E.B.R.O. 497-01 THE ONTARIO ENERGY BOARD IN THE MATTER OF the Ontario Energy Board Act, R.S.O. 1990, c. O.13; AND IN THE MATTER OF an Application by The Consumers' Gas Company Ltd. for approval of an incentive mechanism in relation to the Operation and Maintenance Expense component of its cost of service, effective during the 2000 through 2002 fiscal years, and an incentive mechanism in relation to Demand Side Management. Hearing held at 2300 Yonge Street, 25th Floor, Hearing Room No. 1, Toronto, Ontario on Tuesday, February 2, 1999 commencing at 9:00 a.m. ---------- VOLUME 5 ---------- B E F O R E : H. GAIL MORRISON PRESIDING MEMBER PAUL VLAHOS Member ROGER M. HIGGIN Member 657 A P P E A R A N C E S JENNIFER LEA ) Board Technical Staff EDWARD SWEET ) KATHI LITT ) FRED CASS Enbridge Consumers Gas, Applicant MURRAY KLIPPENSTEIN Pollution Probe ROBERT WARREN Consumers Association of Canada (CAC) TOM BRETT Metropolitan Separate School Board, Ontario Association of School Board Officials and CAESCO BETH SYMES ) Alliance of Manufacturers CAROL STREET ) and Exporters, Canada GUY PRATTE Industrial Gas Users Association (IGUA) DAVID POCH Green Energy Coalition IAN MONDROW HVAC Coalition ELIZABETH DeMARCO Coalition for Efficient Energy Distribution (CEED) THOMAS ADAMS Energy Probe PHILIPPA LAWSON Ontario Coalition Against Poverty (OCAP) MICHAEL MORRISON Ontario Association of Physical Plant Administrators LINDA ANDERSON Union Gas Preliminary Matters 658 ---Upon commencing at 9:00 a.m. THE PRESIDING MEMBER: Thank you. Please be seated. Good morning. Are there any preliminary matters? Mr. Cass? MR. CASS: I have one preliminary matter, Madam Chair. In short, Dr. Fuss has a scheduling difficulty this afternoon. I perhaps have been very naive and had been thinking it would not be a problem because he will be finished before then. I'm still cautiously optimistic, but I thought I should raise it just in case it becomes a difficulty. In the event that we can't finish Dr. Fuss, the next panel, the company's final panel, will be ready to go so we don't lose any time and Dr. Fuss could be stood down and returned tomorrow morning. THE PRESIDING MEMBER: That sounds satisfactory. The Board will not be available between twelve and two either of today, tomorrow or Thursday. So we will try and whisk right along and perhaps take slightly shorter breaks during the morning and the afternoon. Any other matters? Mr. Pratte? MR. PRATTE: Thank you. THE PRESIDING MEMBER: A problem? MR. PRATTE: I hope not, Madam Chair. But I, as amongst ourselves, have decided I would lead on Dr. Fuss Preliminary Matters 659 and I have to leave at two, as well. I don't know how we will do with the remainder of this panel. If we are finished by the morning break I guess we should be all right because at least I could do... But I wonder if we might revisit that issue depending on how far we go with this panel. THE PRESIDING MEMBER: Certainly. We maybe just should stand this panel down and deal with Dr. Fuss and then go back to this panel, if it looks like we are going to go longer than the morning break. MR. PRATTE: Thank you very much. That would be very much appreciated. MS. LAWSON: Madam Chair, excuse me. I am wondering if we should begin with Dr. Fuss this morning. I know I am in the middle of my cross-examination with this panel, but it would be fine with me if you think that's the wisest course to follow. THE PRESIDING MEMBER: How much longer do you think you have? MS. LAWSON: Ten minutes. THE PRESIDING MEMBER: Let's at the very least finish with you and maybe we can think again. Go ahead. MS. LAWSON: Thank you very much. DARRYL SEAL, DAVE CHARLESON, DAVID J. deJONGH, JAMES C. GRANT, STEPHEN McGILL; Resumed. CONTINUED CROSS-EXAMINATION BY MS. LAWSON: Q. Good morning, panel. In your evidence Seal,Charleson, 660 deJongh,Grant,McGill cr-ex (Lawson) on productivity you state that total factor productivity is a sound and rigorous methodology used to determine productivity changes. Dr. Norsworthy uses this approach to measure productivity, do you agree? MR. SEAL: A. That is the prime productivity estimate that Dr. Norsworthy has, that's correct. Q. But you refer to use the multi-factor productivity measure because it corresponds with your targeted PRB plan, I take it; that is, because you are limiting the PRB to operations and maintenance you think the productivity measures should also be so limited, correct? A. That's correct. Q. Okay. I would like to turn to Dr. Norsworthy's evidence. I just have a few questions for you on it. In particular, I would like to start at table 3 on page 18 which is where Dr. Norsworthy compares Enbridge Consumers Gas productivity growth over the past decade with that of the Canadian gas distribution industry. Now, my first question is: Do you have any quarrel with Dr. Norsworthy's numbers? I'm not asking about his arguments based on these numbers, but just the numbers themselves. A. Yes, we do have some issues with them. Perhaps Professor Fuss will deal with them in more detail with the next panel, but we do have some problems with Seal,Charleson, 661 deJongh,Grant,McGill cr-ex (Lawson) some of his numbers. Q. Okay. I'll put that over to Dr. Fuss, then. Am I correct that you also take issue with his focus on productivity growth without reference to what might be called static productivity in the base year and in particular Consumers Gas' operations and maintenance expense per customer in relation to the rest of the industry starting, let's say, in 1987? A. I'm not sure exactly what you mean, Ms. Lawson. Perhaps you could rephrase the question. Q. Do you take issue with Dr. Norsworthy's comparison of productivity growth over the past ten years between Consumers Gas and the rest of the industry? A. We take issue with it in terms of the numbers that he is showing here. As I have indicated, we have some difficulty with some of the methodologies that he has used in this calculation. So we have problems with it from that perspective. Q. Okay, thanks. Do you disagree, however, that in terms of productivity growth over the last decade Consumers Gas under-performed the industry? A. As I've mentioned, yes, we do have problems with that because of the way they have been calculated. Now -- Q. But even on your numbers, do you disagree with that conclusion? A. Our numbers, specifically referring to Dr. Norsworthy's calculation of our numbers? Seal,Charleson, 662 deJongh,Grant,McGill cr-ex (Lawson) Q. I'm referring to however you would correct Dr. Norsworthy's numbers? A. I guess I would correct Dr. Norsworthy's numbers by saying that our numbers are correct, at least in terms of the FMP estimate for Consumers Gas. That's how I would correct them. Q. Let me ask my question again. Whatever your numbers are, do you agree with me and Dr. Norsworthy that Enbridge Consumers Gas underperformed the Canadian gas distribution industry in productivity growth over the period 1987 to 1992 which is the period that we have data for the industry? A. Perhaps I can answer that question by referring to our own evidence. As I've said, I don't agree with the numbers so I can't draw conclusions from Dr. Norsworthy's evidence. But within our own evidence or at least the evidence of Dr. Fuss, the comparison we make is with Statistics Canada data over comparable periods and the growth rates of productivity are different between the Stats Can and our number. Q. I don't think you have answered my question, sir. My question was quite simple. It was: Did Enbridge Consumers Gas under-perform the Canadian gas distribution industry in productivity growth over the period 1987 to 1992? A. The only reference I have is -- the reference that I know the numbers and know exactly how they are Seal,Charleson, 663 deJongh,Grant,McGill cr-ex (Lawson) calculated is using our own numbers that we calculated, the .63, and the number that is in Dr. Fuss' evidence where it does show that using the Stats Can data the productivity growth over the comparable period was higher than Enbridge Consumers Gas' estimated productivity growth for the same period. Q. So the answer-- A. Sorry. Q. --is yes? A. The answer is yes, based on those numbers and there are reasons given in Dr. Fuss' evidence as to why that growth could be measured as being higher. Q. Is it the case, based on your numbers, that, in fact, Enbridge Consumers Gas' productivity growth performance was worse over this period when gas costs were excluded? A. Again, we exclude gas costs from all our calculations. So our calculations are without gas costs. Q. Then let me rephrase the question. Is it the case that ECG's productivity growth is comparatively better in relation to the industry when you include gas costs? A. We don't include gas costs in our estimate. Q. So you didn't do that comparison? A. No, we didn't because we don't believe gas costs should be included in the productivity estimate. Q. Okay. I take it then since you have some issue with Dr. Norsworthy's numbers that you don't agree Seal,Charleson, 664 deJongh,Grant,McGill cr-ex (Lawson) with the figures shown in tables 3, 4 or 5 in his evidence? A. Certainly the numbers in table 3 we have issue with. Q. And you are going to explain those issues through Dr. Fuss? A. I believe Dr. Fuss will explain some of reasons that we believe that those numbers are incorrect. I can explain one of them right off the bat is that we believe that customers is a proper measure of output, not volumes, which is what Dr. Norsworthy uses. Table 4 -- Q. Insofar as tables 4 and 5 are derivative of table 3 you would presumably take issue with them? A. I'm not sure that they are, but I think they are simply showing some numbers, some company numbers. Q. Well, let's move on, sir, to page 23 and table 6 of Dr. Norsworthy's evidence where he compares company and industry trends in capital and labour per unit of gas service over the period 1987 to 1994. Q. Now, do you take issue with these numbers? A. Yes, we do. Q. Okay. Can you explain? A. I'll briefly explain. I'm sure Professor Fuss will do a much better job than I. But basically, again on the output side of things, Dr. Norsworthy is using volumes as his growth measure. We believe customers is the appropriate growth Seal,Charleson, 665 deJongh,Grant,McGill cr-ex (Lawson) measure -- or output measure, I'm sorry, to be using. And on the capital side, my understanding is that the calculations of the capital input that Dr. Norsworthy has used for industry are different than the way he has calculated it for the company. And, in fact, the way he has calculated it for the company is not standard practice, if you will, and therefore the numbers that come out of it are not correct. That's my semi-layman terms. Q. And what do you mean by "not standard practice"? Can you explain? A. In terms of the way that capital should be accounted for in productivity measures, there are standard practices and ways of measuring capital, and that the way that Dr. Norsworthy has used it for the company is not the standard established practice and is different from what is used for the industry. Q. I guess I'm just wondering in what way is it different from the standard? What is the standard practice and how is Dr. Norseworthy's approach different? MR. CASS: Excuse me, Madam Chair. I will be asking Dr. Fuss for this very explanation when he's on the stand. I believe, if I'm not wrong, that Mr. Seal is just attempting to paraphrase what Dr. Fuss will be saying. If Ms. Lawson doesn't mind, I think it will be better to put the question to Dr. Fuss. MS. LAWSON: Sure. I will wait for Dr. Fuss then, Madam Chair. THE PRESIDING MEMBER: Thanks. Seal,Charleson, 666 deJongh,Grant,McGill cr-ex (Lawson) MS. LAWSON: Q. Would you agree, sir, with Dr. Norseworthy's conclusion that there are similar output trends between the company and the industry in respect of outputs generally and in respect of labour per unit of output? I know you take issue with the numbers, but do you take issue with those conclusions? MR. SEAL: A. The conclusions drawn from his numbers, I guess, no, I don't take issue. Q. And do you agree with him that while there are similar trends in output and labour, there is a markedly different trend between the company and the industry when it comes to capital? A. I don't believe his numbers to be correct and therefore I can't agree with that conclusion. Q. Right. But based on your numbers? A. We haven't done a comparison of capital for the company versus industry, so I can't comment. Q. Okay. So you can't comment on Dr. Norsworthy's statement that while the industry generally was using capital more efficiently, the company was doing the opposite; it's capital efficiency was declining? A. I don't believe that to be true. Q. But you haven't done the calculation? A. I have not calculated because we have calculated a multifactor productivity that does not include capital that goes in our formula. Q. Okay. Thanks. Seal,Charleson, 667 deJongh,Grant,McGill cr-ex (Lawson) Now moving on to Table 7 of Dr. Norsworthy's evidence on page 26. I think we all understand that you take issue with his measurement of output, and I don't want to go through that. I think we have had enough discussion over that already. I am just wondering, assuming that Dr. Norsworthy's output measure is used, do you take issue with his numbers in Table 7? A. I guess I'm not sure what you're asking me. If you're asking me if I assume that the numbers in that table are correct, that the growth rates that fall out of it are the ones that he has calculated... Q. No. I'm asking if, instead of using customers as your output measure, you use revenues -- volume weighted by revenues, as Dr. Norseworthy did. A. If I was going to do that calculation, I would do it a little bit differently in that I would take out gas costs in the revenues, which I believe Dr. Norseworthy has included. Q. So you would come up with different numbers? A. Possibly. Q. But otherwise, you don't take issue with the numbers? A. Other than the difference that I have just noted, if I was doing that calculation, then he has done the calculation the way I would do it. Q. Okay. Thanks. MR. GRANT: A. The only other comment that I Seal,Charleson, 668 deJongh,Grant,McGill cr-ex (Lawson) would provide on this table, and that I don't know the details of his calculation, but just looking at the table, in my mind when it says "unweighted gas volume" on Table 7 and I look at the numbers, it doesn't appear to be weather normalized, so it would give some rather spurious results in my mind. Q. If it's not weather normalized? A. Yes. I don't know whether it is or isn't. I'm just looking at the numbers. It would appear that it isn't. The volumes sort of bob around over the years. Q. Okay. Thank you, Mr. Grant, for that. Finally, gentlemen, I would like to turn very briefly to the issue of the deflator that you use. And, again, this has been covered by other counsel and so I really just want to ask you to look at Table 8 and again tell me what disputes, if any, you have with Dr. Norsworthy's numbers in this table. MR. SEAL: A. Table 8 deals specifically with the deflator used to deflate the materials price input in the productivity measure. So it is different from the inflation variable we're using in our formula. I just want to say that from the start. In terms of this particular table, I think I mentioned yesterday that Dr. Norsworthy has assumed that the materials price index for the gas industry is the correct one to use for deflating the company's material expenses. And as I said yesterday, there is no evidence Seal,Charleson, 669 deJongh,Grant,McGill cr-ex (Lawson) to suggest that is the correct one to use for Enbridge Consumers Gas. Q. Well, I know about that dispute. That's why I'm just asking you about the numbers. A. I haven't checked the materials price index against what came out of Stats Can, so I guess I can't comment on whether they're accurate or not. Q. Can't comment. Okay. Well, I did notice that Dr. Norsworthy's calculation of the Ontario CPI using Stats Can data over the period 1987 to 1997 is 2.9 per cent; whereas, in your evidence, you state it as 3.1 per cent, and I am just wondering if you can explain that difference. A. Well, one possible explanation is that Dr. Norsworthy is using calendar year data; whereas, our own data uses fiscal year data. I don't know that for sure, but that's one suspected difference. Q. Thanks. Now, you stated earlier that there are effectively two inflation measures used in this formula; one for inflation itself and the other for the deflator of the company's inputs. I'm wondering, in what sense are those different? I thought you were using the Ontario CPI for both. A. Yes. Our proposal does use Ontario CPI to deflate the materials index in our productivity estimate and we do use Ontario CPI as our proposed inflation Seal,Charleson, 670 deJongh,Grant,McGill cr-ex (Lawson) measure in the PBR formula, so we are using it in both of those. There are other deflator indexes, if you will, used in the productivity estimate. For example, on the labour side, we're using our actual labour costs as the deflator to put things in real terms. I'm just trying to differentiate between the Ontario CPI that we've used to deflate the materials price index in our productivity estimate from the inflation variable within our PBR formula; that's all. Q. But you are using the same measure for both? A. Yes, we have. MS. LAWSON: Okay. Thank you very much. Those are all my questions, Madam Chair. Thank you, gentlemen. THE PRESIDING MEMBER: Thank you, Ms. Lawson. Ms. Lea, do you have any idea how long you would be? MS. LEA: Madam Chair, most of my questions are reconciliation-type questions, clarification-type questions. I should think that I will probably be 20 minutes or so. THE PRESIDING MEMBER: Okay. Go ahead, Ms. Lea, please. MS. LEA: Thanks. If I'm exceeding that estimate, I can always stop in the middle and start again, you know, after, so.... Good morning, gentlemen. Seal,Charleson, 671 deJongh,Grant,McGill cr-ex (Lea) CROSS-EXAMINATION BY MS. LEA: Q. I wonder if we could begin with having a look at Exhibit C, section 4, page 2, please. That's Exhibit C, section 4, page 2. And what I would need to understand is a couple of differences between the original white page filing and the latest blue page update. I think these differences also occurred in the first blue page update but I'm looking at the original filing, 98-07-21, and the most recent filing, 99-01-22. Thanks for the explanation you gave in part yesterday on this. I would like to ask you a couple of other things. In the original filing we noticed that the one-time adjustments for the 1999 Y2K expenditures were 11.7-million. By the time we get to the January '99 update, they're 6.2-million. Can someone help me with why that difference exists? MR. CHARLESON: A. Yes, the Y2K expenditures were adjusted from the 11.7-million to 6.2-million as a result of the Board's decision in EBRO 497 where the allowed expenses in '99 were reduced. Q. Thank you. And what about the Legacy system, this is another thing that I wasn't sure about. In the original filing you talk about the Legacy system being a one-time adjustment but by the updates it no longer appears? A. Yes, and again that's for similar reasons. Seal,Charleson, 672 deJongh,Grant,McGill cr-ex (Lea) In the Board's EBRO 497 decision, the Legacy System upgrade costs were removed. Q. Thank you. Now, the removals that you've made from the base O&M, the O&M base year, were these items also removed from the data set used to estimate productivity? MR. SEAL: A. Neither of those two items that you just mentioned were in the data that were used to estimate productivity. Q. Okay. And what about the other removals from O&M for the base year calculation, were they also removed from the data set for productivity? A. Yes, they were all removed from the data used to calculate productivity. Q. So we then have consistency between the O&M base year data set and the productivity data set? A. Correct. Q. Thank you. I wonder if we could have a look for just a moment at the growth parameter. You'll notice my questions jump from topic to topic with about two in each. We heard a lot about the fact that the growth parameter is estimated on number of customers but what, gentlemen, is number of customers? Is it number of bills; is it number of burner tips; what is it? How do you count the customers? MR. CHARLESON: A. The number of customers is done looking at the average number of bills. Seal,Charleson, 673 deJongh,Grant,McGill cr-ex (Lea) Q. Average number of bills. And is this in your mind the best estimator, the best count for number of customers and if so, why? A. From our experience it's felt that the average number of bills is the best estimator. The rationale being we've looked historically at the impact in terms of other measures such as number of unlocks and we've found that the average number of bills tracks truer to, say, the number -- it would be more representative of the number of customers as some of the other measures can include a number of double counting issues or issues where items being missed so I guess, from our experience, it's more -- we've deemed it to be the best driver. Q. Can we make the further inference -- oh, sorry? Go ahead. MR. GRANT: A. The only thing I would add is that if you think of it in an operational sense, a greater number-- Q. That's what I was going to ask. A. --a greater number of bills can generate a greater number of phone calls, for example, a greater workload, which is affecting the O&M costs so that's why we feel it's a better measure. Q. So, Mr. Grant, your evidence would be that the number of bills as the way of counting the number of customers is also the best predictor or driver of O&M costs among the ways to count customers? A. Yes, that's right. Seal,Charleson, 674 deJongh,Grant,McGill cr-ex (Lea) Q. Thank you. I wonder if we could, please, look at Exhibit C, section 4, page 11, please, which is the Table 4-3, Customer Gas, Historical Productivity. And I'm looking at now the third column I suppose under Growth, and we see here a number of percentage increases in growth over the years and the average that is on the bottom here is in 3.96. And I'd ask you to comment on whether this figure 3.96, in fact, reflects the downward trend that we've seen in growth since 1990? For one thing, I might say is that 3.96 is a larger number than any number in that column since 1990. Does not the use of this average bias it in a sense towards growth that you have not seen for the last nine years? MR. SEAL: A. The 3.96 is a simple average of the growth over those ten years, '87 to '97. And I guess I would comment first of all that it was approached in 1997, very close 3.91 per cent. But I don't believe -- the numbers that we use here are calculated on a year-to-year basis as the output growth index which is then matched up with the input growth index for each of those years so that you calculate the productivity index from that and then the growth in productivity. So I don't believe that the trend biases anything in particular. What it's reflecting is the customer growth year-to-year. The input index is reflecting the input growth that matches with that customer growth and the productivity is what falls out of that. Seal,Charleson, 675 deJongh,Grant,McGill cr-ex (Lea) So I guess the short answer is, no, I don't believe it biases anything. Q. Did you consider using a weighted average, that is, giving more weight to the years most recent? A. No, we didn't. We took an average over the period as indicated because we thought that taking an individual year wasn't appropriate because you can have swings from year to year and I'm talking in a general sense here, not just the growth but all of our index numbers. But taking an individual year might not be representative. Taking an average over a period of time smooths out some of those year-to-year changes and also choosing a period of time that roughly approximates a business cycle, if you will, is appropriate for measuring these. Q. What about a rolling average, a five-year rolling average, for example? A. Again we didn't do that because we thought that taking this average over the ten years was a good approximation or a good estimate of what the growth of both the input and output indexes is. Q. Okay. I think anything further would just be argumentative. Thanks. When you take that number, the 3.96, and plug it into the formula, do I understand the formula correctly, in that as the customer growth number increases, all other things become equal, the O&M budget will increase? If all Seal,Charleson, 676 deJongh,Grant,McGill cr-ex (Lea) other numbers in that formula remain the same, an increase in the customer growth number results in an increase in the O&M total budget? A. That's correct. Q. Thank you. Earlier in your evidence, and I think this was in an answer to Mr. Brett, someone - and I'm sorry I cannot remember the exact quote or even the person - talked about the fact that there was a balance of the growth factor and the CPI, one over-estimated slightly, one under-estimated slightly and those things balanced out. Can someone help me with a little more elaboration on that, please? MR. GRANT: A. Yes, I think it was my testimony. When I was discussing this with Mr. Brett, my recollection is that I was suggesting that the growth variable, if I were to calculate a ratio - he was talking about 1-to-1 ratio - in the equation and I was suggesting that over the three-year period it would approach a 0.7-to-1 ration, recognizing that in the very short term it's only around about a .5-to-1 ratio. And I was comparing and contrasting that with the CPI situation where the CPI is used to try to capture inflation that affects the company and yet historically we know that roughly 60 per cent of O&M which is labour and benefits tends to be roughly half a per cent I think is what the evidence shows higher than CPI over time. So that's what I was balancing, those two Seal,Charleson, 677 deJongh,Grant,McGill cr-ex (Lea) particular aspects. I think that later on Mr. Seal was sort of approaching it as well from an overall statistical or econometric point of view in the sense that he was looking at the whole equation and saying if you look at everything there's balance in the whole equation. I think his testimony was touching on the CPI -- or the decision to take CPI as an appropriate inflation index and he was taking that in the context of the balancing issue that I had discussed. Q. Mr. Grant, when you were assessing this balance that might occur, did you actually assess this quantitatively; that is, did you go through calculations and numbers or did you merely qualitatively assess how these things might occur? A. My assessment was qualitative. Q. Has anybody done quantitative work? MR. SEAL: A. The quantitative work I guess that was done was -- it described how we came up with our inflation index and that was, we first of all estimated productivity. We put that in the equation, we estimated -- we knew what our customer growth was. So you've got two parts of your equation and then you try and fit the third part, the inflation variable to your actual O&M, after adjusting for productivity and inflation. So that the -- sorry, productivity and customer growth. So that the resultant adjusted O&M number, if you will, for productivity and growth is what was matched up against numerous inflation indices to determine which one Seal,Charleson, 678 deJongh,Grant,McGill cr-ex (Lea) was the closest match. That's where we determined that Ontario CPI was a good match. So that's the statistical work that went behind it. Q. Also yesterday during your testimony, probably Mr. Grant, but I don't specifically remember who, someone said that about 50 per cent of O&M is explained by the number of customers. Does anybody recall that evidence? MR. GRANT: A. Yes, that was my evidence and it was in the context, again, of what we just discussed. That reflected what the Board was saying in the EBRO 497 in establishing the appropriate O&M level for fiscal 1999. Q. Can you tell me what explains the rest then? If number of customers explains 50 per cent of growth in O&M, what explains the rest of it? A. The number of customers explains 50 per cent of the short term, the immediate year-out growth. The rest of it would be a combination of factors; wage rates and other inflation that affects our cost structure. Then, obviously, there would be a productivity offset. There's always been some productivity that we've driven out. Q. Just to explore this short-term matter, perhaps we can look at the administrative and general costs for a moment. Would I understand things correctly that administrative and general costs do not change continuously on a smooth curve? Seal,Charleson, 679 deJongh,Grant,McGill cr-ex (Lea) It is more of a lumpy or step-wise pattern in the expenditures or the O&M expenditures related to this factor. In other words, although number of customers may grow steadily, administrative and general moves in a step-wise fashion as you need more of those services; would that be correct? A. I think that's true for the vast majority of A&G costs. There are probably a few specific costs within A&G that do move in lock step, but I think that's a fair statement in general for A&G. Q. Does this relate to the question of short-term or long-term drivers of O&M in terms of numbers of customers? A. Yes. Really what it means is that within the administrative part of our business you can get by, if you will, as you add customers you can -- in the short-term without adding significantly to A&G for that reason, for the reason of customer growth. But in time what happens eventually is that these are support departments for other sort of front-line departments and you need to then staff up or do something to meet the increased demands. Q. Can you tell me, Mr. Grant then, given the characteristics of A&G, should A&G be included as an O&M base year amount in this short-term PRB proposal that we have before us, or should it be included as part of the base year and part of the productivity data set if it's going to move up step-wise and may not, in fact, reflect Seal,Charleson, 680 deJongh,Grant,McGill cr-ex (Lea) the growth in number of customers in the short-term? A. I think A&G has to be part of the base. That's what we're starting with. So my impression of your question is that if it's in the base, then moving from there forward, should it somehow be accounted for differently-- Q. Mm-hmm. A. --given its characteristics. I think the answer is no, because what you have to do is look at the overall equation that we are using for the overall O&M and understand that there will be imprecisions if you take apart O&M and look at specific amounts. But on balance you're capturing the basic drivers of total O&M. I think that's what you have to be satisfied with. Q. Yesterday one of my friends pointed out, I think it was Ms. DeMarco, something that is obvious from the interrogatory which she was referring you to. It was either 143 or 148. I don't think you have to turn it up. That was on a statistical correlation basis the closest match between O&M expenses was kilometres of mains as opposed to number of customers. I don't wish to rehash the discussion as to why that measure was not used, but did you consider using a growth factor that involved two or more drivers; that is, weighted partly for customer growth and weighted partly for kilometres of mains, for example; and if not, why not? MR. SEAL: A. No, we didn't consider using a two-part driver, if you will, for O&M growth. Seal,Charleson, 681 deJongh,Grant,McGill cr-ex (Lea) We felt that the customer number was an appropriate, easy to understand number to use in this formula. One of our objectives was to have a formula that was easily understood, not debateable because of an infinitesimal degree of detail in it or not subject to debate because of infinitesimal detail in it, but just a simple formula with the most appropriate drivers in it. So that is why we chose customers and that's why we didn't look at multiple component drivers for any of the variables. Q. Okay. One moment ago when I was asking you what else drove O&M expenses besides number of customers, have you ever identified a system growth that is independent of price changes? We talked about labour costs and so on. Is there any component of it that is independent of price changes? A. I'm sorry, could you ask that question again, Ms. Lea. I'm not sure I understood. Q. Yes. Have you identified any factor which affects the growth of your O&M that is independent of price changes? We've talked about number of customers; Mr. Grant talked about the change in labour costs. Are there any other factors that are independent of price changes that you've identified as being a driver for O&M costs? A. Productivity, which we have in the formula, is also one of the drivers, if you will, of O&M costs. I Seal,Charleson, 682 deJongh,Grant,McGill cr-ex (Lea) can't think of any others. Q. All right. Fine. That's fine. Turning to the question of the inflation factors used in the formula, you would agree with me, I presume that, if we're going to estimate productivity accurately, the inflator that you use in the formula must be relevant to productivity? A. I think I agree with that, yes. Q. Does it concern you that the CPI, whether Ontario or Canada, is a general inflator and is not or may not be completely related to those factors which drive your productivity? A. It doesn't concern me in the sense that our estimates of that particular variable within the equation are robust. In other words, that the matching over that 10 period was quite good. Q. Does it concern you that there is always the problem with Consumers price indices that they are based on a basket of goods? Consumers are presumed to continue to buy the same things in their basket and this is not, if I can call it, rebased very often and that this may lead it to be less and less relevant with time. Does that concern you? A. Not generally. The CPI is rebased every five years. It is believed to be an accurate indicator of consumer price inflation and I don't think -- you will find some people have debates with whether it is the best measure, especially more so in the United States than in Seal,Charleson, 683 deJongh,Grant,McGill cr-ex (Lea) Canada, but I think people will agree that it is a good measure of inflation. MR. VLAHOS: Mr. Seal, it used to be updated every ten years. When did that change? MR. SEAL: I believe the basket is updated every five years. It certainly is rebased every five years. MS. LEA: Thank you. Q. I wonder if I could run through with you briefly the inflation estimation technique that you used. As we understood it the process was somewhat circular, but I wanted to ask you about that. Let's take materials price inflator for a moment. The actual spending of materials is a -- a data set is derived from 1986 to 1997 and this series had growth factored out by using the number of customers; is that correct? MR. SEAL: A. The O&M data that we had was adjusted to account for the productivity estimate that we had and then the customer growth number. In other words, we take took the formula, the PRB formula that we are proposing, and moved the customers and productivity over to the left-hand side and derived what the inflation number would be and then that was matched against various independent indexes for inflation. Q. Well, let me try and tell you what I had written down for sides of the equation. You took the O&M numbers that you had, you substracted growth first; you subtracted productivity and that give you inflation? Is Seal,Charleson, 684 deJongh,Grant,McGill cr-ex (Lea) that about it in a very simplistic fashion? A. We would have subtracted growth and added productivity. Q. Added productivity, okay. That gave you inflation? A. That's right. Q. An inflation estimate? A. Yes. Q. Then did you take the data set -- and now I'm trying to get to the point where you got to productivity, okay, now we want productivity on the right-hand side of the equation. You start with the data set, you remove growth, you use that inflation estimate that you just derived and either add or subtract it as appropriate and that gives you productivity as a result? A. No. In fact, productivity is estimated -- it's the first step in the estimation. The productivity estimate is estimating the company's productivity, trying to estimate it but using physical data instead of price data, price adjusted data, if you will. So customers is our output measure for productivity. The input measures are labour and materials. On the labour side, we've taken -- we've converted our labour into full-time equivalents and that's -- there is a growth in full-time equivalents is the labour input. On the material side, because the materials category is very diverse and we don't have data on all the Seal,Charleson, 685 deJongh,Grant,McGill cr-ex (Lea) individual materials categories we've had to rely on a dollar estimate of materials. For that we take the total O&M, subtract the labour component and that's the materials. Q. Thank you for that explanation. What I was trying to get at was, the way we understood the calculation it seemed as if you derived an inflation estimate and then used that inflation estimate in deriving productivity. In other words, that it was somewhat circular, that productivity was used to derive inflation and the inflation that you used derived productivity. A. And that's not the way it was done. As I said, productivity is the first step in the process. There's a relationship obviously between the productivity variable in the equation and the other variables in the equation. You want them to have a relationship with one another. The productivity is estimated first. If I can sort of delineate the steps, you estimate the productivity first using the method I just described. Then you have your customer numbers in your equation, you have your productivity number in your equation. Then you can derive what the O&M inflation was. Then you match that up against some other number. Q. But how do you get to productivity the first time? Do you use an inflation estimate there? A. The only inflation estimate that we used, we Seal,Charleson, 686 deJongh,Grant,McGill cr-ex (Lea) did assume that the materials price index, to put it in real terms, used the Ontario CPI to put it in real terms. So there was an assumption there. Q. And when you went to determine what the best inflator was, did you use the productivity number that you had derived using the Ontario CPI? A. Yes, with the Ontario CPI being one component of that productivity, the derivation of that productivity. Q. But would it not be almost inevitable that in attempting to determine what was the best inflator, that the one that you had used in coming up with your productivity number would be the one that would be the best inflator because it's already factored into the equation once already? A. No. I don't believe that follows. Again, the Ontario CPI in the productivity estimate is strictly used to deflate the materials component of the input index which also included the labour. So it doesn't necessarily follow that because I am using that to deflate my materials component of the input into my productivity, that Ontario CPI is the best fit for the productivity or the PBR formula. Q. Okay. Thank you. MS. LEA: Madam Chair, I have 15 minutes more. Do you wish me to continue or do you wish to break and bring on Dr. Fuss and so on? THE PRESIDING MEMBER: I guess it's very Seal,Charleson, 687 deJongh,Grant,McGill cr-ex (Lea) difficult to tell how quickly things will go with Dr. Fuss. Why don't we try and finish your cross-examination, Ms. Lea? I think we will still be left, I think, with some Panel questions. MS. LEA: Panel questions, yes. ---Off the record discussion. THE PRESIDING MEMBER: I think we will go over to Dr. Fuss and -- MS. LEA: At this time? THE PRESIDING MEMBER: Yes. MS. LEA: Okay. THE PRESIDING MEMBER: And then come back. MS. LEA: All right. THE PRESIDING MEMBER: Just in case we start to run on a bit and... MS. LEA: Well... [Laughter] THE PRESIDING MEMBER: Not you; us. [Laughter] MS. LEA: That's okay. I'll accept part of the blame. THE PRESIDING MEMBER: So is that all right, Mr. Cass? MR. CASS: Yes, it is. Thank you, Madam Chair. THE PRESIDING MEMBER: Sorry, Panel. We will see you back here again. ---(Witness Panel withdraws) MR. CASS: Madam Chair, in addition to Dr. Fuss, the other member of the next panel will be Mr. Seal. MELVIN A. FUSS; Sworn, DARRYL SEAL; Resumed. Fuss,Seal 688 dr ex (Cass) DIRECT EXAMINATION MR. CASS: Q. Dr. Fuss, if I may, I will begin by just addressing a few questions to your qualifications. I understand that you are and have been for quite a number of years a Professor of Economics at the University of Toronto. Is that correct? DR. FUSS: A. Yes. That's correct. Q. And, as well, you're an affiliate of the Law and Economics Consulting Group. Is that correct? A. That's correct. Q. You have a B.Sc. degree and also an MA in economics from the University of Toronto. Is that right? A. Yes. My B.Sc. degree is in mathematics and physics. Q. You also have a Ph.D. in economics from the University of California at Berkeley, I think? A. That's correct. Q. Okay. In addition to your duties as a Professor of Economics at the University of Toronto, I believe you were Chairman of the Economics Department from 1985 to 1990. Is that right? A. That's correct. Q. You have also been a research associate at the Institute for Policy Analysis since 1972? A. Yes, correct. Q. As well, you have been a research associate for the National Bureau of Economic Research since 1983. Is that right? Fuss,Seal 689 dr ex (Cass) A. That's correct. Q. And also for a period from something like 1992 to 1995, you were editor of the Journal of Productivity Analysis. Is that right? A. That's correct. Q. And I think, as disclosed in your Curriculum Vitae at Exhibit A, section 5, you have written quite extensively on productivity analysis. Is that right? A. That's correct. MR. CASS: Madam Chair, I would ask that the Board accept Dr. Fuss as an expert in productivity analysis. THE PRESIDING MEMBER: The Board accepts him, Mr. Cass. MR. CASS: Thank you, Madam Chair. Q. Then just a few other questions, Dr. Fuss. First of all, there is an evidence list at Exhibit A, tab 8, schedule 1, showing the evidence responsibility of various witnesses. Can you confirm that you were responsible for the evidence shown opposite your name in that exhibit, Dr. Fuss? DR. FUSS: A. Yes, I can. Q. Thank you. And was that evidence for which you were responsible prepared by you or under your direction and control? A. Yes, it was. Fuss,Seal 690 dr ex (Cass) Q. Do you have any corrections that need to be made to your evidence? A. No. Q. Is the evidence accurate to the best of your knowledge or believe? A. It is accurate to the best of my knowledge and belief. Q. All right. Thank you. Now, to date in the hearing, Dr. Fuss, we have had a little bit of discussion about what's called the Averch Johnson effect, if I have correctly characterized it. Can you provide your comments on that Averch Johnson effect, please? A. Yes, I will. I have noticed reference to it from time to time in some of the previous transcripts that I have read. Basically, the Averch Johnson effect is named after Harvey Averch and Leland Johnson, two authors who published a paper in 1962, a paper which became quickly quite influential among regulatory economists. And basically what they demonstrated was that under a certain type of rate of return regulation in a certain setting, which I will describe shortly, the regulated firm has an incentive to substitute capital for labour. And this is the source of the notion that a regulated firm might become too capital-intensive based on Fuss,Seal 691 dr ex (Cass) this incentive. Now, it's sort of important just to understand the assumptions of this model because they then became the subject of some debate. It's a simple, relatively simple static one-period model. The type of regulation is stripped of many of the real-world features. Essentially what the regulator does is set the rate of return on capital and for this incentive to work, it has to be above the cost of capital. And then basically allows the firm to choose all its inputs and prices as it sees fit, and the firm maximizes profits subject to its rate of return constraint. So it's as if, for example, this Board were to hold a hearing on ROE and then walk away from the firm, say, 'This is your allowed rate of return. Go do what you want to do.' Q. If I might just stop you, Dr. Fuss. I apologize. Are you now describing the assumptions in the Averch -- A. Yes. I'm now describing the assumptions of the original Averch Johnson model, and the one that's referred to I would believe in, for example, Dr. Norsworthy's testimony and perhaps Dr. Bauer's testimony. Now, that did create quite a sensation, as I said. Most academics said, 'Gee, this is a great idea, why didn't I think of it,' and started to pursue some of the implications of the model. And there are really two strands that were Fuss,Seal 692 dr ex (Cass) pursued. On the one hand, the setting of the model was considered more carefully and on the other hand, some economists tried to empirically test the implications of the model; that is, over-capitalization. I might mention with regard to the theory side of it; that is, I noticed again in the transcripts a reference to gold plating, and the use of the Averch Johnson effect to say that it predicts gold plating. That actually is incorrect if, by "gold plating", one means what I think it means and that is the acquisition of unproductive capital - plush offices, expensive cars, et cetera. The Averch Johnson model has as one of its basic assumptions that there will be no gold plating; that is, that the firm will produce efficiently whatever it produces from a technical perspective. The inefficiency comes in because it doesn't choose the cost-minimizing mix of inputs. It chooses too much capital and too little labour. But given the amount of inputs it chooses, it produces the maximum amount of output that's possible. So gold plating comes from somewhere else; it does not come from the Averch Johnson model. That sort of is an aside. To go on now, what happened with the Averch Johnson model? Well, on the theory front, there was a series of papers, perhaps the most influential one by Paul Joskow, a professor at MIT in the early 1970s, arguing Fuss,Seal 693 dr ex (Cass) much the way I have just implied, that the assumptions of the model are unrealistic in any cost of service setting; that is, the regulator does not walk away from the regulated firm. And, in particular, the regulator has oversight over capital and considers carefully applications for additions to the capital base, rate base. The other aspect of the model which was questioned was the fact that the model assumes continuous binding of the rate of return constraint; that is, the firm always earns exactly the rate of return and it doesn't take into account regulatory lag and the possibility that the firm might, for short periods of time, be able to earn more than the allowed rate of return just because of the lag in the process. And both of those aspects - and I think for this particular jurisdiction, the fact that the regulator has oversight over capital is the important one - both of those imply that you would predict the impact to be less; that is, you wouldn't find as much likelihood of an over-capitalization effect. In later years - most of this work was done, as I said, in the '70s - in later years, actually the emphasis is turned the other way towards looking at conditions under which, in a cost of service setting, a regulated firm might actually acquire too little capital. And that's actually been more the dominant theme in the '90s. And again, probably the most influential paper is one by Gilbert and Newbery in 1994, I believe, in Fuss,Seal 694 dr ex (Cass) which they argued that because regulated firms have a lot of sunk capital - and I'm not suggesting that what I am about to say occurs in this jurisdiction - they can be held hostage by a regulatory commission who might decide to have an allowed rate of return below the cost of capital or might decide that some of the investment is, in fact, not very useful and shouldn't be allowed in the rate base. And that would then lead a firm to actually under-invest in capital because it would be risky to take a chance on that. And they refer to some of the proceedings in the 1980s, '70s and '80s in electricity generation in the United States as their example of how that might happen. On the empirical testing front, there were a rash of studies in the early 1970s, going through into the middle and late 1970s. Some studies found some effects; some studies found no effect. I myself conducted a study along with my colleague, Leonard Waverman, I guess who is a former member of this Board, in 1981 looking at Bell Canada to try to find evidence of over-capitalization due to the Averch Johnson effect. We couldn't find any at the time. And it's fair to say that the conclusion of that literature is that if the effect is there it's not very important empirically. And since the early '80s there's really been no attempt to continue that testing. It's really a dead issue. Fuss,Seal 695 dr ex (Cass) Q. Thank you, Dr. Fuss. I believe you were in the room this morning when Ms. Lawson put some questions to Mr. Seal about Dr. Norsworthy's evidence and in particular here I'm referring to the questions about Table 6 on page 23 of Dr. Norsworthy's evidence. You probably heard me say that I that going to be asking you a question about this. Could you, Dr. Fuss, comment on the data which Dr. Norsworthy has used in this table for the purposes of his capital comparison? A. Yes, perhaps just a second of background. I have, of course, been advising Consumers Gas that to construct total factor productivity estimates and the appropriate capital that goes with it is a very difficult task, difficult and time consuming. So I was somewhat surprised when I first got this document to see that Dr. Norsworthy had, in fact, produced such estimates, given what I would guess is the amount of time he had available to him to work on this particular project and I remained puzzled until I came to Table 6 when I looked to Footnote 13. And Footnote 13 says that the capital comparison is based upon rate-based data for the company and perpetual inventory capital stock for the industry. And I should probably explain what those concepts mean. Of course, you know what rate-based data is. I don't have to explain that in this setting. But for the purpose of doing productivity Fuss,Seal 696 dr ex (Cass) analysis and for a productivity analyst, other names for rate-based data are original cost data or book value of capital data. And productivity analysts believe that that is an inappropriate definition of capital for the purpose of doing productivity studies. Hence it would been an inappropriate concept to capital for the kind of comparison that's done in this table and for any of the productivity calculations that Professor Norsworthy has performed which are TFP in nature. Let me explain a little bit about what the difference is. In my evidence, I describe the appropriate concept of capital as being replacement value of capital, sometimes called market value of capital. That is, a construct of capital which takes into account inflation and what it does to purchases of capital in dollars. What we're looking for is an approximation to physical capital. That's what we mean by productive capital. And it turns out that in cases in which capital is long-lived and depreciates slowly, in which there is not a lot of technical change that using book value of capital can be very misleading, and rate-based capital can be very misleading. And I think that's true in this industry because the bulk of capital is in services and means and when I looked at some of the depreciation studies of the company, I think the lives are somewhere in the neighbourhood of 50-plus years, slow depreciation rate, so that fits. Fuss,Seal 697 dr ex (Cass) And I think I can explain it better through a simple example. Again this will be simplified from reality but it will give the idea. Suppose in 1975 Consumers Gas purchased one unit of a particular type of capital, that it was the only unit in that capital category, to make things simple. And twenty years later it decided to purchase another unit of that capital. Now, over 20 years, let's say inflation has doubled the price of the capital, so that in book value terms, that's going to be two capital. And when you add the two together, two plus one is three, so that's three capital. And that's what book value accounts would show. So it would appear as though you had gone from one capital to three capital, a tripling of the capital. Now, what would a productivity analyst do with that kind of information? What the product analyst would do would be one of two possible steps, both giving the same answer. First of all, I think you realize that the correct answer is doubling. You add one unit of capital, then you add another unit of capital. From a production account set-up, you should be doubling not tripling and so the tripling is an overestimate of the rate of growth. That's what book value does in these settings. What would a productivity analyst do? He would convert you to the 1975 capital or the 1995 capital to the other year's prices. So, for example, if you're going to measure it in '75 dollars, you would deflate the '95 number by the inflation index of two and so you would Fuss,Seal 698 dr ex (Cass) reduce two K to one K and then you would be adding one to one and you would get the correct doubling. Alternatively you could bring the earlier year's capital forward and that would mean multiplying the first 1975 capital by two and then you would be going from two capital up to four capital and then it redoubling again. So it would be consistent. But what's not consistent is the use of the book value concept. Now, it becomes particularly a problem when you mix the two, comparing apples and oranges. On the one hand, Statistics Canada uses what is generally agreed to be the correct way of measuring capital; that is, on a market value methodology. And, in fact, the most common way of doing - it is not the only way - but the most common way is to use what's called the perpetual inventory method. And I actually described the perpetual inventory method in my -- it's in the answer to one of the interrogatories that asked exactly how would you do this calculation properly. I won't go back over that but it's in one of the interrogatories. So if you think about government statistics, Statistics Canada produces total factor productivity numbers. The Bureau of Labour Statistics in the United States produces total factor productivity numbers. They all use the perpetual inventory method. Nobody uses the rate base method. I'm familiar with studies in front of the CRTC Fuss,Seal 699 dr ex (Cass) going back to the 1970s in telecom and there all the studies use the perpetual inventory method or another variant of the replacement value of capital method in a long stream of productivity studies that have been put forward in front of that Commission. They would not accept, I'm sure, the rate base methodology. Q. Thank you, Dr. Fuss. And just one last question. There has been discussion here about the choice of measure between customers and volumes. I wonder if you could provide any comment on that particularly in relation to what Dr. Norsworthy has said? A. Basically Dr. Norsworthy says that the appropriate -- one of the things that Dr. Norsworthy says is the appropriate way to aggregate output is to use cost weights to do so. I agree with that. And it's particularly correct when one is doing a methodology which involves cost recovery, which is what we're talking about here. Where weighting by revenues is appropriate is in price caps types -- versions of comprehensive plans. That's not what the company is proposing so that what the company should be doing is weighting by cost, cost drivers. And in discussions that I had with the company they convinced me that, in fact, a reasonable approximation, if you were to think, well, you've got two kinds of output, you've got access to the distribution system which is reasonably measured by customers and you Fuss,Seal 700 dr ex (Cass) have volume, and they're both outputs of the system. But as a cost driver volume does not contribute very much; whereas customers contribute quite a bit and it's the relative contributions that matter. It led me to believe that a reasonable simplification was to put the weight on customers and not to put any weight on volume in order to keep the formula simple and to avoid having, in the growth part of it having to deal with a lot of controversy. MR. CASS: Thank you, Dr. Fuss. Those are my questions, Madam Chair. Thank you. THE PRESIDING MEMBER: Thank you, Mr. Cass. Mr. Pratte? CROSS-EXAMINATION BY MR. PRATTE: Q. Doctor, my name is Guy Pratte. I represent Industrial Gas Users Association. DR. FUSS: A. Good morning, Mr. Pratte. Q. Would you, first of all, turn up your CV which is, I believe, at Exhibit A, section 5? A. I have it. Q. First of all, it's with some trepidation that I approach this because I see that at page 10 of 12 that you have, in fact, given litigators advice as to how to cross-examine economists. Now, in terms of your experience with gas local distribution companies I see from your CV that you've been retained as a consultant by Consumers and as well by Union Gas? Fuss,Seal 701 cr-ex (Pratte) A. That's correct. Q. And I see that from page 6 of 12, you've listed the consultancy assignments? A. Yes, that's correct. Q. Without giving any details in terms of the Union Gas retainer, when was that retainer; is it still ongoing? A. I'm not certain, to be perfectly honest. I've had very little contact with Union Gas. Q. Personally you were not involved in that consultancy arrangement? A. Yes, I was. What I meant was that I have not had any contact with Union Gas for some months now. Q. Okay. Did it relate to PBR? A. Yes. Q. Did you examine in that context Union Gas' own productivity, for instance? A. No. Q. And the consultancy with Consumers Gas started when? A. In August of 1997. Q. And it related to PBR? A. Yes. Mr. Pratte, I don't want to mislead you, the consultancy with Union Gas also related to PBR but not... Q. But in that -- A. I did not measure Union Gas' productivity. Q. Thank you. And have you written, aside from Fuss,Seal 702 cr-ex (Pratte) the evidence you've filed, articles or books relating to the productivity of gas utilities? A. No. Q. Have you ever until this particular assignment compared the relative productivities of gas utilities in Ontario, for example? A. No. Q. And is it fair to say that your experience in productivity analysis relates mainly to the telecom industry? A. I would say it relates to the telecom industry and somewhat earlier to electricity generation but also more broadly to manufacturing firms and service firms, I've done a fair amount of productivity analysis. Q. Maybe I can sum up my understanding of your expertise to saying that until this assignment, that is, with Consumers Gas, you had never studied the productivity of gas utilities? A. That is correct. Q. And will you agree with a comment, and I believe it's also in the evidence, but also Mr. Grant made, that there are differences between industry in productivity analysis? A. I'm sorry, could you repeat that question? Q. Would you agree with Mr. Grant's comment yesterday that when you compare productivity levels one has to be careful in importing standards from one industry to the other? Fuss,Seal 703 cr-ex (Pratte) A. I think he was referring to comparing different utilities within the same industry but would you like to give me a specific? Q. Would you think, for instance, that one could apply the productivity offset factor to the telecom industry readily to the gas industry? A. Not the specific number but the methodology. Q. Okay. Now, in terms of your retainer with Consumers Gas, you said it started in August 1997? A. That's correct. Q. And did you participate directly in the elaboration of the recommendation made by Consumers Gas to its executive about a targeted PBR plan? A. No, that was not part of my retainer. I was retained to advise the company on productivity and inflation, not on the policy decision as to what kind of a plan to adopt. Q. Okay. So I've got that clear, you were never asked whether a targeted PBR would be better than a comprehensive PBR? A. That's correct. I was just asked, depending on which the company decides to choose, how would you go about measuring the parameters of the... Q. And if I looked at your evidence, Madam Chair, which is found at Exhibit C, Appendix B, page 3 of 8, and I'm looking at your answer to question No. 7 and specifically the fourth -- I'll just wait till you get there. Page 3 of 8, doctor? Fuss,Seal 704 cr-ex (Pratte) A. Yes. Q. I'm looking at the fourth paragraph under Answer, the one which begins 'In general'; do you see that? A. Yes. Q. Without reading the paragraph into the record, am I correct to understand that paragraph as saying that in general a comprehensive PBR plan is preferable to a targeted one? A. From my perspective, if you have reasonable knowledge of all the parameters that go into it and here I'm referring specifically to having reliable TFP estimates, the answer is yes. Q. And do I understand your evidence correctly from a moment ago that when you were retained you were retained to advise on productivity and inflation? A. That's correct. Q. Did that involve as well advising Union Gas (sic) as to the appropriateness of selecting an internal benchmark for productivity? A. Did you mean Union? Q. No, I'm sorry, I meant Consumers Gas. I'll come back to Union later. I'm sorry. Let me repeat the question, if I might. A. Please do. Q. Did the scope of your retainer include advising the company as to the appropriateness of choosing as a benchmark for its productivity offset its internal Fuss,Seal 705 cr-ex (Pratte) numbers as opposed to an industry or external benchmark? A. I was involved in the search for data that might be used for productivity analysis, that is correct. Q. Let me just understand your answer correctly, is it upon your advice that the company decided to use its internal productivity data as the benchmark in the formula, or was that decision effectively made and you simply assisted them thereafter? A. I guess I'm not exactly sure of the cause and effect here, but I did look at every source of data. There seem to be two possibilities. One was the company's own internal data; the other was the Statistics Canada data. I believe since Statistics Canada data ended in 1992 it would not be recent enough to be very relevant. And so I certainly would have advised the company to use its internal data. I do not know the weight that they put on that. Q. At page 4 of 8 of your evidence in answer to question 8, you say that Consumers Gas has used a state-of-the-art methodology in calculating its historical O&M productivity growth. Do you see that? A. Yes. Q. Just to understand the import of that statement. Is this only meant to convey that the calculation for period 1997 to 199 -- sorry, '87 to '97, which I believe is found at - excuse me a second - Section 4 of Exhibit C, page 11 of 12. Perhaps you can, Fuss,Seal 706 cr-ex (Pratte) first of all, confirm that that is the calculation you are referring to? A. Yes, that is the calculation. Q. And when you say it is a state-of-the-art methodology, you are only talking about the calculation of the numbers? You're not saying that it is a state-of-the-art methodology to use those numbers as the exclusive measure of the company's productivity? A. I think the answer to that is yes. What this refers to is state-of-the-art indexing procedure which -- conditional on whatever the numbers are. The method of getting a productivity estimate is state-of-the-art. Q. But you are not saying that's state-of-the-art to use those numbers as the soul benchmark for the productivity that the company should achieve in the future? A. Not in this statement, no. Q. You agree with the statement I just made, Doctor? A. Not in this statement. Q. Thank you. Now, later on you talk about the choice of the internal measure. This is later on, on page page 4 of 8. I'm sorry, it's page 5 of 8 in answer to question 10. A. Yes. Q. You give two examples -- or you give one example of a PRB plan for Teleglobe which includes historical productivity as a growth factor, correct? Fuss,Seal 707 cr-ex (Pratte) A. That's correct. Q. And that is from a telecommunications industry perspective? A. Yes, I guess I -- Q. Then you give another one where an industry average is used, Stentor Alliance, same industry? A. Yes. Q. What examples do you know of that exclusively rely on historical productivity for the very company you're trying to benchmark in the gas industry? A. I've not studied the PRB plans in the gas industry. It was not part of my retainer to do so, other than to try to find out whether there was any data that could be used that would shed light on productivity estimates. So I guess I can't really answer the question. Q. Are you aware of any targeted PRB plan relating to the gas industry that relies exclusively on internal measures of productivity? I guess the answer is no. I'm talking to you now, Dr. Fuss. A. The answer is I am not aware. Q. Thank you. Later on in answer to question 10 you talk about having benchmarked Consumers Gas' O&M productivity levels to the industry as a whole. Do you see that in the last paragraph? A. I do. Q. Now, I am just trying to understand, first of all, what it is we're measuring. Is one of the variables Fuss,Seal 708 cr-ex (Pratte) you are looking at when you measure productivity is the O&M costs of the LDC per customer? A. Customers, the customers are viewed as the output of the process. So you could convert -- you could convert the comparison into that. That's not the way I did the comparison, but you could convert it to that. Q. You're aware - and I don't know if you actually participated in this evidence, I wasn't clear - but you are aware that a number of places in Consumers' evidence they compare their own productivity on the basis of O&M per customer? A. This is not the same calculation at all. Q. But you're aware of that, aren't you? A. Yes. Q. Let me take you there, for example, at Section 5, page 3 of 6. Are you familiar with that table? A. Not very, no. But I'm looking at it at the moment. Q. You've never read this evidence before? A. Yes, I have. Q. Oh. A. But some time ago. Q. You were here the last two days when we discussed parts of this evidence or the -- A. Yes, that's correct. Q. Okay. I will give you a moment to just look at it. Did you see this evidence before it was filed, Fuss,Seal 709 cr-ex (Pratte) Doctor? A. I certainly saw estimates of -- comparative estimates of O&M costs per customer. I'm not sure whether it was these specific numbers. Q. Okay. Have you had a chance to look at the table? A. Yes. Q. Okay. Do you agree that -- or is it your view that looking at O&M costs per customer is a reliable measure of relative productivity when you compare various utilities? A. No. Q. So are you telling me that looking at that table I can't tell anything in terms of the relative productivity of the various companies listed there? A. That's correct. You can tell something about the cost per customer. Q. But that doesn't mean anything about productivity or efficiency? A. Not by itself. Q. Similarly, if you go to the previous page, at page 2 of 6, this is a comparison now of Consumers Gas to a number of American companies? A. Yes, I see this. Q. And would you -- if you look at the -- sorry, start on page 1 of 6 and going over to that page, the company's evidence is basically that when you look at the cost, the O&M cost per customer, Consumers Gas compares Fuss,Seal 710 cr-ex (Pratte) favourably in terms of productivity? A. Yes. Q. And from what you just told me that does not follow. You can't derive any productivity conclusion merely by the fact that their O&M costs may be lower than some other utilities? A. That's right. You can only do so if you assume that the factor prices are the same across the comparisons and the mix of inputs are the same across the comparisons. Then it is a reflection of productivity. Q. Do you recall as well that there was an interrogatory that was pointed out to me, and I think it is schools number... MR. VLAHOS: 45. MR. PRATTE: Yes, 45. Q. Have you seen that? This is the comparison between Consumers Gas and B.C. Gas? DR. FUSS: A. No, this one I have not seen. Q. Okay. But you see at the bottom of it, at page 2 of 4, lines 6, O&M per customers, Consumers Gas is slightly below B.C. Gas? A. That's correct. Q. And, again, would you say that that doesn't tell us anything about the relative productivity or efficiency of the two companies? A. That's right. You have to make the assumption that the prices they pay for their factors of production are the same. If you make that assumption then Fuss,Seal 711 cr-ex (Pratte) it will give you a comparison of productivity. MR. PRATTE: Can I just have a moment, Madam Chair. THE PRESIDING MEMBER: Yes. ---Off the record discussion. MR. PRATTE: Too many binders in front of me here. I apologize, Doctor. I will be back to you in a second, if I can find it. Thank you. DR. FUSS: I have enough binders here myself. THE PRESIDING MEMBER: Mr. Pratte, are you switching topics or should we... MR. PRATTE: I can break whenever it is convenient to you, Madam Chair. I'm not going to be much longer, but probably 15 minutes. THE PRESIDING MEMBER: I think we will take our break now. MR. PRATTE: Certainly. THE PRESIDING MEMBER: 15 minutes. ---Recessed at 10:30 a.m. ---On resuming at 10:50 a.m. THE PRESIDING MEMBER: Please be seated. One comment I think from the Board. We think it is unlikely that we would move on to the other panel should we finish Dr. Fuss and the panel we are already working on today, if that can assist anybody in making arrangements. Because we are breaking between twelve and two, and I think it is highly unlikely that we would want to start the panel on performance standards this Fuss,Seal 712 cr-ex (Pratte) afternoon. MR. CASS: Thank you, Madam Chair. That's helpful. THE PRESIDING MEMBER: Sorry, Mr. Pratte, go ahead. MR. PRATTE: Thank you, Madam Chair. Q. Dr. Fuss, can I conclude from the discussion that we just had that in justifying the company's reliance on its historical performance, which you do at page 5 of 8, you did not rely on the relative differences in O&M per customer costs? That was not part of your analysis in justifying the company relying on its internal historical productivity? DR. FUSS: A. I certainly used estimates of O&M and estimates of customers. In that sense I did rely on that data. Not necessarily the specific data that appears or the specific numbers, since I presume it is based on the same data that appears in the testimony. Q. Now, in your own analysis which is referred to at the last paragraph, first of all am I right to understand that your evaluation of relative productivity among utilities ceased in 1992? A. That is correct. Q. And so you have no idea what happened to the relative productivities of the various utilities since that time? A. That's correct. Fuss,Seal 713 cr-ex (Pratte) Q. So that it could very well be that the relative advantage that you talk about in the last paragraph could have decreased or increased? You don't know? A. No, I do not know. This data -- the Statistics Canada data extends only through 1992, so this particular comparison cannot be pushed beyond that date. Q. You note in there that the relative advantage of the company went from 1986 at 21 per cent down to 9 per cent at 1992? A. That's correct. Q. It could have continued on that downward trend or switched around back to its higher advantage? A. I suppose anything is possible. Q. Okay. And perhaps you can explain to me how you've measured or from the three sources you have identified: Statistics Canada, gas utilities and Consumers Gas, what measure of productivity you looked at in order to measure that relative advantage? A. I computed what productivity economists call a level comparison of productivity. I use the word "benchmarking" here just because I felt it would be more familiar to the Board, but productivity economists call it level productivity calculations. It essentially uses the same formula as the growth rate calculations with the exception that instead of doing a calculation from year "T" to year "T" minus 1, you do a calculation from unit one to unit two. Fuss,Seal 714 cr-ex (Pratte) Q. Is the information you require to perform those calculations, would it typically be filed in a rate case for the utilities that are regulated by this Board? A. For the O&M calculation of level -- productivity levels, yes, possibly so. Q. Such that if someone looked at the rate cases for these three utilities up until 1998, one could have actually performed that calculation for the three Ontario utilities, for example? A. I guess I have to ask clarification because I don't know the answer to this. If wage rates are regularly filed, then the answer to that -- and employment levels are filed, then the answer to that would be yes. If they aren't filed, then the answer is no. Q. I take it from the uncertainty implicit in your answer that you did not ask Consumers whether or not that kind of evidence was filed and whether or not it would have been possible to extend the model or to, in fact, construct your entire calculation on the basis of the rate case data that is regularly filed with this Board? A. My analysis really isn't based on these levels comparisons. They won't play a role in the formula in any way. This was done in order to try to understand why it might be the case that the numbers obtained from Statistics Canada were quite different from the numbers for the company itself as was brought out earlier this Fuss,Seal 715 cr-ex (Pratte) morning. Q. I'm trying to understand the import of that. Does your calculation actually measure the relative productivity of companies like Union Gas and Centra Gas specifically? A. No. In theory it could, but it just takes -- compares the average as reflected in the Statistics Canada data base with Consumers. Q. And what companies are included in the Statistics Canada list? Are they only gas utilities? A. This is the gas distribution industry as defined by Statistics Canada. It is to be only those that distribute gas. They all filed information with Statistics Canada? Q. Are you saying there that of all utilities in Canada, Consumers is the most efficient? A. No. I'm just saying that it is -- it was more efficient than the average. Q. But there are some that are more efficient? A. Possibly, yes, but you can't tell that from this information. Q. One might have been able to tell that by looking at the rate filings of a selection of utilities? A. I wasn't interested in that matter at all. Q. Well, are you -- A. It plays no role in the productivity calculation and that was what my retainer dealt with. Q. Then I misunderstood an earlier answer. Fuss,Seal 716 cr-ex (Pratte) Would it have been possible to measure relative productivity in terms of looking at the O&M of other utilities by looking at the rate filings, by gathering the information filed in the rate case? A. You are now referring to productivity growth? Q. Yes. A. If the kind of detail that Consumers Gas had available, which was quite a detailed breakdown of labour, which is very important, I doubt very much if that kind of detail has been filed in rate hearings. Q. You know that, for instance - just to take a for instance - Union and Centra gas are regulated by this Board in the same manner that Consumers Gas is? A. Yes. Q. And have been so for many, many years? A. Yes. Q. You know that. Is there any reason to think that the Board would have approved over the years for those companies less efficient O&M expense budgets than for Consumers? A. No. Q. On that basis, is there any reason to believe that the Board would have approved -- let me rephrase my question. Are you assuming in your analysis, sir, that Consumers and Centra -- Centra and Union are less efficient, less productive than Consumers Gas? A. No. Fuss,Seal 717 cr-ex (Pratte) Q. No. How about B.C. Gas? You're not saying anything about B.C. Gas either? A. You are referring specifically to this calculation? Q. Or anything you have done. A. No. Q. So nothing in your evidence and particularly in this paragraph... Let me just direct you again. I will direct you to Section 5 of Exhibit C, page 3 of 6. A. Would you repeat that, please? Q. It's at Exhibit C, section 5, page 3 of 6. A. All right. Q. Are you there? A. Yes, I am. Q. If you just forget the numbers for a moment, but just looking at the list of the companies there, there are five Canadian companies. Do you see that? A. I do. Q. Is there anything in your evidence that would suggest that Consumers Gas is more efficient than any of the other four listed there? A. My evidence does not relate specifically to any of the utilities other than Consumers Gas, hence the answer must be no. Q. In other words, nothing in your evidence enables me to say that Consumers Gas is more efficient than either Northwestern Utilities, BC Gas, Union Gas or Centra Gas? Fuss,Seal 718 cr-ex (Pratte) A. My evidence is uninformative on this matter. Q. I wonder if you could just answer my question directly? A. I thought I did. Q. Nothing in your evidence enables me to say that Consumers Gas is more efficient than Northwestern, BC Gas, Union Gas, Centra Gas; correct? A. Repeat that again, please? Q. Nothing in your evidence enables me to conclude that Consumers Gas is more efficient than Northwestern Utilities, BC Gas, Union Gas or Centra Gas? A. That is correct. Q. Thank you. Now you say, Doctor, again at the bottom of page 5 of 8 that: 'To require a productivity growth factor...' Sorry, let me rephrase that. You conclude that the growth factor should be 0.6 per cent, and to do so otherwise would unfairly penalize the company; correct? A. That's right. Now, all I'm doing here is comparing two numbers, the number from Consumers Gas' own experience with the number that comes out of the Statistics Canada database. There's no broader implications in this particular comparison. Those are the only two numbers that were on the table; choose one, choose the other. I say choose the 0.6. Q. Okay. And the 0.6 falls from where? A. That's my rounding of the company's evidence. Fuss,Seal 719 cr-ex (Pratte) Q. And that's what we find at page 11 of 12? I'm sorry, section 4, page 11 of 12, the table that I took you to? A. I think the number is 0.63. Q. 0.63. Okay. And what you're saying is any other number than that would unfairly penalize the company; any other number than 0.63? A. That's not what this says. What this says is if you use 1.6 instead of 0.63, that would unfairly penalize the company. Q. What if I use 0.7? A. You need some reasons. I'm just making a comparison here. Q. Well, Doctor, you're telling me -- A. Well, let me put it this way: Yes. For the productivity offset part of the formula, 0.63 is what I recommend. Q. And am I correct in understanding that looking at that table, you're saying extending this forward to future years at 0.63 is the only way to ensure the company will not be unfairly penalized? A. For the productivity offset part of any formula. Q. Okay. Fair enough. I appreciate... For the productivity. And you would suggest then that that would apply Fuss,Seal 720 cr-ex (Pratte) to the years - if you look at the table, it might be helpful - for the years 1998 and '99, would that also be the case? The calculation finishes in '97. A. Yes. Q. And so it follows from your logic that you should do that for all future years, beginning in 1998? A. As an average indication; that's right. Q. Now -- A. Any one year may differ substantially. Q. You know that the company has calculated that as a result of EBRO 495 and 497, you will see that at page 4 of 6 in section 5, the growth factors would be 10.89 and 1.54? A. I do remember those from yesterday. Q. And that is substantially above the 0.63 you recommend; correct? A. That's correct. Q. And is the company unfairly penalized by those growth factors that follow from the Board-approved decision? A. There was a one-time substantial reduction in O&M that is outside of the application of the productivity factor; that is, that's more in the -- in these kinds of calculations, that would be more in the form of a "Z" factor. Q. Are you talking about 1998 or 1999? A. I believe -- the one that leads to 10.9. Q. All right. What about 1.54 in 1999; does Fuss,Seal 721 cr-ex (Pratte) that unfairly penalize the company? A. If one believed that that number was sustained over the full period of the plan, yes, it would. Remember that this average includes a number. One of the numbers is 4. Other numbers are negative. What we are talking about here is applying the average as a best guess for the future. It's not expected that in any one year it will exactly match the company's experience. Q. The average for those two years, 1998 and 1999, is roughly 6.2 per cent, if you average those two numbers? A. My understanding is those are not actual numbers; those are forecasts. Productivity analysts don't use forecasts. Q. The numbers in the formula turn out to be a forecast, aren't they? The productivity offset is something you extend into the future as the best estimate? A. That's correct. But what I'm referring to was calculations of productivity offsets for... Best guesses into the future use actual numbers. I don't know whether the company will, in fact, achieve these kinds of productivity growth rates. If they do, then of course that's another story. But at this point, I don't know that they have actually done so. Q. Let me put my question this way, Doctor. Don't the growth factors implicit in the Board's approved O&M budgets for 1998 and 1999 suggest that Fuss,Seal 722 cr-ex (Pratte) greater productivity could be achieved in those two years? MR. CASS: Pardon me, Madam Chair. Just for clarity of the record, Mr. Pratte has I think twice referred to the 1999 number as a Board-approved one. I think if one even looks at the filing date at the top of this page, it would be apparent that that's not the case. MR. PRATTE: Well, let's deal with 1998 then. Do you quarrel with that? MR. CASS: (Shaking head) MR. PRATTE: Q. 1998; doesn't that number suggest, Doctor, that the Board-approved O&M budget had room for greater productivity improvements relative to prior years? DR. FUSS: A. It does indicate, if it's achieved, that there could be a one-time increase in the productivity level. But it has no real implications for future rates of growth. Q. Well, let me suggest to you that your reliance exclusively on the 0.63 past average could underestimate the actual productivity improvements that could be made in subsequent years? A. Could underestimate it; could overestimate it. MR. PRATTE: Thank you. Those are all my questions. I'm grateful to you, Doctor. THE PRESIDING MEMBER: Thank you, Mr. Pratte. Fuss,Seal 723 cr-ex (Lawson) MR. BRETT: Madam Chair, Ms. Lawson has asked if I could precede her and I don't know whether -- I think -- I don't know if Mr. Warren has questions. I have no problem with Ms. Lawson going before me. She has to get away apparently. THE PRESIDING MEMBER: Thank you, Mr. Brett. Mr. Warren, do you have questions? MR. WARREN: I thought we agreed that Mr. Pratte would cover off this, as I will do the next panel rather than him. THE PRESIDING MEMBER: Thank you, Mr. Warren. Ms. Lawson, would you like to go ahead then? MS. LAWSON: Thank you, Madam Chair. CROSS-EXAMINATION BY MS. LAWSON: Q. Dr. Fuss, good morning. DR. FUSS: A. Good morning, Ms. Lawson. Q. I would like to just begin by reviewing the differences that you have with Dr. Norsworthy on productivity calculations, and I have gone through and I identify five. Let me just put those to you and see if I'm missing any or mischaracterizing any. A. All right. Q. The first would be that he measures total factor productivity and you would agree with the company that we should be looking only at multi-factor productivity, specifically O&M productivity. Is that correct? Fuss,Seal 724 cr-ex (Lawson) A. Should I respond to each one individually? Q. Sure. A. The company has put forward a proposal which is a targeted PBR proposal. It is multi-factor productivity which is consistent with that proposal. Q. Right. And therefore you don't think we should be looking at capital or gas, for example? A. Not if what's on the table is the company's proposal. Q. All right. That's No. 1. No. 2, as you explained in your direct examination, you believe that if you are looking at capital, that the deflator should not be based on a book value measure. It should be based purely on a market value measure; correct? A. The capital itself should be based on a replacement value or market value. This is not deflator; it's the capital itself, productive capital. Q. Okay. I'm sorry. I misunderstood that. But that's another difference you have with Dr. Norsworthy? A. That's right. In his testimony, he has based his productive capital estimates on book value, and I disagree with that. I believe it should be replacement value. Q. But insofar as you're saying we shouldn't even be looking at capital, how is that relevant? A. It's not relevant to the company's proposal. I agree with that. It's relevant to an understanding of Fuss,Seal 725 cr-ex (Lawson) Dr. Norsworthy's testimony. Q. Thank you. A third point would be that, am I correct, that Dr. Norsworthy's focus on productivity growth does not account for what you have described as this level productivity comparison between ECG and the industry? A. Not in terms of how a formula would work. The formula depends only on growth calculations. But in understanding why there might be a difference between the company's productivity growth rate and a measure that's derived, say, as an average of the gas distribution firms in Canada, one needs to look at possibilities that they started at different levels. I might point you to Table 1 of Dr. Norsworthy's testimony. Actually, in my copy it says Table 3, but I assume it's Table 1. Q. Page 13? A. Yes, page 13, in which he provides evidence that utilities which start out at the beginning of a period with high productivity levels tend to have lower productivity growth rates. And that's consistent with the evidence that I have been presenting. Q. Okay. So you may not actually disagree on that point then. A. (No verbal response) Q. Okay. Then the fourth point is that the output measure in calculating productivity should, in your view, and that of Consumers, be customers; whereas, Dr. Fuss,Seal 726 cr-ex (Lawson) Norsworthy is saying it should be volume weighted by revenue; correct? You disagree on that? A. Again, the answer isn't quite that simple. For the purpose of a PBR plan which involves cost recovery, customers, because it is the cost driver, is the appropriate measure of output. In a comprehensive plan which would be of a price caps type, then revenue-weighted volume is the correct measure. So, once again, it depends which plan you are looking at. For the plan that's on the table, which is the company's PBR plan, customers is the correct measure of output. Q. Thank you. And finally, by my reading of the evidence, the final area of disagreement has to do with the correct deflator to use in the measure of the "X" factor. Whereas, you are saying that the Consumers' price index is fine and appropriate in that regard; whereas, Dr. Norsworthy is saying that a specific price index on the materials inputs of Enbridge Consumers Gas should be constructed or at least on the inputs of the gas industry? A. Okay. Here is where we get into some testimony that I think will be quite surprising to everybody here. It doesn't matter. The reason it doesn't matter is that if you set up the proper formula for a PBR plan, it actually has an extra term in it which appears in Dr. Norsworthy's math - Fuss,Seal 727 cr-ex (Lawson) and I won't take you to the math, I promise - which the company has not included for a particular reason. It's called usually in this literature the input price differential. But more accurately what it is is the difference between rate of growth of inflation that's used as the external marker of inflation and the rate of inflation of the input prices for the company. Now, when the two are the same, of course that term disappears. And one of the reasons why Consumers chose the Consumers Price Index as its external inflation index is because it simplified the formula. It meant that this last term didn't have to be included at all because it was zero and simplicity is a virtue in these kinds of situations so that seemed like a sensible thing to do. Besides it was an intuitively a sensible measure on other grounds. But if you then switch to Dr. Norsworthy's price deflator for materials, what you will do is create an input price differential that didn't exist before. No problem with that. But it will exactly offset the change in the productivity index and the factor -- with the so-called "X" factor that goes into these plans will remain unchanged. It will be identical. But I see by the look on your face you're as surprised as I thought you would be. Q. I'm not sure I fully understand but I'm not going to ask you to go over it again, Dr. Fuss, so... A. You can ask Dr. Norsworthy about it. Q. Yes, I'll review the transcript, thank you. Fuss,Seal 728 cr-ex (Lawson) I did want to ask you a few questions on the issue of the deflator but before you move to that I just want to make sure that my understanding of your differences with Dr. Norsworthy on this issue of productivity for Consumers Gas are complete. Is there anything that I haven't mentioned that you take issue with in Dr. Norsworthy's methodology? A. Well, I guess -- I think we've covered the most important points. I suppose in general the most important point is the difference I have with Dr. Norsworthy on the construction of the capital stock. Much of his evidence flows from that construction, in particular, his comparison in Table 6 of the company trends in capital labour per unit output flow from that. And if you'll notice, in that table he estimates that the capital per unit output has increased four and a half per cent a year for Consumers and declined by 1.67 per cent for the industry as a whole which means the differential is over 6 per cent per year. Now, over an eight-year period that compounds to 50 per cent which is saying that, if they started with the same capital intensity at the beginning of this period, Consumers Gas is 50 per cent more capital intensive than the average industry. Taking off from a question I just had - maybe it was Mr. Pratte perhaps who said - does anyone believe that this Board would allow Consumers Gas to have a 50 per cent Fuss,Seal 729 cr-ex (Lawson) higher capital intensity than the industry norm? I don't and it doesn't. The problem is in the numbers. The data are wrong. Q. And your explanation is the use of book value versus product value? A. That's right. Q. So have we covered then-- A. I believe we have. Q. --the key differences. Great. Okay. Maybe I'll just deal with the deflator issue since we were discussing it so recently. Now, if I look at your testimony on this issue, it's Question and Answer 11 at page 6 of Appendix B. In the very last sentence you state that: 'Both the Ontario CPI and the Input Price Index for Consumers Gas have increased at a rate of 3.1 per cent per annum.' And I understand from Board Staff Interrogatory No. 50, that's Exhibit D, section I, point 50, that you did construct an inflation index for the company's O&M expenditures; is that correct? You did that calculation? A. I guess that's part of the productivity calculation. It's a byproduct of doing the productivity calculation. Q. Right. So you calculated an input price index for Enbridge Consumers Gas? A. It was computed from the productivity database which has labour and materials as sort of the two Fuss,Seal 730 cr-ex (Lawson) inputs. Materials is quite aggregate. It isn't detailed in categories but it's quite aggregate. In fact, it's just one number, the cost of materials. And then that was deflated. MR. SEAL: A. Could we have one moment. Could we have the question again, Ms. Lawson, because I'm not sure we're talking about the same thing. Q. Maybe I'm confused here. In your testimony at page 6 you talk about -- you refer to an input price index for Consumers Gas. Is that a different measure from - I'm looking at page 2 of Board Staff Interrogatory 50 - what you referred to as an inflation index for the company's O&M expenditures? DR. FUSS: A. It is a different... Q. Oh okay. I'm sorry, then I was confused. Let's just go back then. Forget about Board Staff Interrogatory 50 and let's talk about the Input Price Index for Consumers Gas that you have referred to in your evidence. Did you calculate such an index? A. Consumers Gas calculated it and I guess it doesn't appear in the evidence. Q. So you just relied on the company's calculation? A. Yes, I did oversee the calculation, so that I can vouch for the methodology but not necessarily the arithmetic. MR. SEAL: A. Maybe I can just help, Ms. Lawson. Fuss,Seal 731 cr-ex (Lawson) That Input Price Index is the one that I was speaking of earlier this morning. It's the prices - the O&M inflation that occurs after you account for productivity and customer growth. That's what we're talking about here. Q. Okay. Did you, Dr. Fuss, construct a similar price index for the Canadian gas industry? A. No. Q. You didn't think that it might be worthwhile to do that and compare that to Consumers Gas? A. Now, we're talking about the materials price deflator again I believe, are we not? Q. Yes. A. Yes, remember I just said it doesn't matter. Q. Okay. But you do agree that prima facie a deflator based on inputs purchased by the industry or by the company in particular is a better measure than a consumer price index, a general consumer price index, for deflating a company's non-labour inputs? A. Not necessarily. There are many problems with constructing such an index and I might refer you to Dr. Norsworthy's data on this matter. I have actually gone through Dr. Norsworthy's spreadsheet that was kindly provided to us. Looking at the data that he used to construct his productivity estimates for the '87 to '95 period, because I was particularly interested in the period after '92 when no published data exist. Fuss,Seal 732 cr-ex (Lawson) And I looked at some various measures and I was struck by the materials price index that happens to be in that data set. Now, I should clarify that this material's price index is a subset of the materials that Consumers Gas uses because Statistics Canada has a slightly different category. It has services and energy and materials and those are all lumped together in Consumers Gas, what's called materials. But just looking at the materials price index in that data set, I note that over the 1987 to '95 period it increases by 27 per cent per annum. And in one year it goes up 140 per cent. I guess I wouldn't say that that was necessarily a great measure of Consumers Gas price index of materials. Q. It appears that, Dr. Fuss, consistent with what you've just told us and from what you say in response to the School's Interrogatory 41 -- no, not 41, sorry. I've lost that interrogatory reference, I'm sorry. Actually I think it's a Board Staff Interrogatory. Sorry, I can't find the interrogatory reference. But in any case, it's my understanding that it's generally thought that the consumer price index rises faster than producer price indices; is that not a generally accepted notion? A. Yes, that's right. Q. But you're disagreeing with it in this case? A. No, I'm just saying -- I'm sorry, you should give me the context. Fuss,Seal 733 cr-ex (Lawson) Q. Well, is it possible that the Ontario Consumers Price Index overstates inflation in Enbridge Consumer Gas's purchases of non-labour materials and services inputs? A. It is possible but it wouldn't make a difference. I'm sorry, I should expand it. It would not make a difference as long as the formula is calculated correctly so that the input price differential is included if there's a divergence. Q. Okay. Thank you. I'd like to move to the issue of the output measure-- A. Fine. Q. --for a moment, and in answer to Question 9 in your testimony at page 5, you note just at the end of your answer there that: 'According to Statistics Canada data O&M productivity for the aggregate Canadian gas distribution industry grew at an annual rate of 1.6 per annum whereas Consumers Gas's O&M productivity growth rate calculated from its database was 0.6 per cent per annum.' Now, what measure of output is the Statistics Canada measure of O&M productivity based on? A. I should give you two answers to this. First of all, this comparison is based on a common measure of output. Customers were used in both calculations. Q. How did you do that? Fuss,Seal 734 cr-ex (Lawson) A. Well, I removed the Statistics Canada output growth rate and substituted the customer growth rate for it. But to answer your question directly, Statistics Canada uses volume as its output measure. I should also mention that it is doing total factor productivity calculations. Q. Are there not other differences between the Statistics Canada measure and yours? A. Well, the all of the data are difference in some sense. Q. It's a measure -- I'm thinking of the methodology. A. No, the methodology is the same except for that one difference. Q. I take it that you take issue with that approach in respect of a targeted PRB? A. By "that approach", you mean using weighted volumes. Q. Yes. A. Yes, that's right. Q. Have you ever advocated the use of customer output measure in a productivity calculation? A. I've actually never done a productivity calculation for a targeted plan before. Q. Would you agree, Dr. Fuss, that the productivity growth rate for Enbridge Consumers Gas' O&M would be lower than the 0.6 per cent if you had used Fuss,Seal 735 cr-ex (Lawson) revenue weighted volume as the output measure? A. Yes, I agree. Q. In fact it would have been negative. I think you provided a number in an interrogatory-- A. That's right. Q. --response? Finally, Dr. Fuss, I think you said in your answer, A10 on page 5, that normally performance-based regulation measures productivity using industry wide productivity benchmarks and you gave the Stentor example? A. I think you are putting words in my mouth here. Q. But -- A. I gave two examples. I said you can find both kinds of... I might mention that for the industry wide and telecommunications, there is a little bit of semantics here because this was put forward by the Stentor Alliance as a single productivity number for that group of firms. It was one submission by a group of firms using all the data. It was an internal submission from the Stentor Alliance. So in both cases internal data were -- historical data were used. Q. But several different companies? A. That's right. A number of companies were averaged. Q. I'm just interested in the Teleglobe example, as Teleglobe being your example of a PBR formula using the Fuss,Seal 736 cr-ex (Lawson) company's own historical productivity growth-- A. That's -- Q. --to determine the -- is it not the case that there is no easily identified comparable firm for Teleglobe? Teleglobe stands on its own as an overseas wholesale telecommunication supplier in Canada? A. That's correct. Q. Whereas in the gas distribution industry there are lots of other comparable firms? A. Well, "comparable" is a relative term. There is lots of difference in geography, for example. That's why one has to do some level comparative analysis before jumping to conclusions as to what rates are reasonable. MS. LAWSON: Thank you very much, Dr. Fuss. Those are all my questions, Madam Chair. THE PRESIDING MEMBER: Thank you, Ms. Lawson. Mr. Brett. MR. BRETT: Madam Chair. CROSS-EXAMINATION BY MR. BRETT: Q. Good morning, Dr. Fuss. My name is Tom Brett and I act for-- DR. FUSS: A. Good morning. Q. --instutional customers of the company. I only have a couple of areas of questioning for you. Just before I get into those, in one of your exchanges with Mr. Pratte you said that, when you were talking about a growth rate unfairly penalizing Consumers Gas for its past relative efficiency, this is your -- I Fuss,Seal 737 cr-ex (Brett) think you said as between .6 and 1.6 per cent. A. Yes, but the context here is that I was -- I had really two data sets that one could possibly use to generate this number. Q. Right. A. The Statistics Canada data set and the company's own data set. So I was really choosing between the two numbers. Q. I understand. My question really is -- what I want to put emphasis on is that you said as between the company's number of .6 and the Stat Can of 1.6 it would be inappropriate to use -- if you had to choose between those two it would be inappropriate to use the .6. I take it you were not saying necessarily that it would be inappropriate to use some number in between those two, 1.2, 1.1, something that sort? A. No, that's too high. I would say that it is inappropriate for the productivity offset part of the formula to use a number which is significantly above the .63. Q. Your basis for saying that is that the -- that your level productivity study that you have done comparing the average is the basis for that comment, the study that you have done that you elude to in the bottom paragraph of page 5 of your evidence that suggests that the relative advantage of Consumers in productivity terms over the average in the industry is 9 per cent. Is that what you're basing on? If not, what would you base that Fuss,Seal 738 cr-ex (Brett) on? A. What we're looking for here is some productivity growth rate that the company can reasonably assume to aspire to as a hurdle in the future. Q. Right. A. I don't think anybody knows what that number is. Q. Right. A. But I believe it's close to historical experience. If you find that a company seems to be reasonably efficient in its past, then the historical experience is a pretty good guide. But if you discover that that company was inefficient in the past, then you might want to move the number up considerably because there is a lot room for improvement. If you don't believe there is a lot of room for improvement, then you want to keep the number pretty close to the historical number. So I was trying to look at that issue. Q. I take it you don't say that the number has to be identical to its historical experience? You would accept the proposition that it could be -- there could be, to use a layman's term, a bit of stretch in the number to induce the company to greater effort? A. Yes. Actually I have given this some thought because I assumed it would come up. And actually I do Fuss,Seal 739 cr-ex (Brett) depart with the company policy on this matter and I do agree with you, and I would recommend a stretch of 10 to 15 basis points. Q. Tell me, in arriving at your opinion that the company was relatively efficient in the past, I assume you based that on what you referred to earlier as your level productivity study that you refer to in the bottom of page 5 here? A. That's correct. Q. And could you just -- I understand -- I read the words here: you talk about a productivity advantage of 21 per cent in '86 decline to 9 per cent in '92. Can you tell me in layman's terms, if you might, just how that works? I take it a level productivity study is not talking about productivity growth-- A. That's right. It is an attempt -- Q. --for example, from year to year? It is a snapshot of the year? A. It is an attempt at a comparison between two groups or two firms. More difficult to do. Basically what this calculation tries to do is to take the O&M expenses, and you could define this as O&M expenses per customer, and take out the factor prices so that you could be left with approximations to physical comparisons. And that process then leads to this relative number. Q. So you started off with the O&M -- you start with the O&M expenses per customer? Fuss,Seal 740 cr-ex (Brett) A. That's right. You could -- Q. Your point is that you don't end with the O&M customers, O&M expenses per customer? A. That's right. Q. That's what you said to Mr. Pratte several times -- A. That's right. Go one step further. Q. If I could just finish my question. A. I'm sorry. Q. It would be inappropriate to use the O&M expenses per customer as themselves in an untreated form, if I can put that way? That would be terribly misleading? You have to adjust them? A. Well, not necessarily, again. It depends on the situation. Basically what you have to decide, if you are going to use the O&M expenses per customer to measure relative productivity is whether you think the factor prices in the two situations are approximately the same. Q. The factor prices being the prices the -- A. Wage rates. THE PRESIDING MEMBER: Excuse me. DR. FUSS: Excuse me. THE PRESIDING MEMBER: Can you each talk one at a time. The reporter is having a terrible time. MR. BRETT: I'm sorry. I'm sorry to interrupt you. A lot of this terminology is-- DR. FUSS: I understand. Fuss,Seal 741 cr-ex (Brett) MR. BRETT: --is not known to me. I'm not an economist. DR. FUSS: I understand. MR. BRETT: Q. When you say a factor price, I take it you mean the price of wages and the price of equipment and supplies that the utility is using? DR. FUSS: A. Not equipment. Not capital, but-- Q. Services? A. --labour, service, materials. The inputs that go into the O&M category. Q. Okay. You're saying you start with the -- in this level comparison, in your case you were comparing the bundle of utilities that compromise the -- or compose the Stat Can gas distribution set of numbers and Consumers Gas, and you start with the O&M numbers per customer, dollars per customer and then you take out these various factor -- you adjust for any variations in the factor prices between those two pieces you are comparing; is that it? A. That's right. Q. Having done that, is that then your final answer? A. That's the final answer. Q. When you were doing your comparison did you find significant differences in factor prices and, if so, which ones? A. I actually did this comparison two ways. I did it assuming -- in all cases, I assumed that the Fuss,Seal 742 cr-ex (Brett) materials price was the same in both -- both for the distribution industry as a whole and Consumers. I did two calculations and in one I assumed that the wage rate was the same, hence essentially reproducing the O&M expenses per customer. When I did that, Consumers Gas was still more efficient in both '86 and '92. So that, in fact, I should have told Mr. Pratte that, that actually means that in this particular instance the calculations that Consumers did, just looking at O&M expense per customers, was an accurate reflection of efficiency. When I did the other calculation, which is the one reported here, the qualitative results did not change, but the quantitative results changed somewhat, and I did estimate that Consumers Gas paid somewhat lower wages than the average in the industry. Q. So that the -- all right. In terms of the prices, did you do the same exercise with respect to materials and other service inputs or did you simply assume that they were -- A. I assumed they were the same. Q. Okay. So that effectively, then, you're not getting very far away from -- when you assume they're the same, when you talk about either wages or materials and inputs, and I take it that the balance between labour and the other inputs, the mix as it were, did you adjust that in any way in the comparison or did you assume the same-- A. No. Fuss,Seal 743 cr-ex (Brett) Q. --mix? A. It is taking into account, whatever the mix is. Q. Okay. A. It roughly is -- it turns out it is about the same. It's about, if I remember correctly, it's about 70 per cent labour as a cost share and 30 per cent materials and that was true both for Consumers and for the industry as a whole. Q. So that -- having done this, you do this analysis and come to the conclusion -- In some sense you are not getting too far removed from the dollars per customer? A. That's true in this case. It's not -- it doesn't take us very far away from the dollars per customer. Q. Would you agree with me, I know you are not an expert in the gas industry, per se, but would you agree with me that the physical configuration, the quantities of people and the quantities of materials required on a per customer basis to run a gas business may well differ considerably from company to company based on such things as characteristics of its franchise; I think you mentioned geography earlier, age of the plant, configuration of customers as between various types of customers? A. I would agree with all of that. That's certainly true. That's why these kinds of calculations have to be viewed as approximate. Fuss,Seal 744 cr-ex (Brett) Q. I just want to -- I don't want to go on very much with this, but I did want to just refer you to a study that was done by Mr. Milne. This is Exhibit D, 5.19. This is an interrogatory response to the Consumers Association of Canada. It was done by Peter Milne and Associates for the Ontario government which was back in 1994. I just want to refer you to page -- little page vii there, seven Roman numerals, as part of this executive summary. A. Okay. That's the last page of the Roman numerals? Q. Last page of the executive summary. A. Mm-hmm. Q. And the first bullet, the first square bullet, I just want to read a passage from that. You see there where this study assesses the comparative costs of a group of utilities and deals with various characteristics of them and how that might impact on their cost structures. He says in the first: 'Natural gas distribution utilities are highly heterogeneous or unique organizations in terms of their integrated structures, financial structures, market structures, and natures of the franchise. It is unlikely that any two large North American LDCs could be considered identical or even similar across all of these key characteristics.' Fuss,Seal 745 cr-ex (Brett) Does that strike you as a fair characterization? A. Well, certainly not similar or, I'm sorry, certainly not identical. "Similar" may reflect the circumstances and things you're trying to compare. Q. All right. Could I just ask you then to go briefly to Dr. Bauer's evidence? If you could have a look at that, I want to talk a little bit about the use of an individual productivity rate as opposed to a general rate. And if I could ask you to turn up Dr. Bauer's evidence or have someone put that in front of you. Do you have that? A. I do have that, yes. Q. Okay. If you look at page 11, the issue I want to address here is the use of a historical number of productivity. If you look at the second paragraph, the first full paragraph on page 12 of Dr. Bauer's evidence, it starts off, the second sentence: 'Similar arguments apply to the proposed measure of expected productivity developments. Reliance on the historical performance of the utility cannot be considered an appropriate performance target.' A. I am sorry. I don't have it. Q. Have I lost you? A. Yes. Are you on page 12? Q. I'm on page 12 of 21. A. Okay, fine. I was on 11. Fuss,Seal 746 cr-ex (Brett) Q. Sorry. I think I mentioned both pages. I apologize. A. Would you start again, please? Q. Yes. I'm at the first full paragraph on the page: 'Reliance on the historical performance of the utility cannot be considered an appropriate performance target.' He goes on to say: 'One of the basic premises of PBR is that traditional regulation contributes to 'X' inefficiencies.' I take it he means productivity inefficiencies or inefficiencies in running of the business. A. Yes. Q. 'These are inefficiencies that do exist, but cannot be exactly pinpointed. The prefiled evidence accepts this notion of 'X' inefficiencies when arguing for the advantages of PBR. Thus, it also acknowledges that the historical status quo is characterized by such hidden inefficiencies. A logical consequence is that the historical record of the utility is not a good benchmark for expected future productivity gains. Rather, as has been argued above, the overall industry trend should be used for that purpose.' Could you comment on that? I mean, do you agree with that Fuss,Seal 747 cr-ex (Brett) as a general...? A. Well, I don't agree with the punch line, which is the last line, necessarily. I mean, after all, he's comparing two types of historical records. To the extent that there might be inefficiencies, they could be in either group or they may not be in either group. One may have them; one may not. Q. Isn't he, in addition to comparing two types of historical records, also commenting on the fact that you are moving to a new form of regulation and that he assumes in this statement - and I take you don't disagree with his assumption here in the early part of the paragraph - that inefficiencies typically would exist under cost of service regulation which might, in respect of which the point of PBR is, in part, to root these out by giving the company a longer period and more incentive to do this? A. Conceptually, I agree with that and that's why I agreed with you that a stretch factor of 10 to 15 basis points would be appropriate. Q. How did you arrive at the 10 to 15 basis points? A. Actually, from my telecom experience. I looked at the stretch factors that were used there and I think it's reasonable to use proportionality as the means of determining what it ought to be in the sense that the TFP growth rates - and TFP was used there - are much higher because of the opportunities for Fuss,Seal 748 cr-ex (Brett) technological advance in that industry and they were using - well, give you an example - Teleglobe used a 1 per cent stretch factor on a base of 7 per cent growth rate. So 7 per cent, you might realize a difference in industries. In U.S., AT&T used 0.5 on a base of 2.5 or it was used for AT&T. In Canada, for The Stentor Group, a 1 per cent stretch was used on a base of a little over 4 per cent. As a proportion, those are roughly increases of anywhere from 15 to 25 per cent, so 20 per cent increase in Consumers Gas's historical record would be roughly 12-and-a-half basis points, so that's why I say 10 to 15. Q. Can I ask you, going back to Dr. Bauer for a moment and his comment that the overall industry trend should be used as opposed to the company trend because you want an external benchmark, do you have any further comment on that? A. Let me present the following case that I wouldn't agree to, but Consumers might. Suppose that the industry was much more efficient than the company and so, because of that, it had a very low growth rate over time because there just wasn't much it could do. Well, you might say if we're going to have a rigid rule, use the industry growth rate, we will use that growth rate, and maybe it's going to be zero. As I say, Consumers may agree to that. I certainly wouldn't. Q. Your point is, I guess, in sum that you need Fuss,Seal 749 cr-ex (Brett) to look at all of the circumstances-- A. That's right. Q. --surrounding the choice you're being asked to make; look at the industry experience, look at the experience of the company, and make your decision based on that? A. I think that's right. The Board has to look at all of this kind of information in making its decision. There are no rigid rules here. Q. Could I ask you to look at for a moment Exhibit D.4.35? That's a response to an interrogatory from my client. You will have to have the binder, Mr. Seal will have to give you the binder. That's a decision of the California Public Utilities Commission on Southern California Edison, and it's Item 9.1 in that binder? MR. SEAL: A. Sorry. Can you give me the interrogatory? Q. Yes. D4.35. MR. BRETT: Madam Chair. Excuse me, just while they're looking that up. I don't know how tight your situation is, if you want to get out of here more or less at twelve, I would suggest that I'm going to be a few minutes. I'm not going to be long, but I don't want to compromise your situation. And I don't -- THE PRESIDING MEMBER: We have a little flexibility, Mr. Brett, but not very much. Fuss,Seal 750 cr-ex (Brett) MR. BRETT: Okay. THE PRESIDING MEMBER: Do you think you will be more than another 10 minutes? MR. BRETT: No, I don't think so, no. But I... MS. LEA: If it's of assistance, I have one question. THE PRESIDING MEMBER: Okay. Well, let's press on for a few minutes. I will take the liberty of cutting you off, if it looks -- MR. BRETT: Yes. Just cut me off because, for my part, I don't want to focus in a way that's misleading. MR. BRETT: Q. I want to have you look at page 14. DR. FUSS: A. I'm sorry. I think we're having a little trouble finding it. Q. All right. Well, this is the decision dated 96-09 of '92... Sorry. It's dated September 20th, 1996. ---Off the record discussion. MS. LEA: September 20, 1996? MR. BRETT: Yes. It's the decision on which I guess you base your summary of the Southern California Edison case in your evidence. It's a decision written by the Chairman or the former Chairman of the Utilities Commission, Fessler. MR. BRETT: Q. You may be aware of him, Dr. Fuss. You may have even testified there, but he's -- I want you to turn to the second paragraph on page 14. He's talking here about the productivity. This is his section Fuss,Seal 751 cr-ex (Brett) on the role for PBR regulation, and he says: 'To make this update of utility rates independent of the utility's cost, the price and productivity value should come from national or industry measures and not from the utility itself. The independence of the update rule...' The update rule is his CPI minus 'X' which is the inflation minus the productivity factor. '...independence of the update rule from the utility's own costs allows PBR regulation to resemble the unregulated market where the firm faces market prices which develop independently of its own cost and productivity.' And so he sets that out as a basic principle, which -- do you see that? Do you disagree with that? DR. FUSS: A. I think the basic principle for choosing the productivity offset is that it cannot be manipulated by the firm. That's what I think of as independence and so -- and the historical record, of course, can't be manipulated by the firm during the PBR period. Secondly, with regard to trying to simulate market forces, that's a great idea in theory, but when you're dealing with a number of local monopolies, they're not going to compete with each other. BC Gas and Union -- not Union Gas, but BC Gas won't compete with Consumers Gas certainly. And so when you're looking at the kind of industry that's a Fuss,Seal 752 cr-ex (Brett) statistical industry; not an economic industry, but a statistical industry, you're not replicating in any way the forces of a competitive market. Q. No, you may not be. But aren't you trying to simulate them? Aren't you trying to create pressures on a company that are -- A. Well, I just gave you an example of a case where you wouldn't be creating any pressures on the company by using a very low productivity number taken from the industry average. Q. All right. In that case, you wouldn't in the case you specified. But in the other case, in the other way around, you would; right? A. Well, if you always choose a large number, you do. But I have no reason for choosing a large number. Maybe the industry will give me a small number. But if you decide that the industry number is always going to be large and therefore you want to go that way, I suppose you can do so. But there isn't really an economic theory which tells you that those numbers will simulate competitive pressure. Q. Well, I guess we're getting close to argument here. You would agree that if you set a target higher than the historical experience, you will, other things being equal, induce the company to try and behave more like a competitive firm in this new environment than it Fuss,Seal 753 cr-ex (Brett) has in the COS regulated historical period? A. No, not necessarily. I don't know how it would behave as a competitive firm. Q. Isn't the theory of PBR to simulate competitive conditions? A. The theory of PBR is to induce the company to be more efficient. The competitive process is believed to be the best way to do so, but you have to have a competitive process to start with. Q. You don't have a competitive process, Dr. Fuss -- A. That's right. And so -- Q. But you have a series -- but you're trying to get these still monopolies in the sense of at least their distribution service to behave as if there were a competitive process? A. And what I'm trying to explain to you is that the numbers that are generated by the statistical industry are not the result of competitive pressures; they are the result of a series of local monopolies just doing whatever they do. And, hence, looking for that number as the way to simulate, competition is not logical. Q. All right. You're saying you might want a number that's higher than, for example, the historical record for the industry? A. Or lower. I don't know the circumstances. You have to look at the circumstances. Q. But one might assume... One would be Fuss,Seal 754 cr-ex (Brett) inclined to assume that the productivity experience under cost of service regulation historically would be less than one might achieve under PBR and that's what are we all doing here? A. I have agreed with that at least twice. Q. You have agreed to it and it's a question of magnitude, I guess? A. That's right. Q. How much more or less? A. That's right. Q. It would lead to a slim advantage? A. Right. And you would like a large one, so that's fine. Q. All right. Well, then on that note we can argue. But I want to ask you to just, on one further point, if you turn down that same page, the bottom of 14, the last paragraph where Dr. Fessler says: 'We note that improved efficiency can arise from three sources: Adopting more efficient technology and meeting current demand, realizing economies of scale when expanding the operation or reducing existing inefficiencies in the current operation.' Do you see that? Does that look like a reasonable summary of how one can improve efficiency? A. Yes. Q. It goes on to say: 'In the non-generation business...' Fuss,Seal 755 cr-ex (Brett) And he's speaking here of distribution: '...the first source of productivity may contribute only selectively toward greater efficiency and lower rates. The incentive of this PBR should discover the opportunities to increase the efficiency of the current operation and thereby lower rates.' Now, I want to take you over the page to page 15 and the second paragraph, and I want to have you look at the last sentence, the last two sentences in the paragraph, in the second paragraph, in the first paragraph, in the third -- first sentence in the third paragraph. Let me just read those, starting with the last two sentences in paragraph 2. It says: 'The utility' -- this is talking about PBR now: 'The utility can increase short-run profits through reducing variable costs but without revenue sharing such cost reductions will not lower rates. Moreover, such reductions not only can affect staff immediately but the service quality impact may only appear much later. In this PBR for Edison we balance these interests by requiring a progressive sharing of net revenue between shareholders and ratepayers...' Leave that aside for a moment because there's no sharing in this case. '...and by having both the productivity and service quality measures increase over the Fuss,Seal 756 cr-ex (Brett) duration of the PBR.' And in the last sentence in that same paragraph, if you will: 'Allowing time to discover such opportunities for cost reduction could defer some of the rate reductions for ratepayers but we believe that when such reductions occur they will represent more permanent efficiency gains.' So what he's saying there as I understand it is there should be a progressively -- you should raise the bar as you go forward, that the productivity offset in here -- and this is what they've done here, the California Edison case is - I don't have the numbers at my fingertip but they're in the evidence - it's a five-year program and the numbers increase over the five years. My question really is: Do you agree with the proposition that that's a sensible way to proceed to gradually increase the productivity offset over a multi-year term? A. I think if you had a plan as long as five years you might entertain such a procedure. I think for a three-year plan with rebasing at the end of the third year that's overly complicated. Q. What's so complicated about it, Dr. Fuss? A. Well, there will be quite a lot of debate over what the numbers should be, what the progression should be, when should you start the progression. Q. So it's really some choices that have to be Fuss,Seal 757 cr-ex (Brett) made about the selection of the target for each year? A. That's correct. Q. Beyond that there are no systemic problems that arise and the company presumably can work to those increasing targets. A. Well, in that sense it's not much different than picking a target, I agree. MR. BRETT: Thanks very much. Those are my questions, Madam Chair. Thank you. THE PRESIDING MEMBER: Thank you, Mr. Brett. We'll return at two o'clock and have Board Staff questions and Board questions after that. Mr. Warren, I see you have a... MR. WARREN: I'm just wondering whether or not, Madam Chair, we're likely to get to the service quality people. I'm beginning to feel a bit like Mr. Klippenstein's cardboard cutout. THE PRESIDING MEMBER: I think I said just as we came back - perhaps you weren't here - that we won't go on this afternoon. MR. WARREN: I'm sorry, I missed that. Then if I might be excused and I'll return tomorrow morning. MR. PRATTE: Madam Chair, I just wanted to say that I'd like to be excused for the balance of the hearing. Mr. Thompson will reappear tomorrow morning so I'm grateful for the Panel's and the company's attention. Thank you. THE PRESIDING MEMBER: Thank you, Mr. Pratte and Fuss,Seal 758 cr-ex (Lea) thank you for your assistance. Mr. Cass? MR. CASS: Yes, Madam Chair, just for clarity, Dr. Fuss will return tomorrow morning and the first panel will reappear after lunch. THE PRESIDING MEMBER: Yes, I think that makes sense although, forgive me, I wonder if we might take Ms. Lea's one question and see if the Panel has questions. Perhaps we should just try another few minutes because it would be inconvenient to have to bring Dr. Fuss back. DR. FUSS: I appreciate that. MS. LEA: Thank you very much. Are there any other counsel who had questions for Dr. Fuss? MS. DeMARCO: No. I wonder, Madam Chair, if I might be excused until tomorrow morning's Service Quality panel as well? THE PRESIDING MEMBER: Yes. CROSS-EXAMINATION BY MS. LEA: Q. I have to confess I now have two questions but I trust they'll be brief. Dr. Fuss, during examination with other counsel you've been talking about the sorts of measures that you select as factors. And one of the things that seemed to be important was independence from manipulation by the company. In other words, a publicly available index that was not subject to company manipulation. Is that a factor that you considered in your Fuss,Seal 759 cr-ex (Lea) choice of indicators? DR. FUSS: A. Yes. Q. Can you tell me then why it is appropriate to use the company's own historic productivity numbers which have clearly been affected by company action in the past? A. It's not really -- that's not really what one means by manipulation. What one means by manipulation is the ability to manipulate the data by actions with the PBR rewards clearly in mind. When this historical data were, in fact, produced, there was no PBR proposal and so that data remains free of manipulation as I've defined it. Q. All right. In that context, thank you. I think earlier in questions by counsel you talked about the input price differential being zero between CPI and -- what was the other, productivity? A. Between the CPI and the actual input price inflation of the -- Q. Right. Thank you. Can you perhaps by way of undertaking bring to the Board any evidence to support the conclusion that this input price differential has been zero in the past and is likely to be zero in the future? A. With regard to the past, the evidence I think is probably here among the testimony. It's the evidence that the company's own input price has grown at 3.1 per cent per year and the CPI is 3.1 per cent per year growth. So with regard to the past I believe that somewhere in this evidence appears that material. Fuss,Seal 760 cr-ex (Lea) Q. That's an average number, is it not, sir? A. As it appears in the formula, just as productivity is an average figure and inflation -- not inflation, I'm sorry -- productivity is an average figure, so is this input price differential. It's not meant to change from year to year. With regard to the future, no, nobody can guarantee that it will not change in the future. Q. All right. So in coming to that conclusion then you've relied on the company's evidence to which you've just referred? A. That is correct. MS. LEA: Thank you. Those are my questions. THE PRESIDING MEMBER: Thank you, Ms. Lea. The Board has one question. MR. VLAHOS: Dr. Fuss, just one question in this line of examination by Ms. Lea. EXAMINATION BY MR. VLAHOS: Q. Now, you said numerous times that it doesn't matter whether you use a specific material price index or the Ontario CPI, for that matter. I'm sorry, you didn't extend that to the Canadian CPI, just the Ontario one. I must confess that I'm not totally clear as to how exactly the formula works so that you're indifferent. And if you tell me it's all in the evidence I will go back and I will read that and I will take my calculator and... DR. FUSS: A. That is not in evidence. That's why I said that that would be a big surprise today. Fuss,Seal 761 ex (Vlahos) I could as an undertaking - although I hesitate because it is math - demonstrate that fact, if you wish. Let me just explain that we're talking here only about deflating the materials' price by some deflator. We're not talking about the inflation rate that's in the formula. What I'm referring to is the fact that the productivity -- the formula would normally include an additional term which was not brought forward by Consumers Gas because in their submission that additional term is zero. It does appear in Dr. Norsworthy's evidence, the additional term in his math, again, unfortunately. And what it means is that when you change the price index for deflating the material, you change the measure of productivity growth because you're using a different price to deflate current dollars. But you have to take account of that in this input price differential. And when you take account of it, they cancel out. So it doesn't make any difference. Q. Right. So if you were to use a different index other than the Ontario CPI to deflate - let's call it index Y, okay - then by definition if you use an index to deflate something then you can -- at the end of the day you're going to have the same index that will give you the same predictive value? A. Yes. What would have happened if, for example, suppose Consumers Gas had used the other index that's proposed here in evidence, Dr. Norsworthy's numbers Fuss,Seal 762 ex (Vlahos) taken from Statistics Canada. It would have gotten a different productivity growth rate as its ten-year historical experience. But then it would have looked and said, well, we now have an input price differential. We better bring forward to the Board another term which includes this input price differential. What would happen is that the sum of the two would have been the .63. Q. So if there was another index that was a lot higher and parties may be attracted to that and say, well, why don't we use this index because that is going to generate a larger O&M base... let me finish. But then you're saying that in order to -- but that's only half of the equation. You have to go back and you have to adjust the productivity growth factor, so within the same formula you may have a larger inflation index but you're going to have a different productivity index? A. That's right, yes. Q. Is that what it is? A. Basically the differences will cancel out. Q. Okay. Thank you. A. It's not quite the same in the sense that it doesn't affect the inflation rate that changes from year to year. What it does reflect is the single number that's brought forward for the Board to consider which we've called the productivity factor. It would have been called the productivity plus input differential. And it would have been the same number .63. 763 MR. VLAHOS: Okay, thank you, Doctor. DR. HIGGIN: Yes, I have a question regarding the methodology and assumptions that would underlie Exhibit C, section 4, page 11. That's the calculation of productivity growth. I'm hoping that Mr. Seal would be able to answer the question without the assistance of Dr. Fuss and, therefore, that he could answer it on the next panel. That's the point. Just about what you did, what the assumptions were and the methodology. So if that's the case, I would defer that question with that assumption that I could ask him with respect to the next panel and Dr. Fuss would not be needed. Is that reasonable? MR. CASS: Yes, I believe, Dr. Higgin, that Mr. Seal is indicating that he can answer that. DR. FUSS: Thank you. THE PRESIDING MEMBER: Mr. Cass, did you have any re-direct? MR. CASS: No, I do not, Madam Chair. Thank you. THE PRESIDING MEMBER: Thank you, Dr. Fuss. DR. FUSS: Thank you very much for having me here. ---Witness withdraws. THE PRESIDING MEMBER: We will return at two or very close to two and we will then resume with Panel 1. MS. LAWSON: Madam Chair, at the risk of keeping you ten more seconds, I just wanted also to request to be 764 excused for the remainder of the hearing and to advise you that Mr. Janigan will be appearing only behalf of OCAP for the remainder of this proceeding. Thank you. THE PRESIDING MEMBER: Thank you very much for your assistance. Until two then. ---Luncheon recess at 12:15 p.m. ---On resuming at 2:20 p.m. THE PRESIDING MEMBER: Please be seated. Sorry again to be late. I guess we are beginning with Ms. Lea, unless there are some preliminary matters. MS. LEA: Thank you. DARRYL SEAL, DAVE CHARLESON, DAVID J. deJONGH, JAMES C. GRANT, STEPHEN McGILL; Resumed CROSS-EXAMINATION BY MS. LEA (cont'd): Q. Mr. Seal, just to, I hope, finish our debate about the inflation factor and to make sure I understand what your evidence was, if in coming up with the productivity number you had used a different inflation sort of -- if you used a different inflation index like the GDPPI, for example, in coming up with your productivity number, would you have then derived the Ontario CPI, do you believe? Would it have matched the Ontario CPI as well? Would the correlation have been as good when you came to calculate your inflation inflator? MR. SEAL: A. I think you have got the crux of Seal,Charleson, 765 deJongh,Grant,McGill cr-ex (Lea) what I was saying. It might have been different. I can't say exactly what it would have been. But the point I was making is if the productivity number changed because that is the first thing that's calculated and I'd use that number, taking away the growth, take away the productivity from the O&M number to come up with my inflation, O&M inflation number, then match it against an index, yes, it likely would have been different so that Ontario CPI might not have been the most close measure. Q. Thank you. That's helpful. A couple of questions about "Z" factor and off-ramps. Can somebody, I think it was Mr. Grant probably who was dealing with this earlier. I want to explore slightly more the question of the materiality of the "Z" factors. You've set the materiality criterion at $100,000. I think I understand that this has been the historical factor used for affiliate transactions and so on. Do you believe that this is still an appropriate materiality level when we're talking about a $240-million O&M budget and we're talking about a PRB scheme here which is different, of course, than a cost of service scheme? My suggestion is it might be too low. MR. GRANT: A. I don't disagree with you that it's a low threshold. And typically in the past we have, for regulatory purposes and presenting information to the Board, we have used that as sort of the rounding point. Seal,Charleson, 766 deJongh,Grant,McGill cr-ex (Lea) So I guess it was in that context that we were suggesting that that could be a threshold on "Z" factors. It wouldn't be inappropriate, I don't think, to have a higher threshold than $100,000 for "Z" factor. Q. If the Board were to believe that the threshold was too low and that it would lead to unnecessary applications and whatnot, what threshold could you be comfortable with or would you like to think about that and give us a reply by undertaking? A. I would say that it would be fair to set a "Z" factor threshold that was between 500,000 and $1-million. Q. Thank you. The increasing -- the materiality level of the "Z" factor, what does that do with respect to risk, if we did increase the materiality level? A. Well, the figures that I just gave you, that -- I don't think that it would materially alter risk. Amounts that were above those figures, I think we'd really have to start thinking about whether it does affect the risk. Q. Thank you. With respect to the off-ramps, the second off-ramp, CPI greater than six per cent or less than zero per cent, can someone indicate how that range was selected? MR. SEAL: A. What it reflects, it's somewhat arbitrary, but it does reflect what we believe is a reasonable range for CPI, both historically over the historical period that we've estimated the various Seal,Charleson, 767 deJongh,Grant,McGill cr-ex (Lea) components of the plan and a reasonable range over the forecast period. We don't expect it -- or I don't expect that it will be breached over the expected PRB period. It's a reasonable CPI range. Q. No. 3 on the list which was described as extreme volatility in financial markets, I would suggest that's a little bit vague. Can you tell me what constitutes extreme volatility in financial markets? MR. GRANT: A. I guess what we mean by this is that extreme volatility in the financial markets that is having an impact on your ability to raise funds at reasonable rates. So, in other words, it is not just a generic statement of extreme volatility in financial markets. It has to have some impact on the utility before it becomes an off-ramp. I haven't got any particular thoughts in mind as to what the threshold would be in this area, but... Q. Okay. So if you felt that you had a situation, if that situation existed you would come to the Board and ask the Board to determine whether this volatility, external volatility, was affecting the company to a degree that warranted a cessation of the PBR plan? A. Yes, that's fair. Q. In No. 5, proposed off-ramps, which is impending our actual downgrading credit rating, would it be possible to change the word "downgrade" to "change" in Seal,Charleson, 768 deJongh,Grant,McGill cr-ex (Lea) credit rating so that it reflects both upside and downside risks? I don't remember what your credit rating is. I am sure it's very good, but how would you feel about changing the word "downgrade" to "change"? A. This particular item -- what we're trying to capture here is a situation where, once again, we would have trouble raising funds in the market. If for some reason the credit rating went up, then, of course, that would be in -- our targeted PRB plan would be reflected in ongoing rates cases. In other words, we would reflect the lower cost of raising those funds in a rates case. I suppose the same thing could be said about it going the other way, though, too, in the sense that if there was a downgrade and it resulted in higher cost of funds, that would be dealt with by the Board in the normal course of business; that is to say by way of a rates case. Q. All that I was interested in, Mr. Grant, was that if you had an improvement that the cost savings related -- resulting from that improvement would be shared with ratepayers. So that it's a-- A. That's right. Q. --mutual share? A. I think that's my point, that it is. Q. So your evidence is that because this will be dealt with in rates cases, then we don't need to change that word because the results of a change will be passed Seal,Charleson, 769 deJongh,Grant,McGill cr-ex (Lea) through in that fashion? A. Yes. Can I have a moment, please. Q. Yes, please. ---Off the record discussion MS. LEA: Thank you. MR. GRANT: I think in dealing with the implications of a change in credit rating as it relates to the cost of the funds that are raised by the company, that's going to be dealt with in the normal course of events. In other words, in the normal rates case. So in that regard you don't really, from the point of view of passing on to ratepayers any change in the credit rating, you don't need to have an off-ramp in our PBR mechanism. Having said that, though, if for whatever reason there was a downgrade in credit rating simply because of the PBR plan, that's something that we would want to bring forward as an issue. MS. LEA: Yes. I can see that being different. Q. Could I just clarify then with you, of these six off-ramps an occurrence of the first two really requires a change in the formula, whereas occurrence of the last four requires a reconsideration of the wisdom of the PBR plan et al. A. Yes, I think that's fair. Q. Thank you. One moment. ---Off the record discussion MS. LEA: Q. I'm not sure that I -- perhaps I Seal,Charleson, 770 deJongh,Grant,McGill cr-ex (Lea) need some assistance with this. For Nos. 3, 5 and 6, would you need to return to some kind of cost of service regulation if those occurred, or would it be a comprehensive PBR or can you tell at this juncture? MR. GRANT: A. I must say for the items that -- for the ones that you have raised here it's most applicable in a comprehensive PBR environment as opposed to our environment. Q. That's, I guess, a question I asked you or many people have asked you already. Why are these off-ramps in a targeted PBR plan when they do apply to comprehensive PBR, but I don't want to rehash that. A. No, I understand. I'm just trying to find -- I'm just trying to make sure that we have -- if these are issues for us as a company, that they are dealt with in an expeditious way and in a reasonable way. So that if we could be assured that these would be dealt with in the normal course of events; that is in the rates case, then they don't need to be off-ramps for this targeted PBR. I think they should be off-ramps for a comprehensive one, but really our objective is to ensure that if these situations ever arise that they are dealt with appropriately. Q. All right. So for Nos. 3, 5 and 6, you're saying they don't need to be off-ramps for this PBR proposal as long as they are dealt with in some fashion? A. That's right. As long as we are given some assurance that they will be dealt with. Seal,Charleson, 771 deJongh,Grant,McGill cr-ex (Lea) Q. Can you think of a way, Mr. Grant -- we are kind of trying to find our way as to how the Board deals with all this and it is new to us. For Nos. 3, 5 and 6, in what way would you propose that they be dealt with outside of moving to cost of service, outside of moving to comprehensive PBR? What would we do? A. I think for item No. 6, it would be dealt with by way of an opening up of the ROE formula, as an example. That could be one way that the Board deals with it. For items -- that may, in fact, be the case for item 5 as well. It may necessitate a reopening of that particular mechanism. The other way that we could deal with it is simply by way of explaining it in our rates case. If any of these things were having an impact on debt costs, we would bring that forward, explain it and have the Board deal with it in the rates case. Q. Okay. Thank you. THE PRESIDING MEMBER: Mr. Grant, can I just clarify, essentially none of these items 3 to 6 -- or at least 3, 5 and 6 really relate to O&M? Isn't that the crux of the matter? MR. GRANT: Yes, it is, Madam Chair. MR. VLAHOS: Although O&M may impact the interest coverage ratios? MR. GRANT: Yes. O&M is one of the elements of Seal,Charleson, 772 deJongh,Grant,McGill cr-ex (Lea) that calculation, that's correct. But there may be other, more significant things that affect those issues, and I suppose that's why we're a little uncertain here today because, on the one hand, you sort of want to have them in as off-ramps in any PBR approach that you take, but on the other hand we recognize that there is a lot of other things outside of O&M that affect these issues. THE PRESIDING MEMBER: Thank you. MS. LEA: Q. Mr. Grant, several cross-examiners have asked you about the lack of a sharing mechanism in this proposal. I understood from some of the evidence from the company that you developed off-ramps and "Z" factors in part by surveying off-ramps and "Z" factors in other jurisdictions, in other PBR plans. Can you tell me two things. First, in other jurisdictions where sharing mechanism's adopted or what was a percentage or proportion of those plans that had sharing mechanisms; and, secondly, can you summarize why you believe a sharing mechanism is not appropriate for this particular plan? Whoever, please. MR. deJONGH: A. I think the short answer to the first question is that most of the examples that we have included in our table and, in fact, many of the comprehensive PBR plans that are operating today do have some form of sharing mechanism. So in terms of percentage I'd say it is on the high side. Seal,Charleson, 773 deJongh,Grant,McGill cr-ex (Lea) Q. So you would see yourself as an exception to that general rule, then? A. In terms of our limited targeted PBR proposal, yes, we would be an exception. Q. Why do you believe you should be an exception? MR. GRANT: A. Mainly because it is most appropriate. The sharing mechanisms are most appropriate in comprehensive PBR environments, not a targeted PBR environment. Having said that, I still believe that there is a sharing mechanism, if you will, in our proposal. If you look at it from beginning straight through to end there is a sharing between ratepayers and shareholders of the productivity savings. Q. Thank you. I wonder if we could look for just a moment, please, at Exhibit C, tab 4, page 11 which is the productivity escalator. I just wanted a clarification on the calculation of the numbers which we have perhaps been viewing too simplistically. Can someone tell me, under the column Growth of Output, we see Percentage Changes there. Can somebody tell me how that number is derived and whether it then flows into the output index in the following year? MR. SEAL: A. I'll try without being mathematical. The growth number with the output index is Seal,Charleson, 774 deJongh,Grant,McGill cr-ex (Lea) calculated as the 1987 index number over the 1986 index number. Actually, it's the log of that. Q. Okay. Well, that, I think, gives most of the answer-- A. Answers the question? Okay. Q. --because we were just using simple. We hadn't thought of using the log function. A. The common way of doing it in productivity analysis like this is using logs. Q. Okay. And then it's factored into the next year's output index? A. No. The indexes are all calculated. The growth rates simply fall out of those index numbers. Q. Thank you. That's helpful. Right. Just one moment, please. I wonder if someone could help me with what this proposal means in terms of the timing of productivity improvements. Under cost of service regulation with a yearly review, the company has undertaken certain productivity improvements. But would it be true to say that the fact that you have a one-year timeframe has acted, in a sense, as a damper on productivity improvements because those sorts of improvements often require a longer timeframe to plan, execute and to realize or reap the benefits of them? Can someone comment on that? MR. GRANT: A. Yes, I think I can. I think that does act as a damper, the one-year Seal,Charleson, 775 deJongh,Grant,McGill cr-ex (Lea) period, successive one-year periods in cost of service regulation, because what it does is limit people's consideration of what is possible within, say, a three-year time frame. For instance, it may be good business to spend dollars, spend additional operating and maintenance expense dollars in Year 1, let's say, to, in for example in the IS area that would affect other areas in Years 2 and 3 and allow you to save dollars elsewhere. And if you're in a cost of service environment, it makes it more problematic in pursuing those things. It really allows for freer thinking in the timeframe and we're hoping that it will invoke a change in the culture because of that and provide a real incentive to our employees to go out and think broadly and think in a little longer term horizon when it comes to productivity. Q. If this proposed plan then will remove some of the dampers on productivity improvements, should we not be increasing the productivity measure or targets above historical levels? A. Well, I think to do that really presupposes that a certain level of productivity is going to be driven out and we don't know whether that's the case as yet. What we need is the incentive to go out and do that and keep those savings in the three-year period. So I would think that anything that is above the historic level of productivity which has been delivered does start Seal,Charleson, 776 deJongh,Grant,McGill cr-ex (Lea) to change the mix of sharing, if you will, between ratepayers and shareholders in the PBR period, and probably dampens to a certain extent anyway the incentive, the financial incentive, the pure financial incentive. MS. LEA: Thank you. Those are my questions. Thank you, Madam Chair. THE PRESIDING MEMBER: Thank you, Ms. Lea. DR. HIGGIN: Thank you. EXAMINATION BY DR. HIGGIN: Q. I would like to start, Mr. Seal, by asking the question I would have asked you both you and Dr. Fuss, had there been time. If you could turn up again Table 4.3 which is on Exhibit C, section 4, page 11 of 12, and also turn up Exhibit C, Appendix B, page 8 of 8 which is in Dr. Fuss's evidence and gives the equation I think, first question, that underlies the calculation; that is, the equation shown as A6. Is that correct, first of all? Am I making the correct connection here? MR. SEAL: A. I believe that is correct. Q. Thank you. The question is quite straightforward, and that is in calculating the historic data, did you attempt to get to "a pure distribution utility" or did you include the non-utility and ancillary program activities within the inputs on labour and materials? A. We did, indeed, attempt to remove all the Seal,Charleson, 777 deJongh,Grant,McGill ex (Higgin) non-utility components. Q. What about the ancillary program? A. And the ancillary programs, as well, to the best of our estimates. Q. Because otherwise, I think you would understand that the shares of labour and materials might change; specifically, 500 people moving out might have an impact post-separation? A. Post-separation. Q. I'm trying to determine whether there may be a disconnect between the shares of labour and materials due to unbundling. Now, if you had factored them out in your calculations, then there may not be such a disconnect in the share of the SL and the SMs? A. Right. And we did try to factor those out, as I said, to the best of the estimates that we could get for those. Q. You see what I'm getting at? A. Yes, I do. And I don't believe -- Q. That would change the future historic -- there would be a structural change in the formula which underlies productivity which would affect the future value; correct? A. I believe you're correct, yes. Q. But you think you have taken account of that and therefore the future will be like history in this case? Seal,Charleson, 778 deJongh,Grant,McGill ex (Higgin) A. I believe we have taken account of that, Dr. Higgin, to try and take those out of that to eliminate that effect, that possible effect. Q. But you haven't actually projected yet the productivity based on your latest estimates for 1998, 1999 and 2000 given the unbundling or have you done that to see how it compares with the historic levels? A. I haven't done that. The only 1998 and 1999 figures that we've done are the ones that are included in section... Q. Yes. I think you pointed us to that exhibit. A. Where it shows the bridge and the test year numbers. Q. Yes. And what did they -- do you remember just off, what they showed? A. Sorry. The productivity estimates that come out of that are the 10.89 and the -- Q. And the 1.5 which you could say were due to extraordinary circumstances. The problem that I'm having particularly is even if you factored out the ancillary program, it was probably on a marginal cost basis which would have a difference of 30-million in the number between the marginal and the fully allocated cost number that you would take out for O&M, for example. Is that right? A. I guess I'm not sure on what basis the estimates for those particular programs that we took out. I could undertake to find out, though. Seal,Charleson, 779 deJongh,Grant,McGill ex (Higgin) Q. Yes. I think it would be helpful. In broad sense, we're not interested in the detail; we're just trying to get an understanding that there is no disconnect in the mix - I think this is the key - in the structural mix of share of labour and materials going into the O&M from the historic to the future, given the unbundling plan. That's the essence of the question. So the details in terms of quantitative, I'm less interested in, in getting some feel that you've looked at that and you have some reasonable feel that 0.63 average is still a good number after unbundling. That's the essence. MS. LEA: Is that an undertaking then? DR. HIGGIN: Yes, please. MS. LEA: Thank you. 5.1. ---UNDERTAKING NO. H5.1: Enbridge Consumers Gas undertakes to look at [productivity calculations] so that Enbridge Consumers Gas has some reasonable feel that 0.63 average is still a good number after unbundling and provide, in a broad sense, an understanding that there is no disconnect in the structural mix of share of labour and materials going into the O&M from the historic to the future, given the unbundling plan. DR. HIGGIN: Thank you, Mr. Seal. Could you just help me a bit and be very explicit with me about the proposed numbers that will be used in the formula for the Years 2000, 2001, 2002 for customer growth, productivity, offset and inflation based on the Ontario CPI? Seal,Charleson, 780 deJongh,Grant,McGill ex (Higgin) Just turning to the pure numbers now, what are you proposing to use? Do you have the numbers yet or are they still to be forecast? MR. SEAL: A. I will maybe approach them each individually. The productivity number that we propose to use in all three of those years is the 0.63 per cent, so that remains unchanged in all three years. Q. Okay. A. The customer growth number; we propose to use a forecast of customer growth as calculated, I believe, just prior to the giving of each fiscal year. Q. That will coincide with your revenue forecast? A. That's right. With our annual rates case. And then we will update that or true up, if you will, at the end of each year for the actual customer growth. Q. Right. A. Similarly, for the CPI or the inflation index, we propose to use a forecast of Ontario CPI produced by the four forecasting agencies, external forecasting agencies as produced by them as close to the beginning of the fiscal year as possible, as well, and then again true up for actual CPI at the end of the year, each year. Q. Now, the last numbers that you had, I believe, if I looked them up right, are customer growth at Seal,Charleson, 781 deJongh,Grant,McGill ex (Higgin) 3.9 per cent in the EBRO 497. Am I correct? That would be for 1999. A. The numbers shown in Exhibit C5.0, page 4 are growth customer growth of 3.79 in '98 and 3.86 in '99. Q. Eight six rounded up to... Yes. I think the Board's decision shows 3.9. We took the liberty of rounding. And as far as inflation then, what number did you use in 497 for 1999, that is, Mr. Seal? A. I believe in the ADR process -- Q. It says 1.9-- A. That's right. Q. --but that's not the Ontario... A. Well, that was the, I believe, agreed-upon Ontario CPI forecast for 1999, I believe. Q. That was the forecast? A. Yes. A forecast. Q. 1.9. I think it's a bit higher now? A. Sorry? Q. I think it's a little bit higher now. A. Our forecast? Q. No. The latest forecast for '99 is a bit higher? A. It might be. Q. It's about 2 or slightly more than 2. A. The 1.9 was an agreed upon number in the ADR. Q. Okay. Thank you. What this is leading to, I would like to get your Seal,Charleson, 782 deJongh,Grant,McGill ex (Higgin) assistance in doing a calculation that would allow me to compare what the '99 O&M base - emphasize the word "base" - budget would have been under three scenarios. One is the company's request. So can we turn to the 497 decision and look at page 34, please? We may have a spare here if you're missing one. Do you have one? MR. CHARLESON: A. We have one. Q. That's page 34, Mr. Seal. The whole section 3.2 discusses the O&M expense. So the first thing I'm trying to get, see if we have an agreement on the company's requested base. I emphasize the word "base" for 1999. I would interpret it as follows: That you would agree that within 0.1-million, the 1998 base, as said here, is 250-million. Is that -- do you agree -- it's actually 250.1. MR. SEAL: A. We'll take the 0.1. Q. Yes, all right. Now, to that then, you see in paragraph 3.3.2, if the Board has interpreted the company's request correctly in the decision, you would like to have added 2.9 per cent or 7.2-million and you've included customer growth, inflation, et cetera, and offset some productivity saving. So you would have liked to have added 7.2-million in that category, correct? A. Correct. Q. And then if you look at the next bullet point, you also wish to reinstate some service levels that Seal,Charleson, 783 deJongh,Grant,McGill ex (Higgin) had dropped to 2.9-million. So I think that would be legitimate to say that was part of the company's request for a total of 10.1-million. Now, the other two bullet points on that page, I would interpret those as falling within Category 6 of your "Z" factors. Can we at least maybe make that assumption that these are improving service levels and one-time costs? So they wouldn't be part of the base. They're still part of the request and they're not part of the base request? A. I can certainly see the Y2K numbers. I'm not sure about the CIS numbers. Q. Okay. I was saying improved service levels was the key which seems to fit with your Category 6. Anyway... Okay, so would you let me know which way you would say that was part of the base request or part of the incremental or "Z" factors? It's either 10.1 or 14.7 depending on which way that goes. That was your request for the base? MR. GRANT: A. Yes, we're now focused on that third bullet then, the 3.6-million? Q. Yes, I assumed it was going to be off the table for base purposes? A. That's correct; it is. Q. It is. A. Yes. Q. So Mr. Grant has just confirmed that for base Seal,Charleson, 784 deJongh,Grant,McGill ex (Higgin) purposes it's off the table. So it's 10.1-million was your request. Now, if you could look at the Board's decision at paragraph 3.2.9 and the Board - this is under cost of service - the Board's decision quotes your 3.9 per cent increase. We had this question of how much was directly -- anyway 4.9-million for customer growth and then add to that 1.9 per cent for the inflation factor and then the Board used its own productivity offset of 1-million in paragraph 3.2.10, correct? A. Yes, that's correct. Q. So we could calculate that -- I think it's relatively straightforward. I think it's about 8.75-million the Board awarded. So what I'd like you to do now is take the -- so those are the derivative of the numbers. Could you put all these onto an undertaking which the third column is application of the PBR formula; if you took 1998 250 and then you applied your formula using the numbers you just gave me of 3.86, 0.63 productivity offset and inflation of 1.9 per cent, what would have been the company's base 1999? This will somehow allow me to compare the cost of service approach versus - that's my own interpretation - versus what the formula will -- admittedly only for the one year. We will accept it's just for that one year. But unfortunately I don't have any other way to do it over a longer term period. These numbers just Seal,Charleson, 785 deJongh,Grant,McGill ex (Higgin) happen to be available. So if you would do that by way of undertaking then. On one undertaking show what the 1999 O&M would have been, the company's request, the Board-approved base and the PBR formula base -- again emphasize base for those. Thank you. MS. LEA: H5.2, please. ---UNDERTAKING NO. H5.2: Enbridge Consumers Gas undertakes to compare the cost of service approach versus the formula for one year, showing what the 1999 O&M would have been, the company's request, the Board-approved base and the PBR formula base. DR. HIGGIN: Thank you. That's all my questions. Thank you, Madam Chair. THE PRESIDING MEMBER: Thank you, Dr. Higgin. Mr. Vlahos? MR. VLAHOS: Thank you, Madam Chair. EXAMINATION BY MR. VLAHOS: Q. Panel, I only have two questions. Mr. Grant, I believe you can answer both of them. You talked about the company's options in the event that your proposal is rejected. There was a lot of discussion about that in the last few days. There hasn't been any discussion as to what would happen in the event the Board were to approve your proposal with some modifications to the formula. Can you assist me with that at all? MR. GRANT: A. We haven't had any detailed Seal,Charleson, 786 deJongh,Grant,McGill ex (Vlahos) discussions on that point with the exception of in preparation for the ADR negotiations we did talk about that issue at a high level of generality but it wasn't any detailed discussion at that point. Q. Can you help me more? A. If the Board felt that it was necessary to modify the formula, in other words, the Board approved the PBR, approved the period that we were going to operate it in, yet had some modifications to it to correct what the Board felt were either weaknesses or to improve upon the formula in some way, I think if those changes were made to the formula in the Board's decision, we would certainly live with it and carry on and drive out as much productivity as we could. We're committed to PBR and we're committed to moving forward in a logical step fashion that I discussed the other day. And that's really what this proposal is all about that's before the Board. Q. On that same issue is the nature of the application such that the Board will/can order you to implement a targeted PBR albeit with a modified formula or is the company proposing that here is a PBR proposal and if we like the decision, we'll implement it; otherwise we'll consider cost of service for the time being? A. No, it's the former, Mr. Vlahos. We're committed. Q. Okay. Thank you for that. One last question, would you agree with me that Seal,Charleson, 787 deJongh,Grant,McGill ex (Vlahos) from a regulatory perspective if the Board were to consider a sharing mechanism - and we heard about that sharing in the last few days also - versus the imputation or the application of a stretch factor that from a regulatory perspective, the stretch factor would be the easiest route? A. From a regulatory perspective I think that it would because it would eliminate debates that may arise out of the actual sharing that may happen within the PBR period. Q. First of all, you have to quantify the sharing, right? A. Indeed. Q. So that would have to take a cost of service review? A. Some sort of review and it would be fairly detailed I would think. So again from a regulatory perspective the stretch factor would make more sense. MR. VLAHOS: Okay. Thank you, Mr. Grant. EXAMINATION BY THE PRESIDING MEMBER: Q. I just have a couple of questions, Mr. Grant, on the inflation factor. I think a couple of times I heard you say over the past couple of days and I've checked the transcripts and haven't been able to just land on the right phrase but I've heard you say that the CPI is underestimating and I gathered when I listened to you say that that it was because of the labour inputs being a Seal,Charleson, 788 deJongh,Grant,McGill ex (Presiding Member) somewhat inflating or somewhat higher rate than the Ontario CPI would go; is that correct? MR. GRANT: A. Yes, I was making reference when I said that to an interrogatory response. Bear with me a second. I'm told it's Exhibit 10.29. Q. Can you just hold it for a second while I reshuffle. Yes, thanks. A. That would be page 2 of that Exhibit. Q. Yes. A. My eyes went to the bottom of that page where the Ontario CPI is compared against Ontario average weekly earnings. And that's why I drew the conclusions that I did. I also know that from past experience we've discussed in previous rates cases the fact that when we do go ahead and implement wage increases for our employees typically by the time you figure out the overall increase it comes in roughly a half per cent higher than the CPI or -- I'm sorry, roughly half a per cent higher than the base wage increase which tends to track the CPI. Q. Okay. If I go to your table in section 4, Exhibit C, in which you set out the various CPIs and GDP deflator, et cetera, in your company O&M inflation column, that would include your labour inflation, right? A. I'm sorry, I'm not at the right page. What's the reference? Q. Section 4, page 6 of 12. Seal,Charleson, 789 deJongh,Grant,McGill ex (Presiding Member) A. Yes, I believe it does. Q. If I run down a comparison between company O&M inflation and Ontario CPI and do the calculations year by year rather than doing the average as you've done it, a great deal will depend on how many years I go but, for example, if I did the first three years I would come up with a fair difference between the result of using company figures and CPI. And the difference would certainly not be in the direction you're suggesting, Mr. Grant. It would be the other way by I think perhaps as much as about $10-million because of the large differences between Ontario CPI and the company O&M inflation the first four years there say '87 to '91; would you agree with that? It certainly would be directionally different than what you're arguing for? A. I agree that those figures are directionally different for those particular years. Q. And even if I go all the way to the bottom doing that calculation one year at a time which is my simpleminded way of doing it, instead of averaging the numbers but just applying the two numbers each year and then applying it again to the number that that results from, I end up with a higher number if I use the company's or -- sorry, if I use the CPI than if I use the company's number; would you agree with that? This is my compounding interest one year at a time approach to CPI. Do you understand the calculation Seal,Charleson, 790 deJongh,Grant,McGill ex (Presiding Member) that I've done, Mr. Seal? MR. SEAL: A. If I interpret correctly I believe you're taking a base year dollar amount and inflating it by the Ontario CPI. Q. Each year. A. Right and also inflating by the company O&M inflation if you will. Q. Right. I'm doing two columns, one the company's, one the CPI. A. And seeing where you end up. Q. I start with $100 and I see what I have at the end. And directionally by the time I'm finished that year by year calculation, I'm still out ahead in a direction opposite to what Mr. Grant has been suggesting. I can't quite reconcile that because I understand the company's inflation to be including labour. So have I conceptually got the wrong idea? A. I think maybe two points. One is that it could depend on your starting point. You could do that calculation starting each year if that brings out the same. So that might be one of the issues. If you look at the average over those periods they're quite close, the average growth rate. So that would suggest that on average if you did it over that whole period you should end up very similar. Q. They're not far wrong. A. Right. Q. My understanding is that CPI is somehow Seal,Charleson, 791 deJongh,Grant,McGill ex (Presiding Member) considered to be not quite as good a number as one that would include the labour, and that if I do those calculations I'm not sure that's the way it works out. I wanted to confirm that your labour was already included in your company O&M, but if that's the case I'm satisfied? A. The labour is in the company O&M as well as the materials. All O&M costs are in that O&M number, yes. MR. GRANT: A. Pure labour probably makes up -- I think our evidence is that labour plus benefits makes up about 60 per cent of that. So, obviously, there's other things that are acting being within the O&M, the total O&M -- Q. Right. And on a sort of up and down basis? A. Yes. Q. That is to say one year the company's inflation is quite a lot higher than CPI and another year it's quite a lot lower? A. That's right. Q. Although you come out with the same 3.1 per cent on the average, if you do your calculation year by year they are quite different. So, for example, over the first -- as I say, over the first four years they are quite different because the CPI there is substantially higher each year than the company's inflation? A. Yes. Yes, I agree with that. I agree that that is the arithmetic that comes out. It may -- Seal,Charleson, 792 deJongh,Grant,McGill ex (Presiding Member) Q. You still think that the CPI understates the -- I just can't quite reconcile what you said about the CPI when I do that calculation. A. I think it's fair to say that the CPI understates the increases in that 60 per cent category that I was speaking of, wages, salaries and benefits. That's a fair statement. As to whether it understates in total the O&M, I think that's point that you're getting at-- Q. And it doesn't. A. --and it doesn't. Q. At least it doesn't appear to. A. It doesn't appear to do that. The reason for that would be that there would be special offsetting factors happening in each of those years within the O&M offsetting the labour component, but I wouldn't want to lose the point that that 60 per cent on an ongoing basis has historically and would tend to in the future exceed inflation. Q. Well, I guess I couldn't agree that historically these numbers are right, but I just want to take your statement about the relationship between CPI and the labour as -- I want to understand a bit what we're to take from that, that ongoing into the future the CPI is going to be slightly less favourable to the company than some other index would be? Is that the statement and, if so, that doesn't appear to me to be the case from this table. Seal,Charleson, 793 deJongh,Grant,McGill ex (Presiding Member) A. I think perhaps the best thing for you to do is to relate my statements comparing CPI to wages only to the 60 per cent of the O&M that I discussed. So it's really relevant to that section of the O&M budget. Q. Okay. Thank you for that. Mr. Cass, do you have any questions? MR. CASS: I have one question and maybe two, Madam Chair. RE-EXAMINATION BY MR. CASS: Q. The first question arises out of Ms. Lea's examination during which, Mr. Grant, I think you discussed a possible threshold for "Z" factors and you suggested a range I believe of 500,000 to $1-million. Were you speaking per "Z" factor or cumulatively? MR. GRANT: A. I was speaking cumulatively because obviously we have a number of "Z" factors which would add up to a significant amount if it is taken individually for each "Z" factor. I was really trying to put a number around what we would consider to be material in the context of "Z" factors in general. And I was taking her point that $100,000 was a pretty low threshold, but I wanted to also bookend it, if you will, with a threshold that would not materially affect our risk profile. MS. LEA: I don't understand something now. Was the $100,000 intended to be cumulative, then? I guess I completely misunderstood your answer. MR. GRANT: Well, I think our $100,000 was Seal,Charleson, 794 deJongh,Grant,McGill re-ex (Cass) something that we were suggesting could be applied in any situation. We may not have all "Z" factors coming together in one year. It may be one "Z" factor. I think we were just -- I was really just throwing that out as sort of a practice that we had had in the past. THE PRESIDING MEMBER: Mr. Grant, can I ask, too. MR. GRANT: Certainly. THE PRESIDING MEMBER: I thought that $100,000 was sort of the threshold before a "Z" factor could kick in. I don't understand any kind of word like cumulative to relate to this. I might just not be getting the idea. MR. GRANT: Well, when I answered the question that was posed to me, I was not thinking that we would be in a situation where we would be a million dollars out for, say, four or five or six "Z" factors all in one year. That's a substantial amount of money. So I wasn't meaning to convey that we were going to be at risk for that amount of money. I think what my assumption was in answering the question is that we really wouldn't have many "Z" factors coming forward all at one time. We may only have one. For instance -- and the one "Z" factor that's really important to keep in mind -- well, a number of them are, but the one "Z" factor that's important has to do with certain cases that -- a case that is before the courts now. And I would think that in that instance we would Seal,Charleson, 795 deJongh,Grant,McGill re-ex (Cass) be wanting to have an opportunity to seek relief and that we would operate with a threshold that was relatively low. In other words, it wasn't any -- we would be prepared to accept the $500,000 threshold, but anything before that would be problematic. That's what I was trying to convey. THE PRESIDING MEMBER: Thank you. I think I understand what you are saying. Thanks. Any other questions, Mr. Cass? MR. CASS: Just one other area Madam Chair. I am going to try to wade into this area that you were just examining Mr. Grant on without leading him and see if I can succeed in bringing anymore clarity to it. MR. CASS: Q. What I might do, Mr. Grant, is take you back to the table 4-1 at Exhibit C, section 4 that you were looking at with the Chair. Page 6 of Section 4, Exhibit C. MR. GRANT: A. Yes, I have that. Q. I'd you ask to read the note at the bottom of the table and then to comment on whether you were addressing O&M on a total basis or on a per customer basis? A. Well, I understood the Chair's questions to be focused on the total O&M basis. So that's what my answers were based on. MR. CASS: That's all I can do, Madam Chair. That's my re-examination. I'm sorry. THE PRESIDING MEMBER: Thank you, Mr. Cass. 796 Mr. Cass can I -- sorry. Can I ask you what the status is of the various undertakings that we have outstanding. Are we expecting written responses or is there anything -- any update you can give us on those? Not the ones we just gave. MR. CASS: Madam Chair, to begin with, I think there have been difficulties because many of the undertakings involve Mr. Grant who has been on the stand for a long time. THE PRESIDING MEMBER: I understand. MR. CASS: He did have a break this morning and I do believe he was working very hard on undertaking responses. I believe that most of them, if not all, are in progress. The one that there is still a bit of an issue around was one given to Mr. Vlahos that concerned the -- I believe it is correct to call it the draft code, and as of the last time we checked that was still not available. Subject to that, I believe that they are all in progress and we'll get them as quickly as we can. THE PRESIDING MEMBER: That's fine. Thanks, Mr. Cass. So, as we said, we won't address another panel today. We will begin again at nine o'clock tomorrow. And we're trusting that we're not going to have a problem getting through a panel tomorrow and a panel on Thursday. Any other concerns or comments? ---(No response) 797 Thank you very much, panel, for your patience and your assistance. ---Whereupon, the proceedings were adjourned at 3:20 p.m., to be reconvened on Wednesday, the 3rd day of February, 1999, at 9:00 a.m. 798 I N D E X o f P R O C E E D I N G S Page No. Preliminary Matters . . . . . . . . . . . . 658-659 DARRYL SEAL, DAVE CHARLESON, DAVID J. deJONGH, JAMES C. GRANT, STEPHEN McGILL; Resumed. 659 Cont'd Cross-Examination by Ms. Lawson 659 Cross-Examination by Ms. Lea 671 MELVIN A. FUSS; Sworn, DARRYL SEAL; Resumed. 687 Direct Examination by Mr. Cass 688 Cross-Examination by Mr. Pratte 700 Cross-Examination by Ms. Lawson 723 Cross-Examination by Mr. Brett 736 Cross-Examinatino by Mr. Lea 758 Examination by Mr. Vlahos 760 ---[Luncheon Recess 12:15 p.m. - 2:09 p.m. ] 764 DARRYL SEAL, DAVE CHARLESON, DAVID J. deJONGH, JAMES C. GRANT, STEPHEN McGILL; Resumed. 764 Cross-Examination by Ms. Lea (cont'd) 764 Examination by Dr. Higgin 776 Examination by Mr. Vlahos 785 Examination by the Presiding Member 787 Re-examination by Mr. Cass 793 799 L I S T O F U N D E R T A K I N G S No. Description Page No. H5.1 Enbridge Consumers Gas undertakes to look at [productivity calculations] so that Enbridge Consumers Gas has some reasonable feel that 0.63 average is still a good number after unbundling and provide, in a broad sense, an understanding that there is no disconnect in the structural mix of share of labour and materials going into the O&M from the historic to the future, given the unbundling plan. 779 H5.2 Enbridge Consumers Gas undertakes to compare the cost of service approach versus the formula for one year, showing what the 1999 O&M would have been, the company's request, the Board-approved base and the PBR formula base. 785 MC LJ BV [ Copyright 1985]