ARCHIVÉ -  Transcription

 

 

 

 

 

 

 

              TRANSCRIPT OF PROCEEDINGS BEFORE

             THE CANADIAN RADIO‑TELEVISION AND

               TELECOMMUNICATIONS COMMISSION

 

 

 

 

             TRANSCRIPTION DES AUDIENCES DEVANT

              LE CONSEIL DE LA RADIODIFFUSION

           ET DES TÉLÉCOMMUNICATIONS CANADIENNES

 

 

                          SUBJECT:

 

 

 

Review of price cap framework /

Examen du cadre de plafonnement des prix

 

 

 

 

 

 

 

 

 

 

 

 

 

HELD AT:                              TENUE À:

 

Conference Centre                     Centre de conférences

Outaouais Room                        Salle Outaouais

140 Promenade du Portage              140, Promenade du Portage

Gatineau, Quebec                      Gatineau (Québec)

 

October 16, 2006                      Le 16 octobre 2006

 

 

 

 

 

Transcripts

 

In order to meet the requirements of the Official Languages

Act, transcripts of proceedings before the Commission will be

bilingual as to their covers, the listing of the CRTC members

and staff attending the public hearings, and the Table of

Contents.

 

However, the aforementioned publication is the recorded

verbatim transcript and, as such, is taped and transcribed in

either of the official languages, depending on the language

spoken by the participant at the public hearing.

 

 

 

 

Transcription

 

Afin de rencontrer les exigences de la Loi sur les langues

officielles, les procès‑verbaux pour le Conseil seront

bilingues en ce qui a trait à la page couverture, la liste des

membres et du personnel du CRTC participant à l'audience

publique ainsi que la table des matières.

 

Toutefois, la publication susmentionnée est un compte rendu

textuel des délibérations et, en tant que tel, est enregistrée

et transcrite dans l'une ou l'autre des deux langues

officielles, compte tenu de la langue utilisée par le

participant à l'audience publique.

               Canadian Radio‑television and

               Telecommunications Commission

 

            Conseil de la radiodiffusion et des

               télécommunications canadiennes

 

 

                 Transcript / Transcription

 

 

 

 

              Review of price cap framework /

          Examen du cadre de plafonnement des prix

 

 

 

 

 

BEFORE / DEVANT:

 

Richard French                    Chairperson / Président

Helen del Val                     Commissioner / Conseillère

Elizabeth Duncan                  Commissioner / Conseillère

Andrée Noël                       Commissioner / Conseillère

Stuart Langford                   Commissioner / Conseiller

 

 

 

 

ALSO PRESENT / AUSSI PRÉSENTS:

 

Marielle Giroux-Girard            Secretary / Secrétaire

Bob Noakes                        Staff Team Leader /

Chef d'équipe du personnel

Stephen Millington                Legal Counsel /

Rachelle Frenette                 Conseillers juridiques

 

 

 

 

HELD AT:                          TENUE À:

 

Conference Centre                 Centre de conférences

Outaouais Room                    Salle Outaouais

140 Promenade du Portage          140, Promenade du Portage

Gatineau, Quebec                  Gatineau (Québec)

 

October 16, 2006                  Le 16 octobre 2006

 

                           - iv -

 

           TABLE DES MATIÈRES / TABLE OF CONTENTS

 

 

                                                 PAGE / PARA

 

AFFIRMED:  TREVOR ROYCROFT                       1148 / 7825

Cross-examination by The Companies               1149 / 7837

Cross-examination by MTS Allstream               1318 / 9061

Cross-examination by BCOAPO                      1341 / 9265

 

 

AFFIRMED:  GREG MATWICHUK                        1373 / 9440

Examination-in-chief by The City of Calgary      1373 / 9440

Cross-examination by The Consumer Groups         1374 / 9450

 

 

                           - v -

 

               EXHIBITS / PIÈCES JUSTICATIVES

 

 

No.                                              PAGE / PARA

 

CONSUMER      Table 6 - (corrected) of the

GROUPS‑2      summary of DSL impact on

TFP Growth                        1146 / 7800

 

CONSUMER      Amended p. 69 of the

GROUPS‑3      testimony of Trevor R. Roycroft

on behalf of the Consumer Groups  1146 / 7801

 

THE           Dr. Roycroft's recommendations

COMPANIES‑11  for calculating the x-factor

source: table 8 p. 58 of

Dr. Roycroft's testimony          1172 / 8015

 

THE           Dr. Roycroft's Rebuttal

COMPANIES‑12  Testimony before the Indiana

Utility Regulatory Commission

- Cause No. 42405, 13 January 2004 1179 / 8065

 

THE           The IPD Trend

COMPANIES‑13  Source: p. 27 of

Dr. Roycroft's testimony          1190 / 8156

 

THE           The IPD - Labour component        1206 / 8280

COMPANIES‑14

 

MTS           International Journal of

ALLSTREAM‑4   Development Planning Literature

Vol. 16 Numbers 1 & 2             1319 / 9068

 

TELUS‑13      Decision dated August 24, 2006

before the Public Utilities

Commission of the State of

California                        1383 / 9514

 

TELUS‑14      Pages 90, 91 and 92 of

Telecom Decision CRTC‑2002‑34     1383 / 9514

 

TELUS‑15      Economic and financial indicators

from the Economist dated

Sept. 16, 2006                    1383 / 9514

                           - vi -

 

                 UNDERTAKINGS / ENGAGEMENTS

 

 

No.                                              PAGE / PARA

 

MTS           Dr. Roycroft to verify whether

ALLSTREAM‑1   the Average Impact of DSL on

TFP Growth (2000-2003) of

0.57% reflects an overall

weighted average over the

four‑year period or whether

it reflects an average

annual impact                     1340 / 9255

 

THE           Dr. Roycroft is requested

COMPANIES‑3   to replace Dr. Roycroft's DSL

input data with the data

provided by The Companies in

The Companies' exhibit #2

(capital and expense) to

verify the results shown

in The Companies Exhibit #2       1381 / 9496

 

                          - vii -

 

                           ERRATA

 

 

Transcript:  October 13, 2006

 

PAGE / PARA.  DESCRIPTION

 

956 / 6460    "MR. LAWFORD:" should read "MR. DANIELS:"

956 / 6462    "MR. LAWFORD:" should read "MR. DANIELS:"

956 / 6464    "MR. LAWFORD:" should read "MR. DANIELS:"

956 / 6466    "MR. LAWFORD:" should read "MR. DANIELS:"

 

 

Transcript:  October 16, 2006

 

The proper description for Undertaking No. The Companies‑3 is found at page 1381, paragraph 9496:

"Dr. Roycroft is requested to replace Dr. Roycroft's DSL input data with the data provided by The Companies in The Companies' exhibit #2 (capital and expense) to verify the results shown in The Companies Exhibit #2"

 

                  Gatineau, Quebec / Gatineau, Québec

‑‑‑ Upon resuming on Monday, October 16, 2006

    at 0859 / L'audience reprend le lundi

    16 octobre 2006 à 0859

7773             LE PRÉSIDENT:  À l'ordre, s'il vous plaît.

7774             Madame la secrétaire...?

7775             THE SECRETARY:  Good morning, everybody.

7776             We will start the week with the witness on behalf of the Consumer Groups, Dr. Roycroft.

7777             Mr. Janigan, if you want to present your witness.

7778             MR. JANIGAN:  Thank you very much, Madam Secretary.

7779             Mr. Chair, Dr. Roycroft is seated at the witness panel.  Assisting him will, of course, be Mr. Lawford.

7780             MR. JANIGAN:  Dr. Roycroft, are your qualifications correctly set out in the appendix to your testimony of July 10, 2006?

7781             DR. ROYCROFT:  Yes, they are.

7782             MR. JANIGAN:  With respect to the evidence entitled "Testimony of Dr. Trevor R. Roycroft on behalf of Consumer Groups, July 10, 2006", did you prepare this evidence and does this evidence fairly reflect statements that are true to the best of your knowledge and belief?

7783             DR. ROYCROFT:  Yes, it does.

7784             MR. JANIGAN:  Did you also prepare or review and supervise the preparation of the interrogatory responses to your questions based on your evidence from other interested parties in this proceeding?

7785             DR. ROYCROFT:  Yes, those questions specifically directed at my evidence.

7786             MR. JANIGAN:  All right.

7787             Are those true to the best of your knowledge and belief?

7788             DR. ROYCROFT:  Yes.

7789             MR. JANIGAN:  I understand that there are some corrections to your evidence that we would like to put on the record today and that the corrections have been put in the form of amended pages to your testimony?

7790             DR. ROYCROFT:  That is correct.

7791             MR. JANIGAN:  These sheets have been given to the Hearing Secretary, as I understand.

7792             I wonder if we could deal first with the two‑page sheet with Table 6 at the top of the page.

7793             What corrections are involved in this?

7794             DR. ROYCROFT:  The corrected sheets replace values, specifically values that are in the third column of Table 6, which reflect calculations that I made on the average impact of DSL lines on total factor productivity growth.

7795             In the course of responding to discovery, I uncovered an error within one of the spreadsheets that led to a slight difference in the calculation outcome and have replaced the original values in Table 6 with the corrected values.

7796             That correction also caused Table 7 to require an update and the values and discussion associated with Table 7 have been updated to reflect the correction.

7797             MR. JANIGAN:  Mr. Chair, I don't know whether or not the correction itself should be marked as an exhibit.

7798             THE CHAIRPERSON:  Neither do I.

7799             MR. JANIGAN:  Perhaps we will, just to make sure that it is on the record.

7800             THE SECRETARY:  For the record, the two exhibits that were filed before the Commission, "Table 6, Summary of DSL Impact on TFP Growth" is Consumer Groups Exhibit No. 2.

EXHIBIT NO. CONSUMER GROUPS‑2:  Table 6 - (corrected) of the summary of DSL impact on TFP Growth

7801             THE SECRETARY:  The "Amended Telecom Public Notice CRTC 2006‑5" is Consumer Group Exhibit No. 3.

EXHIBIT NO. CONSUMER GROUPS‑3:  Amended p. 69 of the testimony of Trevor R. Roycroft, on behalf of the Consumer Groups

7802             MR. JANIGAN:  I wonder if you could deal with the single page correction, Dr. Roycroft.

7803             DR. ROYCROFT:  Yes.  That page is page 69, which is the very last page of the testimony.

7804             Once again, when responding to discovery I uncovered an error with regard to the description of how I estimated dial equipment minutes associated with the productivity studies that I undertook and have created a replacement page that restates the estimation method.

7805             MR. JANIGAN:  Thank you.

7806             That page is marked as Exhibit No. ...?

7807             THE CHAIRPERSON:  Three.

7808             MR. JANIGAN:  Three, thank you.

7809             THE CHAIRPERSON:  Mr. Janigan, could I ask your indulgence, just to be sure I have this right.

7810             Does the page being replaced begin with the words "data on local and interstate toll"?

7811             MR. JANIGAN:  You are referring to Exhibit No. 3?

7812             THE CHAIRPERSON:  Yes.  I am just trying to figure out what it replaces.  I think my colleague has just ‑‑ I'm still a little confused.

7813             Does the first sentence of the original last page remain or does the entire last page get replaced by this Exhibit 3?

7814             DR. ROYCROFT:  The entire last page is replaced.

7815             THE CHAIRPERSON:  So the sentence that says:

"Data on local and interstate toll dial equipment minutes is available from FCC monitoring report for the years 1985‑2000."

7816             That is gone as well?

7817             DR. ROYCROFT:  Yes.

7818             THE CHAIRPERSON:  Thank you very much.

7819             Sorry to interrupt.

7820             MR. JANIGAN:  That's okay.  Thank you.

7821             Subject to these corrections, then, your evidence is true to the best of your knowledge and belief?

7822             DR. ROYCROFT:  Yes, it is.

7823             MR. JANIGAN:  Thank you.

7824             Mr. Chairman, Dr. Roycroft is available for examination.

7825             THE SECRETARY:  Before we do so, I would like to affirm our witness, please.

AFFIRMED:  TREVOR ROYCROFT

7826             THE SECRETARY:  Thank you very much.

7827             I now invite counsel Denis to proceed on behalf of The Companies.

7828             MR. HENRY:  We are on a first‑name basis.

7829             THE CHAIRPERSON:  You are a regular.

‑‑‑ Laughter / Rires

‑‑‑ Pause

7830             THE CHAIRPERSON:  Counsel Denis, would you, for the benefit of the webcast audience, introduce yourself and your colleagues?

‑‑‑ Laughter / Rires

7831             MR. HENRY:  I will, for all those in webcast land particularly.

7832             I am Denis Henry, representing Bell Canada and Saskatchewan Telecommunications and Bell Aliant this morning.

7833             With me, on my left is Dr. Patrick Owens of Bell Canada.

7834             On my right I am assisted by Dr. Hariton.  I must say, Mr. Chairman, I think I am actually justified this morning in calling him Doctor, because I have a feeling before I'm through here this morning I am going to need to rely on his Ph.D. in Mathematics.

7835             THE CHAIRPERSON:  That is not good news, Mr. Henry.

‑‑‑ Laughter / Rires

7836             MR. HENRY:  It was kind of meant to be a warning.

‑‑‑ Laughter / Rires

CROSS‑EXAMINATION / CONTRE‑INTERROGATOIRE

7837             MR. HENRY:  Good morning, Dr. Roycroft.

7838             DR. ROYCROFT:  Good morning.

7839             MR. HENRY:  I would like to start with some questions of clarification on one aspect of your evidence that I have to say we remain somewhat unclear on.

7840             Before I do that, let me just recap the essential elements.

7841             You were proposing an I‑X formula, an "X" of 6 percent and a price cap period of four years.

7842             Is that correct?

7843             DR. ROYCROFT:  That is correct.

7844             MR. HENRY:  The part I'm unclear about is the effect of forbearance, because you do talk about it in your evidence.  What happens when areas are forborne pursuant to the Commission's forbearance decision and the effect of that on price cap.  I wanted to just explore that with you.

7845             Under today's regime, could you confirm that when services are forborne their revenues and demand are removed from the price cap indices?

7846             DR. ROYCROFT:  That is correct.

7847             MR. HENRY:  As I understand it, you recommend what you call a "hold harmless" provision.  That would keep PES, primary exchange service, within the price cap framework following a forbearance grant?

7848             DR. ROYCROFT:  Yes, it would keep it within the framework, also recognizing that the Commission has assigned a special status to primary exchange service within forbearance areas with regard to a ceiling requirement associated with the service.

7849             MR. HENRY:  That would be for a period of five years after the grant of forbearance?

7850             DR. ROYCROFT:  The hold harmless provision?

7851             MR. HENRY: Right.

7852             DR. ROYCROFT:  Yes.

7853             MR. HENRY:  Does that mean that price changes for primary exchange service in forborne areas would need to comply with the price cap constraints for that five‑year period?

7854             DR. ROYCROFT:  It would mean that they would be counted within the calculation of the actual price indices that are used to check compliance for the overall price cap.

7855             Within the forbearance area the primary exchange service is under an alternative constraint as defined by the Commission in the final order 2006‑15.

7856             MR. HENRY:  But does it mean that the revenues subject to the I‑X formula would include the forborne revenues?

7857             DR. ROYCROFT:  The recommendation that I make is that the primary exchange service quantities remain in place within the calculation of the actual price index, and the corresponding prices charged for primary exchange service would presumably be used in that calculation as well.

7858             When you speak to revenues, I would then surmise that the impact of evaluating both the quantities and the prices would then feed revenues into the equation.

7859             MR. HENRY:  Yes, because demand times price equals revenues.

7860             DR. ROYCROFT:  Yes.

7861             MR. HENRY:  Let me give you an example.

7862             If I had a million dollars in revenue that was forborne and inflation were at 2 percent and your X was 6 percent, then do I have to reduce my revenues attributable to those revenues by $40,000?

7863             DR. ROYCROFT:  I'm sorry.  Could you give me those values again.

7864             MR. HENRY:  A million dollars in revenue in a forborne area and ‑‑

7865             DR. ROYCROFT:  This is $1 million in primary exchange service rates.

7866             MR. HENRY:  Right.  And I‑X is 4 percent; I is 2 and X is 6.

7867             Then do I have to apply the I‑X to the million dollars, which would result in a $40,000 revenue bogey that I would have to shed to meet the price cap?

7868             DR. ROYCROFT:  I don't think that you would from the standpoint that the impact within the overall basket coming under the compliance would give you the ability to decide where you would be reducing those revenues to come into compliance under the cap.

7869             MR. HENRY:  But don't I have to apply my demand from the forborne revenues to the price to calculate the index?

7870             DR. ROYCROFT:  It seems that you are assuming that the cap is being applied directly to the forborne area as opposed to a broader basket, as it is today, with regard to covering the overall set of primary exchange service offerings across rate bands.

7871             That does not imply that you must reduce rates within each rate band area.  You have a degree of flexibility.

7872             MR. HENRY:  I'm still confused.

7873             Could you turn perhaps to Interrogatory Consumer Groups‑The Companies 8 August, No. 11.  And I think it is at page 16 of the pile you gave us.

7874             Do you have that?

7875             DR. ROYCROFT:  Yes, I do.

7876             MR. HENRY:  It is the (b) part of the question I'm interested in.  It is on page 17.

7877             Do you see that?

7878             The question starts with "If residential stand‑alone PES".

7879             Do you see that?

7880             DR. ROYCROFT:  Yes.

7881             MR. HENRY:  Okay.  Let me just read the question:

"If residential stand‑alone PES in forborne areas were to remain within the existing price cap framework following a forbearance grant, as Dr. Roycroft suggests, would Dr. Roycroft advocate that these PES be included in the calculation of the actual pricing index such that a decrease in the rate for these PES would impact the calculation of this index?"

7882             And your answer was a simple "yes".

7883             DR. ROYCROFT:  Yes.

7884             MR. HENRY:  Which led me to believe that the revenues were part of ‑‑

7885             DR. ROYCROFT:  The question that is posed here is different than the question you were asking just now, in that this implies a price adjustment made at the company's discretion within the forbearance area, which would then feed into the calculation of the actual price index.

7886             Within the context of this question, that price reduction would then allow the company to have higher offsetting price increases within the overall cap from the standpoint that the revenues are going down within the basket associated with the prices in the forbearance area.

7887             MR. HENRY:  Let me understand then.

7888             If the ILEC responds to competition in the forborne areas, which I think you say is more likely because those would be the ones that were more competitive by reducing prices, and leaves prices untouched in the non‑forborne areas, your proposal would create head room in the non‑forborne areas to raise rates.

7889             DR. ROYCROFT:  That is what the proposal does, yes.  It continues to treat the primary exchange service within the forborne areas as if it was in the overall basket.

7890             MR. HENRY:  So wouldn't this allow prices to be higher in the non‑forborne areas than they would with the current system where you remove the revenues?

7891             DR. ROYCROFT:  Well, given the structure of the actual price index, I don't know the exact answer to that question from the standpoint if you remove primary exchange service from the actual price index, as I have indicated through an example in my testimony, there may be an impact on the overall compliance level implied by the actual price index.

7892             The approach that I'm suggesting here holds harmless both sides of the equation from the standpoint that it prevents an adjustment resulting from an artifact of the actual price index either harming consumers or affecting the company.

7893             MR. HENRY:  I would suggest that an artifact from your proposal would be that as prices fall in the forborne areas, it would give you head room and allow you to increases prices which you otherwise wouldn't be allowed to do in today's system in the non‑forborne areas.

7894             I think we agreed on that.

7895             DR. ROYCROFT:  That is correct.

7896             MR. HENRY:  By the way, another curiosity.  I notice that you say that this treatment of leaving these PES demand from the forborne areas in for five years after a grant of forbearance ‑‑ how would that work?

7897             You have a four‑year price cap period.  So if in year four, 2010 were forborne, you still carry on for five years to 2015 counting these PES demand in the forborne areas?

7898             I don't follow that.

7899             DR. ROYCROFT:  I think I'm proposing that the application of the hold harmless apply for a five‑year period.

7900             MR. HENRY:  Your price cap period is four years.  So even if everything were forborne on day one, you would continue this beyond the end of the four‑year price cap period.

7901             DR. ROYCROFT:  Yes.

7902             MR. HENRY:  Okay.

7903             I want to understand a little bit better your perception of the context within which you made your proposal.  I take it your proposal is based in large measure on your view that competition falls short of the level necessary to discipline prices.

7904             DR. ROYCROFT:  That is correct.

7905             MR. HENRY:  In fact, you say that there is generally little evidence of entry in Canadian residential markets.

7906             DR. ROYCROFT:  That is my testimony.

7907             MR. HENRY:  I take it you did not see competition progressing sufficiently during the next price cap period to warrant removal of your I‑X price cap formula.

7908             DR. ROYCROFT:  No, I don't believe that there is a reasonable expectation that a blanket removal of I‑X within the next price cap period is a reasonable expectation.

7909             MR. HENRY:  And we are talking out to 2011.

7910             DR. ROYCROFT:  Yes.

7911             MR. HENRY:  You base your conclusions on the state of competition primarily ‑‑ I know not exclusively ‑‑ on the documents that you cite of the CRTC's 2005 monitoring report as well as the forbearance decision?

7912             DR. ROYCROFT:  Yes, I used information contained in those documents, as well as my own expertise and experience based on observing competition unfold in U.S. markets.

7913             MR. HENRY:  So you would not be of the view that local competition is more deeply rooted than it was at the time of the local forbearance decision?

7914             DR. ROYCROFT:  The local forbearance decision of earlier this year?

7915             MR. HENRY:  Yes.

7916             DR. ROYCROFT:  When you say "more deeply rooted", that seems to imply a significant change.  There may be marginal changes in the intervening period but none that would warrant me changing my perception.

7917             MR. HENRY:  And you wouldn't agree that growth in the residential local VoIP services is resulting in significantly stronger competition in the local exchange market, more than what was anticipated at the time of the local forbearance decision?

7918             DR. ROYCROFT:  Local VoIP being defined as over‑the‑top, VoIP?

7919             MR. HENRY:  Yes.  Well, either.

7920             DR. ROYCROFT:  Well, I view the ‑‑

7921             MR. HENRY:  Significantly stronger competition than the local exchange market as a result of both kinds of VoIP since the local forbearance decision.

7922             DR. ROYCROFT:  No.  With regard to over‑the‑top VoIP, I don't see that technology is providing a reasonable alternative for a broad section of the marketplace.

7923             With regard to cable VoIP, there is no question that cable companies are in the process of planning and expanding their service offerings based on the evidence that I've reviewed. 

7924             Whether it's reasonable to expect that those cable service offerings will be an alternative for a broad section of the market is not a reasonable expectation at this point.

7925             MR. HENRY:  Are you aware that the Commission is now re‑examining its local forbearance decision and, in fact, the two propositions I've put to you are exactly the Commission's words, that competition is more deeply routed than it was at the time as a local forbearance decision and that local VoIP is resulting in significantly stronger competition?

7926             I guess you weren't aware of that, you didn't study those decisions?

7927             DR. ROYCROFT:  No, I did not.

7928             MR. HENRY:  Are you aware that the Government of Canada ‑‑

7929             MR. JANIGAN:  Sorry.  It would be the notices, right; it wouldn't be the decisions?  You didn't study those decisions.  I assume that he studied the forbearance decision, but it would be the notices that you're referencing to?

7930             THE CHAIRMAN:  No.  There was a VoIP pre‑consideration decision and there was a public notice, Mr. Janigan, and I think the witness has said to us that he didn't read either.

7931             MR. JANIGAN:  Sure.

7932             THE CHAIRMAN:  I think we all know what we're talking about specifically here.  It's fair enough to add that one of the quotes may have been from a public notice.  I don't know whether you've got a TELUS counsel?

7933             MR. HENRY:  One was from Decision 2006‑53 and the other one may have been from the public notice, but ‑‑

7934             THE CHAIRMAN:  Fair enough, but I think the point is that two official Commission documents came out and the witness hasn't had the opportunity to examine them.

7935             MR. HENRY:  And that the witness disagrees with the proposition in them.  I think he's said that.

7936             Are you aware that the Government of Canada in its order to the Commission to reconsider VoIP, its VoIP decision, has stated that VoIP has transformed the nature and extent of competition in the local telephony?

7937             DR. ROYCROFT:  Your statement was their order to the Commission to reconsider VoIP.  Are you saying they've ordered to the Commission to reconsider the forbearance order?

7938             MR. HENRY:  No, the VoIP decision.

7939             DR. ROYCROFT:  No.

7940             MR. HENRY:  And I take it you would be ‑‑ you would disagree with that statement?

7941             DR. ROYCROFT:  Could you re‑read the statement?

7942             MR. HENRY:  That the VoIP ‑‑ VoIP has transformed the nature and extent of competition in  local telephony.

7943             DR. ROYCROFT:  I would agree that it has the potential to.  Whether it has done that as of today's date, I don't believe it's done that.

7944             MR. HENRY:  Okay.  So, you disagree with that rule.

7945             Let's look forward a bit.  Are you aware of the cable companies projections in this proceeding about the intensity of competition?

7946             DR. ROYCROFT:  Yes, I have reviewed their statements.

7947             MR. HENRY:  And you're aware that they've said that the intensity of competition in the local residence market by May 2007 should be sufficient to put pressure down or pressure on ILEC rates?

7948             DR. ROYCROFT:  Well, I understand their statements, but I also approached my analysis of this market recognizing that what we wind up with if our alternative is cable is a duopoly market. 

7949             Duopoly markets are not known to perform particularly well with regard to price reduction outcomes and I don't think that it's a reasonable expectation that just because you have two firms providing service in the marketplace that we will have an outcome that is sufficient to discipline market power and that it's entirely reasonable to expect that the cable companies would like to see prices go up rather than go down and not necessarily compete in a manner that will result in a competitive outcome.

7950             MR. HENRY:  And this duopoly that you anticipate excludes VoIP over‑the‑top provider wireless loop resellers?

7951             DR. ROYCROFT:  As I indicated earlier, the over‑the‑top VoIP alternative is one that is not widely available for a number of reasons, the first being that consumers have to have a broadband connection and not all consumers have broadband, not even a majority of consumers have broadband as of my last look at the statistics for Canada.

7952             Beyond that, there are a number of limitations associated with over‑the‑top technology that make it an inferior choice that doesn't necessarily suggest that it will be capable of replacing voice usage of voice services for a broad section of the population.

7953             You mentioned wireless.  As well, wireless services have limitations, the first of which being the costs associated with providing wireless service that also limits their substitute ability for a broad section of the telecommunications marketplace, especially in the residential market.

7954             MR. HENRY:  Are you aware in Canada of how many consumers have access to broadband services from cable?

7955             DR. ROYCROFT:  I don't recall the exact percentage, but I know it is quite high, but access does not necessarily imply the same constraint as consumption does and the consumer who does not take broadband will evaluate the purchase price point based on the expenditures associated with the broadband connection as well as an over‑the‑top VoIP connection or service if that's what we're talking about and they will weigh differences in the service quality and other limitations associated with the VoIP ‑‑ over‑the‑top VoIP as well.

7956             MR. HENRY:  In preparing for this hearing, did you make yourself familiar with any of the projections of any of the various well‑known Canadian consulting firms in the telecommunication space as to their projections as to how this will unfold in the next year or two or three?

7957             DR. ROYCROFT:  I may have.  I am not sure of specifically who you are referring to.

7958             MR. HENRY:  Well, people like M.B.I., Michaelson Associates.  Are you familiar with them?

7959             DR. ROYCROFT:  Generally familiar, yes.

7960             MR. HENRY:  Are you aware of their estimates that are on the record of this proceeding?

7961             DR. ROYCROFT:  No, I don't believe I've seen those, no.

7962             MR. HENRY:  How about Lemay‑Yates?

7963             DR. ROYCROFT:  I'm sorry?

7964             MR. HENRY:  Lemay‑Yates, have you heard of them?

7965             DR. ROYCROFT:  Yes.

7966             MR. HENRY:  Are you aware of their predictions?

7967             DR. ROYCROFT:  I may have seen predictions in the past.  If you're referring to specific predictions that are now in the record in this proceeding, I don't believe I've seen those.

7968             MR. HENRY:  Convergence Consulting Group?

7969             DR. ROYCROFT:  I may have.  I can't say for sure.

7970             MR. HENRY:  I take it you would disagree.  You can accept this subject to chat, but people like that have projected anywhere from 27 per cent loss of lines by the ILECS in as early as 2008 and in the case of Lemay‑Yates, I think it's 40 per cent by 2010, which is still well within your four year price cap period.

7971             I take it you would disagree with them.

7972             DR. ROYCROFT:  I would have to see their specific recommendations with regard to how they're referring to lines as to whether they're talking about switched access lines or whether they're including broadband connections and so forth.

7973             So, I couldn't speak specifically to their projections.

‑‑‑ Pause / Pause

7974             MR. HENRY:  Now, in discussing the state of competition, you refer to VoIP services and you say that VoIP does not provide a reasonable substitute for basic telephone service.  Correct?

7975             DR. ROYCROFT:  Over‑the‑top VoIP are we talking about at this point?

7976             MR. HENRY:  Well, you tell me.

7977             DR. ROYCROFT:  Well, if you would like to talk about over‑the‑top VoIP first, that is my position, yes.

7978             MR. HENRY:  But it's not your submission with respect to access dependent VoIP?

7979             DR. ROYCROFT:  Such as that offer by a cable company?

7980             MR. HENRY:  Right, that is a substitute.

7981             DR. ROYCROFT:  Yes, it has a much higher degree of substitute ability, especially if the cable company takes care to provide functionality such as battery back‑up for the service.

7982             MR. HENRY:  Now, I'll confess I don't know if this is a fair question to you, so maybe I'll ask and you can let me know, but I gave some submissions ‑‑

7983             DR. ROYCROFT:  Can I do that for each question?

‑‑‑ Laughter / Rires

7984             MR. HENRY:  I gave some submissions to your counsel of the Consumer Groups on this VoIP issue made over the last couple years and I have to say I'm a bit confused about the Consumer Groups position because in the original VoIP proceeding in 2004, they took the position that VoIP was a substitute for and in the same market as basic telephone service, and they said that this applies for all categories.

7985             Then, in 2005, in the local forbearance proceeding, they took the position that VoIP was not a substitute for residential PES.

7986             But, then, in June of this year, in the VoIP reconsideration proceeding, they took the position that VoIP was a substitute and in the same market as res PES, but then a month later they filed your evidence, which says that VoIP is not a substitute.

7987             I just wondered if you know whether the position you express in your evidence, that VoIP is not a substitute, is just your own view, or does it also represent the views of your client, at least in October?

7988             DR. ROYCROFT:  I think I am best qualified to speak to my own evidence, and to also clarify your statement with regard to my position on VoIP.

7989             We are talking about over‑the‑top VoIP with regard to the limitations associated with substitution.

7990             MR. HENRY:  Right, and they were, too, but if you don't want to get into that, that's fine.  Perhaps Mr. Janigan and I can deal with it in argument.

7991             In any event, you would be aware, would you, that the CRTC has found that VoIP ‑‑ all flavours of VoIP ‑‑ are substitutes for and in the same market as residential primary exchange service?

7992             DR. ROYCROFT:  Given that finding, I believe it is still reasonable to consider the impact of the various flavours of VoIP on market power, and that is the key issue here when we are talking about lifting pricing constraints, in that if we have evidence that consumers can readily substitute among a number of alternatives that are approximately comparably priced, then there is a greater chance that market power will be eroded and that regulation can step aside.

7993             At this point the market share evidence and the technology alternatives that I am familiar with, based on my evaluation and research for this proceeding, still point to a high degree of market power, significant segmentation in markets, based on geography and on customer class and ‑‑

7994             MR. HENRY:  That is not really my question.  My question is, is it in the same market?

7995             I take it that you say it is not and you disagree with the Commission, which says that it is.

7996             Is that fair?

7997             It is all right to disagree with the Commission.  We have been known to do it from time to time.

‑‑‑ Laughter / Rires

7998             DR. ROYCROFT:  I think I have testified several times here already that I don't believe that over‑the‑top VoIP is in the same market as standard primary exchange service.

7999             MR. HENRY:  That's fine.

8000             You mentioned wireless.  Again, you also don't think that is a substitute.

8001             DR. ROYCROFT:  That is correct.

8002             MR. HENRY:  This is a position you have taken in the U.S., as well.  Correct?

8003             DR. ROYCROFT:  Yes.

8004             MR. HENRY:  In fact, you took that position fairly recently in California, didn't you?

8005             DR. ROYCROFT:  Yes, I did.

8006             MR. HENRY:  As I understand it, the California Public Utilities Commission concluded, despite your evidence, that wireless services are in the same market as basic wireline voice service.  Correct?

8007             DR. ROYCROFT:  Yes, that was one of their erroneous conclusions.

8008             MR. HENRY:  In fact, they found that landline and mobile services are substitutes and not mere complements, as you had tried to convince them.

8009             DR. ROYCROFT:  That is what the Commission's conclusion was, yes.

8010             MR. HENRY:  Thank you.

8011             Now come the numbers.

8012             I want to go through your X factor.  Before I do, I have prepared a summary document, which is mostly from your evidence, with a tiny bit more information on it.

8013             It is at Tab 2 of your binders, Mr. Chairman, and it is called "Dr. Roycroft's Recommendations for Calculating the X Factor."

8014             THE CHAIRPERSON:  Mr. Henry, just give the Secretary a minute to number this exhibit.

8015             THE SECRETARY:  I am giving it Companies Exhibit No. 11.

EXHIBIT NO. THE COMPANIES‑11:  Dr. Roycroft's recommendations for calculating the x-factor source: table 8 p. 58 of Dr. Roycroft's testimony

8016             MR. HENRY:  I want to go through, very quickly, the arithmetic, and then I promise you that I will come back and we will go through each number.

8017             As I understand it, you started with ‑‑ and we might want to keep this out throughout the cross‑examination, because, as I say, we will keep coming back to each of the numbers.

8018             You started with a historical TFP, that is the Total Factor Productivity number, as a starting point.  You used that rather than the marginal cost studies that the Companies did.  Correct?

8019             DR. ROYCROFT:  The marginal cost studies that the Companies had previously done ‑‑

8020             MR. HENRY:  And in this proceeding.

8021             DR. ROYCROFT:  Yes.

8022             MR. HENRY:  This TFP number of 4.2 percent is based on previous TFP results filed with the Commission that showed historical TFP that was for the period 1988 to 1995.

8023             Is that correct?

8024             DR. ROYCROFT:  That is correct.

8025             MR. HENRY:  When you use TFP as a starting point, the accepted methodology for an X, as I understand it, is to then subtract the economy‑wide productivity growth, which you calculate at 1.06 percent, and that is Canadian data from the period 1995 to 2004?

8026             DR. ROYCROFT:  That is correct.

8027             MR. HENRY:  Again, I think the accepted methodology when you start with a TFP is to add back an input price differential, which is the difference between the rate of inflation in economy‑wide prices and the rate of inflation in prices paid by the ILEC for its inputs, and you calculate that at 2 percent.

8028             DR. ROYCROFT:  Yes, with the caveat that the differential is best reflecting the economy‑wide input prices as opposed to just the economy‑wide prices.

8029             MR. HENRY:  Fair enough.  I should have said that.

8030             The inflation in the economy‑wide prices compared to the rate of inflation in the ‑‑

8031             DR. ROYCROFT:  Inflation in the economy‑wide input prices.

8032             MR. HENRY:  Right, input prices.

8033             And then you add a stretch factor of 1 percent, which you justify, in part, on your findings that economies of scope resulting from the provision of DSL services can be expected to add .57 percent to productivity.

8034             DR. ROYCROFT:  No.  I am suggesting that the stretch factor is to reflect the overall scope economies and how well they are reflected ‑‑ or how well the Commission feels they are reflected in the total productivity study.

8035             MR. HENRY:  Fair enough.  You use an example with DSL to get .57.  You are saying there may be other economies of scope.

8036             Is that correct?

8037             DR. ROYCROFT:  Right, and the example with regard to DSL is illustrative with regard to the ‑‑

8038             MR. HENRY:  Sure.  Understood.

8039             DR. ROYCROFT:  ‑‑ fact that it is based on U.S. data.

8040             MR. HENRY:  I understand.

8041             But using that .57 percent, you get a resulting ‑‑ the math is a resulting X factor range of 5.14 percent to 6.14 percent, and then you suggest 6 percent.

8042             DR. ROYCROFT:  You linked this, once again, to the .57 percent.  The 5.14 to 6.14 is based on whether or not there is any stretch factor added, with the upper limit being suggested at 1 percent, which would give you the 6.14.  A stretch factor of zero would lead to the 5.14.

8043             MR. HENRY:  Fair enough.

8044             Let's go back and look at each of these numbers.

8045             Let's go back to the top.  The 4.2 percent ‑‑ the TFP that you use as a starting point, it measures productivity for the company as a whole, rather than for regulated services only, does it not?

8046             DR. ROYCROFT:  Depending on how the study is structured, yes.

8047             MR. HENRY:  Would you also agree that if the Commission were to develop an X factor, it would be applied to the regulated capped services?

8048             DR. ROYCROFT:  That is correct.

8049             MR. HENRY:  And, to the extent possible, the X factor should represent the productivity growth associated with the regulated services of the company.

8050             DR. ROYCROFT:  With the caveat that some of those regulated services may be sharing inputs that are unregulated and, if there is productivity associated with those unregulated services, it is fair to at least recognize that these alternative sources of productivity are available and that that productivity can have a positive impact on the overall structure of the productivity offset.

8051             MR. HENRY:  Those services, to the extent that they increase the productivity of the other service, as a result of economies of scope, should be reflected.

8052             DR. ROYCROFT:  Correct.

8053             MR. HENRY:  You are not suggesting that productivity within the other service itself should be reflected in the regulated service, but to the extent that the other service ‑‑

8054             DR. ROYCROFT:  When you are talking about shared inputs, it may be difficult to allocate or assign the productivity ‑‑

8055             For example, when DSL is added to the overall service set and it shares significant investments associated with local loop, that, in itself, then expands overall productivity and it is difficult to say that 5 percent of this or 95 percent of this should be assigned to a specific service.

8056             If you ignore the productivity gains that are associated with expanding services such as the addition of DSL, then you are implicitly saying that they should all go to the new service and that the residual services, those that have relied on the local loop prior to the introduction of DSL, should not enjoy and of those productivity benefits.

8057             MR. HENRY:  I assure you we will come to that, but we will do it when we get out of the stretch factor.  Let's stay with the 4.2 for a moment.

8058             You are aware that in the last price cap decision of this Commission, the Commission rejected an approach based on TFP for the company as a whole and concluded that the productivity offset should be based on service‑specific marginal costs in order to reflect the actual productivity gains that are likely to be achieved for individual capped services ‑‑ capped baskets?

8059             DR. ROYCROFT:  Yes.  And that approach, at least based on my ability to review information on how that calculation was made, implicitly assumes that all of the costs of shared inputs such as the local loop are solely attributable to primary exchange service and therefore prohibits any sharing of productivity gains from the introduction of new services like DSL and the corresponding increase in economies of scope.

8060             MR. HENRY:  Let's turn to another document at Tab 3 of your binder that I gave to your counsel on Friday.  That is your rebuttal testimony filed with the Indiana Utility Regulatory Commission in January 2004.

8061             DR. ROYCROFT:  That is correct.

8062             MR. HENRY:  Could you turn to page 26, starting at line 13.

8063             THE CHAIRPERSON:  Which tab?

8064             MR. HENRY:  Tab 3.

8065             THE SECRETARY:  That would be Exhibit No. 12.

EXHIBIT NO. THE COMPANIES‑12:  Dr. Roycroft's Rebuttal Testimony before the Indiana Utility Regulatory Commission - Cause No. 42405, 13 January 2004

8066             MR. HENRY:  If you will bear with me I just want to read your testimony here, starting at line 13 on page 26.  You start by asking yourself a question:

"Dr. Currie indicates that when measuring productivity the totality of an industry's outputs must be measured.  Do you agree?"  (As read)

8067             And you answer:

"I would agree as long as the scope of the productivity study is first appropriately defined.  If one is interested in measuring the productivity growth associated with the totality of the industry, then measuring the totality of an industry's output is appropriate.  However, the focus of the study that I performed was on the regulated operations of the RBOC industry and as a result I believe that it is important to try to capture the best representation of the RBOC industry's regulated operations.  Inclusions of all aspects of the RBOC industry would not be appropriate for rate‑making purposes.  The RBOCs engage in a wide variety of operations from regulated telephone service to foreign investments, commercial real estate, thus narrowing the focus to regulated operations is appropriate and thus is the approach that I utilized in my study.  Inclusion of the totality of RBOC operations could result in a biased TFP calculation."  (As read)

8068             So you yourself have recognized and advocated an approach to productivity that is confined to the regulated activities of the company.

8069             Is that correct?

8070             DR. ROYCROFT:  The study that I performed associated with this testimony focused on the regulated activities, although by discussion in this testimony, especially in the direct filing of the testimony, discussed the impact of DSL and other technologies such as wireless on the productivity potential of the companies.

8071             So it is quite clear that the specifics of my argument directed at Dr. Currie is of a much more general and broad nature, focusing, as indicated in the passage that you read, on operations outside of the United States, foreign investments, commercial real estate ventures that have nothing to do with the question at hand, the question at hand being productivity growth associated with the regulated telecommunications operations.

8072             MR. HENRY:  Was DSL a regulated service?

8073             DR. ROYCROFT:  Pardon me?

8074             MR. HENRY:  Was DSL a regulated service?

8075             DR. ROYCROFT:  At the time, yes, although it was regulated in a different jurisdiction.

8076             MR. HENRY:  Your statement here is to focus on the regulated.  Okay.  Thank you.

‑‑‑ Pause

8077             MR. HENRY:  As we established, the 4.2 percent starting point of TFP, that is based on historical Canadian telco data from the period 1988 to 1995.

8078             You are not using anything more recent?

8079             DR. ROYCROFT:  I did not have the opportunity as part of the procedure to conduct discovery or otherwise address the calculation afresh of TFP for the Canadian companies.

8080             The United States regulatory set‑up offers some access to publicly available information that is not available in Canada, so I could not go about constructing such a study.

8081             I will note that the period in question here is one where technological change was much more limited than it is today, so I would view the 4.2 percent reflecting a period where we have conservative technological change and declines in input prices that were not as ‑‑

8082             MR. HENRY:  Not as much productivity in the time period?

8083             DR. ROYCROFT:  That is correct.  There certainly was ‑‑

8084             MR. HENRY:  So you have assumed that that is conservative?

8085             DR. ROYCROFT:  That is correct.

8086             MR. HENRY:  But wouldn't that period include outputs for that period that would include high‑growing services such as long distance that were very high‑growing in those days?

8087             Wouldn't that contribute to high productivity?

8088             DR. ROYCROFT:  It would depend on the overall share in the output mix.

8089             MR. HENRY:  You haven't looked at that?

8090             DR. ROYCROFT:  No, the studies were not available to me.

8091             MR. HENRY:  What about data services?  Wouldn't those have been high growth in those days?

8092             DR. ROYCROFT:  Well, depending on what sort of data services you are referring to.  The period in question has a part of the uptake of data services.  I would say that you had very rapid expansion in data services beginning in the mid‑1990s.  Certainly there was expanding data service provisioning and demand in that period, but the most robust growth has been after the commercialization of the Internet.

8093             MR. HENRY:  But in any event, more importantly, these services are not regulated today, are they?

8094             DR. ROYCROFT:  That is correct.

8095             MR. HENRY:  So let's turn to the next factor, No. 2 on the list, Canadian economy‑wide multi‑factor productivity.

8096             That is the economy‑wide productivity growth and back on the table we see that factor is 1.06, which is subtracted from the 4.2 starting point.

8097             As you show on your table, that data is from Statistics Canada for the period 1995 to 2004, which is not the same period, is it, on which the TFP of 4.2 percent is based, is it?

8098             DR. ROYCROFT:  No, it is not.  And I selected a more recent time span to bring up to date what I could.  I certainly would have liked to have brought a more up to date TFP calculation to my testimony, but I did not have that capability.  I didn't see any reason not to reflect the more recent periods economy‑wise multi‑factor productivity.

8099             MR. HENRY:  On the TFP you make an assumption that it's a proxy for today and then you use the most recent data for the economy‑wide?

8100             DR. ROYCROFT:  I assumed that the industry‑wide TFP growth is one that provides a conservative approximation of what would be capable today.

8101             MR. HENRY:  All right.  An assumption.

8102             Let's turn now to the input price differential, which is No. 3 on the list.  Yes, No. 3 on the table that we handed out.

8103             You calculate a +2 percent adjustment there and let me just understand.

8104             The input price differential, as I understand it, is a measure of the difference between input price inflation and the economy as a whole and inflation in the price of inputs used by the telephone company.

8105             Is that correct?

8106             DR. ROYCROFT:  Yes.

8107             MR. HENRY:  So your proposed input price differential of 2 percent means that on average you expect the prices paid by the telephone company for the goods and services it uses as inputs to produce its goods and services will be 2 percentage points less than the economy‑wide input price inflation rate.

8108             Do I have it?

8109             DR. ROYCROFT:  That is correct.

8110             MR. HENRY:  So if the economy‑wide input price inflation were to use 2 percent, you would expect the telephone company to experience input price inflation of zero?

8111             DR. ROYCROFT:  Yes.

8112             MR. HENRY:  You have used GDPPI as one proxy for the economy‑wide plant input price differential.

8113             Is that correct?

8114             DR. ROYCROFT:  I report the information on the GDPPI.  The primary basis of my comparison was with the economy‑wide input price inflation for the non‑farm business sector, which is shown in the table in my testimony.

8115             MR. HENRY:  And you say that was the primary one.

8116             Did you say that in your evidence?

8117             DR. ROYCROFT:  I'm not sure that I did, but it is certain ‑‑

8118             MR. HENRY:  It is 20 percent, and I kind of thought the primary one was the other one, the 1.92.

8119             But it doesn't much matter.  You used two different inflation measures, and one gave you 2.96 and one gave you 1.92.  And you picked 2.0.

8120             DR. ROYCROFT:  Right, which is between the two.

8121             MR. HENRY:  Is it fair to say that when we are looking at this factor, the input price differential, what we are looking at is the rate of change in unit prices only, independent of quantities of input?

8122             It is the prices.

8123             DR. ROYCROFT:  No.  The indices involved are standard price indices which also reflect quantity.

8124             MR. HENRY:  Isn't the quantity taken care of in the TFP and now you are looking at price?

8125             DR. ROYCROFT:  We are looking at ‑‑

8126             MR. HENRY:  Let me give you an example.

8127             I understand this is the common way to do this in all textbooks, as some of my friends are fond of saying, but let me give you a simple example.

8128             If I use two pieces of equipment at a total cost of $10 one year, the average price that year is $5.00.  Correct?

8129             DR. ROYCROFT:  Two pieces of equipment...

8130             MR. HENRY:  For $10.  The average price is $5.00.

8131             DR. ROYCROFT:  Right.  And you couldn't have figured out the average without the quantity.

8132             MR. HENRY:  That's why I said two is the quantity.

8133             DR. ROYCROFT:  Right.

8134             MR. HENRY:  If I use three pieces of equipment at a total cost of $18 the next year, my average price has increased to $6.00.

8135             DR. ROYCROFT:  Yes.

8136             MR. HENRY:  In that example, the input price inflation would be $1.00: six minus five; in other words, one over five, 20 percent.

8137             That would be the input price differential in that example, would it not?

8138             Not the input price differential, the input price.

8139             DR. ROYCROFT:  I'm sorry, it would be what?

8140             MR. HENRY:  It would be the input price inflation.

8141             The $5.00 unit price would have gone up to $6.00.  That would be $1.00, or 20 percent.

8142             DR. ROYCROFT:  Yes.

8143             MR. HENRY:  Okay.

8144             You say in your evidence that you have attempted to capture this input price differential on a forward looking basis.  Correct?

8145             DR. ROYCROFT:  Yes.

8146             MR. HENRY:  As I understand it, you have conducted a study of telecom input price for the RBOCs in the U.S. over the period 1995 to 2004?

8147             DR. ROYCROFT:  Yes.

8148             MR. HENRY:  Could you turn to an exhibit at Tab 4 that I provided to you counsel.

8149             It may not be labelled.  It is at Tab 4 in your book.  I think we have labelled it this morning "Input" ‑‑

8150             THE SECRETARY:  Excuse me, Mr. Henry.

8151             I note in the binder that Tab No. 4 and Tab No. 5, the attachments are identical.  So I just want to make sure.

8152             MR. HENRY:  Actually, I can explain that, Madam Secretary.

8153             THE SECRETARY:  That's okay.  So we are talking ‑‑

8154             MR. HENRY:  I don't fault you for coming to that conclusion.  I hope it is only the first page that is identical and that we have changed the title, and that the data and pages behind are quite different.

8155             THE SECRETARY:  That will be Exhibit No. 13.

8156             MR. HENRY:  Thank you.

EXHIBIT NO. THE COMPANIES‑13:  The IPD Trend

Source: p. 27 of Dr. Roycroft's testimony

8157             MR. HENRY:  The first page of this is just a photocopy of Table 3 of your evidence where you summarize your data.  I have attached it really just for convenience here so we don't have to keep flipping forward to two books.

8158             This table, as I understand it, presents your data for the plant portion of the inputs.

8159             On Line A at the end, you have an average of minus .04 percent for the telco input plant price inflation.

8160             Is that correct?

8161             DR. ROYCROFT:  Yes.

8162             MR. HENRY:  If I understand correctly, if we go down to the last row of the table, we see one calculation of the telecom plant input price differential, which is the difference between telecom plant input prices and economy‑wide inflation for each of the years 1995 to 2004 for that period.

8163             DR. ROYCROFT:  Yes.

8164             MR. HENRY:  The last row is the difference between the economy‑wide measure of GDPPI and the telecom plant price index you calculate in Line A.  So the bottom corner 1.92 percent is the difference between the 1.88 percent, which is the GDPPI a couple of lines up, and the minus .04 for the telecom.

8165             The difference between that is 1.92.  And that is one of the ways you calculated input price differential.  I recognize you did the other way as well.

8166             DR. ROYCROFT:  That is correct.

8167             MR. HENRY:  I am curious.  If we are just looking at telecom plant differential, why did you compare the telecom plant price index to the economy‑wide measures such as GDPPI or the non‑farm inflation index?

8168             Both of those things, as I understand it, include more than plant.  They include labour, they include materials and other things.

8169             DR. ROYCROFT:  They were benchmark measures that were available, and the Federal Communications Commission, when it was conducting price cap regulation, relied on the non‑farm input price inflation differential as the benchmark to compare telecom price changes.

8170             So I found it to be a recognized benchmark to utilize.

8171             MR. HENRY:  They used it to compare to telecom plant?

8172             DR. ROYCROFT:  To telecom input price inflation.

8173             MR. HENRY:  To telecom input price inflation.

8174             DR. ROYCROFT:  Yes.

8175             MR. HENRY:  You are comparing this to telecom plant input price inflation.  Did the FCC do it that way?

8176             DR. ROYCROFT:  No.

8177             MR. HENRY:  Okay.

8178             DR. ROYCROFT:  But I separately evaluated the impact of labour and materials.

8179             I generally assumed that given that we are interested in a differential between price trends, the contribution of materials prices would be small from the standpoint that if one of the RBOCs like SPC is buying a pick‑up truck or buying paper clips, it is likely to be facing prices that are similar to other large corporations that are making similar purchases.

8180             MR. HENRY:  Another assumption you made.

8181             DR. ROYCROFT:  It is a hazard of the trade.  If you are an economist, you run into a few assumptions from time to time.

8182             MR. HENRY:  Fair enough.

8183             As I understand it, the 1.92 percent on the bottom corner represents an average of the rates of change in the plant input price differential that you have calculated for each of the years 1995 to 2004.

8184             So we add up those and we get an average of 1.92 percent on the last row.  Correct?

8185             DR. ROYCROFT:  That is correct.

8186             MR. HENRY:  When we looked at that row, it seemed to us that the differential peaked in 1999 at 4.07.  Do you see that?

8187             And then it steadily declined after that every year until 2004.

8188             DR. ROYCROFT:  If we are looking at the GDPPI alone, that is a fair assessment of the trend exhibited.

8189             MR. HENRY:  Well, we are looking at the differential.

8190             DR. ROYCROFT:  The differential based on GDPPI as opposed to the differential above that, which has slightly different characteristics.

8191             MR. HENRY:  Fair enough.  So the 2.96 might be different.

8192             Rather than the average of 1.92 percent, we attempted to project the trend line out into the future beyond 2004.  So we performed an extrapolation of the data rather than using an average, as you did.

8193             The data is on page 2, but I think the fastest way I am hoping to do this ‑‑ I think it is most illustrative if we turn to the third page of that exhibit where it is graphed.

8194             What we did was do an extrapolation using a common technique ‑‑ well, common to my associates anyway ‑‑ least squares regressions to extrapolate the data into 2005, 2006, 2007 and 2008.

8195             Let me just explain these three lines.

8196             The line with the boxes is the raw data, the one that peaks and then falls.

8197             The flat line with the diamonds is a representation of what you did.  It is the 1.92 percent average of that.

8198             The straight sloped line with the triangles, I guess they are, represents our extrapolation.

8199             When we extrapolated that way, we found that out in 2007 and 2008, the first two years ‑‑ in our proposition, the only two years of the next price cap proceeding ‑‑ we got an average between those of 1.17, instead of your 1.92.

8200             Would you agree with me that that's another way to do this?

8201             DR. ROYCROFT:  I would agree; however, I think your approach is incorrect, for a number of reasons, the first reason being that it focuses on only one of the differentials, and the differential which has the more favourable trend from your perspective.  It ignores the data that is available on the differential for the non‑farm input price inflation in the row above that.

8202             The other points that I would make with regard to your projection would include the fact that when you make a progression based on a regression analysis, it is typically the protocol to provide some summary statistics about your regression and to provide a confidence interval associated with your projection, and I don't see that information here.

8203             Lastly, I would note that, in determining your average, the 1.17 percent that you just highlighted, that average is not based on your projection, but rather on a subset of your projection, in that you only look at two of the points, and I see no reason to exclude the other data points that you have calculated.

8204             MR. HENRY:  We just projected out to the two years of the price cap.

8205             In fact, we probably did that based on what you said in paragraph 47, that the components have to be forward looking.

8206             DR. ROYCROFT:  And the series of data points that I have identified in my Table 3 are data points that I believe are consistent with the estimation of a forward looking differential.

8207             In calculating the 1.17 percent average that you show on the second page, you are ignoring your own data and your own calculations.  If you were to include those, then you would have a higher average projection of 1.27 percent rather than 1.17 percent.

8208             Once again, that projection is only based on partial information that is presented in Table 3, and you have ignored the data that doesn't show a declining trend over time, uniformly.

8209             The non‑farm input price inflation row, which is the second row from the bottom in Table 3, shows, too, a peak in 1999, then a decline, and then an increase, which is different from what is in the final run.

8210             MR. HENRY:  Fair enough.  I am only looking at one of your two, but I am trying to suggest that, depending on your assumptions ‑‑ you had a 1.92, and then you had this 2.96.  If we just look at the 1.92, you could just as easily conclude that it is 1.17.

8211             I am not suggesting that this is necessarily the only way to do it, but just eyeballing the line, it looks better than yours, and, at least, is an alternative to yours.

8212             You have assumed it's an average; why can't we assume it's a trend?

8213             DR. ROYCROFT:  You raised the "assumption" word here, and, of course, you are free to assume whatever you like.  My point is, you have ignored specific information that is provided here to draw your conclusions.  Not only do you ignore the information that was contained on my table, you also ignore information that you have generated.

8214             And in calculating your average for 2007 and 2008, you throw out two data points that, apparently, you feel are not representative.  If those data points aren't representative, then why are the ones that you have homed in on and selected representative?

8215             MR. HENRY:  Dr. Roycroft, we looked at your TFP studies.  Do you recall something called "Dial Equipment Minutes" in your .57 calculation?

8216             DR. ROYCROFT:  Yes, Dial Equipment Minutes.

8217             MR. HENRY:  We noticed something very similar to this.  You had to project holding times out into 2001, 2002, 2003 and 2004, because you had a historical series, and the line was going like "this", and, interestingly enough, you didn't take the average, you projected it out on, actually, quite a steep scale.

8218             Do you recall doing that?

8219             DR. ROYCROFT:  I did not use a linear regression to do that, I used averages to do that.

8220             MR. HENRY:  You used averages to do that.  You did not do an extrapolation?

8221             DR. ROYCROFT:  I don't believe that I did a linear regression.

8222             MR. HENRY:  Can you confirm that for us, exactly what you did?

8223             In fact, I am informed that you did not ‑‑ again, subject to check on your part ‑‑ you did not assume a flat trend.  You did not assume a constant average.  You projected out by assuming a continued rate of growth.

8224             I didn't think I was going to have to go into this.  We did plot it on a graph.  I could leave it with you and ask you to undertake to confirm that we did it correctly.

8225             DR. ROYCROFT:  You could ask me to undertake ‑‑

8226             MR. HENRY:  Undertake to confirm that we did it correctly.

8227             DR. ROYCROFT:  Did what correctly?

8228             MR. HENRY:  Took your data and just plotted how you did your projection.

8229             Could I leave that with you?  I didn't expect to have to go there ‑‑

8230             COMMISSIONER LANGFORD:  Mr. Henry, from my point of view, you have asked him a question, and, at least, we should give him the opportunity to answer the question as to how he did his work.

8231             MR. HENRY:  Okay.

8232             COMMISSIONER LANGFORD:  Then, if you want to leave something with him following from that ‑‑

8233             MR. HENRY:  That's fine.

8234             COMMISSIONER LANGFORD:  That is only a personal opinion, but I am sitting at the edge of my chair waiting for his answer.

‑‑‑ Laughter / Rires

8235             MR. HENRY:  So am I.

8236             I can give you the references.

8237             In Consumer Groups/The Companies 8(j), and then your discussion at pages 68 and 69 of your evidence, you make reference to this.

‑‑‑ Pause

8238             MR. HENRY:  I am reminded that the exhibit you filed this morning talks about, for the period 2001 to 2003, projected DEMs per call.  Projected DEMs.  I am suggesting to you that you projected those DEMs on an extrapolation basis.  You did not assume a constant average, you did what we did, or something very similar.

8239             DR. ROYCROFT:  The projection with regard to the Dial Equipment Minutes was done on a basis of average growth over a period of time.  I don't have the electronic version of the document, but I believe it was for a four or five‑year period prior to the last data point.

8240             I did not use regression, and I did not take the approach that you have identified here, but I did not project it in the way that you suggest, that I was just using the last data point.

8241             MR. HENRY:  Whether you used a regression or not, there are a number of ways to extrapolate.  My point is, you did not use a flat line average, you projected it out, and on a steep curve.  You took the last point ‑‑

8242             The graph goes like "this", and then you have put it out.

8243             As I say, I have a graphic representation.  If you want to take it away and study it, and give us an undertaking to look at that, I would be quite happy with that.

8244             DR. ROYCROFT:  The actual data that I did have in the DEM projection was going on an upward incline, and I took an average based on a subset of periods toward the end of the overall number of data points and used that to project forward.

8245             MR. HENRY:  Okay.  Fine.  I think you have confirmed what I was trying to confirm.  Thanks.  We don't need to go there.

8246             Now, so far we have been looking at telecom plant price inflation versus economy‑wide plant price inflation.  But as you quite rightly point out in your evidence, there is another important component of input prices and that is the labour component or wage inflation.  So I want to turn to that component.

8247             Again, just as was the case for plant, with labour we are looking at the rate of change in the labour rate or wages only.

8248             Is that correct?

8249             DR. ROYCROFT:  I looked at the total compensation which would encompass wages, benefits, insurance and also reflect the quantity of labour at a particular point in time.

8250             MR. HENRY:  I thought it was accepted that you are trying to get a price differential here and you are including quantity.  Again, let me give you the example out of the textbooks I have been reading.

8251             Another simple example:  I have three employees at a total wage of $150,000 one year, the average year would be $50,000 per year.

8252             Is that correct?

8253             DR. ROYCROFT:  The average wage rate, yes.

8254             MR. HENRY:  If I had two employees the next year at a total wage of $120,000, my total wages may have gone down, but my average wage would have increased from $50,000 to $60,000 and so the wage inflation rate would be $10,000 or 20 percent.

8255             Is that correct?

8256             DR. ROYCROFT:  That is correct.  However, your total compensation would have a declining rate of inflation from the standpoint that your wage bill has gone from $150,000 down to $120,000.

8257             MR. HENRY:  Well, that is because you have included the quantities, but here we are supposed to be just looking at price.  You confirmed that earlier with equipment, you confirmed it with wages.

8258             TFP does the quantity.  The input price differential is supposed to be price.  That's how you did it with all the other components.  It's price, unit price.

8259             DR. ROYCROFT:  But to calculate an index it has to have quantities associated with it.  That is the way that indices are calculated, while spares or partial methods use quantities.  You can't get away from it if you want to create an index.

8260             MR. HENRY:  But quantities are weights only, not the total.

8261             Is that correct?

8262             DR. ROYCROFT:  The weights are certainly important to acknowledge from the standpoint that it recognizes how the market basket that you are focusing on changes with regard to the ability of the consumer or the firm to substitute for the various components of the basket.

8263             MR. HENRY:  But you are weighting prices.

8264             DR. ROYCROFT:  Right.  And the weights are based on quantities.

8265             MR. HENRY:  Well, do you deny the example we just went through, that the wage inflation in the example I gave you was $10,000 or 20 percent?

8266             DR. ROYCROFT:  No, with the caveat that the compensation inflation was a decrease, a negative amount because the total bill went down.

8267             MR. HENRY:  All right.

8268             Well, let us say that we did it on a wage inflation basis, as I have put to you.

8269             DR. ROYCROFT:  Just looking at wage ‑‑

8270             MR. HENRY:  At the wage price, just like you did with the plant, just like you did with everything else, it was on a quantity unit‑price basis.  Let's do the same thing with labour.

8271             You state in your evidence that you also analyzed the labour cost transfer, the U.S. private business sector and the RBOC industry and found that the input labour prices for the RBOCs were growing less rapidly than those in the general business sector and that the difference was also about 2 percent, which is the same number you estimated for the plant component.

8272             Is that correct?

8273             DR. ROYCROFT:  Right.  Compensation is the focus of my study, looking at the difference in total compensation.

8274             MR. HENRY:  Right.  So I want to look at this 2 percent labour component a little bit with you and see how it was calculated.

8275             Could you turn to the exhibit, please, at Tab 5 which we have labelled "Input Price Differential Labour Component"?

8276             Again, the first page is just the same page from your evidence.

8277             On reflection, we didn't need to attach this one here because it is not really relevant to labour.

8278             Mr. Chairman, before your eyes glaze over ‑‑

8279             THE CHAIRPERSON:  Actually, we need a number too, I think.

8280             THE SECRETARY:  Exhibit No. 14.

EXHIBIT NO. THE COMPANIES‑14:  The IPD - Labour component

8281             MR. HENRY:  There is a lot of data here.  It was given to Dr. Roycroft to give him an opportunity to see what we had done.

8282             What I would like to do, and I'm hoping I can do it this way, is just go to the punch line, the second to last page, describe our conclusions and his conclusions and then let him comment on that and then see if we can go from there.

8283             Is that a fair way to do it?  We will give the conclusions and then you can tell us why we are right or wrong.

8284             DR. ROYCROFT:  It sounds reasonable.

8285             MR. HENRY:  So that is on the second last page.  it looks like this.

8286             If we turn to the second line, the first line with the numbers on there, if we look at the first row of numbers, that represents a summary of your results.  You looked at the total economy‑wide U.S. wage inflation rate for two time periods and you got 3.262 percent and 3.89 percent.  Then you compared it to the RBOC wage inflation rate which you calculated at 1.6 and 1.73 percent, or again the two different time periods, and the difference between those shows that the RBOC wage inflation exceeds the economy‑wide U.S. average by 2.02 percent and 2.15 percent for the two different periods.

8287             That's what you did?  So far, so good?

8288             DR. ROYCROFT:  That is correct.

8289             MR. HENRY:  Now the next line of numbers is what we did.

8290             Again, based on theories of input price differentials as we understand them, and as I think you agreed earlier in our elementary examples, we took your data for the RBOCs and we took U.S. Bureau of Labour statistics for the U.S. average, and all we did was take the total compensation and divide it by the total employees, because what we found was the total employees was changing, dropping significantly for the RBOCs, sometimes in half over this period, but the total compensation was moving in a much different direction.

8291             So what we got as a result of that analysis was, for the U.S. average we got 2.34 percent and 3.68 percent, depending on the period.  This is on a per‑employee basis.

8292             Then when we did the RBOC averages we found something interesting.  We found the RBOCs were 5.8 percent and 8.13 percent, again depending on the period.  So what that shows is that the RBOCs wage inflation was actually considerably more rather than less, as you had calculated, than the U.S. economy as a whole by ‑3.46 percent and ‑4.5 percent.

8293             That is a swing of 5.5 to 6.5 percentage points from yours, isn't it?

8294             DR. ROYCROFT:  It is certainly a different number.

8295             MR. HENRY:  What is wrong with it?

8296             DR. ROYCROFT:  The approach that you have utilized is essentially abstracting from the conventional methods used to create price indices from the standpoint that you are focusing stricture on the price and not on the weights associated with the creation of the index.

8297             MR. HENRY:  Well, when you did your plant ‑‑ let me go back to the plant one ‑‑ was that prices or was that quantities?  Did you just take the total quantities they spent on switches and compare it to the total quantities of plant in the economy as a whole or did you convert it to a price?

8298             DR. ROYCROFT:  It was reflective of weights associated with the plant categories and those weights essentially allocate the price increases based on the quantities reflected in the RBOCs RMIS data.

8299             MR. HENRY:  Back in my example, I thought you confirmed to me ...

‑‑‑ Pause

8300             MR. HENRY:  Is the telephone plant the total expenditures on telephone plant or is it the prices weighted by the quantities?

8301             DR. ROYCROFT:  Can you take me to something specific as far as your reference?

8302             MR. HENRY:  When you did your telephone plant price index, was it just the total expenditures year‑over‑year, or did you weight it by the quantities?

8303             DR. ROYCROFT:  It was weighted by quantities.

8304             MR. HENRY:  With your total compensation, isn't it just the total compensation?  You didn't weight it by the quantities as near as we can tell.

8305             DR. ROYCROFT:  The total compensation reflects the quantities.

8306             MR. HENRY:  Well, of course it does, because it is the total compensation.  But it is not weighted by the quantities.

8307             DR. ROYCROFT:  It is a different type of index from the standpoint that it only has one series in it as opposed to the multiple series that were included in the plant index.

8308             So the weighting process doesn't look the same.

8309             MR. HENRY:  Okay.  I think we understand what you have done.  We can leave this for argument.

8310             One last thing.  Could I turn to the last page of the exhibit.

8311             We wanted to get a sense of ‑‑

8312             DR. ROYCROFT:  This is No. 5?

8313             MR. HENRY:  The same exhibit, yes, the very last page with the number 28 on the bottom.

8314             There is a whole bunch of numbers with the number 28.

8315             DR. ROYCROFT:  Yes.

8316             MR. HENRY:  All we did there is we were just trying to get a sense of what proportion of any telco input price differential could reasonably be attributed to wage inflation.  So we went to the data in your TFP studies which you did for 14 U.S. telcos for the purposes of your economies of scope discussion, which we will be coming to, and we found, based on your data, that about 28 percent of total telco input volumes in that study were labour.

8317             We have set out how we did that on the last page.

8318             As I say, it is a weighted average of 28 percent.

8319             Does that sound about right?  Does it look like we did it right?

8320             DR. ROYCROFT:  You take the average of the entire period for each state and then average those averages to come up with the .28.

8321             MR. HENRY:  That's a reasonable approximation of the weight of labour in the input prices, based on your own data.

8322             DR. ROYCROFT:  It is one approach to take.  It certainly is taking a snapshot across a broad period.  If you are looking at a forward‑looking perspective, you might want to focus on a different subset of time.

8323             I do not dispute the calculation.

8324             MR. HENRY:  It is your data.

8325             DR. ROYCROFT:  But it is your calculation.

8326             MR. HENRY:  You wouldn't deny that labour is a fairly significant component to telco input prices?

8327             DR. ROYCROFT:  It is certainly non‑trivial, but the telecommunications industry is a highly capital intensive industry that has been shedding labour significantly over the past ten years.  Therefore, its importance has declined.

8328             MR. HENRY:  And capital has been dropping too, hasn't it.

8329             DR. ROYCROFT:  I'm sorry?

8330             MR. HENRY:  And capital has been dropping too, hasn't it.

8331             DR. ROYCROFT:  Capital as a factor share?

8332             MR. HENRY:  Right.  You have no evidence to suggest that 28 percent isn't a reasonable approximation of the total input price.

8333             DR. ROYCROFT:  Right, but understanding that the 28 percent in this is presumably going to contribute to the overall factor shares, which adds up to one.

8334             MR. HENRY:  Absolutely.

8335             DR. ROYCROFT:  So when you substitute out labour, something else has to be going up because your factor shares always have to add up to one.

8336             So if you are using less labour, you are using more capital and/or more materials within the context of these studies from which you have drawn the numbers.

8337             MR. HENRY:  Fair enough.

8338             THE CHAIRPERSON:  Mr. Henry, did the word "last" prefacing your last question imply that this might be a time to take a break?

8339             MR. HENRY:  It's a perfect time, Mr. Chairman.  I have one area left.

8340             THE CHAIRPERSON:  That's even better.

8341             MR. HENRY:  That's perfect.

8342             THE CHAIRPERSON:  So because you have only one area left, I will say a quarter to 11:00.

8343             MR. HENRY:  Thank you.

8344             THE CHAIRPERSON:  We will rise and sit again at a quarter to 11:00.

8345             Thank you.

‑‑‑ Upon recessing at 1025 / Suspension à 1025

‑‑‑ Upon resuming at 1044 / Reprise à 1044

8346             THE CHAIRPERSON:  Order, please.

8347             Mr. Henry.

8348             MR. HENRY:  Thank you, Mr. Chairman.

8349             If we go back to that table again, I want to turn to the last factor, the stretch factor.

8350             DR. ROYCROFT:  I'm sorry, which table?

8351             MR. HENRY:  The original table that I handed out.  We have been through each of the first three numbers.  We are on the last one, the 1 percent stretch factor.

8352             DR. ROYCROFT:  Yes.

8353             COMMISSIONER LANGFORD:  Excuse me, Mr. Henry.  There have been so many tables.  Could you just tell us where it is in your book.

8354             MR. HENRY:  It is at Tab 2.

8355             COMMISSIONER LANGFORD:  Tab 2; thank you.

8356             MR. HENRY:  That is what you call a stretch factor of 1 percent.

8357             In recommending this factor, as I understand it, you recommended in part, agreed in part, based on your assertion that economies of scope should be reflected in the X factor.

8358             One of the examples you use is the study that you did in the U.S. where you calculated economies of scope relating to the provision of DSL service to contribute .57 percent to productivity.

8359             Is that right?

8360             DR. ROYCROFT:  I think you are combining two aspects of my overall analysis and recommendations.

8361             The stretch factor is provided as a mechanism for the Commission to assess the total factor productivity number that they ultimately go with and to what degree they believe it reflects the economies of scope which are, I would argue, not reflected in the historical number, the 4.2 percent, above.

8362             So the 5.7 percent ‑‑

8363             MR. HENRY:  Point 57.

8364             DR. ROYCROFT:  I'm sorry.  The .57 percent number that resulted from my study is an illustrative example of the magnitude of the scope economies and its impact on total factor productivity group within a sample of U.S. firms.

8365             MR. HENRY:  Fair enough.

8366             As I understand your approach, you estimated productivity for 14 state level holding companies, which are subsidiaries of AT&T and Verizon, based upon two scenarios which are then compared to each other.

8367             Let me make sure I understand the scenarios.

8368             The first scenario estimated productivity in relation to the production of telephone services and is based on the period 1988 to 2003.  Then, as I understand it, you calculate productivity under a second scenario which adds DSL outputs and inputs to the production mix based on data for 1999 to 2003?

8369             DR. ROYCROFT:  That is correct.

8370             MR. HENRY:  Then you calculate the difference in TFP growth between the two scenarios.

8371             DR. ROYCROFT:  Yes.

8372             MR. HENRY:  And that difference is the .5 percent which you assume is attributable to economies of scope.

8373             DR. ROYCROFT:  It is the weighted average result of the increase in productivity that arises from the study.

8374             MR. HENRY:  Right.

8375             I want to give you a very simple scenario and ask for your conclusions.

8376             Suppose you had a firm with two products, Product A and Product B, and suppose the productivity of Product B is increasing much faster than the productivity of Product A.

8377             If you were to measure the productivity of the company as a whole and compare it to the productivity of the company's Product A, the productivity of the company as a whole would be greater than the productivity of the company's Product A.

8378             Is that correct?

8379             DR. ROYCROFT:  If I'm understanding your example, it sounds like the two products are produced in entirely disparate and separate production processes; that there are no shared inputs.  Therefore, you can specifically identify productivity with the individual lines of business.

8380             MR. HENRY:  Well, it may or may not.  Let's take one where there are economies of scope.

8381             Isn't that what you did?  You took the productivity of the company with telephone service and then the productivity of the company with telephone service and DSL and compared the two and subtracted on average, weighted average, and got .57 percent?

8382             So you could do this exercise whether or not there were economies of scope.

8383             DR. ROYCROFT:  Well, my approach is a little bit different than what you described it.  And I may have misheard your example or some of the details.

8384             It sounded like you were separately calculating ‑‑ I calculated total factor productivity for two scenarios: one with and one without DSL.  It sounded as if you were separately calculating for individual product lines.

8385             MR. HENRY:  No.  I did it with Product A, so that's ‑‑

8386             DR. ROYCROFT:  but you said that Product A's productivity was growing much less rapidly than Product B, if I'm recalling correctly.

8387             MR. HENRY:  Right.

8388             DR. ROYCROFT:  That is what raised my clarification or need for clarification as to whether you are separately ‑‑ if you have pure separability across your production processes.

8389             It sounds like you are calculating separately the productivity of one product line and then the productivity of the other product line.

8390             MR. HENRY:  No, I was calculating the productivity of one product, and then I was calculating the productivity of the company as a whole, with two products.

8391             DR. ROYCROFT:  But, then, to do that ‑‑

8392             MR. HENRY:  There may or may not be economies of scope.

8393             THE CHAIRPERSON:  But if there are, Mr. Henry, how do you calculate the productivity of the first product?

8394             MR. HENRY:  Isn't that what Dr. Roycroft did with the telephone service, and then you added in the DSL?

8395             DR. ROYCROFT:  But I calculated the total factor productivity for the entire company in each case, I did not separate out the product lines in the way that your example is suggesting.

8396             MR. HENRY:  Suppose that the two products were totally independent.

8397             DR. ROYCROFT:  So no economies of scope?

8398             MR. HENRY:  No economies of scope.

8399             Wouldn't you get the same mathematical result?

8400             You would calculate the productivity of the company as a whole, and the company with the two products, and the company as a whole with the one product, subtract the difference and attribute that to economies of scope.

8401             DR. ROYCROFT:  But you said there were no economies of scope.

8402             If there are no economies of scope ‑‑

8403             MR. HENRY:  That's my point.  You would assume that the difference was economies of scope.

8404             DR. ROYCROFT:  I'm sorry; if there is a question, I am not hearing it.

8405             MR. HENRY:  You have a company with two products, with no economies of scope.  You calculate the TFP.

8406             Then, you have a company with one product ‑‑ the same company with one product.  You calculate the TFP.

8407             You would subtract the difference and say that's economies of scope, and you attribute it all to the first product.

8408             DR. ROYCROFT:  I'm sorry; I am getting confused.

8409             I thought the starting assumption was that there were no economies of scope, and if there are no economies of scope, then the company's total factor productivity would simply be reflective of the weighted average productivity across the product lines.

8410             If there are no economies of scope, then the resulting calculation doesn't generate any differential which represents economies of scope.

8411             MR. HENRY:  My point is, how do you know that all of the difference between the two scenarios is attributable to economies of scope, as opposed to economies within the DSL product itself?

8412             DR. ROYCROFT:  So we are moving away from a hypothetical Product A and Product B.

8413             MR. HENRY:  Sure.  We will come back ‑‑

8414             DR. ROYCROFT:  We are coming back to a specific discussion of my study?

8415             MR. HENRY:  Right.

8416             DR. ROYCROFT:  How do I know that there are economies of scope associated with the production of DSL?

8417             MR. HENRY:  I understand that you have calculated the difference, and that the company with the two products is more productive than the company with the one product.  I accept that.

8418             I don't see where you have proven that the .57 has anything to do with economies of scope.  It may have something to do with that, but it could also have a lot to do with economies within the DSL service itself.

8419             DR. ROYCROFT:  If we were to artificially force the company to produce DSL through a fully separate subsidiary, that would require that it would build its own loops to provide DSL service.  In order to build its own loops, it would incur substantial investments and the associated costs with those investments.

8420             Loops are just one part of the picture.  There are a number of other aspects of their operations, including the physical floor space, rack space, customer support facilities that are all shared or shareable when the production of DSL is added.

8421             So the fact that the alternative to producing DSL jointly with voice services, using the existing facilities, is a complete replication of the plant, that certainly would result in much higher costs, and, ultimately, much lower productivity, if you were to be calculating how you would add that service under that scenario.

8422             MR. HENRY:  My main point is that all of that or some of that or none of that may be true.

8423             There is a difference between the two productivity scenarios.  How do we know how much of that difference is attributable to efficiencies within the DSL service itself?

8424             Do you not think there would be any within the DSL service that would have nothing to do with the loop?

8425             DR. ROYCROFT:  The calculation that I make evaluates the growth in outputs and the growth in inputs, and therefore is, by definition, generating the total factor productivity associated with DSL.  I don't assume that the production of DSL results in the ability of the firm to sell anything else but DSL, which is a conservative assumption from the standpoint that the company may be able to add value to the DSL with other service offerings, be they Internet related or gaming or even video‑related services; my focus is on the production of DSL alone and the sale of DSL alone.

8426             MR. HENRY:  Would you agree that, at the early stages of a product's introduction, productivity gains can be expected to be easier to achieve than at later stages?

8427             DR. ROYCROFT:  It would depend on scale effects, from the standpoint that, with many products, rolling them out results in scale effects that result in lower unit costs over time.

8428             MR. HENRY:  Sure.

8429             The data used for your DSL results was for the period 2000 to 2003.  That is a period when DSL was in the early days of being rolled out.  Correct?

8430             DR. ROYCROFT:  Yes, it was in the relatively early period of DSL adoption.

8431             MR. HENRY:  Isn't it fair to say that you could expect some productivity to be realized within the DSL service itself, things that have absolutely nothing to do with economies of scope?

8432             Let me give you some examples.

8433             Fewer truck rolls.  In the early days of DSL, I think it is common knowledge that there were many truck rolls.  As time went on, there became more plug‑and‑play.

8434             Do you know how many truck rolls would have been involved in the year 2000 versus today, for DSL itself?

8435             DR. ROYCROFT:  No, I don't, but if there are fewer truck rolls, that would indicate that the production process is leading to productivity improvements associated with DSL.

8436             MR. HENRY:  And would you expect it to drop substantially?

8437             DR. ROYCROFT:  I certainly believe it is reasonable within any production process to assume that learning by doing and the scale effects that I mentioned earlier could lead to further productivity gains over time, from the standpoint that it becomes easier and cheaper to provision this technology that shares the existing infrastructure which provides the other services.

8438             MR. HENRY:  Things like less activity at help desks?

8439             Would you expect that to happen, get better, fewer activity at help desks?

8440             DR. ROYCROFT:  Help desk activity is a function of sales.  If sales are growing, then you would expect to see activity associated with that.

8441             If you are referring to the expertise associated with the help desk and being able to troubleshoot, that would likely have a life cycle associated with technology modifications and the way the service operates over time.

8442             MR. HENRY:  And better utilization rates for DSLAM ports as traffic and customers increase?

8443             You would get better utilization levels within the DSL service itself?

8444             DR. ROYCROFT:  Right, and that would result in improved productivity.

8445             MR. HENRY:  Exactly.  My point is, how much of that .57 percent productivity differential, which is the difference between Scenario A with DSL and Scenario B without DSL, is attributable to those kinds of efficiencies which have nothing to do with economies of scope?

8446             I am not saying there aren't economies of scope, but the ones we have talked about have nothing to do with economies of scope.

8447             DR. ROYCROFT:  As far as the numerical value, the assumptions that go into the creation of the input costs are, I believe, conservative and reflective of telephone company practices that were existing at the start of the overall process, and to the extent that there were improvements in those processes, I did not include that dynamic.

8448             I assumed constant, for example, maintenance and operating expenses associated with DSL over the period.  If those were declining, which I did not assume, then the number would actually be higher.

8449             MR. HENRY:  When you estimated productivity for the DSL scenario, did you use any data provided by telephone companies themselves in the U.S.?

8450             DR. ROYCROFT:  Yes, some of the data is based on telephone company reporting in the U.S.

8451             MR. HENRY:  Volume of DSL customers?

8452             DR. ROYCROFT:  Volumes of DSL sales, yes.

8453             MR. HENRY:  Yes.  You used that from telephone company data?

8454             DR. ROYCROFT:  Yes.

8455             MR. HENRY:  And prices they charge for DSL?

8456             DR. ROYCROFT:  Yes.

8457             MR. HENRY:  How about capital dollars?

8458             DR. ROYCROFT:  Capital dollars...?

8459             MR. HENRY:  Spent on DSL over that time period?

8460             DR. ROYCROFT:  No.  The estimates that I made were based on information that I was able to obtain through trade press reports on the costs of DSL‑related equipment.

8461             MR. HENRY:  Expense dollars?  Did you get those from the telephone companies?

8462             DR. ROYCROFT:  I used an approach that was based on telephone company cost studies with regard to central office equipment, cost factors from TSLyric studies that were associated with provision of components of central office equipment and therefore related to telephone company experience.

8463             MR. HENRY:  Could you turn, please, to the next exhibit found at Tab 6.

8464             This one was not given an exhibit number,  Madam Secretary.

8465             THE SECRETARY:  The Companies Exhibit No. 15.

8466             MR. HENRY:  It is already on the record.  Right.

8467             This was an undertaking provided as a result of a discussion between your counsel and Dr. Hariton.

8468             On page 1 of this exhibit we set out, in the top half of the page, your assumptions about telco capital and expense dollars required to provide DSL service per subscriber.  You can see, the capital costs are $218.75 declining to $118.75 in 2003.

8469             The little formula on the side basically says that what you assume ‑‑ take 99 ‑‑ $155 for a DSLAM piece of equipment, you divide it by a .8 utilization ratio and then you added an extra $25 to represent unknown capital costs.

8470             Then, as I understand it, the annual expense per subscriber you assumed was $1.50 per subscriber per month, which is the $18.

8471             I was wondering, how many truck rolls do you think you would get for the $25, and help desk inquiries do you think you would get for $1.50 a month?

8472             DR. ROYCROFT:  Well, not every customer requires a truck roll and not ever customers calls the help desk.

8473             MR. HENRY:  No, I mean on average.  On average.

8474             DR. ROYCROFT:  The estimated expense was drawn from cost studies that I have reviewed for regional Bell operating companies in the United States associated with central office equipment and also associated with help desk ‑‑

8475             MR. HENRY:  For DSL?  The $18 was based on DSL?

8476             DR. ROYCROFT:  No.

8477             MR. HENRY:  No.

8478             DR. ROYCROFT:  As the cost studies were associated with basic local exchange service I looked ‑‑

8479             MR. HENRY:  Basic local exchange service?  We are introducing a brand new DSL service and you assumed $1.50 a month for DSL?

8480             DR. ROYCROFT:  The process of introducing DSL involves placing DSLAMs into central office space and those DSLAMs are sitting right alongside other components of central office equipment which are used to provide basic exchange service and other services that are sold by the telephone company.

8481             I believe it is reasonable then to assume that the operating expense factors of electronic equipment that looks very similar and acts very similar to the electronic equipment that is already in place would be captured in a cost factor associated with the existing electronic equipment.

8482             MR. HENRY:  Another assumption.

8483             DR. ROYCROFT:  Certainly.

8484             MR. HENRY:  Do you know if these capital expense numbers bear any relationship to any Canadian telephone company who was introducing DSL over this period of time?

8485             DR. ROYCROFT:  Well, the market for equipment associated with the provision of DSL is one where the electronics associated with that are readily available from multiple supply sources.  This equipment primarily comes from East Asia.  It is considered ‑‑

8486             MR. HENRY:  Well, the DSLAM does.

8487             DR. ROYCROFT:  Pardon me?

8488             MR. HENRY:  the DSLAM does.  You just assumed ‑‑ even if I take that, you assumed another $25 for other equipment without ‑‑ you just kind of picked it out of the air as near as I can tell.

8489             DR. ROYCROFT:  With regard to other equipment, I'm not aware of significant impediments in input markets that are associated with telecommunications companies that would drive huge cost differentials between U.S. ILECs and Canadian ILECs.

8490             MR. HENRY:  Well, but you don't ‑‑ where does the $25 come from in terms of the U.S. ILECs?  that was an assumption, wasn't it?

8491             DR. ROYCROFT:  I added additional capital costs to provide a degree of conservative estimation.

8492             MR. HENRY:  Another conservative estimation.

8493             But it's just an assumption, it is an American assumption, and then you assume that an American assumption is a valid assumption in Canada.

8494             Is that correct?

8495             DR. ROYCROFT:  No.  I think you are misunderstanding my testimony.  My testimony focuses on an example in the United States where I have data and the information there is representative of telecommunications carriers in the United States and is based on dollar values associated with U.S. markets.

8496             The process of transplanting DSL technology into Canadian telecommunications firms does not present any undue or unusual obstacles, rather the operations are going to ultimately be similar.

8497             In the final analysis, the analysis that I present here is reflective of U.S. experience, but it is one that I believe is reasonable to assume would be similarly enjoyed by Canadian companies.

8498             MR. HENRY:  Well, let's look at that.

8499             When Dr. Hariton was on the stand he was asked if we had looked at it in terms of Canadian values and in the transcript he described Bell Canada's costs over that period and they are on the bottom half of the page.

8500             You can see in Canadian dollars ‑‑ and again I remind everybody this was the very early days of DSL rollout ‑‑ our capital dollars were $1,541 in '99, falling to $645 in 2003.  We converted that using purchasing power parity to U.S. dollars and then we used an annual expense of $100 rather than $18 and I think if you go to the transcript Dr. Hariton testified that was a very conservative number.

8501             Then what we did was, we re‑ran your study using all of your data, but with the Bell Canada numbers for DSL.

8502             If you flip the page you will see that.

8503             Basically the .57 disappears, doesn't it?  To be fair, it is .03 percent instead of .57 percent?

8504             DR. ROYCROFT:  That's what your number says if I am ‑‑ I have not seen the underlying work papers.  It's difficult to audit exactly what has gone on here, but what you seem to be ‑‑

8505             MR. HENRY:  Sorry, can I just stop you on that?

8506             DR. ROYCROFT:  Pardon me?

8507             MR. HENRY:  Can I just stop you on that?

8508             We have given the assumptions on the previous page, so if you haven't been able to could you just rerun your study with those numbers and confirm that these are the results you get?

8509             DR. ROYCROFT:  I could engage in that undertaking.  It would take more time than I have had given that I received these ‑‑

8510             MR. HENRY:  That's fair enough.  Well, I think the dates are October 26th.

8511             Can you get it done by then?

8512             DR. ROYCROFT:  Yes, I could.

8513             MR. HENRY:  All right.  So I have your undertaking to verify these numbers, verify that we have done your calculation correctly based on our numbers?

8514             DR. ROYCROFT:  Yes, I will do that.

8515             MR. HENRY:  Okay.  Thank you.

8516             MR. JANIGAN:  I wonder if Dr. Roycroft can complete his answer with respect to the assumptions associated with the numbers that he was about to give.

8517             THE CHAIRPERSON:  Yes.  If you have anything to add, Dr. Roycroft, feel free.

8518             DR. ROYCROFT:  Yes, I do.

8519             The assumptions that have been supplied here result in, as is clearly shown on page 2 of 2 of Companies Exhibit No. 2, as is marked here, that the impact of adding DSL is essentially a wash; that the companies involved here don't really experience any productivity improvement as a result of this.

8520             During this time period, as part of my normal consulting operations I keep track of what telephone companies such as these here tell their investors and read the transcripts of investor conferences.  During this period, as is still the case today, these companies speak very favourably of their DSL operations.

8521             Investors at these conferences I think are very astute individuals.  If they believed that adding DSL had no impact on productivity, I think they would call the persons conducting the investor conference on their numbers, saying:  Isn't it true that your costs of providing DSL are so high that you are not able to recover those through revenues?  Isn't it true that your productivity is actually flat as a result of adding this new line of service?

8522             I simply don't believe that this is a reflection of any sort of reality with regard to adding DSL.

8523             MR. HENRY:  I am content to let the Commission decide whether the numbers that come from Bell Canada are to be preferred to your American assumptions based on catalogue prices.

8524             Thank you very much, Dr. Roycroft.  Those are all of my questions.

8525             THE CHAIRPERSON:  You have concluded your questions?

8526             MR. HENRY:  Yes.

8527             THE CHAIRPERSON:  I have a couple of questions on the subject we have just discussed.

8528             To be clear, Dr. Roycroft, is it possible ‑‑ and I'm asking this as someone who clearly doesn't have the expertise of the three doctors on either side here.

8529             Is it possible that the years in question would have been years in which the productivity improvements had not yet appeared and they would appear in subsequent years, or be expected to appear in subsequent years, and therefore justify the investment notwithstanding the ‑‑ I'm putting a word in here that may be inappropriately used ‑‑ the relatively static picture that you get from those introductory years?

8530             DR. ROYCROFT:  Certainly the companies when they make their investments anticipate that they may have opportunities to sell additional products and services over that time period.

8531             My assumptions and my approach use the best data that I had available to evaluate what was happening right then and there, and it is certainly possible that over time the productivity improvements associated with the investments will expand further, especially with regard to once you have the opportunity to sell ‑‑ once you have the broadband platform, then the ability to add additional products, value‑added products, to that platform becomes relatively easy.

8532             You have the opportunities associated with expanding scope economies and expanding productivity in the future.

8533             THE CHAIRPERSON:  Would it be possible that equally notwithstanding the neutral or wash effect that comes out of these figures, companies might regard themselves as doing the right thing if they could reduce term by adding a product to the basic PES?

8534             DR. ROYCROFT:  During this particular period?

8535             THE CHAIRPERSON:  Or make themselves less vulnerable to competitive inroads?

8536             DR. ROYCROFT:  Certainly they could view this as having a positioning impact on their overall operations and market position.

8537             THE CHAIRPERSON:  So overall it is conceivable at least, forgetting about this particular case, that in theory relatively unattractive numbers for productivity associated with incremental investment in a telecom business could be perfectly compatible with a rational business strategy that could be defensible to investors.

8538             DR. ROYCROFT:  Well, ultimately if investors have a long‑term point of view, I think investors tend to require the performance to be satisfactory on a relatively short‑term basis and to see rewards from that as well.

8539             If you read the companies' filings with their investors, they tout the additions that they are getting and the revenues that they are getting from those additions in a fashion that indicates to me that they think they are coming out ahead as a result of those investments, even in the near term.

8540             THE CHAIRPERSON:  Do you think that the RBOCs and their Canadian homologues had any choice with respect to the introduction of broadband into the legacy PSTN?

8541             DR. ROYCROFT:  There certainly have been variations in their strategies, particularly in the U.S., with regard to how rapidly they've rolled out this service.

8542             For example, BellSouth has been a relative laggard in good portions of its service territory and now that it is being merged with the old AT&T is looking to expand its broadband footprint as a result of the merger.

8543             So they certainly have exhibited behaviour that indicates that choices have been made with regard to how they are deploying this new technology.

8544             THE CHAIRPERSON:  Thank you.

8545             Commissioner Noël.

8546             COMMISSIONER NOËL:  Dr. Roycroft, you mentioned early in your testimony this morning that you saw the telecommunications market in Canada more as a duopoly than as a really competitive market because the telcos are facing another body of corporations, namely the cablecos.

8547             You are certainly aware that cablecos' footprints are very limited regionally and that the same telco ‑‑ let's take, for example, Bell Canada in Quebec and Ontario ‑‑ will face numerous cablecos' competitors with various pricing strategies.

8548             Do you think that this is really a truly duopoly market rather than one competitor, namely Bell Canada, facing a number of various competitors in different regions and having to face more than one other price, or one other strategy?

8549             DR. ROYCROFT:  There is no question that they will face multiple strategies depending on the location of the cable company and its footprint.  The competitive process is driven by the consumer's ability to make choices.  If a consumer is residing in a particular cable company's service area, they don't have the ability to take cable service from some other company.

8550             So the choices are limited and the market response recognizes the consumer's limitations.

8551             When companies know that in a particular area they face only one other firm, there is a natural tendency to try to avoid price competition.  This has been evidenced in numerous examples in a variety of markets, including those in the telecommunications industry.

8552             If you look at what happened in the United States where we had a cellular duopoly in the early days of wireless, the market performed very poorly.  Prices were very high.  Service quality was low.  And then through changes in policy which enabled the introduction of up to six new wireless licences in a particular market area, suddenly you had multiple firms providing services and offering them to the same customers that previously just had two choices.

8553             When you started to have six, seven, eight firms competing for those customers' businesses, the consumers enjoyed lower prices, improving service quality and a greater consumption of the service as a result of those lower prices.

8554             COMMISSIONER NOËL:  Apparently some time today you will reach in the U.S. 300 million people as the population of the U.S.  Here we are about 33 million and we have a territory that's at least as large as the territorial U.S. or continental U.S., as you call it.

8555             Do you think we have the luxury of having eight competitors with eight infrastructures?

8556             DR. ROYCROFT:  In the wireless environment or the ‑‑

8557             COMMISSIONER NOËL:  Or the wireline environment.

8558             DR. ROYCROFT:  That, to me, gets to the core of the policy issue here with regard to how to proceed in addressing this market from a regulatory perspective.

8559             If you do have a duopoly and the duopoly is going to behave as duopolies normally do, then market power continues to be an issue even though you have two firms supplying in the market.

8560             Even in the United States with our large population base, we are seeing a similar development there, with tremendous consolidation on the local exchange carrier side, with the CLEC industry essentially ramping down and either disappearing through a merger or exit from the market, leaving the primary alternative as the cable companies.

8561             It appears that a duopoly is a market environment that is emerging across the borders and, in my opinion, consumers need protection in such an environment, and price cap regulation provides a reasonable set of protections, especially when it is quite clear that the competition from cable is tending to be targeted at a specific segment of the market, namely, the higher end of the market.

8562             If there are uncontested customers, vulnerable customers, they need protection even in this duopoly environment, and I think that price cap regulation emerged as a theoretical construct to address the presence of competition in telecommunications markets.

8563             COMMISSIONER NOËL:  And that is what we are discussing here, price cap regulations.  We are not discussing forbearance.

8564             DR. ROYCROFT:  That is correct.

8565             COMMISSIONER NOËL:  Thank you.

8566             THE CHAIRPERSON:  There is a lot more to be said about all of this, and I wish we could discuss it now, but we will find a moment.

8567             You have completed your questions, counsellor?

8568             MR. HENRY:  I have.

8569             THE CHAIRPERSON:  Thank you very much.

8570             COMMISSIONER LANGFORD:  Mr. Chairman, perhaps I could ask a question.  I know that normally we would wait, but this flows right out of the last exhibit that Mr. Henry put before us, and it seems probably more appropriate now than to try and revive it in three hours' time.

8571             This is difficult stuff if you don't have a Ph.D. in mathematics or a Ph.D. in economics, and, unfortunately, I don't have either of those, so I may not couch my questions as clearly as I wish I could.

8572             I think when we left the last exhibit that you and Mr. Henry were discussing you had agreed to get out your calculator and make sure that he didn't make any arithmetical errors on the last page.  That is, I suppose, useful to us.  If there are errors, we should know about them.  But I wonder if you can go any further with what we have here.

8573             I don't feel that I heard an explanation or even a guess from you ‑‑ which is not a criticism, I just didn't hear it.  It might have gone on, as I say, at a level that I missed, but there is this wide gap between the figures based on all of the companies you looked at that you came up with and the figures that apply to Bell Canada DSL costs ‑‑ the figures that you supplied at the top of page 1 of Exhibit 2, and the figures that Bell Canada has supplied at the bottom of page 1 of Exhibit 2.

8574             It seems to me, that will require more than work with a calculator.  Where I am lost is ‑‑ can you hazard any guesses as to what might explain these disparities?

8575             We are looking at numbers that, in the first year, are five or six times higher, one‑time capital costs, and in the fifth year they are four or five times higher.  If I had a calculator I could come closer.

8576             Could you hazard a guess, from your experience, as to why these numbers seem to be so far out of line with yours?

8577             DR. ROYCROFT:  I don't know what the company's numbers include or what they reflect, as far as what are the investment items behind that.  My focus was on DSLAM, which is the electronic equipment that essentially adds the DSL functionality and an additional adder to account for other investment.

8578             What I know is that the DSLAM market is one, as I was indicating in my earlier testimony, that is a market for electronic equipment.  There are multiple providers, multiple vendors.  This equipment comes from Asia and is purchased by both users and by telecommunications firms.

8579             Press reports that I have identified for this period, for example, for 2003, identify prices ‑‑ list prices ‑‑ that are significantly below the prices that are indicated here, and they reflect the prices that I have identified in my assumptions.

8580             As a result of that, the numbers are substantially below Bell Canada's numbers.

8581             To me, if Bell Canada is talking about DSLAMs, these numbers are really not credible.

8582             I found a reference to an individual talking about Canadian prices in 2002 ‑‑ and I provided this in response to a discovery to Bell Canada, I believe ‑‑ where it identifies an individual named Ted Yeap, who is a researcher at the University of Ottawa School of Information Technology and Engineering, and in 2002 he discusses Canadian DSLAM list prices, I believe, in the range of $250 to $300 per port.

8583             If you translate those numbers using exchange rates, which I believe is more appropriate than purchasing power parity, as purchasing power parity is designed to reflect the non‑fluid nature of certain aspects of markets and here we are talking about DSLAMs, which are like a commodity, and therefore are likely to be priced very similar based on exchange rate ratios, his value is much closer to the value that I identify in 2002 than the value that the company has identified.

8584             The company says $829; this individual says $250 to $300 Canadian per port.

8585             I can't explain it.

8586             There is other evidence, as well, in the trade press talking about major vendors and their prices per port, and they are identifying prices that are consistent with mine, because that is the method I have used to develop the prices, and they are way out of alignment with the capital costs associated with Bell's data.

8587             I don't know the source of the discrepancy; I can only explain the logic behind my estimates.

8588             COMMISSIONER LANGFORD:  All the companies you have listed, which you have done research into, do they basically look at the DSLAM price, or are some of them different?  Do some of them add other factors in when they are doing their costing?

8589             DR. ROYCROFT:  When it comes to evaluating the addition of DSL, the major increment to existing technology is going to be the DSLAM.  The DSLAM is going to pick up the copper loop, through some sort of physical connection to the loop, and then transfer on the other side of the DSLAM some of the information that comes over the loop to the voice switch, and the rest to an Internet service provider, or possibly the company itself, if it is acting as its own Internet service provider.

8590             The DSLAM itself is the major cost driver when you are talking about the addition of DSL service, given the sharing of many of the other facilities, given their pre‑existence and use for providing the whole host of voice services.

8591             COMMISSIONER LANGFORD:  Thank you very much, Dr. Roycroft.

8592             That was my question, Mr. Chair.

8593             THE CHAIRPERSON:  Madam Secretary.

8594             THE SECRETARY:  Thank you, Mr. Chair.

8595             Thank you very much, gentlemen.

8596             I was informed this morning of a mutual agreement that took place between MTS Allstream and TELUS Communications that they would trade places in the order of appearance.

8597             Therefore, I would invite TELUS counsel to come forward.

8598             THE CHAIRPERSON:  Commissioner Langford suggests that we take a 10‑minute break to make sure the witness has an opportunity to prepare.

8599             We will start again at 20 minutes to 12.

8600             THE SECRETARY:  Yes, with Mr. Ryan.

‑‑‑ Upon recessing at 1130 / Suspension à 1130

‑‑‑ Upon resuming at 1139 / Reprise à 1139

8601             THE CHAIRMAN:  Order, please.  À l'ordre, s'il vous plaît.

8602             We have order but we do not have madame la secrétaire.  Elle s'en vient.  Elle s'en vient en courant.

--- Laughter / Rires

8603             THE CHAIRMAN:  That is right.

8604             Mr. Langford points out that there cannot be order without the Secretary.

--- Laughter / Rires

8605             THE SECRETARY:  I am so sorry.

8606             Mr. Ryan, please proceed on behalf of the company TELUS Communication.

8607             MR. RYAN:  Thank you, Mr. Chairman.

8608             And for the record, again, my name is Michael Ryan.  I appear on behalf of telecommunications company and with me this morning is Dr. Jeffrey Bernstein.

8609             Good morning, Dr. Roycroft.

8610             DR. ROYCROFT:  Good morning.

8611             MR. RYAN:  Dr. Roycroft, it has already been established as a result of your cross‑examination by Mr. Henry that you filed testimony in a recent proceeding before the California Public Utilities Commission, I understand.

8612             DR. ROYCROFT:  That is correct.

8613             MR. RYAN:  And since the filing of your evidence in this case, the California P.U.C. has rendered its decision in that other case?

8614             DR. ROYCROFT:  Yes, it has.

8615             MR. RYAN:  And I've provided you yesterday through your counsel with a copy of the decision of the California P.U.C. in that proceeding, which you've probably already had some familiarity with in any event, but did you receive a copy?

8616             DR. ROYCROFT:  Yes, I did.

8617             MR. RYAN:  And that document that I provided to you, is that ‑‑ is that indeed the decision of the California P.U.C. in the proceeding in which you've testified?

8618             DR. ROYCROFT:  Yes, it is.

8619             MR. RYAN:  Mr. Chairman, you should have a largest document in front of you, a couple of hundred pages, which is the document that doctor Roycroft and I have been referring to.

8620             THE SECRETARY:  I have this.  Yes, and it's referred to as "TELUS Exhibit number 13".

8621             MR. RYAN:  And just to orient ourselves at the beginning, doctor Roycroft, could we go to page 2 of the decision?

8622             DR. ROYCROFT:  Roman numeral 2 or ‑‑

8623             MR. RYAN:  Numeral 2, please, the page that begins "Summary".

8624             DR. ROYCROFT:  Yes.

8625             MR. RYAN:  And the Commission says at the outset of its decision:

"This decision evaluates both statutory guidance and market conditions in determining whether we may rely more heavily on competitive forces to produce just and reasonable rates for California's telephone consumers."

8626             They go on to say:

"We grant carriers broad pricing freedoms concerning almost all telecommunication services, new telecommunications products, bundles of services, promotion and contracts."

8627             And towards the end of the paragraph, they say:

"With few restrictions we permit carriers to add services to bundles and target services to specific geographic markets."

8628             Is that a reasonable shorthand summary of the conclusions of the P.U.C. in this proceeding?

8629             DR. ROYCROFT:  Yes, with the caveat that they also address the limitations associated with their decision with regard to high cost areas and lifeline programs which are treated differently under the framework.

8630             MR. RYAN:  And which are referenced in the next paragraph of the summary, I believe.

8631             DR. ROYCROFT:  Yes.

8632             MR. RYAN:  Now, on who's behalf did you testify in that proceeding?

8633             DR. ROYCROFT:  I testified on behalf of TURN, which stands for "The Utility Reform Network".

8634             MR. RYAN:  And who is or was TURN?

8635             DR. ROYCROFT:  TURN is a independent non‑profit organization that focuses on utility matters in the State of California.

8636             MR. RYAN:  Could we go to page 113?  And here, at least with regard to one particular issue, the California P.U.C. appears to be summarizing the position of TURN and I'm looking at the subheading on the page TURN and just initially, that first instance, the P.U.C. says:

"TURN maintains that ILECS continue to have significant market power."

8637             And I understand that that conclusion, the TURN, or that proposition that TURN was advanced, it was based on the observation that ILECS are very large market shares; is it not?

8638             DR. ROYCROFT:  That contributed to the evaluation of the market ‑‑ market shares were evaluated as well as the ability of firms to substitute between alternative sources of supply, so the evaluation looked at more than strictly market share evaluated alternative technologies, evaluated observed consumer behaviour on the market and drew that conclusion.

8639             MR. RYAN:  So, in the next couple of paragraphs, for instance the P.U.C. says:

"With respect to Sure Western Varies and TURN states that its analysis found that market concentration, a little change from pre‑1996."

8640             It goes on to say that:

"TURN asserts that Frontier ‑‑",

presumably another carrier in California:

"‑‑‑ maintains a pure monopoly position and, finally, that AT&T retains an overwhelming market share for both the residential and small business market segments."

8641             Are you with me?

8642             DR. ROYCROFT:  Yes.

8643             MR. RYAN:  And I note that those statements, footnotes 464, 465 and 466, citing TURN, referring TURN to comments that you've made in the proceeding, you mentioned in the footnotes?

8644             DR. ROYCROFT:  That is true.

8645             MR. RYAN:  So that indeed reflects the evidence that you put before the Commission, I take it then, that's been ‑‑ that's been cited there?

8646             DR. ROYCROFT:  Yes and I mean, for example, with regard to Frontier Telecommunications, they are a company that serves primarily rural areas, the evidence indicated that among their ‑‑ I believe it was 30 wire centres, only one wire centre had evidence of a cable alternative provider. 

8647             The rest were strictly provided by Frontier and Frontier had not sold a single unbundled network element to any Select in the period since 1996, which reflected a dominant market position which, from the consumer's perspective looks like monopoly.

8648             MR. RYAN:  And, of course, you go on to mention AT&T in your evidence as well and its position in the market?

8649             DR. ROYCROFT:  That is correct.

8650             MR. RYAN:  And that would be the former S.B.C. effectively that's been referred to there?

8651             DR. ROYCROFT:  Yes, after ‑‑ after S.B.C. acquired AT&T, which was S.B.C.'s largest competitor in the California market, S.B.C. took AT&T's name, I guess to the Victor Gelvis(ph) spoils, and the evidence indicated that the acquisition of AT&T not only took residential consumers off of AT&T's network and placed them on S.B.C.'s service offerings, but S.B.C. also acquired significant facilities that AT&T was ‑‑ had deployed in primarily business sectors in a large number of wire centres around the State.

8652             MR. RYAN:  Of course, your evidence in this proceeding is analogous to the position that you advanced in California, to the extent that you're concerned here also that the ILECS continue to have significant market power?

8653             DR. ROYCROFT:  Yes.

8654             MR. RYAN:  And if we look at the second full paragraph on page 113, the California P.U.C. says:

"TURN's analysis of market to competition, however, relies on its conclusion that the relevant market only includes wire‑line circuits switch telecommunication services."

8655             Again, citing to comments filed by you.  Am I correct?

8656             DR. ROYCROFT:  That is what it says.  It mischaracterizes my evidence and analysis.

8657             MR. RYAN:  All right.  Would you explain to us to what extent it mischaracterizes your evidence?

8658             DR. ROYCROFT:  Well, the analysis that I conducted evaluated market share and the market share calculations were based on E‑911 counts from E‑91 data basis that included both cable provided VoIP services as well as circuit switch services offered by the incumbents and by the ‑‑ any Selects still remaining in the market.

8659             So, the statement that I only considered wire‑line circuit switch was erroneous with regard to my evaluation of market share. 

8660             However, my analysis also evaluated wireless services as well as over‑the‑top VoIP services, with the distinction that the over‑the‑top VoIP services, at that time, were not included in the E‑911 data basis whereas cable VoIP services would be included.

8661             So, my evaluation was much more broad ranging than circuit switch telecommunication services and evaluated a variety of supply alternatives which consumers might consider and made a determination based on that evaluation.

8662             MR. RYAN:  But in point of fact, the thrust of your evidence before the California P.U.C. was that wireless services, for instance, is not a substitute for wire‑line service.  Is that correct?

8663             DR. ROYCROFT:  The evaluation that I presented the Commission certainly acknowledged that for certain customers wireless services has begun to provide an alternative, but that for the overwhelming majority of consumers, wireless has both a higher price and characteristics of functionality that make it inferior to wire‑line offerings and as such then, does not provide a cheque on market power that would be consistent with removing the regulatory constraints to the extent that the Commission in California has done.

8664             MR. RYAN:  Well, let me see if I understand.

8665             Are you saying that wire‑line and wireless are in the same market?  I'm talking now of California.

8666             Were you saying, shall we say, in your testimony in California that wireline and wireless were in the same market or that they were not?

8667             DR. ROYCROFT:  My testimony was that they were generally not in the same market for most consumers, that the decision‑making process did not include a reasonable alternative from wireless to replace wireline given existing functionalities associated with wireline that are not available from wireless and given higher wireless prices when compared to services available over the wireline network.

8668             MR. RYAN:  Turning then to VoIP, was it your testimony in California or was it not that VoIP is not in the same market as wireline telephone service?

8669             DR. ROYCROFT:  There again I have made the distinction between over‑the‑top VoIP and VoIP that would be provided by facilities‑based service provider such as a cable company with regard to over‑the‑top VoIP, both the limitations on the availability of broadband and the consumption of broadband, as well as limitations associated with the technology itself lead me to conclude that it does not provide a service which is capable of constraining ILEC market power.

8670             MR. RYAN:  Another way of saying that is it is not in the same market?

8671             DR. ROYCROFT:  Yes.

8672             MR. RYAN:  So when I go back to the second full paragraph from the California decision and the Commission says:

"TURN's analysis of market competition however relies on his conclusion that the relevant market only includes wireline, circuit‑switched telecommunications services."

8673             Which you said was a mischaracterization of your evidence.

8674             I have to say, it sounds to me like a fairly fair characterization of what you have just told us, because that same market we have just heard you say does not include wireless or over‑the‑top VoIP.

8675             DR. ROYCROFT:  But it does include cable‑provided VoIP and therefore the comment saying "only .. wireline, circuit‑switched telecommunications services" is not correct.

8676             MR. RYAN:  All right.  So it is your testimony here today that cable‑provided VoIP is in the same market as circuit‑switched telephony services?

8677             DR. ROYCROFT:  Yes.  Especially ‑‑ well, depending on the cable company's business strategy that would define the relevant segment of voice services.  Cable companies typically are focusing on bundled service offerings to target it at the higher end of the market.

8678             MR. RYAN:  Could we go next to page 120 of the decision, please?

‑‑‑ Pause

8679             MR. RYAN:  I'm looking now at the top of the page and the Commission is stating some of its conclusions, as I understand them.

8680             They say:

"We agree that the build out of wireless carrier's network since this Commission's last major telecommunications regulatory review 18 years ago has made wireless technologies a close substitute for land line services.  This evidence is a significant factor in this decision."  (All read)

8681             As I understand them, they, on this point, rejected your evidence and preferred an alternative, that wireless technology is indeed a close substitute for wireline service.

8682             Is that correct?

8683             DR. ROYCROFT:  Yes.  Given the evidence presented to the Commission in many cases it was the only alternative that the Commission considered from the standpoint that it has concluded that each and every residential customer in California, regardless of their location, is facing the same level of competitive forces which will lift the Commission's responsibility to ensure directly just and reasonable rates and allow market forces to govern that.

8684             As I indicated earlier, in wide swaths of California ILEC service territory, the only alternative to the ILEC services are what are available from wireless carriers.

8685             MR. RYAN:  So you remain, as it were, unrepentant in terms of your position on this issue in light of the California PUC's decision on the evidence they had before them?

8686             DR. ROYCROFT:  Yes,  To give you a personal experience that helps me reach this conclusion, among extensive other analysis, I live in an area that is not a high density area and while I can receive telephone service over my wireless phone if I step outside, I cannot receive wireless service indoors.  In fact, wireless carriers do not design their networks or guarantee the delivery of service indoors and, as such, their service is certainly not comparable to basic exchange services or even bundles of exchange services which are designed to be received indoors.

8687             In the United States, wireless carriers fall over themselves to explain to their customers that when you buy this service there is no guarantee that it's going to work anywhere, especially indoors, that it is primarily a service that is designed for outdoor operation.

8688             MR. RYAN:  I presume these are all matters that you put before the Commission as part of your evidence in that case.

8689             DR. ROYCROFT:  I don't recall if I presented them my personal experience, but in general the fact that the wireless carrier networks are not designed to function in a manner consistent with wireline services certainly was, as well as the fact that wireless and wireline carriers are oftentimes the same companies have a vested interest in maintaining consumer subscription to wireline and therefore do not encourage substitution of wireless for wireline services.

8690             MR. RYAN:  Could we look at the next full paragraph on page 120, because the Commission goes on then to deal with the question of VoIP.

8691             It says:

"In addition, Verizon's evidence, especially when coupled with data produced by AT&T, convincingly establishes that a competitive threat is offered by the new VoIP technologies."  (As read)

8692             Would you agree with me that the Commission there was essentially accepting the idea that VoIP is in the same market as wireline services and essentially rejecting the view that you had put forward that VoIP over‑the‑top service was not in that market?

8693             DR. ROYCROFT:  That is the Commission's position, yes.

8694             MR. RYAN:  All right.

8695             Can go to page 126 next, please?

‑‑‑ Pause

8696             MR. RYAN:  Will you agree with me one of the criticisms that the California PUC had of your evidence was that it paid undue attention to the question of market share to the neglect of other factors that needed to be considered in evaluating whether there was market power present or not?

8697             DR. ROYCROFT:  The Commission was generally not inclined to evaluate market share information.  Which in my opinion is a mistake, as I am an economist and market share has been demonstrated to matter in a wide variety of market situations

8698             MR. RYAN:  Let me just read you the passage that I have in mind from the PUC's decision and invite your comment on that.

8699             The PUC says:

"Substantial legal precedent discusses the dangers of relying on market share as a measure of competition in regulated markets.  Such dangers are well recognized by the courts, the FCC and this Commission.  For example, in dismissing a claim that a cellular telephone company with a 100 percent share of the wholesale market exercised market power, the Ninth circuit stated..."

8700             And I'm going over to the next page and I will continue and read that quote, if I may:

"Blind reliance upon market share divorced from commercial reality can give a misleading picture of a firm's actual ability to control prices or exclude competition.  Reliance on statistical market share in cases involving regulated industries is at best a tricky enterprise and is downright folly where, as here, the predominant market share is the results of regulation.  In such cases, the court should focus directly on the regulated firm's ability to control prices or exclude competition."  (As read)

8701             I will just give you the opportunity to comment as an economist on the validity of the California PUC's comment as I have just read it to you.

8702             DR. ROYCROFT:  Well, if I might start with the quote from the Ninth Circuit, you will note that that quote is from 1989 and therefore the context of the market share being the result of regulation is one that I would point out is in a very different context given that this is prior to 1996.

8703             If market shares remain high after entry barriers are removed, and enabling the entry has been pursued by a Commission through the availability of unbundled network elements at cost‑based rates, if market share still persists, something must be driving that high market share and continuing entry barriers are a likely candidate.  Where you have entry barriers you have market power and market power needs to be managed to ensure that the outcome is in fact just and reasonable rates.

8704             The other aspect of the quote that I would want to comment on is that my approach was neither blind ‑‑ rather, it was a broad‑ranging evaluation of not only market share, but also of alternative sources of supply and a thorough evaluation of those technologies and how consumers are observed to respond to them in the marketplace ‑‑ nor was it divorced from commercial reality, given the backdrop to the market environment that the Commission was evaluating: one where the Commission in the State of California had observed the complete collapse of CLEC industry, the merger of the number one and number two CLECs with the respondent companies in this case and therefore, if anything, pointed to a retrenching and consolidation in the industry which was not consistent with expanding competition.

8705             MR. RYAN:  To sum up, as I understand it, the California PUC's concern was that you relied unduly on market share in coming to your conclusions, and indeed your calculation of market share was high precisely because you excluded competition from wireless and VoIP service providers in calculating market share.

8706             Is that a fair summary of the situation?

8707             DR. ROYCROFT:  No, I don't believe that it is.

8708             I certainly evaluated market share and I used the best tools that I had available with regard to market share.  I used the Herfindahl‑Hirschman Index approach to evaluate market share, which as a means of assessment focuses on the sum of the market shares, each share being squared.

8709             What this does is result in a numerical measure that allows you to focus on what is important from the theoretical perspective, and that is the market shares of the major players.

8710             I did not dispute before the Commission that it was likely that some percentage of consumers had cut the cord and gone wireless only.  Given the structure of the market, however, that fact resulted in each of those wireless carriers having a very small market share, and therefore the fact that I did not include it in the calculation, because the data was not available to me, did not have a significant impact on that aspect of my assessment.

8711             The market share information that I presented to the Commission was only one aspect of my overall analysis.  I evaluated demand side conditions and supply side conditions to see whether or not there was evidence of consumers' ability to freely substitute among alternatives in a manner that would be consistent with the constraint on the incumbent's market power.

8712             MR. RYAN:  Is it fair to say that the Commission was particularly critical of your use of the HHI index that you just referred to, the Herfindahl‑Hirschman Index?

8713             DR. ROYCROFT:  I didn't rate their criticisms on a scale of 1 to 10, so I couldn't give you the ‑‑

8714             MR. RYAN:  They were very critical of your use of that index, though, were they not?

8715             DR. ROYCROFT:  They indicated that they did not think that the market share information generally mattered.

8716             If you have specifics that you want to show to me, I would be happy to comment on them.

8717             But I think it is a fair statement, as has already been said in my discussion with you, that they do not pay any attention to market share even in instances where the market share is still 100 percent and line growth is consistent with historical line growth in areas of the state; in other words, that walk and talk just the way they did before 1996 and the removal of entry barriers.

8718             The Commission says that that does not matter; that that is the same market as one where we have a cable company providing service in a wire centre and that wire centre has a cable market share of 20 percent now.

8719             So the Commission in its wisdom does not see any difference between those market conditions, which I disagree with very strongly.

8720             MR. RYAN:  You testified in California, again as you have testified here ‑‑ and I'm thinking of page 37 of your evidence here ‑‑ that wireline and wireless are actually complements; that is, that increased use of one leads to increased use of the other.

8721             Is that correct?

8722             DR. ROYCROFT:  Complements in the sense that they are consumed by consumers and not viewed as outright substitutes for one another.

8723             MR. RYAN:  Well, complements in the sense that they are consumed by consumers.  You would have to explain that to me.

8724             What I thought a complement was, was when there was complementarity that increased use of one service would lead to increased use of another.

8725             Is that not a proper understanding of what complementarity involves?

8726             DR. ROYCROFT:  I'm not sure.  It sounds like you are alluding to a crossed elasticity evaluation with regard to complementary products so that if there is a price relationship, a price occurrence with regard to one product or service that there is a reaction on the part of the other product or service.

8727             The fact that wireless subscription has expanded so dramatically in the United States while wireline subscription has not dropped correspondingly indicates that consumers are using both wireless and wireline services; that they have not done a one‑to‑one substitution across the two service sets.

8728             MR. RYAN:  We are not just talking about substitution here.  We are talking about complementarity and the propensity to use more of one service as a result of your use of another.

8729             That coincides with the normal understanding of complementarity, does it not?

8730             Why don't we go to page 37 of your evidence and you can tell us what you mean by complementarity here.

8731             MR. JANIGAN:  Mr. Chairman?

8732             Mr. Chairman?  Over here.

8733             THE CHAIRPERSON:  Mr. Janigan; sorry.

8734             MR. JANIGAN:  I have no wish to interrupt my friend's cross and there are certain hooks or certain relevance that this line of cross may have to the design of a price cap framework, but I wonder if it might be possible for us to connect the dots a little bit quicker on this.

8735             The exploration of some of these aspects is probably best confined to a forbearance proceeding rather than to a price caps proceeding.

8736             I appreciate the fact that there are some relevant aspects of this, but this is ‑‑

8737             THE CHAIRPERSON:  At the moment, Mr. Ryan, you appear to wish the witness to agree with your analysis that complementary products imply that the consumption of the one increases the consumption of the second.

8738             I think the witness is saying to you that the importance of a complementary ‑‑ and he will confirm or deny this.  His analysis of complementary is that the consumption of one does not necessarily diminish the consumption of the second; i.e., it is not a substitute.

8739             Is that a fair characterization?

8740             DR. ROYCROFT:  Yes.

8741             THE CHAIRPERSON:  And I don't think he is going to agree with you on the other proposition.  So let's leave it there.

8742             MR. RYAN:  Can we look at page 37 of your evidence.

8743             I am bearing in mind your comment, Mr. Chairman.  I hope I'm not transgressing here your advice or direction.

8744             You say on page 37:

"It is reasonable to expect that wireless and wireline will continue to be utilized in a complementary fashion in Canada."

8745             Are you with me there?

8746             DR. ROYCROFT:  Yes.

8747             MR. RYAN:  And perhaps just so I understand correctly what your definition of complementarity is, could you explain to me what you mean in that statement.

8748             DR. ROYCROFT:  What I am characterizing here is that consumers will not substitute wireless services for wireline services.

8749             The reference point of Canada here is also reflecting the fact that wireless services are currently much more expensive than wireless services in the United States, which suggests a further impediment on substitution.

8750             MR. RYAN:  Would you agree with me that lack of substitution is not the same as complementarity?

8751             DR. ROYCROFT:  I think, as I indicated earlier, there is certainly an economic definition of complementary relationships based on cross price elasticities.

8752             I think there is also a more casual usage of complementary relationships as opposed to substitute relationships.

8753             I think that the latter is being characterized in this particular instance.

8754             MR. RYAN:  So it is that more casual, as you put it, understanding of the concept of complementarity that you are referring to on page 37 of your evidence?

8755             DR. ROYCROFT:  Yes.

8756             MR. RYAN:  Could we go to page 129 of the California decision?

8757             I will give you an opportunity, given the parallels between your evidence on this, particularly the notion of complementarity, before this Commission and before the California PUC, to comment on this observation by the California Commission.

8758             At the first full paragraph on page 129 they say:

"Neither TURN nor DRA..."

‑‑ which was another intervenor, I take it:

"...presents adequate evidence in support of their contention that wireless services complement wireline services."  (As read)

8759             And skipping a sentence:

"TURN attempts to address changes in the number of landline telephone lines, but its analysis is not convincing.  TURN only generates an increase in the number of wirelines by treating a single voice plus data line as two lines.  This double counting of lines that provide both voice and DSL services is inappropriate.  It is a poor and misleading analysis to argue that line losses are indeed line increases, and then conclude that wireline and wireless services are complements."  (As read)

8760             I understand that you haven't conducted the sort of count that is referred to in this passage from the California decision as part of your analysis of complementarity for the purposes of your evidence in this proceeding, but would it be fair to say that the analysis of complementarity on which you based your California evidence is the foundation for your statement in this proceeding that the two are complements; that is, wirelines and wireless?

8761             DR. ROYCROFT:  As the passage that you read indicates, the Commission is speaking broadly here and brings DSL into this discussion.  I would say that my testimony in California analyzed the evidence with regard to the substitution of wireless for wireline circuits, and I am not sure of the exact point of reference that the Commission is using that word here, but portions of my evidence in California were addressing whether or not wireless and wireline were substitutable for one another.

8762             MR. RYAN:  Is it fair to say that that evidence was the subject of particular criticism, the reasoning that led you to that conclusion in California?

8763             It was referred to as poor and misleading analysis.

8764             DR. ROYCROFT:  The analysis that the Commission appears to be critical of here, as you read, is the interpretation of line loss associated with DSL substitution, and here the Commission reaches the conclusion that if a telephone company loses a second line, but continues to sell that customer DSL service, that is the same as the company losing the second line and not selling DSL service.

8765             The Commission is focusing on a line loss analysis that was presented by multiple parties to the Commission which pointed out that you can't evaluate the impact of competition on the market while you are observing second line losses and treat that as if that is a competitive loss, when, in fact, those second line losses are simply the company rolling over the technology onto DSL.

8766             MR. RYAN:  All right.  We will move on.  Thank you.

8767             Could we go next to page 46 of your evidence?

8768             I am going to set aside the California decision for the moment, Mr. Chairman.

8769             COMMISSIONER LANGFORD:  Mr. Ryan, I wonder if you would be kind enough to get me, at least, paragraph references to Dr. Roycroft's evidence.

8770             For some reason my photocopy lacks page numbers.

8771             We are tracking down the culprit, and there will be a public execution on Confederation Square at two o'clock this afternoon.

‑‑‑ Laughter / Rires

8772             MR. RYAN:  I hope that madame la secrétaire is not implicated.

‑‑‑ Laughter / Rires

8773             MR. RYAN:  I would say paragraph 84.

8774             COMMISSIONER LANGFORD:  Thank you.

8775             MR. RYAN:  You start with the heading, Dr. Roycroft, posing yourself the question:  "Have you performed a productivity study which illustrates the impact of the production of new services on TFP growth?"

8776             Of course, you have covered a good part of this subject matter already with Mr. Henry today.

8777             Are you with me?

8778             DR. ROYCROFT:  Yes.

8779             MR. RYAN:  As I understand it, to illustrate your theory that there are economies of scope associated with new services, you set out to measure the impact on TFP of adding new services to the product mix of the 14 U.S. telephone companies that we have already heard referred to.

8780             Is that true?

8781             DR. ROYCROFT:  Yes.

8782             MR. RYAN:  And the new service that you added, of course, was DSL.

8783             DR. ROYCROFT:  Yes.

8784             MR. RYAN:  Could we go to paragraph 85 of your evidence?

8785             In calculating TFP growth for the U.S. telecommunications carriers you utilized data for the years 2000 to 2003.

8786             Is that correct?

8787             DR. ROYCROFT:  Starting in 1999 and going to 2003, yes.

8788             MR. RYAN:  However, you had relevant data going back to at least 1985 for RBOC TFP, did you not?

8789             DR. ROYCROFT:  Yes.

8790             MR. RYAN:  Are you not omitting important information when you choose to use only the period 2000 to 2003 for the purpose of your analysis when you have data for the timeframe beginning with 1985 available to you?

8791             DR. ROYCROFT:  I do not have the data on broadband available to me prior to 1999, so I could not include that.

8792             I believe that is not an unreasonable omission from the standpoint that DSL was not widely pursued in that period, and that the statistics that I have are representative of the rollout period beginning in the late 1990s.

8793             MR. RYAN:  Could we go to Table 8 of your evidence, which is associated with paragraph 95.

8794             On Table 8, the first entry is "Industry TFP Growth," and you say, "From previous plan."

8795             The previous plan, is that a reference to Decision 97‑09?

8796             DR. ROYCROFT:  Yes.

8797             MR. RYAN:  Can you confirm that the source of the 4.2, if it is not already clear to everyone that you cite there, is from Decision 97‑09?

8798             DR. ROYCROFT:  That is correct.

8799             MR. RYAN:  Decision 97‑09 was, in turn, based on data for the years 1988 to 1995, was it not?

8800             DR. ROYCROFT:  Yes.

8801             MR. RYAN:  Given that that information is over a decade old, are you not concerned about the reliability of it for the purpose of this analysis?

8802             DR. ROYCROFT:  I think that the value is conservative from the standpoint that technological change and expanded ability to produce inputs is much greater in a more recent period.  I think it would be preferable to have TFP growth for a more recent period, but, as I indicated in my earlier testimony, that was not available to me, nor was the opportunity to prepare such a study.

8803             MR. RYAN:  Could we go to the transcript for Friday at page 880?

8804             Do you have that page?

8805             DR. ROYCROFT:  Yes.

8806             MR. RYAN:  I am going to look at the bottom of the page, and I will give you a chance to catch up with me.

8807             Essentially, that passage relates to a cross‑examination that was conducted by Mr. Lawford of the MTS Allstream panel on Friday, and in this passage, beginning here ‑‑ and I will take you through it ‑‑ Mr. Lawford criticized MTS Allstream for using data that ended in 2001.

8808             I will give you a chance to catch up with me.  I'm looking starting at page 880, line 5898, the question is to ‑‑

8809             DR. ROYCROFT:  I'm sorry, which page?

8810             MR. RYAN:  Page 880, line 5898 at the bottom.

8811             DR. ROYCROFT:  All right.

8812             MR. RYAN:  Question 5898.  Mr. Lefebvre says:

"The issue here is, when you look back at the marginal cost trend data that has been used, it was for the period 1988, I believe, to 2001..."

8813             Then going over to the next page, question 5904 at the bottom, Mr. Lawford says:

"It seems that you are basing your position of the productivity factor ‑‑ you are inclined to do that on future estimates, and yet the productivity numbers you are using are, in a sense, derived from more ancient data, if I could put it that way.

Wouldn't it be better, if we are going to be consistent with your position at paragraph 35, to use more recent productivity numbers?"

8814             Are you with me?

8815             DR. ROYCROFT:  Yes.

8816             MR. RYAN:  Given that your 4.2 TFP figure was based on data that ends in 1995, I thought it curious that your counsel was criticizing MTS Allstream for using data that ended in 2001 and referring to that as "ancient data".

8817             By that measure, the data that you are using is practically prehistoric, isn't it?

8818             DR. ROYCROFT:  I would have to carbon date it, but the ‑‑

‑‑‑ Laughter / Rires

8819             DR. ROYCROFT:  As I just testified a few moments ago, I would prefer to have more recent data if that data was available, and this was the most recent data that I had available for the Canadian situation.

8820             MR. RYAN:  Well, it is the most recent data, but that doesn't necessarily qualify it as reliable data for the purposes that you are using it, does it, Dr. Roycroft?

8821             DR. ROYCROFT:  As I indicated, the technological change in the intervening period as well that is going to have a strong impact on input usage and the ability to provide a broader set of services certainly would have an impact on productivity measured more recently, and because of that I believe that the value stated here is conservative as it does not include those expanded opportunities.

8822             MR. RYAN:  Well, can we move on to further on page 53.  You say:

"The X factor you calculate is based on TFP analysis."

8823             I'm still on paragraph 95.

"The X factor you calculate is based on a TFP analysis."

8824             Is that right?

8825             DR. ROYCROFT:  The X factor is based on the information contained within Table 8 which we have just been discussing.

8826             MR. RYAN:  It starts with the TFP growth factor of 4.2 ‑‑

8827             DR. ROYCROFT:  Right.

8828             MR. RYAN:  ‑‑ that we have already referred to.

8829             TFP measures, of course, the difference between output growth in all services provided by a firm versus growth in all factor inputs used by the firm.

8830             Is that fair?

8831             DR. ROYCROFT:  Yes.

8832             MR. RYAN:  Since the X factor that you have produced here is based on a measure of the productivity of the firm as a whole, it is not specific to Res PES, is it?

8833             DR. ROYCROFT:  No, it is not.  And the total factor productivity of an individual product line, especially when you have a multi‑product firm with significant shared inputs is something that is logically impossible to calculate, so it's not done.

8834             MR. RYAN:  I'm sorry, are you saying it is logically impossible to calculate the productivity for a specific subset of services?

8835             DR. ROYCROFT:  You would have to make some sort of set of decisions about how the inputs were going to be allocated across the services, and therefore the approach, as is indicated by total factor productivity is to evaluate the operations of the firm and then that is the answer that you get as a result.

8836             MR. RYAN:  So we are agreeing that total factor productivity isn't an approach that can be adapted to use for measuring the productivity of a specific subset of services?

8837             Is that correct?

8838             DR. ROYCROFT:  Right.

8839             MR. RYAN:  Right.  Okay.

8840             DR. ROYCROFT:  From the standpoint of you can only calculate the total number.

8841             MR. RYAN:  Right.

8842             MR. RYAN:  Could we go next to another document that I passed to you through your counsel, and that is some selected pages from Decision 2002‑34, which you may recognize as the decision of this Commission in the previous price cap case.

‑‑‑ Pause

8843             MR. RYAN:  Do you have a copy of that in front of you?

8844             DR. ROYCROFT:  Yes, I do.

8845             MR. RYAN:  Could we look at paragraph 629, please?

‑‑‑ Pause

8846             MR. RYAN:  I would just like to read a couple of paragraphs from this decision to you and invite your comment on them.

8847             The Commission says, in paragraph 629:

"In part of this decision the Commission determined that the productivity offset will be applied directly to certain baskets of services and individual rate elements.  The Commission notes that this is significantly different from the initial price cap regime in which a productivity offset was imposed on cap services overall.  The Commission considers that in applying an X factor to certain baskets of services and individual rate elements, a TFP‑based approach would be more comprehensive than a marginal cost‑base approach.  However, the Commission notes that the ILECs were not able to calculate a TFP‑based X factor on a service‑specific basis."  (As read)

8848             The Commission continues in paragraph 631:

"In these circumstances the Commission does not consider the continued use of a TFP‑based X factor to be appropriate.  In the Commission's view, the basic productivity offset for the next price cap regime should be based on service‑specific marginal costs in order to reflect the actual productivity gains that are likely to be achieved for individual capped baskets."  (As read)

8849             So if is fair to say that the approach you are advocating before the Commission here, which is a TFP‑based approach, is at odds with what the Commission decided to do in the previous price cap proceeding?

8850             DR. ROYCROFT:  Yes, certainly it is different.  The application of a total factor productivity based number on specific service sets is something that has been done in other jurisdictions.  For example, the Federal Communications Commission calculated total factor productivity for the RBOCs and then applied that number to the interstate services that were governed by the Commission.

8851             The problem with going with marginal cost‑based approaches, especially for primary exchange service, given my review of studies performed by Bell, is that those marginal cost studies assume that 100 percent of the local loop is attributable to primary exchange service, when it is quite clear that the local loop is capable of delivering a broad set of services.

8852             So that the marginal cost‑based approach then, in my view, unfairly burdens primary exchange service customers with the costs of providing the shared infrastructure that enables a wider variety of outputs to be provided.  When a company builds a loop they can provide local services, they can provide long distance services, vertical services, and now DSL‑type services.

8853             So the approaches are certainly different and it is my opinion that the marginal cost approach is a weaker one than the total factor productivity based approach.  The Commission indicates that it at the time thought that a TFP‑based approach would be more comprehensive, but decided to go otherwise.

8854             MR. RYAN:  Right.  The considerations you identified were certainly considerations the Commission would have had in mind the last time around in price caps 2 when it made its choice as to whether a TFP‑based approach or a marginal‑cost approach would be a superior way of identifying the appropriate X factor for a specific subset of services.

8855             Is that fair?"

8856             DR. ROYCROFT:  Yes.  The Commission changed its mind once and I imagine the Commission could change its mind again.

‑‑‑ Pause

8857             MR. RYAN:  Could we look next at your TFP‑DSL study, if I can refer to it that way, which again you have already touched on, as I said, with Mr. Henry.

‑‑‑ Pause

8858             MR. RYAN:  I'm going to go back to my copy to make sure I can provide some paragraph references as well as some page references.

‑‑‑ Pause

8859             MR. RYAN:  I am going to say page 49, which is in and around paragraph 88.

8860             Are you with me?

8861             DR. ROYCROFT:  This is the page with Table 6 on it?

8862             MR. RYAN:  That's correct, in and around those pages.

8863             This is where you discuss your TFP growth study for U.S. telecommunications carriers.  Correct?

8864             DR. ROYCROFT:  Yes, I do.

8865             MR. RYAN:  You already discussed many aspects of this study with Mr. Henry and I won't go over that territory again.  But as I understand your study, you calculated TFP growth for all services without DSL and then you proceeded to calculate TFP growth for all services including DSL.

8866             Is that essentially what you did?

8867             DR. ROYCROFT:  Yes.