding that the only change he made in deriving his 24 discount figures was one change from the Postal Service's 25 presentation and that was to assume that labor costs vary 19776 1 100 percent with volume? 2 A Yes. 3 Q So he didn't make any change, did he, in the 4 delivery costs? 5 A Delivery costs. 6 Q Are we agreed that he did not make a change in 7 delivery costs? 8 A Yes. That was my recollection, but I wasn't going 9 to say it without checking. 10 Q Thank you. Now, looking at lines 7 through 8 of 11 your rebuttal testimony on page 24, isn't your third 12 criticism of Dr. Clifton that he proposed a test year 13 increase in the benchmark that Witness Hatfield used to 14 determine cost savings for work-shared discounts? 15 A Yes. 16 Q Did Mr. Bentley propose any increase or change in 17 the benchmark that USPS Witness Hatfield used? 18 A I haven't figured that out. 19 Q Perhaps I can help you. Would you look again at 20 the excerpts from transcript Volume 21 in this case, page 21 11235. Do you see that the Postal Service asked Mr. Bentley 22 whether he didn't in fact use the same bulk metered mail 23 benchmark as did the Postal Service? 24 A But be also conditions that by saying, "For this 25 reason, I did not specifically accept or reject the Postal 19777 1 Service's use of the bulk metered mail as the appropriate 2 benchmark for measuring First Class automated letter cost 3 savings. 4 Q Yes. But wasn't the question, and I quote, 5 "Please confirm that in developing your letter automation 6 proposals, you used bulk metered mail as the benchmark, as 7 did Witness Fronk, in developing the Postal Service 8 proposal." And his answer, to begin with, was, -- 9 A Confirmed. 10 Q -- quote, "Confirmed." close quote. 11 A Confirmed. 12 Q So you would agree, he used the same benchmark? 13 A Yes. 14 Q Looking at line 9 on page 24 of your rebuttal 15 testimony, isn't your fourth criticism of Dr. Clifton based 16 on his reduction in the cost coverage for First Class 17 work-shared letter mail, that you say is based on subjective 18 considerations of efficiency and equity? 19 A I'm sorry. I missed the question. 20 Q Please look -- 21 A Line 9, fourth adjustment -- 22 Q Yes, look at line 9 through line 11 on page 24 of 23 your testimony, beginning with the last two words on line 9. 24 Isn't your, quote, "fourth" criticism of Dr. Clifton based 25 on his reduction in the cost coverage for First Class 19778 1 work-shared letter mail which you say is based on subjective 2 considerations of efficiency and equity? 3 A Yes. 4 Q Did Mr. Bentley derive his proposed discounts by 5 first making a reduction of cost coverage based on 6 subjective considerations of efficacy and equity? 7 A No, he did not. 8 MR. LITTELL: Mr. Chairman, I want to thank 9 Witness Andrew and his counsel for being very cooperative 10 and showing me his workpapers this afternoon and allowing me 11 to confer with him. 12 That concludes my cross examination. 13 CHAIRMAN GLEIMAN: Mr. Baker. 14 MR. BAKER: Thank you, Mr. Chairman. 15 CROSS-EXAMINATION 16 BY MR. BAKER: 17 Q Good evening, Mr. Andrew. Let's turn to page 12 18 of your testimony on the paragraph that appears at the top 19 of that page. Are you there? 20 A Yes, sir. 21 Q And focusing on lines 3 and 4 where you discuss 22 Ms. Chown's use of the piggyback factor to allocate indirect 23 costs to identifiable institutional costs, do you see that? 24 A Yes. 25 Q Are you familiar with the concept of piggyback 19779 1 costs? 2 A Yes. 3 Q Could you state your understanding of them, 4 please? 5 A It's the portion of the costs that are considered 6 volume variable that vary with the labor but they are not 7 labor itself, so it is supervision that changes as the 8 amount of labor required changes and similar kinds of they 9 are costs that are attributable or volume variable, but they 10 are not, excuse me, measured -- I shouldn't say they are not 11 measured. 12 They are added to the labor costs or piggybacked 13 on top of the labor costs. 14 Q They are indirect costs in other words? 15 A Yes. 16 Q Do you know -- 17 A But they are not fixed costs, they are -- 18 Q Do you know whether they include employee 19 benefits? 20 A I believe they do. 21 Q Do they include a portion of costs such as space 22 and utilities that are used to provide the functions that 23 the employee is engaged in? 24 A I believe they do. 25 Q Okay. Let's take a look at mail processing, for 19780 1 example, for each dollar of attributable labor cost of mail 2 processing is it your understanding that the Postal Service 3 also computes a piggyback factor that measures these 4 additional indirect costs that are added to the directly 5 attributable costs? 6 A Yes. 7 Q Okay -- and so let's assume for example that for 8 every dollar of mail processing costs 90 cents are 9 attributed. 10 A Yes. 11 Q For example -- and would each of those 90 cents of 12 direct labor costs cause the Postal Service to incur these 13 indirect supervisory and employee benefit and other costs? 14 A Well, the 90 cents wouldn't but the labor that's 15 associated with that 90 cents would. 16 Q Now let's consider the remaining 10 cents of labor 17 costs that are deemed institutional. 18 Does the mail processing employee get paid 19 benefits on the basis of his or her salary -- full salary -- 20 or only based upon the portion of his efforts that are 21 attributable? 22 A Full salary, I believe. 23 Q Okay, and does his or her supervisor supervise the 24 mail processing employee only when the employee is 25 performing work that can be attributed or does the 19781 1 supervisor supervise the employee at other times as well? 2 A I don't know. 3 Q If the portion of mail processing costs that were 4 attributable declined so that 80 cents out of every dollar 5 is attributable instead of the 90 cents we assumed 6 previously, but total labor costs remained the same, would 7 the cost of the employee benefits decline? 8 A No. 9 Q Do you know whether the amount of supervisory time 10 would decline? 11 A I don't know. 12 Q Okay. Thank you. 13 Mr. Andrew, could you please turn to your Exhibit 14 MOAA et al. RT-1B. 15 A Page? 16 Q 1. Do you have it? 17 A Yes, sir. 18 Q At the -- let's see -- do you show here a 19 calculation of institutional cost contributions using a 20 marginal cost metric and assuming equal markups at the top 21 part of that exhibit? 22 A Yes. 23 Q And setting aside the assumed equal markups, is it 24 your testimony that the method shown in the top half of this 25 exhibit matches the current method of assigning 19782 1 institutional costs? 2 A Subject to your caveat that it's not equal 3 markups, it's up to these gentlemen here to make the 4 markups, yes. This was just an illustrative example. 5 Q I understand. Right. And under the current 6 method the Commission would look at the attributable costs, 7 select what it believes are reasonable markups, and apply 8 those markups to the attributable costs. Correct? 9 A That's my understanding. Yes. 10 Q Okay. And if the Commission were to do that in 11 this example here on the page of your exhibit, it would 12 assign the total amount of institutional costs which are if 13 I am correct 150? 14 A Yes, sir. 15 Q Okay. Now directing your attention to the bottom 16 half of the exhibit, here you show how the institutional 17 cost contributions are calculated using the Chown method as 18 drawn from her testimony. Correct? 19 A Yes. 20 Q And in this example using weighted attributable 21 costs once again -- well, have you applied equal markups 22 here as well? 23 A Yes. 24 Q And under the proposed Chown method that you used 25 here, is the sum of the weighted attributable costs equal to 19783 1 the sum of the unweighted attributable costs? 2 A It is forced to do that by the nature of her 3 metric. 4 Q Right. 5 A As we proved in the -- 6 Q Yes. So under the -- 7 A Exhibit. 8 Q Proposed Chown method the Commission would 9 continue to assign the total amounts of institutional costs 10 which in this example still come to 150; correct? 11 A Yes. 12 Q Okay. So in each method the Commission would 13 decide on the markup and assign total amount of 14 institutional costs based on its consideration of the 15 relevant factors; correct? 16 A Starting from what point? 17 Q Well, starting from either point. Wouldn't the 18 Commission -- 19 A Okay. That's what I wanted to make sure that we 20 understood, that one of them would be starting from true 21 marginal cost or a proxy, namely attributable costs. The 22 other one would be this weighted attributable costs. 23 Q Okay. 24 A That we don't know what it is. 25 Q Could you please turn to page 17 of your 19784 1 testimony? 2 A Yes. 3 Q I have a question on table two that appears on 4 that page. Is this a modification of one of Ms. Chown's 5 examples? 6 A Yes. We added $100 of system-wide institutional 7 costs to see what impact it would have. That's the only 8 modification. 9 Q Is this based on -- I hate to flip you back and 10 forth -- back to Exhibit RT-1-B, is this based on page two 11 of that exhibit? 12 A Yes. 13 Q Could we compare this, back on page 17, table 2, 14 with table one on page 15, to see how the institutional cost 15 contributions change as a result of the addition of the $100 16 of system-wide institutional costs that you have added to 17 table two? 18 A I'm sorry. You are fading, sir. 19 Q Can I compare table two on page 17 to table one on 20 page 15 to see how the institutional cost contributions 21 change as a result of your adding $100 of system-wide 22 institutional costs? 23 A Yes. 24 Q Previously, I had distributed to your counsel a 25 proposed cross examination exhibit by the cumbersome name of 19785 1 NAA/MOAA et al-RT-1-XE-1. Did you receive that? 2 A Yes, I have that. Comparison of institutional 3 cost contributions? Base case, case one? 4 Q Yes, sir. 5 [Cross-Examination Exhibit No. 6 NAA/MOAA et al.-RT-1-XE-1 was 7 marked for identification.] 8 MR. BAKER: Thank you, Mr. Littell, for your 9 assistance. 10 BY MR. BAKER: 11 Q Referring you to my exhibit that was just handed 12 to you, at the top of this page, this exhibit shows 13 institutional cost contribution for each class of mail under 14 the base case of your Exhibit RT-1-B, using the marginal 15 cost method and the Chown method; correct? You might take a 16 look at our Exhibit 1-B, page one. 17 A I'm sorry. I missed the question. 18 Q Could you turn to Exhibit 1-B, page one of four? 19 A Yes. 20 Q Where you have behavioral characteristics of the 21 Chown metric base case. 22 A Sir, I'm sorry. I cannot -- I'm a little bit hard 23 of hearing. 24 Q I apologize. 25 A I'm sorry. 19786 1 Q Looking at Exhibit RT-1-B, page one, which is the 2 base case, and my question is whether you are able to verify 3 that the mark up or institutional cost contribution figures 4 on my cross examination exhibit correspond with those 5 provided on your Exhibit 1-B, page one? 6 A Yes, they do. 7 Q Is one difference that my exhibit uses percentages 8 and yours uses cost coverages? 9 A Yes. Well, yours uses percentage of the -- 10 Q I stand corrected. 11 A -- of the marginal costs or the attributable 12 costs. 13 Q I have shown on my cross examination exhibit the 14 percentage of institutional costs born by each sub-class of 15 mail under both the marginal cost method and the Chown 16 method; correct? 17 A Yes. 18 Q Again, assuming equal mark up's? 19 A Yes. 20 Q Now, on the bottom half of the cross examination 21 exhibit labeled case one, it shows the institutional cost 22 contributions assuming an additional $100 of system-wide 23 institutional costs; correct? 24 A Yes. 25 Q And do these figures correspond with those 19787 1 provided in your Exhibit 1B, page 2? 2 A Yes, they do. 3 Q And, again, have I shown on my cross-examination 4 exhibit, the percentage of institutional cost contributions 5 borne by each subclass of mail under each method? 6 A Yes, except your Chown method, the percentage 7 totals 100, not zero. 8 Q I accept that correction. Now, I notice that 9 under each method, each -- on my cross-examination exhibit, 10 each subclass pays the same percentage of the institutional 11 costs both before and after the additional $100 of 12 institutional costs, is that correct? 13 A That is correct. 14 Q So does this exhibit show that the use of either 15 method, with equal markups, results in no change in the 16 share of institutional costs borne by each subclass of mail, 17 is that correct? 18 A That is correct. With the assumptions that this 19 example has. 20 Q Now, if we compare my cross-examination back to 21 your Table 2 on page 17, on Table 2 has your -- does Table 2 22 show a change in the markups or coverages of the classes of 23 mail after the system-wide increase of $100 of system-wide 24 institutional costs? 25 A Yes. 19788 1 Q Okay. And those markups, those cost coverages 2 differ, correct? 3 A That is correct. 4 Q Okay. Is it one of the points of Ms. Chown's 5 testimony that markups based on unweighted attributable 6 costs may be misleading, at least in her opinion? 7 A I think that's one of the things she is concerned 8 about, yes. 9 Q Now, I would like to move to a slightly different 10 example that I have worked up in my next cross-examination 11 exhibit which has the similarly awkward label, NAA/MOAA et 12 al.-RT-1-X-XE2. 13 A Is that page 1 of 2, or page -- 14 Q It's a two page exhibit. 15 A Okay. 16 MR. BAKER: I need to make sure Mr. Todd has his. 17 [Cross-Examination Exhibit No. 18 NAA/MOAA et al.-RT-1-XE-2 was 19 marked for identification.] 20 BY MR. BAKER: 21 Q Do you have that before you, Mr. Andrew? 22 A Yes, sir. 23 Q Okay. This is a two page exhibit in which the 24 second page attempts to show how some of the numbers on the 25 first page were derived. Were you able to follow that? 19789 1 A Yes, sir. 2 Q Okay. Now, is this a variation of your Case 1 3 from RT-Exhibit 1B, page 1, with the difference that the 4 increased $100 of institutional costs are directly related 5 to the provision of function 2? 6 A Yes. 7 Q And so this is a bit different from your Case 1? 8 A Yes. 9 Q Have you reviewed the figures in this Exhibit 2 10 and the related page in which we derive these results using 11 the Chown method? 12 A Yes. 13 Q Are the figures calculated correctly? 14 A I believe so. 15 Q And referring to page 1 of that exhibit, Exhibit 16 2, I notice that under the marginal cost -- 17 A Would you hold -- 18 Q Page 1. 19 A No, would you hold just a minute? 20 Q Oh, excuse me. 21 A I'm ready, sir. Thank you. 22 Q And I notice that on page 1 of Exhibit 2 under the 23 marginal cost method the rates for each subclass rise by 25 24 percent compared to the base case when the additional $100 25 of institutional costs were incurred associated with 19790 1 Function 2, is that correct? 2 A Yes. 3 Q And I notice that under Ms. Chown's method the 4 rates for Class A rise 25 percent and the rates for Class C 5 rise by 45.5 percent but the rates for Class B remain 6 unchanged, is that correct? 7 A That's correct. 8 Q Now is the reason why the rates for Class B remain 9 unchanged under Ms. Chown's method that the institutional 10 costs we have added are related to the provision of Function 11 2 only and Class B does not use Function 2? 12 A Mathematically that is what happens, yes. 13 Q Referring you to the marginal cost side of this 14 exhibit, do you believe it is equitable for Class B to pay 15 25 percent more to cover an increase in cost for a function 16 that the class does not use? 17 A Yes. When you have economies of scale and scope, 18 yes. 19 Q Do you regard this as an area where you and Ms. 20 Chown will simply disagree? 21 A Yes. 22 Q And I will not be able to talk you out of your 23 position tonight? 24 A I doubt it. It's early in the evening though. 25 [Laughter.] 19791 1 BY MR. BAKER: 2 Q All right. Could you turn now to page 20 of your 3 testimony and look at Table 5? 4 A Yes? 5 Q This changes the hypothetical analysis that we 6 have been following because here you show the effect of a 7 reduction of attributable costs, correct? 8 A Yes. 9 Q Okay, and that reduction occurs because our Class 10 A engages in work-sharing in Function 2, correct? 11 A That's correct. 12 Q And are we still using an assumption of equal 13 mark-ups? 14 A Yes, we are. 15 Q And just as an aside, can we agree that that is as 16 an unreasonable assumption now as it was all along? 17 A Yes, but it's something to look at. I have done 18 these with varying rates but it gets messy, so -- 19 Q Okay, and I notice that under your marginal cost 20 method on the left side of Table 5 the rates for Class B 21 increased by 4 percent because Class A has changed its use 22 of Function 2, is that correct? 23 A That is correct. 24 Q Now in this hypothetical example, Class B is the 25 class that does not use any of Function 2, correct? 19792 1 A That's correct. 2 Q Why should the rates for Class B be affected in 3 any way by the use Class A makes of Function 2? 4 A Because it was changing the overall institutional 5 cost of the system. 6 Q How are we changing the institutional costs of the 7 system? 8 A Because there -- well, I should say we are 9 changing the base on which they're distributed. We are 10 reducing the amount of attributable cost in the system. 11 Q Right. The change from this example -- the 12 difference between this table and the ones we have discussed 13 previously is this one we have reduced attributable costs in 14 the system? 15 A That's right. 16 Q And as a consequence of this reduction of 17 attributable costs, Class B sees a rate increase, is that 18 correct? 19 A Yes. 20 Q In a sense is class B being made to pay a share of 21 the institutional costs associated with function 2 that 22 class A is no longer paying? 23 A Yes. 24 Q Okay. And under Ms. Chown's method on the right 25 side of table 5 of your page 20, I see that if class A 19793 1 reduces its use of function 2, the rates for class B are not 2 affected. Is that correct? 3 A That's correct. 4 Q And that's because -- again because class B does 5 not use function 2. Correct? 6 A That's correct. But you also have a major 7 increase in class C, and even a more drastic decrease in 8 class A, which was what she was trying to improve upon. 9 Q Well, now, class C, after this reduction in 10 attributable costs, isn't it true that class C will be the 11 heaviest user of function 2? 12 A Yes. 13 Q Now we've walked through a comparison of the Chown 14 metric and the marginal cost metric in a number of examples. 15 And I notice that you -- well, on page 23 of your testimony 16 you conclude that the marginal cost metric responds to 17 changes in a reasonable and predictable manner. Is that 18 correct? 19 A Yes. 20 Q And you contend that the Chown approach does not? 21 A It does not. 22 Q For each of the changes we have discussed, do you 23 think one could have predicted the results that have been 24 achieved using the Chown method? 25 A Not the major variations that have occurred; no. 19794 1 Q You don't think so? Well, that may be another 2 area where you and she will have to agree to disagree? 3 A I think so. 4 Q Okay. 5 A Because through time it would be very difficult to 6 track these as changes occur in the base attributable costs 7 and the institutional costs. 8 Q All right. Could you turn now, shifting topics, 9 to page 37 of your testimony, where you have some testimony 10 regarding Dr. Clifton. 11 And just quickly I'd like to refer your attention 12 to lines 15 and 16. 13 MR. BAKER: Actually, Mr. Chairman, I suppose 14 before I do that I should move into the record my 15 cross-examination exhibits. 16 CHAIRMAN GLEIMAN: Do you want them transcribed? 17 MR. BAKER: Yes, sir. 18 CHAIRMAN GLEIMAN: Are you going to provide copies 19 to the reporter? 20 MR. BAKER: I believe -- has Mr. Littell done 21 that? 22 CHAIRMAN GLEIMAN: Can I ask you a question with 23 respect to Cross-Examination Exhibit No. 1? 24 MR. BAKER: Yes. 25 CHAIRMAN GLEIMAN: My recollection is that the 19795 1 witness pointed out that there may have been an error in the 2 last -- 3 MR. BAKER: Yes. 4 CHAIRMAN GLEIMAN: Number on the bottom right-hand 5 side. Did you correct that, or we'll just assume that it's 6 corrected in the transcript? 7 MR. BAKER: It has been corrected in the 8 transcript, but I would be happy to mark it if -- 9 CHAIRMAN GLEIMAN: I think that that would make 10 the record clearer. 11 And with that change on Cross-Examination Exhibit 12 No. 1, then we'll have Cross-Examination Exhibit No. 1 and 13 Cross-Examination Exhibit No. 2 transcribed into the record 14 at this point. 15 MR. BAKER: Thank you. 16 CHAIRMAN GLEIMAN: There was no motion to 17 introduce? 18 MR. BAKER: Oh, I'm sorry. I'd like to move them 19 as evidence as well. 20 CHAIRMAN GLEIMAN: Without objection, transcribed 21 and entered into evidence. 22 [Cross-Examination Exhibit Nos. 23 NAA/MOAA et al.-RT-1-XE-1 and 24 NAA/MOAA et al.-RT-1-XE-2 were 25 received into evidence and 19796 1 transcribed into the record.] 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19800 1 MR. BAKER: My apologies, Mr. Chairman. 2 Back to page 37 of the witness' testimony. 3 BY MR. BAKER: 4 Q You state there by definition cost coverage for a 5 given subclass of mail is the ratio of revenue to volume 6 variable cost. That's not exactly correct the way the 7 Commission has done it, is it? 8 A Well, are you quibbling with volume variable 9 costs? 10 Q Yes. They use -- 11 A It's attributable cost, yes, right. 12 Q Which is not under the Commission's practice 13 necessarily the same as volume variable. 14 A Not exactly. 15 Q Okay. To the next page, page 38, and beginning in 16 this section but in the text at line 21 you are discussing 17 the arithmetic phenomenon that cost coverages for mail 18 increase as work-sharing increases. And I want to draw your 19 attention to the sentence beginning on page 23, where you 20 state that this increase in cost coverage for First Class 21 work-shared mail is not an issue of equity and efficiency as 22 suggested by Witness Clifton. Rather, it's a matter of 23 arithmetic. 24 A Where are you? 25 Q Page 38, line 23. I'm sorry. 19801 1 Do you see that sentence? 2 A Yes. 3 Q Okay. Now Dr. Clifton's testimony contended that 4 the second and third ounces of work-shared First Class mail, 5 that the current extra ounce charge works out to a markup 6 of -- or cost coverage of approximately 920. Do you recall 7 that part of his testimony? 8 A Yes. That was using his cost. 9 Q His analysis. 10 A His analysis of cost. 11 Q Yes. That is -- do you have an opinion of whether 12 when a cost coverage approach is 920 there, an equity issue 13 may arise? 14 A Generally, the cost coverages are not used except 15 at the class or sub-class level. 16 Q You would say there is no equity issue there? 17 A I don't have an opinion. 18 Q Do you think that when a cost coverage approach is 19 920 in the context in which Dr. Clifton discussed it, any 20 efficiency issue arises? 21 A Not necessarily. It depends on the conditions. 22 Remember, that 920 comes out of using his costs and the 23 current rates. 24 Q Is there any percentage point or cost coverage 25 level at which you would begin to be concerned that an 19802 1 equity or efficiency issue does arise? 2 A Yes, it would be contextual though. 3 Q Contextual. Can you give me a number? 4 A No, not without the context. 5 Q It's your testimony that in the context in which 6 Dr. Clifton calculates the 920, that there is not an equity 7 or efficiency issue; is that correct? 8 A I can't agree with the 920 because of his 9 calculation of costs. 10 Q If his 920 were correctly calculated, would you 11 think there was an equity or efficiency issue? 12 A I would be concerned if I was him, yes. 13 Q Would you be concerned if you were you? 14 A Yes. 15 Q Turn to page 42 of your testimony. At the section 16 beginning under the caption subheading B, you are 17 criticizing here Dr. Clifton's use of the term "incremental 18 costs and revenues" in the context of the proposal; correct? 19 A Yes, sir. 20 Q I will agree with you that the second ounce of a 21 piece of first class mail is indeed part of a two ounce 22 piece of mail, and it's not a separate thing. However, the 23 second ounce does pay a separate charge, 23 cents, that is 24 above that paid for an one ounce piece. Can one think of 25 that 23 cents charge as an incremental charge or a marginal 19803 1 charge? 2 A No, sir. 3 Q You would not? 4 A No, sir. 5 Q If I increased the weight of a mailing from first 6 class letter from one ounce to two ounce and I go to the 7 Post Office and discover that I must add a 23 cents stamp, I 8 have not incurred an incremental charge? 9 A You now have a two ounce letter instead of an one 10 ounce letter. These incremental charges are product -- 11 incremental costs are product related. You have to consider 12 the entire product. The method of Clifton's computing 13 incremental costs, as I've shown on table 15 on page 43, 14 when he's making a comparison with standard A, will never 15 ever pass the cross subsidy test because he's got it loaded, 16 because in the case of standard A mail, the incremental 17 revenue, if you like, of that second ounce, we don't charge 18 any more. The first ounce picks up the entire cost. No 19 matter how much difference between revenue and costs there 20 are in standard A, it would always show a subsidy, the way 21 he has used to calculate it. 22 Q You are now talking about subsidy. I was asking 23 still back on whether the charge for the second ounce is an 24 incremental revenue or not. 25 Is an one ounce letter a different product from a 19804 1 two ounce letter? 2 A Yes. 3 Q Will you accept, subject to check, that a first 4 class carrier route automation one ounce letter today in the 5 current rates is charged 23 cents a piece? 6 A Yes, subject to check. 7 Q And can you confirm that the second, the extra 8 ounce rate for that piece would be 23 cents, or would you 9 need to take that subject to check? 10 A No, that's accepted. 11 Q That you know. Okay. So this is an instance 12 where the rate of piece doubles with its weight, correct? 13 A Yes. 14 Q So when the First Class letter that we just 15 discussed moves from its one ounce to -- it got heavier, it 16 became a two ounce piece, it paid an extra ounce of 23 17 cents. That's an incremental revenue, if you will, when you 18 go from the once to the two ounce. Is that correct? 19 A No. 20 Q No? There is not an incremental revenue if you go 21 -- 22 A It's a different product. 23 Q It's a different product you say. Is there, at 24 least in theory, a marginal or cost difference between the 25 one ounce piece and the two ounce piece? 19805 1 A There is a cost difference, in theory. 2 Q Now, a Standard A mailer can not mail a second 3 ounce without mailing the first ounce as well, correct? 4 A Correct. 5 Q Okay. And is it your testimony that the one ounce 6 Standard A piece and the two ounce Standard A piece are 7 different products? 8 A Yes. 9 Q And because of the difference in the rate 10 schedules between First Class and Standard mail, when the 11 First Class mailer moves to a two ounce piece, he pays an 12 extra 23 cents, but while the Standard A moves to a two 13 ounce piece, he will continue to pay the same rate as he 14 pays for the one ounce piece, is that correct? 15 A That is correct. 16 Q Could you -- 17 MR. BAKER: I have one more line of questions and 18 then I am done, Mr. Chairman. 19 BY MR. BAKER: 20 Q Dr. -- or Mr. Andrew, could you turn to page, back 21 to page 40? 22 A Yes. 23 Q And here you will educate me. Because on line 17, 24 you use a ratio of revenue to cost to test cross-subsidy. 25 Do you see that? 19806 1 A Yes. yes. 2 Q Why do you use the ratio rather than simply the 3 difference between revenue and cost? 4 A Because it gives me an idea statistically of how 5 much error I can have in the data that is going into this 6 thing and still be free of subsidy? 7 Q And then can you tell me why the ratio minus one, 8 expressed as a percent, is positive, indicates that the 9 amount can be tolerated. Is that a statistical concept? 10 A Yes, it's a statistical concept. 11 MR. BAKER: Mr. Chairman, that concludes my 12 questions. 13 CHAIRMAN GLEIMAN: Is there any follow-up? 14 MR. COOPER: Yes, Mr. Chairman, I have a few 15 questions. 16 CHAIRMAN GLEIMAN: Mr. Cooper. 17 CROSS-EXAMINATION 18 BY MR. COOPER: 19 Q Would you refer to Mr. Baker's Cross-Examination 20 Exhibit Number 2? I'll skip the lengthy title. It's the 21 two page exhibit. 22 A Yes, sir. 23 Q Now would you agree with me that in this example 24 Class B does not cause the cost, the additional cost 25 associated with Function 2? 19807 1 A Yes, I would agree. 2 Q Would you also agree with me that Class A doesn't 3 cause the additional cost associated with Function 2? 4 A Yes. 5 Q And similarly, Class C does not cause them? 6 A Yes. 7 Q I think you mentioned along the way that you were 8 not troubled by the fact that under the marginal cost method 9 each of these three classes bears some of the burden of 10 these additional institutional costs, is that correct? 11 A That is correct. 12 Q And you mentioned that the presence of economies 13 of scope was one reason why you weren't troubled by that 14 outcome, is that right? 15 A Yes, and scale also. 16 Q With respect to the economies of scope, are you 17 saying that Class B benefits from the presence of Classes A 18 and C? 19 A Yes. 20 Q How does it benefit? 21 A Well, as a practical matter, it benefits from the 22 overall organization and the functions that it does share -- 23 the functions that are shared. 24 Now this is a toy example when you look at the 25 entire spectrum of the Postal Service. You have a lot of 19808 1 economies of scope with products, with the multiproduct 2 production. 3 Q And you also mentioned economies of scale a few 4 moments ago. 5 A Yes. 6 Q How did the economies of scale enter into this? 7 A Because of the more overall activity. 8 Q Does Class B benefit from the economies of scale? 9 A Yes. 10 Q How does it benefit? 11 A If there were less units of A and C, for example, 12 it's likely that one or more of the functions could not 13 afford to have as much automation. That would be an 14 example. 15 Q So you are saying that the unit costs of Class B 16 are affected by the economies of scale? 17 A Yes. 18 MR. COOPER: I have no further questions. 19 CHAIRMAN GLEIMAN: Mr. Baker? 20 FURTHER CROSS-EXAMINATION 21 BY MR. BAKER: 22 Q Were you in the exchange with counsel for the 23 Postal Service or -- when you referred to the unit costs of 24 B, were you referring to attributable costs or total -- you 25 know, basically the price of B? 19809 1 A Would you restate it? 2 Q Yes. Where you just referred to unit costs of B. 3 Is that attributable cost? 4 A Yes, attributable cost. 5 MR. BAKER: I think I will stop here, Mr. 6 Chairman. 7 CHAIRMAN GLEIMAN: Is there any further follow-up? 8 There are no questions from the bench. 9 That brings us to redirect. Mr. Todd, would you 10 like an opportunity to consult with your witness? 11 MR. TODD: Yes, please, Mr. Chairman. 12 CHAIRMAN GLEIMAN: All right. Ten minutes? 13 MR. TODD: That should be ample -- or at least 14 sufficient. 15 [Recess.] 16 CHAIRMAN GLEIMAN: Yes, sir. 17 MR. TODD: Mr. Chairman, and not just because only 18 you of the Commissioners have come back, we have no 19 redirect. 20 CHAIRMAN GLEIMAN: I hadn't noticed. That's what 21 happens when you control the buzzer. They never know when 22 it's going to happen. 23 If that is the case then, Mr. Andrew, that wraps 24 it up for you tonight. We appreciate your appearance here 25 today and your contributions both the first time around and 19810 1 this time for our record, and if there's nothing further, 2 you're excused. 3 THE WITNESS: Thank you. 4 [Witness excused.] 5 CHAIRMAN GLEIMAN: Our last and very patient 6 witness, appearing on behalf of CTC Distribution Services, 7 Mr. Clark, is already under oath in this proceeding, and, 8 Mr. Olson, if you want to introduce your witness and enter 9 his rebuttal testimony, we can hopefully move right along. 10 I notice that after counsel finishes with their 11 last witness of the day they seem to move out of the room 12 faster than they move in. I don't understand that. 13 Mr. Olson, whenever you and your witness are 14 ready. 15 MR. OLSON: Mr. Chairman, we would call John Clark 16 to the stand, and I guess we can proceed. 17 Whereupon, 18 JOHN L. CLARK, 19 a witness, was called for examination by counsel for CTC 20 Distribution Services, L.L.C. and, having been previously 21 duly sworn, was examined and testified as follows: 22 DIRECT EXAMINATION 23 BY MR. OLSON: 24 Q Mr. Clark, I'd like to present you two copies of 25 what has been entitled the Rebuttal Testimony of John L. 19811 1 Clark on Behalf of CTC Distribution Services, LLC, 2 designated CTC-RT-1, and ask you if you can review those and 3 tell us whether they were prepared by you or under your 4 supervision. 5 A They were prepared under my supervision. 6 Q And do you have any changes to that testimony? 7 A Upon review I'd like to eliminate or delete one 8 sentence from the testimony. It's on page 8, lines 7 and 8. 9 UPS -- it says in quotes, UPS has offered no sound basis for 10 questioning the cost estimates in this proceeding, end of 11 quote. I feel that might have overstated my position with 12 regard to cost estimates, so I'd like to delete that 13 sentence. 14 Q And with that change, do you adopt this as your 15 testimony? 16 A Yes, I do. 17 MR. OLSON: Mr. Chairman, we'd move the admission 18 of this testimony into evidence. 19 CHAIRMAN GLEIMAN: Are there any objections? 20 [No response.] 21 CHAIRMAN GLEIMAN: Hearing none, Mr. Clark's 22 testimony and exhibits are received into evidence, and I 23 direct that they be transcribed into the record at this 24 point. 25 [Rebuttal Testimony and Exhibits of 19812 1 John L. Clark, CTC-RT-1, was 2 received into evidence and 3 transcribed into the record.] 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19823 1 CHAIRMAN GLEIMAN: The only participant who has 2 requested oral cross-examination of Witness Clark is the 3 United Parcel Service, and there's nobody else left in the 4 room. So I think -- except for OCA and the Postal Service, 5 of course. 6 MR. OLSON: And the Commission. 7 CHAIRMAN GLEIMAN: I suspect we can proceed with 8 cross-examination at this point, Mr. McKeever. 9 THE WITNESS: Before we begin I'd like to thank 10 Mr. McKeever and the Commission for allowing me to change my 11 date of appearance to accommodate my personal schedule. I 12 appreciate it very much. 13 CHAIRMAN GLEIMAN: Mr. Clark, you were one of the 14 easy ones when it came to changing schedules. 15 CROSS-EXAMINATION 16 BY MR. McKEEVER: 17 Q Mr. Clark, could you turn to page 6 of your 18 testimony, please? 19 A Rebuttal testimony. 20 Q Yes, your rebuttal testimony. And I'd like you to 21 take a look at lines 15 to 20. 22 A Okay. 23 Q There you state that while you have not received 24 any formal education as an economic or financial analyst, 25 you believe that much of what UPS Witness Luciani has said 19824 1 about the Postal Service's cost estimates is, to use your 2 words, subject to question. Is that right? 3 A Yes. 4 Q You've given as an example there his statement 5 that 100-percent passthroughs of estimated avoided costs 6 assume perfect execution, which to use your words is 7 impossible to attain. Is that right? 8 A Yes. 9 Q You then say that, assuming that is so, could not 10 the same be said about every cost estimate? Is that 11 correct? 12 A Yes. 13 Q So I take it that your testimony is that every 14 cost estimate is uncertain, or do you believe that the 15 Postal Service's avoided cost estimates are 100-percent 16 accurate? 17 A Well, since this case began, I have received all 18 of the testimony of all of the witnesses, I've done my best 19 to review that, I've looked at mostly those cases that have 20 to do with anything related to parcels and the things that 21 I'm interested in. But this whole proceeding by its very 22 nature is kind of a war of estimates, even tonight as we sit 23 here listening to the various parties go back and forth 24 interpreting the meaning of statistical data, the 25 applications of statistical data and so forth. So it's up 19825 1 to the Commission apparently to decide who has the greatest 2 ability to accurately portray the reality of the situation 3 you're dealing with. 4 I found in Luciani's testimony a number of 5 instances that didn't comply with what I believe to be 6 reality, and that's why I thought this rebuttal testimony 7 would just point out a few of those. 8 Now I presume that the Commission's role in these 9 rate hearings is to decide who has the most valid analysis 10 in its application. Now the Postal Service has submitted, 11 and I have read the information that they have submitted, 12 and they really have a better track record than UPS 13 witnesses have all the way going back to 1990. 14 I know I'm kind of getting off the track here, but 15 I just wanted to kind of put my, you know, where I'm coming 16 from in putting this together. 17 Q I think you did that in your rebuttal testimony, 18 but I'd like to get down to some specifics. 19 A Okay. 20 Q Do you believe that the Postal Service's avoided 21 cost estimates are 100-percent accurate? 22 A It seems to me they have the best chance of being 23 accurate because they are the ones that are operating the 24 system, and I guess I find it very difficult to believe I 25 read for example Luciani says in his testimony that he 19826 1 visited an ASF, he visited this place and that place, but 2 how could his ability surpass the ability of the Postal 3 Service to really know what their costs are? 4 Q So your basis for your testimony is that the 5 Postal Service must know more than anybody else about it and 6 so that therefore when it makes a cost estimate, that should 7 be accepted? 8 A Well, the whole approach of Luciani is to create 9 an air of uncertainty and that is the basis for him, for 10 example, saying that we don't think you can execute this to 11 perfection and we think that parcels that come in from DBMC 12 are less dense, we think that transportation costs are 13 affected by DBMC. 14 All of these things are designed to create an air 15 of uncertainty that would cause the Rate Commission to 16 recommend some different kind of an approach. 17 I say I believe that the Postal Service is more 18 certain of their proposals. 19 Q Well, do you believe that there is certainty about 20 their avoided cost estimates? 21 A I would say that if I had to -- if I were sitting 22 in a seat up here I would be more inclined to accept their 23 proposals than any others that I have seen. 24 Q And that is because they are the Postal Service, 25 right? 19827 1 Is that what you said a few minutes ago? 2 A Well, assuming that they are not dishonest, which 3 I think is appropriate -- 4 Q They can't make mistakes? 5 A -- that they make an honest approach to it. They 6 are more qualified and have a better record of performance 7 than anybody that I have seen in this case. 8 Q Is it your testimony that they can't make 9 mistakes? 10 A Absolutely not. In fact, in some of the rebuttal 11 testimony they have come forth and said, you know, here is 12 where we made a mistake in the past and here is how we are 13 correcting it. 14 Now Luciani doesn't bother to do that, by the way. 15 I mean he made the most colossal mistake you could imagine 16 in 1990 with Witness Hall, talking about, you know, that the 17 DBMC is a total disaster -- 18 MR. McKEEVER: Mr. Chairman, I move to strike any 19 remarks about Mr. Luciani in R90. He was not even a witness 20 in that case. 21 THE WITNESS: But he put into his 22 qualifications -- 23 MR. McKEEVER: It is nonresponsive, Mr. Chairman, 24 and it is certainly inaccurate. Mr. Luciani wasn't even a 25 witness in that case. 19828 1 I move to strike. 2 MR. OLSON: Mr. Chairman, I think it is 3 responsive. I think counsel asked the question and is stuck 4 with the answer and it certainly is that Witness Luciani has 5 testified that he assisted with -- 6 CHAIRMAN GLEIMAN: Was Mr. Luciani a witness in 7 R90? 8 MR. OLSON: He testified that he assisted the UPS 9 in presenting and preparing their testimony. 10 MR. McKEEVER: Mr. Chairman, may I also mention 11 that the question was is the Postal Service capable of 12 making a mistake. 13 CHAIRMAN GLEIMAN: As everybody who has been 14 around these proceedings knows, the Commission is generally 15 reluctant to strike and I will deny the motion to strike in 16 this case. 17 However, we will read the testimony and it will be 18 given the appropriate weight in light of the facts of the 19 R90 case. 20 Now I am going to ask the witness to -- I know he 21 has strongly held views. All of us probably do on one 22 subject or another, but I think it is important to try and 23 confine your responses, to the extent practicable, to the 24 questions that are asked by counsel who is doing the cross 25 examining. 19829 1 There is an opportunity under redirect if your 2 counsel and you decide that there is other information you 3 want to get in that follows up on cross examination, and I 4 think that is the appropriate place for it to come. 5 THE WITNESS: I apologize. 6 CHAIRMAN GLEIMAN: There is no need for an 7 apology. This is -- you know, this goes on in here. It's 8 not only a war of estimates, it's a war of a lot of 9 different natures, and, you know, we'll sort it all out and 10 hopefully come up with something that makes sense to most 11 people before it is all over. 12 Could you please continue, Mr. McKeever? 13 MR. McKEEVER: Thank you, Mr. Chairman. 14 BY MR. McKEEVER: 15 Q Mr. Clark, in fact, didn't you testify that in the 16 real world, cost estimates rarely hit the mark? 17 A Well, an estimate by definition is an estimate. 18 Q And they rarely hit the mark; isn't that correct? 19 Isn't that what you stated in your testimony? 20 A That would be a fair statement; yes. 21 Q On page seven at lines seven to nine, you refer to 22 Mr. Luciani's testimony, that the DBMC entry savings 23 estimated by the Postal Service are high, and you state 24 there that he offers no contrary proof or explanation. Do 25 you see that? Page seven, lines seven to nine. 19830 1 A Yes, I see it. 2 Q You cite two pages of Mr. Luciani's testimony in 3 support of that statement. Did you read all of his 4 testimony? 5 A Well, I know he put in a couple of supplements. 6 He had his original and then he had his corrections that 7 came later on. 8 Q Did you read all of his direct testimony? 9 A I think so; yeah. 10 Q Now, did you read his testimony -- matter of fact, 11 Mr. Chairman, may I approach the witness and furnish him a 12 copy of the transcript in which Mr. Luciani's testimony is 13 reproduced? 14 A I think I have that. 15 Q You do have it? Okay. Thank you. Did you read 16 his testimony -- you do not have the transcript version; is 17 that correct or do you? 18 CHAIRMAN GLEIMAN: It appears as though the 19 witness has the version that was submitted and not the 20 transcript version. 21 BY MR. McKEEVER: 22 Q Did you read his testimony beginning on page seven 23 as it was originally submitted and now appears at transcript 24 page 14293, that the Postal Service double counted platform 25 acceptance costs in estimating the costs avoided by DBMC? 19831 1 Did you read that? 2 A Yes, I read it at some time. 3 Q Did you read his testimony on pages 12 to 14, 4 which appears in the transcript at pages 14298 through 5 14300, that the Postal Service overestimated the costs 6 avoided by DSCF by overstating the number of parcels that on 7 average on in a sack or in a general purpose mail container? 8 Did you read that? 9 A Yes. 10 Q Did you read his testimony on page 15, which 11 appears at transcript page 14301, that the Postal Service 12 overstated DSCF avoided costs by assuming that Postal 13 Service personnel would not assist drop shipping mailers in 14 unloading drop shipments? Did you read that testimony? 15 A You're on page 15? 16 Q I'm on page 15, yes. 17 A Yes. 18 Q In fact, you testified when you were here 19 previously, at transcript page 10195, that when CTC drop 20 ships to a BMC or, in fact, to a SCF, you testified at page 21 10194, the Postal Service employees unload the vehicle, 22 assisted by the driver when requested, isn't that correct? 23 A Yes. I might comment, I am not sure -- 24 MR. McKEEVER: Mr. Chairman, I am going to move to 25 strike if -- he has responded to the question. He has 19832 1 completed an answer. An appreciable amount of time passed. 2 And if we are going to get out of here at a reasonable hour 3 tonight, -- 4 CHAIRMAN GLEIMAN: Well, I am not all that 5 concerned about the reasonableness of the hour, and I can be 6 cavalier about it on everybody else's behalf. I think that 7 we have to operate with a rule of reason here. On the one 8 hand, I have admonished Mr. Clark about going on to speak 9 about matters that weren't the subject of the question, and 10 I would respectfully request that we give Mr. Clark an 11 opportunity to give a complete answer within the confines of 12 my earlier admonition. I think it is not unreasonable. 13 You know, this gentleman is not a professional 14 witness. To his credit, he is probably not an economist or 15 an econometrician or any of the other kind of people that we 16 see coming through here. And, you know, let's just try and 17 take a deep breath and calm down, and I think we'll -- I'll 18 stop talking, then we will get out of here at a reasonable 19 hour. 20 THE WITNESS: So could I make a comment on that? 21 CHAIRMAN GLEIMAN: A short one. 22 THE WITNESS: When I read this part of the 23 testimony, I recall in the original, at the original times, 24 at the times when the times when these were being originally 25 proposed, in talking with Virginia Mayes, that the whole 19833 1 assumption was that anybody that utilized Destination 2 Sectional Center Facilities or DDUs would be required to 3 deliver packages in the same way that they were presently 4 being received from the Bulk Mail Centers, that was the 5 premise of the program. 6 So when you start to bring in all of these 7 considerations about shaking out sacks and helping with 8 unloading and so on, we are to do no more or no less than is 9 presently being done by a Bulk Mail Center. It is supposed 10 to be transparent to the system who brings those packages 11 in. And so that was, you know, part of the perspective that 12 I had. 13 CHAIRMAN GLEIMAN: Mr. Clark, I am going to offer 14 another comment here that might be helpful. 15 THE WITNESS: Yes. 16 CHAIRMAN GLEIMAN: And that is that, you know, 17 discussions that took place during the developmental phase 18 of Postal Service proposals are not a matter of record here. 19 And, as a matter of fact, in most cases, and as a matter of 20 fact, many of the kinds of issues that you just spoke to 21 aren't matters before us. The Postal Service has reserved 22 unto itself a great deal of flexibility as to how it 23 actually goes about implementing. 24 And, consequently, it is not unreasonable for 25 counsel to ask questions that go to different views of how 19834 1 things are actually going to work. Because, quite frankly, 2 it is not all locked up with every little I dotted and T 3 crossed when the proposal comes to us from the Postal 4 Service. 5 Now, let's -- counsel, if you want to move on now. 6 BY MR. McKEEVER: 7 Q It is currently the case that when you bring a 8 drop shipment to the Postal Service, the Postal Service 9 employees unload the vehicle, assisted by the driver when 10 requested? 11 A That is correct? 12 Q Okay. Incidentally, have you ever testified 13 before in any proceeding? 14 A I was here in '95. 15 Q Have you ever testified in any other forum before? 16 A Regulatory? 17 Q No. Any forum, ever, any court? 18 MR. OLSON: Mr. Chairman, I think I would object 19 to that as being irrelevant. If it is regulatory and as a 20 witness, that is one thing. But in any time, on any matter, 21 I would object to that. 22 CHAIRMAN GLEIMAN: Would you like to modify your 23 question or limit to something that might relate -- 24 MR. McKEEVER: No, Mr. Chairman, I think it is 25 relevant, the extent to which this witness has testified 19835 1 before. I think he knows what testifying is about. 2 THE WITNESS: Well, I apologize if I have offended 3 you. 4 MR. McKEEVER: No, you haven't offended me. 5 THE WITNESS: You can go ahead. 6 MR. McKEEVER: Mr. Clark, I do this for a living, 7 I don't get offended by it. 8 THE WITNESS: Okay. 9 CHAIRMAN GLEIMAN: Mr. McKeever, do you -- 10 MR. McKEEVER: Mr. Chairman, I'll withdraw the 11 question. 12 CHAIRMAN GLEIMAN: Thank you. 13 BY MR. McKEEVER: 14 Q Mr. Clark, you state in your rebuttal testimony, 15 and you indicated this earlier this evening, too, that Mr. 16 Luciani made an argument that DBMC parcels are different 17 from other parcels. Do you remember that? 18 A Yes. 19 Q That's in your rebuttal testimony on page 7 at 20 lines 10 to 11, is that right? 21 A Yes. 22 Q And, again, as I said, you made reference to that 23 really earlier tonight. In your rebuttal testimony, you say 24 that Mr. Luciani's statement is, quote, "simply posited with 25 no proof or quantification of any kind". Do you see that? 19836 1 A Yes. 2 Q Did you read Postal Service Witness Hatfield's 3 testimony on transportation costs in this case, by the way? 4 A I probably just glanced through it. 5 MR. McKEEVER: Mr. Chairman, with your permission, 6 I'd like to supply the witness with a copy of page 14 from 7 Mr. Hatfield's transportation testimony, USPS-T-16, I 8 believe. 9 CHAIRMAN GLEIMAN: Please proceed. 10 BY MR. McKEEVER: 11 Q Mr. Clark, I've handed you, as I mentioned, a copy 12 of page 14 from Postal Service Witness Hatfield's testimony 13 which shows a cube weight relationship by rate category for 14 parcel post for intra-BMC, inter-BMC and DBMC. Do you see 15 that? 16 A Yes. 17 Q Do you notice the DBMC line is the one at the top? 18 A Yes. 19 Q It's different from the other two; is that 20 correct? 21 A Yes. 22 Q Thank you. Does the Postal Service accept 23 shipments at CTC plants or is the acceptance done at a 24 Postal facility? 25 A At our plants. 19837 1 Q Postal people go to your plant? 2 A DMU, yes; detached mail unit. Your question was 3 parcels, right? 4 Q Yes. 5 A We have other types of mail as well. We have some 6 advertising mail, some letter mail that we commingle, so to 7 the extent that we pay the postage on that, it would be 8 included as well. 9 Q Mr. Clark, could you turn to page four of your 10 testimony, rebuttal testimony? There you state at lines 15 11 to 16 that UPS "is still far and away the dominant player in 12 the parcel delivery market." Do you see that? 13 A Yes, sir; I do. 14 Q Is it your testimony that UPS is the dominant 15 carrier in the case of residential deliveries? 16 A Residential deliveries? 17 Q Yes. Is that your testimony? 18 A Yes, for what's defined as a parcel. 19 MR. McKEEVER: Mr. Chairman, I would like to 20 provide the witness with a copy of the Postal Service's 21 Household Diary Study, Fiscal Year 1996, the Executive 22 Summary, which is a document published by the Postal 23 Service. 24 CHAIRMAN GLEIMAN: Please approach the witness. 25 BY MR. McKEEVER: 19838 1 Q Mr. Clark, could you turn to page 28 of that 2 document, please? 3 A I have it. 4 Q Now that page is headed, "Packages," is that 5 correct? 6 A Yes. 7 Q There is a Section B entitled "Overall Volume and 8 Carrier Used" -- do you see that? 9 A Yes. 10 Q And the first sentence in that paragraph says, 11 quote, "There was an increase in the number of packages 12 received per household per week during 1996." 13 Do you see that? 14 A Yes. 15 Q And then the third sentence in that paragraph 16 states, quote, "Of all packages received, 85.1 percent were 17 delivered by the Postal Service, which represents a 18 relatively large increase over the 1987 figure of 77.7 19 percent." 20 Is that correct? 21 MR. OLSON: Mr. Chairman, may we have time to 22 examine whether there is a definition of package in the 23 document before the witness has to answer all these 24 questions about this, which he hasn't seen? 25 MR. McKEEVER: Mr. Chairman, the witness can take 19839 1 as much time as he would like. 2 CHAIRMAN GLEIMAN: Yes, you can have time to 3 examine whether there is a definition. 4 Quite frankly, I don't know whether there is a 5 definition in the summary. We do have the entire study if 6 you find it necessary to look through that. 7 MR. McKEEVER: And I would be happy to have the 8 entire document here as well, Mr. Chairman. 9 THE WITNESS: My own personal experience is that 10 this -- that this doesn't relate to the market we are 11 talking about here at all. This would include all Third 12 Class, books -- 13 BY MR. McKEEVER: 14 Q It includes all packages. 15 A Priority mail. It's not Fourth Class Parcel Post. 16 Q I see. Your testimony is only with respect to 17 Fourth Class Parcel Post? 18 A That's what I made clear -- yes -- in the original 19 testimony. 20 Q Well, that is not what you say in your rebuttal 21 testimony and I don't have your original testimony available 22 right now but you say, "It is far and away the dominant 23 player in the parcel delivery market." 24 Now you are only talking about some parcels, is 25 that correct? Is that your statement? 19840 1 A Well, I think you are quibbling here over what we 2 are talking about. 3 We are talking about work-sharing discounts. 4 That's the context of the whole rate hearing. Why bring in 5 issues about Third Class records, books, all kinds of 6 packages? 7 Q Mr. Clark, you stated in your testimony that UPS 8 is far and away the dominant player in the parcel delivery 9 market. 10 A That's my opinion. Yes, sir. 11 Q And that is the reason I brought this document up. 12 A Well, my own opinion is it's not an appropriate 13 representation or comparison for this case. 14 Q You don't believe that a package -- you don't 15 believe that a package is a package? 16 A Well, I guess we could ask another question. Does 17 UPS compete for the delivery of Third Class packages? 18 Q Do you know -- first of all, I ask the questions, 19 Mr. Clark. I apologize, but that is how it works. 20 A Well, I am just trying to get to the truth here. 21 Q Is it your testimony that UPS will refuse to 22 deliver a package under one pound? 23 A No, it's not. 24 Q Okay. They will, won't they? 25 A They sure do. 19841 1 Q And they will deliver books and records, is that 2 correct? 3 A Absolutely. 4 Q They also deliver packages as part of Second Day 5 Air, is that correct? 6 A I presume so. 7 Q They also deliver packages that could be sent by 8 the Postal Service as First Class mail; is that correct? 9 A That could be sent as First Class? 10 Q Could be sent as First Class mail by the Postal 11 Service; yes. 12 A I would imagine they do; yes. 13 Q Is it your -- how was this diary put together, by 14 the way? 15 A All I can tell you, Mr. Clark, is it's an official 16 publication of the United States Postal Service. 17 MR. McKEEVER: Mr. Chairman, that concludes my 18 examination. I would like to move that the executive 19 summary of the household diary study be transcribed in the 20 record and also admitted into evidence as an official 21 document of the United States Postal Service. 22 CHAIRMAN GLEIMAN: Mr. Olson? 23 MR. OLSON: Mr. Chairman, I guess I would think 24 that page 28 would be an appropriate page, since that's what 25 we discussed, and perhaps even 29 and 30, which deal with 19842 1 packages, but I can see no reason for including the whole 2 document. If it's an official postal document, it can 3 probably get in some other way besides cross-examination of 4 this witness. 5 MR. McKEEVER: Mr. Chairman, I have no objection 6 to moving into evidence only the cover page and the three 7 pages mentioned by Mr. Olson. 8 MR. OLSON: I'd certainly object to it being moved 9 into evidence. As a cross-examination exhibit I have no 10 problem. 11 CHAIRMAN GLEIMAN: You -- I'm not quite sure I 12 understand. You object to it moving into evidence -- 13 MR. OLSON: Yes. 14 CHAIRMAN GLEIMAN: Period. You object to it 15 moving into -- being transcribed and admitted into evidence. 16 MR. OLSON: I think it being transcribed as a 17 cross-examination exhibit is a good idea. I would object to 18 putting into evidence. I cannot testify as to authenticity 19 or Mr. McKeever certainly cannot. 20 MR. McKEEVER: Mr. Chairman, that's exactly why 21 there is an exception for public documents, so that you 22 don't need someone to testify as to authenticity. Rule 23 31(d) of the Commission's rules specifically indicates that 24 a public document such as an official report -- go ahead, 25 Mr. Chairman, I apologize. 19843 1 CHAIRMAN GLEIMAN: Mr. McKeever and Mr. Olson, you 2 had an exchange before about the extent to which you would 3 limit the number of pages. I just want to make sure before 4 I direct that certain pages be transcribed into the record 5 and admitted into evidence that we have the right pages. 6 Did we have an agreement on the cover page and pages 28, 29, 7 and 30? 8 MR. McKEEVER: That's acceptable to us, Mr. 9 Chairman. I was putting the whole document in just for the 10 sake of having the complete document in, because of course 11 in the case of a public document when a party offers part, 12 anyone else is free to put it all in, and so I was 13 attempting to put the entire executive summary in. But I'm 14 happy to have only those three pages put in, together with 15 the cover page. 16 CHAIRMAN GLEIMAN: You want to stick by your 17 request that we limit it to the cover page and the three 18 pages, Mr. Olson? 19 MR. OLSON: At this time I do. I would say that 20 not knowing this was going to be put into the record, I'd 21 have to review it and see if there was some reason to burden 22 the record with more pages, and if there was, I guess I'd 23 make an appropriate motion, but at this time I don't. 24 CHAIRMAN GLEIMAN: Well, in that case I'm going to 25 direct that four pages, the cover page and pages No. 28, 29, 19844 1 and 30 of the Executive Summary of the Household Diary Study 2 for Fiscal Year 1996 -- the Executive Summary, I did say, of 3 the United States Postal Service, which was apparently 4 issued, it's got a date of September 1997 on it, be 5 transcribed into the record and admitted into evidence at 6 this point. 7 [Excerpt from the United States 8 Postal Service Household Diary 9 Study, Fiscal Year 1996, Executive 10 Summary, was received into evidence 11 and transcribed into the record.] 12 13 14 15 16 17 18 19 20 21 22 23 24 25 19849 1 MR. McKEEVER: That concludes our 2 cross-examination, Mr. Chairman. I can furnish the court 3 reporter with two copies of those pages in just a few 4 moments. 5 THE WITNESS: Can I make a comment here? You 6 asked about the definition, the top of page 29. 7 CHAIRMAN GLEIMAN: We'll have a chance to do that 8 on redirect. 9 Is there any followup? 10 There's no followup. 11 That brings us to redirect. Mr. Olson, would you 12 like a few minutes with your witness? 13 MR. OLSON: Yes, Mr. Chairman. 14 CHAIRMAN GLEIMAN: Five, ten? 15 MR. OLSON: Five. 16 CHAIRMAN GLEIMAN: Five. Okay. 17 [Recess.] 18 CHAIRMAN GLEIMAN: Mr. Olson, you have some 19 redirect, I take it? 20 MR. OLSON: Yes, Mr. Chairman. 21 REDIRECT EXAMINATION 22 BY MR. OLSON: 23 Q Mr. Clark, I would ask you to look at your 24 testimony on page 7. I believe it is line 7 through 9, 25 although I don't have it in front of me. Mr. McKeever had 19850 1 asked you some questions with respect to your criticism of 2 Witness Luciani about certain inexplicable changes, and I 3 wonder if you can tell us what you meant by that testimony. 4 A Well, I was really referring to line 16, line 16 5 through 21 on page 26, where he simply commented that the 6 Postal Service had changed the cost of -- the magnitude of 7 cost estimates is unexplained. In fact, I thought that they 8 did explain that. So just because it is unexplainable 9 doesn't mean it is -- in his opinion, unexplainable, doesn't 10 mean it was uncertain. It doesn't create necessarily an 11 uncertainty. 12 Q So that part of your testimony dealt with that 13 particular point that Witness Luciani reference, right. 14 What page of Witness Luciani's testimony is that? 15 A On page 26. And it follows, you know, the first 16 comment that he made about imperfect execution, it's all in 17 that area. 18 Q And, in fact, those are the pages, 26 and 27, that 19 are cited in your testimony right after your point, correct? 20 A That is correct. 21 Q And so when Mr. McKeever walked you through 22 Witness Luciani's criticisms of the Postal Service costing 23 in the early part of his testimony, that was not what you 24 were referencing, correct? 25 A That's right. 19851 1 Q And, in fact, you did have a line that you deleted 2 at the beginning of the testimony. Can you tell us why you 3 deleted that? 4 A Well, I didn't -- you know, I don't -- I feel that 5 I overstated my position as far as my qualifications. I 6 really -- I am not qualified to war -- be at war with the 7 cost estimates in the proceedings, I never pretended to be 8 that. 9 Q Okay. Secondly, let me ask you to take a look at 10 the pages of the Household Diary Study which counsel for UPS 11 asked you some questions about, particularly 28 and 29, and 12 ask you if you have any further thoughts about UPS being the 13 dominant carrier with respect to residential delivery? 14 A Well, I think that the question came up as to what 15 the definition, the appropriateness of this comparison was, 16 and on page 29, it says the percentage of packages delivered 17 by the Postal Service includes parcels delivered at all 18 rates. The largest share of parcels was Third Class bulk 19 rate, 32.8 percent of the total, following by First Class 20 mail and priority, which was a combined 26 percent. 21 Later on, on that same page, it goes by content. 22 The Postal Service delivered 92.4 percent of records, tapes 23 and CDs, 94.6 percent of books, reflecting the use of 24 special Fourth Class and Bound Printed Matter rates. UPS 25 was a preferred -- UPS was preferred by many mailers. It's 19852 1 largest shares of delivery were in the catalog order and 2 store order categories, respectively, 27.1 percent, 29.9 3 percent. 4 Q Okay. And let's go back to page 4 of your 5 testimony, which was the quotation that I believe counsel 6 for UPS used to set up the question. Can you tell us what 7 you mean the phrase "parcel" on line 15 and 16 of page 4? 8 A Well, throughout my testimony, I talk about parcel 9 post, on line five, line seven, line 11. I talk about the 10 DBMC program on line 14. Line 15, nationwide parcel 11 delivery. Line 16, parcel delivery market. I pretty well 12 limited my testimony on comparisons in this case to that 13 marketplace, and that considered books, records, tapes, CD's 14 and all those different things. 15 MR. OLSON: Thank you, Mr. Chairman. That's all 16 we have. 17 CHAIRMAN GLEIMAN: Recross? 18 MR. McKEEVER: Yes, Mr. Chairman. 19 RECROSS-EXAMINATION 20 BY MR. McKEEVER: 21 Q In your rebuttal testimony, you didn't use the 22 term "parcel post" when you were talking about the parcel 23 delivery market, did you? 24 A I just cited on page four, the number of times I 25 did use that. 19853 1 Q In your rebuttal testimony? 2 A Yes, sir. 3 Q You said "parcel." You didn't say "parcel post." 4 That's my recollection. 5 A Well, look on page four, line five, line seven, 6 line 11. 7 Q But that's not where you were talking about the 8 delivery market; is that right? When you were talking about 9 the delivery market, you said the parcel delivery market. 10 A Imperfect syntax, I guess. 11 Q As the household diary summary sentence that you 12 pointed out states, the packages being talked about there 13 includes parcels delivered at all rates; is that correct? 14 A That's what it says; yes, sir. 15 Q So those figures include among packages delivered 16 in third class, it includes parcel post packages as well; is 17 that right? 18 A I think so; yes. It says that in section C there. 19 Q Right. In fact, the Postal Service has a 20 significantly large -- has a significantly large share of 21 packages carried in third class and books and records, 22 because their rates are far cheaper than United Parcel 23 Service's rates; isn't that the case? 24 A I would think that would be the primary driver; 25 yes. 19854 1 Q Did you read Mr. Jellison's testimony in this 2 case? 3 A Yes. 4 Q He gave some figures about parcel post shipments 5 to residences and the share delivered by UPS and the share 6 delivered by the Postal Service; is that correct? 7 A I don't recall it specifically. I'll take your 8 word for it. 9 MR. McKEEVER: That's all I have, Mr. Chairman. 10 CHAIRMAN GLEIMAN: Is there any further redirect? 11 MR. OLSON: Just one, Mr. Chairman. 12 FURTHER REDIRECT EXAMINATION 13 BY MR. OLSON: 14 Q Mr. Clark, isn't it true when Mr. McKeever just 15 identified the items that were packages sent -- received, 16 according to the household diary study, that perhaps he 17 should have also listed first class mail and priority mail? 18 A It's in there; yes. 19 MR. OLSON: Nothing else. 20 CHAIRMAN GLEIMAN: Mr. McKeever? 21 FURTHER RECROSS-EXAMINATION 22 BY MR. McKEEVER: 23 Q Taking into account all those markets, the 24 household diary study shows that the Postal Service by far 25 delivers more packages than anyone else to residences; is 19855 1 that correct? 2 A That's what it reports. 3 MR. McKEEVER: Thank you, Mr. Chairman. 4 CHAIRMAN GLEIMAN: Anyone else? 5 [No response.] 6 CHAIRMAN GLEIMAN: If there is nothing else, then 7 Mr. Clark, I want to thank you. We appreciate your 8 appearance here today and your contributions to our record, 9 and if there is nothing further, you are excused. 10 [Witness excused.] 11 CHAIRMAN GLEIMAN: I was supposed to have a short 12 ending sentence here saying these hearings are adjourned. 13 They are adjourned pending Postal Service response to my 14 ruling on ANM motion to compel, and the scheduling of the 15 hearing on Witness Shenk's survey relating to the documents 16 in question, that hearing is tentatively scheduled for 17 Monday, March 30th. 18 I hope you all have a good evening and a nice 19 weekend. 20 [Whereupon, at 8:57 p.m., the hearing was 21 recessed, to reconvene at 9:30 a.m., Monday, March 30, 22 1998.] 23 24 25