AMERICAN STATISTICAL ASSOCIATION (ASA) MEETING OF THE AMERICAN STATISTICAL ASSOCIATION COMMITTEE ON ENERGY STATISTICS WITH THE ENERGY INFORMATION ADMINISTRATION (EIA) DAY ONE Washington, D.C. Thursday, October 5, 2006 2 1 P R O C E E D I N G S 2 (8:40 a.m.) 3 MR. HENGARTNER: Good morning, all. 4 It seems I need a gavel here to call us to 5 order. Of course, I don't have it. 6 Good morning. Welcome all. This 7 is a -- I'd like to open the meeting. And I 8 want to do that by reminding you that this is 9 an ASA and not an EIA committee, and it meets 10 periodically to provide input and guidance 11 and advice to EIA. 12 This meeting is open to the public. 13 I hope there is someone here. And public 14 comments, of course, are welcome. There's 15 time set aside for public comments at the end 16 of each day as well as the opportunity to 17 submit written comments either at the end of 18 the meeting or throughout the year. Your 19 comments can be sent either to the ASA or the 20 EIA. 21 All attendees, including guests and 22 EIA employees, should sign in at the register BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 3 1 in the hall and include your e-mail address. 2 Restrooms are at the end of the 3 hall. You go out that way or this other way, 4 and you know where they are. If not, I'll 5 show you. 6 There's a telephone in this room. 7 It is a shared number. Let me give you that 8 number in case you need to be reached. It's 9 (202) 586-6273. Again, (202) 586-6273. 10 Kathleen Wert and Donna Renee 11 Arrington are with the ASA committee. Are 12 they here? No, they're outside manning the 13 table. And if you have any questions about 14 expenses or travel reimbursements or so 15 forth, please address those questions to 16 them. They will be glad to help you. 17 In the past, we had someone writing 18 a transcript. We are still doing this, so I 19 please ask you to speak up, and if you talk 20 from the seats in the back, go to the 21 microphone. That will help us all to get the 22 transcript. There was an indication how to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 4 1 use a microphone. I will skip this, but I 2 will keep the opportunity to remind you that 3 if you don't do it right, I'll go and read 4 this explicitly. 5 Finally, I'd like to 6 announce -- before we start, I'd like all of 7 us to go around the table and introduce 8 ourselves as well as the members of the 9 audience, and I will start with myself. 10 Again, please use the microphone. 11 So, my name is Nick Hengartner and 12 I'm from Los Alamos and the chair of this 13 committee. 14 MS. KIRKENDALL: I'm Nancy 15 Kirkendall. I'm director of the Office of 16 Statistics and Methods Group, whatever office 17 I'm in. 18 MR. CARUSO: Guy Caruso, 19 administrator of the EIA. 20 MS. KHANNA: Neha Khanna, State 21 University of New York at Binghamton. 22 MR. BURTON: I'm Mark Burton, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 5 1 University of Tennessee. 2 MR. SINGPURWALLA: Darius 3 Singpurwalla, I'm at Klemm Analysis. I got a 4 new job. 5 MS. SEREIKA: Susan Sereika, 6 University of Pittsburgh. 7 MR. NEERCHAL: Nagaraj Neerchal, 8 UMBC. 9 MR. CLEVELAND: Cutler Cleveland, 10 Boston University. 11 MR. EDMONDS: I'm Jae Edmonds at 12 the Joint Global Change Research Institute 13 and Pacific Northwest National Laboratory. 14 MR. FEDER: Moshe Feder, RTI. 15 MS. FORSYTH: Barb Forsyth, WESTAT. 16 MR. HILL: Walter Hill, St. Mary's 17 College of Maryland. 18 MR. McDOWNEY: Preston McDowney, 19 Statistics and Methods Group. 20 MR. HENGARTNER: And the audience, 21 if you want to please walk to -- 22 MR. HELKIE: Bill Helkie, senior BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 6 1 advisor at EIA. 2 MS. HOJJATI: Behjat Hojjati, EIA. 3 MR. TSENG: Phillip Tseng, EIA. 4 MS. MILLER: Renee Miller, EIA. 5 MS. WAUGH: Shawna Waugh, 6 Statistics and Methods -- 7 MS. BLOMBERG: Carol Blumberg, 8 Office of Oil and Gas, EIA. 9 MS. WAUGH: Shawna Waugh, 10 Statistics and Methods Group, EIA. 11 MS. JENNINGS: Alethea Jennings, 12 EIA. 13 MR. BROON: Tom Broon, SMG at EIA. 14 MR. BRADSHER-FREDERICK: Howard 15 Bradsher-Frederick, EIA. 16 MS. SWEENEY: Amy Sweeney, EIA. 17 MR. STROUD: Lawrence Stroud, 18 Statistics and Methods Group, EIA. 19 MR. RASMUSSEN: Eric Rasmussen, 20 EIA. 21 MR. RUTCHIK: Bob Rutchik, EIA. 22 MR. WEINIG: Bill Weinig, EIA. I'd BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 7 1 like to mention that we have a constant 2 ringing in the room. We know about it. It's 3 not reparable and I'm sorry for the 4 inconvenience. It's with the air 5 conditioning system, so I think we're going 6 to have to suffer through it. I'm sorry, but 7 that's just reality. 8 MR. HENGARTNER: Okay. Thank you 9 very much, everybody. I'd like to inform you 10 that under the terms of the DOE grant to the 11 ASA, Nancy Kirkendall may chair, but must 12 attend each meeting. 13 And she's authorized to adjourn 14 this meeting if she determines this to be in 15 the best public interest. Must approve all 16 meetings of the advisory committee and every 17 agenda. She may also designate a substitute 18 in her absence. 19 Overall, I think we're going to 20 have a very good meeting again. This meeting 21 seems to be -- I mean, I looked at the 22 agenda. I was interested by seeing what's BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 8 1 going to be on the plate. There seems that 2 there's a lot of room for discussion this 3 time around, and I hope that we will all feel 4 free to comment and elaborate on the points 5 that are brought forward to us. So just to 6 let you know, I think it's going to be good. 7 The first session this morning is 8 going to be by the EIA administrator, Guy 9 Caruso, who is going to tell us about the 10 general affairs at the EIA. 11 Welcome, Guy. 12 MR. CARUSO: Thank you, Nick. Good 13 morning. Thanks again, Nick. I always start 14 my remarks by saying how much we really 15 appreciate the ASA committee's work. And I 16 think it's -- you know, my -- this is I guess 17 about the eighth meeting for me, and every 18 time, I feel as though that EIA is a 19 significant beneficiary of this committee's 20 work and the advice you've given us. 21 And as you point out, Nick, the 22 agenda for this meeting is no exception in BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 9 1 that it lays out a number of areas where we 2 can really seek the advice of this committee 3 and the membership and goes beyond just this 4 meeting. Clearly we're going to hear a lot 5 from your members in the sessions themselves, 6 but I also want to thank you and the members 7 for the continuing relationship we have with 8 you outside the formal meetings, and that we 9 continue to seek the advice of the committee 10 and the membership. So thanks very much. 11 Good news I think for bringing you 12 all up to date as to what's been happening at 13 EIA is that we're a lot less -- it's been a 14 much calmer six months this October than when 15 I stood here last October right after the 16 aftermath of Katrina and Rita, which so 17 dominated our own work and the Congress' and 18 the Administration's demands on EIA, and that 19 tended to overshadow a lot of the ongoing 20 core activities. 21 The last six months, we've I think 22 been able to get back to doing the things BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 10 1 that we need to do to follow up on many of 2 the suggestions that we've heard from you. 3 And you're going to hear 4 immediately after my remarks, and Nancy 5 brings you up to date with the activities, 6 where we are on probably the most exciting 7 new development in EIA on the forecasting 8 side in a couple decades. And that is we 9 have received the approval and support of DOE 10 management to do a complete redesign of the 11 National Energy Modeling System. 12 And of course, budgets being what 13 they are, this is highly dependent on our 14 getting the monies that we've asked for, but 15 we're optimistic. And I'll talk a little bit 16 about specifically on the budget, but I 17 mentioned this the last time, but we are six 18 months closer to actually implementing the 19 redesign. And Susan Holte will go into a lot 20 more detail about that. And I think it's 21 very much an activity which we can benefit 22 from, this committee and your membership, so BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 11 1 we're real excited about that. 2 We've been talking about the 3 reporting on greenhouse gases for a number of 4 meetings now, and the redesign of the 5 voluntary reporting survey is well underway 6 now and we're hoping to implement that in 7 '07. And as you know, it's been about a 8 two-year process, and again, benefiting from 9 your advice. 10 And then there is always the 11 ongoing energy markets and how they impact 12 what EIA does. And just a couple of 13 activities, a couple of events that highlight 14 the effect on our work, and that is 15 the -- we've been involved to some extent in 16 this whole debate about peak oil that's been 17 underway in the analytical community now for 18 several years. 19 And we at EIA, as I think you've 20 heard from John Wood and others of John 21 Wood's staff in Dallas, are of the view that 22 the resource base, whether it be oil or BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 12 1 natural gas, is sufficient to meet the 2 demands that our forecasting models indicate 3 will be there, whether it be domestically or 4 internationally. 5 And of course, we follow new 6 developments. And the most recent one on the 7 upstream side has been this very deepwater 8 discovery and the lower tertiary trend in the 9 Gulf of Mexico, the so-called Jack 2 well 10 operated by Chevron. And our models have had 11 the reserves from that lower tertiary trend 12 in our long-term outlook for some time now. 13 So in a way, it kind of supports our view, 14 which doesn't necessarily mean it's 15 ultimately where it's going to come out. 16 And then going to the other extreme 17 of the spectrum on oil is the downstream. 18 Not only have we had this transition from 19 MTBE to ethanol, which is significant for EIA 20 because of the data quality issues with 21 respect to collecting and understanding 22 what's going on with blending of ethanol, and BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 13 1 that we certainly did not have in our system 2 when EIA was founded -- so that's been an 3 issue; we'll talk about that -- but other 4 things like the lower sulfur gasoline, the 5 ultra-low-sulfur diesel, and then the 6 increased emphasis on flexible-fuel vehicles, 7 such as biodiesel, all have important 8 implications of the work of EIA and 9 particularly the data quality on, in this 10 case, oil. 11 I'll also mention where we are on 12 the FY 2007 budget, which is, unfortunately, 13 Congress went out without passing an '07 14 budget for energy and water, where our budget 15 resides. So we have to live with the 16 continuing resolution at least through 17 mid-November, but the smart money in 18 Congress -- insiders are saying it won't be 19 until the new Congress in January before they 20 act on an '07 budget, which is difficult for 21 an organization that has as its core function 22 collecting data, which uses about 50 percent BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 14 1 of the budget. And we can't stop doing that 2 or even diminish it, so it means other things 3 really get squeezed. 4 We talked about the external study 5 team that was headed by Denny Ellerman. We 6 have, since our last meeting, gotten the 7 formal report, and those reports are I think 8 available if you haven't gotten one. And 9 we'll talk a little bit more about the 10 recommendations they've made, some of which 11 we'd like your advice on about how to 12 implement that most effectively. 13 And we are beginning, believe it or 14 not, another five-year strategic plan. That 15 means that I've been here almost five years, 16 because I was here when we started the last 17 one. 18 You know, Susan's going to go into 19 a lot more detail about the NEMS system, the 20 redevelopment or redesign. And some of you 21 were around and remember when this current 22 version was started in 1991. And you think BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 15 1 about what's happened in the energy markets 2 since '91, there's been an enormous amount of 3 change, restructuring, deregulation. And as 4 I mentioned, in oil and gas, there were 5 virtually no biofuels in 1991. It didn't 6 exist in our database. 7 On the gas side, there's been an 8 enormous change. In 1991, coal bed methane 9 was less than 1 percent of our gas supply; 10 it's 10 percent today. Other unconventional 11 gases, if our model is correct, will 12 represent 40 percent of our gas by 2030. 13 So clearly, you try to keep up with 14 these changes by making adjustments in the 15 algorithms and add-ons, but oftentimes when 16 you do that in an existing large-scale 17 mathematical model, as many of you know much 18 better than I do, it creates a lot of 19 unintended consequences and you come out with 20 results that, quite frankly, don't make any 21 sense unless you go back and relook at the 22 whole system. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 16 1 And a few of us were at the energy 2 economists meeting in Detroit, and Bill Hogan 3 made an excellent luncheon talk. And he made 4 some of these points about how energy 5 modeling can be useful, but clearly, in his 6 case, he related it to electricity 7 deregulation, there are a number of things 8 that we need to do in that area to bring NEMS 9 up to date, and many others, as I've talked 10 about. 11 Where we are now is Susan Holte, 12 who will speak to you at 9:30, is heading up 13 our internal design and planning team for the 14 redesign of NEMS. And we've sent letters out 15 to key stakeholders who are interested in 16 NEMS, and asked for their input. And 17 obviously by virtue of this meeting, we're 18 asking for yours. And we think it's probably 19 about a three-year project. We're hoping 20 that if all goes well by maybe the 2010 21 Annual Energy Outlook, the new NEMS would be 22 up and running. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 17 1 Something that's very near and dear 2 to the heart of Jae on this committee, the 3 voluntary reporting on greenhouse gases is 4 underway and been redesigned. And I think if 5 all goes well, that will begin collecting 6 that data in '07. And thank you all for your 7 help in the work leading up to this. 8 I already mentioned the deepwater 9 discoveries. And I think these are events 10 that we have to respond to, whether it be 11 because we're asked to analyze what does it 12 mean for energy markets -- but in the case of 13 the deepwater discovery, it brings up another 14 point which is directly related to the 15 redesign of NEMS, and more specifically, to 16 forecasting in general, and that is 17 technological change. 18 Now we try to do our best in 19 capturing technological change in every 20 aspect of the industry, and we have utilized 21 independent expert groups to come in and say 22 are we doing the right thing in trying to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 18 1 capture what's going on in whatever sector it 2 happens to be. But I don't think anyone 3 could have anticipated the drilling in 12,000 4 feet of water and producing from the lower 5 tertiary trend in 1991, when NEMS was 6 designed. 7 So it's that kind of thing that 8 we'll be looking to all of the -- not only to 9 this committee, but to all of those I 10 mentioned that we've sent letters out to 11 asking for input how best to capture 12 technological change. I mean, we've been 13 using I think a reasonable system to do that 14 in terms of recent trends, what's going on in 15 each sector, but we'll certainly be looking 16 for outside advice on that as well as in the 17 downstream sector. 18 You know, we talked about the 19 Energy Policy Act of last year, and some of 20 the implications that has had for a number of 21 things. I mean, certainly, the one you hear 22 the most about is biofuels and things like BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 19 1 particularly ethanol and the subsidies and 2 incentives for renewables, and we certainly 3 will capture all of that in our Annual Energy 4 Outlook 2007. But many of these items that 5 were in the EPAct 2005 were authorization 6 language, but no appropriations. So indeed, 7 many of the things that we think are going to 8 occur, we can't really model without knowing 9 what the actual budgets and appropriations 10 will be. 11 But we are working with others 12 within the federal government, such as 13 Agriculture, to try to understand better 14 what's going on in the production of biofuels 15 and what it means for our projections. For 16 example, we don't collect on a regular basis 17 the ethanol production, but Agriculture has 18 some collections on their system, so we're 19 working with them on that. 20 And clearly we need to improve 21 that, since ethanol, we think, will represent 22 about 8 percent of total gasoline demand in BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 20 1 our long-term forecast. So it's extremely 2 important that we get the -- we capture all 3 that blending that's going on outside of 4 refineries. Our systems were designed to 5 capture data at the refinery, at bulk 6 terminals, but we never anticipated -- you 7 know, I think there's about 2,500 blenders at 8 last count that are not in our -- that are 9 increasingly -- we're trying to capture that 10 data, but it's increasingly important and 11 necessary. 12 So clearly, our intention is to 13 improve both the oil and gas data quality in 14 both the areas of ethanol and biodiesel. We 15 are proposing in the '08 budget, which is now 16 at OMB, a system to collect the biodiesel 17 data, beginning with the '08 budget. 18 Bringing you up to date, as I 19 mentioned, Congress left town without having 20 voted on the energy and water budget, and our 21 request was about $90 million. The Senate 22 actually added $3 million to that in their BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 21 1 subcommittee. The House approved the $89.8, 2 so we're hoping the Senate prevails when they 3 come back and they get in conference 4 committee. 5 One of the issues that came up 6 during the budget process was that at that 7 roughly $90 million budget, we said that we 8 couldn't keep doing everything and keep our 9 quality up, as we've talked about before this 10 committee previously. 11 So we proposed the dropping of 12 three surveys, petroleum marketing surveys 13 mainly affecting the price or the pricing of 14 both the first purchases of crude oil in the 15 domestic market and imported prices for 16 crude, as well as electricity Form 767. 17 Interestingly enough, even though those three 18 forms are relatively small in terms of the 19 total budget, we estimate about $1 million to 20 do those three. 21 It kicked off kind of I think a 22 reasonably strong pushback from users of that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 22 1 data, including some within the federal 2 government, such as IRS, which uses the 3 pricing data; EPA, which objected to 4 the -- they use the data from the electricity 5 survey. So we got some pretty strong support 6 for keeping these surveys. And indeed, that 7 was one of the reasons the Senate added 8 $3 million in addition to it. The two 9 reasons the Senate gave the Senate Budget 10 Subcommittee on Energy and Water for adding 11 $3 million was to restore these three surveys 12 and to add collection data on ethanol and 13 biofuels. 14 So we're hopeful, again, that that 15 works out, because as you know, the trend in 16 the EIA's budget that we've talked about for 17 a number of meetings here has not been 18 favorable, but we hope this is the beginning 19 of the turnaround and the increase that we 20 hope that will come to be passed for the '07 21 budget. And we have reason for optimism for 22 '08 as well. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 23 1 We've talked about that external 2 study team, which -- the design of that team 3 actually was very much influenced by your 4 advice in that we model it after kind of a 5 visiting -- outside visitors kind of program, 6 as they do in many of your universities. 7 And Denny Ellerman, heading that up 8 from MIT, and then an excellent supporting 9 cast that we talked about previously, I think 10 it really worked really well, in our view, 11 because it forced us to pull together the 12 information they needed to do their 13 assessment or their study and kind of forced 14 us to do almost an internal review before we 15 could prepare for the external review. And 16 there are three main recommendations that 17 come out of the study. 18 And again, the first one is that we 19 should do more analysis. And they said that 20 it was obvious when you look at that, what 21 suffered in that long downward trend in real 22 prices was less analysis. We had to focus BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 24 1 obviously on the core data collection. And 2 up until recently, we haven't dropped any 3 surveys, and as I mentioned, we decided that 4 couldn't continue. We had to either drop 5 some things or the quality would continue to 6 decline. What they're saying is that there's 7 a linkage between the data collection and the 8 analysis. And if you kind of shortchange the 9 analysis, you actually don't keep up with 10 what data you need to keep the analytical 11 function current. So that was one 12 recommendation. 13 And then the IT recommendation was 14 in response to a trend in at least the 15 federal government that there be a 16 consolidation of IT functions within each 17 department under one CIO, chief information 18 officer. And they thought in EIA's case that 19 didn't really work well because we needed to 20 maintain that independence, and we certainly 21 agreed with that. 22 And then the one where I think we BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 25 1 may be asking -- I think for sure we'll be 2 asking you advice in this: Making greater 3 use of the external market information. And 4 the team thought that one of the areas where 5 EIA may not have kept up to date as well as 6 we could have is taking advantage of the 7 information that's out there in the 8 marketplace and using that in our analysis. 9 And we think we have some 10 ideas -- and they gave us some on how we 11 might do that, but it's an area where we 12 certainly will seek your advice, among the 13 many other areas that we'll be talking to you 14 about these two days and in future meetings. 15 And finally, we are launching -- I 16 think in two weeks, we'll have our first 17 meeting to begin our next five-year strategic 18 plan. And again, we're trying to use this as 19 a way of, one, implementing some of the ideas 20 that have come out of advice from groups like 21 yours and from the external study committee, 22 as well as the recognition and need to keep BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 26 1 up with the change that is always underway 2 with respect to what EIA does. 3 So again, I want to close by 4 thanking you all for holding this meeting and 5 being willing to spend some time in advising 6 us. And we looking forward to the next day 7 and a half. Nick, again, thank you. And I'm 8 certainly available for questions now or over 9 the next two days. 10 MR. HENGARTNER: Any questions? 11 Jae? 12 MR. EDMONDS: Yes. Guy, I was 13 wondering if you'd take a couple minutes and 14 elaborate a little bit more on the Ellerman 15 committee findings. 16 I think the first and the last 17 intrigued me. The first was do more 18 analysis. And what does that mean? 19 And then the third one was use 20 external data sources more, and I'm not quite 21 sure what that means. So if you could 22 elaborate a little bit on what the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 27 1 implications for EIA are, I'm sure that Denny 2 put a few more pages than that on the topics. 3 MR. CARUSO: Yes, and I think the 4 full report's available, yes, to everyone 5 here. What they were reacting to is that 6 when they looked back on the last -- well, 7 especially the last 10 years, they saw that 8 the EIA used to publish a lot of standalone 9 analyses as either working papers or, up 10 front in many of our publications, there used 11 to be an opportunity for EIA staff to publish 12 analytical work on that particular -- like, 13 for example, in the Petroleum Supply Monthly, 14 you might have an article on deepwater 15 drilling or, on the other extreme, we'd have 16 things on the refinery sector, individual 17 analytical pieces that could either be part 18 of a statistical publication or a standalone. 19 And there had been a significant decline in 20 that. That's what they meant. 21 MR. EDMONDS: And these were issues 22 that EIA was taking on itself as opposed to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 28 1 Congress asking EIA do a special study on X? 2 MR. CARUSO: Yes, exactly. Yes, I 3 should have mentioned that, Jae. It was 4 self-initiated analytical work as opposed to 5 Sen. X saying please do a study on carbon 6 restrictions or on the NCEP. 7 The third recommendation about 8 using information that's out there in the 9 marketplace, they felt that maybe that's one 10 area where EIA has not kept up as well as we 11 might have. And that is because so much 12 information's available now out -- and they 13 were specifically -- and the examples that 14 came up during the study team were things 15 like futures and forwards markets options, 16 and whether or not we're capturing -- when we 17 do our short-term energy outlook, whether 18 we're really making use of that kind of 19 information in trying to project, in this 20 case, particularly oil, oil prices. 21 And we're saying, yes, it's a great 22 idea, but we're not sure how to do that. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 29 1 That's why I said we'll be looking for 2 advice, and what's the implications of hedge 3 fund activity and the commodity price index 4 funds. We all talk about it, but no one 5 really knows how to incorporate that into our 6 projections. 7 MR. CLEVELAND: Guy, just to follow 8 up on the first point, I feel like the 9 decline in discretionary analysis is a real 10 loss, I thought, for oil and gas in 11 particular. And what we only get now is 12 whatever's blowing in the current political 13 winds oftentimes. And whereas there used to 14 be -- in the oil and gas supply module, for 15 example, you'd often -- it was a way to get 16 into the guts of NEMS when someone working on 17 it would think about, well, how do you model 18 tautological change? And so there might be a 19 paper on that or what are the trend and find 20 development costs? That'd be a way of not 21 having to get into the code of NEMS, but to 22 start -- a way of users, at least some of us, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 30 1 thinking about some of the issues that we 2 struggle with in building these models. And 3 I think that's a significant loss, and I'm 4 sure it's true across every sector of NEMS. 5 MR. CARUSO: Well, I think Denny 6 and Paul Joskow, I think they're the two that 7 felt the most strongly about exactly what you 8 just said. And at least in my experience of 9 four years here is that it was -- we had a 10 choice between continuing to try to put the 11 resources on the core data collection and do 12 as much of the analysis as possible, and that 13 always was the discretionary analysis, to 14 follow up on Jae's point, that suffered. And 15 I agree. And then we certainly -- if we get 16 additional money, we'd certainly like to 17 spend more resources on that. 18 MR. HENGARTNER: Susan? 19 MS. SEREIKA: On the work that you 20 did here or the work that the external team 21 do, did any of this come through your request 22 for funds for next year? Like, for example, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 31 1 that first point. 2 MR. CARUSO: Well, yes, I think 3 if -- particularly for in '08, in the 4 budgetary request that we submitted, if 5 everything goes as requested, we would 6 definitely have more money for analysis in 7 addition to a redesign of NEMS, and the third 8 area where a lot of that additional funding 9 would be expended would be improving data 10 quality, particularly on oil, where as I 11 mentioned, things like biofuels and other 12 changes really were not captured in our 13 system. 14 MR. NEERCHAL: Coming from an 15 economic background, our department, for 16 example, is going to go through one of these 17 external reviews this year. 18 And it went through one of those 19 seven years ago. Usually the reports usually 20 sit on a shelf somewhere. So I was wondering 21 what is the next steps for this report and 22 who is the audience? Who is going to get it? BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 32 1 And if we feel that we need to support some 2 of these points, how do we forward our 3 support to those points? 4 MR. CARUSO: There are two things 5 very specific. One is that we have, as I 6 said, at least a little -- some effect 7 already is in our request for the '08 budget. 8 Secondly, the strategic plan that we're 9 launching in the next couple weeks will use 10 some of the recommendations in there to guide 11 us in that. 12 But there are a couple of other 13 things that they recommended. One is there 14 were a couple of areas that they felt that 15 really weren't essential to EIA's mission. 16 One was the financial reporting system, and 17 they recommended that we drop that. I didn't 18 include that in these highlights, but that 19 has been a controversial program, for 20 example. And I think there have been three 21 attempts to drop that since it was in the 22 authorizing legislation of 1977, and all BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 33 1 three met stiff resistance from Congress. 2 And we thought that if we were to propose 3 that, let's say in our FY '08 budget, to drop 4 financial reporting of energy companies at a 5 time when we're not too far removed from 6 Enron and other activities, that it really 7 wasn't politically feasible. 8 And then the collection of the 9 renewable transportation -- 10 MS. KIRKENDALL: Alternative-fuel 11 vehicles -- 12 MR. CARUSO: Alternative fuel. 13 There was another form -- alternatively 14 fueled vehicle form. They said that that 15 really probably belonged in EPA or was not 16 really an EIA function, not in energy. But 17 again, at this time when so much emphasis on 18 biofuels and renewables that's probably not 19 in the cards. So we aren't necessarily 20 following up on -- implementing -- planning 21 to implement everything. 22 MR. SINGPURWALLA: Did they give BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 34 1 any tips on implementing the stuff or was it 2 just recommendations? 3 MR. CARUSO: It was mostly 4 recommendations. You know, I mean, we had 5 more in the offline conversations with Denny 6 and Paul and Jay. Jay Hakes, the former 7 administrator, of course, has been through 8 this and understands the difficulties, but 9 was extremely valuable to have him there. 10 And they all said that they would be happy to 11 continue discussing this. 12 MR. SINGPURWALLA: Do you think 13 it's primarily a manpower issue? I was 14 looking at the top point again specifically. 15 MR. CARUSO: That one is, yes. 16 Yes, I think it's a combination of having 17 enough people and then the right people. And 18 as we've talked about in this group 19 previously, there is a significant turnover 20 of people with -- about one-third of our 21 staff at the senior level turning over in 22 '06, '07, '08. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 35 1 MS. KHANNA: Well, yes, sort of 2 related to that, which is does this affect in 3 any way the hiring priorities? 4 MR. HENGARTNER: You stole my 5 question. 6 MS. KHANNA: I mean, if you talk 7 about analysis, you're talking about a 8 different kind of group as opposed to people 9 who are really good at collecting data or 10 collating data. 11 MR. CARUSO: I think so. I think 12 we would hope to be able to get more 13 analytical positions, particularly I think 14 where it's -- if you look at -- and I don't 15 want to -- since there's a lot of staff here 16 I don't want to single out, but it's fair to 17 say that the office that suffered the most is 18 the oil and gas office in terms of the 19 analytical -- the personnel available to do 20 analysis. I think they're about one-half of 21 where they were -- certainly from '96. So in 22 a 10-year period, they're about one-half. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 36 1 MR. HENGARTNER: Related to hiring, 2 the one thing I didn't see in your 3 presentation is actually something that came 4 up several times in the last six years is the 5 turnover, a lot of people retiring and 6 getting new people into EIA. And I was 7 curious to see if we made progress or, if 8 not, if that is something that one should 9 focus in the next five-year plan? 10 MR. CARUSO: Oh, definitely, that's 11 critically important. The workforce planning 12 is the buzzword, but we lost 16 senior people 13 in '05. We're probably looking at around 20 14 in '06; a similar number, maybe a little more 15 in '07. And we have been I think reasonably 16 successful in attracting high-quality 17 applicants in the new -- so we have this 18 bimodal demographic distribution developing 19 now in EIA. We actually have some people in 20 their twenties again, and then a large group 21 in the fifties and even in their early 22 sixties. And so what we want to do is BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 37 1 continue to attract good young people with 2 analytical skills, obviously 3 statisticians -- that we've talked about in 4 this group about how difficult it is to get 5 math stat -- high quality, qualified 6 applicants. 7 And then to the extent that people 8 who are eligible for retirement would stay a 9 bit longer, that would help, too. So it's a 10 combination of making it -- retaining people 11 a bit longer and attracting young people, and 12 then the mentoring end activities of 13 transferring that knowledge to the new 14 people. 15 We had a very active intern program 16 this summer, the most active -- I think it's 17 got to be -- in EIA history. We had more 18 than 20 interns in the summer of '06, and we 19 think it was very successful. 20 How many of the interns would 21 ultimately come to work for EIA? Hard to 22 say, but it's useful in recruiting as well BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 38 1 because some of them go back to their 2 universities and say we had a good time 3 working there, they gave us some interesting 4 work. It's not a bad place, you know? All 5 those old people aren't so bad after all. 6 So yes, I definitely should have 7 mentioned that because that's one of our 8 great challenges. As I said, probably a 9 significant -- at least one-third turnover in 10 three years. Nancy? 11 MS. KIRKENDALL: You might talk 12 about we've got the big recruiting -- they're 13 doing a lot of work right now. We have a 14 contractor helping us to improve our 15 materials for recruiting on the web, because 16 it's awful. 17 MR. CARUSO: Yes. 18 MS. KIRKENDALL: It has been awful, 19 and they're trying to put together better 20 descriptions of the jobs we have. They're 21 trying to make it sound like it's more 22 interesting. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 39 1 MS. FORSYTH: Potential for 2 analysis. 3 MS. KIRKENDALL: So there is an 4 effort and I think they've made some 5 progress. At least the things I've seen look 6 pretty promising. So you might occasionally 7 check the web and see if you think it looks 8 any better. 9 MR. CARUSO: Yes, we have hired a 10 recruitment coordinator. When you think 11 about it, 10 years ago, EIA was in decline. 12 So we hadn't hired -- there was a long period 13 there where there was a hiring freeze. And 14 to go from that to someone who actually 15 that's what they do full-time is -- and it 16 came out of our strategic plan and it 17 certainly will be an important part of the 18 next five-year plan. 19 MS. KHANNA: Actually, follow-up to 20 Nancy. The one thing I noticed, occasionally 21 either from you or from Bill, we'll get an 22 e-mail saying, hey, we have a job available BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 40 1 and please advertise it to some of your 2 colleagues and students. But at least for 3 economists, I mean, if I'm looking for a job, 4 I go to the central marketplace for 5 economists, which is job openings for 6 economists. 7 If you are trying to recruit, 8 recruit at least economists, that would be 9 the place to create a link to, because people 10 don't go anywhere else, literally. I mean, 11 if you're looking for an academic or a 12 federal government-type job or even private 13 jobs with an analytical component. And it 14 could be I think really important to get 15 hooked up to that. 16 SPEAKER: Get the advertisement in 17 the right spot. 18 MS. KHANNA: It comes out five or 19 six times a year and it's really the central 20 marketplace. 21 MR. CARUSO: Thank you. 22 Susan? BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 41 1 MS. SEREIKA: For the 2 statisticians, you should go to the joint 3 statistical meetings, because they have lots 4 of opportunities for recruitment there. 5 That's where more statisticians will go. 6 MR. CARUSO: I know we always have 7 at least several people attending those. 8 MS. SEREIKA: Do you? 9 MS. KIRKENDALL: We don't always 10 recruit them, so that's a good point. 11 MS. SEREIKA: Yes, a lot of 12 students go there for recruitment. And those 13 interns, where are they in terms of their 14 training? Are they toward the end of their 15 training for their degree or are they 16 mid-way? 17 MR. CARUSO: They range from 18 juniors in college to some that were in 19 graduate school, and at least two of them 20 were the JPSM, right, Nancy, from the 21 University of Maryland's Joint Program of 22 Statistical Method? BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 42 1 MS. KIRKENDALL: Yes. 2 MS. SEREIKA: Because I would 3 imagine if you caught them in their last 4 year, that's probably ideal. Because they're 5 right at that point where they're starting to 6 think about where they want to go next. And 7 if you give them that wonderful opportunity 8 to do the internship, that they might -- you 9 know, might capture them. I know that like 10 in engineering, when they do their 11 cooperative experience, it's in that last 12 year and a lot of people transition right 13 into that job. 14 MR. NEERCHAL: I think the issue 15 here I think is that you are competing with 16 many other recruiters, also. 17 MS. SEREIKA: That's true. 18 MR. NEERCHAL: I think we sent one, 19 Katherine (?), and she had a wonderful time 20 and she gave a very good report, but she 21 wants to do her Ph.D. in Iowa, so that's -- 22 MS. KIRKENDALL: That's what BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 43 1 happens when you get good ones. 2 MR. NEERCHAL: Yes. So I think 3 this is about competition. So I think going 4 to JSM and putting a recruiting booth is a 5 good idea, even when you have one opening, 6 but you kind of make your presence felt there 7 because then you know the recruiters from 8 year to year to year, the EIA's (inaudible) 9 recruiter. 10 MR. CARUSO: Thanks. 11 MR. WEINIG: I might mention to the 12 committee that at 3:40 this afternoon, we do 13 have a discussion chaired by the deputy 14 administrator, Howard Gruenspecht, on the 15 external study team. It will not be -- it's 16 a pretty bulky report, but we are printing 17 copies of that so that you'll be able to see 18 it later today. 19 MR. CARUSO: Thanks, Bill. I 20 should have mentioned that in response to 21 Jae's question. There will be an opportunity 22 to really dig into that a bit more. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 44 1 MR. HENGARTNER: Okay, thank you 2 very much for your patience with us. 3 MR. CARUSO: Thank you, Nick. 4 Thank you, everyone. 5 MR. HENGARTNER: Next, I'd like to 6 invite Nancy Kirkendall as the director of 7 the EIA's Statistics and Methods Group to 8 tell us what is happening to the advice and 9 input we provided to her over the last year 10 or so. 11 MS. KIRKENDALL: Actually, what 12 we're going to do is give out our thank yous 13 for our departing committee members, and then 14 I'll go. 15 The first one is for Susan Sereika. 16 Susan has been a member of the committee for 17 the last three years, and we'd like to thank 18 you for your advice. 19 MS. SEREIKA: Well, thank you for 20 having me. 21 SPEAKER: And a couple of things, 22 Susan. The certificate and the letter from BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 45 1 Sec. Bodman. 2 MS. SEREIKA: Well, thank you very 3 much. 4 MS. KIRKENDALL: A certificate and 5 a letter from Sec. Bodman. Okay, our next 6 one is Mark Burton. Mark has been a member 7 of the committee for six years. 8 SPEAKER: Forever. 9 MS. KIRKENDALL: Forever, yes, it 10 seems like. We've seen him through two jobs. 11 SPEAKER: Nancy, how about giving 12 him a (inaudible)? 13 SPEAKER: Sell it fast. 14 SPEAKER: I'd rather take 60 bucks. 15 SPEAKER: You're becoming an 16 economist. 17 MS. KIRKENDALL: Finally, Nick 18 Hengartner, who has also been a member of the 19 committee for six years and has served ably 20 as our chair for the last two years, and has 21 kept us entertained. 22 MR. HENGARTNER: Thank you, Nancy. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 46 1 This is the end of the era. 2 SPEAKER: And the beginning of a 3 better one. 4 MR. HENGARTNER: Thank you so much. 5 SPEAKER: Right, now we can do some 6 analysis. 7 MS. KIRKENDALL: All right, this is 8 just a little summary of what's happened 9 since the last meeting in terms of some of 10 the suggestions you gave us. 11 We had a session, just a 12 wide-ranging discussion about how modeling 13 could suggest data needs. And the intention 14 on this was to try to get some advice 15 on -- since EIA sort of uniquely have data 16 users as well as our data providers in-house, 17 so we thought maybe we could get some ideas 18 on how to establish better communications 19 between those two groups. Because the 20 modelers sit over here and complain about the 21 data side, but we don't -- you know, and this 22 is an oversimplification. We do talk to each BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 47 1 other. It's just -- it could benefit from 2 being more formal. 3 But the discussion ended up being a 4 little bit more trying to bring in outside 5 ideas on data needs, which is probably 6 useful, too, because we haven't really done 7 enough of that. So some of the ideas that 8 came up were engaging the energy modeling 9 forum in some kind of a formal way, perhaps 10 special issues in the energy journal, and 11 then outreach efforts. 12 One of the things that senior staff 13 has actually talked about -- we've had our 14 NEMS conference every year for a number of 15 years trying to get input from users, and 16 that's worked pretty well. The 17 recommendation is, well, why don't we have 18 EIA Week or something, just expand that 19 conference to a couple days to bring in the 20 data side, too? In recent years, it's been 21 more than just a NEMS conference. They've 22 brought in STEO issues. So it has been BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 48 1 expanding, and I think that that's something 2 that senior staff seems to be interested in 3 doing, too. So I think you will see more on 4 the user conference idea as we go forward. 5 That'll show up in strategic planning, I'm 6 sure. 7 We had a session on the new 8 representation for the SAGE Model, and you 9 had some suggestions. You're going to hear a 10 talk by our contractor tomorrow at 8:30, so 11 you'll hear more about what we have learned 12 on that. This is all work-in-progress, so 13 you're going to be hearing things as we go 14 forward. It's one thing about giving you 15 information -- we try to pressure people into 16 talking to you before they're really ready 17 to, but at least that way we are more 18 interested in listening to advice. 19 And on the SAGE Model, the 20 committee commented the model was too complex 21 and we're still working on a more empirical 22 approach. And hopefully, that'll come back BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 49 1 at a future meeting. 2 The 2006 manufacturing energy 3 consumption survey, you had some good 4 suggestions. As you know, the Census Bureau 5 is our data collection agent on that, and we 6 work very carefully with them so that we're 7 working through considering inexpensive 8 alternatives for incentives. They're 9 examining non-response follow-up letters. 10 They're trying to improve things in general. 11 We had a session on respondent 12 cutoff dates for electricity data 13 collections, and a very good discussion. And 14 what that has turned into now is you're going 15 to be hearing about Electricity 2008. That's 16 the proposals for going forward with the 17 electricity surveys, and the piece of that 18 that relates to this session is that now 19 they're talking about doing a sample of the 20 smaller respondents. 21 The 860 and the 861 are Census 22 surveys. They've been run every year. And BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 50 1 so what they're talking about doing is taking 2 the smaller respondents and putting them on a 3 three-year cycle instead of having it be 4 annual. So in the course of three years, 5 you'd have a census, but you'd have -- every 6 year, you'd do a sample of about a third of 7 them. And so that's the idea that's related 8 to this cutoff date thing, to try to reduce 9 the burden on getting in late response, 10 non-response for all these little tiny 11 companies. 12 The IDC. We had a good discussion 13 on the IDC, the functional requirements 14 document. They tell me they are very close 15 to having it done. They've incorporated a 16 lot of recommendations. This has actually 17 been a really good effort in EIA. I think 18 everybody is serious about trying to make it 19 happen and moving to a common way of doing 20 Internet data collections. 21 We are going to be implementing the 22 greenhouse gas program in this new IDC BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 51 1 system. And the only little challenge we 2 have to work through is that we have legacy 3 Internet data collection systems, and we have 4 to figure out how do we bring them along and 5 move them into the new system. So there's 6 still some things we have to work through to 7 try to actually end up with a good, 8 consistent IDC technology. 9 MR. EDMONDS: Nancy, what is the 10 greenhouse gas program? 11 MS. KIRKENDALL: What is it? 12 MR. EDMONDS: Yes. 13 MS. KIRKENDALL: It's the voluntary 14 survey of greenhouse gases that they're going 15 to put in this IDC. 16 MR. EDMONDS: Okay. 17 MS. KIRKENDALL: And then finally, 18 this was a study, a co-integration analysis 19 between crude oil and natural gas prices, and 20 you recommended that we continue to do 21 similar analysis. We have published the work 22 you saw. The authors, Jose and Fred, do plan BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 52 1 to look at that trend term. They haven't 2 managed to do that yet. 3 However, that study did generate 4 further interest in EIA. Andy Kydes has had 5 an intern work on it over the summer. They 6 did an analysis of crude oil, natural gas, 7 and coal prices, and trying to figure out 8 what does the model tell you and what doesn't 9 it tell you and what can you learn from it. 10 So we're interested in the 11 technology and trying to figure out how to 12 use it. And so it's always slow trying to 13 figure out how to do something new. You have 14 to bring it in and use it and work through 15 all the details. 16 So that is what I have. Any 17 questions? 18 MS. KHANNA: This may be a question 19 that needs discussion separately, but I was 20 just curious why is the EIA collecting the 21 greenhouse gas data as opposed to, say, the 22 EPA? BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 53 1 MS. KIRKENDALL: Well, that was 2 sort of what the external study team asked. 3 I don't really know how we got tapped with 4 it, but Congress decided that we should do 5 it. Maybe they trust us. 6 MR. CLEVELAND: Well, I mean, it's 7 not that unreasonable, I mean, given that 8 it's intimately tied to energy production and 9 consumption. 10 MS. KHANNA: But there's plenty of 11 stuff that the EIA -- the EPA does. I mean, 12 it's just at a time when budgets are 13 declining and we're having to turn off 14 service, which only the EIA can probably 15 provide. 16 MR. CLEVELAND: Right. 17 MS. KIRKENDALL: Well, the other 18 thing is it's a voluntary survey. It doesn't 19 represent anything. Whoever wants to send in 20 stuff can send it to you. 21 MS. KHANNA: Thank you. 22 MS. KIRKENDALL: You know, so what BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 54 1 does it really mean? You know, it's a way 2 for somebody to get on record of having done 3 something. 4 MS. KHANNA: Well, that's even more 5 reason the EPA should be doing it, because 6 they have a huge voluntary emissions database 7 which has variable quality. 8 SPEAKER: That was charitable. 9 MR. NEERCHAL: Yes, that's what I 10 was going to say, have you seen the data EPA 11 collected? I think that in fact, most of 12 EPA's data is self-reported I think one way 13 or the other. I think part of (inaudible) 14 compliance and things like that, I think they 15 really do not do service like you do. I 16 think that maybe -- 17 MS. KIRKENDALL: They may do some, 18 but, yes. 19 MR. NEERCHAL: But I want to just 20 ask and make a comment. You mentioned the 21 legacy system. One of the things I think 22 sometimes people don't talk about -- what BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 55 1 you're building right now is going to be 2 legacy some day. And I think that dealing 3 with a past legacy, many times you are 4 dealing with the paper information that you 5 are trying to put up, and I think it's like a 6 one-time something you have to deal with. 7 But it's even more important to really look 8 at the system you're building right now and 9 be aware that one day, the technology is 10 going to change, and make sure that it does 11 not become as much of a burden as the legacy 12 of today. 13 MS. KIRKENDALL: Right. 14 MR. NEERCHAL: I think 15 that's -- you know, it is kind of one of 16 those megatrend kind of an issue and I think 17 it's very important to keep in mind. 18 MR. FEDER: Nancy, I just wanted to 19 comment that even voluntary service can be 20 made representative by use of some weighting, 21 such as propensity scores and things like 22 that. People do that with web-based surveys BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 56 1 where you have -- I mean, which are also 2 voluntary. I know Harris Interactive is 3 doing it with their surveys. 4 MR. EDMONDS: Yes, but these data 5 are just public relations. Let's face it. 6 SPEAKER: That's right. 7 SPEAKER: They're biased. 8 SPEAKER: Which is probably why 9 it's Energy and not EPA. 10 MR. EDMONDS: I was there, I was 11 involved with the original creation of this 12 stuff and it is -- there's just -- I mean, it 13 has no content. 14 MR. FEDER: But Jae, do you think 15 that the greenhouse emissions survey can be 16 made representative using propensity scores? 17 If you post-stratify by industry type and 18 things like that, you can at least improve 19 the -- 20 MR. EDMONDS: See, it's not -- see, 21 it's about emissions reductions, not about 22 emissions. And so it's a delta based on BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 57 1 heaven knows what. 2 MR. NEERCHAL: Estimates. 3 MR. EDMONDS: Right. And then how 4 does that fit into the larger accounting? I 5 mean, that's why I sense some frustration is 6 that it doesn't -- I mean, I can't think of 7 any way you could use it even if it were 8 totally accurate. 9 MS. KIRKENDALL: The only thing 10 that might be interesting, I understand many 11 electric companies report on it and we could 12 match to our electric power data, so you 13 might be able to do something interesting 14 with that, but other than the electric power 15 industry, I don't think that you have enough 16 to do anything with. 17 MR. FEDER: That's a weighting 18 technique which we commonly use in service by 19 using auxiliary data, so yes. 20 MR. HENGARTNER: Well, thank you 21 very much. It's nice to see that our advice 22 is taken seriously. Thank you, Nancy. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 58 1 I think we are -- unfortunately, 2 we're half an hour behind schedule. This 3 doesn't mean -- that's all right, I think 4 it's good to have those discussions. 5 I'd like to segue into our first 6 presentation, which is done by Susan Holte, 7 which is one of the technical advisors to the 8 EIA. And she's going to talk to us about 9 energy modeling systems. And I think this is 10 one of those issues that I hope will generate 11 discussion within the committee. And so, if 12 need be, we'll just stretch out this session 13 until we're all talked out. 14 MS. HOLTE: Does that mean I blow 15 off my 10:30 meeting? 16 MR. HENGARTNER: Oh, you have a 17 hard (inaudible). 18 MS. HOLTE: Well, that I can do, 19 but I think I do have to be out of here at 20 11:00. 21 MR. HENGARTNER: Okay, thank you 22 very much. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 59 1 MS. HOLTE: I'm sorry. This is 2 late. Anyway, thank you. Is this coming 3 through okay? Okay. 4 I've been asked to head a small 5 team of modelers and analysts who consider 6 the future development of the National Energy 7 Modeling System, and this effort has been 8 underway for about six months. We are 9 seeking to get input from interested 10 stakeholders on modeling and analytical 11 issues. Given that input, our own 12 assessments and those of the NEMS analysts, 13 we will prioritize (inaudible) development 14 projects for the next two fiscal year budget 15 cycles and beyond. And we may also initiate 16 some research projects on our own. 17 In addition to me, the team members 18 are: Bill Helkie, who joined EIA this year 19 from the Federal Reserve Board; Jeff Jones, 20 who's a lead modeler for electricity markets 21 in NEMS, and he has more than 25 years' 22 experience in EIA modeling and analyzing BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 60 1 electricity markets; Andy Kydes is the senior 2 technical advisor for the Office of 3 Integrated Analysis and Forecasting, and he 4 has nearly 30 years of experience in energy 5 analysis and modeling, with some private 6 sector experience in natural gas markets; and 7 Dan Skelly has worked with EIA both as a 8 contractor and as a federal employee for more 9 than 25 years in a number of areas, and more 10 recently has done work with the industrial 11 and transportation modules of NEMS, and he's 12 responsible for the overall system 13 configuration of the model as well as the 14 greenhouse gas emissions and analysis. 15 Personally, I've worked on the 16 design and development of all the major 17 integrated models in EIA and its predecessor 18 organizations going back to Project 19 Independence, which has everybody under 35 20 saying why are you still here? Referring to 21 the demographic issues. 22 Anyway, NEMS was developed in the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 61 1 early 1990s, and the call for NEMS arose from 2 the National Energy Strategy, which was 3 released in 1991. The model that EIA used at 4 the time was largely econometric and was 5 deemed inadequate to analyze many potential 6 energy policies. 7 As a result, there was a lot of 8 publicity about a new, biggest and best 9 energy model that EIA was going to build with 10 public statements by the Secretary and the 11 Deputy Secretary. The Office of Policy, 12 Planning, and Analysis in DOE commissioned a 13 National Academy of Sciences study to provide 14 recommendations on a model. And there was 15 some activity by the Secretary of Energy 16 Advisory Board. 17 Finally, the administrator, Calvin 18 Kent, decided to create a NEMS project office 19 in late 1990. And this consisted of the 20 then-technical assistant, Bill Skinner; John 21 Conte, who is now head of the Office of 22 Integrated Analysis and Forecasting; me under BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 62 1 a different name; and John Holte, if you get 2 the connection there. 3 The mission of the project office 4 was to research models and methodologies and 5 plan the design of NEMS. And the mission 6 also included representing the project in the 7 National Academy and other organizations, and 8 make some budget and organizational 9 recommendations. 10 Then, EIA created the Office of 11 Integrated Analysis and Forecasting in 12 October of 1991, with a mission of developing 13 NEMS, producing the Annual Energy Outlook and 14 the International Energy Outlook, but leaving 15 the short-term energy outlook out of that 16 office's mission. OIAF developed various 17 design papers, including very detailed 18 component design reports on each of the 19 modules and submodules of NEMS, which were 20 circulated for review, both in and outside of 21 the department. 22 And OIAF also started working group BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 63 1 meetings in each subject area so there would 2 be periodic interaction with interested 3 people in the department, the national 4 laboratories, and some outside organizations, 5 and this practice continues today. 6 We're undertaking the project a bit 7 differently this time, for several reasons. 8 First, we have an office in EIA with the 9 central responsibility for midterm energy 10 modeling and analysis, and that was not the 11 case 15 years ago. We didn't have a 12 centralized organization. 13 Second, we believe that EIA has a 14 functional model for many analytical 15 requirements, although more functionality is 16 required. 17 Finally, it seemed unnecessary and 18 impractical to start from scratch. In an era 19 of tight budgets and many requirements upon 20 EIA, the cost would be very high. So we do 21 not have a separate formal project officer at 22 this time. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 64 1 Much has been made of NEMS being 15 2 years old, but the model has been extensively 3 developed and adapted to maintain relevancy 4 beyond just the simple updates and periodic 5 extensions of the forecast horizon. As 6 background information, we asked the NEMS 7 analysts for a list of the model changes that 8 have been made since the first incarnation of 9 the model, and received a very long list of 10 projects. 11 And I've listed a few here: 12 Restructuring of the electricity and natural 13 gas markets, mercury emissions, greenhouse 14 gas, the ability to put limits or prices on 15 greenhouse gases, technology learning, 16 additional standards and regulations, more 17 technological detail, integration of the 18 entire macroeconomic model within NEMS. So 19 these are just a few, but the model has 20 certainly been extensively adapted over the 21 15-year period. 22 At this time, the team has had BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 65 1 meetings with interested parties in the 2 department to ask for their comments and 3 suggestions on NEMS. I should note that many 4 people in the department are far more 5 interested in NEMS now than they were 15 6 years ago. 7 And this is largely due to the 8 Government Performance and Results Act of 9 1993. As part of their budget submissions, 10 many department programs are required to show 11 the expected outcome of their programs on 12 energy markets by analyzing the program 13 impacts through NEMS. So there's a very good 14 reason for them to be interested. 15 Inquiries were also sent to 16 relevant congressional staffers. Not 17 surprisingly, we have received no comments. 18 They use our analysis and our data, but would 19 not typically be interested in modeling down 20 at this level. Guy Caruso's also sent 21 letters to a variety of energy trade and 22 research organizations, and inquiries are BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 66 1 going out about now to some of the national 2 laboratories. 3 We have also asked for suggestions 4 from the modelers within OIAF looking for 5 projects that may be beyond their ongoing 6 model development. So interest does vary 7 widely, and suggestions do vary widely 8 throughout the department. 9 After meeting with us, the Office 10 of Energy Efficiency and Renewable Energy 11 provided a 15-page paper of suggestions 12 ranging from outreach and communication to 13 broad topics of consumer behavior modeling 14 and technological change to some very 15 module-specific recommendations. 16 Fossil Energy was also very 17 interested with a list of recommendations, 18 including improved representation of all 19 alternative liquid fuels, improve our 20 representation of oil and natural gas 21 production technologies and distributed 22 generation, and transportation advances in BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 67 1 vehicle designs. 2 The Office of Electricity Delivery 3 and Energy Reliability has a more narrow 4 focus. Their issues are basically related to 5 the difficulty that they have in representing 6 their programs within NEMS, such as 7 transmission reliability, load shifting, and 8 storage. Some of their concerns may be 9 nearly impossible to address in the absence 10 of a transmission model. Many of their 11 metrics for their benefits analysis concern 12 congestion pricing and the cost of electric 13 outages. Nuclear Energy is primarily 14 interested in expanding the number of nuclear 15 technologies represented and the uranium fuel 16 cycle from supply to reprocessing and 17 storage. 18 Policy really didn't provide us 19 with much feedback. As a somewhat different 20 suggestion, they appeared more interested in 21 some scenarios in the three-to-six time frame 22 for which NEMS is really not suited. And the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 68 1 Office of Science is not interested in 2 weighing in, but wanted more information on 3 NEMS for some new staffers. 4 I'm not going to really dwell on 5 this slide, but just point out that the 6 combined vision for NEMS is always rather 7 lofty: Do everything, have all sorts of 8 diagnostics and uncertainty and everything, 9 and, oh, by the way, solve it in a wink of an 10 eye. But working from that lofty vision, 11 there are some overarching topics that we 12 must consider. 13 EERE and Fossil Energy are very 14 interested in extending the model horizon. 15 Even if we have no desire to extend the 16 projections in the Annual Energy Outlook, we 17 could -- and I repeat that is a big 18 "could" -- consider doing this to service the 19 needs of the department. Currently, a 20 contractor does that for them for their 21 benefits analysis. 22 Extending the model horizon raises BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 69 1 many issues. Obviously, one that always 2 comes to people's minds first is resource 3 depletion, but there's also the question of 4 how to represent technologies in such an 5 extended time frame. Generic technologies, 6 which is always sort of a fallback in 7 longer-term models, generic technologies may 8 not be very useful for the department 9 programs because they have to analyze very 10 specific programs. There's also a question 11 of how relevant a lot of the detail in NEMS 12 becomes as you stretch the forecast horizon 13 further out. 14 There were several issues raised by 15 people within the department about improving 16 the feedback between the energy and 17 macroeconomic models. There were a wide 18 range of infrastructure issues that are of 19 interest to the department, including 20 transmission, pipelines, G terminals, and 21 water use, which was of particular interest 22 to Fossil Energy. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 70 1 Technological choice and 2 technological improvement are topics of 3 continuing interest to us and to the 4 department, and how consumers choose 5 technology and the related topic of foresight 6 are paramount. 7 In addition to these broad issues, 8 there are some very specific model 9 improvements that are potentially of interest 10 to us and others. EERE is sponsoring the 11 development of a hydrogen market module for 12 NEMS. EIA is part of that process, so that 13 it is done within our modeling structure. 14 The process for adopting that model into our 15 baseline version of NEMS has yet to be 16 determined. 17 We would like to consider improving 18 representation of world energy markets, one 19 goal being to represent the competition among 20 energy sources and energy uses outside the 21 United States to better improve our domestic 22 projections. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 71 1 Not to pick on any one area, but 2 offshore wind is another technology we're not 3 currently representing in NEMS. There are 4 sure to be many others, such as gas hydrates. 5 I think there is a serious question 6 about whether we can capture some of the 7 important points about electricity 8 transmission, reliability, and congestion 9 without incorporating a transmission model. 10 Also, the solution time of NEMS has 11 become a problem. This is actively being 12 addressed through some algorithm changes. 13 There were other topics related to 14 NEMS and NEMS analyses that we do not view as 15 strictly modeling issues. In our meeting, 16 several people raised the issue of 17 transparency. This can run the gamut from 18 diagnostic tools to documentation in order to 19 understand the model, and how changes in 20 assumptions filters through to the results. 21 However, when pressed for further 22 information, nobody in the department came BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 72 1 back with any really specific ideas. 2 Our discussions with both EERE and 3 Fossil Energy would indicate that some real 4 investment on their part in learning the 5 model is really necessary. 6 NEMS relies on information about 7 technology characteristics from external 8 experts in various areas. An issue raised by 9 several people concerns the equitable 10 treatment of technologies, the so-called 11 level playing field. There is a real 12 question about whether all the experts are 13 evaluating the technologies with the same 14 level of optimism, and that technology 15 evaluation may be a tradeoff between 16 consistency and expertise. 17 So I guess that leaves a question 18 about whether there are guidelines that can 19 be offered in order to solicit this input 20 that's very necessary for the modeling. 21 There are various policies and 22 programs that cannot be treated now in NEMS. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 73 1 Local policies and deployment programs are 2 two examples. 3 Further outreach was also 4 recommended as a two-way street: Us 5 providing more information about the model 6 and the results, and us contacting more 7 experts to solicit input. Similar to many 8 recommendations, this is subject to available 9 resources, money, and sometimes even more 10 scarce, time. 11 So next steps? We will have 12 further discussions with stakeholders, and 13 the team will be looking at all the proposed 14 projects, reviewing and prioritizing them, by 15 very early next spring. 16 There are several major design 17 projects underway within OIAF. The oil and 18 gas supply module has been redesigned, and 19 the design addresses some of Fossil Energy's 20 recommendations. The peer review of that is 21 scheduled for early December. 22 A redesign of the petroleum market BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 74 1 module, that's our refining portion, has been 2 initiated. And there's some talk of doing 3 similar for the natural gas transmission and 4 distribution module. 5 A workshop on consumer 6 decision-making is in the planning stages. 7 Additional module reviews may be conducted. 8 And addressing a solution time, we've 9 initiated testing of another optimization 10 package. 11 Thank you. 12 MR. HENGARTNER: Well, Cutler? 13 MR. CLEVELAND: Thanks, Susan. 14 It's an enormous, daunting task given the 15 complexity of the model and the way it's 16 been. It's like the air conditioning system 17 in my building. It's been renovated and 18 re-renovated about 15 times the last 20 19 years. And every time they do it, some new 20 contractor comes in and puts in a new switch 21 or a blower or something. And when you call 22 when it doesn't work right, which is pretty BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 75 1 much all the time -- that's not true with 2 NEMS -- but when it doesn't work right, you 3 call the VU (?) staff that come in and no one 4 knows how the whole system works; they know 5 their little blower part. And I know it's a 6 daunting task, but I'm glad you're doing it. 7 I think the transparency, from the 8 user's side, the transparency in 9 documentation is I think the really important 10 issue. It's really hard to find out a lot 11 about the details of most models. You either 12 have kind of the report for the public that 13 you all produce and then some PDF buried 14 somewhere written in contractorese. 15 And there has to be some way to 16 present to users in some kind of 17 understandable, digestible way a little bit 18 more of the details than what you see in the 19 annual outlook or whatever the publication 20 is. And I think that would go a long way to 21 improving the transparency issue, which I 22 know always dogs you. And the users can help BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 76 1 do that because we know what we want and what 2 would help them. 3 One of the things that I didn't see 4 up there in terms of some of the big issues 5 was the connection to international markets. 6 And I'm wondering -- was it up there? 7 MS. HOLTE: No, I did mention that, 8 yes. 9 MR. CLEVELAND: Okay. I must have 10 been writing or not paying attention. 11 MS. HOLTE: Yes, that is really a 12 big issue. I'd like to enhance that. Right 13 now, the various import-export markets are 14 handled in a very -- you know, sort of 15 independently by fuel -- coal exports, 16 natural gas imports, oil, you know? But as 17 you go out -- especially as you go out 18 further in the time horizon, we anticipate 19 more need to consider the competition among 20 those fuels out there as it relates to it. 21 MR. CLEVELAND: Who wants to go 22 past 2050? BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 77 1 MS. HOLTE: What? 2 MR. CLEVELAND: Who's pushing for 3 the really long forecast? 4 MS. HOLTE: Well, right now, NEMS 5 goes to 2030. 6 MR. CLEVELAND: Right. 7 MS. HOLTE: So even 2050 is an 8 extension. 9 MR. CLEVELAND: But who wants it? 10 MS. HOLTE: Basically EERE and 11 Fossil Energy. They have -- some of their 12 programs, they view as -- well, definitely in 13 2030, just sort of on the edge or even beyond 14 2030. So when they're asked to do the 15 program benefits, they're sort of out of 16 luck. The contractor, as I said, does extend 17 the model, extend NEMS to 2050 in order to do 18 that analysis for them. 19 MR. CLEVELAND: Two other quick 20 points. One, you mentioned a lot and I just 21 wanted to reiterate how important I think it 22 is, modeling technological change. I think BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 78 1 you quite clearly understand that. I think 2 it's represented in a pretty ad hoc way 3 across many of the different NEMS models. 4 And I think there's a large 5 literature out there in the economics world. 6 In fact, this month's issue of Energy Policy 7 has a whole special issue related to how do 8 you represent technological change in carbon 9 abatement. And then you have the whole 10 engineering literature of learning curves and 11 experience curves. And it would be good if 12 you guys could somehow address this issue in 13 a more systemic and consistent way, not 14 necessarily that every module is going to be 15 represented the same, but it's pretty ad hoc. 16 And it would be good -- it would 17 give users more confidence if the 18 rationale/discussion of how technological 19 change is represented, for example, in oil 20 and gas finding costs, if that was connected 21 to this ongoing debate in the literature 22 amongst economists and engineers and between BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 79 1 the two groups on how best to do this. It's 2 not easy, I know, but I think you could do 3 better in that area. 4 And the other issue is statistical 5 methods used in some of these things. I 6 think, again, you need to make sure, as I 7 think you try to do with varying degrees of 8 success, stay up to speed with what's 9 happening in terms of the latest statistical 10 techniques, particularly when you're dealing 11 with forecasting issues, time series data, 12 just to make sure that you're up to speed on 13 what's happening out there in the forefront 14 of forecasting. 15 MS. HOLTE: Did you want to weigh 16 in, Andy, on technological change or not? 17 I'm not trying to put you on the spot. 18 MR. KYDES: Where we have 19 technological -- 20 MR. HENGARTNER: Sir, could you go 21 to the microphone, please? Thank you. 22 MS. HOLTE: We do have several of BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 80 1 the NEMS analysts here in the audience if 2 they want to weigh in. 3 MR. KYDES: Can you hear me? 4 MS. HOLTE: Yes. 5 MR. KYDES: Yes. The model 6 actually is not a single framework. We try 7 and build the model in each area -- the 8 decision-making -- based on what we think how 9 the market responds, and also all the 10 available data that's available to us. In 11 the electricity markets, they're lumpy 12 investments. They're easy to identify. 13 There's a collection that we can do. 14 And I've been doing technology 15 learning since I guess about 20 years now, so 16 I'm aware of the literature you're talking 17 about. And it is the thing that we 18 incorporate within a market where we can 19 identify significant investments of that 20 kind, lumpy and large. 21 In the area of, say, buildings, the 22 technologies are a little more muddy. There BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 81 1 are about 90 million, for example, 2 heaters -- space heaters, I mean, gas-fired 3 heaters -- and so the technology learning 4 curve in terms of learning by doing in some 5 of those things, it really kind of flattens 6 out. What you're really seeing is the 7 incorporation of new features: Dual fans, 8 dual speed, that kind of stuff. And that's 9 what's really playing, and so we do try and 10 incorporate that where it is a new 11 technology. 12 Now, in the area you're interested 13 in, I know oil and gas, the problem there is 14 that we don't have explicit technologies 15 there represented. And so in that area, we 16 haven't figured out how to incorporate tech 17 learning except as changes to the parameters 18 that relate to finding rates, for example, 19 cost per unit, and so on. So we're still 20 studying it, but if you have ideas, we'd like 21 to hear them. 22 MR. CLEVELAND: Well, I guess BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 82 1 there's this very interesting debate, you 2 know. You use learning curves, for example, 3 with these new technologies. And of course, 4 the engineers love learning curves, but the 5 economists have a very different way of 6 looking at it and would suggest different 7 ways of trying to model technological change. 8 So it might be worth kind of 9 engaging some of those folks and talking 10 about how best to represent technological 11 change. I'm not necessarily endorsing how 12 they do it, but for example, if you look at 13 this latest issue of Energy Policy, which are 14 mostly economists, they would approach how do 15 we forecast the impacts of new technologies, 16 like wind or something, on carbon abatement 17 from a very different perspective than the 18 learning curve perspective. Just something 19 to be aware of. 20 MR. EDMONDS: Is this the Michael 21 Grubb special issue? 22 MR. CLEVELAND: I don't know if it BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 83 1 is or not. I'm not sure. 2 MR. EDMONDS: I reviewed some of 3 those papers and they're scary. Yes, what's 4 in some of those models is just scary stuff. 5 And I think Andy's right in the 6 sense that it does require sort of a 7 fairly -- you really need to get some 8 perspective on technology, because learning 9 is a real phenomena. I mean, it's been 10 reproduced in study after study. But there's 11 a really nice paper, and I don't know if 12 you've seen it or not, but it's by Clark (?) 13 and Wye (?), which kind of goes through how 14 the models do -- and then Clark has a new 15 paper coming out where he goes through all 16 the different approaches and then goes 17 through the economic literature on it. 18 And what it boils down to is that 19 it's a really tough problem. I mean, 20 particularly as you're trying to take on the 21 whole energy system, as EIA is, because 22 technological change comes in through so many BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 84 1 different pathways. Particularly as you're 2 starting to go out toward 2050, what you find 3 is that the spillover effects become 4 increasingly dominant; that is, if you sort 5 of look at some of the big things that have 6 happened in energy. You look at, for 7 example, in production of oil, this 8 combination of 3-D seismic imaging with 9 directional drilling, and they've really 10 expanded that. 11 And sort of it you trace that back 12 and that comes through medicine where you've 13 got these CAT scans and then that came over 14 into the production side, and that, in turn, 15 was made possible all this computational 16 science. 17 And so you get all of this very 18 complicated set of interactions that are 19 going on. And that's going to be really 20 tough to model explicitly, and you may -- you 21 know, I wouldn't try to package 22 everything -- the vogue is to package BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 85 1 everything up into a learning curve and to 2 say that this relationship between a 3 declining price and cumulative deployment is 4 the whole story, and that the causality was 5 cumulative deployment led to the lower price. 6 You know, there's another hypothesis that 7 lower prices lead to greater deployment. 8 It's probably a much more complicated story, 9 and I think it is a much more complicated 10 story. 11 So I guess my advice is, yes, and I 12 think you probably are on top of the 13 literature, but that literature is 14 sophisticated. And probably, in the final 15 analysis, going through and just thinking 16 through the what-ifs may be as important a 17 contribution as you can make. 18 Trying to mechanize it so that you 19 actually are going to predict technological 20 change, I don't know that I would necessarily 21 recommend that you go fully down that path. 22 You know, it's just my two cents. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 86 1 MR. KYDES: Just two points. I'm 2 essentially confirming your point. The 3 spillover effects, I mean, who would have 4 thought, for example, that the PC, it's 5 accelerated development in the '80s and '90s 6 would lead to basically such great abilities 7 to do analysis of data that you previously 8 had to spend enormous amounts of money for? 9 MR. EDMONDS: Sure. 10 MR. KYDES: So there's no effect 11 that maybe might have been hard to anticipate 12 in advance, and just a point I think that 13 confirms your point. 14 The other area that learning -- and 15 we've looked at, is learning through R&D, 16 which the Europeans are heavily into. The 17 fundamental problem as far as I can tell is 18 that you kind of have to assume the parameter 19 or the success of what are the parameters are 20 to get significance on the other. 21 So in order to get -- if you put 22 them both in the experience portion and then BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 87 1 the R&D portion simultaneously, you don't get 2 a good fit. But if you, on the other hand, 3 sort of fix the learning through experience 4 of cumulative builds, and then estimate the 5 R&D component, you got a significant 6 (inaudible). 7 MS. HOLTE: Let me just comment on 8 one thing Andy said very early on about the 9 different parts of the model being handled a 10 bit differently in terms of technology. One 11 of the suggestions that we received, I 12 believe from EERE, had to do with 13 standardizing consumer behavior across the 14 model. Well, we think that's exactly the 15 wrong thing to do. We don't think that the 16 residential homeowner uses the same process 17 in making some decision about energy-using 18 equipment as does the person who's making it 19 for a big hospital complex as does an 20 electric utility. So we think that 21 standardizing in that way is exactly the 22 wrong thing to do. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 88 1 MR. HENGARTNER: Jae? 2 MR. EDMONDS: Let me also just 3 reiterate what Cutler said. I think it's 4 really great that you're doing the sort of 5 stepping back and thinking about what is it 6 you're doing and where you're going and just 7 how does this all fit together. I really 8 think that's a wonderful thing that you're 9 doing. 10 I think in doing that, one of the 11 things that I didn't hear, and presumably 12 it's because you don't have enough time to go 13 through all this stuff, but what you 14 obviously want to do is go back and ask 15 yourself the questions of what is it that I'm 16 trying to do here. Who are my customers? 17 What are the questions that I really need to 18 answer? And what is it that this model is 19 really going to have to do? 20 Because I'm sure that you're going 21 to get requests from congressional districts, 22 I would like to know what in 2043 the gas BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 89 1 supply to my district will be, and what's the 2 price, by the way? 3 MS. HOLTE: We got that just sort 4 of comment from EERE. 5 MR. EDMONDS: Exactly. 6 MS. KIRKENDALL: They would love to 7 do congressional districts. 8 MR. EDMONDS: You know, just sort 9 of sorting through exactly what is it that 10 you're going to -- you know, what issues are 11 you going to take on and which ones aren't 12 you going to take on? I think those are 13 important things. 14 MS. HOLTE: That's why we're 15 getting this input from the people around the 16 department about what is it that you would 17 like to address. 18 And we've got other input that 19 we're soliciting from outside. 20 MR. EDMONDS: Okay. 21 MS. HOLTE: I'm not the 22 decision-maker, but I think it's unlikely BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 90 1 that we're going to spill over into health 2 impacts of energy and emissions, you know? 3 That's not an area that we have expertise in. 4 EIA does not. So there has to be boundaries 5 like that drawn. 6 MR. EDMONDS: And I'm curious about 7 this equitable treatment of technology issue 8 that you raised. 9 MS. HOLTE: Yes. 10 MR. EDMONDS: What on earth is 11 that? 12 SPEAKER: Oh, come on, you know 13 what that is. 14 MS. HOLTE: Are you treating my 15 technology as fairly as you're treating your 16 technology? 17 MR. EDMONDS: Did you assume it got 18 as good as I'd like it to get? Is that what 19 the question is? 20 MS. HOLTE: Oh, of course. What 21 they're really after is my technology isn't 22 grabbing -- BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 91 1 MR. EDMONDS: Enough of market. 2 MS. HOLTE: As big a part of the 3 pie as I want. But -- so that's always an 4 issue that arises. Are we treating all 5 technologies with the proverbial level 6 playing field? And as I said, EIA goes out 7 to experts in the individual areas. And I 8 think it's a very real question. Is 9 everything being treated -- you know, when we 10 get back, our contractor reports on these 11 technology characteristics, the cost and 12 efficiencies and how they're going to change 13 in the future. Is that being done with the 14 same degree of optimism? I think that's very 15 important. 16 I don't view that really as 17 modeling. That's the input information that 18 goes into the model. 19 MR. EDMONDS: Right. 20 MS. HOLTE: And it is a very 21 serious concern. 22 MS. FORSYTH: Right. So my BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 92 1 follow-up question is I presume there are 2 standard protocols for collecting those 3 expert inputs. 4 MR. CLEVELAND: Guy's Rolodex. 5 MS. FORSYTH: That would be one way 6 of ensuring that they're treated fairly is by 7 standardizing -- it's really data, it's the 8 data collection. 9 SPEAKER: Well, when somebody -- 10 MS. FORSYTH: It's not -- 11 MS. HOLTE: It's opinion. It's 12 opinion. 13 MS. FORSYTH: Am I misunderstanding 14 how it's used in the model? 15 MS. HOLTE: It really isn't data. 16 SPEAKER: It's in the future how 17 good do you assume the technology becomes? 18 So that's the question -- I think that's the 19 question on the table and how big a share of 20 the market does it get. 21 MR. CLEVELAND: But wouldn't people 22 look at NEMS and they see (inaudible) this BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 93 1 tiny little thing 20 years out and they go 2 what do you mean? We're going like 3 gangbusters. 4 MS. FORSYTH: Oh, okay, I 5 misunderstood. 6 MR. CLEVELAND: We're going to take 7 over the world. 8 MS. FORSYTH: It's not input to the 9 model, right? 10 MR. EDMONDS: It is an input, but 11 it's a future state. 12 MS. FORSYTH: Yes, okay. 13 MS. HOLTE: You know, it's the 14 future efficiencies, the capacity packages, 15 all the costs that are involved, and these 16 are the things that go into the technological 17 choice within the model. So it's really not 18 data. 19 MS. KHANNA: And I have two 20 comments, one of which got stolen, but I'll 21 just reword it. I think this question of 22 equitable treatment of technologies is really BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 94 1 important. I don't know what you mean by 2 "level playing field" in terms of a 3 technology treatment within a model. 4 However -- however you resolve this issue, 5 you should be really aware that things like 6 NEMS assumptions have much better effect 7 across the policy arena. 8 For example, the Asian Development 9 Bank, the World Bank, various think tanks 10 across the world look at what are the 11 assumptions incorporated in NEMS. And then 12 those pick up the same assumptions that go 13 into the Kenya Energy Study for Greenhouse 14 Gas Abatement, exactly the same numerical 15 parameter values. Unfortunately, that's the 16 reality. 17 So it is actually a broader 18 implication, and I think stating up front how 19 you actually treat technologies and what are 20 the assumptions or what are the impacts of 21 the oil technologies assumptions on NEMS 22 results is going to become really key. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 95 1 That's my first comment. 2 The second was a question and a 3 comment. You talked about extending the 4 horizon to 2050. And you said -- or at least 5 there's a desire to extend the horizon to 6 2050. And I'm wondering how much that desire 7 is linked to galloping or what were galloping 8 oil prices. 9 MS. HOLTE: How much it's linked? 10 MS. KHANNA: Well, I'm -- 11 SPEAKER: Motivated. 12 MS. KHANNA: Yes, I mean, if oil 13 prices do come back to the 40- to $45 ranges 14 EIA predicts, will that pressure to in fact 15 extend the horizon to 2050 decline with that? 16 MS. HOLTE: I don't think so. 17 MS. KHANNA: Okay. 18 MS. HOLTE: No, because -- 19 MS. KHANNA: So then what's the key 20 reason for wanting to extend the model? 21 MS. HOLTE: Because the two 22 departmental elements that really brought it BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 96 1 up, Energy Efficiency and Renewable Energy 2 and the Office of Fossil Energy, they have 3 the benefits analysis that they have to do. 4 They've got some programs and technologies 5 that -- you take our 2030 version in the 6 model right now -- and those things are just 7 starting to become market players by the end 8 of that 2030 horizon. 9 And they -- for example, hydrogen 10 is one. And they're trying to say, yes, fund 11 my program, you're going to get results, and 12 you just see this teeny, little blip out 13 around 2028 or 2029 when they do their 14 benefits results. So that's where the 15 motivation is. 16 MR. HENGARTNER: Going out to 2050 17 is very interesting, but one always has to 18 ask ourselves the question how much 19 uncertainty do we have in those predictions? 20 And as you're talking to a bunch of 21 statisticians here who, of course, say, well, 22 it's just a point estimate without some BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 97 1 measures of uncertainty may be a little bit 2 misleading. And so maybe putting in some 3 thought on actually how to model and bring 4 out those uncertainties and being up front, 5 we talk about how technology changes and how 6 that's impacted. 7 But in fact, including the 8 uncertainties, oh, this one might work or 9 this might and so forth, there's a framework 10 I think. I don't want to (inaudible) 11 statistics, but there's a whole edifice that 12 allows, in a principled fashion, to actually 13 go and pick experts and get their opinion and 14 form something like a informed or relevant 15 prior and then that goes in. 16 I mean, there are actually ways 17 that I think we can bring to the table some 18 tools that may be useful at this particular 19 endeavor. So that's just a statistician's 20 point of view. 21 And last but not least, I'm looking 22 at the stakeholders. And I saw you mentioned BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 98 1 Dashiell (?) Labs, and I want to reiterate 2 that indeed, we talked a lot about energy 3 security. That is a big issue. And we 4 actually would like to have actually 5 international models for energy because we 6 think it's related to homeland security. 7 We think that understanding our 8 energy security is fundamental to ensuring 9 the prosperity of our country. And so you 10 have maybe a much broader set of stakeholders 11 that -- or maybe you are aware, but then just 12 reiterating the fact that there are many 13 people interested in the outcome of these 14 models as input to other things they want to 15 do. 16 Thank you. 17 MS. HOLTE: I'm in trouble now. 18 Howard? 19 MR. GRUENSPECHT: I just want to 20 say a couple things from those things you 21 raised. I mean, there are -- like, first of 22 all, we did just find out about your letters BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 99 1 to the labs at various levels, so we are very 2 interested in that interaction. 3 You know, your point about the 4 uncertainty, 100 percent agree with it. It's 5 a big -- you know, there's an issue I think 6 of modeling and there's an issue of 7 presentation, and those are related, but 8 they're -- you know, Washington, 9 unfortunately, the bandwidth is so narrow 10 that even to get one number is like a 11 big -- I think about that old system they had 12 to communicate with submerged submarines to 13 launch missiles, and I think the state of 14 Wisconsin was the antenna. Like four 15 characters was I guess all they needed 16 because that was the signal for the end of 17 the world. 18 But there really is a bandwidth 19 problem, even if technically you can do it. 20 For some reason, people want very simple, and 21 wrong, things to work with. And we try 22 to -- you know, Jae mentioned this point BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 100 1 about trying to do alternative -- you know, 2 certainly (inaudible) we don't know why. The 3 bottom line is -- who knows what the state of 4 technology is going to be in the future. And 5 one of the things we've done in some studies 6 is looked at the outcomes under alternative 7 states of technology, not trying to link the 8 policies. 9 You know, people like to link their 10 policies to driving technology. We have a 11 hard time doing that, frankly, and we don't 12 say that, but we do say, well, we can look if 13 technology was different, you know? We're 14 not going to say that because you enacted the 15 Energy Policy Act of 2005, technology is 16 different. But we'll say if technology is 17 different for whatever reason, this is what 18 it would -- you know, so there are ways we 19 try to do stuff. 20 But there's always this question of 21 presentation of bandwidth, that even if 22 there's a way to do it, even communicating BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 101 1 the simplest concept -- I mean, that a gallon 2 of ethanol and a gallon of gasoline have 3 different energy content -- I mean, you would 4 think that that would be like -- but I swear, 5 people talk about them interchangeably, you 6 know. 7 We're going to have about 5 million 8 gallons of -- you know, barrels a day of 9 ethanol. Well, that isn't the same as 10 5 million barrels a day of gasoline, but 11 you'd be surprised how many -- you know, so 12 it's almost daunting intellectual stuff, and 13 then there's this -- even after -- if you 14 could even win on that -- if it would be 15 smart enough to -- I mean, it doesn't say we 16 shouldn't try to do better. That's the whole 17 purpose of the project is to try to do a lot 18 better. 19 But it's also frustrating that you 20 could -- if what you bring forward is too 21 rich, then you run into an issue of how you 22 get it through this pipeline that's real BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 102 1 skinny. I mean, part of that is not the 2 model. I mean, part of that we're trying to 3 do already I think. Hopefully, our reports 4 are better than they have been historically 5 in making clear some of the uncertainties, 6 that we say this is the effect of the bill. 7 We don't mean the whole bill, we mean the 8 parts of the bill we thought we could model, 9 or showing how -- you know, and try and 10 emphasize that the referenced case in the 11 Annual Energy Outlook is not our forecast. 12 It is a reference, you know? 13 And we used to talk about forecast 14 error. And this year's thing on our web site 15 on performance -- because sometimes it's not 16 a forecast because supposedly we don't 17 consider changes to policy. You know, for 18 one thing, we know policy's going to change, 19 and there's probably some probability of 20 policies changing that we would incorporate 21 if we were trying to do a forecast, but we 22 don't even try to. So there are all kinds of BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 103 1 things. 2 And I think making -- part of it is 3 how we -- I mean, we have to work on making 4 the framework better, more modern -- you 5 know, all the things that Susan raised. But 6 part of it, there's this whole parallel track 7 of how you deal with the people who are 8 looking for this information. 9 MR. BURTON: To what extent does 10 EIA have a responsibility for not pushing 11 this beyond the place where you're confident 12 that this is usable? I mean, there's 13 reliable, there's crazy, and there's some 14 borderline in between. What responsibility 15 is there for making sure that you stay on 16 this side of crazy? 17 MS. HOLTE: Have we done that? I 18 don't know really how to answer that. 19 MR. GRUENSPECHT: I think, you 20 know -- 21 MS. HOLTE: You have to look 22 at -- consider available data. And certainly BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 104 1 there's a lot of things people would like to 2 do, but the data are not there to support a 3 credible effort. 4 MR. GRUENSPECHT: Right, the idea 5 of soliciting people's requirements is not 6 because we're going to concatenate them and 7 try to satisfy them. 8 MS. HOLTE: Oh, no. 9 SPEAKER: It'd be awfully 10 interesting (inaudible). 11 MR. GRUENSPECHT: I mean that would 12 be very much of a null set exercise when you 13 concatenated them. I mean, on the other 14 hand, people care about -- pick one. I don't 15 know, it said only pick one example. 16 A lot of people are -- the 17 transportation thing in the current National 18 Energy Model is sort of implemented on a 19 nationwide side, the demand side for motor 20 fuels for light-duty vehicles. In fact, if 21 there would be significant use of ethanol, 22 whether one thinks that's likely or not, my BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 105 1 guess is that might be regionally 2 distributed. There may be -- in terms of the 3 infrastructure, there may be reasons to use 4 ethanol disproportionately in certain 5 regions. So if what you have is sort of an 6 assumption that that comes out of motor 7 vehicles being fueled by petroleum products 8 and sort of petroleum products or generated 9 through the same processes in all regions, 10 and that there is some regionalization by the 11 West Coast is a little different, but 12 fundamentally it's sort of treated -- the 13 trends in fuel economy, the types of 14 vehicles, sort of you have a national fleet, 15 looking forward to a world in which a lot of 16 questions are being posed about biofuels. 17 You know, maybe that national 18 structure of the demand side of light-vehicle 19 transportation is not something that's -- it 20 isn't just a question of changing a parameter 21 of how much a (inaudible) ethanol plant 22 costs. I mean, parameter changes you can BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 106 1 always do. You know, you can fight with the 2 nuclear people over what's the overnight cost 3 of a nuclear plant. 4 And they want to say it's this, and 5 some expert will say it's that and people 6 have these views all over the place. But on 7 this transportation side, it isn't really a 8 question of what's the cost of an ethanol 9 plant, which is sort of a number that you 10 can -- but structurally, the question is do 11 you have a national representation of the 12 demand for light-vehicle travel, or do you 13 have a regional representation -- if there 14 are going to be region-specific kind of 15 developments? 16 And that's -- I mean, the parameter 17 things will always be a discussion, but the 18 structural things I think are what we, 19 certainly in this round, want to address. 20 But it's not that because people want to do 21 stuff, we're going to do it. I mean -- but 22 we think it is important to ask people. And BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 107 1 this is not about changing parameters. It's 2 more about trying to deal with structural 3 issues. I don't know if that helps, but -- 4 MR. CLEVELAND: I think a lot of 5 the issues -- I actually think NEMS does a 6 pretty good job of dealing with uncertainty 7 and it's not true on sorting the statistical 8 set. There's really -- we have no idea, so 9 we're going to run -- do the scenario 10 analysis. And I think that in general, NEMS 11 does a pretty good job at trying to bound 12 the -- not get out of the crazy mode. 13 I think the issue -- a lot of what 14 we're talking about is transparency. And the 15 problem is you are trying to be transparent 16 to a staffer and the Hill is never then 17 transparent to me. 18 And so from my perspective, which 19 is a difficult task that you face, from my 20 perspective, I would like to see more 21 explicit -- it's either -- for me it seems 22 like it's either -- it's bimodal. You either BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 108 1 have the Annual Energy Outlook or some 2 85-page PDF document that was written by a 3 contractor that describes the model. And 4 something in between which deals with these 5 assumptions about learning curves or 6 technology choice to a different set of users 7 where it doesn't have to be quite as 8 distilled would help a lot, at least in some 9 user communities. But I know that takes 10 resources to produce that extra 11 documentation. 12 MR. GRUENSPECHT: I mean, one of 13 the issues -- it's interesting, there's an 14 issue in EIA that's been discussed a lot, 15 which is like the web site. You know, we've 16 hired -- it sounds like an odd thing. But 17 like the Annual Energy Outlook has some 18 examples, like pieces in it -- I don't know 19 if people are familiar with it, but like 20 issues -- you know, there are the numbers 21 that people -- oh, they say it's going to be 22 45 this year or not, you know. For a long BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 109 1 run 45, not 70 or whatever, and a lot of 2 people focus on that, but there are these 3 issues in focus articles. 4 Now, it's interesting because I 5 think many people would find those useful, 6 but they're now buried in PDF of the entire 7 Annual Energy Outlook. And there are many 8 people who don't want to read the entire 9 Annual Energy Outlook who may care about 10 those issue discussions, like something about 11 the greenhouse gas -- you know, how you deal 12 with the regional policies for greenhouse 13 gases or how you're dealing with -- there are 14 different ones each year. 15 And one option that we're thinking 16 of is kind of taking some of those 17 discussions, still including them in the AEO, 18 but also having them -- have, like, places on 19 the web page where people who are looking for 20 discussions of those issues instead of 21 knowing, oh, there are five-page discussions 22 of seven issues in this AEO this year, that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 110 1 no one who is not -- someone might be 2 interested in those issues and not interested 3 in the AEO. And so taking those articles and 4 kind of making them accessible in a different 5 way and writing other articles like 6 that -- but there is -- it's a real -- you 7 know, some of it I think is I don't want to 8 say repackaging, but I think there is a lot 9 of stuff that is buried in places where some 10 piece of a PDF taken out and made available 11 more directly for someone searching on that 12 issue might be a more valuable thing. 13 MR. CLEVELAND: I think that would 14 be very useful. It would get a lot of -- a 15 lot of people would read those things that 16 don't read them because they're very -- and I 17 don't think you were here this morning. We 18 were talking about the external committee's 19 report. And one of the big things was this 20 decline in sort of independent analysis by 21 EIA as a real sad outcome of the decline in 22 funding, and then the need to cut that to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 111 1 deal with the core mission of data. And this 2 would go a long ways, given what you 3 currently have, of trying to meet some of 4 that need. 5 MR. HENGARTNER: I'd like to have 6 our speaker go to her 10:30. And so I'll 7 take the last question for Walter. He's been 8 so patient here. 9 MR. HILL: Okay, thank you. I have 10 two points. I'll try to be quick. One of 11 them -- the long time frame. I've seen 12 academic papers, maybe we all have seen them. 13 We have a 50-year time horizon, with claims 14 like we believe the oil data for the first 30 15 years, but if you go out another 10 years, 16 something else comes online and a couple 17 scenarios. If there's a lot of oil, this 18 happens. If there's no oil, this happens. 19 So in some sense, it's not 20 unreasonable that you might be interested in 21 that, although it might not be appropriate 22 for this group to look at it. So that's item BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 112 1 one. 2 The second one is on Cutler's early 3 statement about the international issues. 4 That may have been the early slide that you 5 saw about the policy -- international policy 6 people being relatively uninterested. And 7 maybe what's going on is the experts in the 8 senior caucuses, people who are experts in 9 the caucuses turn out to be econometricians. 10 So maybe that's why you're not getting 11 responses. I assume what the group does is 12 you look at a couple scenarios to see whether 13 or not there's a supply disruption someplace, 14 or whether that has an effect on the model. 15 In a sense, you get point estimates and you 16 want spreads within the model. Maybe that's 17 how you compensate for lack of information 18 from the international specialists. 19 MR. HENGARTNER: Okay, well, thank 20 you very much. 21 MS. HOLTE: Thank you very much. 22 MR. HENGARTNER: Before we break BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 113 1 for a fast coffee and then for the breakout 2 session, I'd like the members in the room who 3 have not introduced themselves this morning 4 to state who they are. 5 MR. RUTHERFORD: Tom Rutherford, 6 I'm a member of the committee. 7 MR. HENGARTNER: Okay, off to the 8 breakout session. 9 (Recess) 10 MR. HILL: I was deputized to run 11 this session. We are to go to about 50 12 minutes past. We're running a bit behind, so 13 that'll be the target time. We might not 14 need that. We'll see how it goes. And I 15 believe -- 16 MR. FRENCH: Maybe I shouldn't say 17 that because I'm known for talking on at 18 length, so who knows how long this session 19 might go. Can everybody hear? Everybody 20 ordinarily can. More? 21 SPEAKER: More. 22 MR. FRENCH: Something wrong with BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 114 1 this microphone? Is it deadening the sound? 2 SPEAKER: There's a ringing. 3 MR. FRENCH: Must a result of the 4 recent renovations. Should I begin now? 5 MR. HILL: Go ahead, yes. I'm 6 sorry. 7 MR. FRENCH: Good morning. And I 8 want to thank the committee for putting up 9 with me once again. My name is Dwight 10 French, and I do direct the Energy 11 Consumption Division of EIA. I have been in 12 front of this committee, I don't know, six, 13 seven, eight times in the past. 14 This will be my final time. I'm 15 going to be leaving federal service in the 16 very near future, before the next committee 17 meeting, and so in this occasion, I probably 18 am going to do something a little bit 19 different than I've done in previous times. 20 I've been before the committee before talking 21 about a variety of statistical methodology 22 and operational methodology issues regarding BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 115 1 the consumption area. I'm going to be a 2 little bit more philosophical this time. And 3 I certainly do want to thank the committee 4 for their efforts on our behalf over the 5 years. 6 For those of you who have gotten 7 the background information, may have had a 8 chance to go through it, you might be 9 puzzling what in the world is this Chapter 3, 10 Chapter 16 business? For the last two years, 11 I have been involved in the process of 12 writing a history of the consumption area of 13 EIA. 14 And these two chapters I thought 15 might add something to the discussion I'm 16 going to bring here. 17 And it was interesting -- the 18 discussion about a midpoint between a very 19 superficial discussion of things and a very 20 detailed discussion of things coming up about 21 10 minutes ago at the end of the NEMS 22 discussion. That really is the purpose of BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 116 1 what I am writing in the history. It's to 2 try to have a concise discussion of the 3 origins and some of the early decisions and 4 so forth that were made in the consumption 5 area of EIA, as a help maybe to some people 6 when they come on board later on and perhaps 7 for some users that would be able to look at 8 things in 5 pages rather than look at 1,000 9 pages of technical documentation somewhere. 10 But anyway, onward. If I can work 11 the machine here. First of all, the 12 necessary disclaimer, and you've seen it 13 before, so I won't dwell on that. 14 When I'm talking about history, 15 let's take just a little brief look at the 16 consumption area of EIA. Our responsibility 17 is to deal with the demand side of the energy 18 throughput in U.S. society. You can do that, 19 to some extent, on supply side surveys based 20 on delivery statistics, but of course, when 21 you do that, you tend to get sectoral totals, 22 you don't get any description of sector. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 117 1 And so the real purpose -- core 2 purpose of the consumption area is to provide 3 information and analysis on the areas of 4 energy consumption in the various sectors of 5 the country, along with a description of the 6 characteristics of the users so that you 7 have, instead of simply an accounting, an 8 understanding of what is going on. 9 If I had to characterize this 10 presentation, I would say it would be 11 characterized as if you can't do everything, 12 what can and should you do and why? And what 13 I'd like to do maybe is have you folks here 14 put yourselves in a position sort of like the 15 external review team when they were thinking 16 about EIA's responsibilities. And I'll talk 17 about the consumption area, but some of the 18 things I have to say really have a wider 19 application than just the consumption area. 20 Obviously, we have a responsibility 21 to cover the demand side of the equation in 22 EIA. But like other parts of EIA, we don't BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 118 1 do it all. We've never done it all. We 2 don't have the money, we don't have the 3 resources to do it all. Our data 4 collections, because of the nature of 5 consumers, are separated by 6 sector -- residential, commercial, 7 transportation, pieces of industrial -- and 8 across energy sources, whereas your supply 9 surveys are generally across sectors but 10 within individual energy sources. When 11 you're surveying consumers, it's a different 12 story. 13 However, we don't do all sectors. 14 We don't do all energy sources. And although 15 we collect a wide variety of characteristics 16 about the consuming entities, we don't 17 collect everything by any means. It would be 18 impossible to do so. 19 Thinking about limited sectors for 20 a moment, what do we cover? Households, 21 commercial buildings, manufacturers basically 22 with our own surveys. We used to have a BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 119 1 survey of household vehicles. However, it 2 was discontinued after 1994, and our latest 3 effort in dealing with household 4 transportation energy use is a derived data 5 set based on the Department of 6 Transportation's National Household Travel 7 Survey conducted in 2001 and 2002. 8 The sectors we don't cover are: 9 Transportation other than household vehicles, 10 although I will tell you that at present we 11 are investigating the possibility of using 12 the 2002 VIUS, Vehicle Inventory and Use 13 Survey, to strip out the -- it's actually a 14 truck database, strip out the residential 15 vehicles out of that and have a complementary 16 non-household highway energy data set. 17 The sectors we don't cover: 18 Transportation other than household vehicles, 19 as I said, except for VIUS. 20 In industry, we cover 21 manufacturing; we do not cover agriculture, 22 mining, and construction. In the commercial BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 120 1 area, we cover buildings; we do not cover 2 non-building service activity, things like 3 streetlights, parking lights, plug load in 4 parks and that sort of thing, billboards, 5 pumps, drawbridges, all those types of 6 non-building service activity. 7 And of course, even within the 8 sectors we cover, we have a myriad of 9 exclusions: Commercial buildings less than 10 1,000 square feet; buildings and households 11 on military bases that we can't access; 12 vacant and seasonal households; manufacturers 13 that are very small and not on the Census of 14 Manufacturers mail file. Our Manufacturing 15 Energy Consumption Survey only covers those 16 on the CM mail file, which is about 17 two-thirds of manufacturing establishments or 18 so, certainly greater than 95 percent of 19 manufacturing energy consumption. And we 20 don't cover things like motorcycles, motor 21 homes, and other types of exotic riding 22 equipment in household transportation. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 121 1 MS. KHANNA: Can I ask a really 2 quick question? 3 MR. FRENCH: Sure. 4 MS. KHANNA: Sorry to interrupt. 5 MR. FRENCH: No, that's quite all 6 right. Interrupt at any point. 7 MS. KHANNA: Do you cover use of 8 energy by military at all? 9 MR. FRENCH: No. 10 MR. FEDER: In transportation, what 11 about aircraft? I mean, all the commercial 12 airlines. 13 MR. FRENCH: All of the 14 non-highway, no, we do not cover it, never 15 have. 16 MS. FORSYTH: Only residential 17 airplanes. 18 MR. FRENCH: Only -- no, no 19 residential airplanes either. 20 MR. FEDER: My helicopter does not 21 count? 22 MR. FRENCH: No boats, no BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 122 1 recreational boating, no nothing. If you 2 want information on that, the best place to 3 go is the Transportation Data Book, which 4 some of you may be familiar with. 5 On the energy side, we're limited 6 also. With households, we collect 7 electricity, natural gas, fuel oil and 8 kerosene, LPG, and we do get some wood 9 information. We don't do coal, gasoline, or 10 propane for grills or anything like that or 11 other exotic fuels. 12 In commercial buildings, you see 13 what we cover. District heat is an 14 interesting one. It represents steam, hot 15 water, chilled water that is brought to 16 individual building sites either from offsite 17 or maybe from a central plant in some types 18 of commercial facilities. But we do not 19 cover LPG, coal, or other exotics. 20 And we'll get to why we're not 21 doing this in a minute, by the way. I'm just 22 showing you the limits right now. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 123 1 For manufacturers, we cover 2 virtually everything because a manufacturer, 3 if they think it's an energy source, they can 4 fill it in and account for it. 5 On household vehicles, we cover 6 basically gasoline and diesel, which accounts 7 for virtually everything now. 8 On the characteristics side, sure, 9 we're limited. We have a lot that's not 10 known about the engineering of housing units 11 that affects energy use, a lot on technical 12 characteristics of systems and equipment. We 13 have some information on use of particular 14 types of equipment, number of loads of 15 laundry done in a week, number of showers 16 taken in a week, and so forth, but there's a 17 lot more potentially we could do. 18 On commercial buildings, we do not 19 know a lot about the engineering of the 20 building structure. 21 We try to get what we can. We 22 don't know a lot of things about number and BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 124 1 type as well as technical characteristics of 2 systems and equipment, and especially, we 3 have a shortage of information about 4 equipment that contributes to commercial 5 building plug load. For example, in this 6 building, there are who knows how many 7 refrigerators sopping up energy, but we don't 8 know, and generally, we don't know. 9 Currently, with household vehicles, 10 what do we know about the vehicles 11 themselves? Given what NTHS does, we know 12 nothing other than make, model, and year of 13 the vehicle. We know a lot about the travel, 14 the trips associated with that vehicle, but 15 we don't know about the physical 16 characteristics of the vehicle. Why? 17 Because the Department of Transportation 18 collects the survey for trip purposes and us 19 borrowing that survey means we take what they 20 do for their purposes. 21 Now, why do we have limited 22 sectors? Well, you're going to see cost on BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 125 1 every one of the following slides, so you 2 might as well realize that that's the case. 3 Other sectors are not explicitly 4 required in law. And frankly, one of the 5 reasons that we were able to discontinue the 6 Residential Transportation Energy Consumption 7 Survey is that it's not explicitly required 8 in law either. EIA, of course, has enabling 9 legislation that specifies, directs the 10 administrator to collect information on 11 supply and storage and imports and exports 12 and transmission and so forth, and 13 consumption is in there. But as far as 14 explicitly requiring a particular survey, 15 RECS, CBECS, MECS, the manufacturing, 16 commercial buildings, and household surveys 17 are the ones that have been explicitly 18 pointed out in law as being necessary to run; 19 others are not. 20 There is perhaps a perception of at 21 least relative unimportance of some of the 22 other sectors. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 126 1 The United States consumes about 2 100 quads of energy. If you look at 3 households, commercial buildings, and 4 manufacturers, together they represent about 5 55 percent of that total consumption. If you 6 add in residential transportation, you're up 7 to 70 percent or maybe slightly more. That 8 is what we historically have covered. The 9 big hole, obviously, is non-residential 10 transportation. It would be a very complex 11 and enormously difficult and costly hole to 12 cover if we were to ever try to do it. 13 And that gets me down to the final 14 point of this slide, which is difficulty in 15 accessing the sector. In addition to 16 non-residential transportation, obviously 17 things like construction would be very 18 difficult to access also, because the energy 19 use takes place in disparate places over 20 perhaps relatively short time periods, 21 especially in the circumstance of residential 22 construction, and perhaps in ways that aren't BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 127 1 very collectible by field workers. 2 Why don't we get all the energy 3 sources? Well, there's cost again. It is 4 difficult to obtain the information for some 5 of the exotic fuels, because a lot of them 6 are cash-and-carry or their delivery by 7 people and organizations that maybe don't 8 maintain their records all that well. We 9 struggle with fuel oil and kerosene when we 10 get information about it. 11 And in addition, some of these 12 fuels are not used very much. Our sample 13 sizes are fairly small, and so the sample 14 sizes for our surveys don't support quality 15 measurement of these rarer energy sources. 16 And let's face it, these sources don't amount 17 to much relative to the major ones. When you 18 are putting together a program, you always 19 have this balancing act. What do you 20 include? What do you exclude? You want to 21 try to exclude the things that are most 22 inconvenient for you but at the same time BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 128 1 have minimal impact on the core information 2 you're trying to collect, which in our case, 3 of course, is energy consumption and 4 expenditures. 5 What about limited characteristics? 6 Well, I put cost down here, but actually 7 that's somewhat of a red herring. In and of 8 itself, collecting more detailed 9 characteristics of consumers would not be all 10 that expensive because our field work, most 11 of those costs is getting to the respondent. 12 Adding some more to the questionnaire really 13 cost-wise is not that big a deal. And with 14 the processing advances that have come about 15 in the past 10, 15, 20 years, processing 16 additional information is not the incremental 17 cost that it used to be either. But we do 18 have issues of burden control. 19 We certainly have issues of 20 respondents not being able to answer the 21 questions because they're ignorant of 22 technical issues -- and this is not a BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 129 1 criticism of the respondents; they just don't 2 know -- or because data are simply not 3 readily available. And even if we had more 4 information, relating it to energy use 5 sometimes is a demanding derivation or 6 modeling exercise, and sometimes things don't 7 work out all that well. 8 We have someone who just left our 9 division this past week, Robert Latta, who 10 did a lot of end-use modeling for our 11 division and was a right-hand person to me 12 for the past 26 years in EIA. He would add 13 things, for example, to the Residential 14 Energy Consumption Survey questionnaire 15 because he was modeling end-use for 16 residential households: The heating, the 17 water heating, the air conditioning, 18 appliance load, and so forth. And sometimes 19 variables that we would add in would add to 20 the model, sometimes they wouldn't. Anyway, 21 sometimes our information is self-limiting in 22 that way. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 130 1 Okay. So we have all of these 2 limits on what we can do and what we do do. 3 And we're continually, in our own minds, 4 asking ourselves do we want to take the next 5 step? Do we want to consider new 6 initiatives? Do we want to expand 7 questionnaires? Do we want to collect 8 different information? 9 Certainly we do have initiatives 10 that change in response to new technologies, 11 new issues of the day, and so forth. But we 12 always have this underlying we can't do it 13 all, and we're thinking about should we do 14 things differently. Should we do things in 15 addition to what we're doing? Do we need to 16 cut out some things in order to change what 17 we're doing? 18 Now, how could we address the 19 limitations that I talked about? Well, 20 obviously, you could think about something 21 like a new household transportation survey, a 22 non-residential transportation survey, a BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 131 1 survey of construction, a survey of mining, 2 et cetera, et cetera. 3 You could think about new questions 4 on existing surveys. In fact, we do, all the 5 time. 6 You could conduct research studies 7 and engineering assessments and try to 8 incorporate that type of information, which 9 might be able to get you some more technical 10 characteristics information, along with the 11 more standard-type data that we collect in 12 our surveys. And in fact, we have done that 13 in the very distant past. The problem, of 14 course, there is cost because engineering 15 assessments or other types of technical 16 assessments, case-for-case, tend to be pretty 17 expensive. So you're really in a 18 circumstance where you have to do a limited 19 number of cases and try to model out effects 20 of what you might get from that information. 21 We certainly can collect and 22 compile pertinent information from other BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 132 1 studies and do derivation. And that's what 2 Mark Schipper, who's eating a bagel back 3 there, did in the residential 4 transportation -- I'm sorry to embarrass you, 5 Mark -- did in the transportation area and is 6 one of our newest products up on the web 7 right now. There's some dirty laundry 8 involved with that, by the way, and you 9 can -- we try to be relatively transparent 10 about the dirty laundry along with the 11 information that we give. 12 We can do specialized derivations 13 and assessments for energy measures. And all 14 of these types of things have some sort of 15 resource requirements, from relatively small 16 to humongous. And when I say "humongous," 17 I'm talking multimillion dollars, which is a 18 big chunk of change for an agency whose 19 budget is $80- or $90 million a year, as 20 EIA's budget level is at this point. 21 So now I turn it over to you folks. 22 What do you think? You've heard about some BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 133 1 of the things that we collect and what we 2 don't collect. What do you think are the 3 most important areas for the consumption area 4 to collect and publish? Are we really 5 missing something? Is it important enough to 6 try to expand in other areas that we back off 7 in some areas that we are in now? Given that 8 we have no real reason to expect much 9 additional funding in the relatively near 10 future, you have to think of those 11 limitations as well. 12 Now, before I go any further, I 13 need to thank Howard Gruenspecht for a 14 moment, because a few years ago, the 15 consumption area was headed for a wall 16 financially. Our appropriations were not 17 what they needed to be to continue even the 18 program that we had. And I had told people 19 that we were headed for a brick wall, and the 20 beginning of calendar year 2006 was the time 21 to make some decisions. And Howard and the 22 EIA management saw fit to add some money to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 134 1 our appropriations so that we can continue 2 the program through the next cycle of surveys 3 in manufacturing, households, and buildings 4 that we have scheduled. 5 And now the tough nut to crack. 6 I'm not asking for money at the expense of 7 other places in the EIA. That wouldn't be 8 fair. But what I would ask the committee is 9 this: If EIA as a whole is going to make 10 choices in the future regarding its programs, 11 what do you think should be the criteria that 12 EIA should use to set its priorities as the 13 future unfolds among survey programs, among 14 data programs, among analysis programs? 15 You've heard a little bit earlier 16 this morning about how EIA will be shifting 17 its focus a little bit toward the analysis 18 side. And we heard some comments from the 19 committee about analysis may be taking a hit 20 over time a little bit, and EIA is going to 21 attempt to bring that back somewhat, and it's 22 probably overdue for something like that. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 135 1 But other than that, put yourself in the 2 position of an external review team, not 3 saying what EIA should do necessarily at this 4 point, but what are the criteria that EIA 5 should use in order to determine what its 6 priorities should be. 7 I leave you with those questions. 8 And I'll be willing to entertain any 9 questions and discussion you have, and let's 10 see what happens. 11 MS. FORSYTH: I have a question for 12 clarification. 13 MR. FRENCH: Sure. 14 MS. FORSYTH: When you talk about 15 programs, are they different -- like is 16 consumption a program? So you're talking 17 about competing interests at that level, or 18 are you talking within consumption competing 19 surveys? 20 MR. FRENCH: I would say both, 21 okay? 22 MS. FORSYTH: Okay. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 136 1 MR. FRENCH: You heard what we do 2 and what we don't do. In the consumption 3 area, it may be a time for shifting of 4 priorities within the consumption area, given 5 a fixed amount of resources. 6 I'm not asking about is it time for 7 program reassessment across programs in EIA, 8 but what criteria should EIA be using to make 9 such an assessment either within programs or 10 across programs in the organization? 11 And yes, I think the consumption 12 area is a self-contained division. I think 13 of it as a program area within EIA. Now, 14 others may not exactly think of it that way, 15 but that's the way I think of it. 16 MR. HILL: All right. Thank you 17 very much. 18 MR. FRENCH: Any comments or 19 suggestions about what the division should be 20 looking at in the future relative to what 21 it's looking at right now? 22 MR. HILL: We're supposed to take BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 137 1 questions. Go right ahead, questions from 2 both the panel and the audience. Go ahead, 3 you're on. 4 MR. FEDER: Sorry. My main 5 statistical interest is usually in measuring 6 change rather than measuring level. What I 7 would like to say here is if there are 8 certain things that are more likely to be 9 changing in the future, I would put more 10 emphasis there rather than on just estimation 11 of how much energy is consumed. If some 12 things are a given and are not going to 13 change, I think they're of less interest 14 sometimes with things that are going to be 15 changing because of new technologies or 16 because of the changes in the market. 17 Now, another consideration would be 18 where it's cheaper to collect data, certain 19 things, like if you can get data from 20 producers. But I guess that's on the supply 21 side, so, yeah. 22 MR. FRENCH: It is on the supply BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 138 1 side. Again, there are sectoral totals based 2 on supply information, but you don't have the 3 sectoral detail. 4 With regard to the comment about 5 what's changing most rapidly, what would you 6 think would be changing most rapidly over the 7 next 5, 10, 15 years, if you had to guess? 8 You're the external review team now. 9 MR. FEDER: But I'm not an energy 10 expert, so I would yield to those who 11 probably know more. But just as a consumer, 12 I would -- I don't know. Actually, I thought 13 I knew until about a week ago, when I heard 14 that, in fact, people are going back to SUVs, 15 so I don't even know that. I don't know. 16 Neha, maybe you know. 17 MS. KHANNA: I don't know. I can 18 only guess. 19 SPEAKER: Well, you get to go 20 beyond 2050. 21 MS. KHANNA: Oh, so much easier. 22 MR. FRENCH: Therein lies the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 139 1 difficulty, you see, to know exactly what 2 change is going to be occurring. I think 3 there's probably -- we seem to be on some 4 cusp of change in the transportation area, 5 but how it develops depends on how supply 6 versus demand develops. 7 MS. KHANNA: See, a part of it, 8 though, wouldn't it be determined by the 9 policy initiatives that are coming through? 10 Because if there is a push, for example, on 11 the renewable energy portfolio, then it means 12 that at an institutional level, there's going 13 to be an increased push at that end, and so 14 that should be something that -- and even on 15 the consumption side, perhaps we need to 16 focus on it. 17 So that would, for me, be one 18 criteria is to look at where the policy is 19 headed now. Of course, you can't project 20 policy beyond -- the direction which policy 21 might go beyond a shorter time frame because 22 as administrations change, policies might BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 140 1 change. But perhaps as an expectation that 2 renewables is a reasonable thing to expect to 3 be on the radar screen. 4 MR. FRENCH: And of course, the 5 problem with something like covering 6 renewables and consumption is that they start 7 out at a low level and our limited sample 8 sizes give us difficulty. 9 MS. KHANNA: Right. 10 MR. FRENCH: Not everything 11 penetrates, and not everything is the success 12 story that, let's say, microwave ovens were 13 in the 1980s, where the Residential Energy 14 Consumption Survey measured the penetration 15 of microwave ovens from -- over a decade 16 essentially about 1 percent of the households 17 in the country to 85 percent of the 18 households in the country. 19 Now, if you knew that something was 20 taking over a significant share of the 21 vehicle fuel market or the vehicle technology 22 market over the next 10 or 15 years, you BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 141 1 could put it in. The first or second time 2 around, you wouldn't get a lot and you 3 wouldn't be able to say a lot, but at the 4 end, if something really penetrated, you'd 5 have something. 6 But it's an interesting guessing 7 game to play, especially in transportation, 8 where we're literally talking about half a 9 dozen fuel/technology options or more that 10 are on the table with nobody really playing 11 out and saying, well, we're going to pick a 12 winner now or whatever. The argument at this 13 point is let the market pick a winner rather 14 than let somebody pick a winner. 15 MS. KHANNA: And they will. The 16 market picks -- they just -- several 17 technologies co-exist. It just may not be 18 that you get one big winner. Then what would 19 we do? 20 MR. FRENCH: An interesting 21 question in an arena where we don't have 22 large sample sizes. But actually, this is an BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 142 1 interesting issue for transportation in 2 particular, because when Mark did his work 3 with the National Household Travel Survey, 4 yes, we don't have the technical information 5 about vehicles we used to have. And that 6 really hurt us in deriving the energy 7 consumption and expenditures information for 8 technical reasons which I don't have time to 9 explain here. 10 But what it did give us in exchange 11 is a much larger sample size, because the 12 NHTS goes out to about 25,000, 24,000 13 households, whatever it is, nationwide, 14 whereas the old RTEC Survey went out to 3,000 15 households, 6,000 vehicles. 16 So you're talking about a sample 17 size eight times of what you used to have. 18 That sample size might be enough if 19 we were to attach, in the future, to the 20 survey and try to get more information built 21 into NHTS relevant to our needs. We could 22 borrow that sample size and maybe get more BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 143 1 information about emerging technologies as 2 they occur. 3 However, I'll get Howard stirred up 4 again. We've got the DOT folks talking to us 5 about supporting the next NHTS. And we'd 6 like to get VIN numbers from the vehicles and 7 other types of things. 8 And they're willing to do it for 9 seven figures, and so there you are. 10 SPEAKER: That's about how long a 11 VIN number is, right? 12 MR. FRENCH: So any other comments, 13 questions? 14 MR. EDMONDS: I'll just for a 15 response to your questions that I guess as a 16 general principle, as a user of data, I'd 17 like to be able to work from kind of these 18 macro aggregate statistics back down through 19 the services to technologies to as great a 20 degree as possible. So that in some sense is 21 the guiding principle that I would offer up 22 to the degree possible. You know, for BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 144 1 example, in the building sector, you'd like 2 to have shell characteristics, you'd like to 3 know how much is going into the HVAC systems, 4 and how much of it is heating, how much of it 5 is cooling? How much is lighting? How much 6 is cooking the usual suspects? 7 And I think your point about it, it 8 is difficult, but my recollection is that the 9 biggest, fastest -- the fastest-growing 10 component is "other." And your "other" is 11 growing like gangbusters, but it's not -- you 12 know, it's a whole package of these things 13 like microwaves, like the fact that people 14 will plug in their little recharger units and 15 leave them going day and night. 16 MR. FRENCH: Coffee makers. 17 MR. EDMONDS: The coffee makers, 18 all these little odds and ends that are 19 growing very rapidly. It's not one thing, 20 but it is a -- you know, it turns out to be a 21 big deal. 22 Over there in the industrial BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 145 1 sector, you'd kind of like to know how much 2 is going to raise steam, how much is -- you 3 know, is it process heat? You know, how much 4 is in motors? How much is in specialized 5 services like direct reduction? 6 And then in transportation, just 7 like you were saying, you'd like to be able 8 to trace back to what's the vehicle fleet and 9 what's the character of that vehicle fleet? 10 Because I think you're absolutely right, you 11 know? We haven't been in this spot for 100 12 years, where it isn't clear exactly what 13 technologies are going to dominate 30 years 14 from now. And I think that -- you know, it 15 is an exciting time in the transportation 16 area, where there's a lot of potential that's 17 out there that people are thinking about. 18 So that's my two cents. 19 MR. FRENCH: Okay. In fact, we do 20 model end uses in residential. And I'll say 21 loosely we model end uses in the commercial 22 area. However, the engineering-based model BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 146 1 that we used to have to model in the 2 commercial area is no longer supported and is 3 really out of date and out of commission. 4 And we're in the process of thinking about 5 approaches for developing a new 6 engineering-based model so that we can try to 7 deal with commercial end uses. But that's 8 something that's -- hopefully will be coming 9 in the relatively near future. We've got to 10 get that restarted, because you're right, 11 this is something that people are interested 12 in. 13 By the way, manufacturers supply 14 their own end uses: The processed heat 15 versus the steam versus the machine drive, et 16 cetera, et cetera. So we get that from the 17 manufacturers because they can estimate it 18 far better than we could ever measure it, 19 model it, or approximate it in any other way. 20 So yeah, we do have concerns. 21 Other than the consumption and 22 expenditures themselves, I would say the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 147 1 end-use breakdowns are probably the most 2 important energy-related statistics that we 3 put out, and have a great deal of call for 4 them. 5 SPEAKER: Go ahead. 6 MS. COX: Dwight? 7 MR. FRENCH: Yes. 8 MS. COX: I would say the hardest 9 thing for you is that what you really need is 10 a crystal ball, so that you could look ahead 11 in the next 5 or 10 years and say, okay, what 12 are going to be our pressing policy issues? 13 You know, what is there right now that might 14 suggest is coming out? And it may be more 15 than just levels of consumption, which is a 16 lot of what's focused on. It might be how 17 can we shift consumption from one energy 18 source to another, or from energy sources 19 that are less moving to energy sources 20 that -- in other words, how can we change 21 behavior? Not so much, yes, measuring 22 consumption, but also saying how can we BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 148 1 affect consumption? 2 MR. FRENCH: Interesting that you 3 mention a crystal ball. Part of our 4 difficulty in dealing with activities and 5 technologies and so forth as they change is 6 that we only run our surveys once every four 7 years. And so if we miss the mark or 8 something happens a couple of years later, we 9 can't catch up to it until sometime down the 10 road. 11 MS. COX: Five years is a long 12 time. 13 MR. FRENCH: Sometimes it isn't. 14 Sometimes things don't change that rapidly. 15 But you're right, there are times when things 16 do change rapidly. And for example, 17 transportation could change rapidly over the 18 next 5 to 10 years. Could. 19 Who knows really? 20 MR. HILL: (inaudible) Census data 21 with the "other" category and over time, the 22 "other" category becomes something more BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 149 1 specific. 2 MR. FRENCH: Yes. 3 MR. HILL: Race data is black, 4 white, other and now the Hispanic is now a 5 significant feature. You don't know in 1960 6 to sample -- to break Hispanics out of that 7 category. 8 MR. FRENCH: Yes. You mean -- 9 MR. HILL: But in some -- 10 MR. FRENCH: You mean a race/ethnic 11 background designation in our variable? 12 MR. HILL: Oh, no. Is "other" 13 growing as a portion? Is "other" becoming 14 more important over time in your residential 15 consumption data? 16 MR. FRENCH: Residential, 17 commercial, yes. Manufacturing, not so much 18 because it's always been process-dominated. 19 And there's some left over for the building 20 environment in some industries, but not all 21 that much. 22 But yes, "other" is becoming -- and BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 150 1 we have, especially in the residential area, 2 tried to give increasing information about 3 some of the components of that, maybe not by 4 modeling out specific pieces, but by 5 presenting in conjunction with our end-use 6 estimation information about different 7 appliances and from other engineering sources 8 how much the average appliance can consume, 9 so that analysts can start speculating and 10 modeling out the potential breakdowns within 11 that "other" category. 12 And perhaps that's something that 13 if we had more resources, we might want to 14 take a crack at, too, but there's limits to 15 everything. 16 MS. KHANNA: An overall question. 17 You said that household energy consumption 18 and transportation together account for about 19 70 to 75 percent of total U.S. energy 20 consumption. 21 MR. FRENCH: No, household, 22 commercial building, manufacturing, and BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 151 1 household transportation. 2 MS. KHANNA: Thank you. 3 MR. FRENCH: Together account for 4 70 percent or so. 5 MS. KHANNA: Okay. Is that share 6 over time changing, that combined aggregate 7 share, or does it stay roughly at about 8 70 percent? 9 MR. FRENCH: It's not changed that 10 much. 11 MS. KHANNA: Okay. 12 MR. FRENCH: Perhaps you have a 13 little relative loss in manufacturing and a 14 little relative pick up in commercial. 15 Transportation has stayed remarkably stable 16 even though the number of vehicles has 17 increased. Our transportation energy use 18 hasn't gone up that much, because efficiency 19 gains have sort of offset increased usage and 20 some structural shifts, as it were, so that 21 we're really pretty stable in that. 22 And household is the same way, by BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 152 1 the way. Household energy use has been 2 really remarkably stable considering you've 3 got 30 million additional housing units 4 relative to a quarter of a century ago, and 5 all these appliances that everybody's 6 complaining about that drain electricity and 7 so forth. But the building shell and better 8 windows and more efficient furnaces, et 9 cetera, have helped to mostly counteract 10 that. 11 MS. WAUGH: One of the issues you 12 raised is that cost is important. And I 13 think (inaudible) a three-year cycle to a 14 four-year cycle -- 15 MR. FRENCH: Back in the mid-'90s, 16 yes. 17 MS. WAUGH: Is there any 18 possibility of either going to a five- or 19 six-year cycle and having some kind of 20 mini-survey in between? For example, on the 21 MECS, you might just do this mini-survey of 22 the NAICS codes that are energy-intensive BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 153 1 consumers. 2 MR. FRENCH: Yes, this is another 3 way of expanding yourself by limiting 4 yourself, as it were. 5 And it -- there's been some slight 6 thought given to it. There -- anytime you 7 think about doing this, it's not just 8 aggregate dollars, it is when the resources 9 are required to do something like this -- and 10 do you back up something first and then use 11 those resources to create something in the 12 middle, or do you try to get extra money to 13 do something in the middle first and then 14 back up later? Because you see, it takes a 15 while. We don't operate anymore on funding 16 our surveys by the time that the survey is 17 actually out in the field. We are much 18 closer to just-in-time funding. 19 And so if you were to try to insert 20 something between a MECS survey -- and this 21 is the place where it's been discussed more 22 than any other place -- mini-MECS, as it BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 154 1 might be called. 2 SPEAKER: I think of Austin Powers. 3 MR. FRENCH: If you did something 4 two years from now, I wouldn't necessarily 5 have the money to do it. But if I delayed 6 the next MECS a little bit and then did a 7 mini-MECS after that, the money that I saved, 8 I might have money to do something. 9 MS. WAUGH: And if you did try that 10 and it was successful, you might actually 11 consider it as a model for the other two if 12 it also shows to be more cost-effective and 13 useful. 14 MR. FRENCH: Yes, and I guess the 15 appropriate thing then would be that we would 16 only go to the millionaires' houses in the 17 mini-RECS or the federal office buildings who 18 are obviously the biggest energy hogs 19 among -- actually that's not true, of course. 20 The biggest users among commercial buildings, 21 per floor space, are hospitals and 22 supermarkets and restaurants, all for their BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 155 1 own reasons. 2 MS. WAUGH: Well, maybe another way 3 to look at this is which data in the 4 publications the users are the most 5 interested in as opposed to which ones are 6 the most energy -- 7 MR. FRENCH: An interesting 8 philosophy, but if you are interested in 9 collecting just certain pieces of information 10 because those are the data that you're 11 interested in, but you don't limit your 12 population that you're trying to cover, that 13 has implications for sample and it has 14 implications for overall costs. The beauty 15 of doing a mini-MECS in certain industries is 16 that you can limit your sample size and you 17 can limit then perhaps your cost to some 18 great extent; much more difficult to do in 19 the residential or commercial areas. 20 There also has been a suggestion to 21 go to continuous sampling, which could be a 22 consideration in the residential and/or BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 156 1 commercial sectors. 2 SPEAKER: That would be 3 interesting. 4 MR. FRENCH: Well, it's 5 interesting, but there are a lot of 6 logistical challenges to that. 7 MS. WAUGH: I have -- 8 MR. FRENCH: Yes, I know, and she's 9 just itching for something like that. 10 MS. WAUGH: Oh, I'd love that. No, 11 I was thinking they do that for the 12 government. 13 MR. FEDER: How would you do it 14 continuously? Would you move the geographic 15 focus over time? 16 MR. FRENCH: Well, that's the 17 question, because our surveys are area-based 18 in residential and commercial. So do you do 19 one area of the country each year, in which 20 case you may have a problem with time biases 21 correlated with regional emphases from year 22 to year, or do you do a national subsample BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 157 1 every year, which of course then inflates 2 your field costs because you're covering the 3 whole country every year even though it's a 4 smaller number of observations? 5 MR. FEDER: Well, there's two 6 issues. I know the French are moving towards 7 a rotating census of population, but it 8 depends on the granularity of your geographic 9 aggregation. You can still move your sample 10 around, but do it in a way that you cover at 11 least the regions every time. 12 MR. FRENCH: When you say "the 13 regions," you mean like the four Census 14 regions? 15 MR. FEDER: The U.S. is divided 16 into nine regions or four regions, depending 17 on how you disaggregate the country. So you 18 cover the Atlantic Coast, Pacific Coast, the 19 mountains, but not necessarily every state 20 every year. 21 MR. FRENCH: And of course, we only 22 go down -- except for some estimates for the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 158 1 four largest states for RECS, we only go down 2 to the Census Division level on any of our 3 data now. So I mean, we could think about 4 going to the regional level or even just 5 doing national estimates from year to year. 6 MR. FEDER: Then you would have 7 this confounding between original changes 8 and -- 9 MR. FRENCH: Well, if you're saying 10 stick to particular regions in a year. I was 11 thinking what you meant was slim down your 12 number of primary sampling units, which is 13 what we use. We use an area sampling 14 approach as the basis for our surveys in 15 residential and commercial. Slim down the 16 primary units so you have less accuracy, but 17 you offset that by not estimating it as low a 18 final level of geographic detail. 19 MR. FEDER: I was thinking of an 20 approach where each year, you sample like a 21 high level of aggregation from different 22 areas, but you move it around from year to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 159 1 year, so hopefully, you break some of the 2 potential confounding between your sampling 3 design and changes in trends that are 4 regional. 5 MS. COX: That's an interesting 6 idea. One of the things that the consumption 7 survey faces is that there's this 8 extraordinary variation due to -- it's not 9 really geography so much as climate 10 conditions and things like that. But you 11 need such penetration to get a full 12 understanding, and RECS kind of pushes it to 13 the limit. By that I mean -- 14 MR. FRENCH: 4000 to 5,000 15 households isn't enough to -- 16 MS. COX: Yes, that's kind of what 17 I was going to -- you really -- but you kind 18 of understand why. Because you have this 19 real need for this fine level, and that's 20 expensive for a household survey. But the 21 way -- a continuous survey sounds interesting 22 to me, you would -- to make it feasible, what BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 160 1 you would have to -- it would have to be 2 whole PSUs that were being rotated. In other 3 words, the expense is more attached to the 4 PSU, so it's an interesting thought of having 5 a survey that in order to do your fine level 6 of detail, you might have to combine over 7 three years. 8 MR. FRENCH: Yes, there is this 9 idea that if you could get a continuous 10 measurement approach started, what you would 11 end up with is an average over four years 12 that you would start publishing after four 13 years, if it were on a four-year rotating 14 cycle. And you would then have a rotating 15 measure every year rather than a snapshot 16 every four years, which is what we do now. 17 There are, as I say, logistics 18 issues to be considered. There are money 19 issues to be considered. When do you start 20 this? You know, if we were really going to 21 start something like this, for example, in 22 RECS, we couldn't really start it until the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 161 1 year before we would start a RECS anyway, 2 because it would take that much time for us 3 to collect enough money to do that. So we 4 would not have a full RECS comparable 5 estimate for seven years, until you build up 6 four years' worth of data and you can 7 estimate. 8 MS. FORSYTH: I'm not sure of the 9 sources of the cost savings from a continuous 10 sample as opposed to every seven years. 11 MR. FRENCH: Every four years. 12 MS. FORSYTH: Is it because you 13 could use -- I mean, I'm sorry, already I'm 14 pushing it out. Is it because you have a 15 constantly trained staff so you don't have to 16 worry about those, like, ramp-up costs? Or 17 what are the efficiencies? 18 MS. COX: Well, I wasn't pushing it 19 for cost savings. 20 MS. FORSYTH: Precision. 21 SPEAKER: No, like information -- 22 MS. FORSYTH: Right, okay. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 162 1 MR. FRENCH: I don't necessarily 2 know that you would have a cost savings when 3 you look at a per four-year, per current 4 cycle basis. What I'm trying to do is figure 5 out how you can keep the costs down to the 6 per four-year cycle basis and not have more 7 because of the need to train interviewers in 8 all parts of the country every year. 9 MS. FORSYTH: Right, right. 10 MR. FRENCH: Perhaps you can save 11 some on training costs if you do have year 12 after year and you have some overlap of 13 interviewers. As long as your questionnaire 14 doesn't change all that much from year to 15 year, you can get by with a minimum of 16 training for interviewers and maybe save some 17 money that way. 18 MS. FORSYTH: Well, there are lots 19 of other ramping up kinds of costs. 20 MR. FRENCH: Yes. 21 MS. COX: The in-house cost, I 22 would think that at some point, you would BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 163 1 reach a deduction level. In other words, the 2 next year, you're just repeating what you did 3 the previous year, so that there would be a 4 cost savings from that extent. 5 MS. FORSYTH: But it's also a 6 constraint. You can't address emerging 7 things very well except in the off years. 8 MS. COX: Well, you can't address 9 emerging things very well -- 10 MS. FORSYTH: Anyway. 11 MS. COX: Every four years if they 12 emerge the previous three years. So in a 13 way, it does give you more of a sense of 14 emerging things, because you would 15 have -- emerging things often just don't 16 happen this year -- sometimes they do -- but 17 a lot of times they kind of creep up on you 18 and all of a sudden you realize we got a 19 problem. 20 MR. FRENCH: I'm behind the 8-ball, 21 yes. 22 MR. FEDER: I'd like to give an BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 164 1 example from my Canadian experience with an 2 entire different survey. It was the 3 International Travel Survey. And what we 4 wanted to do, we sampled from different ports 5 of entry and sampled in practically every 6 time of the year. 7 So what do you do? You cannot 8 sample in every port of entry every day, so 9 it was staggered across time and across 10 ports. So every month or so, there was some 11 sampling in some port, but it was done using 12 some way so you had a good distribution 13 across ports and across months, but not 14 combined, because that was too much. 15 So you had a good way to capture 16 the seasonality in travel and to capture the 17 variation across ports, but of course, there 18 was the odd chance that maybe people somehow 19 tricked you and entered through ports when 20 you were not sampling, but that's not likely 21 right. So that was the approach. 22 And maybe it's the same idea behind BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 165 1 continuous sampling here is to have a good 2 idea about changes across time without having 3 to sample everybody every year. 4 MR. FRENCH: We're about at 10 of 5 12:00, and I know that was what you were 6 aiming for. 7 MR. HILL: Yes, it looks like some 8 of our colleagues are drifting in, so maybe 9 the other group also has finished. Is there 10 a last question out there? And in fact, we 11 haven't had anything from the audience. 12 Go ahead. 13 MR. FRENCH: By the way, this is 14 our relatively new manager of the Residential 15 Energy Consumption Survey, Eileen, Eileen 16 O'Brien. 17 MS. O'BRIEN: I'm fairly ignorant 18 on energy topics, but I did attend the recent 19 ACEEE conference. And what I heard there was 20 a lot of interest in barriers to adoption of 21 new emerging technologies. We measure 22 characteristics of the households, some BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 166 1 behavioral questions. We don't measure 2 attitudes which are impediments. Do you see 3 any role for measuring attitudes which shape 4 shifts in transportation choices or household 5 technologies and so on? 6 MR. FRENCH: Are you asking the 7 committee or are you asking me? 8 MS. O'BRIEN: I'm asking the 9 committee because -- I mean, I don't know if 10 this would be something we would throw on a 11 survey if it weren't going to help in the 12 modeling of end uses or address some of the 13 adoption issues that seem to be of interest 14 to the energy sector. 15 MR. HILL: Go ahead. 16 MS. KHANNA: Just a reaction to 17 that. I think trying to elicit data on 18 attitudes is a whole different ballgame than 19 trying to elicit data on actions. And the 20 kind of data that I think EIA traditionally 21 collects is what do you do, and that's the 22 question. That's looking at just -- it's BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 167 1 actually reporting fact, whereas reporting 2 attitude is try to figure out what would you 3 do or why don't you do it, but that's a 4 completely different kind of survey issues, a 5 completely different kind of data collection 6 issue. And I'm not sure you want to actually 7 combine the two. I think attitude is one of 8 those places where maybe what the Review 9 Committee said looking at market information 10 or others' data that are available out there, 11 that may be something that EIA could draw on. 12 MR. GRUENSPECHT: Sort of following 13 on this and maybe a little bit on that, I'm 14 sort of curious. In lighting, do we go in 15 and ask people if they have compact 16 fluorescent lights? I don't know if most 17 people know what they were, but you could 18 imagine you could bring a regular one and 19 whatever, and say do you have any of these in 20 your house, you know? 21 SPEAKER: Yes, there's one. 22 MR. FRENCH: Yes, there is some BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 168 1 information collected on -- 2 MR. GRUENSPECHT: There is on that. 3 MR. FRENCH: Right. 4 MR. GRUENSPECHT: So you're trying 5 on penetration and trying to figure out, you 6 know -- 7 MR. FRENCH: Well, not a whole lot 8 on how many do you have or what proportion of 9 your lighting is in -- 10 MR. GRUENSPECHT: No one would know 11 that, what portion of the bill, right. 12 MR. FRENCH: Well, but that -- I 13 mean, if you really want to figure out what 14 the penetration is, it isn't just the number 15 of bulbs. I have quite a few bulbs in my 16 house that are compact fluorescent. A lot of 17 them are in my basement. You know how often 18 I turn the lights on in my basement? 19 Oh, I'm getting older -- 20 MR. GRUENSPECHT: It's exactly the 21 opposite. The economics says you put them in 22 the ones that you use most frequently. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 169 1 MR. FRENCH: Ah, but -- 2 MR. GRUENSPECHT: And in your 3 closet, you add a closet, you leave the 4 incandescent bulb because you don't turn that 5 one on, so it's good to know that you are 6 balancing out -- 7 MR. FRENCH: As a consumer, I will 8 tell you that I don't think that 9 understanding is correct. 10 MR. GRUENSPECHT: All right. 11 MR. FRENCH: I think that one of 12 the big reasons that people use compact 13 fluorescent bulbs -- 14 MR. GRUENSPECHT: Is they don't 15 want to change them. 16 MR. FRENCH: They don't want to 17 change them and they put them in applications 18 where access is -- 19 MR. GRUENSPECHT: Where they don't 20 care about the quality of light, okay. All 21 right. I'm going to start this again. But, 22 I mean, these kinds of issues I think are, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 170 1 you know -- I mean, whether you actually ask 2 people their attitude toward new technologies 3 or -- 4 MR. FRENCH: And by the way, that's 5 where attitude intersects with behavior. 6 MS. KHANNA: Right. 7 MR. FRENCH: In something like 8 that, which is an interesting -- 9 MS. FORSYTH: And it's all 10 self-reported; right? 11 SPEAKER: I'm sorry, Barbara? 12 MS. FORSYTH: Well, I was going to 13 say it's all self-report. 14 SPEAKER: Yes, essentially. 15 MS. FORSYTH: So the differences 16 between attitude and behavior are less in a 17 self-report survey than they are when you 18 have records that you go to. They all have 19 the same kinds of errors. 20 MS. KHANNA: But I mean, as 21 economists, we can sort of elicit attitudes 22 from market behavior. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 171 1 MS. FORSYTH: Right, right. 2 MS. KHANNA: Why do we need to go 3 out and actually ask about attitudes? 4 MR. CLEVELAND: Oh, you sound like 5 an economist. 6 MS. FORSYTH: Yes. 7 MS. KHANNA: I rest my case. 8 MR. HILL: Yes, we're getting hints 9 that we should stop at this point even though 10 the discussion -- 11 MR. FRENCH: Again, my thanks to 12 the committee for all their work with me over 13 the years. 14 MR. HILL: Let me adjourn for five 15 minutes or so. 16 MR. FRENCH: You don't have to put 17 up with me anymore I guess is the statement. 18 SPEAKER: Won't have Dwight French 19 to kick around anymore. 20 MR. FRENCH: Kick around anymore, 21 that's right. Thank you. 22 MR. HILL: Thank you. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 172 1 SPEAKER: Are you a crook? 2 (Recess) 3 MR. HENGARTNER: Okay. I know we 4 still have a few people missing out there 5 grabbing coffee. 6 My experience is I can be late for 7 any session except for lunch. 8 SPEAKER: So we're done. 9 SPEAKER: That's my kind of leader. 10 MR. HENGARTNER: So we'll try to 11 wrap it up and we have a deadline for 12:15, 12 if we can. If we're five minutes late, so be 13 it, but we're going to get there. 14 So I'd like to invite the comments 15 and more or less summaries from the different 16 breakout sessions. And I'm going to do it in 17 reverse order and I'm going to ask first for 18 Mark Burton to talk to us, what we heard 19 downstairs in the Using Models to Detect 20 Outliers in Refinery Data. Mark. 21 MR. BURTON: Okay, I have, like, 22 two pages' worth of notes that go about 15 BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 173 1 different directions, and thank you for 2 giving me the chance to look really silly 3 first. 4 MS. FORSYTH: Don't worry, I'll 5 make up for it. 6 MR. BURTON: Sort of the basis for 7 the work, and I think that this is something 8 that is really, really important, and Phillip 9 did a very good job of saying it, and that is 10 that the work that EIA does matters. People 11 pay attention to it. 12 They look at the various products, 13 particularly the short-earned (?) products, 14 and those are very influential in terms of 15 behavior. So for them to understand the data 16 as well as they can possibly understand them 17 is a very important thing. And I think I 18 heard Nancy say that, too. It really -- it 19 matters. They need to understand the data. 20 The exercise then that they 21 described was an attempt to do that. And 22 specifically, it was an attempt to identify BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 174 1 outliers in refinery gain measures and in 2 gasoline production measures. And it was 3 based on some regression equations, and they 4 sort of parsed the data three different ways 5 in a temporal sense to try and see whether or 6 not that yielded any substantial differences 7 in the regression results. 8 Frankly, and a number of us said it 9 a number of different ways, there are things 10 going on that are affecting those data that 11 are not currently part of the model 12 specification. There are I think clearly 13 some specification issues, particularly in 14 the refinery gain equation, that have to be 15 addressed. 16 That having been said, I mean, you 17 have to start someplace, and I wish I had a 18 dollar for every time I have gone through and 19 gone through the painful process of looking 20 at something and going, ew, that's -- just 21 something's -- they had some co-efficient 22 signs that were reverse of what they should BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 175 1 have been. That probably, given the way the 2 models were specified, that probably is an 3 indication, as Nick said, of 4 multi-planarity (?) as opposed to any sort of 5 real causal anomaly, so that there are 6 specification issues. There are probably 7 problems with multi-planarity. Also, just 8 looking at the residuals, there are signs of 9 auto-correlation that has not been treated. 10 So basically what they're trying to 11 do is something that is very useful. It's 12 the sort of thing that EIA should be doing. 13 They've taken a couple of steps down what is 14 likely to be a very -- considerably longer 15 road. 16 They asked the committee really a 17 couple of questions, but the most essential 18 one was: Is this something that we can do 19 with these data? And, I mean, we all 20 probably as a group have somewhat differing 21 opinions on it. For my part, keep after it. 22 You know, maybe it's there, it's just not BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 176 1 there yet. 2 Are there other committee -- people 3 that were downstairs in the breakout have 4 things they want to add to that? 5 MR. HENGARTNER: Okay, thank you 6 very much, Mark. I'd like now to invite 7 Barbara Forsyth to report on the Needs and 8 Priorities for Consumption Information in the 9 Future. 10 MS. FORSYTH: Being a big energy 11 user myself, well, what we got was a 12 presentation of the consumption area surveys, 13 kind of the dimensions around which the 14 surveys are organized and the kinds of areas 15 they are able to cover and the kinds of areas 16 they are unable to cover, and some of the 17 reasons for the constraints in the current 18 programs. 19 And the questions for the committee 20 were: Which areas seem most important? 21 Could we give guidance on where to make 22 tradeoffs? Where, if we want to build data BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 177 1 collection or enhance data collection, would 2 we trade off reduced data collection? And 3 how to meet program priorities making those 4 trade offs. So those are two questions. 5 And then the third was, can we 6 begin to articulate or help them articulate 7 some criteria for choices, making choices 8 among programs, either within the consumption 9 area or possibly across -- I guess program 10 areas in EIA? 11 And so we kind of talked around all 12 of those things. I think we started off 13 talking about criteria for making trade offs. 14 And again, if I missed something, someone 15 will add to it, I hope. But I think some of 16 the criteria we identified were focusing on 17 areas where there's big potential for change 18 and identifying predictors of change, like 19 policy issues or likely policy issues or 20 emerging policy issues. And we identified 21 some specific areas where there is potential 22 for change. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 178 1 In the residential sector, I think 2 we said that "other" is the biggest source of 3 change. And we're not sure what that other 4 is, but there's lots of change there, there's 5 lots of growth there. And transportation is 6 another because of the potential for change 7 in the, I guess, relatively near-term. 8 So I think those are some criteria 9 we identified for prioritizing areas, areas 10 to cover. I don't think we identified any 11 criteria for areas not to cover. 12 In terms of -- let's see, making 13 trade offs, one thing we talked about was 14 trading off the time between data collections 15 with the nature of the data collections. So 16 if we could expand the time between data 17 collections, we might be able to plunk in a 18 special study that might be able to address 19 some of these -- where are the areas for 20 change or where are the policy issues likely 21 to emerge. So there might be some chance for 22 trading off, for a fixed monetary amount, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 179 1 long-term data collections versus specialized 2 short-term data collections. So that's one 3 place where there might be a potential for 4 trading off. 5 And a different kind of trade off 6 we talked about was perhaps drawing on 7 continuous samples rather than one-time 8 longitudinal -- I mean -- 9 MR. FEDER: Periodic. 10 MS. FORSYTH: Yes, periodic 11 samples. And that way, we could get seasonal 12 changes or perhaps more detail about regional 13 variations -- not more detail -- but I guess 14 more specific information, maybe less 15 precise, but more specific information about 16 regional variations as well. And over a 17 longer time frame, maybe instead of four 18 years, maybe seven years, we could get a 19 total representation as well. So that's one 20 kind of trade off, the detailed information 21 versus the time. 22 And I guess related to those BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 180 1 mini-surveys that we talked about planting in 2 between, perhaps extended periods between 3 surveys, that using that as a model perhaps 4 in one sector to see how successful it was 5 might provide impetus for moving it across 6 sectors or perhaps across -- not sectors, but 7 energy sources. I guess we already gather 8 across energy sources -- so perhaps testing 9 it in a small area and then expanding it 10 after proving or -- I guess after proving its 11 worth would be a way of doing it perhaps in a 12 relatively low-cost way, doing some 13 experiments early on. 14 I think that's what I heard, but, I 15 mean, I don't know anything about energy and 16 I'm not really a statistician either, so. 17 MR. HENGARTNER: Oh, I see, so 18 you're just goofing around. 19 MS. FORSYTH: I meant it. I really 20 meant it. 21 MR. HENGARTNER: Anybody else wants 22 to add to either this discussion that Barbara BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 181 1 just summarized or the previous one? Good. 2 I'm taking up your time, I'm sorry. I'm 3 blamed. 4 At this time, I'd like to invite 5 comments from the general audience and 6 questions, if any there are. Good. 7 MR. CLEVELAND: Now we're going to 8 finish early. 9 MR. HENGARTNER: No, you guys, 10 that's what they call efficiency. Before we 11 break, though, we have very important 12 business. Namely, we have to decide who's 13 going for dinner, right? So I'd like to have 14 a show of hands so that Bill can make 15 reservations for tonight. 16 Where are we going, Bill? 17 MR. WEINIG: We're going to the Old 18 Ebbitt on 15th Street, a very nice place. 19 SPEAKER: Very fancy place. 20 MR. HENGARTNER: The Old Abbey. 21 SPEAKER: Ebbitt. 22 MR. HENGARTNER: Ebbitt. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 182 1 SPEAKER: Old Ebbitt Grill. 2 SPEAKER: It's a good restaurant. 3 SPEAKER: It's very good. 4 SPEAKER: Yea. 5 MR. HENGARTNER: Okay, thank you 6 very much. 7 MR. WEINIG: Reservations at 6:00 8 at the Old Ebbitt in the atrium, which is 9 noted to be a less noisy place. 10 SPEAKER: (inaudible) we get loud. 11 MR. HENGARTNER: And will we have 12 directions on how to go there? 13 MR. WEINIG: Yes. I would 14 recommend either the Metro Center and exit 15 west. 16 SPEAKER: On 13th Street. 17 MR. WEINIG: On 13th and then walk 18 two blocks to 15th and one block down to 675 19 on 15th Street. 20 SPEAKER: It's right across from 21 Treasury. 22 MR. WEINIG: For those of you who BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 183 1 know Washington well, it's right across the 2 street -- on the other side of 15th Street is 3 the Treasury Building. And this is where 4 Pennsylvania Avenue does its dog leg to go 5 from the Marriott Hotel up to the White 6 House, depending on who's in favor in town. 7 MR. HENGARTNER: Okay, thank you 8 very much. And I just want to remind you 9 that this is the only vote the committee does 10 every year or every meeting. Good. So if 11 there's no more comments or discussions, I'd 12 like to adjourn for lunch. The committee has 13 lunch downstairs on the first floor. I think 14 we all know where it is. 15 SPEAKER: Yes, we do. 16 SPEAKER: Same place. 17 MR. HENGARTNER: Same place as 18 usual. And we will reconvene in this room 19 for further proceedings at 1:15. 20 Thank you very much. 21 (Whereupon, at 12:10 p.m., a 22 luncheon recess was taken.) BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 184 1 A F T E R N O O N S E S S I O N 2 (1:25 p.m.) 3 MR. HENGARTNER: Let's start to 4 resume our proceedings. I'd like to 5 introduce to you Robert Schnapp. Make sure I 6 pronounce it right. I apologize if I messed 7 it up. 8 He's going to be talking to us 9 about the electricity and 2008, and in 10 particular about what the broad plans are for 11 the electricity review. So welcome, Robert. 12 MR. SCHNAPP: Thank you, I 13 appreciate it. Lawrence, do you want to put 14 the slides up there? 15 All right, thank you very much. My 16 name is Bob Schnapp. I'm the director of the 17 Electric Power Division in the Energy 18 Information Administration. And today I'd 19 like to talk to you about our project for 20 renewing our electric power data collection 21 forms. 22 Every three years, all the surveys BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 185 1 have to be sent to the Office of Management 2 and Budget for review and clearance. On the 3 electric side, we have taken a very stringent 4 view of our forms because of the many changes 5 that have occurred in the industry over the 6 past 10 years. 7 And so we started a project, first 8 of all, Electricity 2002, where we literally 9 threw away all of our forms, all of our 10 surveys, all of our publications, and all of 11 our processing systems, and started from 12 scratch. And so for January 1 of 2002, we 13 had all new systems and all new forms. 14 Then we had a projected called 15 Electricity 2005, where we kind of changed 16 the surveys a little bit around the edges and 17 instructions and the confidentiality policy. 18 And this year, we now have Electricity 2008. 19 And what we plan to do here is to merge some 20 forms. So I'll go into this in a little more 21 detail now. 22 So this is a broad overview of what BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 186 1 our project is. I'll just -- I'll talk loud. 2 The first phase is data requirements. We 3 have to take a look at the forms and see what 4 issues we need to evaluate, and I'll go over 5 those in another slide here. 6 Then we want to talk with all of 7 our stakeholders and figure out are there 8 things that we need to change for them? Are 9 there comments that they have? So we talk 10 with the electric power industry, combined 11 heat and power plant industry, consumer 12 groups, environmental groups, other federal 13 government agencies to see what's required. 14 Then we're also going to take a 15 look at our confidentiality policy and have a 16 couple slides on that. And then we would 17 issue a Federal Register notice in February 18 of 2007 that would be our proposed changes, 19 and that's out for 60 days for comment. 20 Phase 2 is the forms design, and 21 this is actually going along at the same time 22 as Phase 1 is. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 187 1 And that's where we would take a 2 look at the forms and see what we need to 3 consolidate, what revisions we need to make. 4 And then when we come up with those 5 decisions, we would do some pilot testing out 6 in the field and see if these forms are going 7 to work. 8 And then after all that's done and 9 we would evaluate the comments we got on the 10 first Federal Register notice, we would then 11 go to the Office of Management and Budget in 12 September of 2007 to get approval from them 13 by November of 2007. 14 Other things that we have to do 15 along the way is in Phase 3, which is 16 collecting and processing. And so we will 17 have to modify our imputation procedures, and 18 Bob Rutchik actually will talk a little bit 19 more about that. We'll have to modify our 20 data processing systems obviously to accept 21 different data, and then our Internet data 22 collection system, and I'll talk a little bit BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 188 1 more about that as well. And the idea here 2 is that we would have everything operational 3 by the end of December 2007, so that 4 January 1 of 2008, we would be ready. Hence 5 the name of the project, Electricity 2008. 6 Phase 4, again, work on this will 7 be going on while we're in the other phases, 8 but we would be modifying our 9 publications -- well, we don't have 10 publications anymore, dissemination products. 11 So we have an Electric Power Annual, Electric 12 Power Monthly. 13 We have detailed Excel spreadsheets 14 that we put out on the Internet that are at 15 the state level with lots of cuts by fuel 16 type and producer and owners. And then we 17 have the detailed databases, which is the 18 actual data that we collect that we're 19 allowed to give out to the public, so it's at 20 the individual facility level. 21 And then we have a couple of other 22 products, State Electricity Profiles. This BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 189 1 is nine tables that are about each individual 2 state that we have on the Internet. And then 3 Cost and Quality, which is -- particularly 4 hones in on two of the forms, the FERC-423 5 and the EIA-423 forms. And we would have to 6 have all this done by June of 2008, so that 7 we would be ready to put our reports out. 8 Some of them actually have to be done before 9 June 2008, but that's kind of the end date 10 really. 11 And so the key areas that we are 12 focusing on this year, the first one is 13 merging the EIA-906, 920, and EIA-423. Just 14 to summarize, the 906 collects information 15 from utilities and IPPs, or independent power 16 producers, that are not combined heat and 17 power plants, and so that collects 18 information on fuel consumption, generation, 19 and stocks. 20 The EIA-423 collects information on 21 fuel receipts from the non-utilities. There 22 actually is another form that the Federal BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 190 1 Energy Regulatory Commission has, the 2 FERC-423, that's very similar to our EIA-423, 3 and they collect the same information from 4 utilities. And the EIA-920 actually also 5 collects the same thing as the 906, but 6 that's from combined heat and power plants. 7 So we're talking about merging all 8 those forms together so that we would have 9 information on ending stocks, receipts, 10 consumption, and then stocks at the end of 11 that period. And Bob -- I don't want to talk 12 too much about that. Bob will get into more 13 of the details on that. But we think that we 14 might be able to combine all these forms. 15 And last week, we understand that 16 FERC issued a Federal Register notice asking 17 for their FERC-423 form to be renewed for 18 three years, and they were asking comments on 19 it. And two of the questions they asked in 20 there was what if EIA takes over this form? 21 What would be the impact on you? 22 So we have been talking with them BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 191 1 about the possibility of that, but it's 2 certainly not written in stone at this point. 3 We don't know where we're going. From our 4 point of view, we think it'd be really good 5 because we then merge all the utilities and 6 the non-utilities into one form, and we think 7 that it would really save us resources and 8 improve the quality of the data. 9 But one of the things that we're 10 considering is the combined heat and power 11 plants, the industrial and commercial plants, 12 account for 5 percent of generation and they 13 take up 1/3 of the resources because most of 14 them are small. So we're thinking about not 15 collecting any monthly data from them, but 16 collecting it from their annual data once. 17 And as we go through the year, we 18 would impute for them, so we think that would 19 save a lot of money as well. 20 The second major issue that we're 21 talking about is merging the EIA-860, which 22 collects characteristics of a fuel plant. In BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 192 1 other words, what's their capacity, the type 2 of fuel that they would be using, what their 3 plans are for the next five years in building 4 or changing their facilities, combining that 5 with the static information on the EIA-767. 6 The 767 collects environmental 7 information. So if we pick up what 8 type -- what are your scrubbers, what are 9 their capacity, where are they located, we 10 could put that onto the 860 and possibly drop 11 the operational information. That's 12 something that we're still trying to figure 13 out if we should do from the user point of 14 view. We understand there's a lot of people 15 that use the 767, but we're trying to kind of 16 gauge what that is. 17 Also, and another way of saving 18 resources, we think that we might be able to 19 save a lot of resources by looking at the 20 sample. Right now, the 860 goes to -- I 21 think it's 2,800 respondents. About 700 or 22 so of them account for over 90 percent of the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 193 1 capacity. So if we took the remaining 2,100 2 and split them up into three years, we would 3 then have 1,400 respondents each year for 4 each three years, and we would go to all of 5 them at least once over that time. There 6 would be certain little variations on who we 7 would pick each time, but we think we could 8 do that and save money as well. And then 9 we'd also actually have kind of a constant 10 burden on the contractor over time as well, 11 so it wouldn't be going up and down. 12 Then we'd have to go and clarify 13 our instructions. We've been having many 14 meetings going over each individual form, 15 whether we need to change the instructions or 16 modify the question a little bit. We would 17 assess our confidentiality policy, and then 18 we would have to go and improve our online 19 edits and quality assurance reports. 20 And in doing this, we have support, 21 particularly from the Statistics and Methods 22 Group, who are helping us a lot this year BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 194 1 around hiring a contractor. They have a lot 2 of folks that are going to be able to help 3 us. And so they are the ones that are 4 designing the merged forms, and they're going 5 to be presenting to us what -- draft copies 6 of the hard copy forms and then what they 7 might look on the Internet. Right now our 8 hard copy forms and our Internet forms look 9 the same. What I think they may be doing is 10 moving more toward kind of a tax return way, 11 where you're asked questions and it fills the 12 data in for you and moves it wherever it has 13 to go. So we'll be seeing some of those 14 designs in the next few weeks here. 15 They'll also be going out and 16 testing those forms in the field for us. And 17 they're also providing the funding and the 18 oversight for -- I believe it's five 19 independent expert reviews. And what they're 20 reviewing is there's a set of data that we do 21 not publish in any summary form in any of our 22 publications, but we do release the data at BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 195 1 the individual facility level. So we're 2 asking them is there other data that we don't 3 publish at the summary level that we should 4 be publishing and how would we do that? 5 They're also going to be giving us feedback 6 on what the merged forms might look like. 7 And the last thing is the combined 8 heat and power methodology. We've briefed 9 the committee here several times on the 10 problem of splitting the fuel that's used by 11 combined heat and power plants between fuel 12 for electricity generation and fuel for use 13 of thermal output. And we've redesigned the 14 form. The form hasn't been as successful as 15 we had hoped, so we're asking some experts 16 for what their opinion is on how to proceed 17 from here. 18 So actually, we've gotten two of 19 the reports so far and we're having a meeting 20 with them onsite on Tuesday next week, so 21 we're looking forward to the rest of that 22 input. Anyways, the excellent Statistics and BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 196 1 Methods Group is providing support for us on 2 that. 3 Where is Stan? He told me to say 4 that? You'll tell him I said that, right? 5 MS. KIRKENDALL: I'll tell him. 6 MR. SCHNAPP: Okay. On 7 confidential data, right now, we have really 8 a very limited amount of data that we hold 9 confidential as opposed to most of the rest 10 of the EIA whose data is confidential. Most 11 of this is on the price side, not all of it, 12 but a good portion of it is. 13 The first is the fuel costs of fuel 14 delivered to non-utilities. Then there's the 15 fuel stocks on the 906. Actually on the 920, 16 also. Monthly data for the energy service 17 providers, only we would hold their retail 18 sales, revenue, and number of customers. But 19 what's interesting is that at the 20 annual -- for an annual report from these 21 same respondents, that information is not 22 confidential, and I'll address that in just a BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 197 1 minute here. 2 Latitude and longitude, which is of 3 the facility and then of the scrubbers on our 4 two forms, that's confidential, and then the 5 maximum tested heat under full load 6 conditions. 7 MR. RUTHERFORD: You give the 8 county location in the EIA-860. Doesn't it 9 tell the county that the plant's located in? 10 MR. SCHNAPP: Yes, yes. So what 11 we're talking about is, dropping down to the 12 bottom, is to remove the confidentiality on 13 latitude and longitude. That information's 14 in the public domain. 15 SPEAKER: Yes. 16 MR. SCHNAPP: We had withheld it. 17 We started withholding it about 2000. And 18 when we -- for our Electricity 2002 project, 19 we submitted our proposal to the Office of 20 Management and Budget that would have that 21 not confidential on September 11, 2001. And 22 so we called them on I think it was September BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 198 1 13th or 14th, and said do you want us to make 2 that confidential? And they called back and 3 said yes. So we made that confidential then. 4 For Electricity 2005, we thought we 5 would continue to hold it confidential since 6 we didn't want to give a convenient list to 7 terrorists to find all the information. But 8 at this point the information is just so out 9 there that we just decided there was no 10 reason to withhold it. So that will probably 11 be one of our recommendations. 12 SPEAKER: Perhaps they'll Google 13 it. 14 MR. SCHNAPP: We're proposing, 15 though, for most of the other items, the 16 three out of the four remaining items, that 17 we would release that data nine months after 18 the end of the reporting year. We had 19 proposed that actually to OMB for Electricity 20 2005 and they turned us down, so we're going 21 to try and make a stronger case this year. 22 MR. HILL: Was anybody worried BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 199 1 about that confidentiality issue? 2 MR. SCHNAPP: Yes. Oh, yes, there 3 are people -- 4 MR. HILL: It seems to be a big 5 deal. 6 MR. SCHNAPP: Yes, definitely, yes. 7 I mean, companies feel that it would -- 8 MR. HILL: That nine-month window 9 wasn't long enough. 10 MR. SCHNAPP: Affect their ability 11 to compete. 12 MR. HILL: The nine-month window 13 was too short. 14 MR. SCHNAPP: After the reporting. 15 MR. HILL: It was too short for 16 them. 17 MR. SCHNAPP: Well, no, they don't 18 ever want it out in the public domain. 19 MR. CLEVELAND: So who makes that 20 call, the nine-month call? 21 MR. SCHNAPP: Well, as we did last 22 time, if that's what we decide, we would BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 200 1 propose that to the Office of Management and 2 Budget. We make no decisions. The Office of 3 Management and Budget makes the decision 4 whether we can do it or not. 5 MR. CLEVELAND: You make the 6 recommendations. 7 MR. SCHNAPP: Yes. Actually when 8 our proposal goes to them, there's another 9 Federal Register notice issued for another 10 30-day comment period. And if people have 11 problems or comments, they send them right to 12 the Office of Management and Budget. 13 MS. KIRKENDALL: Typically, that 14 Federal Register notice will get comments by 15 anybody who really disagrees with what we're 16 proposing, and so that's what OMB looks at, 17 the comments they get in response to the 18 Federal Register notice. And if you get lots 19 of people saying no, no, no, then they say 20 no. 21 MS. SEREIKA: So you have these 22 data, it's just that they're not published, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 201 1 right? 2 MR. SCHNAPP: We don't put them out 3 in our detailed database. 4 MS. SEREIKA: You don't have like 5 an honest broker who is holding them? 6 MR. SCHNAPP: No. 7 MS. SEREIKA: You actually have 8 them, it's just they're not being published 9 or used. 10 MR. SCHNAPP: Right, right. No, we 11 maintain ourselves, yes. 12 Okay. And so I was talking in the 13 beginning about our Internet data collection 14 system. 15 I'll just tell you real quickly 16 about it. I know, again, we have briefed 17 your committee, but I know there's some new 18 faces here. 19 We collect -- we established our 20 Internet data collection system for 21 Electricity 2002. We now send out all of our 22 requests for data using e-mail. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 202 1 We established the Help Center to 2 field inquiries. This year we got 3,500 3 e-mails and 1,000 phone calls. And out of 4 the about 36,000, 37,000 forms that we 5 process a year, 85 percent of them, or over 6 30,000, came in on the Internet. The monthly 7 forms, 70 percent of them came in, and 8 80 percent of the annual forms. 9 And what this has in it is that 10 when they enter the data, they press an Edit 11 button and it gives them a little report on a 12 comparison. It compares their current data 13 with, say, last month's data. And if it's 14 outside of a range or if it doesn't add up 15 properly or if it it's used in the 16 calculations, they make a heat rate that 17 doesn't make sense, there's an error that 18 shows up. 19 They have to clear up all those 20 edits or press the Override button and 21 explain why the data's correct or it can't be 22 submitted. So that tells you that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 203 1 85 percent, 30,000 forms, came in that passed 2 all those edits. So we don't have to key it 3 in. Most of those edits that they overrode, 4 we talked with them and so some of them were 5 accepted, some of them they had to change. 6 But really only 15 percent of these forms now 7 we key in for them, and we're hoping to get 8 that up even higher. 9 And so the savings that we've seen 10 from the Internet data collection is that our 11 Electric Power Monthly, which had taken us 12 161 days to put out -- that's 161 days from 13 the end of the reporting month, not 161 days 14 after the date it was due, which is usually 15 30 days later. We -- actually I have to 16 update this. We put it out in 63 days the 17 last report. So we have literally saved 100 18 days in putting out Electric Power Monthly. 19 So that's due to the Internet data 20 collection system, our imputation process 21 that's been modernized, and really hard work 22 of our staff. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 204 1 MR. HENGARTNER: How many FTEs do 2 you save with this? 3 MR. SCHNAPP: How much what? 4 MR. HENGARTNER: FTEs. 5 MR. SCHNAPP: Well, I'll show you 6 here what we're saving. We've invested it. 7 We've reinvested it. 8 MR. HENGARTNER: It's hard. 9 MR. SCHNAPP: Electric Power 10 Annual, which took 470 days to produce, is 11 now produced -- again, Friday it came out. 12 It was produced in 272 days. Two hundred 13 days in savings. 14 Because of that, we caught up on 15 our State Electricity Profiles. We put out 16 two editions, the 2003 and the 2004 edition, 17 this year. We put out two editions of the 18 Cost and Quality of Fuels, 2003 and 2004 19 editions. And we created a brand new monthly 20 report, Monthly Flash Estimates of Electric 21 Power Data, which has national-level data for 22 the most part. That comes out just 50 days BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 205 1 after the end of the reporting month. 2 So there's a lot of savings that 3 we've plowed back in and caught up this year. 4 And next week, we'll be having a party on 5 that. And that's the comments I have. 6 I'd be glad to take comments now or 7 questions or we can wait until after Bob 8 Rutchik will be making his comments on the 9 proposed mergers forms. 10 MR. RUTHERFORD: I just have a 11 newbie question. 12 SPEAKER: Newbie? 13 MR. RUTHERFORD: I've looked at the 14 860. I think it's a great -- really great 15 database. I really like it. 16 But I was curious, do you guys 17 produce any information about the 18 distribution network that goes along with the 19 plants? 20 MR. SCHNAPP: No. 21 MR. RUTHERFORD: So we know the 22 exact locations, but we don't know how BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 206 1 they're connected. 2 MR. SCHNAPP: Well, we know what 3 they're connected. We know what service 4 areas. 5 MR. RUTHERFORD: But you don't know 6 what the path is for -- 7 MR. SCHNAPP: No, no. 8 MR. RUTHERFORD: Does FERC produce 9 that information? 10 MR. SCHNAPP: I don't know. Alan, 11 do you think they -- I don't think they -- 12 MR. BEAMON: Yes, they have forms 13 that collect that sort of information. 14 MR. SCHNAPP: Is that the 714 or 15 715? 16 MR. BEAMON: The 715 that they 17 collect that -- 18 SPEAKER: Alan, for your comment, 19 would you come to the mike so that we can get 20 on the recording? 21 MR. BEAMON: Say yes again? I said 22 FERC collects the line information, the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 207 1 transmission line information, on the 714, 2 715. I don't know how detailed it gets down 3 to individual plants and levels, and there's 4 issues with it, I can tell you that. 5 MR. SCHNAPP: When you said 6 "distribution," I thought you meant 7 distribution as opposed to transmission line. 8 MR. RUTHERFORD: Well, I meant 9 transmission line. 10 MR. BEAMON: Transmission. 11 MR. SCHNAPP: Okay. So I'll turn 12 it over to Bob. 13 MR. RUTCHIK: Thank you, Bob. Let 14 Bob come up here and hit for the Boston Red 15 Sox. 16 Good afternoon. I'm Bob Rutchik of 17 Statistics and Methods Group of EIA. And the 18 purpose of my presentation is to outline 19 issues that have arisen for merging the five 20 EIA survey forms into two new surveys. The 21 five surveys, again, are in Bob Schnapp's 22 outline. The three monthly surveys that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 208 1 would be merged into one new form, Form 2 EIA-423, Monthly Cost and Quality of Fuels 3 for Electric Plants Report; EIA-906, Power 4 Plant Report; and EIA- 920, Combined Heat and 5 Power Plant Report. 6 As kind of a refresher, the 7 combined heat and power plant produces both 8 heat and electricity from a single heat 9 source. These are commercial and industrial 10 co-generators, such as paper mills and oil 11 refineries. The second merger will involve 12 the two annual forms that Bob mentioned, the 13 EIA-860, Annual Electric Generator Report, 14 and parts of the EIA-767, Steam, Electric, 15 and Plant Operation and Design Report. 16 The issues that we'd like the 17 committee to think about and on which we'd 18 like to get your comments are the data 19 requirements for both surveys. 20 Is EIA collecting the correct data? 21 Two, combined heat and power plant monthly 22 imputation methodologies. Bob mentioned that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 209 1 we had proposed to discontinue collecting 2 monthly data from the EIA-920 from industrial 3 and commercial combined heat and power 4 plants, but continue to collect annual data. 5 The third issue will be the new 6 sampling cutoffs for both surveys. And I'll 7 get into a little bit more detail on that 8 later. 9 Now, I want to talk about the 10 monthly survey. And on the monthly survey 11 there are about three categories of data. 12 One is plant level data. 13 The second is prime mover level of 14 data, and the prime mover is the engine, 15 turbine, waterwheel, or similar machine that 16 drives an electric generator. And the third 17 category of data is the annual data on 18 sources and disposition of electricity. 19 These are the plant level data that 20 we propose to collect on the merged form for 21 the 423, 906, and 920. The left-hand column, 22 that's the information that EIA currently BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 210 1 collects that we want to carry over to the 2 merged form. The first one is quantity of 3 receipt, how much coal or how much gas came 4 to the plant. The second is the quality of 5 the receipts, the heat content and sulfur and 6 ash content. Sulfur and ash content applies 7 basically to coal and to the oils, like 8 displate (?) oil and residual fuel oil. And 9 the third one -- variable is end-of-month 10 starts of coal, residual, and displate fuel 11 oil, jet fuel oil, and kerosene. 12 The right-hand column lists the new 13 data that EIA proposes to collect on the 14 survey. The purpose of these data in general 15 are to fill in EIA's published data gaps and 16 gaps also in its forecast. In other words, 17 help make more accurate forecasts. 18 The first variable is cost. There 19 are cost indices in the trade press, and this 20 is something that was talked about a little 21 bit this morning in general terms, but we're 22 not sure of the quality. And sometimes when BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 211 1 EIA tries to get private data sets, you just 2 can't get them. They won't let EIA have 3 them. And we also know that there are no 4 official government sources on the price of 5 fuel, so that's one reason we wanted to 6 collect our own. 7 In the cost, we're going to have 8 two components. One is delivered cost, and 9 that will be for all the fuels. Delivered 10 cost is basically a bundled number of all the 11 costs that are involved the invoiced cost of 12 the fuel, the transportation charges, taxes, 13 commissions, and so on. That, again, we're 14 going to collect for all fuels. For coal and 15 for natural gas we want to collect the 16 commodity cost. That's just the price for 17 the lump of coal or the molecule of natural 18 gas. 19 The second variable is 20 transportation mode. And this is really not 21 a new variable for EIA. We -- what we do 22 plan to do is propose to move the coal BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 212 1 transportation mode question on the EIA-6, 2 the Coal Distribution Report Annual, to the 3 merged survey. That question asks for 4 primary and secondary, if there is more than 5 one transportation mode used, of 6 transportation. 7 To give you a simple example of how 8 this works, let's say the electric plant is 9 1,000 miles from the coal mine that supplies 10 it with coal. The first 600 or 700 miles, 11 the coal travels, let's say, by rail or 12 that's the primary mode of transportation. 13 The next, let's say, 390 miles, it's unloaded 14 from the railroad cars onto a barge and goes 15 on a barge down river, up river, or wherever, 16 to -- let's say, 10 miles from the electric 17 plant. That's 390 miles in the barge as the 18 secondary mode. 19 And the last 10 miles, it moves by 20 truck. We're really not interested in the 21 third and final 10-mile leg, the truck data. 22 But you are interested in -- to be able to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 213 1 trace the fuel in the primary and secondary 2 transportation modes. 3 The third new variable is Mine 4 Safety Health Administration, or MSHA, Mine 5 Identification Number. Again, this is just 6 to help us plot out a more complete path and 7 give us a fuller picture of the industry. 8 Bob touched on this a little bit. 9 And the third one is really an adjustment 10 factor or balancing factor, where the ending 11 starts equal the previous month's starts plus 12 receipts minus consumption plus or minus an 13 adjustment. And the balancing check, and 14 most of the EIA data has this, is important 15 to ensure the integrity of the data. 16 Now, currently, there are three 17 separate systems: 906, 924, 923. And to 18 balance all these three systems that has to 19 be done outside the IDC system, and this is 20 very inefficient. It takes up a lot of 21 resources. So one advantage of the 22 consolidated form would be to have the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 214 1 balance done in one system, the merged 2 system, and have it done before the forms are 3 submitted to EIA. 4 So if the numbers don't equate, the 5 left side doesn't equal the right side, the 6 respondent can do one of two things. Go back 7 and recheck its consumption and receipt data, 8 see if they're correct. And if they are 9 correct and still not -- both sides of the 10 equations are different, explain why they're 11 different. An example would be a pile of 12 coal. You have a pile. We thought it was 13 20,000 tons. We inventoried it. It was 14 25,000 tons. 15 And that's basically what the 16 adjustment factor is. It's not so much a 17 mathematical factor, but really a discrepancy 18 explanation of why the two sides of the 19 equation do not balance. So you could see 20 this would make data editing a lot easier and 21 a lot better. The data would improve EIA's 22 data editing and improve the data. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 215 1 Okay, the next category of data is 2 prime mover level data, and these are data 3 that we already area collecting. It should 4 be net and gross generation for all electric 5 plants, gross generation for combined heat 6 and power plants, total consumption for all 7 purposes, again, "all purposes" being heat 8 and steam and electricity combined, 9 particularly for CHP plants. And that would 10 be by fuel type and heat content by fuel 11 type. Heat content can be different from 12 receipts when it's consumed. 13 Again, the coal sticking out, being 14 out there in a pile could get moisture and 15 that would lower the heat content. 16 You noticed again, and, again, this 17 is something that Bob touched upon, there's 18 no fuel used to generate electricity. You 19 know, we're proposing to drop it and we have 20 the IERs (?) out there looking for a 21 surrogate, if there is one. 22 This, again, is another adjustment BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 216 1 or balancing check. And this is determined 2 annually both on the 906 and the 920 with the 3 two (inaudible) differences. Basically, 4 again, you want the source to equal the 5 disposition. What comes in, what's 6 generated, what a plant gets from another 7 source equal what it sends out either through 8 the grid, either through itself, or sells. 9 Again, (inaudible) want these two sides to 10 (inaudible) and the disposition equal itself. 11 It's, again, a check. And so that'll be the 12 third part of the merged form. 13 Okay, here we go, the monthly 14 surveys. And I want to go over the current 15 sampling schemes for a reason because they 16 are directly related in my opinion to 17 understanding EIA's proposed new 18 methodologies for -- sampling methodologies 19 for the merged survey and, also, later on for 20 the annual survey, which I'll discuss. 21 The first one is a 423. It really 22 doesn't use a sampling scheme. It just goes BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 217 1 to the approximately 740 non-utility electric 2 plants, or CHPs, whose total fossil fuel 3 nameplate generating capacity is 50 megawatts 4 or greater. 5 The EIA-906 and 920 do use a 6 sampling scheme. And the 906 goes 7 to -- collects data monthly from a 8 statistical sample of approximately 1,400 9 plants based on fuel consumed and generating 10 capacity criteria. For example, let's say we 11 have all plants that use coal, natural gas, 12 or displate residual fuel oil, or a 13 combination of the three, which would be most 14 of the plants, who have a nameplate generator 15 capacity of 200 megawatts or greater. They 16 would be include in that 1,400 monthly 17 sample. And those others, we'd collect 18 annually from approximately 3,000 other 19 plants that do not file the EIA-906 monthly. 20 Now, for the 920, it, too, collects 21 monthly data from a sample of approximately 22 300 plants in this case and uses the same BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 218 1 criteria of fuel and generation capacity as 2 the 906. And it also collects data annually 3 from the approximately 700 plants in this 4 case that do not file monthly. 5 Now to get to some of the issues 6 that I mentioned. EIA has not formulated a 7 cutoff yet for the merged form, but the 8 principle will be that it'll be -- generation 9 capacity will determine the cutoff between 10 monthly and annual respondents. And the goal 11 is, and Bob I think intimated at this 12 earlier, is to minimize the monthly 13 respondents. More specifically, minimizing 14 monthly respondents in great part means not 15 having respondents provide data that they had 16 not provided before on the three separate 17 forms. This is where a discussion of current 18 sampling cut off supplies. 19 For example, the current cutoff for 20 the EIA-423, as I mentioned, is 50 megawatts 21 or greater -- capacity or greater. If EIA 22 asked on the merged form CHPs and electric BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 219 1 plants that are below the 50 megawatts to 2 complete the merged form, EIA will be 3 increasing the number of monthly respondents 4 and using more resources, not less. This is 5 self-defeating the purpose of Electricity 6 2008. And this is also I think one factor in 7 having the CHPs respond annually. 8 Next is some of the ideas that we 9 had brought up in our brainstorming sessions 10 on the imputation strategies for imputing 11 monthly numbers for the CHPs and on which we 12 would like the committee's views. And these 13 main ideas that we brought up and talked 14 about is, one, benchmark data from 15 independent power producers. 16 Most independent power producers 17 produce electricity, but some of them I 18 believe may produce 5 or 10 percent heat and 19 steam and ship it next door to a paper plant, 20 and I've seen this type of operation. So 21 that would be one benchmark. 22 The other that we talked about is BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 220 1 maybe, possibly collect monthly data from a 2 small sample of the larger CHPs. 3 Okay, now I want to move -- switch 4 gears a little bit and talk about what we 5 call the new EIA- 860. Guy Caruso mentioned 6 this morning, and Bob also reiterated that 7 EIA's fiscal 2007 budget does not have any 8 funds for the EIA-767. So this is how -- our 9 plan to deal with that is that, one, there 10 will be no changes to the EIA-860. That form 11 will stay just as it is. And two, we will 12 merge from it -- from the EIA-767 to it 13 static data on power plant characteristics 14 from the EIA-767. 15 These are things like design or 16 physical characteristics data. Your 17 nameplate generating capacity is one. The 18 capacity of -- design capacity of your 19 various pieces of equipment is another 20 general example. So we'd keep that data and 21 delete the operational data from the EIA-767. 22 This is the data that changes year to year, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 221 1 like, let me think of a real esoteric 2 example, the chlorine you put in your cooling 3 water for a cooling system. 4 And again, a simple example of how 5 this would work is Schedule 6. If you had a 6 copy of the EIA-767 in front of you, 7 Schedule 6 is cooling system information. 8 There are two parts to it. Part A, annual 9 operations. That would be merged to the 10 EIA-860. And Part B -- I mean, annual 11 operations would go, believe it, because 12 that's the data that's variable year to year. 13 And Part B is design parameters, the static 14 information, and that would stay. 15 Now I want to go to the current 16 annual cutoffs for the EIA-860. The EIA-860, 17 again, goes to existing and planned electric 18 generating facilities in the United States 19 with a current total generating nameplate 20 capacity of one or more megawatts. Again, it 21 collects generator-level data from 22 approximately 2,700 companies that operate BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 222 1 5,700 plants, spanning over 16,000 2 generators. The data collected are 3 generator-level data again. 4 The 767 goes to organic fuels or 5 combustible and renewable steam electric 6 generating plants with nameplate generating 7 capacity, 10 or more megawatts. And it 8 collects boiler-level data from about 1,400 9 plants. So that's where we currently are. 10 And for what we propose for the 11 merged survey is, again, generating capacity 12 will determine the cutoff. And the goal is 13 to have about 10 percent of plants that have 14 90 percent of generation respond each year. 15 And that would be -- and we have a proposal 16 out that this would include most plants 17 between 500 and 600 megawatts and above. 18 They would be surveyed annually. 19 And to follow up on a point that 20 Nancy made this morning, EIA plans to do a 21 census for the remaining plants, the 22 so-called small guys. One-third would be BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 223 1 sampled one year, the second third the second 2 year, and the remaining third the third year, 3 so we'd have a complete census every third 4 year. And the next year, the cycle would 5 just start again. 6 Now, I'm going to get to the 7 specific questions which we have for the 8 committee. Do we have the correct data on 9 the EIA-423, 906, and 920 combined form? 10 Second, what should EIA be aware of 11 in formulating an imputation -- I always have 12 problems with the word -- imputation 13 methodology for CHP plants? 14 Third, do we have the correct data 15 on the combined EIA-860 and 767? 16 And finally, what does the 17 committee think of the cutoff methodologies 18 for both surveys? 19 And I don't have a thank-you slide 20 and I should have, but thank you very, very 21 much. 22 MR. HENGARTNER: Okay. Questions? BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 224 1 MR. BURTON: I've got a couple of 2 questions and a comment or two. On the 920, 3 is the distinction between the primary and 4 secondary mode, transport mode, going to be 5 just distance? 6 MR. RUTCHIK: As far as I 7 understand it, yes. 8 MR. BURTON: I do a lot of 9 transportation economics. Is there any 10 chance that you could go ahead and get every 11 mode that they're using even if it seems 12 inconsequential? More is better. 13 And finally, on the 920, I think 14 you're going to run into situations where the 15 respondents are going to have a hard time 16 telling you what their costs are. I mean, 17 more information is better than less, but I 18 would worry somewhat about the reliability of 19 that because there are all sorts of different 20 transportation arrangements. Sometimes 21 they're going to know what transport costs 22 are, sometimes all they're going to know is BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 225 1 what the delivered price is. There are 2 common facilities and you've got to figure 3 out how to cost those out, so that they get 4 included. I mean, collecting it's fine, but 5 I would worry about relying on it too 6 heavily. 7 And finally, on the imputation 8 matter for the -- if you look simply at the 9 larger CHPs, I would want to know one piece 10 of information first and that is whether the 11 economics for the smaller ones that you're 12 not looking at are similar or dissimilar from 13 the larger ones where you are collecting 14 information. If the smaller ones are 15 different, then I would probably look for a 16 different method. 17 MR. HENGARTNER: Thank you very 18 much. Before we go around the table, 19 actually, I realized we actually had prepared 20 statements by Nagaraj and Jae, and maybe we 21 should get them on the record before the rest 22 of us talk. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 226 1 SPEAKER: Oh, well. 2 MR. HENGARTNER: Oh, well. It was 3 my mistake. 4 MR. RUTCHIK: Thank you, Mark, for 5 your comments. 6 MR. HENGARTNER: So Jae, do you 7 want to start? 8 MR. EDMONDS: Yes. Actually, I had 9 down here it was Johnny Blair, and so my 10 comment was I was going to agree with him. 11 Before the fact. You have my agreement. 12 MR. NEERCHAL: We'll let him know. 13 MR. EDMONDS: So, yes, this was 14 actually helpful. I went through the 15 presentations before I came here and I think 16 it is much more helpful to hear this in 17 person. I guess I have three reactions. 18 You know, the first is I guess the 19 one to just start off with is that the whole 20 notion of going back and rethinking the data 21 collection with an eye toward merging and 22 consolidating, automating, and, in some BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 227 1 instances, expanding the data collection 2 seems like a good idea. And just looking at 3 the plan of attack to get to 2008, that looks 4 like a very rational and well thought out 5 plan. And so all of that seems to be to the 6 good. 7 There's this discussion in 8 the -- the second point is that there's all 9 this discussion about minimizing the number 10 of monthly respondents. 11 And of course, that, you know -- a 12 minimum, of course, is zero. So the question 13 is minimizing subject to what? 14 I presume it's a constrained 15 optimization, and so my presumption is that 16 it's budget. But I'm -- no one said that and 17 I'm guessing that the cutoff numbers are 18 given the budget we have, these are the 19 numbers that emerge from trying to fit the 20 form into budget. And then the question is 21 is that the right budget? 22 And there it comes down to asking BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 228 1 yourselves questions like Mark was asking, 2 which is the guys that are getting sampled 3 less how do they look compared to the ones 4 that are being sampled more? Obviously 5 they're small. Therefore, there's a 6 fundamental difference in the engineering of 7 these small units relative to the big ones 8 and often in the way they're operated. And 9 so the question is what are you losing when 10 you cut down the respondents? 11 So on the one hand, consolidation 12 and adding new information seems to be 13 increasing the quality of the product the EIA 14 is producing, but minimizing the number of 15 monthly respondents seems to be, in fact, 16 moving in the opposite direction. And so you 17 have to ask what am I giving up and what am 18 I -- and why am I giving that up? And I 19 don't know that I have the answer to that, 20 but I just put that on the table. 21 And then the third point that I 22 have, it's perhaps a small point, but this is BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 229 1 the second time this notion of CHP and the 2 disaggregation. You know, with combined heat 3 and power you're producing a joint product. 4 You're producing processed heat and you're 5 producing electricity, and you've got an 6 input into that process. 7 And then there's this desire to 8 say, well, a certain fraction of that 9 actually went to raising the steam and a 10 certain fraction of the fuel input went to 11 producing electricity. But to my mind it's 12 like why are you going to pretend like 13 they're two separate activities when, in 14 fact, they're a joint product? You know, 15 you're actually -- I mean, presumably, 16 there's some gains to conducting it as a 17 joint product relative to conducting this 18 activity as two separate activities, that 19 there's actually some efficiency gains in 20 doing combined heat and power. 21 And of course, if you go over to 22 EIA that's why they want to encourage CHP is BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 230 1 that there are efficiency gains in the 2 process. 3 So trying to go back and act as if 4 they're two separate activities never -- you 5 know, I never understood that. And so why is 6 the question that occurs to me. So those are 7 my thoughts. 8 MR. HENGARTNER: Thank you very 9 much. Nagaraj? Are you going to agree with 10 Jae? 11 MR. NEERCHAL: I think pretty much. 12 MR. EDMONDS: No, you're going to 13 agree with Johnny. 14 MR. NEERCHAL: Yes, I think we had 15 two Bobs speaking. I think Nick was looking 16 for a Bob from the committee and he didn't 17 find one, so I said (inaudible) I'm Bob for 18 this talk, I think, these comments. You 19 know, I agree with Jae I think. 20 No, a couple of comments. I think 21 one thing. The talk from Bob Schnapp, I 22 think he -- the improvement using Internet BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 231 1 data collection's just remarkable. I think 2 that's a slide to be remembered forever I 3 think. It's fantastic. And I was just 4 wondering just being a statistician I'm 5 supposed to ask some negative questions, is 6 there any downside to it? 7 You know, for example, are there 8 some people who say I don't have access to 9 the Internet, so I can't send you data, for 10 example? So there's some response rate, you 11 know -- are there any downsides to it? Are 12 you improving it so much that I think we can 13 tolerate a lot of disadvantages for those 14 kind of advantages? But I was kind of 15 wondering here are there any disadvantages at 16 all? 17 A couple of other things that kind 18 of got to me. I think on the cutoff, 19 determining cutoff, I think you're really 20 using the generation capacity for the cutoff. 21 I think that one of the questions I had in my 22 mind here, and I think Bob answered that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 232 1 question, but something about 90 percent of 2 the total is accounted by 10 percent of the 3 people or some number like that. 4 But what about some other 5 co-variates? Perhaps maybe if you ordered 6 this universe or this population by some 7 other co-variates and used, like, some other 8 size measure. I don't know, it would be 9 total sales, I don't know, whatever it could 10 be, number of employees, something else, 11 whether that could be more useful for 12 imputation, for example. 13 And then again, I had that question 14 that Jae had, that's why I agree with him. I 15 think that -- I remembered the CHP issue from 16 a couple of meetings ago. And it's a very 17 difficult thing to do. And when you're 18 trying to do something very difficult, you 19 want to make sure that it is useful. So I 20 think that will be a very interesting 21 question. 22 And finally, I think that every BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 233 1 third year, I think, once again, it's a 2 really very cute idea. But I was wondering, 3 every third who are being surveyed this year, 4 is it possible to ask them to give data from 5 the previous two years, for example, without 6 burdening them too much? You know, what do 7 you remember from last year? And even though 8 that may not be very reliable information, 9 that could be useful. That way, you 10 know -- I don't know, that's -- I don't know 11 the real parameters of burden reduction here 12 now. 13 I think EPA is going through this 14 discussion on TRI right now. They're talking 15 about burden reduction. They're not thinking 16 about EPA. The burden reduction on the 17 respondents, which is a totally different 18 burden reduction. So is it possible to have 19 them give us data from last year, for 20 example? Will that be reliable information? 21 And those are the couple of things 22 I had jotted down which was not covered by BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 234 1 either Mark or (inaudible). 2 MR. HENGARTNER: Thank you very 3 much. I did cut off Neha previously, so 4 please fill us in. 5 SPEAKER: She agrees. 6 MS. KHANNA: Okay. I have I think 7 three comments. 8 One, the initial one goes to both 9 Bobs, but maybe Bob S. can answer it. You 10 mentioned that you're able to move from 11 monthly to annual data collection with CHPs. 12 And at least the two reasons that you 13 mentioned in your presentation were they're 14 only 5 percent of the total and savings in 15 costs. But if I understand it correctly, 16 CHPs (inaudible) a lot of promise for the 17 future. We expect it to be a pro sector (?) 18 So that will be something -- I'm 19 not sure, I mean, savings in cost is a great 20 thing, but if it's going to be (inaudible) 21 going to have greater penetration we don't 22 really want to go there, you know. Cutting BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 235 1 off something is easy. Getting it back 2 online is harder. So that's just a question 3 to throw out there. You probably have a good 4 sense of constraints, but I want to throw 5 that into the mix. 6 Two other comments. One, I'd like 7 to follow up on what Mark said, total 8 transportation. It seems like it's important 9 not only for this sector, but in the morning 10 we had a discussion on the consumption side 11 and one of the big areas they don't cover is 12 non-residential transportation. And it seems 13 like there is a scope here for collaboration 14 across different groups. Maybe they'd be 15 interested in having those data as well or 16 access to those data or working in trying to 17 collect those data. So that seems like 18 something to do. 19 And the last comment is a reaction 20 to what Nagaraj said about asking those 21 two-thirds that don't get interviewed or 22 surveyed every year. I would worry very much BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 236 1 about combining those data with actual 2 annual -- you know, the Europe survey data 3 because of recall issues. And in developing 4 economics we have a huge issue about recall 5 (inaudible). If we do go down that road it 6 has to be very clear which one is recall and 7 how much is the recall period, because that 8 can affect the quality and variability of the 9 data. 10 MR. HENGARTNER: Thank you. I 11 actually have a question for Barbara on this 12 one-third. She's our sampling expert. 13 MS. FORSYTH: Yes, and on 14 everything else. 15 MR. HENGARTNER: Yes. 16 MS. KHANNA: And actually, I don't 17 want to undermine your authority, but I think 18 somebody wanted to react to you. I heard a 19 mumble from (inaudible). 20 MR. HENGARTNER: Did I see his tent 21 go up? 22 MS. KHANNA: He doesn't have a BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 237 1 tent. 2 SPEAKER: He doesn't have a tent. 3 MR. HENGARTNER: Oh, oh, oh, sorry. 4 MR. RUTCHIK: Let me use the 5 microphone before Bill gets really mad. On 6 the -- you talked about his idea of possibly 7 when we ask every third year getting the 8 other two years, and you talked about recall. 9 These are establishments. Would there also 10 maybe be a problem that the records for those 11 prior two years might not be that good? 12 Because when I think of recall I think more 13 of household surveys than establishment 14 surveys. 15 MS. KHANNA: But this is 16 information they're regularly keeping anyway. 17 MR. RUTCHIK: Yes, so I'm just 18 wondering if the data would maybe degrade 19 over time a little bit. 20 MS. KIRKENDALL: Maybe that'd be 21 something to ask about when you do your 22 number. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 238 1 MS. FORSYTH: It's all empirical. 2 It's all empirical. I mean, that's something 3 you can evaluate as you set up the program. 4 And in fact, I'd recommend it for just the 5 reasons. 6 MR. HENGARTNER: I mean, my 7 reaction to the one-third, and that was the 8 question to Barbara, is the following, is 9 that you are thinking or you're in the mode 10 of thinking that by cutting it in one-third, 11 the first year one-third, the second 12 one-third, the third year you're going to get 13 the census every three years. And that's 14 correct if the frame is constant. 15 Unfortunately -- 16 MS. FORSYTH: Right, and response 17 is perfect. 18 MR. HENGARTNER: Right. Now, of 19 course, the frame is moving over time and the 20 question I have is how much does the fact 21 that the frame is not constant over the 22 three-year period impact their ability to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 239 1 actually carry through this strategy? And 2 it's a question, not a comment. Yes, Bob? 3 SPEAKER: (inaudible)? 4 MR. HENGARTNER: Yes. 5 SPEAKER: Or do we wait 'til the 6 end? 7 SPEAKER: No, go ahead. 8 MR. HENGARTNER: No. 9 MR. HILL: No, I sort of have an 10 answer. Maybe he's got it, too. I remember 11 some of the discussion last time on this 12 cutoff. I'm actually glad to see you're 13 going to a sample. I remember that was part 14 of our discussion. 15 I think the unstated issue is that 16 the response -- the number of responses from 17 the smaller producers is less likely, so that 18 you're getting good responses from the large 19 producers, but fewer responses from the 20 smaller group. Yes, so by cutting it in 21 thirds you're getting -- of the group that's 22 less reliable, you're getting less data from BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 240 1 that group. 2 MS. FORSYTH: Per year. 3 SPEAKER: Working harder to collect 4 that data. 5 MR. HILL: Per year. Per year. 6 SPEAKER: Yes. 7 SPEAKER: But you're getting better 8 data. 9 MR. HILL: Yes, there might be some 10 sort of optimal way of trying to predict 11 where that cutoff occurs, if there's an 12 associated cost or a benefit to the surveys. 13 And I don't know if you can do that or not. 14 And you may be able to come up with an 15 optimal -- I guess that's part of your 16 comment, that there is I guess zero if you 17 don't want to do as few as possible, zero is 18 the (inaudible) as few as possible under some 19 sort of constraints. So the question is what 20 sort of constraint can you find that'll give 21 you an optimal? 22 MR. SCHNAPP: As I say on my stamp, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 241 1 I'm not a statistician, I'm just a poor 2 economist, so I kind of look at things a 3 little bit differently than everybody else 4 does. But let's see, starting at the top 5 here, merging is a good idea, very good. 6 Okay. 7 Is the budget an issue? I mean, 8 certainly it's always an issue, but the idea 9 is we want to be able to get back to doing 10 analysis. Our group really hasn't done any 11 serious analysis, feature articles, analysis 12 reports, in about five years. And that's 13 because we have to move -- because we 14 increased the amount of data that we were 15 collecting, we lost resources, and so 16 everyone was moved over to the data side and 17 we're trying to carve back out resources to 18 do analysis. So whatever we can save 19 anyplace will really be a plus we think to 20 all of EIA because we'll be able to do those 21 kinds of independent analyses again. 22 Why separate out the fuel and the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 242 1 CHP plants? 2 Because when we get asked the 3 question by Congress how much -- among 4 anybody else, how much fuel do you 5 use -- does the United States use to produce 6 electricity, we feel we need to split out the 7 amount of fuel for electricity and the amount 8 for use of thermal output. Is it elegant? 9 No, but we think that we need to do that in 10 order to be as accurate as we can to explain 11 how much fuel is used for electricity 12 generation. Our efficiency's gone up when 13 you're a CHP plant versus just an electric 14 power plant obviously, and that gets shown in 15 the data. It's certainly there. But we 16 think we need to split out those fuels into 17 the two subcategories. 18 Let's see here. Okay, the downside 19 of the Internet data collection system. It's 20 not quite as easy as it might sound. We used 21 to stuff 35,000 envelopes. And now you press 22 a button and you send out 35,000 e-mails, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 243 1 except that e-mail -- when you sent out that 2 form, if it didn't go to the right person, 3 that hard copy piece of paper would bounce 4 around from office to office and finally find 5 a home. Okay. We need to know the exact 6 e-mail address. If you're off by one letter 7 or it goes to somebody that doesn't work 8 there anymore and they don't have this 9 bounce-back because some them don't they 10 don't do away with the e-mail, they just 11 leave it there, it's out in cyberspace. And 12 so that has been a problem for us. 13 MR. HENGARTNER: How many did 14 bounce back from your 35,000? 15 MR. SCHNAPP: Well what we're told 16 is that about 20 percent of the respondents 17 typically change over a year, and that's 18 about what we had. But we're trying to do a 19 better job of that this year, contacting them 20 earlier. We had some other problems as well 21 going along, but that certainly is a problem 22 keeping those e-mail addresses up to date. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 244 1 MR. NEERCHAL: And following up on 2 those bounce-back. 3 MR. SCHNAPP: Right, yes. And we 4 call them back and we can get a phone number 5 into the organization and find out who it is. 6 But we'll say that our response rate, and 7 this kind of goes to a question by Mr. Hill 8 before, which was that the nonresponse rate 9 for all of our forms put together is 10 certainly less than 1 percent. 11 I mean, on the 860, I think we had 12 less than 40 non-respondents out of 2,700. 13 On the 861, sales and revenue, out of 3,200, 14 I think we're missing 19. On 906 and 920, 15 which had a total of 3,200 respondents, I 16 think we're missing 20 or 25. Yes. 17 SPEAKER: You put the fear of 18 (inaudible)? 19 MR. SCHNAPP: Yes. 20 MS. FORSYTH: Does that include the 21 things -- 22 MR. SCHNAPP: And I'll explain that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 245 1 in a moment, yes. 2 MS. FORSYTH: Does that include the 3 ones that are, like, lost? 4 MR. SCHNAPP: This is everybody. 5 Everybody that's in the frame. 6 MS. FORSYTH: (inaudible) answer? 7 MR. SCHNAPP: Yes, yes, yes. Well, 8 we have -- well, we're trying to get down our 9 nonresponse strategies. So we send them an 10 e-mail and then, two weeks before the form is 11 due, we send them another e-mail saying, hey, 12 don't forget us. The day after it was due we 13 send them you forgot us, you need to get it 14 in. Two weeks later, we send an e-mail from 15 me to a higher level person saying that your 16 data was due and for every day it doesn't 17 come in, you're liable to incur a fine of 18 $2,750. 19 Four weeks later, we send a letter 20 from my supervisor to that last person's 21 supervisor, so now we're up about the vice 22 president, general counsel, or president, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 246 1 sometimes mayor of the city, telling them the 2 same thing, that there could be a fine. And 3 I've got to tell you, you get response when 4 you say that. We don't make up those words. 5 It's on the front of every form that we have, 6 EIA has. 7 And plus, on top of that, the 8 contract staff and the federal staff are just 9 tenacious. They will not let go, so they 10 continue to follow up. 11 Let's see. Sure. 12 MR. SINGPURWALLA: Do you send them 13 a link to like a web site that they put the 14 answers in and then push the button? 15 MR. SCHNAPP: Yes. 16 MR. SINGPURWALLA: Push them 17 (inaudible). 18 MR. SCHNAPP: Right, right, yes. 19 Let's see here. So there is a lot of cost 20 anyway on the other side, on Internet. It 21 really is, it is. And then you have to 22 modify your edits and keep them up and, I BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 247 1 mean, there's a lot of stuff going on there. 2 But there's still a lot of payoff then 3 because you can evaluate the edits 4 electronically and figure out what to change, 5 which particular 861 staff didn't, so they 6 saw that there were a lot of failures of an 7 edit because the number of customers changed 8 by more than 20 percent. 9 Well, they investigated it and saw 10 that these were mostly the small distributors 11 of electricity. And so if they had 10 12 customers and now they had 13 customers, that 13 was outside the 20 percent ban. So we went 14 back and we changed that to have -- so that 15 if you were -- you know, we defined what 16 "small" was. If it was small, it could 17 change by 100 percent. If you were medium, 18 it could change by 50 percent, and then a 19 large one by 20. So that cut down on a whole 20 host of edits that were failed and that they 21 overrode. 22 MR. NEERCHAL: And most importantly BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 248 1 I think, you're using the data entry type of 2 labor into improving quality of data. 3 MR. SCHNAPP: Right, right, 4 exactly. Right, because we don't have to key 5 it in. 6 MR. NEERCHAL: Yes. 7 MR. SCHNAPP: And the quality then 8 with it coming in means it's higher. I tell 9 them that my goal, when I do talk with 10 someone and they call to either complain or 11 they call to say, hey, I like this, I tell 12 them our goal is never to speak to you again. 13 We want to send you an e-mail. We want you 14 to send me the data, press that button, pass 15 all the edits, we're done. We don't ever 16 have to speak again. 17 Let's see, the next item was, well, 18 there are different kinds of imputation we're 19 talking about here. One's on the 860, which 20 is the three-year thing we're talking about, 21 and the other is the 906, 924, 23, we're 22 talking about doing away with the CHPs. So BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 249 1 let me address each one of them. 2 One the -- by doing the three-year 3 process, processing on the 860, we're talking 4 about static data. I mean, they don't have 5 to remember anything. We're talking about 6 capacity. Their nameplate capacity probably 7 hasn't changed over the past three years. 8 Even if it changed by a little bit, it's 9 really not going to affect our data much and 10 it's not going to affect any of your uses of 11 the data much. Again, we're talking about a 12 third -- you know, we'd be doing that for 13 two-thirds of the respondents, so by 14 imputation -- you know, it's the poor man's 15 imputation. I look back at last year's 16 number. That's all that we would be doing. 17 It's actually the imputation 18 process we put in this year for the 40 that 19 didn't respond; we looked back at last year's 20 number. How far could we be wrong? I mean, 21 they accounted for.00001 percent of capacity. 22 So that's why we don't think that that's such BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 250 1 a bad idea. 2 On the CHP side, by not going to 3 the -- you know, again, so these are small 4 generators. We just don't think we're going 5 to be losing that much. 6 It's not that we're not going to 7 collect it from them. We would collect it 8 from them on an annual basis, so we'd still 9 have their data. And even if there's a 10 growth over time, we would capture that. 11 Again, in three years we could revisit 12 whether we would go to them on a monthly 13 basis. But now, they really -- and I got to 14 tell you, their growth has been 0 over 10 15 years. So we don't think it's such a bad 16 idea, but I mean, certainly something -- 17 MS. KIRKENDALL: But how are you 18 going to pick up seasonal patterns? Because 19 some of them do have some seasonality, don't 20 they? 21 MR. SCHNAPP: Well, they account 22 for 5 percent of generation, you know. I BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 251 1 mean, if we're off by -- what imputation does 2 is it figures out what the -- again, in an 3 economist's term, my little terms here, okay, 4 the growth rate is 4 percent. Okay, so 4 5 percent growth on 5 percent is .02, something 6 like that. Okay. If we're off by 7 50 percent, I mean, how much is it going to 8 change -- how much off on generation are we 9 going to be in the whole scheme of things for 10 that month? Now, that's why we're not too 11 worried about it, only because they're such a 12 small part of the frame, you know. 13 Jim Knaub's imputation process does 14 account for seasonality. I mean, we do 15 include that and look at the others. 16 MS. KIRKENDALL: Well, you do, but 17 now you have data from guys who are in the 18 same industry and you could say that maybe 19 the seasonality is the same for guys in the 20 same industry. And now you're going to get 21 rid of monthly data from people in the same 22 industry, so you won't have any information BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 252 1 of the seasonality for the same industries. 2 MR. SCHNAPP: Right, right. Well, 3 it's something that -- 4 MS. KIRKENDALL: So you really have 5 a lot less information to be making estimates 6 from. 7 MR. SCHNAPP: Right. We're 8 still -- it's still an issue that we're 9 discussing, and I might be a majority of one 10 at the moment in that group. We're still 11 discussing that one. I think that answers 12 most of the questions. I kind of cut across 13 everybody there. I'd be glad to answer any 14 others. 15 MR. HENGARTNER: Thank you very 16 much. 17 MS. SEREIKA: But the seasonality 18 part, what if the -- this, again, this is the 19 one-third that we're talking about that's 20 sampled every -- 21 MR. SCHNAPP: No, no. 22 MS. KIRKENDALL: No, actually BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 253 1 that's -- 2 MR. SCHNAPP: That's the CHPs. 3 MS. SEREIKA: Oh, CHPs. 4 MS. KIRKENDALL: These are the 5 5 percent of total generation. 6 MS. SEREIKA: I mean, I would 7 imagine if you sampled different people at 8 different times of the year you might capture 9 it, but not that same group every month, you 10 might capture some of that seasonality. 11 MR. SCHNAPP: Well, I think that we 12 would have to -- we would want somebody to be 13 in the sample as opposed to coming in and out 14 of the sample during the year. It's really 15 difficult enough just to follow them, but 16 coming in and out is really unbearable. 17 MS. SEREIKA: Okay. 18 MR. BURTON: Can I beat a dead 19 horse? The point that Jae originally brought 20 up about allocating -- 21 MR. HENGARTNER: Mark, can you 22 speak in the -- I can read you the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 254 1 instruction for the microphone if you want. 2 MR. BURTON: After the break. Just 3 to go back to the point that Jae brought up 4 originally about allocating out across the 5 different products of the CHPs. Not only 6 do -- I'm not sure that I completely accept 7 the purpose for it because these guys are so 8 little, then it really shouldn't matter that 9 much to the total fuel consumption. But I 10 can't -- it's just like any other case of a 11 common cost in economics. You can bound 12 where the volumes might be by the incremental 13 cost of conducting each activity separately, 14 but there is absolutely no logical way to 15 refine that allocation beyond those two 16 boundaries. So, I mean, in a sense, there's 17 no way to do it right. 18 MR. SCHNAPP: We agree. 19 MR. BURTON: Okay. 20 MR. SCHNAPP: And (inaudible) a 21 hard time. 22 MR. BURTON: Well, the truth is BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 255 1 probably right now the volumes are so small 2 that it doesn't really matter, but if this 3 does become a big deal in subsequent decades, 4 then this is a problem that's really going to 5 be pretty vexing. 6 MR. SCHNAPP: Yes, yes. 7 MR. BURTON: Was that okay? 8 MR. HENGARTNER: Thank you very 9 much, Mr. Burton. Excellent. Any more 10 questions? Yes? 11 MR. KNAUB: Well, what about in the 12 case of a few of the states where CHPs 13 may -- yes? 14 SPEAKER: Please identify yourself. 15 MR. KNAUB: Thank you, sorry. Jim 16 Knaub, CNEAF. What about in the case of a 17 few of the states where some of the CHPs 18 might be more important if we just estimate 19 it for a few of the more -- of the few 20 selected states where they were more 21 important? That maybe that would give 22 us -- keep a foot in the door and still not BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 256 1 very much burden, as an example. 2 MR. HENGARTNER: A good point. 3 Okay, thank you very much. I'd like to move 4 on to our next -- 5 MS. FORSYTH: Yes. Actually I 6 might be able to respond to that. 7 MR. HENGARTNER: Okay, sorry. 8 MS. FORSYTH: In a completely 9 different context, we were doing work for the 10 IRS that focused on specific localities, 11 doing extra data collection in those 12 localities. And the localities reacted very 13 strongly against the extra data collection 14 being focused on them. So there might be 15 local political reasons to avoid that kind of 16 strategy. It did impact at least some 17 aspects of the data quality. 18 MR. CLEVELAND: One minor point on 19 the joint product issue. It's not going to 20 go away for a lot of political reasons. We 21 went through this in the -- there is no 22 common -- there's no right way to do it. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 257 1 It's completely arbitrary. But in the '60s, 2 when the federal government was heavily 3 regulating interstate gas regulation and they 4 were trying to determine how to cost 5 interstate versus intrastate, they wanted to 6 do it on a cost basis. So the federal 7 government had years of meetings where they 8 brought people in and said, well, we need to 9 know the different -- the cost of producing a 10 cubic foot of gas. Well, of course, a lot of 11 gases produce jointly with oil. You drill a 12 well, out comes both. 13 Again, it's completely arbitrary. 14 Yet, they had to, so everyone went through 15 back handsprings to come up with what it 16 would cost to produce the gas. 17 And they had lots of fights over 18 what the right way was to do it. But if it 19 does become increasingly important, which it 20 will, you're going to be stuck with having to 21 grapple with these issues. 22 MR. EDMONDS: Well, just -- you BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 258 1 know, you might just leave it as a separate 2 category. In a sense it's a CHP and, in one 3 sense, you're not going to get any 4 electricity unless you raise heat. So in a 5 sense it's a one-for-one. I mean, it's all 6 going -- you know, in a sense, it's all going 7 to electricity in one sense. In another 8 sense, I'm raising the heat so it's all the 9 electricity is free. So you might just want 10 to keep it as a separate category. 11 Right now it isn't a big deal, but 12 I think once you begin to get into these 13 arbitrary distinctions, you may or may not, 14 in fact, be doing full service, but if you 15 keep it separated out and you always have a 16 star by it, then at least you've been honest 17 with them it's not straight up. 18 MR. HENGARTNER: Good. Can I go 19 now? Thank you very much. I'm sorry to have 20 cut you off. 21 I'm trying to keep on time. So our 22 next presentation is on emerging BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 259 1 technologies, data, and NEM modeling by Fred 2 Mayes from CNEAF and Chris Namovicz. 3 MR. NAMOVICZ: Okay, I guess I'll 4 go ahead and get started here. Bill told me 5 that you guys were interested in learning a 6 little bit about the renewable energy data 7 needs that we have in NEMS, in particular the 8 wind energy area. So that's what I'm going 9 to focus on today. I'll spend a very brief 10 amount of time talking about the wind energy 11 market projections that come out of NEMS and 12 sort of the importance of why wind is 13 something worth looking at. And then I'll 14 spend the bulk of the presentation talking 15 about our key assumptions and the data needs 16 behind those. 17 You can see the blue line, the dark 18 blue line there is the history and the 19 reference case, AEO projection from AEO 2006. 20 The wind industry has seen a lot of growth in 21 the last 5 to 10 years, going from about 22 1,200 megawatts in the late '90s, to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 260 1 approaching 10 megawatts today, and by the 2 end of next year probably closer to 1,500 3 megawatts. However, you see that after 2007, 4 we show in the reference case subsiding 5 substantially and flattening off. That's 6 primarily due to the expiration of the 7 federal production tax credit, which has 8 really been a key driver behind most of the 9 growth that we've seen in the last 5 to 10 10 years for the wind industry. 11 You can see the red line -- the red 12 line shows a case that we ran on last year's 13 AEO assuming a 10-year extension of the 14 production tax credit. And you can see 15 throughout that 10-year extension period that 16 the growth pretty much continues on the same 17 trajectory that it is seeing today. 18 A similar growth pattern is seen if 19 the Congress would -- if there were a 20 national cap on greenhouse gas emissions of 21 some significance. We've done several 22 analyses of these over the last several BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 261 1 years. This one particular one that we did 2 for Sen. Salazar last year, and you can see 3 that it's a pretty significant cap. And the 4 contribution of wind to meeting that carbon 5 reduction goal produces a similar growth 6 trajectory as if you extended the production 7 tax credit for another 10 years. 8 I take from -- I sat in on Susan's 9 session this morning, so most of you are, I 10 take it, fairly familiar with NEMS. The main 11 thing I want to point out for how we do these 12 projections is wind energy. 13 Wind is an electricity resource. 14 It's modeled within our electricity capacity 15 expansion model. It competes against all 16 other forms of generation -- nuclear, coal, 17 natural gas, whatever -- for growth and to 18 meet growing load in the electricity market. 19 And it's based on the 13 regions 20 that are loosely based on the NERC regions. 21 Like I said, that was going to be 22 brief. Now I'm going to spend the rest of my BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 262 1 time talking about our key assumptions in 2 particular. I'll divide it into two parts: 3 The technology assumptions, which manifest 4 themselves within the model in terms of 5 technology costs and technology performance, 6 but there's also particular to wind, because 7 of its unique operating characteristics and 8 grid interaction assumptions that I'll also 9 be talking about as well as some of the 10 resource limits, the resource assumptions 11 that go into NEMS. 12 Now, these also manifest themselves 13 in terms of the performance and cost data 14 within the model, but they're trying to get 15 at a slightly different aspect of the 16 technology or the resource. 17 The current costs that we're 18 modeling for wind energy and NEMS is based on 19 an analysis I did a couple years ago, the 20 average cost from, say, the late '90s to the 21 early 2000s. It turns out to be about $1,100 22 to $1,200 a kilowatt. There is, however, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 263 1 significant anecdotal evidence since 2003 2 that costs have gone up substantially, 3 perhaps as much as 25 percent. Reports now 4 are somewhere around the $1,500 to $1,600 a 5 kilowatt range. 6 There's not very strong agreement 7 on why that very short-term increase 8 has -- or why that big increase has occurred 9 over such a short period of time. There's a 10 lot of people posing their own pet theories. 11 It's increased materials cost; foreign 12 exchange rates most -- when technology is 13 manufactured in Europe or outside the United 14 States, and the market is, in fact, a very 15 international market even beyond where the 16 technology is manufactured; supply chain 17 bottlenecks, both in the United States and 18 the world, but industry has been growing 19 quite substantially and it's up. 20 MR. FEDER: Chris, that's a 21 one-time expense? 22 MR. NAMOVICZ: This is the capital BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 264 1 cost, the initial investment cost, yes. 2 So because the industry's been 3 growing so fast, it's been difficult perhaps 4 for the industry to maintain manufacturing 5 production capacity to meet the level of 6 growth. And part of that may be due to 7 subsidy policy, both because the subsidies 8 are propelling such fast industry growth, but 9 also because of perhaps the structure of some 10 of the subsidies, especially in the European 11 market, might not encourage cost reduction or 12 at least price reduction. And then there's 13 also some people, especially the developers, 14 complaining about price gouging by the 15 manufacturers to take advantage of all these 16 factors. 17 My own sense is that there's 18 probably some truth in all of these and there 19 may even be some factors that I don't know 20 about or that haven't really been widely 21 explored in the literature that are causing 22 this. The bottom line, though, is that we BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 265 1 are assuming that these costs, that this 2 25 percent cost increase is transient, and 3 especially given the context over model where 4 the PTC goes away and growth goes back down. 5 We're assuming that by that time period, 2008 6 or so, the cost will return to something 7 closer to their long-term average of $1,100 8 or $1,200 a kilowatt. 9 I'll talk a little bit about how I 10 got that number. Starting in about 2001, we 11 started to do the Form 412 survey, which is 12 capacity cost essentially, what we had been 13 sending I guess just to the public power 14 folks, to what are called the independent 15 power producers. And most of -- about 90+ 16 percent of the wind industry are represented 17 by independent power producers, so suddenly 18 we actually had real installation cost data 19 from this population that previously we had 20 relied on sort of anecdotal industry reports 21 project announcements to kind of glean a 22 sense of where capital costs were. We now BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 266 1 had actual data. 2 There are some limitations to the 3 analysis. 4 We had -- within the wind industry, 5 and I don't know about the survey frame at 6 large, but within the wind industry there's a 7 pretty poor response rate for many years 8 until last year. 2001 was the year of record 9 for when the power installations and for 10 plants in that vintage coming online. That 11 year, we had 1 respond representing about 80 12 megawatts out of 1,700 megawatts installed in 13 that year. So obviously not only could you 14 not get statistics from a sample of one, but 15 you can't even report costs because then 16 there's confidentiality in the response and 17 you can't report a single guy if you've got 18 to keep his identity secret. 19 The other problem is the questions 20 were really designed for the conventional 21 industry, the steam plants, the combustion 22 turbines. And in some cases, it doesn't BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 267 1 matter, but there are -- especially for the 2 operations and maintenance costs and things 3 like that. It was very clear when I started 4 looking at the data that there was no 5 consensus within the respondents of what the 6 question meant because the data was just all 7 over the place. And essentially it was 8 useless because it was clear that -- and 9 after reading the question carefully it was 10 pretty clear that if I were a wind power 11 operator, the only one I know about wind 12 plant operations, I'd have trouble knowing 13 what to put in that field as well. 14 The biggest problem with the Form 15 412 is that it's -- we don't do it anymore. 16 So I'd love to deal with these other two 17 problems in more depth, but right now I can't 18 deal with any of them. 19 So the other thing we were able to 20 get, fortunately, the way the Form 412 is 21 taken, we were able to get essentially 22 retroactive installation cost data, more or BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 268 1 less, so that we could get a historical cost 2 pattern for the wind industry, which feeds 3 into our learning curve assumptions model 4 that we have in NEMS. You can see from the 5 dark blue diamonds going down here our 6 learning curve that I developed off of our 7 412 data. As the wind industry frequently 8 touts, they had, in fact, had a very 9 substantial cost decline from the early '80s 10 when they were a new industry. You can see 11 over the first 1,200 megawatts or so capacity 12 quite a substantial cost decline. 13 However, the other trend that they 14 don't talk about a whole lot is really since 15 the late '80s, when the current boom started, 16 capital costs, at least installation costs, 17 have been very flat, if not somewhat 18 increasing. You can see the sort of upward 19 end of the hockey stick if you just look at 20 the diamonds and not look at the best fit 21 line there. And that, keep in mind, does not 22 include the data post-2003, when there are BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 269 1 substantial anecdotal evidence that costs 2 have gone up even further than that. 3 As far as NEMS is concerned, for 4 the wind industry based largely on this 5 analysis we're assuming that wind capital 6 costs declined by about 1 percent for every 7 doubling of installed capacity. However, 8 capital costs doesn't really tell the 9 complete story for wind, and I'm going to 10 talk about -- spend a couple slides talking 11 about performance, which we model in terms of 12 wind plant capacity factor. 13 MR. CLEVELAND: Chris, before you 14 go on, can I just ask a quick question. One, 15 have you seen -- I mean, why not just use 16 European data where they've got a rich data 17 set on learning curves for Spain, Denmark, 18 and Germany? 19 MR. NAMOVICZ: Well, 20 that's -- before I was able to get the 412 21 data, I was relying on European data, which 22 essentially because the 412 data was in BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 270 1 pretty close agreement with what -- the 2 trends I was seeing in Europe -- 3 MR. CLEVELAND: Yes, I think they 4 have progress ratios of like 85 to 5 90 percent. 6 MR. NAMOVICZ: Right, but they have 7 the same hockey stick where starting in the 8 late '90s things started to -- 9 MR. CLEVELAND: Right, that's why I 10 asked the question. So they see the same 11 leveling off? 12 MR. NAMOVICZ: That's why I say 13 leveling off and not just going up because 14 the German data is going up, the Danish data 15 was leveling off, the Spanish data when I 16 looked at it wasn't complete enough to get a 17 sense since their industry really didn't pick 18 up until the early 2000s. But looking at 19 Germany and Denmark, which were kind of the 20 center of the industry for most of the '90s 21 and early 2000s, it's a very consistent story 22 up through about '97, '98, costs going way BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 271 1 down and then hit the wall and have either 2 stayed the same or started to go up. 3 So from a developer's perspective, 4 the end user customer for the wind turbine, 5 they're not only interested in how much it 6 costs, they're interested in how many 7 kilowatt hours they can amortize the cost of 8 that plant over, which is primarily a 9 function of the capacity -- annual average 10 capacity factor for the plant. 11 This, of course, varies greatly 12 from site to site. There are a lot of 13 factors that go into it. There's the brand 14 of turbine that you use. Even within a 15 brand, there are a dozen different models 16 made by GE, Vestas, the two largest 17 manufacturers, have models that are designed 18 for particular sites. 19 And how you arrange the turbines 20 within the site affects the capacity factor. 21 Obviously the climate of the region as well 22 as the interannual variation in weather, how BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 272 1 the climate actually plays out, makes a big 2 difference. 3 We're able to use the Form 906 to 4 actually get a good sense for the U.S. 5 capacity, what kind of -- the range, the bulk 6 range of capacity factors, which shows a 7 range, even for newer plants built within the 8 last couple of years, annual averages from 9 under 20 percent to over 40 percent, with a 10 pretty high concentration between, say, 25 11 and 35 with a significant number in the 35 to 12 45 range. 13 The main problem we have is we have 14 a good theoretical model of what the wind 15 resources look like and engineering estimates 16 of what capacity factors we can expect out of 17 particular resource areas. But we don't yet 18 have a good way of matching individual sites 19 to our theoretical model to say, oh, well, a 20 plant in what we call a Class 4 really gets 21 in this range and a plant in Class 5 really 22 gets in that range. So that's an area that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 273 1 there's probably data available to do this if 2 we could afford to -- you know, if somebody 3 wanted to go out and actually do it, and it'd 4 be useful to do. 5 The capacity factor, however, also 6 changes over time. Many people wonder if 7 capacity factor is just a function of wind 8 resource, how is it possible that capacity 9 factors are really improving over time? And 10 in fact, capacity factor is a design 11 parameter, an engineering function. The only 12 real hard theoretical limit is you can't have 13 more than 100 percent capacity factor. 14 But in theory, if you took, say, a 15 blade off of a 2-megawatt machine, which is a 16 pretty standard machine nowadays, but 17 attached it to a 1-kilowatt turbine, any 18 slight movement of that blade is going to 19 fully load that 1-kilowatt generator. And if 20 you made the blade light enough and the 21 gearbox light enough, you could get the blade 22 to move in any slight wind and you could BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 274 1 easily see where you could be getting a 90 or 2 a 95 percent capacity factor out of a 3 theoretical situation. Obviously, though, 4 it's highly uneconomic to be spending so much 5 money on a blade and a gearbox to get such a 6 small amount of electricity production. So 7 that's why the economic balance tends to be 8 in the 35 to 45 percent range. 9 However, we do model capacity 10 factor growth as a function of the capacity 11 installment, similar to the way we model 12 capital cost declines. So we're trying to 13 capture this as well. What we don't have a 14 good model of and would be a nice data 15 analysis to try and do, and something I've 16 thought about a lot in the last couple years, 17 is there's this internal -- trying to model 18 the internal economic trade off that goes 19 into determining whether you take an 20 improvement in blade technology and you pass 21 along the savings in terms of reduced capital 22 costs or you pass along the savings in terms BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 275 1 of improved performance, which really seems 2 to be the way the industry has gone in the 3 last few years. 4 The final set of technology 5 characterizations I'll go into is how the 6 technology interacts with the grid. A lot of 7 people use the term "intermittent resource" 8 to talk about wind power. The wind industry 9 doesn't like the term "intermittent" anymore. 10 They want to use the term "variable." All 11 capacity is variable and all capacity is 12 intermittent for that matter. The key here 13 is that it's not -- unlike the intermittency 14 and variability of a natural gas-fired 15 combustion turbine, this is not under 16 operator control. It's based on the weather 17 obviously. 18 We model three main system impacts. 19 The other thing that people like to talk 20 about is the cost of intermittency. And 21 these impacts are not really what you'd think 22 of as costs. They can manifest themselves as BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 276 1 costs only to the extent that you're willing 2 to pay to bring the value of your product to 3 be comparable with the value of a 4 dispatchable commodity. Really what they are 5 is reductions in value of your electricity 6 relative to the value of other generated 7 electricity. 8 MR. CLEVELAND: So Chris, can I 9 just stop you there? So then you're not in 10 the business of trying to estimate what these 11 ancillary -- I mean, there's a lot of people 12 who are trying to actually measure all those 13 ancillary costs. 14 MR. NAMOVICZ: The bottom line is 15 that it's a reduction in value. The only way 16 you have a cost is when you try and make the 17 assumption that you have to make the value of 18 the wind the same as the value of everybody 19 else, and in the real world that hasn't been 20 the case. There are some things that are 21 sort of, okay, well, if you need to have 22 more -- and I'll talk about them a little BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 277 1 bit, a little bit, regulating reserves, maybe 2 there's a cost. Those actual costs in the 3 sense of how you would have to necessarily 4 reconfigure system operations the literature 5 shows are fairly small. But the big picture 6 things that really have an impact on how you 7 model wind are really value. And that's, in 8 essence, how we model them, as impacts on the 9 value of wind. You know, reduced relative to 10 the value of dispatchable generation. 11 Now, all of these impacts sort of 12 increase in magnitude as the wind gets higher 13 and higher penetration on the grid. The 14 first one I'll talk about is the reduced 15 value of energy. This is really I think the 16 biggest impact of all and it's probably one 17 that's talked about least, but it's also, 18 fortunately, the one that's probably modeled 19 the best and we have the best data on. 20 For those of you who are not 21 familiar -- oh, and by the way, the reduction 22 in value of energy is due to the fact that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 278 1 the wind isn't always blowing when there's 2 demand for power. 3 For those of you who are not 4 familiar with the load duration curve, that's 5 essentially what I've established here based 6 on a model run of several years ago. The 7 idea with the load duration curve is if you 8 take the hourly load from every hour of the 9 year and then sort of the X axis, instead of 10 it being time-wise, to be sorted from highest 11 load to lowest load. You get something that 12 looks like that. And I superimposed that for 13 each hour that we model, essentially the 14 dispatch stack, which shows from lowest value 15 to highest value or lowest cost to highest 16 cost, the types of capacity that might be 17 dispatched to meet that demand. 18 And you can see the big gray blob 19 in the middle is coal. The very light green 20 at the very bottom is wind. Most of the wind 21 power, and this is really where I want this 22 to work, is if you look from about 50 percent BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 279 1 to 100 percent of the hours, that's where 2 most of the wind power is being generated in 3 this year. And that's all coal setting the 4 value. Coal has a very low fuel cost, which 5 is really the value going to the wind is the 6 fuel cost of what's being dispatched. 7 There's a significant amount of 8 wind being dispatched with this natural gas. 9 But the other thing you notice, if you 10 imagine doubling the magnitude of that green 11 bar at the bottom, the light green bar, 12 you're going to displace what's at the top. 13 And so this area's sort of -- towards the 14 left, but not at the very left, that light 15 pink area is going to go away when you double 16 the wind value. 17 So all of a sudden, you have even 18 less of the value going to wind being 19 determined by natural gas fuel as you 20 increase penetration. And that's what I mean 21 by not only does wind have less value than 22 other dispatchable fuels because it's not BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 280 1 blowing when you want it to, but as you 2 increase penetration on the grid you're 3 further reducing the energy value of wind on 4 the grid. 5 The next one I'm going to talk 6 about is contribution to reserves. This is 7 frequently capacity credit. People often 8 wonder, well, if wind is an under-operator 9 control, how can you say that it has any 10 contribution to grid reliability operations? 11 And the bottom line is no type of capacity 12 has perfect reliability. Combustion turbine 13 might be available 95 percent of the time, 14 which means that 5 percent of the time when 15 the grid operator calls the plant operator up 16 and says, okay, we want you to turn on, he'll 17 say, sorry, we've got a problem and we can't 18 turn on. Wind, it just so happens, instead 19 of being 5 or 10 or 15 percent of the time, 20 it's 50 or 60 percent of the time. But 21 still, that means that there's 40 or 50 22 percent of the time that wind is available BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 281 1 when it's needed. 2 The problem with wind is as you 3 increase wind penetration on the grid, you 4 sort of are decreasing that value. These 5 normal distributions that I've put in order 6 show sort of the total system availability 7 distribution curve for the total system of 8 this, you know, not-so-imaginary region as we 9 model it in NEMS. And you can see with no 10 wind, the dark blue curve on the left, you 11 know, has maybe a 48 gigawatts of average 12 availability for the system. 13 And as you increase wind, you know, 14 2.5 percent of capacity, et cetera, you're 15 essentially proportionally increasing the 16 peak of that curve. But the grid operator 17 from a reliability perspective doesn't care 18 about the peak. They don't want the lights 19 to be on 50 percent of the time when you flip 20 the switch. They want the lights to be on 21 99.999 percent of the time when you flip the 22 switch. So what they're really interested is BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 282 1 this far left tail. 2 And you can see as you go from the 3 blue dotted line to the green dotted line to 4 the orange and red dotted line, which are 5 almost essentially indistinguishable because 6 they're so close to each other on that, those 7 get compressed as you get more and more wind 8 penetration. That's because these 9 distributions are getting flatter and 10 flatter. 11 And the reason for that is because 12 the first two wind plants on the system can 13 be as far away from each other as they can 14 within this region, and there's not much 15 chance that they're going to see the same 16 weather front pass by these two plants at the 17 same time or within a time frame of interest, 18 say within two or three hours of each other. 19 You put the third plant in, well, it's 20 necessarily going to be closer to either one 21 of those two. And as you get more and more 22 plants, they're necessarily going to be BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 283 1 spaced closer and closer together, increasing 2 the odds that within a time frame of interest 3 you have a weather front passing over two or 4 more plants at the same time, thus creating, 5 you know, outages and increases at the same 6 time. And your standard deviation goes up 7 and you flatten out your distribution. 8 We have a decent model for this. 9 What we need is better data. We're kind of 10 basing our data on the spatial-temporal 11 relationship on, you know, some wind that was 12 collected for one state. It'd be nice to get 13 it for 50 states. 14 The flip side to the peak load 15 reliability issues are the off-peak liability 16 issues. This is what I call surplus wind 17 curtailment. This is actually an issue that 18 pops up in Denmark, which is sort of the 19 leading country from penetration of wind or 20 at least was a couple of years ago. I think 21 it still is in terms of the utility control 22 area. In Denmark, they actually have a lot BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 284 1 of wind and they have a lot of CHP units that 2 are providing district heat at their local 3 villages. So on a cold, windy night you have 4 these CHP units running full blast because 5 everyone wants to keep warm, but other than 6 that you don't have much electrical load. 7 And then if you have the wind pick 8 up, all of a sudden you have a lot more 9 electricity on the system than you need. 10 What they do in Denmark is they're tightly 11 integrated with the Scandinavian grid and 12 they just back down the hydro plants 13 operating in Norway and everything's 14 hunky-dory. 15 But in the U.S., or at least as far 16 as NEMS is concerned, we're modeling more 17 self-contained electricity markets, where 18 there might be limited opportunities for this 19 kind of exchange. But in the U.S., you don't 20 have a lot of the CHP units, but you do have 21 a lot of steam units and coal and nuclear 22 plants that, you know, can't operate below a BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 285 1 certain operating threshold, and you don't 2 want to shut them off very quickly because 3 then they take forever to turn back on. And, 4 you know, to accommodate a 3-hour surge in 5 wind you have to take a steam plant offline 6 for 12 hours. It's probably not such a good 7 idea. 8 What we're modeling is the decision 9 the operator's going to make is to flip the 10 switch on the wind. It's no big deal to flip 11 the breaker on the wind plant and take, you 12 know, this turbine offline or that whole 13 plant offline for two or three hours, then 14 you switch it back on and they can operate 15 again. The cost to the system is really the 16 cost to the wind operator and the opportunity 17 cost. 18 He's got all this potential to be 19 generating this power and can't put it 20 anywhere. And so we use essentially the same 21 spatial-temporal data that we use for the 22 reliability model to model this aspect as BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 286 1 well. 2 There are a couple of other 3 grid-related issues that we don't model 4 explicitly in NEMS at least. The one that 5 people talk about is power quality. This is 6 something that's really being addressed by 7 the technology that's available and has been 8 available for the last five or more years. 9 So, in effect, we're implicitly including the 10 solution to this problem in it since we're 11 using actual historic cost data for wind 12 power. It's assumed that, you know, the 13 technology that will address this is included 14 in the cost at the wind plants. 15 The issue that was raised here, 16 regulating reserves, the idea that you're 17 going to have to change the real-time 18 operations. Unfortunately, that's at a time 19 scale that we're not really equipped to model 20 within NEMS. The studies that have been done 21 that I've seen suggest that at least moderate 22 or even significant penetration, the costs BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 287 1 are modest enough that I would still 2 ultimately like to get some model in here, 3 but that's not one of my higher priorities, 4 you know, in terms of modeling. 5 Now I'm going to focus some more on 6 wind resources. I like to call them the 7 three Rs of wind resources are remoteness, 8 roughness, and restriction. The upper map 9 here shows the sort of original wind resource 10 model that was developed by DOE maybe 20 11 years ago. They've since updated that. 12 We're using most of their updated 13 data, but that gives you a good idea of sort 14 of how it looks. It hasn't changed except 15 it's more detailed than that. 16 And then the bottom one is the 17 transmission data that we use, and I'll talk 18 about that on the next slide because that's 19 related to the remoteness of wind. 20 With coal or nuclear or natural 21 gas, not only can you bring the fuel to the 22 plant, but you can also bring the plant to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 288 1 the load if you need to, if that's what's 2 cost-effective. There are certainly plenty 3 of mine-mouth coal plants that, you know, 4 they bring the electricity from the plant to 5 the load or from the mine to the load, but 6 there are also plenty of coal plants and 7 other plants that they bring the fuel to the 8 load for when you don't have a choice. You 9 can't decide what's economic to do. You've 10 got to generate the electricity where the 11 wind is. 12 That's going to require a couple of 13 things. One, it's going to require that you 14 need to built a new transmission line to get 15 from wherever your ridge with the wind is to 16 the existing transmission. And then there 17 may be spare capacity in that existing 18 transmission line, but there's not unlimited 19 spare capacity. At some point you're going 20 to saturate that. You're going to have to 21 build more long-distance transmission to get 22 from that existing line to the load. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 289 1 We have a pretty good handle on 2 modeling this first step of getting to the 3 transmission line. 4 And right now we've sort of just 5 started a study this summer with a contractor 6 to try and use some work that the National 7 Renewable Energy Laboratory has done to model 8 this long-distance transmission issue. So 9 I'd like to say we have a good handle on it. 10 It's probably more accurate to say we're 11 trying to get a better handle on it right 12 now, but the work isn't done yet. 13 The roughness issue, you can see 14 this picture of myself and some of my 15 colleagues at a site in West Virginia. 16 Hopefully, it helps to illustrate some of the 17 roughness ideas. The best wind sites tend to 18 be on mountain ridges, hills, slopey land. 19 It's difficult and expensive to bring heavy 20 construction equipment up to these areas, set 21 them up so that they're level and stable, and 22 build your plant. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 290 1 The other issue is simply that, you 2 know, it's one thing -- the blades on modern 3 machines are literally as big as the fuselage 4 of a 747, and they're only getting bigger 5 soon. You know, a couple years ago, it was, 6 well, one blade was as big as the wing on a 7 747, now it's they're as big as the body, and 8 soon it'll be one blade is as big as the 9 wingspan of a 747. And they're not modular. 10 The blade has to be manufactured and 11 transported whole. 12 And I actually saw one of these 13 parked along the side of an interstate once. 14 They're bulky to transport even on an 15 interstate, but you get on a small mountain 16 road, especially a small town that you might 17 have to turn a corner in, and suddenly you've 18 got to worry about moving all the 19 streetlights back 100 feet from the side of 20 the road temporarily so you can turn the 21 corner and then paying to move them back. 22 And you have 50-ton equipment, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 291 1 50-ton cells, which is where the generator is 2 located, that have to be transported across 3 these one-lane rural bridges, so you might 4 have to actually pay to put in a new bridge 5 to accommodate your -- you know, heavy load. 6 There's not a lot of analysis that 7 have been done on this, at least that's 8 useful, in terms of NEMS modeling, but it's 9 something theoretical. I think there might 10 be data available. It might just be a 11 question of sort of going out and trying to 12 find the best available data, you know, and a 13 good analytic technique to make it useful to 14 NEMS. 15 The final R I'll talk about is 16 restrictions. There are a number of legal or 17 technical restrictions that we model 18 explicitly in NEMS by excluding certain 19 lands, including, like, national parks, areas 20 near airports, urban areas, highly sloping 21 lands, like 20 percent grade lands. They're 22 just outright excluded from our land resource BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 292 1 database. However, there's plenty of other 2 remaining land that doesn't fall under one of 3 these legal or technical exclusions. 4 And our working assumption is that 5 it's not going to be the same cost. There's 6 not a uniform cost to exploit all the 7 remainder of this land. You know, anecdotal 8 evidence has shown that there's a 9 regional -- in some places there's a regional 10 economic pull to get wind power in a region. 11 These tend to be rural areas with depressed 12 economies or, you know, not-growing economies 13 that they really want to have a good, you 14 know, steady -- the farmers want to have a 15 good steady income stream. They want to have 16 some jobs in their community. 17 There are also some, you know, 18 regions that have shown a significant 19 economic push. These tend to be what I'll 20 call the exurban areas, where you get a lot 21 people that just moved out from the cities or 22 the near-in suburbs and wanted to enjoy the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 293 1 nice bucolic setting and don't really 2 appreciate having these industrial-grade, you 3 know, 100-meter tall towers looking out their 4 back window. 5 And related to that there are also 6 the areas, like the one on the previous page, 7 near West Virginia, which is near a touristy 8 area that people go to commune with nature or 9 whatnot. 10 And once again, there's some 11 question as to whether it's good for the 12 tourist industry to have these big 13 industrial, you know, very prominent machines 14 out there, whether that's going to attract or 15 repel tourists. There are some people that 16 in fact claim that these wind farming areas 17 have turned into tourist traps. People come 18 out to see them. Of course one wonders also 19 what the saturation point on that is. You 20 know, if we have a 100 gigawatts of wind in 21 the United States and they're everywhere, 22 there's not going to be much tourist value in BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 294 1 a wind plant in your backyard. 2 The other thing that gets a lot of 3 attention is noise and bird kills. These are 4 more site-specific issues. The bird and bat 5 issue is something that has popped up in a 6 few prominent places, but given the almost 10 7 gigawatts of wind power that's out there in 8 America, it might be an issue affecting maybe 9 500 megawatts of capacity. 10 Noise is something that there are 11 rare instances where people have complained. 12 If you've been to a modern wind farm, they're 13 not noisy, but I could certainly not argue 14 with somebody that lived nearby. And even a 15 very faint, whoosh, whoosh, whoosh 16 constantly, day and night, might be a little 17 bit annoying. 18 There have been actually quite a 19 number of studies on a lot of these, but they 20 don't provide a lot of good statistical 21 basis. They don't really look at costs, per 22 se, and that's what we really need to put in BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 295 1 our model is cost. 2 So how do we model these Rs? Well, 3 like I said, there are a couple of them, like 4 the distance to grid, some of the legal or 5 technical restrictions that we have a pretty 6 good handle on modeling explicitly. The 7 remainder we have this sort of catch-all 8 thing called a long-term cost adjustment 9 factor. These are based on some 10-, 10 15-year-old studies that really are just 11 designed by some region or state to get a 12 general idea of what the economic potential 13 of their state was. And we kind of 14 guesstimated from that. Okay, well, this 15 must be -- if they're not economic, what are 16 the reasons why and what does the 17 non-economic supply look like? 18 Right now we're trying to get a 19 real good handle, like I say, on these 20 long-distance transmission costs because we 21 feel that of this sort of implied supply 22 curve that represents the low-end cost that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 296 1 will really be coming into play with the next 2 10 or 20 gigawatts of wind capacity. 3 And it's also, fortunately, the 4 area where there's existing data and 5 analytical techniques that we're able to take 6 advantage of. But in some of these carbon 7 constraint scenarios we might be getting into 8 areas of the curve where we're concerned 9 about some of these other factors such as the 10 roughness or the restrictions, the non-hard 11 restrictions that it'd be nice to get some 12 good data on. 13 So that's about all I have to say. 14 The bottom line is wind power isn't very big 15 now. It's been growing a lot. We don't show 16 a lot of growth in the reference case, 17 primarily because of policy, but we do have a 18 lot of policy cases that we have run or are 19 asked to run where wind is a significant 20 player. So it's important that we try and 21 get a better handle on some of these cost 22 impacts. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 297 1 Unfortunately, some of the things, 2 like the NIMBY factor, it might be impossible 3 to get data on until after the fact. You 4 don't really know that a community's going to 5 want or not want a wind plant until you go 6 out and propose one. 7 And so until you have a good sample 8 of not only hundreds, thousands of proposals 9 and rejections or acceptances, but somehow 10 have been able to track all of these and sort 11 of figure out even for the ones that are 12 accepted what cost did they have to give to 13 the community to mitigate their impacts. 14 Those are the kinds of things that are real 15 difficult data challenges. 16 And so I thank you for listening to 17 me and look forward to any input you can 18 provide on helping me get a better handle on 19 some of the data issues. You know, take your 20 pick, there's a lot to choose from. 21 MR. HENGARTNER: Thank you very 22 much, Chris. We have two ASA discussants, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 298 1 Walter Hill and Neha Khanna. Walter, you 2 want to go? 3 MR. HILL: Yes, although I don't 4 have -- it turns out I have very little to 5 say from the paper. There were a couple of 6 times you have comments that there's not 7 enough data, so you have to collect more data 8 and then the presentation I think that's come 9 out, pick one or two times when I've noted 10 that. 11 You mentioned (inaudible) doing 12 performance standards, you're looking for 13 more data. 14 If it comes, that's good. Well, if 15 it doesn't come, that's bad. Yes, if it 16 comes, that's good. 17 You also note points of generation. 18 And actually maybe you did mention it in the 19 talk, wind plants in Texas. You're looking 20 for data there. You're not sure what works 21 or what doesn't work. If you get a better 22 model, that's fine. If you have a better BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 299 1 example, that's fine, too. So you're looking 2 for data at that point? 3 MR. NAMOVICZ: Yes. 4 MR. HILL: And the third -- 5 MR. NAMOVICZ: We're looking for 6 lots of data. 7 MR. HILL: Yes. And similarly, 8 you've mentioned it here whether or 9 not -- you put up a wind farm at one point. 10 Maybe people like that, maybe they don't, so 11 you mentioned maybe it's a tourist 12 attraction, which means that some people like 13 it. Maybe it's a NIMBY, which means some 14 people don't like it. You won't know one way 15 or the other until you start having the farms 16 up there. 17 MR. NAMOVICZ: Right. 18 MR. HILL: I can even imagine in 19 some of the local areas, in the local 20 counties, that you might want -- having the 21 plants there would prevent or reduce the 22 amount of other development there, so it BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 300 1 might be preferred over condominiums and -- 2 SPEAKER: Jails. 3 MR. NAMOVICZ: Yes, that's 4 certainly, you know -- 5 MR. HILL: Or jails. Yes, jails if 6 you're -- 7 MR. NAMOVICZ: Which sort of 8 reinforces this trend of sort of the rural 9 areas have the pull and the exurban areas 10 have the pleasure because the exurban areas 11 have a large population that want to see more 12 development like they came to enjoy, which 13 are sprawling suburban estates. And they're 14 not -- wind farms are not compatible just 15 because you need sort of minimum setbacks for 16 occupied structures and roads from the 17 turbines to avoid ice throw and stuff like 18 that. 19 MR. HILL: But those numbers might 20 be good. 21 MR. NAMOVICZ: Yes. 22 MR. HILL: You might actually want BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 301 1 the -- 2 MR. NAMOVICZ: Oh, yes, and there 3 are certainly a lot of areas, especially 4 you've got all the wind plants that are out 5 there. And go on a site visit and talk with 6 the people that work there, the locals in the 7 community, and they like it, but then it's 8 essentially by example because they already 9 accepted the plant in their community. 10 MR. HENGARTNER: Thank you, Walter. 11 Our other discussant is Neha Khanna. 12 MS. KHANNA: Well, let me say you 13 have a tough nut to crack because essentially 14 what you're doing is you're trying to model 15 something for which there is almost no data. 16 MR. NAMOVICZ: Right. 17 MS. KHANNA: At least not in the 18 United States. My sincere apologies. 19 MR. HENGARTNER: Let me find 20 this -- 21 SPEAKER: When you see him walking 22 toward you, you know you're in trouble. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 302 1 MR. HENGARTNER: How to use the 2 microphone. 3 MS. KHANNA: So to me it seems that 4 there are two -- well, three big issues that 5 I don't have any solution for, I just have 6 recommendations for you to probably keep your 7 finger on. And honestly, it's not going to 8 solve your immediate problem. And I'm sorry, 9 but your problem is here to stay for now. 10 But the one thing that I thought 11 was very interesting actually was your 12 increase in the capital costs since 13 post-2002, 2003. And what was even more 14 interesting, which I didn't realize when I 15 read the paper, but I realized when you 16 presented it, that this is not unique to the 17 United States. It seems to be -- 18 MR. NAMOVICZ: Right, it's a global 19 industry. 20 MS. KHANNA: To a greater degree or 21 a lesser degree it's there in Europe as well, 22 which tells me this is -- even if this is BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 303 1 just a transient phase for now and your 2 presumption is correct within NEMS, it may be 3 a really important thing to keep track of. 4 Because if it's just material costs that are 5 driving this up, this could always come back 6 again. So we really want to know, try and 7 find out, what was it that caused this 8 increase in cost in terms of future upgrades 9 to your module within NEMS because that's 10 going to be important. Let's not forget it 11 even if it just goes away. Because, like I 12 say, it may come back, especially as wind 13 penetration increases. So that's just -- 14 like I say, it's not solving your problem, 15 it's a recommendation. 16 The only source of data it seems 17 right now that you may have is the 18 international market. You know, you've got 19 some historical evidence from Denmark, 20 Germany, Spain. Suddenly India is becoming 21 another place where production of wind 22 turbines is actually a big -- it's becoming BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 304 1 one of the biggest leaders in manufacturing 2 wind turbines. They've got a plant now, an 3 Indian company in Minnesota, and the 4 manufacturing plants for China. It may be 5 something worth keeping track of. I mean, 6 what's going on there? 7 In India, I don't know China as 8 well, I can say a little bit about India, 9 what makes wind competitive is that it goes 10 where -- it actually is like a point 11 generator. It's off-grid and they sort of 12 create a small local grid for a community and 13 hurray, we've got electricity where we never 14 had it before. 15 MR. NAMOVICZ: Yes, that's sort of 16 on the tangent. It's a probably that we have 17 in NEMS in general, but it's particularly 18 problematic in wind is that it's such an 19 international industry, both in terms of 20 supply. As you pointed out, not only are 21 Denmark and Germany supplying a lot of 22 turbines; Spain and now India and potentially BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 305 1 China are all potential suppliers. But in 2 terms of demand the U.S. is a significant 3 player in the world, but it's by no means the 4 biggest demand center in the world. 5 MS. KHANNA: Right. 6 MR. NAMOVICZ: And so -- 7 MS. KHANNA: So maybe the growth 8 may be coming from what are currently small 9 demand centers, but it may be worth watching 10 what happens to the technology in those 11 sectors to see how you could incorporate some 12 of those trends in NEMS as a predictor for 13 what might happen in the United States. 14 Again, given the fact that you're dealing 15 with something for which you really don't 16 have much information. 17 The third issue, which I think 18 actually may be really important in the 19 United States, is the issue of land 20 valuation, agricultural communities and 21 aesthetic valuations. And I'm going to have 22 to resort to anecdotal evidence here for now, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 306 1 which is the town where I lived there was 2 talk of them. There was -- the university 3 wanted to set up a small wind (inaudible). 4 The Kyoto (inaudible) Group on campus was 5 very active and the president had reached an 6 agreement with the students and they were 7 going to -- Cornell was going to try and meet 8 the Kyoto commitments and so on. And so one 9 of the things they wanted to do was set up a 10 small wind farm. And the community turned 11 out in hordes against it. 12 The main reason? Aesthetic. The 13 biggest thing was aesthetic. 14 The second one was bat and avian 15 issues, which I don't know actually 16 applicable to this particular site, but in 17 general it's a big thing. And the Audubon 18 Society has made big statements about it, so 19 everyone latches onto that. 20 And the third thing, which maybe 21 because this was a university town with lots 22 of people with advanced degrees and ways to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 307 1 spend those degrees, was subaudible noise. 2 They had all kinds of research on how 3 subaudible noise is very detrimental to human 4 health, potentially. And voom, Cornell 5 immediately cut back on its study -- on its 6 proposal. 7 It's anecdotal evidence, but I 8 think actually, in the United States, these 9 might become big issues. 10 MR. NAMOVICZ: Yes. 11 MS. KHANNA: And it may be worth 12 tracking. 13 MR. NAMOVICZ: Yes, I've been 14 tracking the Ithaca one when I heard there 15 was some movement there. But there 16 is -- when we first started working on this, 17 there was a whole web site by this woman in 18 Wisconsin that was just livid about the 19 proposal to put in some wind turbines near 20 her house and they raise all the same sort of 21 issues. 22 MR. CLEVELAND: Well, the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 308 1 penultimate example is the Cape Wind farm in 2 Massachusetts, which is -- you know, I've 3 been following closely because it's in my 4 backyard. So this is 130 turbines, first 5 major offshore wind farm in the United 6 States, private developers. 7 MR. NAMOVICZ: It won't be the 8 first. 9 MR. CLEVELAND: What? 10 MR. NAMOVICZ: It won't be the 11 first when it happens, if it happens. 12 MR. CLEVELAND: That's right. 13 MR. NAMOVICZ: There'll be a couple 14 ahead of it I think. 15 MR. CLEVELAND: You know, and you 16 have a state where you have the two senators 17 who have about the best environmental voting 18 records in the Senate, one who came out 19 opposed to it, adamantly -- Kennedy -- and 20 the other guy who's running for president, 21 who waffled his way through the whole 22 campaign about it. Because the fact that, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 309 1 you know -- I mean, Teresa Heinz has a house 2 in Nantucket Harbor, you know. So they'd be 3 sitting out there looking at these turbines, 4 and it's all about aesthetics and sighting. 5 And in the United States, that's going to be 6 the issue, the sighting issue. 7 MS. KHANNA: Right, and just a last 8 thing and I'll end my comments. What was 9 interesting about the Ithaca -- the episode 10 was Tompkins County now has on its 11 legislature a wind ordinance about all the 12 things that a commercial developer -- they 13 defined "commercial" as anyone who serves. 14 If I set up a little turbine in my backyard 15 and sell a little bit of my electricity to my 16 neighbor, I become a commercial developer, or 17 I could have more than I think 5 megawatts 18 and I'd just be using (inaudible) for the 19 university, so I'm not selling it to anyone. 20 But there's a whole ordinance now 21 about what you must do, all the hoops you 22 must jump through in order to set up a wind BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 310 1 farm, including view sheds, affecting view 2 sheds, including informing neighbors about 3 potential health hazards, including -- I 4 mean, it's just this huge -- very ad hoc, I 5 might add. But they're actually promoting 6 that ordinance to other counties within New 7 York state because of the renewable energy 8 portfolios and the push for wind in New York. 9 But these are issues are that are 10 going to affect you, which you're going to 11 have to keep track of somehow, and then 12 change things in NEMS. But I think, again, I 13 think the international market might be the 14 way to start off right now to see what the 15 experience was, but keeping in mind that 16 preferences do vary spatially, and especially 17 Scandinavia I think tends to be quite 18 different in terms of its preferences than 19 United States. 20 MR. CLEVELAND: But it strikes me, 21 looking at their graphic, the economics and 22 the technology is interesting, but this is BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 311 1 all about policy. Without the subsidy to the 2 federal government, this industry would be 3 dead in the water. Without the feed-in 4 tariffs in Germany and Denmark, those 5 industries would be tiny. And you show that 6 in your graph, right. Your model shows once 7 the subsidy goes away, it essentially goes 8 into stasis. It actually declines and is the 9 fraction of the total generation, right. 10 MR. NAMOVICZ: Yes. 11 MS. KHANNA: But that's in the 12 absence of a price of carbon. 13 SPEAKER: Right. 14 MR. CLEVELAND: That's what I mean. 15 That's what I'm saying, the real interesting 16 story here is not so much in the technology 17 and the economics. It's on the policy 18 instruments and the effect that that's going 19 to have. 20 And this guy, Jim Gordon, who was 21 trying to build this wind farm in 22 Massachusetts, he spent $20 million of his BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 312 1 own dollars. I had him up to BU last year to 2 give a talk and the last question of the 3 thing was, well, if it weren't for this 4 1-1/2-cent a kilowatt hour subsidy would you 5 be trying to do this? He said, oh, of course 6 not. 7 MR. BURTON: From a NEMS 8 standpoint, do capital costs capture the land 9 acquisitions costs as well? 10 MR. NAMOVICZ: No. 11 MR. BURTON: No? 12 MR. NAMOVICZ: Because -- I think 13 at one point we said that it did, but if you 14 look at this Form 412, as much as I could 15 figure out, and this is one of the big 16 questions about it not being designed for the 17 wind industry, it's not clear on the 412 how 18 they want wind guys to -- because they don't 19 buy their land typically. They lease their 20 land and they either lease on a royalty basis 21 or they lease it on an absolute annual 22 payment basis, and none of that is accounted BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 313 1 for in the 412. 2 But the costs that we're basing it 3 off the 412 does not seem to include land 4 costs. So I'll say that no, we don't include 5 that in the capital costs. 6 MR. BURTON: Let me ask a follow-up 7 question then. As there are more of these 8 facilities, as available lands for them 9 become fewer and fewer, and as people have to 10 be induced to put up with the aesthetic or 11 health or whatever, is it foreseeable that 12 over the next generation that land 13 acquisition costs for the purpose of wind 14 generation are going to escalate measurably? 15 MR. NAMOVICZ: Yes, and one would 16 assume that that's sort of the biggest way 17 that financially at least these -- you know, 18 the cost of these alternative land value uses 19 will come into play. You got to assume that 20 the guys buying land right now are getting 21 the cheapest sites first and will get the 22 more expensive sites later on. But the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 314 1 question is what goes into the cost as the 2 costs go up. There's the absolute market 3 value of the land, just a farmer values it 4 because he can produce so many acres of corn 5 on it or whatever. 6 MR. BURTON: Right. 7 MR. NAMOVICZ: But then there's 8 also the sort of non-economic value of -- the 9 farmer might only give it that value, but his 10 neighbors down the hill value having that 11 pretty bucolic view, and that's not in the 12 market captured in the cost that the -- you 13 know, they're paying it for their lease. 14 SPEAKER: It's a positive 15 extra (inaudible). 16 MR. NAMOVICZ: But it's at some 17 point going to be captured either as a delay 18 in sighting or as a lack of being able to 19 sighting or as a payment that you make to the 20 town in lieu of taxes or in addition to taxes 21 hush money if you want to call it. And 22 that's been known to happen. In fact, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 315 1 there's issues with TV reception. There was 2 this other plant in Wisconsin where they 3 actually had to pay for some houses nearby to 4 get cable because the rotation of the blades 5 was interfering, ghosting their TV reception. 6 SPEAKER: That's terrible. 7 MR. NAMOVICZ: There are all kinds 8 of costs that can come up that are hard to 9 even imagine how you capture that on a sort 10 of national scale. 11 MR. NEERCHAL: I think your comment 12 about the wind turbines being tourist 13 attractions is probably on target. There are 14 more than 100 dams on the Columbia River, and 15 I have actually gone to more than 15 of them. 16 Other than Grand Coulee, none of the others 17 are really attracting any significant 18 tourists, I can guarantee that. I think 19 there are 150 dams I believe down the 20 Columbia River. And I think Grand Coulee is 21 the only that is a significant tourist 22 attraction. And I don't know the numbers, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 316 1 but I think there aren't as many people there 2 as there are at Niagara Falls, I can tell 3 you. 4 MR. HENGARTNER: Okay. Two more 5 comments. Jae? 6 MR. EDMONDS: Just two points. 7 One, storage units seem to be an issue and 8 I'm presuming that you didn't have time to 9 cover it. But there's a -- one way to deal 10 with a variance is storage. And so you've 11 got pump storage. You've got -- you know, 12 you've got a variety of different mechanisms 13 for storage. And so I'm going to presume 14 that you've dealt with that. 15 But I have a question that there is 16 this winds model that Walter Short has put 17 together. 18 MR. NAMOVICZ: Right. 19 MR. EDMONDS: And I was just 20 wondering how the NEMS component, wind 21 component, compares with the winds model. 22 MR. NAMOVICZ: Well, what Walter BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 317 1 has done is a transmission model that he 2 superimposed wind on. And we don't, as Susan 3 mentioned this morning, we don't have a 4 transmission model. And I think it's an open 5 question, but I'm skeptical of that because 6 of the run time it takes just to do Walter's 7 simple transmission model. I understand it 8 takes several hours. You know if we tried to 9 iterate that over all of our -- 10 MR. EDMONDS: No, no, I wasn't 11 thinking of it of just dropping it in, but 12 rather to look in terms of -- 13 MR. NAMOVICZ: That's essentially 14 what we're, hopefully, going to do if we can 15 get buyoff from EE and NREL. You know, we've 16 got buyoff from NREL. It's now a question of 17 getting buyoff from EE. Our contractor's 18 going to work with Walter to kind of figure 19 out how we can sort of parameterize some of 20 his model costs and substitute them in for 21 our long-term costs, multipliers at least as 22 related to the transmission. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 318 1 There are other aspects of his 2 model that we sort of at the same time, in 3 parallel, but not really working together 4 came up very similar solutions to, like, the 5 capacity credit. And he models it probably, 6 at least spatially, at a much finer detail 7 than we do. Temporally I think we both have 8 the same problem, that we're dealing with 9 very big chunks of the year and we'd like to 10 be dealing with much finer chunks of the 11 year. 12 MR. EDMONDS: Because if you do 13 that comparison, that goes a long way in 14 addressing the question do you have a level 15 playing field? 16 MR. NAMOVICZ: Perhaps. The 17 comment I get in terms of the level playing 18 field is not so much that we're doing too 19 much to model wind, it's that the other guys 20 aren't doing enough to model the other 21 things. Since I don't model the other things 22 I won't say that that's true or false, but, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 319 1 as Susan points out, it's sort of a question 2 of working with what's there. And maybe I 3 just have -- despite the fact that I don't 4 have a lot of data, I have a lot of 5 analytical tools that I've been able to 6 deduce and put into the model. And I don't 7 know that it's not fair -- is it a question 8 of equity to say that if these tools aren't 9 available for other technologies, of course, 10 they're not using them. 11 And the other -- nuclear doesn't 12 have a winds model. There's not a nukes 13 model that I'm aware of. Maybe there is, but 14 the issues are different with nuclear anyway, 15 so. Some of them are. 16 MR. HENGARTNER: Last comment? 17 Tom? 18 MR. RUTHERFORD: Yes, I mean, this 19 is just a question. Maybe I've learned that 20 I should go and have a look at this winds 21 model because I'm kind of curious. If you 22 subsidize a lot of copper to go out to where BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 320 1 the wind fields are, which looks like the 2 upper plains, do you have to regulate that 3 you're not going to permit coal producers to 4 ship electrons along those wires? I mean, 5 because if you only have 25 or 30 percent 6 capacity utilization on the wind, you're just 7 going to let the copper sit there and not do 8 anything. 9 MR. NAMOVICZ: That's a good 10 question. The winds model does model the 11 electric sector, so they model the trade off 12 that coal guys are going to want to use the 13 tower lines as well, if that's economic. You 14 know, somehow the wind -- the EE program 15 people are always saying, oh, well, we'll 16 just go in with the coal guys to build these 17 transmissions. I give them the same thing. 18 I'm like, well, is the coal guy really going 19 to want to pay for a line that he's only 20 going to get to use 60 percent of the time? 21 He's got to use that line 80 percent of the 22 time to make his money back. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 321 1 MR. RUTHERFORD: The question is 2 what the instrument is. I mean, one thing is 3 you can subsidize construction of the line, 4 but then you also have to regulate access to 5 the line afterwards, which is a different 6 game then. Other than building the line and 7 then selling it to the highest bidder, I 8 can't imagine that the coal guy wouldn't be 9 the highest bidder. 10 MR. NAMOVICZ: Well, yes and no. 11 The coal guy can certainly bid higher if he's 12 able to produce when the value is high, but 13 the wind guy, when he's producing, presumably 14 will be able to outbid the coal guy any day 15 because he's got a lower marginal cost. And 16 that means that the coal guy isn't going to 17 want to pay for a significant part of that 18 line that he's not going to have guaranteed 19 access to because he's going to want 20 all-the-time access whereas the wind guy's 21 going to want to be able to knock the coal 22 guy off whenever the wind is blowing. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 322 1 MR. HENGARTNER: Thank you very 2 much. Very interesting talk. We're 3 fortunate we still have something like a 4 10-minute break, so please help yourself to 5 coffee and sweets. Thank you. We'll 6 reconvene at -- 7 SPEAKER: We have sweets? 8 MR. HENGARTNER: I hope. 9 (Recess) 10 MR. WEINIG: Ladies and gentlemen, 11 for the last session today we're going to get 12 started with a moment with our fearless 13 leader and then on to other things. 14 MR. HENGARTNER: Thank you, thank 15 you. I was saying I needed a drum roll to 16 get our attention. Thank you, Bill. 17 The last session of the day we have 18 the honor of having Howard Gruenspecht, the 19 deputy administrator of the EIA, giving us a 20 talk on suggestions -- or rather a report on 21 the external study team that I think we're 22 all looking forward. And essentially, I hope BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 323 1 that there's going to be some lively debate 2 and discussion of the recommendations and the 3 results of that study team. 4 If anything is an indication, this 5 morning's talk from Guy Caruso should show us 6 that indeed there's an interest for this 7 committee to hear that report. 8 So welcome, Howard. 9 MR. GRUENSPECHT: Thank you, glad 10 to be here. We didn't coordinate, Guy and I, 11 so I hope we're telling you the same story. 12 But this is the report of the 13 external study team, which I think was 14 distributed to you. Are there members of the 15 public here? Important question. 16 I think people in this group kind 17 of know a little bit about this report, 18 probably a little background. There was a 19 program, this program rating assessment tool 20 that OMB uses. And one -- when EIA did this 21 for the first time, I think back in 2005, one 22 thing that OMB came up with is you have a lot BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 324 1 of independent expert review of particular 2 reports you work on or particular -- you 3 know, using the right statistical technique 4 here or using the right -- but no one's 5 really, in a long time, looked at are you 6 doing the right things? And that was 7 something they thought was necessary. And, 8 frankly, although I think a lot of the times 9 we get things from OMB you kind of think, 10 well, we have to do this because OMB requires 11 it, I think we thought about this internally 12 about how to do it. 13 And you always have a choice, I 14 guess, as to whether you answer the mail. 15 And there are definitely things where you 16 just do it in some kind of very perfunctory 17 way and answer the mail and meet the 18 requirement. And they have a lot of red 19 light, yellow light, green light scoring 20 things. 21 And you can try to get the green by 22 answering the mail or you can try and say, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 325 1 well, maybe this would really be a useful 2 thing to do or we could get a lot out of it, 3 in addition to satisfying OMB. 4 And in the case of this 5 external -- or the notion that we hadn't had 6 external review, the question was how do 7 you -- you know, do we want to just arrange 8 something that meets the OMB requirement or 9 do we really want to have a very serious look 10 at it or have someone take a serious look at 11 it real people that we respect, and have, you 12 know -- and roll the dice, you know? Not 13 just go out for the -- sometimes when I read 14 the literature, there was a description of 15 this edition of Energy Policy coming out and 16 it sounded -- someone said, oh, is that the 17 Michael Grubb issue? 18 And I always think, like, they 19 ought to have journals in academia called the 20 Journal of Sycophantic Results where people 21 who believe the same thing review each 22 other's work and publish it. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 326 1 MR. CLEVELAND: Well, that happens 2 a lot, believe me. 3 MR. GRUENSPECHT: I know that. My 4 shot at academics, having been one, but then 5 they say it's peer reviewed, you know. Peer 6 reviewed does not equal good. And could 7 easily have gotten the review of EIA that one 8 could have been assured what the results 9 would be, but I don't think it would have 10 been particularly useful, you know. So the 11 decision was made. 12 Now, we really want a real peer 13 review. We're willing to roll the dice, so 14 to speak, not the (inaudible) house model, 15 the other dice. 16 And so came up with this idea, 17 well, how do we do this? And obviously one 18 thing, well, go to the National Academy. And 19 the answer to that was no. One, too 20 expensive. EIA's much too cheap even though 21 we give you cookies. 22 Two, I'm not convinced the National BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 327 1 Academy stuff is very good. Anyone on 2 National Academy here? Anyone else I can 3 offend? 4 But really, so we thought -- but we 5 didn't want to pick out our own reviewers 6 because that seemed like a bad idea because 7 that seems kind of silly. That's like the 8 Journal of Sycophantic Results. 9 So we came up with the notion that 10 we'd find someone -- we'd pick a chair. You 11 know, you needed a prime mover, but we 12 wouldn't pick the committee. We'd let the 13 chair pick the committee and create this 14 external study team, give them a really broad 15 charter, try to have a person who would -- 16 again, it wasn't like we were looking for 17 modeling experts to tell us we're 18 doing -- you know, that this -- in the next 19 thing we're looking for modeling experts, but 20 in this thing it wasn't like you're doing 21 your model right or wrong or you're doing 22 this -- this survey is done right or wrong. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 328 1 It was more like is the breadth of the 2 program the right stuff? 3 And we thought we needed people, 4 you know -- ideally we talked with -- by the 5 way, I asked -- I approached Denny Ellerman 6 at MIT to do this and he said no to me. But 7 then this is the advantage of -- Guy, of 8 course, is more persuasive than me. So he 9 approached Denny Ellerman and Denny Ellerman 10 said yes, he would chair it. So 11 that's -- which was great. And that actually 12 turned out to be a very inspired choice 13 because Denny -- 14 MR. FEDER: It could be the order 15 effect, too. 16 MR. GRUENSPECHT: Huh? 17 MR. FEDER: The order effect, too. 18 MR. GRUENSPECHT: The older effect? 19 MR. FEDER: Order, order. 20 MR. GRUENSPECHT: Oh, odor. 21 MR. FEDER: Maybe if you had a 22 second, maybe it would agree to you. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 329 1 MR. GRUENSPECHT: Oh, the order, 2 not the odor, okay. I thought, hey, maybe 3 they're telling me something here, you know. 4 But anyway, Denny agreed to do 5 this, which was very good, I thought. And he 6 picked -- you know, we said -- again, we 7 wanted to get people who had experience in 8 using our stuff. You know, we definitely 9 didn't want people who just used the 10 modeling, didn't want people who just used 11 the data, didn't want people who had some 12 government experience, people who had some 13 energy industry experience. 14 You go find people who'll do this, 15 who will do this for -- you know, we'll 16 provide them sort of a minimal honorarium 17 because, again, oh, we're not going to do 18 that at the -- NAS is like cheapest thing in 19 the store is half a million dollars half a 20 million and up for the National Academy. And 21 we just were not in that league. We were 22 more in the 100,000 and under league for the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 330 1 whole deal. 2 And so we kind of said find -- and 3 he fought the committee and persuaded 4 his -- the members he got were Paul Joskow 5 from his institution. He got Phil Sharp, who 6 had been chairman of the Energy and Commerce 7 Committee, Subcommittee on Energy and Power, 8 back in the '80s, now the chairman of 9 Resources for the Future, who was at the time 10 at Harvard. And he agreed to do this, also. 11 Kathy Cooper, who was the undersecretary of 12 commerce for economics and statistics, 13 basically in charge of the Census Bureau, the 14 Bureau of Economic Analysis, also had worked 15 for ExxonMobil, so had industry experience; 16 and Jay Hakes, ex-administrator of EIA, who 17 had long experience in government, at state 18 government, and a whole lot of -- so I 19 couldn't even imagine that he could get these 20 people and persuade them to do this. And he 21 did and it was great. And they put a 22 tremendous amount of time into it. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 331 1 Again, the National Academy, 2 sometimes I feel like the report is written 3 by the academy staff, not by the members. 4 No, I keep on insulting the National Academy 5 and he's not here. 6 But in any event, it was sort of 7 like we thought it would be significantly 8 cheaper than National Academy, and it was, 9 and we'd get more real value out of it, and 10 we did. The members took it really seriously 11 and they came up with this report, and that's 12 the report that you have. 13 And we shared it with OMB and we 14 shared it with the department. We shared it 15 with our appropriations committees. And I 16 guess -- I mean, I'd really be here a long 17 time, but I'll just focus a little bit on 18 the -- you know, a summary. You have these 19 seven which is really written as a very tight 20 Executive Summary, I guess; has these I guess 21 seven recommendations or seven points. 22 And I guess in terms of other than, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 332 1 you know -- we're gratified that they 2 concluded that EIA does a lot of good things 3 and that we have a lot of authority in terms 4 of what we do within the broad area because 5 we really operate under this quite 6 extraordinary general authority. And they 7 felt that given the resources that the 8 selection of the "what" is they thought quite 9 reasonable. And that was gratifying because, 10 again, given who these people were we thought 11 that was -- I was pleased to get that. 12 A real interesting question came up 13 about our analysis -- our involvement in 14 policy analysis and forecasting, and that's 15 some of what you've been discussing today. 16 And frankly, my sense is that some of 17 the -- like, someone like a Kathy Cooper, who 18 ran a statistical agency at commerce, you 19 know -- what's interesting, I mean, EIA is 20 part of the statistical -- yes, we're part of 21 the Interagency Council on Statistical 22 Policy, so we go to their meetings. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 333 1 You know, there are 13 federal 2 statistical agencies. You know, obviously 3 Census, but then there's the Bureau of 4 Justice Statistics and the Bureau of Health 5 Statistics and lots of national -- NASS, the 6 National Agricultural Statistics Service. 7 And what's interesting is most of them, they 8 really -- BLS, Bureau of Labor Statistics, 9 they just do statistics. You know, they do 10 some maybe analysis, but they certainly don't 11 a lot of forward-looking projections, and EIA 12 is set up to do that. And the group had a 13 rather spirited discussion of that. 14 And the second, I guess, point was 15 they concluded the involvement in policy 16 analysis and forecasting is appropriate even 17 though unusual among federal statistical 18 agencies. And it is unusual, frankly. But 19 it was -- you know, again, I thought that was 20 an interesting discussion. And part of that 21 feeds into our desire to pursue the -- if 22 they think we should say involved in it -- in BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 334 1 other words, if you own a tractor and you're 2 getting out of farming and the tractor's real 3 old and kind of creaky maybe you'll just put 4 up with it. But if you decide that you're 5 supposed to be in farming, then maybe you 6 need a new tractor. And so, in some sense, 7 some of the discussion you've had earlier 8 today about work on modeling sort of -- they 9 think we should stay in this business and so 10 we're staying in the business. 11 And then on the general discretion 12 they were pretty happy, on the EIA. We're 13 doing the right things. But then, since this 14 wasn't the review of sycophantic reviewers, 15 they said there were three EIA products that 16 have been specifically mandated by 17 Congress -- financial reporting system, 18 voluntary reporting of greenhouse gases 19 survey, and the alternative-fuels vehicle 20 survey -- that they thought should either be 21 discontinued or transferred out of EIA. And 22 they think they're inappropriate activities BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 335 1 for EIA and they divert resources from higher 2 priority activities. 3 And it's interesting because I 4 guess the mental model of that is that you 5 have a pool of resources and you can identify 6 things that you're spending that pool of 7 resources on that might be low priority, and 8 you could then take those items off the table 9 and you could then divert those resources to 10 a higher priority activity. But often 11 sometimes the resources are actually tied to 12 the particular activity. 13 Now, I saw what this group was 14 doing, but something like the 15 voluntary -- I'll just give one example. The 16 voluntary reporting of greenhouse gases 17 survey, that is an unusual survey for us. 18 That's where companies take actions that they 19 claim reduce emissions below what they would 20 otherwise be and file reports with EIA. And 21 it's a program that the administration has 22 put a great deal of weight on. And the way BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 336 1 the program is set up, the department's 2 policy function writes the guidelines for 3 this reporting program, but we're the 4 implementing agency to receive these reports. 5 And they kind of said, gee, this isn't like 6 either running NEMS or it's not like 7 collecting oil or electricity data. 8 And they're right, but the issue of 9 that comes up. And when they were wrestling 10 with it, it's not clear -- it's clear to us 11 that the administration wants this thing to 12 continue, that they value it pretty highly, 13 that if it were to be moved the money to 14 carry it out would move with it, so it isn't 15 like you'd be able to divert the resources to 16 higher level things. So, again, this is not 17 taking issue with the recommendations. But 18 one of the interesting things, when you get 19 recommendations you have to think about what, 20 you know -- that it's not just, first, we 21 determine the level of resources and then we 22 determine the level of resource allocation. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 337 1 So we are struggling with some of these 2 recommendations, but we're taking them all 3 very seriously I think. 4 MR. CLEVELAND: On that point. So 5 would EIA be worse off if we dumped the 6 greenhouse gas thing and the money that went 7 with it? 8 MR. GRUENSPECHT: I don't know, 9 that's an interesting question. I mean, 10 that's why I asked about -- I mean -- 11 SPEAKER: You can't (inaudible) 12 been mandated. 13 MR. GRUENSPECHT: That's right, 14 it's sort of been mandated. It's sort of --- 15 MR. CLEVELAND: Well, let's 16 say -- but if you pushed hard and got it 17 shifted to some other agency, I mean, would 18 you be worse off for it? I mean, no one 19 likes to have a lower budget. 20 MR. GRUENSPECHT: But we were not 21 be worse off if it were shifted. I mean, 22 this is like stepping out of the EIA hat, but BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 338 1 I can be on the record. I mean, if it were 2 shifted out of the department -- the 3 reason -- I happen to know why it was put in 4 EIA. It was before I was in EIA. It was in 5 1992, when they passed the Energy Policy Act. 6 And there was a jurisdictional 7 fight between the Senate Energy and Natural 8 Resource Committee which wanted authority 9 over climate issue, which is not exclusively 10 an energy issue, but it's clearly very tied 11 in terms of greenhouse gas emissions to 12 energy -- although, again, there are other 13 greenhouse gases that come from other 14 sectors -- and the Environment of Public 15 Works Committee, which has the jurisdictional 16 authority over the Clean Air Act and other 17 things. 18 At that time, it was perceived by 19 people in the energy industry and people in 20 the Energy Department and people in the Bush 21 41 administration that given their 22 perspective they would be happier if BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 339 1 jurisdiction or some jurisdiction over 2 climate was in the Energy Committee. So in 3 the Energy Policy Act of 1992, this Title 16, 4 which deals with climate change, was put in. 5 Now, that committee, when they 6 write legislation, it has to be about the 7 Department of Energy. When the other 8 committee writes legislation, they write 9 about EPA. 10 Now, interestingly enough, the 11 renewable fuels program is in EPA even though 12 that's an energy thing, sort of like the same 13 thing in reverse. That thing was attached to 14 the Clean Air Act the requirement to put 15 oxygenate in fuel, which was the original way 16 that ethanol was put into fuel. So there's a 17 lot of, like, stuff that doesn't make any 18 sense, but it's hard to move stuff and it 19 would be giving up jurisdiction that some 20 people want energy, of which EIA is a part, 21 to have. 22 So that's all beyond I think my pay BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 340 1 grade, but it's sort of a non-answer I guess 2 to your question. But it's awkward, you 3 know. It's an awkward thing and we have to 4 deal with it somehow. But for now, we still 5 have it. You know, it was in our budget. 6 The alternative-fuels vehicles 7 thing, I mean, the question there -- their 8 thought there in more detail, and there's not 9 a lot of detail in this, but you could spend 10 a whole day discussing this report, was that, 11 gee, most of your other programs you collect 12 information on fuels, not on energy, not on 13 the capital -- you know, we don't have 14 a -- although we have consumption surveys we 15 don't take a census of refrigerators or, you 16 know -- the issue is not how many fuels are 17 out there that can burn ethanol. I mean, we 18 know from listening to the radio commercials. 19 I don't know if they run these 20 commercials only in Washington or elsewhere 21 in the country, but GM is running some set of 22 commercials talking -- you know, classifying BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 341 1 vehicles that can burn gasoline or ethanol, 2 which a lot of them were built because they 3 got extra-special credits under the CAFE 4 requirements, but they're now classifying 5 those as alternative-fuel vehicles. But 6 counting up all these vehicles as vehicles 7 that can burn ethanol, even though none of 8 them do their point was why are you wasting 9 your time with this vehicle census? Why 10 don't you try to figure out more about 11 renewable fuels? 12 And I think that is actually 13 compelling. And unlike the greenhouse gas 14 one, I'm probably really hanging myself here 15 we might be able to -- all I want to say is 16 this. All these recommendations are very 17 thoughtful. We have to think about which 18 ones we can kind of do on these three 19 specific recommendations. 20 Then there were more general 21 recommendations about performing 22 discretionary analysis and greater BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 342 1 interaction with the broader energy research 2 community is also recommended. And I think 3 that's something that's come up in this group 4 a fair amount. And making more of our data 5 available for academic researchers, and 6 that's something that I think we will be able 7 to do and we need to do. 8 MR. FEDER: I have a question. I 9 just -- 10 MR. GRUENSPECHT: I'm 11 filibustering, but you can have the floor 12 anyway. 13 MR. FEDER: How many hours do 14 people put into this? I mean, how much work 15 has gone into preparation of this? 16 MR. GRUENSPECHT: A lot of work 17 went it, amazingly much. I mean, this is, 18 like, to me -- I think they came in for sort 19 of an organizational meeting. This was the 20 year, by the way, that Denny Ellerman 21 was -- he had one semester in Oxford and one 22 semester at the Sorbonne, but this was his, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 343 1 like, reward from MIT for having been the 2 director of their Energy and Environment 3 Policy Center and the joint -- he also was 4 director of the Joint Center on the Economics 5 and Science of Climate Change. He was 6 director of both those centers, executive 7 director of both those centers for, like, 12 8 years or something. Something like that. I 9 remember since the early '90s. And he's 10 still associated with them, but he -- they've 11 now brought in someone else to do that. 12 And he came in and they had an 13 organizational meeting, like a day's worth of 14 meetings, kind of made assignment among 15 themselves. 16 SPEAKER: I think it was in July. 17 MR. GRUENSPECHT: In July. And 18 then they did obviously a lot of e-mailing 19 back and forth. And then they came in for a 20 two-day meeting before Thanksgiving and did 21 that and had pretty intensive discussions 22 among themselves. And then again, were BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 344 1 constantly communicating and working on the 2 report. So there's been a lot of effort. 3 MR. FEDER: But that's just Denny, 4 right? What about -- 5 MR. GRUENSPECHT: No, the others. 6 MR. FEDER: Oh, the others. 7 MR. GRUENSPECHT: No, no, they 8 didn't come in to meet with us. They -- we 9 provided space for them to meet with each 10 other. This was not our committee. And, 11 frankly, some of the meetings they invited us 12 into to ask us questions, but it was their 13 meeting, not our meeting. It wasn't like we 14 were meeting with our consultants. It was -- 15 Denny set the -- 16 MS. KIRKENDALL: Denny set the 17 agenda (inaudible). 18 MR. GRUENSPECHT: The agenda, when, 19 who should -- you know, if they wanted us to 20 be there so that we could -- it was a meeting 21 among them. In some cases, they wanted us 22 various EIA people in there to ask us BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 345 1 questions. But it wasn't our meeting, it was 2 their structure. And Denny, I guess, worked 3 with his fellow committee members to -- 4 MS. KIRKENDALL: His fellow 5 committee members didn't do as much, but 6 Denny did a lot. Denny really worked hard. 7 We had an EIA support team that provided 8 anything they wanted and we did draft the 9 first report, but he totally rewrote it. He 10 did not take our words at all; he wrote it 11 himself. I mean and we tried to do it 12 according to his outline, but those are his 13 words. He did it. 14 MR. GRUENSPECHT: No, it's 15 really -- it's a very -- I thought he really 16 put a lot into it. And so on this -- 17 MR. CLEVELAND: So what bad stuff 18 did they say? 19 MR. GRUENSPECHT: Well, I mean, 20 they didn't like these three surveys. They 21 feel like on the -- they thought the analysis 22 was being underdone. They thought that the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 346 1 interactions with the outside community with 2 the data were underdone. 3 And that the quality of the -- this 4 notion that the analysis -- more analysis 5 kind of helps you, it's not just analysis 6 being -- I don't know how to explain this. 7 My sense of EIA, I've only been 8 here for three years, but my sense is that 9 some people associate the modeling with the 10 analysis, and then the data programs as not 11 having any analysis. And their view was the 12 people involved in the survey collection and 13 survey collection of data ought to be -- even 14 if they're not making 30-year forecasts, they 15 need to be -- you don't understand the issues 16 unless you're -- 17 MR. CLEVELAND: I mean, the people 18 who are using the data -- or collecting the 19 data are cohabitating and working with the 20 people who are also using it for analysis. 21 MR. GRUENSPECHT: Right. 22 MR. CLEVELAND: It's just a much BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 347 1 richer environment. 2 MR. GRUENSPECHT: Right. 3 MR. CLEVELAND: And the two feed 4 off each other and you get a much better 5 product. 6 MS. KIRKENDALL: Yes, and they said 7 that you shouldn't be hiring contractors to 8 do the research. You have your staff do it 9 because you want to train your staff and get 10 them to -- and build capacity in your staff. 11 MR. CLEVELAND: I think that's a 12 big -- I think that's a real issue for you 13 guys. And I know you've -- you know, a lot 14 of it's been budgetary, but -- 15 MR. GRUENSPECHT: And part of it's 16 been a resource issue. And then the question 17 is what do you do? 18 MR. CLEVELAND: The use of 19 contractors really detracts from sort of the 20 cohesiveness of the product envisioned. I 21 see that in the oil and gas thing which I've 22 followed for 15 years. It's become, I BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 348 1 think -- it's been -- the quality of the 2 work, while still very high I think has been 3 diminished over time by the increased 4 reliance on outside contractors. 5 MR. GRUENSPECHT: Right. 6 MS. KHANNA: If I can just further 7 that. I think the point of doing more of 8 this work in- house now, the timing is really 9 good because of the turnover coming up. 10 MS. KIRKENDALL: That's right. 11 MS. KHANNA: You're going to get a 12 bunch of young, new people. If they get 13 trained around this you've got a long-term 14 payback. 15 MR. GRUENSPECHT: Although, I mean, 16 I agree with that. I mean, and the issue is 17 I don't think we'll keep the -- the people 18 who come in will not have the same careers in 19 terms of we have an incredible number of 20 people who have been here 30 years 35 years. 21 That's never going to happen. I still think 22 we should do more research, but it isn't, you BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 349 1 know -- I mean, in that regard I don't think 2 the people are going to be as attached to 3 EIA, but that's no reason not to do it. So 4 that was a place there where I 5 think -- although they put it in a positive 6 way, that was a place where I think they 7 don't think the status quo is very good. 8 And, again, this -- 9 MR. RUTHERFORD: Why aren't people 10 going to make careers at EIA? 11 MR. GRUENSPECHT: I just think the 12 nature of the federal employment system, in 13 the old system in 30 years you'd get 14 60 -- you know, the -- once you were here 15 15 years or something, there's a very strong 16 financial incentive to stay the rest of your 17 career here. I just don't think people's 18 work lives -- it's not an EIA thing. It's 19 not EIA. 20 It's just the federal pension 21 system has changed, so it's much more of a 22 defined contribution -- much larger defined BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 350 1 contribution element, like a 401(k) type 2 element, and it's much more significant than 3 it has been in the past when the defined 4 benefit element was a whole thing. And I 5 think the defined benefit element, after 6 you've been here 10 or 12 years, you've 7 really got a strong incentive to stay with 8 the federal government that the people coming 9 on now will not have. But that's specific to 10 the federal government. 11 I also think throughout the economy 12 generally, but just as a personal impression, 13 the notion of I'll go work for IBM straight 14 out of college and I'll retire from IBM or 15 I'll go to work with GM straight out of 16 college and retire from that, that just 17 doesn't happen anymore. People are -- so 18 that's -- it's a combination of the specifics 19 of the federal -- you know, now maybe EIA 20 could be a uniquely attractive place in some 21 ways. 22 MR. CLEVELAND: Well, that's the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 351 1 whole point. I don't think -- 2 MR. GRUENSPECHT: And we're trying 3 to make it as -- right. 4 MR. CLEVELAND: You don't need to 5 have this longevity to -- if EIA became known 6 as a place where this kind interesting 7 analytical work was done -- 8 MR. GRUENSPECHT: Right. 9 MR. CLEVELAND: Even if people only 10 stayed for 5 or 10 years, you would still 11 attract the sort of person that you want. 12 MR. GRUENSPECHT: Oh, absolutely 13 and that's what we're trying to do, right. 14 MR. CLEVELAND: We don't -- 15 MR. GRUENSPECHT: Right, right. 16 And I think that's right. And, again, part 17 of it is a -- not a game, of convincing 18 people -- I mean, a lot of -- it's 19 interesting, a lot of people want to 20 work -- young people out of college, they 21 want to work in policy. And I -- not the 22 policy office, but they want to work in a BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 352 1 place that does policy. 2 I think what a lot of times people 3 don't understand is that if you go to work as 4 a young person in the place that supposedly 5 does policy, you don't do policy, you know. 6 You carry someone else's briefcase for 7 a -- whereas if you come to a place like EIA, 8 I think a lot of our -- we find a lot of our 9 younger people who really get immersed in the 10 nitty-gritty of the oil and natural gas thing 11 or -- they have incredible opportunities, 12 that coming here for 5 to 10 years is 13 actually a smarter proposition than going to 14 do policy where you're not going to policy 15 for 5 or 10 years. And sort of be more of 16 a -- perhaps more of a dilettante's 17 credentials than a -- so I think 18 substantively that's right. I mean, it's in 19 EIA's interest to make that case. I also 20 personally believe that it's true. 21 And the other thing that's very 22 frustrating is oftentimes working in other BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 353 1 types of functions you do -- can do very good 2 work that ends up going nowhere because the 3 flavor of the month changes monthly. You 4 know, that's why it's the flavor of the month 5 in terms of, well, politically it doesn't 6 work sorry. Whereas our stuff, it gets out, 7 you know. And that's a very positive thing. 8 I think there are very strong things to 9 attract good people, but we need to do that 10 more. 11 But anyway, this -- and then the 12 interaction with the broader energy research 13 community is recommended, and we plan to do 14 that I think. Some of this does tie to 15 resources. And we've talked before about the 16 trade offs between the number of things we do 17 versus how we do them when you want to -- but 18 I think the prospects are much better. 19 The -- if Congress ever acts on the 20 FY 2007 budget, there's a -- the 21 administration request was for a -- which 22 started a couple days ago, the administration BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 354 1 request was for, like, a $5 million increase 2 for EIA, which I think is unusually large for 3 EIA, going from, like, 85 to 90. The House 4 went along with the administration request. 5 The Senate hasn't done a bill yet, but the 6 Senate committee markup, they went for $8 7 million above the 2006. They added another 8 $3 million on top of that. 9 So the two -- the House and Senate 10 positions on the '07 is a $5 million increase 11 on the one hand and an $8 million increase on 12 the other hand. 13 The 2008 thing is just getting 14 going, but I think there's a potential for 15 quite a -- for the administration to suggest 16 a further very significant increase in EIA 17 funding. And that will make some of the 18 things possible, I think. 19 Some of it will be to deliver new 20 things. And oftentimes, when you're trying 21 to get resources, you have to say you can get 22 this. It's like if you had a house or BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 355 1 something, you know. If you said, gee it 2 needs to be cleaned up or something, or the 3 floors need to be scraped or something, 4 that's not very, I don't know, I'll use the 5 word sexy in some ways. So it's hard to sell 6 that within the process. If you say, oh, we 7 can give you this great new thing that's 8 easier to sell sometimes. 9 But I think a lot of what we need 10 to do is actually work on the core. So we 11 ask for more money not because we're going to 12 do five more surveys or because we're going 13 to do -- but we're going to do more analysis; 14 the same number of surveys. We're going to 15 work with the broader academic community, 16 which means we've got to find a way to put 17 our data out in data centers where the 18 academic community can work with it those 19 types of things. I think there has been a 20 receptivity to that, so the resources are 21 coming. 22 MR. FEDER: So if you get another BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 356 1 10 percent, you won't only have problems with 2 retention, you'll need to hire quickly a lot 3 of people to do those programs. 4 MR. GRUENSPECHT: We will need to 5 hire some people. We are hiring. We 6 hired -- in the last year, we probably hired 7 close to 30, I'd say. We are hiring them and 8 we've actually put money in -- frankly, we've 9 actually put money into recruitment in a way 10 that we never had before. I guess our view 11 has been that the announcements you have to 12 write these federal job announcements, I 13 mean, those can't be used as the documents to 14 get people interested. Ultimately, someone 15 who is interested will have to complete one 16 of those -- read and complete the federal 17 application. But we have to prepare sort of 18 an alternative set of recruitment materials. 19 We actually have brought someone on to do 20 that and are doing that and are much more 21 active in that regard. 22 We're recruiting our entry BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 357 1 level -- I mean, this is really off the field 2 of this, but it used to be that each office 3 when they had an entry level job, which we 4 define as, I don't know, the job set -- for 5 those of you who know the federal government, 6 like Grade 7 to Grade 12. Those would be the 7 entry level jobs. Each office, when they had 8 one, they put in a request to recruit for it. 9 Now we recruit three times a year, 10 which turned out to be the max you can do. 11 You know, put out announcements for 12 economists, survey statisticians, 13 mathematical statisticians, and operations 14 research analysts at GS-7 through 12 levels. 15 And our thought is that when we're looking 16 for a GS-7 through GS-12 survey statistician, 17 we're not looking for the electricity office 18 for someone who knows a lot about 19 electricity, you know. At those levels, 20 we'll teach them about electricity. Or the 21 gas office, they don't need someone who knows 22 about gas. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 358 1 So the notion is we put out those 2 announcements as a regular release. We get 3 all the responses in. All the offices look 4 at those certs, which are the lists of people 5 you can hire off of, and we're finding that 6 to be fairly effective. And that justifies 7 this recruitment strategy that we have in 8 mind. 9 So we open the announcements in 10 October for a full month. And then we have 11 to close it for 30 days, because if you don't 12 close it -- no, they're closed for 90 days. 13 But if you don't close it for 90 days, all 14 the people who were on the list before are 15 automatically on the list again. So if you 16 want the system to flush, you have to -- so 17 because of that requirement we can only do it 18 three times a year. So we do it in October, 19 we do it in February, we do it in June. 20 And that's the -- so there's really 21 a plan, you know. It's like madness and I 22 never thought that I'd be involved in BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 359 1 anything like this in my life, but we really 2 are trying to do that. 3 MR. NEERCHAL: I think a couple of 4 ideas generated earlier by the committee in 5 this regard is to go to the, for example, 6 American Statistical Association conference. 7 MR. GRUENSPECHT: Right. 8 MR. NEERCHAL: And put up a 9 recruitment booth there or, similarly, where 10 the -- you know, the national level meetings 11 in economics and energy. 12 MR. GRUENSPECHT: Right. 13 MR. NEERCHAL: And find places 14 there and put -- you know, make your presence 15 felt year after year after year as a very 16 good -- 17 MR. GRUENSPECHT: Yes, and we are 18 headed that way. We are headed that way. 19 But anyway, so that's that. 20 There was an issue about IT, about 21 remaining separate from that of the 22 department. That was a battle, sort of a big BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 360 1 battle over the last few years, but that 2 actually seems to have worked out the way 3 that these guys would prefer it anyway, that 4 it remains separate. 5 And then this notion of integrating 6 data that isn't our own data into our work. 7 Again, that's an area where they feel we 8 haven't done enough, that EIA could be -- an 9 example might be the data that the 10 independent system operators have on 11 electricity. Not necessarily that we bring 12 it in, but that somebody who hits the EIA web 13 site we should have links to the data on 14 energy that we don't collect that is relevant 15 to someone interested in energy data. So the 16 web site would become the clearinghouse not 17 only for our own energy -- the energy data 18 that we collect, but would also point to 19 other sources of -- which, again, we think is 20 a good idea and I think we will -- 21 MR. HENGARTNER: But there is a 22 danger in this because if you put a link on BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 361 1 your web page, you are endorsing that -- 2 MR. GRUENSPECHT: Right. Well, 3 there's a whole OMB thing -- 4 SPEAKER: The link to the NYNEX 5 page is not -- I mean, which is another big 6 source of data, which you don't use, which is 7 out there. 8 MR. GRUENSPECHT: Right. 9 SPEAKER: It's not -- you know -- 10 MR. GRUENSPECHT: It's tricky. No, 11 but you're raising a good point. 12 MR. HENGARTNER: It's highly 13 controversial. 14 MR. GRUENSPECHT: I mean, you can 15 tell someone who's in government versus 16 someone who's not in government. I mean, 17 because really there's always -- and -- 18 MR. HENGARTNER: There's liability 19 here, yes. 20 MR. GRUENSPECHT: And Nancy has sat 21 in on these discussions with OMB about 22 these -- you know, there are these guidelines BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 362 1 that they put out for when you can link to a 2 web site and when you can't and all that. 3 But, again, I'm talking about -- these are 4 the committee's recommendations, not ours and 5 so there are some issues. 6 But so I think we're at the stage 7 where I think the report has been widely 8 disseminated. I think given the nature of 9 the people involved and their experience, I 10 think it's having some traction. I think 11 there's some receptivity out -- it's hard, 12 you talk about how to separate the effects of 13 different things. I mean, there's more 14 interest in what EIA does I think in part 15 because of this maybe or in part because 16 energy prices have been a little high how to 17 separate the effects of what's what or making 18 people think this is more important. I can't 19 say what has played what role, but I think 20 there are lots of good things in here. 21 I mean, maybe Guy should speak to 22 what he thinks. And we don't feel like we BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 363 1 have to slavishly do each one, you know. But 2 I think we are -- I think they gave us a lot 3 of food for thought just like you guys give 4 us a lot of food for thought. 5 MR. NEERCHAL: Could I ask you to 6 comment on the -- point number two, they're 7 talking about increasing market sensitivity 8 of EIA's data releases and the principle 9 economic indicator status for some of the 10 data. Could you comment on it? 11 MR. GRUENSPECHT: Yes, I mean, 12 they're talking about things like -- I mean, 13 energy was pretty much a -- I guess when EIA 14 got started in the mid-'70s or late '70s they 15 really, I mean, weren't big -- there weren't 16 giant futures markets in oil. It was all -- 17 I mean, Bill Helkie knows a lot about this. 18 It was all like, you know -- 19 MR. CLEVELAND: It was just a daisy 20 chain market. 21 MR. GRUENSPECHT: A daisy chain 22 market. There was no -- not a lot of BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 364 1 organized markets. All of the electricity 2 and gas were fully regulated. So you put out 3 data on prices, you were really putting out 4 data on regulated rates, which are no -- I 5 mean, it's important to know how -- what 6 people are paying in State 1 compared to 7 State 2, but there wasn't anything riding on 8 it, you know. I mean, now, for instance, we 9 do -- our weekly numbers on oil come out 10 every Wednesday morning and our weekly 11 natural gas storage numbers, which is not 12 something we even did when EIA was created. 13 And that's really the only data -- on the gas 14 side, that's really the only data there is on 15 sort of the balance in natural gas. 16 You know, you can look at the 17 storage inventory as the level of the 18 reservoir. And so, like, you can't, on a 19 weekly basis, check how much water people are 20 drinking and you can't, on a weekly basis, 21 check how well the streams are running. So 22 you just go look at this reservoir every week BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 365 1 and see how high it is and try to make some 2 judgment about gas prices. 3 And, for instance, natural gas 4 prices it's for $14 a million BTU in 5 December, I guess they're now like $4 a 6 million BTU, and part of that is because of 7 high level of gas storage. So when this 8 report comes out, if the consensus thinking 9 is that it was going to add 90 billion cubic 10 feet, but you add 108 billion cubic feet, and 11 it's happened at a time when storage is 12 already really high, this happened like 2 or 13 3 weeks ago you can see -- you can watch the 14 NYNEX thing and you can see at 10:30 blink 15 it'll drop or it'll go up. 16 And obviously you all read the 17 different parts of the paper, but this whole 18 Amaranth capital hedge fund, which collapsed, 19 I mean, although it wasn't our goal we did 20 it, you know. So, I mean, that's a sense in 21 which in the '70s there was nothing we 22 released that had that kind of an impact BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 366 1 because of markets. Either they were 2 regulated markets or no markets. And now 3 they're really -- you know, the amount of -- 4 but what I find interesting is you can go to 5 the national -- I don't want to -- go across 6 the street to NASS and they have what they 7 call -- National Agricultural, and they have 8 the lockup, which is like a $20 million 9 facility. 10 It's like a vault where they lock 11 the people in when they're releasing the pork 12 bellies or the wheat or the -- you know, they 13 have what they call critical releases. And 14 they literally lock the people in overnight 15 and they have, like, armed guards on both 16 ends. 17 And they have their own 18 telecommunications switch independent from 19 the Department of Energy. And they let the 20 reporters come into this thing so they can 21 read the thing before the release time, but 22 they have the telecommunications switches BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 367 1 disconnected. And so they can write their 2 stories and then at 10:30, or whatever time 3 they allow these people to -- and, like, 4 these markets -- you know, this market for 5 cotton is like this, whereas every week we're 6 putting out numbers on natural gas and oil 7 and the market is like that in terms of how 8 many dollars are in the market. 9 SPEAKER: What about cell phones? 10 MS. KIRKENDALL: However, 11 agricultures are called principle economic 12 indicators and ours aren't. 13 MR. GRUENSPECHT: They take them 14 away. 15 SPEAKER: Oh, okay. 16 SPEAKER: They take it away. 17 MR. GRUENSPECHT: Right. 18 MS. KIRKENDALL: So the 19 recommendation was that we should make ours 20 principle economic indicators. 21 MR. GRUENSPECHT: So the 22 recommendation is we should treat them that BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 368 1 way. One of the problems is we're not -- 2 infrastructure-wise, we're not really set up 3 to -- 4 MS. KIRKENDALL: Well, actually we 5 are for -- 6 SPEAKER: This sure would be a good 7 situation to put in your -- 8 MS. KIRKENDALL: But we are for 9 natural gas. We're doing it -- 10 MR. GRUENSPECHT: We are. We are 11 for natural -- for natural gas, we -- 12 MR. BURTON: Does Congress have to 13 make that? 14 MR. GRUENSPECHT: Congress has to 15 make that. There's a lot of history. Again, 16 like the story of the committee and the wrong 17 committee, I mean NASS's -- 18 MS. KIRKENDALL: No, OMB decides 19 what's a principal economic indicator. 20 MR. GRUENSPECHT: Right, OMB 21 decides what's a principal economic 22 indicator, but what kind of funding you're BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 369 1 going to get for security and the like, it's 2 Congress, you know. 3 MS. KIRKENDALL: That's Congress. 4 MR. GRUENSPECHT: So, I don't know, 5 but that involves that. But effectively, 6 these are some of the most market-sensitive 7 numbers. 8 MS. KIRKENDALL: We have actually 9 requested OMB to make our weekly natural gas 10 storage report (inaudible) a principal 11 economic indicator. It's sitting on Cathy 12 Wallman's desk now. It probably won't happen 13 this year. It might happen in a year. They 14 haven't given that status to very many new 15 series, so this is something they haven't 16 done for a long time either. I don't see any 17 reason why they won't for that one. 18 MR. GRUENSPECHT: But this is an 19 example of something that EIA didn't do at 20 all. We didn't collect that storage data 21 until four years ago. And the committee 22 generally felt that we're doing the right BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 370 1 thing here, but they said you should 2 actually, given the importance of this stuff, 3 you should actually seek the formal 4 designation. That is what the -- 5 MS. KHANNA: One question. One of 6 the things we've talked about in the past 7 when it comes to the market sensitivity of 8 some your data was some of these data, I 9 mean, there are people waiting. There are 10 big financial decisions that are made based 11 on your data. We've at least -- I don't know 12 if we formally recommended or we talked about 13 charging for those data. And saying after, 14 let's say, a 12-hour lag when the financial 15 values data are no longer great because then 16 they're open and available to the public, but 17 for that instantaneous access -- about let 18 the people who value it most pay for it. Did 19 the committee at all bring that up? 20 MR. GRUENSPECHT: No, they -- I 21 mean, it's a lot of -- you know, obviously 22 you raise issues. You know, the economics of BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 371 1 information is a very tricky thing because, I 2 mean, if -- you know, this is always the 3 issue with secondary dissemination of 4 consultant reports, like if Reuters values it 5 the most one nanosecond after Reuters gets it 6 it's value is zero then. It has to do with 7 the issue of intellectual property and the 8 economics of information. And I think there 9 are a lot of interesting questions, but the 10 thought has been nobody -- OMB has sometimes 11 raised that issue, but the committee, I don't 12 think, raised that issue. I guess they kind 13 of took the approach more of the Boskin 14 Commission or something. There was a 15 commission on statistics a while back, I 16 guess in the mid-1990s, suggesting that 17 federal investment in statistics was -- I 18 guess the other -- I don't know. I mean, the 19 answer is no, I think, they did not raise it 20 about selling it to the highest bidder. It's 21 worth a lot, though. 22 MR. RUTHERFORD: Well, I have a BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 372 1 question about the -- I think that the idea 2 of trying to develop better sort of analytic 3 passing a match, this survey work that's done 4 is really a good idea. But I can see three 5 activities that are really useful and 6 worthwhile activities. I basically agree 7 with Denny's assessment, that you're 8 collecting data to make markets work better. 9 You're attempting to provide information 10 about forecasts that statistical -- what can 11 be statistically based forecasts. And you're 12 also trying to produce structural tools for 13 assessing policy questions, which is quite 14 different from statistical forecasts. 15 MR. GRUENSPECHT: Right. 16 MR. RUTHERFORD: And that's one of 17 the challenges I think here is distinguishing 18 one from the other. We have statistical 19 bands. You know, we have these -- we have 20 degrees of uncertainty about forecasts, but 21 then structural analysis for policy is really 22 a matter of -- it's a necessary condition for BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 373 1 a policy argument is that you can produce the 2 model that produces that result and then go 3 for it from there. 4 But just in general about how to 5 return to the days of Hogan the sort of 6 notion of a mission that we're going 7 to -- sort of the Department of Energy's 8 going to help assess the things that matter 9 to the country. I think it's really a 10 good -- that'd be a really good sort of 11 direction to try to move and I think you have 12 resources to do that. 13 The only sort of modest suggestion 14 I might have on that direction is I've 15 been -- Alex Miros (?) turned me on to this 16 book from the 1960s called The Mythical 17 Man-Month. It's about the development of the 18 IBM 360. And it makes this -- it's one of 19 these things that makes these cogent 20 observations, like if a software project is 21 behind, adding more persons to the project 22 will probably just make it go on longer, BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 374 1 right? So I think you have to be a little 2 bit careful about just hiring more people. 3 And there's other ways that you can use 4 resources that might improve the profile and 5 indeed attract -- you know, there's a high 6 degree of unevenness of productivity across 7 people. You want to try to create an 8 environment to attract people. 9 But I think, like, having a seminar 10 series or having regular colloquia focused on 11 issues that are of interest, both to 12 academics and to policymakers in Washington, 13 to provide a forum for interaction between 14 the different groups like that might be a 15 real productive way to use some of the 16 resources. 17 MR. GRUENSPECHT: Right. 18 MR. RUTHERFORD: Because you want 19 to try to create an environment where young 20 people coming out with Ph.D.s say I want to 21 go there because I know it'll be in the 22 limelight. I want to go work on the issues BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 375 1 that matter. 2 I mean, clearly, I don't think that 3 you'd have to work very hard to make them 4 interesting questions. I mean, you're 5 surrounded by them. But the main thing is 6 just to increase your profile. And rather 7 than try to sell yourself better to the job 8 markets, it's better just to have them coming 9 knocking on your door than trying to 10 advertise. 11 MR. BURTON: The FCC does that with 12 telecom issues on a very regular basis and 13 they are highly regarded. 14 MR. RUTHERFORD: Just my modest 15 opinion. 16 MR. GRUENSPECHT: No, it's a good 17 idea. I mean, we have been talking in a 18 preliminary way. You know, on NEMS, we have 19 an annual NEMS -- you know, a day-long 20 conference set up so that people could really 21 fly in and fly out the same day. But CERA, 22 which is a large energy consultancy, has BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 376 1 something called CERA Week each year in 2 Houston. And it's quite a big gathering, if 3 you will, of people. 4 And we have thought -- not 5 necessarily want to do EIA Week, but the 6 thought is some of the data programs and some 7 of the issues and analyses that aren't about 8 NEMS and stuff, the thought is we really 9 could possibly have a two- or three-day sort 10 of EIA Week that went beyond NEMS and looked 11 at a lot of these other types of issues and 12 had different groups looking at different 13 things. 14 MR. RUTHERFORD: I mean, the real 15 challenge I see for EIA is that if you're 16 building a policy relevance or a structural 17 model to make a point, it's always better to 18 work backwards. John Whalley taught me that 19 you first draw the figures you want to have 20 your model produce, then write the tables and 21 put the numbers in the tables. When I first 22 learned this from John, I was appalled. But BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 377 1 I've learned afterwards that this is really a 2 good way to work. 3 You begin back from the story you 4 want to tell, write down the numbers you 5 think are going to come out, and you build 6 the model to see if you can produce those 7 numbers. And then you learn something 8 because you learn your ex ante assessment was 9 different from what actually comes from the 10 numbers. And then -- 11 MR. GRUENSPECHT: Right, but then 12 the question is if you're a consultant, then 13 the answer is that you have to change 14 whatever you came up with to meet the 15 numbers -- 16 MR. RUTHERFORD: Well, then the 17 game is -- 18 MR. GRUENSPECHT: You know, in our 19 case I'd like to think that we don't have any 20 particular place we want to come out because 21 we're not for or against. 22 MR. RUTHERFORD: Yes, that's the BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 378 1 point I'm making. 2 MR. GRUENSPECHT: Right. 3 MR. RUTHERFORD: It's just that 4 you're in a much tougher environment because 5 you don't have an argument you're producing. 6 You're trying to produce tools that will be 7 used generally. 8 MR. GRUENSPECHT: Right. 9 MR. RUTHERFORD: But I think you 10 have to be thinking ahead to what the 11 questions are going to be asked. 12 MR. GRUENSPECHT: That's right, 13 exactly. And that's why in this NEMS context 14 we're going out to -- you know, we went to 15 the -- as Susan described earlier, we went to 16 the program officers because they're one of 17 the groups that ask questions. And we went 18 to the -- I think we -- Guy has sent letters 19 to? 20 MR. CARUSO: The associations. 21 MR. GRUENSPECHT: A lot of 22 associations. We're going out to the labs. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 379 1 I've sent letters out to the labs just in 2 this last week. So -- but, again, I don't 3 think we're going to come up -- I remember 4 the Roald Dahl story, I don't know, "The 5 Rocking-Horse Winner," I think. No, not that 6 one. "The Great Automatic Grammatizator." 7 This, like, machine that could write any 8 story and put all the authors out of 9 business. 10 I don't think we're 11 not -- definitely not going to develop a tool 12 that'll answer all the questions, so some of 13 it will fall off. You can't just concatenate 14 what everyone's -- but I agree that we 15 have -- where we choose to design the tool 16 it's certainly relevant what that space of 17 what people want to look at looks like, 18 although I don't think we're obligated to 19 cover the whole space because I don't think 20 we can. 21 MR. EDMONDS: Yes, I just want to 22 follow up on what Tom was just saying. I BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 380 1 think -- he was talking about the EIA as an 2 exciting place to be because it really does 3 get those really interesting questions and it 4 gets them in a way that there's a direct line 5 to Congress. I mean, there's a special 6 relationship with Congress that EIA has that 7 the rest of the Department of Energy does 8 not. 9 And in a sense of making that a 10 more visible and more present sense, one 11 thing you could do is not just -- if you do 12 EIA Week, don't just do EIA. Actually create 13 a convening power that EIA would have. And 14 so bring in everybody else who is important 15 or has important things to say. 16 And that, in a sense, actually 17 gives you leverage and sort of a greater 18 spotlight because then EIA is the convener 19 and everyone does come to EIA. 20 I guess the other thing that Tom 21 said, which I also think is useful, is that 22 really the models are our story. They're our BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 381 1 story of how things fit together. They're a 2 story of relations and you drop -- I mean the 3 EIA -- even the NEMS model doesn't have 4 everything in the whole world in it. Close, 5 but you don't have it all. And you've 6 dropped things off and you actually do have a 7 set of relations. 8 So I think that's, in a sense, 9 the -- you know, that may be one way of 10 telling the story, that is the graphics. 11 But, in a sense, you really have this story. 12 And thinking through the story of the 13 relationships I think is a very useful thing 14 to do. 15 And starting from this logic, and 16 then actually as you build the model and 17 introduce the data, you learn that that logic 18 was flawed because there are 19 interrelationships that have very powerful 20 influences that you didn't realize when you 21 started off and you change the story. And 22 that's I think your true -- you know, sort of BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 382 1 your unbiased -- you don't change the model 2 to get the story you started off with, you 3 changed the story because the model told you 4 things you didn't know when you started. 5 MR. GRUENSPECHT: Unfortunately, if 6 only unbiased would equal correct, but those 7 two things are very different from each 8 other, as we often find out. 9 SPEAKER: Let's not get too 10 esoteric here. 11 MR. GRUENSPECHT: I don't know. So 12 I'll tell you what, we thought it was 13 really -- we thought we got a lot more than 14 our money's worth out of these folks. It was 15 a very -- I think it was a very positive 16 overall experience. 17 MR. CARUSO: And that's the only 18 thing I would add is that it's not so much 19 what we got out of them, it's 20 having -- forcing us to go through this 21 process of -- especially Nancy, who headed up 22 the support team, had to do a lot of work to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 383 1 pull things together. Because they 2 asked -- you can imagine just looking at that 3 list that they asked tough questions. And we 4 find that the process was very useful. 5 MR. GRUENSPECHT: I mean, these 6 kind of tables that they asked about, like 7 percent for this and that typically in our 8 budget process, frankly, we don't do that, 9 you know. Maybe we should. I might -- they 10 didn't criticize us for not doing it, but 11 implicitly, when they asked for this, we had 12 to try to put it together the way that I 13 think it makes -- you learn a lot from 14 looking at this. We learn a lot from looking 15 at it. They needed to look at it they felt 16 to do their work, so we had to do it. 17 MR. NEERCHAL: Did you keep track 18 of this before or because of the committee 19 you started doing this, this news media, 20 Appendix B? 21 MR. GRUENSPECHT: No. 22 MS. KIRKENDALL: NAIC keeps tabs on BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 384 1 that. 2 MR. GRUENSPECHT: We do keep tabs. 3 MS. KIRKENDALL: They keep tabs on 4 the news media who call in. 5 MR. CLEVELAND: It's really 6 shocking how little is done on analysis in 7 total budget. I'm looking at page 17. 8 MS. KIRKENDALL: Well, that was cut 9 considerably. You know, we used to have an 10 analysis review board and we had planned 11 analysis and collaborated on it, then that 12 went away. 13 MR. CLEVELAND: How do you get -- 14 MS. KIRKENDALL: (inaudible). 15 MR. GRUENSPECHT: I don't want to 16 speak for the other people who were here 17 then. Nancy can, but I guess my sense is 18 there's always a concern with data that once 19 you don't collect it you could not go back to 20 people and say, well, what was your 21 generation of coal-fired electricity three 22 years ago or something. So there's been this BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 385 1 reluctance to -- that somehow data is like 2 the vital organs and that analysis is like 3 the appendix. 4 You know, if something has to go 5 cut out the appendix, don't cut out your 6 liver or something. But in some ways I think 7 the committee said maybe this analysis is 8 more like your liver than like your appendix. 9 And maybe you won't drop dead instantly, but, 10 over time, if you don't do more analysis, 11 you're going to -- your data will suffer, you 12 know. 13 So I guess what they would be 14 arguing is if -- even if the budget were 15 constrained at some fixed level I guess 16 implicitly they're saying they would do less. 17 They'd probably shrink the program a little 18 bit and do more analysis in the part of that 19 was retained. But the philosophy, I think, 20 before Guy's arrival was more that the last 21 thing you give up is a data series because 22 you can never recover that. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 386 1 MR. CLEVELAND: Well, that's also 2 part of the culture, at least in the past, 3 part of the culture of EIA, too, which is 4 this is my report and my data and so 5 you -- internally I think you'd fight like 6 hell, regardless of whether it was useful to 7 have or not, to fight to retain to that. 8 MR. GRUENSPECHT: Well, as you 9 know, we've dropped a couple of things in the 10 recent years and we tried to drop a couple of 11 things. You know, we have a letter 12 from -- interestingly, this fall, I mean, 13 this is not on the agenda, but we had said we 14 would drop a couple of petroleum things. And 15 we got a letter signed by, like, I don't 16 know, wow, like 22 senators or something, you 17 know. It was horrible, you know. I mean, it 18 was good but it was -- but, I mean, one of 19 the problems is, you know -- it's horrible 20 and it was good because it kind of felt like 21 we couldn't drop it. 22 And then we said, well, if they're BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 387 1 really going to give us this money in 2 '07 -- but then, of course, the '07 thing is 3 completely hung up. So the question is what 4 do you do? Because, literally, if you stop 5 it and then try to start it again, that has 6 huge -- you know, you get people who are 7 like -- kind of like galley slaves, you know. 8 They're used to the beating and pulling the 9 oars. And then you just don't hit the drum 10 for a couple of months, and then you say, oh, 11 we want to start up again that's really hard, 12 you know. They're not going to get stuff in 13 on time. 14 MR. CLEVELAND: You sound like a 15 great boss. 16 MR. GRUENSPECHT: No, I meant the 17 respondents. I meant the respondents. I'm 18 not worried about the staff. The staff is 19 very dedicated. But you're dealing with the 20 respondents. You tell them, hey, I don't 21 have to -- you know, if they told you you 22 didn't have to file some form you had to file BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 388 1 every month, and they said, oh, you don't 2 have to file it anymore, and then you try a 3 couple months later, oh, we were wrong, you 4 have to file it again, it isn't like people 5 are going to start filing it at the same rate 6 that they were filing it before you let them 7 off the hook. So if we're going to continue, 8 we definitely don't want to have a break. 9 I mean, again -- and I -- that's 10 part of the logic that we're talking about 11 that we -- sort of I guess you were 12 suggesting was wrong. But in this case if 13 you're going to -- you know, maybe you should 14 stop things if you can't -- but in this case 15 if you're going to start it again, then I 16 think there really is a reason not to stop 17 it. 18 MR. CLEVELAND: Well, sure, if you 19 know you want to do that. The point is there 20 are probably some things that should be 21 dumped. 22 MR. GRUENSPECHT: No, and that's BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 389 1 really what this committee said. I mean, 2 they said three things should be dumped. 3 MR. CLEVELAND: Right. 4 MR. GRUENSPECHT: And I think there 5 are -- you know, try to deal with this at 6 a -- although I guess technically anyone 7 could hear these comments. 8 I mean, a good example of something 9 they said should be dumped, I can't disagree. 10 I picked the voluntary greenhouse one of all 11 things, but the reality of it is it's -- it 12 isn't like you're going to dump it and then 13 use the money for something else, you know. 14 MR. CLEVELAND: Right. 15 MR. GRUENSPECHT: That sure ain't 16 going to happen. And so, again, it's -- I 17 think we're taking this all very seriously 18 and I think you'll see changes in the program 19 that reflect a lot of what's in here. Is 20 everything in there going to happen? I'm not 21 sure. I mean, people are aware of this. 22 This is not, you know -- I mean, we're not BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 390 1 hiding this. We're not -- 2 MR. FEDER: Maybe you should repeat 3 it every few years. 4 MR. GRUENSPECHT: Well, that's a 5 good -- 6 MR. CLEVELAND: Is it up on your 7 home page? 8 MR. GRUENSPECHT: It's on 9 our -- no, but if you ask for it, we'll send 10 it to you. 11 MS. KIRKENDALL: It's on the ASA 12 web site, on our home page. You can get to 13 it that way. 14 MR. CLEVELAND: So it's not on your 15 web site. 16 MS. KIRKENDALL: Well, it is on our 17 web site if you happen to know where it is. 18 MR. CLEVELAND: One of those 19 products. 20 MR. GRUENSPECHT: Look, I mean, I 21 don't know what the answer is, you know. 22 Now, if you want -- if we had produced a BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 391 1 sycophantic report then you'd go out -- I 2 mean, I don't know what you're really going 3 to use that's not going to say -- 4 MR. CLEVELAND: I'm just giving you 5 a hard time. 6 MR. GRUENSPECHT: No, but it's -- 7 MS. KIRKENDALL: In the budget. 8 SPEAKER: Program review is very 9 important. In an academic department those 10 things are really crucial. 11 MR. GRUENSPECHT: I mean, the real 12 question is -- I mean, another example, I 13 mean, (inaudible) this too long, but another 14 example is, look, we know there are lots of 15 flaws in our program, right? You know, so 16 we're sailing the ship, sailing the ship S.S. 17 EIA, and we're going through the oceans. And 18 it's a rough and stormy ocean and we know our 19 ship is not perfect. 20 MS. KIRKENDALL: Right. 21 MR. GRUENSPECHT: And there are 22 other people who, for some good reasons, but BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 392 1 some venal reasons, what to take shots at us, 2 you know. Now, frankly, I could probably 3 take shots at us better than they could but 4 the question is should I help them out by, 5 like, pointing my cannon down at the 6 waterline and blowing it? You know, I don't 7 know what the answer is. 8 MR. FEDER: What kind of energy is 9 the S.S. EIA using? 10 MR. GRUENSPECHT: What? 11 MR. FEDER: What kind of energy is 12 the S.S. EIA using? 13 MR. GRUENSPECHT: A sailing ship, I 14 envision this. I think of this as Russell 15 Crowe -- I think of Guy as Russell Crowe in 16 "Master and Commander." 17 SPEAKER: He's a little young, 18 though. 19 MR. HENGARTNER: But, Howard, this 20 report is very useful to the five-year 21 planning you're going to be doing. 22 MR. GRUENSPECHT: Yes, it is and BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 393 1 that's what feeds into it. 2 MR. HENGARTNER: I mean, I see so 3 much value to actually go through this. And 4 maybe every five years is much. I mean, 5 academic departments every seven years they 6 make a review. 7 MR. GRUENSPECHT: Right. 8 MR. HENGARTNER: But having first 9 some outside input before you're doing your 10 five-year planning is probably beneficial. 11 And I don't know what the outcome of that 12 planning is going to be. You're probably 13 going to tell me next year, you know what, 14 that was a good thing or it didn't work at 15 all. 16 MR. GRUENSPECHT: Oh, I think 17 (inaudible). But, I mean, the real issue is 18 I think do you want to go to people who you 19 know or not, you know? Denny Ellerman, Phil 20 Sharp, Kathleen Cooper they don't dance on 21 Guy's string or my string, you know. They're 22 going to say what they want to say and that's BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 394 1 what -- if you want other people, you can get 2 people who will say tell me what you want to 3 hear and I'll tell you, you know. And in 4 this case I felt, you know -- Guy felt that 5 it was worth, you know -- 6 MR. EDMONDS: Have you discussed 7 this with the Hill? 8 MR. CARUSO: We've certainly given 9 it to them. 10 MR. GRUENSPECHT: I'm sure it'll 11 come up. 12 MR. EDMONDS: I mean, this is 13 obviously something now that you've created 14 it, that needs to be discussed with the 15 relevant committees. 16 MR. RUTHERFORD: You could maybe 17 arrange a swap. You could swap them the 18 voluntary greenhouse abatement for the 19 biofuels, right? 20 MR. GRUENSPECHT: Oh, they want the 21 biofuels. They don't want the biofuels. 22 MR. RUTHERFORD: They don't want to BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 395 1 give it back. 2 MR. GRUENSPECHT: No, the Hill 3 wants the -- the Hill wants us to do 4 biofuels, but not -- 5 MR. RUTHERFORD: No, but you said 6 the EPA has this -- 7 MR. GRUENSPECHT: Right, right. I 8 mean, in terms of the government? Yes, that 9 would be interesting. We're going to do the 10 biofuels. 11 Anyway, no, I really appreciate 12 your -- I think a lot of the input was very 13 thoughtful. 14 MR. HENGARTNER: Okay, thank you 15 very much. Any more questions? We're 16 running 15 minutes over, but I think it was 17 well worth it. Thank you very much, Howard. 18 MR. GRUENSPECHT: Thank you. 19 You're the ones who deserve the thanks. 20 MR. HENGARTNER: At this time, we 21 want to invite any member of the public to 22 make general comments. BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382 396 1 Okay, then let us adjourn this 2 meeting until tomorrow morning at 8:30. 3 (Whereupon, at approximately 4:49 4 p.m., the MEETING was were 5 continued.) 6 * * * * * 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 BETA COURT REPORTING www.betareporting.com 202-464-2400 800-522-2382