10025 1 2 3 4 5 6 7 8 9 NATIONAL FEDERAL MILK MARKETING ORDER 10 PRICING FORMULA HEARING 11 12 DOCKET NO.: 23-J-0067; AMS-DA-23-0031 13 14 Before the Honorable Jill Clifton, Judge 15 16 ---o0o--- 17 18 Zionsville, Indiana 19 December 7, 2023 20 21 ---o0o--- 22 23 24 25 26 Reported by: 27 MYRA A. PISH, RPR, C.S.R. Certificate No. 11613 28 10026 1 A P P E A R A N C E S: 2 FOR THE USDA ORDER FORMULATION AND ENFORCEMENT DIVISION, USDA-AMS DAIRY PROGRAM: 3 Erin Taylor 4 Todd Wilson Brian Hill 5 Michelle McMurtray 6 FOR THE MILK INNOVATION GROUP: 7 Charles "Chip" English Ashley Vulin 8 FOR THE NATIONAL MILK PRODUCERS FEDERATION: 9 Nicole Hancock 10 Brad Prowant Jim Sleper 11 Jeffrey Sims 12 FOR SELECT MILK PRODUCERS, INC.: 13 Ryan Miltner 14 FOR INTERNATIONAL DAIRY FOODS ASSOCIATION: 15 Steve Rosenbaum 16 17 ---o0o--- 18 19 (Please note: Appearances for all parties are subject to 20 change daily, and may not be reported or listed on 21 subsequent days' transcripts.) 22 23 ---o0o--- 24 25 26 27 28 10027 1 M A S T E R I N D E X 2 SESSIONS 3 THURSDAY, DECEMBER 7, 2023 PAGE 4 MORNING SESSION 10,030 AFTERNOON SESSION 10,139 5 ---o0o--- 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 10028 1 M A S T E R I N D E X 2 WITNESSES IN CHRONOLOGICAL ORDER 3 WITNESSES: PAGE 4 Jim Hau: 5 Direct Examination by Mr. English 10,031 Cross-Examination by Mr. Miltner 10,046 6 Cross-Examination by Ms. Hancock 10,057 Redirect Examination by Mr. English 10,058 7 Cross-Examination by Ms. Taylor 10,059 8 Dr. Joseph Balagtas: 9 Direct Examination by Mr. Rosenbaum 10,075 Cross-Examination by Ms. Hancock 10,100 10 Cross-Examination by Mr. Sims 10,131 Cross-Examination by Mr. Miltner 10,140 11 Cross-Examination by Mr. English 10,156 Cross-Examination by Mr. Sleper 10,160 12 Cross-Examination by Ms. Taylor 10,168 13 Dr. Mark Stephenson: 14 Direct Examination by Mr. English 10,189 Cross-Examination by Mr. Rosenbaum 10,204 15 Cross-Examination by Mr. Miltner 10,206 Cross-Examination by Ms. Hancock 10,213 16 Redirect Examination by Mr. English 10,221 ReCross-Examination by Ms. Taylor 10,230 17 18 ---o0o--- 19 20 21 22 23 24 25 26 27 28 10029 1 M A S T E R I N D E X 2 INDEX OF EXHIBITS 3 IN CHRONOLOGICAL ORDER: 4 NO. DESCRIPTION I.D. EVD. 5 432 HAU-001 10,073 6 435 IDFA-61 10,074 10,186 7 436 IDFA-62 10,074 10,186 8 437 NMPF-110 10,103 9 291 Document 10,252 10 438 Stephenson-002 10,188 10,252 11 12 ---o0o--- 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 10030 1 THURSDAY, DECEMBER 7, 2023 -- MORNING SESSION 2 THE COURT: Let's go back on record. 3 We're back on record it is 2023, December 7th. 4 It's a Thursday. It is day 42 of this hearing. And I'd 5 like to ask if there are some preliminary matters before 6 the next witness testifies. 7 I see none. 8 The next witness may come to the stand. I believe 9 that's Mr. Hau, H-A-U. 10 MR. ENGLISH: Good morning, Your Honor. Chip 11 English for the Milk Innovation Group. We had a fair bit 12 of discussion a little after 5 o'clock yesterday, so just 13 to clarify, we -- we handed out Mr. Hau's testimony, and 14 it was marked, I believe, as Exhibit 432. 15 THE COURT: That's correct. It is Exhibit 432, 16 and it's also marked as Hau, H-A-U, 001. 17 MR. ENGLISH: Thank you very much, Your Honor. 18 And as with some prior witnesses, I wish to 19 clarify -- I'll actually do so with Mr. Hau after you 20 swear him in. 21 THE COURT: All right. Would you please state and 22 spell your name? And this will also be a test of the 23 microphone location compared to where you are. 24 THE WITNESS: Okay. My name is Jim Hau, J-I-M, 25 H-A-U. 26 THE COURT: Good. You are loud and clear. That's 27 great. 28 Have you previously testified in this proceeding? 10031 1 THE WITNESS: I have not. 2 THE COURT: I'll swear you in. 3 JIM HAU, 4 Being first duly sworn, was examined and 5 testified as follows: 6 THE COURT: Thank you. 7 MR. ENGLISH: So, again, Your Honor, Chip English 8 for the Milk Innovation Group. 9 DIRECT EXAMINATION 10 BY MR. ENGLISH: 11 Q. So, Mr. Hau, you're here on behalf of Maple Hill 12 Creamery; is that correct? 13 A. That is correct. 14 Q. And Maple Hill Creamery is not a member of the 15 Milk Innovation Group, correct? 16 A. No. 17 Q. And it is not actually presently a member of the 18 International Dairy Foods Association, correct? 19 A. It is not. 20 Q. But you are appearing today, and you asked Davis 21 Wright Tremaine to assist you partially because we have 22 an -- an existing professional relationship, correct? 23 A. That is correct. 24 MR. ENGLISH: Your Honor, Mr. Hau will read his 25 statement, and I believe based upon things he heard 26 yesterday, he may have a few supplemental remarks. 27 BY MR. ENGLISH: 28 Q. So, Mr. Hau, please provide your statement, which 10032 1 is Exhibit 432, also Hau-001. 2 A. Good morning. My name is Jim Hau. I am both the 3 president and the CFO for Maple Hill Creamery and oversee 4 all operations of our small company. And I can confirm 5 Maple Hill Creamery is a small business per the SBA 6 definition based on our number of employees. 7 I have a bachelor's of science in business 8 administration from Marquette University and an MBA from 9 Northwestern University. My career includes over 35 years 10 in the consumer products business working with both large 11 public international corporations and small private 12 startups. My experience with the dairy industry spans 13 just over 20 years and includes Unilever's ice cream 14 business, White Wave Foods' Horizon business, Pantheryx's 15 colostrum -- 16 THE COURT: Now, these words are important to 17 capture in the transcript, so read them slowly, and it 18 would be wise to spell these businesses. 19 BY MR. ENGLISH: 20 Q. And just to be clear, Mr. Hau, we all tend to read 21 faster than we think we are, and so our wonderful court 22 reporter really needs you to read much more slowly. 23 A. Okay. 24 Q. I think the statement people have said is, perhaps 25 like we're reading to third grade school children, some of 26 whom may be in the room. 27 A. So -- so let me back up on the experience. 28 So Unilever's ice cream business, that's 10033 1 U-N-I-L-E-V-E-R; White Wave Foods, W-H-I-T-E, W-A-V-E, 2 Horizon's business; Pantheryx, that one's 3 P-A-N-T-H-E-R-Y-X, colostrum supplement business; and now 4 Maple Hill. 5 Maple Hill is headquartered in Kinderhook, New 6 York, where we have operations spread throughout the 7 country, and I'm based just outside of Boulder, Colorado. 8 Maple Hill is a young company started in 2009, and 9 a pioneer in dairy's fastest growing segment, grass-fed 10 organic milk. While small, with just 18 employees, we 11 have quickly grown to national distribution in over 8,000 12 retail locations, including the largest retailers in the 13 country like Kroger, Safeway/Albertsons, Walmart, Whole 14 Foods, Publix, and Sprouts. Milk from Maple Hill Creamery 15 can be found in every region of the continental United 16 States and Hawaii. 17 Our products include fluid milk, yogurt, kefir, 18 and butter, and are produced in four plants largely in the 19 Northeast. The plants are third party co-packers as we do 20 not have the financial resources to build our own 21 factories. 22 Fluid milk is the vast majority of our revenue and 23 is currently produced in one plant that produces Class I 24 products. Maple Hill Creamery products have retail sales 25 over $85 million across retailers nationwide, and we hold 26 the number two share in our fluid -- in our key fluid milk 27 segment. 28 Today I'm hoping to bring a few key points to the 10034 1 attention of the USDA, of this committee. 2 Key point number one: Our business is critical to 3 the success of our 120 independent small family farms. 4 Importantly, while we have just 18 employees, we 5 represent over 120 small family farms who participate in 6 one of four co-ops but ship directly to Maple Hill. In 7 our business, "small family farm" is not a cliche as our 8 average farm milks just over 50 cows. Grass-fed organic 9 dairy represents not just the most innovative and fastest 10 growing segment of the dairy milk shelf, but it is also 11 the most effective way for the small family farms to be 12 able to compete and co-exist in an increasingly large 13 industrial farm environment. 14 In most cases, Maple Hill is the only dairy 15 customer for the these family farms. The vast majority of 16 our farms are in Upstate New York with a few in 17 Pennsylvania as well, and participate in the Federal 18 Order 1 area. The vast majority of the milk they produce 19 for us goes into Class I fluid milk. 20 Key point number two: The grass-fed organic 21 business represents an important source of innovation to 22 the dairy industry. 23 The grass-fed organic dairy business accounts for 24 just over 1% of the total dairy retail milk sheds. While 25 on a volume basis, the dairy milk business is down 2.5% 26 this year, grass-fed organic has grown 7%. 27 In recent years, plant-based beverages have been 28 able to erode dairy share of the beverage market by making 10035 1 claims around healthier alternatives that are better for 2 the environment. The grass-fed organic dairy segment is 3 the dairy industry's most powerful tool to push back on 4 this erosion. Grass-fed organic dairy is both healthier 5 for the consumer, has the most naturally nutrient-dense 6 dairy product on the market, and healthier for the 7 environment due its most regenerative agricultural nature. 8 This latter point has been recognized by the USDA 9 in its recent award of a Partnership for Climate Smart 10 Commodities grant to our company and its farmers. 11 Key point number three: The USDA has wisely 12 invested in this regenerative space in support of growing 13 the grass-fed organic segment, which enhances the overall 14 dairy industry. 15 The USDA dairy grant -- the USDA grant recognizes 16 the importance of supporting the farmers in this growing 17 category and the importance of helping to market the 18 product given its scarce resources due to higher costs and 19 thin margins. Maple Hill's success to date has come from 20 educated consumers who have done their research and 21 recognize our product's benefits. Our future success is 22 dependent on our ability to market a high quality product 23 at a reasonable price to a broader consumer base. 24 The higher cost of grass-fed farming including the 25 price we pay farmers, which is well above FMMO rates, 26 makes marketing funds scarce in an already thin margin 27 dairy business. It also forces us to price our product at 28 the most premium level in the dairy case, which precludes 10036 1 what would otherwise be higher growth. Those high costs 2 and thin margins represent, I believe, the most 3 significant impediment to our ability to grow this 4 innovative dairy segment. The USDA grant will help with 5 that effort. 6 Key point number four: Pooling is a tax on our 7 business which receives no benefit from the FMMO process. 8 Because we pay our farmers well above the FMMO 9 rates, they do not get any benefit from pooling funds. 10 Maple Hill, while paying into the pool, also does not 11 derive any benefit from pooling funds. We balance our own 12 milk. Our customers and consumers do not benefit from 13 pooling, as our product is already available in every 14 region of the country. 15 From a pricing standpoint, pooling actually 16 increases consumer prices rather than offering a relief. 17 For these reasons, pooling is simply a tax on grass-fed 18 organic and all organic dairy businesses that deprives our 19 companies of much needed funding so the dollars can be 20 distributed to the conventional segment. This adds to the 21 already high cost of grass-fed dairy development and, as 22 previously mentioned, weakens growth. 23 I don't believe that in all the hours of testimony 24 heard to date anyone has disputed these points, likely 25 because we are all in agreement that there is no rationale 26 to dispute them. However, rather than addressing the 27 problem, the issue's been ignored and now threatens to be 28 exacerbated. 10037 1 Key point number five: Pooling is a material hit 2 to profitability of small growing companies and, 3 therefore, impedes innovation even before the proposed 4 increases, which will materially worsen the situation. 5 While it would be appropriate for me to share -- 6 while it would be inappropriate for me to share 7 confidential Maple Hill financial information, I would 8 like to share a theoretical example based on practical 9 numbers of the impact of the pooling tax. 10 A small, early-stage, organic milk company, with 11 revenue of about $25 million, is likely to incur a pooling 12 tax of approximately $600,000. Likely, that startup is 13 losing money and already fighting for investment dollars. 14 However, assuming they are lucky and can garner what would 15 be considered a healthy dairy margin of say 5% EBITDA -- 16 that's E-B-I-T-D-A -- they are getting $1.2 million that 17 is available for taxes, interest, and reinvestment to grow 18 their business. Giving $600,000 of that to conventional 19 farms through pooling has a very material impact on the 20 ability -- on their ability to reinvest and grow. The 21 proposals being considered here today, that could take 22 that up another 80%, would eat up another $480,000. 23 The impact this could have on small businesses 24 entering the space with innovative products cannot be 25 overstated. In many cases, maybe most cases, they simply 26 will not be able to get off the ground. If Maple Hill had 27 incurred this higher pooling tax in the early days, it is 28 likely we would not exist today. Even though we have 10038 1 grown to a stronger and more established company, if the 2 considered increases were to go through, we will likely 3 need to reconsider material aspects of our growth 4 strategy. 5 Let me now explain our opposition to Proposal 19. 6 NMPF's Proposal 19 seeks to increase the Class I price 7 differential from its current range of $1.60 to $6 to a 8 proposed range of 2.20 to 7.90. 9 THE COURT: So that's $2.20. 10 THE WITNESS: To $7.90. 11 THE COURT: Thank you. 12 THE WITNESS: This will compound the problem for 13 organic dairy farms. For this reason, we oppose NMPF's 14 Proposal 19. 15 USDA must reject the proposal for multiple 16 reasons: 17 1. The increase would represent a greater 18 taxation and further slow progress of this important and 19 growing segment of the dairy industry. 20 2. We do not see logic and justification for the 21 current Class I price differential, even remaining at 22 $1.60, given current supplies of fluid milk. 23 3. Based on my experience in the market, there is 24 more than sufficient supply of fluid milk. Raising the 25 Class I price differential will only serve to increase 26 unneeded conventional milk supply, which is the slowest 27 growing segment, dairy segment, and harm organic 28 suppliers, including grass-fed, which is the fastest 10039 1 growing dairy segment. 2 4. The differential increase will have the impact 3 of increasing supply at the same time it decreases demand, 4 because it robs valuable investment in market growth and 5 innovation, which drives demand only to redistribute it to 6 supply. Such an imbalance is not healthy impact on the 7 dairy industry or its farmers, the very people meant to be 8 helped by the increase. 9 We cannot pass on -- 10 Number 5. We cannot pass on a commensurate cost 11 increase as proposed here without risking the loss of 12 customers and/or beverage market share. This result will 13 harm both my company and my farmers, and could mean even 14 fewer Class I dollars in the pool. Moreover, it will 15 impede our ability to innovate and market an important 16 segment of the dairy industry as it works to reverse the 17 losses of beverage market share. 18 Number 6. The NPF [sic] proposals do not 19 represent the best interests of our dairy farmers. 20 THE COURT: Could you start that sentence again, 21 please, number 6. 22 THE WITNESS: Sorry. 23 The NMPF proposals do not represent the best 24 interests of our dairy farmers, and in fact, are openly 25 willing to sacrifice the best interests of our small 26 family farms in favor of other interests they represent. 27 7. The differential change is not necessary to 28 ensure distribution of milk to alleged deficit regions of 10040 1 the county, as Maple Hill distributes milk from the 2 Northeast to all parts of the country without any support 3 of FMMO pricing. 4 8. The only impact these proposals will have on 5 our dairy farms is to reduce investment in our business, 6 thus inhibiting their ability to grow, other small family 7 farms' ability to participate in this growing segment, and 8 hamper the dairy industry's ability to use innovation to 9 expand market share and compete against others in the 10 beverage industry, including the plant-based segment that 11 continues to erode dairy consumption in the U.S. 12 9. In fact, none of the rationale put forth in 13 the NMPF proposal is true for Maple Hill or the 120 dairy 14 farms with which we partner. The increase in Class I 15 differential represents a potential 80% increase in the 16 pooling tax on our business, with zero benefits to our 17 stakeholders, and will do nothing to increase pay prices 18 to our farmers (in fact, it is more likely to have the 19 opposite effect), and it will do nothing to increase the 20 already national availability of milk from those farms (in 21 fact, it is more likely to have the opposite effect). 22 We pay grass-fed organic farms a premium because 23 this type of dairy farming is more difficult. Without 24 that premium, there is little financial incentive for 25 farmers to convert to grass-fed organic. Thus, increasing 26 the pooling tax and inhibiting the ability to pay farmers 27 more undermines the ability to expand grass-fed organic 28 dairy farming. 10041 1 THE COURT: You read it perfectly, but I would 2 like you to read that sentence again. 3 THE WITNESS: Thus, increasing the pooling tax and 4 inhibiting the ability to pay dairy farmers more 5 undermines the ability to expand grass-fed organic dairy 6 farming. 7 10. This increased tax burden will increase 8 consumer prices for consumers who can pay it and reduce 9 the number of dairy consumers where affordability has 10 teetered over the edge of manageability. 11 11. Finally, an increased pooling tax burden 12 supported by the USDA, which is punitive to the 13 regenerative grass-fed organic segment, will 14 philosophically and financially conflict with the USDA's 15 desire to support regenerative agriculture as demonstrated 16 by the recent PCSC grant award. The USDA is in the 17 process of investing $20 million with Maple Hill to expand 18 friendly, organic, grass-fed farming. Now we are here, 19 today, considering a damaging tax on the organic business 20 which will undermine the very rationale for the investment 21 the USDA just announced. 22 In sum, we foresee reduced investment in dairy 23 innovation leading to shrinking dairy share of the 24 beverage market, higher prices for consumers, and 25 potentially lower pay prices for farmers. 26 I travelled a considerable way today to make the 27 Department aware of these issues because it will 28 materially affect the way we move forward and impact all 10042 1 aspects of our business and in a negative way. I'm 2 grateful for the opportunity to make the trip and speak to 3 you today, because this represents critical issues for our 4 ability to compete and move forward as a business. 5 For the sake of all of our farmers, consumers, and 6 stakeholders, I hope I have been clear on the impacts and 7 properly represented the problem this proposal represents 8 for my company and for the 120 small family farmers who, 9 with this growing grass-fed organic segment, have found a 10 way to compete, at least to this point, in an industry 11 that continues to favor large corporate farm entities. 12 Your Honor, that's the -- that's the end of my 13 submitted testimony. As Mr. English said following 14 yesterday's comments, I would like to add something, if I 15 could. 16 THE COURT: Yes, you may. Thank you. And I 17 appreciate that you addressed me directly, but it 18 interferes with everyone else's ability to hear you. 19 THE WITNESS: Got it. Feel free to correct me, I 20 won't be offended. 21 If -- if my testimony should get lost and you can 22 only remember one thing I say today, please remember this 23 point: Milk is not inelastic. Milk is not inelastic. If 24 we raise prices, volumes will go down. If we drop prices, 25 volumes will go up. Milk is not inelastic. This is not 26 ivory-tower stuff based on theoretical data from models 27 from the 1970s or 1980s. This is real world stuff, and I 28 can provide you real world examples. 10043 1 30 days ago we ran a promotion. We dropped our 2 price 7%. I saw a spike in my volume of 30%. When that 3 promotion ended, the price went back up 7%, my volume 4 dropped 30%. Right? It's the reason we run promotions. 5 We know if we change price, volume will change. That's 6 why we promote. 7 We took a price increase, our last major price 8 increase was in early 2022. When we took that price 9 increase, our volumes fell. In fact, we had at least one 10 retailer who came to us and said, we are going to take you 11 off the shelf because your price is too high. That's 12 volume that I don't have because my price is too high. If 13 I move my price, my volume changes. Milk is not 14 inelastic. 15 The last major price increase we took before that, 16 I believe, was 2019. And I wasn't here for that, so I may 17 have the date wrong. But the blowback on volume loss was 18 so great that we had to reverse course and adjust our 19 prices back -- back down in order to keep our business. 20 Milk is not inelastic. 21 THE COURT: And because "inelastic" sounds so much 22 like "elastic," your emphasis is that milk is? 23 THE WITNESS: Milk is not inelastic. That's not 24 unique to Maple Hill. I can assure you it's -- it's most 25 definitely true for Organic Valley and for Horizon, and I 26 know that because they are competitors and I watch them on 27 a daily basis. I -- I have a fair amount of confidence 28 that it's not true for any organic or specialty milk or 10044 1 value-added differentiated milk. 2 I would like just, in sum, to say, I want to make 3 sure that the USDA knows that milk is not inelastic. If 4 our prices go up, our volumes will come down. 5 Thank you. 6 THE COURT: Thank you so much, Mr. Hau. This 7 is -- this is a wonderful viewpoint that we needed to 8 hear. I'm so glad that you are here. 9 There are a number of comments you make that lack 10 the illustrative examples to prove them, so I hope on 11 cross-examination you can bring this more to our 12 attention. And if you are not asked to do that on 13 cross-examination, I'm going to ask you to do that on your 14 own redirect. 15 In other words, if you make a summary statement, 16 what are some of the ways you can persuade people in this 17 room that what you are saying is true? 18 THE WITNESS: I appreciate that opportunity, Your 19 Honor. 20 THE COURT: All right. Very good. 21 Mr. English. 22 BY MR. ENGLISH: 23 Q. Good morning again, Mr. Hau. 24 A. Good morning. 25 Q. So you're not here to oppose the existence of the 26 Federal Milk Marketing Order system, are you? 27 A. I prefer it not exist, but I'm here out of concern 28 that it get even worse than it already is. 10045 1 Q. And your real concern is that you produce a 2 product, organic grass-fed, that doesn't fit into the 3 Federal Order system, correct? 4 A. That's correct. 5 Q. And are you aware that the Milk Innovation Group 6 made a proposal that USDA declined to hear that would have 7 differentiated the treatment of organic milk? 8 A. So I have been told. 9 Q. And USDA declined to hear that, correct? 10 A. Correct. 11 Q. And so instead, what you are here to say is, you 12 can't just keep increasing our price, because if you do 13 that, that money is not going to organic, it's going 14 elsewhere, correct? 15 As the differential goes up, that money, that 16 increased payment that you have to make that you call a 17 tax, isn't going to do anything, in fact, it's going to 18 hurt grass-fed organic, correct? 19 A. And that's correct. I would take that one step 20 further, actually. I'm here today -- if this goes 21 through, Maple Hill will likely survive. We have grown 22 enough and we're strong enough now that I don't think we 23 will go out of business. 24 That said, I will have to change my strategy. I 25 can't -- I can't continue to grow and thrive and flourish 26 the way we are if my costs go up that extensively. 27 The broader point I want to make, though, is my 28 business is not unique here. Innovation that comes into 10046 1 dairy tends to come from small companies like mine. I 2 have had the benefit of a few years to -- to build and get 3 our act together, so to speak, so we'll be okay. It's 4 going to hurt, it's going to mean something different, but 5 we'll be okay. 6 But other innovators, other small companies out 7 there that are trying to do what we have done since 2009 8 who are starting up today aren't going to make it, right? 9 So if this goes through, we will be killing 10 innovation in the dairy industry. You won't kill me, you 11 will wound me, but you will prevent other small companies 12 who have ideas and are trying to get started from getting 13 started. It just won't be affordable. 14 MR. ENGLISH: That's all I have, Your Honor. 15 THE COURT: Who would like to begin 16 cross-examination? 17 CROSS-EXAMINATION 18 BY MR. MILTNER: 19 Q. Good morning, Mr. Hau. 20 A. Good morning. 21 Q. My name is Ryan Miltner. I represent Select Milk 22 Producers. 23 I wanted to re-ask your question Mr. English did 24 because I wasn't clear on your answer. 25 When he asked if you are here opposing the 26 existence the Federal Milk Marketing Order program, did 27 you say you prefer it did not exist? 28 A. That would be my preference. 10047 1 Q. Okay. 2 A. I'm sorry, let me clarify that. 3 Q. Please. 4 A. I would prefer that those of us on the organic 5 side that do not participate in the FMMO process not be 6 obligated to pay into the pooling fund. 7 Q. Now, I understand from your statement that you do 8 not own any processing facilities, correct? 9 A. That's correct. 10 Q. And is all your of your milk purchased from 11 cooperatives? 12 A. Yes. 13 Q. So does -- does Maple Hill itself make any 14 payments into the Federal Order pools? 15 A. We do. 16 Q. You do. I have seen your product on the shelf in 17 my little part of the world in Ohio, which is where I am. 18 I have not tried it. I'm sure it's a fine product. I'm 19 sure it's very, very good. I forget what it's priced in 20 my market. 21 I looked online in Sprouts right now, a Whole 22 Foods type market in predominantly the Southwest, for the 23 purpose of the record. And they have got it listed at 24 $6.99 a half gallon. 25 Does that sound about right for the retail price 26 for your fluid milk products? 27 A. That's correct. You will find it on average 28 between $6.59 and $6.99 for a half gallon. 10048 1 Q. That's a half gallon of whole milk, right? 2 A. Correct. 3 Q. Sprouts also has a 6.99 half gallon of 2% milk. 4 Are they generally priced in alignment? 5 A. That's correct. 6 Q. I don't want specific numbers, unless you want to 7 share them, but when you -- when you contract with a 8 cooperative for milk, are you generally paying a fixed 9 price for a period of time; in other words, you know, 10 we'll pay $40 a hundredweight for milk for the next six 11 months or a year or a something like that? 12 A. To be clear, we contract directly with the 13 farmers. We use the co-ops that they -- they handle our 14 payroll and, you know, help with the producers with farm 15 things and that type of thing, but we contract directly 16 with the farmers. We -- it is a contracted price. There 17 is a formula that all of our farmers participate. 18 Q. Do you use a formula that fluctuates based on the 19 underlying regulated prices? 20 A. No. 21 Q. So I understand that there are -- when you look at 22 a retail shelf price of milk, there's a lot in there 23 that's not milk, in terms of cost, right? You have -- you 24 have packaging, correct? And transportation, you have 25 profit for the processor, and a markup from the retailer, 26 and probably some other things, correct? 27 A. That is correct. 28 Q. Now, when I multiply out a per hundredweight 10049 1 equivalent of your shelf price, though, I get whole 2 milk -- well, really it would be both -- $162.59 per 3 hundredweight on a retail equivalent. 4 Does that sound right? 5 A. I'm not familiar with that number. You have to 6 walk me through how you got to that. 7 Q. Sure. And there's about 11.6 gallons of milk in a 8 hundredweight, correct? 9 A. Okay. 10 Q. So if I take 6.99 and I multiply that times -- 11 well, I took 11.6 times two, right? Because there are 24 12 half gallons, about 24 half gallons or so in a 13 hundredweight, and that's the equivalent hundredweight, 14 per hundredweight price of your product to the consumer. 15 A. I would agree with your assumptions there if 16 that's the number you get to. That makes sense. It's not 17 a number we typically look at. 18 Q. Everything we look at here is a number. 19 THE COURT: Say it again? 20 MR. MILTNER: I said everything we look at here 21 tends to be in hundredweight rather than gallons. 22 BY MR. MILTNER: 23 Q. Again, I don't have a point of reference for what 24 grass-fed organic milk is, and I don't want you to share 25 something you don't want to put on a public record. But 26 what is a -- what is a reasonable return to a farmer who 27 is going to be organic grass grazed if they -- if they are 28 saying these are my costs, right, I need this much a 10050 1 hundredweight to cover my farm costs, what do you think 2 that would be? 3 A. I can't comment on the economics of the farmers 4 because every farmer is different. And because these are 5 small family farms, many of them have side hustles that 6 help keep the family farm going. But predominantly, they 7 rely on the dairy to fund their farm. It's -- without 8 going into specifics of what we pay, or our numbers are, 9 it's not unusual for a grass-fed farmer to expect to get 10 38 to $40 a hundredweight for their milk. 11 Q. Which is a modest increase over, I think, what an 12 organic farmer would be paid, grass-grazed aside? 13 A. I would expect an organic farmer to be somewhere 14 in the $34 range. 15 Q. I want to ask about elasticity because you heard 16 part of this yesterday and we heard testimony about it 17 earlier in the hearing. 18 Would you agree with me that own-price elasticity 19 is not a static number? In other words, if the elasticity 20 at one price point might be different than the elasticity 21 of the same product at a higher price point, would you 22 agree with that? 23 A. So in other words, while I might be willing to 24 spend $0.10 more for a half gallon of milk, I'm not 25 willing to spend $100 more for a half gallon of milk. 26 Q. I think that's part of what I was asking. 27 A. I would agree. 28 Q. And you heard the testimony from the gentleman, 10051 1 Mr. Carson, I think, from United Dairy yesterday? 2 A. Yes. 3 Q. And he said that a gallon of milk in the 2 to $3 4 range at a somewhat different price and purchase response 5 than a gallon of milk at $5 price point. 6 Did you hear that testimony? 7 A. I did. 8 Q. Which kind of is -- I interpreted that as the same 9 point, that the demand is more inelastic at a lower point 10 than at a higher price point. At least that's how I 11 interpret it. 12 Would you have interpreted that similarly? 13 A. I would. I would agree that taking pricing 14 increases on commodity conventional milk is lower risk 15 than taking a price increase on the faster growing, more 16 consumer demanded specialty value-added differentiated 17 products that tend to be priced higher. 18 Q. And, you know, all the lawyers up here don't want 19 to ever get in the position of testifying, but for the 20 purposes of your frame of reference, the cooperative I 21 represent founded fa!rlife and was an equal partner in 22 that until it was sold, so we are familiar with higher 23 price points and value-added milks. 24 Would you -- with your product at essentially a 25 $14 a gallon price point, would you expect that your 26 elasticities would be different than those for someone 27 selling a gallon of milk at 2.99? 28 A. Absolutely. 10052 1 Q. And you described a price change you made in 2 relatively recent times. 3 Was that around the Thanksgiving holidays? 4 A. Our last significant price increase would have 5 been in February of '22. 6 Q. Okay. And I was -- I'm sorry, you did speak to 7 two of those. I was -- I'm referring to the -- 8 A. The promotion. 9 Q. -- the promotion. 10 Was that over a holiday period? 11 A. It was the first week in November. 12 Q. Okay. So in -- within your company, a promotion 13 like that, would you consider that a direct marketing 14 expense? 15 A. We would -- we would account for it as a price 16 discount, so it comes out of our revenue. 17 Q. In addition to lowering the price, do you do 18 any -- any educational work to your customers or your 19 ultimate consumers that, you know, hey, we're offering a 20 special? 21 A. Are you speaking of marketing? 22 Q. Yes. 23 A. We do do some marketing. We don't do a lot of 24 marketing, mostly because it -- we just don't have the 25 funds to do it. That's where the USDA grant is helping 26 quite a bit. 27 When -- when you -- when you innovate -- and we 28 saw it in the organic business when organic started, and 10053 1 we're seeing it now in grass-fed organic as grass-fed 2 organic takes off. Your -- your tip of the spear, your 3 first consumers are your best consumers. They are the 4 most educated ones. They go find you. You don't have to 5 market to them. They are doing their research. 6 In the case of grass-fed organic, they are saying, 7 I want to give something better to my family, so I'm 8 looking at organic and I'm looking at what organic gives 9 me, and as I'm doing my research, I'm learning about this 10 grass-fed organic, which is even better, so now I want 11 grass-fed organic. Who is in the grass-fed organic space? 12 Oh, this Maple Hill looks good. 13 And they come to us. That's terrific. Right? 14 That's how you build a business. You offer that consumer 15 something that they are looking for without a lot of 16 prompting. 17 The way you grow, and the way you thrive, and the 18 way you succeed, is you broaden that base. So now I have 19 got those consumers, and that's great, that gives me a 20 base to start with. Now my focus is, how do I convince 21 more consumers? And those are harder to find, right? 22 Like yourself, as you described, right? I have to go out 23 and tell you. I have to go out and market and get in 24 front of you. That's the hard part. That costs money, 25 right? 26 And that's where it becomes difficult in a thin 27 margin business, like grass-fed organic, or quite frankly, 28 any organic or dairy for that matter, to spend those 10054 1 dollars. But we do do it on a -- what I'll call a gorilla 2 basis. A little social media here and there. If I can 3 get a PR article, you know, somebody to write an article 4 to talk about us, that's what you try to do because it's 5 inexpensive and you can reach a broad base. 6 Q. Thanks for that background. 7 Did you say your discount -- on that last discount 8 was at $0.50 a unit? 9 A. The last promotion? 10 Q. Yes. 11 A. In November? It was about 7% -- it was -- it was 12 about $0.60 on a 6.60 price. 13 Q. Okay. So the -- what would be 6.59, is now 5.99, 14 if I went to the store that week? 15 A. Correct. 16 Q. And your products are, from what I saw, for as 17 good as the internet might be, $0.50 a unit above 18 convictional organic, if that makes any sense? That's not 19 grass-grazed organic. 20 A. Could you repeat that? 21 Q. Yeah. I'll make it a little more clear. 22 If -- if -- from what I have seen, if I went to 23 the store and Organic Valley whole milk was 5.49 a gallon, 24 you would be at -- I'm sorry -- if Organic Valley would be 25 essentially $0.50 lower than you on the shelf, for organic 26 non-grazed milk? 27 A. For their standard organic offering without the 28 promotion, generally about $1 difference. 10055 1 Q. Okay. So if you are trying to target that second 2 layer customer, right, and you -- you know there's a bunch 3 of people out there that might be buying Horizon or 4 Organic Valley or another premium milk, but they are 5 paying $1 less per unit, when you offer a discount and the 6 regular customer's thrilled, but you are looking at maybe 7 the person who would otherwise pick up Horizon is going to 8 give me a shot this time, correct? 9 A. I'm trying to get a trial, correct. 10 Q. And some of those consumers are going to take your 11 product home and try it and love it, and come back next 12 week, and they are going to grab it, and maybe not, it 13 might not matter that the price is -- is back up to 14 regular -- the regular price, correct? 15 A. Correct. 16 Q. Some of them, though, are going to try them and 17 say, yeah, this is fine, but to me, I don't see a 18 difference between this and OV, and I'm going to save 19 myself $1 a gallon, or $1 per unit, and continue 20 purchasing what they have -- they are going to make a 21 price decision and not switch brands, correct? 22 A. They are going to make a value decision that says 23 this might be better, maybe it is not better, but if it is 24 better, is it worth an extra dollar to me? Some will say 25 yes, and some will say no. 26 Q. Okay. When the economists look at that situation, 27 they have to unpack a whole lot more than pricing when 28 they measure that elasticity, don't they? Because it's 10056 1 not just own-price elasticity, it is substitution 2 elasticity, correct? 3 A. I think -- I think I'm with you. 4 Q. Okay. And all I'm trying to say is we have a 5 bunch of economists and folks who do it day to day, trying 6 to figure out what the elasticity is. Own-price 7 elasticity at a conventional price point is only one part 8 of the question, correct? 9 A. I think I'm with you. 10 Q. Okay. I think what I'm trying to say is that your 11 experience, I absolutely believe it, not just because you 12 are under oath, it makes logical sense. But there are -- 13 there are -- when you say there's a 30% change in your 14 demand week to week, you have got a lot going on there 15 besides just the price of our product moved, correct? 16 A. I'm not sure I understand what you mean by that. 17 Q. Okay. I'm trying to figure out how to better 18 articulate it, and maybe I'm just trying to summarize for 19 my own sake the answers you have already given, so I will 20 leave it at that for the time being. If I can think of a 21 better way to phrase it, I may pop back up here. 22 MR. MILTNER: But I thank you for the answers you 23 have given. 24 THE WITNESS: Okay. 25 THE COURT: Who next has cross-examination for 26 Mr. Hau? 27 // 28 // 10057 1 CROSS-EXAMINATION 2 BY MS. HANCOCK: 3 Q. Good morning, Mr. Hau. Nicole Hancock on behalf 4 of National Milk. 5 Are you regulated on Federal Order 1? 6 A. Correct. 7 Q. What is the name of the entity that's regulated on 8 that on that order? Do you know what it's identified as 9 through the USDA? 10 A. I do not. 11 Q. Is it NFO Maple Hill? 12 A. We use -- there's basically four co-ops that our 13 farmers use. The biggest one would be DFA. NFO is one of 14 them as well. 15 Q. What are the others? 16 A. I believe it is United Ag and Producers. 17 Q. Okay. 18 MS. HANCOCK: Thank you very much. 19 THE COURT: Who next has cross-examination before 20 I turn to the Agricultural Marketing Service? 21 Mr. English, if you have something further before 22 I turn to the Agricultural Marketing Service. 23 MR. ENGLISH: I thought it would make sense, Your 24 Honor, in the context that it's a direct follow-up on 25 Mr. Miltner's questions. 26 // 27 // 28 // 10058 1 REDIRECT EXAMINATION 2 BY MR. ENGLISH: 3 Q. I think maybe we got a little confused about what 4 happens in terms of payments. 5 I think you said that the co-ops help you with the 6 payroll, correct? 7 A. Correct. 8 Q. To the extent that there is a producer settlement 9 fund payment due, even though they help you with the 10 payroll, do they send you a bill for that? 11 A. Yes. 12 Q. Okay. So -- so Maple Hill Creamery pays the 13 producer settlement fund payment through the co-ops, 14 correct? 15 A. Correct. 16 Q. And that is a highly variable number every month, 17 correct? 18 A. I believe so. 19 THE COURT: Your response was what? 20 THE WITNESS: I believe so. 21 BY MR. ENGLISH: 22 Q. And I certainly do not want confidential 23 information. Mr. Miltner has asked -- or he's done his 24 little research -- not little research -- real research 25 going out and finding what the retail price of your 26 product was. 27 That's up to the retailer ultimately, correct? 28 A. Absolutely. 10059 1 Q. So whatever their markup is at the retail level, 2 that's their business, correct? 3 A. Correct. 4 MR. ENGLISH: That's all I have. 5 Thank you, Your Honor. 6 THE COURT: Before I ask Agricultural Marketing 7 Service for questions, I would like to take a five-minute 8 stretch break. Everyone may move around. Don't go too 9 far. I intend to go back on record at 8:55. 10 (Whereupon, a break was taken.) 11 THE COURT: Let's go back on record. 12 We're back on record at 8:57. 13 Does anyone else have any questions before I ask 14 the Agricultural Marketing Service for questions? 15 I see none. I invite the Agricultural Marketing 16 Service questions. 17 CROSS-EXAMINATION 18 BY MS. TAYLOR: 19 Q. Good morning. 20 A. Good morning. 21 Q. Thank you for coming to testify and your patience 22 to be here for a couple days to get on this morning. I 23 appreciate that. 24 A. I understand. 25 Q. So the cross did help clarify a few things, the 26 questions you've answered so far, which was very helpful. 27 I did want to ask a few more questions on kind of 28 Maple Hill's milk supply and you -- so you say a vast 10060 1 majority of your milk is fluid, and then it looks like you 2 have other Class II products. 3 Can you give maybe a percentage of how much of 4 your raw milk goes into Class I versus say Class II or 5 other classes? 6 A. I can tell you about 65 to 70% of our products are 7 fluid milk. I can't answer your question directly because 8 it doesn't mean 65 to 70% of the milk goes into fluid 9 milk. But -- but obviously milk -- fluid milk, you know, 10 the Class I products require more milk, so it's -- it's 11 slightly more than that. 12 Q. Okay. And then -- so I think we learned through 13 cross that Maple Hill itself is not a regulated handler. 14 You don't own a plant. So you pool through either the 15 co-op or maybe you get your payment obligation through the 16 pool plant you use to co-pack and that's how you pay your 17 obligation? 18 A. The obligation comes from the co-packer. 19 Q. Okay. 20 A. They are the ones that pay and bill me each month. 21 Q. Okay. Do you know for your manufactured Class II 22 products, is that milk pooled? 23 A. It is not. 24 Q. So you only pool your Class I? 25 A. Right. 26 Q. Okay. You talk on page 4, and Mr. Miltner asked a 27 few questions kind of around this subject, about your 28 already thin margin dairy business. We have had testimony 10061 1 from conventional fluid milk processors about their 2 margins, so I was wondering if you could just talk a 3 little bit about from the organic side, on the grass-fed 4 side, what your margins are, kind of in relation to 5 conventional. 6 A. I would say the dairy industry in general has thin 7 margins, right? We're not semiconductors. It's difficult 8 for me to do the comparison because I'm a growing company. 9 Right? So a startup, you know, back in the day, it's not 10 true today, but, you know, when we were starting up, we 11 were losing money, right, and that's not uncommon for a 12 startup, especially in this space. 13 Looking at it just general industry, my general 14 industry experience, I would expect you are getting 15 slightly better margins the higher up in the hierarchy you 16 go. So if you take conventional premium, super premium, 17 the higher up you go, likely the better margins you are 18 getting. But it really depends on the company. 19 Q. Okay. And it sounded like from what Mr. Miltner 20 was asking, your grass-fed organic milk is pretty high up 21 on the food chain when it comes to the premium? 22 A. It is super premium. 23 Q. Okay. 24 A. I don't think I have the highest margins, but, 25 again, that's based probably more on where I am in the 26 life cycle and the evolution than anything else. 27 Q. That makes sense. 28 And you discuss that you do pay your farmers a 10062 1 fixed price, contracted with them directly, correct? 2 A. Correct. 3 Q. Do you use any risk management tools to help cover 4 some of the risk on that side of the business on the 5 supply side? 6 A. We do not use risk management tools. 7 Q. Okay. You also discuss how you balance your own 8 milk, and I was just wondering if you could just expand on 9 what that means, how you do that, et cetera, since you 10 don't have a plant. So I just wanted you to talk a little 11 bit more about that. 12 A. You hustle. The -- so obviously cows don't milk 13 on consumer demand, and consumers don't demand on cow 14 schedules. So, you know, at times, particularly the 15 spring flush, you have more milk than you want, and at 16 times, like, you know, winter, you have less milk than 17 consumers want. To the extent you can, you try to manage 18 that through inventories. 19 Now, we have -- we are, you know, extended life, 20 extended shelf life product, but still, it's dairy. 21 Right? It goes bad, so you can't -- you can't keep 22 inventories for too long. 23 So then what you do is you look for opportunities 24 to what I'll call lay that milk off. Right? And your 25 first option is customers who have flexibility. So if, 26 for example, maybe you have a customer who uses powder, 27 they are going to dry it, right? They can keep inventory, 28 so they might be willing to do a deal with you when you 10063 1 are heavy on milk, I'll take it; when you are light on 2 milk, we'll hold off. 3 Or you will try to sell it, first as grass-fed 4 organic, if you can find a customer who needs grass-fed 5 organic milk. That can be difficult because, generally 6 speaking, everybody in the industry is -- has the same 7 issues you have: At flush we all have milk, and in the 8 winter, none of us have milk. So that can be difficult, 9 but once in a while you get lucky. 10 If you can't find someone to take it as grass-fed 11 organic, you will try to find an organic buyer. Right? 12 And then maybe instead of getting $40 a hundredweight, I 13 can get $35 a hundredweight. If I can't find an organic 14 buyer, I'll try to find a conventional buyer. If I can't 15 find a conventional buyer, I'll dump the milk and we take 16 the loss. 17 Q. Okay. 18 A. And because of that I tend to underestimate or 19 undershoot for the amount of milk I want. Right? If I'm 20 not sure demand is going to be there. 21 The two -- the two most considered and difficult 22 decisions we make in our company, number two is pricing. 23 Right? Because in case you missed it, milk is not 24 inelastic. 25 THE COURT: Milk is not what? 26 THE WITNESS: Milk is not inelastic. 27 BY MS. TAYLOR: 28 Q. I assure you, I wrote that down. 10064 1 A. So anytime we're talking about pricing, you are 2 taking a significant risk with the business, and we -- we 3 spend a lot of time thinking about that and debating it. 4 The most difficult and considered decision is 5 letting the farm go. Right? So if at all you can, you 6 take the hit, and you absorb the loss. So if I have to 7 dump milk, I will dump milk. If I have a significant drop 8 in demand such that I can't survive, if I have to keep 9 doing that, then I have no choice but to let a farm go. 10 And you never want to be in that position. It is a 11 terrible position to be in. 12 I know you guys appreciate that as much as anyone 13 in the room, because you are going to make some decisions 14 here that will have impacts for those farmers. 15 So you tend to say, if I'm going to need a million 16 hundredweights, maybe I'm going to target 950,000. Right? 17 And you hope that you don't have to try to balance the 18 milk. But when you do have to balance, those are the 19 steps you go through, and you rely on your network of 20 contacts and people in the industry who might need it 21 today or might be able to work with you on something. 22 Q. Thank you. I appreciate that information. 23 And you mentioned sometimes one of our options, 24 which is not first on the list, but an option on the list 25 is to sell the milk to some type of powder plant that can 26 dry that milk; is that correct? 27 A. It might be a powder plant. It might be -- we 28 have sold milk to, for example, ByHeart who is an infant 10065 1 formula company. Right? 2 Q. Okay. 3 A. They dry the milk, so they -- they -- you know, 4 they have had some flexibility in terms of when they can 5 take it, that type of thing. 6 Q. And when you sell it on that, in that way, is it 7 sold to go through the process as organic grass-fed or 8 just conventional prices? 9 A. In ByHeart's case, they want it to advertise 10 organic grass-fed, right? They are trying to sell the 11 highest quality, super-premium infant formula, so their 12 game is to say we're using the highest quality 13 super-premium milk in our product. And that works out 14 very well for us. In fact, they -- they have quoted us in 15 their website, we don't co-brand the product. But in that 16 case, it works out very well. They are willing to pay a 17 premium because they are going to use it as a premium 18 product. If you had a conventional manufacturer, and you 19 are going to use it as conventional milk, great, I'm glad 20 you are giving me the best milk. I can't -- I can't 21 realize a value on that, so I'm going to pay you what I 22 would pay a conventional farmer. 23 Q. Okay. 24 THE COURT: Would you spell the name of ByHeart? 25 I just want to the make sure I got it right. 26 THE WITNESS: Yeah, it's B-Y-H-E-A-R-T. 27 BY MS. TAYLOR: 28 Q. Okay. And you gave an example on pages 4 into 5, 10066 1 talking about a $600,000 pool obligation. 2 And is that an annual number that you calculated 3 there in your example? 4 A. Yes. 5 Q. Okay. And you said, "The proposals" -- in this 6 sentence: "The proposals considered here today, that 7 could take that up another 80%." 8 And I just wanted to kind of ask how you came up 9 with that number? 10 A. Looking at the NMPF proposal, my point is -- is 11 proposed to go from, I believe, 2.20 to $4. 12 Q. And that's -- the plant that packages your fluid? 13 A. Correct. 14 Q. I had another -- my last question, in two 15 different places you kind of talk, sort of the same thing. 16 In one place you talk about how -- and this is in National 17 Milk's Proposal 19 -- would mean fewer Class I dollars in 18 pool, and in summary, you have a point that says 19 "potentially lower pay prices for farmers." 20 And I just wondered if you could expand on why you 21 think those things would happen. 22 A. Okay. So let me first address the potentially 23 lower dollars in the pool. If -- 24 THE COURT: I'm going to ask you to have the mic 25 be a little further from your mouth. 26 Yeah. Perfect. Thank you. 27 THE WITNESS: Is that better? 28 THE COURT: Yeah, so that we don't get -- 10067 1 THE WITNESS: Feedback? 2 THE COURT: Yes. 3 THE WITNESS: I hope everyone heard what I had to 4 say. Do you want me to start over? 5 So -- so the concept here is, let's increase the 6 pooling, and in so doing, there's more money in the pool. 7 That only works if you don't reduce the amount of milk in 8 the pool. And what I'm saying is, if I have to increase 9 my price, and if my peers in the -- in the innovative 10 value-add space have to increase their price, we are going 11 to sell less milk. 12 Now, if you think that milk's going to 13 conventional, then you are fine, there's still money in 14 the pool. I don't think it's going to conventional, and I 15 can give you reasons for that. And therefore, money comes 16 out of the pool. And that's why I say potentially. 17 Now, I don't know how consumers are going to react 18 or how far this thing is going to go and how much prices 19 will go up, so I can't say that with certainty. But 20 that's why I say potentially you have less money in the 21 pool. 22 In terms of farmer pay prices, again, I don't pay 23 the FMMO price, I pay well above the FMMO price. And our 24 ability to pay that price is dependent on our ability to 25 realize profitability in the business. So if the business 26 is growing, then I want more milk, and I'm willing to pay 27 for more milk. And if that means I have to pay a higher 28 price, I will pay a higher price. It is a market driven 10068 1 force. 2 If, on the other hand, I need less milk, or I 3 can't afford to pay for the milk because I can't take 4 prices up higher, then I have no choice but to reduce the 5 pay price to the farmer. 6 Now, again, that's tough to do, because if I 7 put -- if I put farmers out of business, I don't have 8 milk. So reducing prices is generally not the best 9 option. The best option is to increase consumer prices, 10 but I know, I'm already at a threshold where, when I go to 11 do a price increase, I lose shelf space and I lose 12 consumers. So if I have to raise my price, I'm going to 13 lose profitability. Right? 14 And this is the squeeze that businesses like mine 15 will be in here if you -- if you raise any of our costs 16 80%. Either I have to go find some place else to lower 17 the cost. And I can't, right? I have got the farmer. 18 That's my biggest cost. So I can reduce price to the 19 farmers, if I thought they could survive on a lower cost. 20 I could reduce my co-pack cost, but we're small 21 companies. Right? We don't have that kind of leverage. 22 So -- so co-packers generally aren't going to work with us 23 a lot there. 24 I can reduce any staff. I have 18 employees. You 25 know, I'm the president, I'm the CFO, and if you need a 26 lock changed on the door, I'm the guy you call. So, you 27 know, there's not places to go. 28 And so this is the squeeze you get in, and this is 10069 1 why companies go out of business. Right? I can't take 2 price, and my costs are killing me, so I lose 3 profitability, and I have no choice but to close shop. 4 Or I think what will happen in a lot of cases 5 here, I just won't open shop. Right? I'll sit here and 6 I'll look at the economics of, I've got this great idea, I 7 want to launch this new milk, I think there's a great 8 consumer demand for it, but I can't make the cost and the 9 pricing equation work, so I just won't launch, and that's 10 where you lose the innovation. 11 Q. And you answered a question from Mr. English, and 12 that the price that -- you know, you're wholesale, and 13 what the retailers do at the market is their decision, on 14 the grocery store shelf? 15 A. Correct. 16 Q. So can you just talk a bit about do you do a fixed 17 price to those retailers then? 18 A. I do. 19 Q. Okay. And how often is that negotiated? 20 A. Irregularly. So, you know, once I -- once I set 21 the price, it is set until we need to take an increase or 22 a decrease, and then annually we'll negotiate promotions. 23 Right? Because the promotion benefits both of us, you 24 move more volume. So generally speaking, a retailer is 25 going to want to take 30 to 40% margin to cover their 26 costs, and I know that. So when I take a price increase, 27 I have some idea of what my shelf price is going to be, 28 but I don't control it. 10070 1 Q. Okay. 2 A. It might be worth further making the point, when I 3 take a price increase, it's not just my price increase the 4 consumer pays. So, for example, let's say, just to make 5 easy math, I have to take a 10% price increase. Well, 6 let's -- let's walk through the numbers if you will bear 7 with me. 8 Let's say I'm selling my milk for 6.69. 9 Q. When you say that, you are selling it to the 10 retailer? 11 A. Sorry. Very good point. I'm not selling it. The 12 retailer is selling it for 6.69. So it's on the shelf for 13 6.69, and I need to take, let's say a 5% increase. So 14 let's call it $0.35. So now I'm at $7, about. And the 15 retailer is going to say, well, hang on, I need my 40% 16 margin, so I'm going to tack on 40% of that 5%, which is 17 an additional 2%. So now I'm going to be at $7.19. 18 Right? No retailer sells anything for $7.19. Right? You 19 have price points, 7.29, 7.59, 7.99, or whatever it is. 20 So in this case, likely the retailer is going to 21 go to 7.29. So now I need 5% to cover my increase in 22 pooling, but my shelf price just went from 6.59 to 7.29, 23 which is a lot more than 5%, right? Now the consumer's 24 saying, hey, this is an additional 10%, right? Now I'm 25 paying significantly more money, you know, our initial 5%, 26 now I'm paying additionally more money for this product. 27 And now I get to that point, you know, it's not just -- 28 it's just not worth it. 10071 1 And maybe I go to organic. Maybe I go to 2 conventional. Doubtful you go to conventional. But guess 3 what? Organic's going up, too. 4 Now, from an elasticity standpoint -- because I 5 love to talk about elasticity -- the question is, where 6 does that consumer go? The consumer is not going to 7 conventional milk, right? The consumer -- the consumer is 8 already paying $6.69 for my value-added milk, and they are 9 making that value decision. I want -- I want those 10 benefits. I'm not happy with just commodity milk, I want 11 those benefits, whether it be fa!rlife and more protein, 12 or Lactaid with -- you know, with lactose-free, or A2, 13 or -- or grass-fed organic in my case, or whatever it is, 14 I want those benefits. I want something healthier. I 15 want something that's better for the environment, 16 something that makes me feel better about my purchase, and 17 I'm willing to pay for it. But sorry, Maple Hill, I'm not 18 willing to pay 7.29. What are my other choices? 19 You know what? There's oat milk over here that's 20 also good for the environment, or at least they say they 21 are. We'll debate that in a different hearing. But -- 22 so -- so now you are just -- right? This is where the 23 pooling starts to go down. Now I'm selling less milk. 24 Right? Now that consumer is finding alternatives in the 25 dairy industry. I don't just compete with -- with my 26 dairy competitors. I'm competing with everyone in the 27 dairy space, whether that be juice, plant-based beverages, 28 Kombucha, whatever -- you know, whatever it is that the 10072 1 customer is looking for. 2 I'm trying to get to that customer who is most 3 interested in the nutrition -- the nutritional content of 4 their beverage, and in my case, the regenerative aspects 5 that that gives. Right? Better for the environment. 6 It's 100% regenerative agriculture. That matters to a 7 consumer and they are willing to pay for it. If they 8 can't get it from me, they are going to go look for 9 somebody else they can get it from, and that's not 10 necessarily going to be in the milk market. 11 Q. And you sell at a fixed price to your buyers. And 12 we have had discussion, a long time ago, at this hearing 13 about the ability of buyers to hedge if they have a fixed 14 price. 15 Do you know if your purchasers are able to use any 16 risk management tools to lay off risk, since they know 17 what the price is you are going to sell the milk to them? 18 A. Are you speaking to the retailer? 19 Q. Yes. 20 A. I don't believe any retailers hedge milk price. 21 Q. Okay. 22 MS. TAYLOR: That's it from AMS. Thank you for 23 your time today. 24 THE WITNESS: Thank you. 25 THE COURT: Mr. English. 26 MR. ENGLISH: I have no further questions. I just 27 on behalf of the witness move the admission of 28 Exhibit 432. And I thank him for his time. 10073 1 THE COURT: Is there any objection? 2 There is none. Exhibit 432, also known as 3 Hau-001, is admitted into evidence. 4 (Thereafter, Exhibit Number 432 was received 5 into evidence.) 6 THE COURT: And, Mr. Hau, I have no additional 7 questions. Your testimony was excellent, and I thank you. 8 THE WITNESS: Thank you. 9 THE COURT: You may step down. 10 Now I'm looking at the two exhibits that we marked 11 yesterday. These are Exhibit 433, also IDFA Exhibit 57, 12 and Exhibit 434, IDFA Exhibit 58. I am watching as the 13 witness comes to the witness stand. 14 MR. ROSENBAUM: Your Honor, those are not actually 15 going to be the next exhibits. Those are Mr. Brown's 16 exhibits, so we will be doing them, well, perhaps this 17 afternoon depending upon how quickly things move. 18 But at this point I would like to call 19 Dr. Balagtas to the witness stand. 20 THE COURT: Let's go off record for just a moment. 21 (An off-the-record discussion took place.) 22 THE COURT: All right. We're back on the record 23 at 9:22. While the witness' testimony is being 24 distributed and marked, we'll take a ten-minute break. 25 Mr. Rosenbaum, let's come back ready to go at 26 9:35. 27 MR. ROSENBAUM: Thank you, Your Honor. 28 (Whereupon, a break was taken.) 10074 1 THE COURT: Let's go back on record. 2 We're back on record at 9:35. 3 And I have two exhibits that need a number, and I 4 believe our next number is 435. 5 (Thereafter, Exhibit Number 435 was marked 6 for identification.) 7 THE COURT: So will that be for the testimony, 8 Mr. Rosenbaum? 9 MR. ROSENBAUM: Yes, Your Honor. It's Hearing -- 10 it is IDFA Exhibit 61. That should be Hearing 11 Exhibit 435. 12 THE COURT: Thank you. Then I will mark as 436, 13 IDFA Exhibit 62. 14 (Thereafter, Exhibit Number 436 was marked 15 for identification.) 16 THE COURT: I'd like the witness now to state and 17 spell his name. 18 THE WITNESS: Joseph Balagtas, J-O-S-E-P-H, last 19 name B-A-L-A-G-T-A-S. 20 THE COURT: Have you previously testified in this 21 proceeding? 22 THE WITNESS: No, Your Honor. 23 THE COURT: Now, as we -- I'll swear you in, but 24 as we go forward, we'll determine whether you need to 25 scoot your chair a little closer to the mic or make any 26 other adjustment as to how people can hear you. 27 First I'll swear you in. 28 // 10075 1 JOSEPH V. BALAGTAS, 2 Being first duly sworn, was examined and 3 testified as follows: 4 THE COURT: Mr. Rosenbaum. 5 DIRECT EXAMINATION 6 BY MR. ROSENBAUM: 7 Q. Good morning, Mr. Balagtas. 8 A. Good morning. 9 Q. You have before you two exhibits. 10 Is Hearing Exhibit 435 your written testimony 11 today? 12 A. Yes, it is. 13 Q. And is Hearing Exhibit 436 a PowerPoint 14 presentation that you have prepared to summarize and 15 orally discuss that written testimony? 16 A. Yes, it is. 17 Q. All right. I would ask then that we have the 18 PowerPoint presented on the screen, and let's go ahead and 19 go to the second page, if we could, Dr. Balagtas, where 20 you describe your background, and I will let you do that, 21 please. 22 A. Thank you. 23 I am a professor of agricultural economics at 24 Purdue University where I have been since -- on faculty 25 since 2004. I am also the interim director for the Center 26 for Food Demand Analysis and Sustainability, a position I 27 have held since August this year. 28 I have three degrees: A bachelor's of economics 10076 1 from Miami University; a master's of science in 2 agricultural economics from Iowa State University; and a 3 Ph.D. in agricultural economics from UC Davis. 4 In addition to my academic positions, I have also 5 taken temporary leave from Purdue twice, once to serve as 6 a Fulbright Senior Scholar at the International Rice 7 Research Institute, and again to serve as a senior 8 economist at the Council of Economic Advisors in the 9 Executive Office of the President. 10 Q. And just what does the Council of Economic 11 Advisors do? 12 A. It provides objective economic analysis to the 13 President of the United States. 14 Q. Okay. Turn to the next page, please. 15 Tell us about that, please. 16 A. So in my professional work I conduct research. I 17 also teach in the areas of -- in -- on the economics of 18 agricultural and food markets, with a particular expertise 19 in industrial organization. Industrial organization is 20 the study of consumer behavior, competition, and public 21 policy. 22 I have published dozens of research papers, peer 23 reviewed general articles on U.S. and international 24 agricultural food markets, including U.S. dairy markets 25 and policy in Federal Milk Marketing Orders. 26 I have also garnered multiple awards for the 27 quality of my research, and have attracted more than a 28 million dollars in funding to conduct that research. 10077 1 I have served and continue to serve on editorial 2 boards for the top journals in my field, and relative -- 3 relevant to this work in my current focus, current work 4 with the Center for Food Demand Analysis and 5 Sustainability, I focus my work -- I focus on consumer 6 behavior in food markets. 7 Q. And what do you consider yourself an expert in, 8 Dr. Balagtas? 9 A. I'm an expert in consumer behavior, competition, 10 and public policy in agricultural and food markets. 11 MR. ROSENBAUM: Your Honor, I would ask that the 12 witness be declared an expert in those four subject matter 13 areas. 14 THE COURT: Tell me again? I want to make sure I 15 got it right. 16 THE WITNESS: I'm an expert in consumer behavior, 17 competition, and public policy in agricultural and food 18 markets. 19 THE COURT: So I got three, Mr. Rosenbaum. You 20 got four? 21 MR. ROSENBAUM: I think I'll go with the three, 22 Your Honor. 23 THE COURT: Is there any objection to my accepting 24 Dr. Balagtas as an expert in those three fields? 25 There is none. 26 Dr. Balagtas, I am so pleased that someone with 27 your experience is here. I do recognize you as an expert 28 in consumer behavior, competition, and public policy with 10078 1 regard to agricultural and food markets. Did I express 2 correctly your expertise? 3 THE WITNESS: Yes, Your Honor. 4 THE COURT: Thank you so much. 5 THE WITNESS: Thank you. 6 BY MR. ROSENBAUM: 7 Q. If we could move on to the next page of the 8 PowerPoint presentation and tell us what you have done. 9 A. Yes. So I -- I have been hired by counsel for 10 IDFA to evaluate the market effects of Proposal 19. I'm 11 conducting this analysis as a private consultant, not as a 12 representative of my employer. 13 Q. Please continue. 14 A. So first let me state my understanding of 15 Proposal 19. It raises Class I differentials to an 16 average of $4.07 per hundredweight. That's an average 17 across counties. In the United States, it raises Class I 18 differentials in every county, ranging from $2.20 per 19 hundredweight in some counties in Idaho, to as much as 20 $7.90 in parts of Florida. 21 Those proposed increases in Class I differentials 22 average $1.50 -- excuse me -- average $1.50 increase over 23 current Class I differentials, ranging from $0.25 per 24 hundredweight increase to an increase of $2.70 per 25 hundredweight. 26 Relative to the 2023 average Class I price of 27 $19.20, that $1.50 increase represents a 7.8% increase in 28 Class I price. 10079 1 Q. Thank you, Dr. Balagtas. 2 And over the next three pages have you reproduced 3 maps that were actually put into evidence already by the 4 American Farm Bureau Federation that pictorially depict 5 these changes? 6 A. Correct. 7 Q. Can you just march through it briefly as to what 8 each page is. 9 A. So this is a map of current Class I differentials 10 that show low prices in -- in surplus markets. And 11 those -- excuse me -- low Class I differentials, and those 12 increase as we move to the South and Southeast. 13 Q. And this is page 3 of the printed version of your 14 PowerPoint presentation, page 6. 15 Can you go to the next page, please. 16 A. Here's Proposal 19's proposed Class I 17 differentials. A little hard to tell from the colors, but 18 the proposed differentials are higher. 19 If we go to the next map -- 20 Q. That was page 7. Now we go to page 8 of the 21 printed copy. 22 A. Here are, again, American Farm Bureau Federation's 23 representation of the increase in the Class I 24 differentials by county. 25 Q. And if we could go to the next page, tell us what 26 specific questions you have addressed. 27 A. So I -- I do three things. One, I -- I look back 28 at market conditions since 2000, the last time we had a 10080 1 large revision in Class I differentials, and ask whether 2 changes in market -- in the market since that time justify 3 higher Class I differentials today. 4 I then think about if -- what would happen in 5 markets if Class I higher -- if Proposal 19 were 6 implemented. I first look at the fluid milk market: What 7 would happen to fluid milk retail milk prices; what would 8 happen to retail milk consumption; what would happen to 9 Class I utilization; and what would happen to the economic 10 well-being of fluid milk consumers. 11 Third, I look at what the implications of those 12 effects in fluid milk markets are for the rest of dairy 13 markets and also for the overall average farm price of 14 milk, which I -- I take as sort of a representation as a 15 the effect on dairy farmers. 16 Q. All right. Could you take us to your next slide, 17 please. 18 A. So starting with that first question: Do changes 19 on dairy markets since 2000 justify increases in Class I 20 differentials? 21 And here, I think it's important to start with 22 some -- some stylite tests, some data that explain what's 23 happened since 2000. 24 Q. Just we're now looking at page 11 of the printed 25 copy of -- 26 A. And this chart shows U.S. milk production from -- 27 every year from 2000 through 2022. Growth in milk 28 production has averaged 1.8% annually over that time. 10081 1 Today, or in 2022, we had 45% more milk produced in the 2 country relative to 2000. So we have a lot more milk 3 produced in the country. 4 Q. Okay. Now we're on slide 12. 5 A. At the same time, fluid milk consumption is 6 falling. This is a chart from the Economic -- USDA's 7 Economic Research Service, showing average per capita 8 fluid milk consumption in the United States. It's dropped 9 by about half, from approximately one cup per person per 10 day in 1970 to half a cup per person per day in 2019, 11 which was the latest year they had data for when this was 12 published, I think in 2001, if I remember correctly. 13 So a big drop over that 50-year period, since over 14 the last ten years of this data, fluid milk consumption 15 dropped from .62 cups per person per day to .5 cups per -- 16 .5 cups per person per day. 17 Q. Now you've pulled up slide 13. 18 A. This chart uses Federal Milk Marketing Orders 19 statistics data to show quantities of milk -- of producer 20 milk in Class I use in 2001 and 2022. The big -- the 21 overarching story to tell here is that we are using an -- 22 less milk in Class I uses today than we were 20 years ago. 23 Over that time period, fluid milk use -- the quantity of 24 fluid milk -- excuse me -- the quantity of farm milk in 25 fluid uses has fallen by 11%, 10.7%. 26 Q. All right. Take us to the next slide, 14, please. 27 A. Yep. Worth noting, it's fallen also in every one 28 of these nine regions that I show here. 10082 1 Q. That's on slide 13? 2 A. Correct. 3 Q. Yes. Okay. And now slide 14, please. 4 A. This -- this table shows the percentage of pooled 5 producer milk used in Class I for those same nine regions. 6 That Class I utilization rate has fallen by 30% over this 7 20-year period, from 38% in 2001 to 27% currently. Across 8 regions, across marketing order areas, it's fallen in six 9 of these nine areas. Three marketing areas, the 10 Appalachian, the Central and Southeast Orders, have shown 11 increases in Class I utilization rates. 12 Q. All right. Page -- the next slide, please. 13 A. So I got ahead of myself there. So in six of the 14 nine milk marketing order regions shown in those tables, 15 Class I utilization rates have fallen since 2001. 16 I take that as evidence that in these regions 17 there is adequate supply of milk for fluid uses. 18 Presuming that in 2000 when we last made large revisions, 19 we made those revisions to ensure adequate supply of milk 20 for fluid uses. Now we have more milk, less fluid 21 consumption in those regions. So I take that to mean that 22 there's -- there's plenty of milk for fluid uses there. 23 In three of those nine regions, Class I 24 utilization rates have risen and -- since 2001, and that 25 suggests potential for market -- that marketing conditions 26 in -- have changed in the -- in a way that supply is 27 inadequate for fluid uses. So I looked at additional data 28 to evaluate this question for those three regions. 10083 1 Q. Okay. Are we now on slide 16? 2 A. Yes. So first I look at the annual utilization 3 rates could mask over some inter-annual variation. In 4 particular, in the fall, milk production's down, kids are 5 back in school, utilization rates are high. And so what I 6 do here -- and so Class I utilization rates are 7 particularly high in -- in those months. 8 Here I take for each year the peak monthly Class I 9 utilization rate for each of these years, and I look at 10 what's happened to peak monthly Class I utilization rates 11 over time. And if they were trending higher, I -- I would 12 worry that there there's a growing problem of inadequate 13 supply of milk in those regions. And the fact is there's 14 no trend in this data, and the peak Class I utilization 15 rates in these three regions has been fairly steady over 16 time. 17 Next I look at prices as an economist. We -- 18 prices would -- would -- we look at prices to indicate 19 inadequate supply. So there's -- lack of supply would -- 20 would show high prices, would manifest in high prices. So 21 I take Federal Milk Marketing Order data. They collect 22 data on 30 cities and report retail milk prices in 30 23 cities in Federal Milk Marketing Order areas. 24 And I look at retail prices in cities in those 25 three regions, the three high Class I utilization rate 26 regions: Atlanta, Georgia in the Southeast; Louisville in 27 Appalachian Milk Marketing Order; and Miami, Florida, in 28 Florida. And I compare the retail -- average annual 10084 1 retail prices in those -- of milk in those three cities to 2 prices in the rest of the country. 3 If you look at Atlanta, you have Class I -- or 4 excuse me -- retail prices in 2019 that's $3.56 per 5 gallon, slightly -- or higher than the 30-city average, 6 lower than the 75th percentile price in that sample. 7 For the remaining for 2020, 2021, 2022 -- and 8 2022, Atlanta's retail milk price is below the 30-city 9 average. It's higher, again, than the city average in 10 2023, but lower than the 75th percentile. 11 In -- in Louisville, retail prices of milk are 12 below the 30-city average in all five years, these five 13 recent years. 14 In Miami, Florida, we have retail prices of milk 15 that are above the 75th percentile in 2019, and below the 16 75th percentile for the remaining -- for the most recent 17 four years for which we have data. 18 This suggests to me that high Class I utilization 19 rates in these regions are not causing high retail prices 20 of milk in these regions, which suggests to me that 21 there's not a problem of inadequate supply of milk in the 22 Southeast, in Appalachian, in Florida. 23 Q. Now we're on slide 18. 24 A. Let me summarize: Higher Class I differentials 25 are not justified in my view on the basis of inadequate 26 supply of milk for fluid uses. There's growing milk 27 production nationally and in most regions. There's 28 declining milk consumption everywhere, that means less 10085 1 Class I milk, and in most regions Class I -- lower Class I 2 utilization. 3 There is -- there are -- this is a typo. It 4 shouldn't say rising Class I utilization rates -- well, 5 yeah. No, that's correct. There are rising and 6 relatively high Class I utilization rates in Appalachia, 7 Central, and Southeast Marketing Orders, but utilization 8 rates have -- have peaked. Monthly utilization rates have 9 not risen since 2000, and higher utilization rates have 10 not caused high retail milk prices. 11 Q. For that first bullet, should we insert the word 12 words "peak"? 13 A. The first sub bullet on the third big bullet, 14 where it says "utilization rates have not risen," that is 15 a typo or an error of omission. It should say "peak 16 monthly utilization rates." 17 THE COURT: Hold on just a minute. I'd like to 18 make that change on the record copy. I want you to, 19 again, say what you did, and I'm going to ask the 20 Agricultural Marketing Service to note this. 21 We are in Exhibit 436 on page 18, and the third 22 bullet point has two subcategories. We're going to the 23 first subcategory, which is also a bullet point, and I'd 24 like, Doctor, for you to tell us how that should be 25 changed on the record copy. 26 THE WITNESS: Yes. The bullet point that 27 currently reads, "Utilization rates have not risen since 28 2000," should instead read, "Peak monthly utilization 10086 1 rates have not risen since 2000." 2 THE COURT: Done. Thank you. 3 THE WITNESS: You're welcome. 4 BY MR. ROSENBAUM: 5 Q. If you could please continue with your next slide 6 and your presentation. This would be slide 19. 7 A. So next I turn to the effect of that higher 8 Class I differentials, Proposal 19, would have in the 9 market for fluid milk. 10 Demand -- higher Class I prices would cause higher 11 retail prices for fluid milk products. Consumers respond 12 to higher prices by reducing consumption, an effect which 13 we quantify -- economists quantify by the own-price 14 elasticity of demand for milk, which I define here as the 15 percentage change in quantity consumed for a given 16 percentage change in the price. 17 Proposal 19 would contribute to declining milk 18 consumption, and the magnitude of this effect depends on 19 this elasticity of demand for milk. 20 Q. All right. Next page, please, page 20. 21 A. So this elasticity of demand for milk, we have 22 talked a lot about it in the couple of days that I have 23 been here, is -- is a crucial parameter. And we want to 24 know how consumers will respond to higher retail milk 25 prices. And not just any consumers, we want to know how 26 consumers will respond -- how the consumers who live 27 through the higher milk prices will respond to that. So 28 if that happens in 2025, how will consumers in 2025 10087 1 respond to higher milk prices? 2 We don't know exactly what that will look like, 3 but we could draw estimates using data, historical data, 4 all this historical data, to think about what that 5 response would be. In fact, there's large literature in 6 my field that tries to estimate agricul- -- demand for 7 fluid milk, dating back some 60 years. 8 The typical finding in that literature is that 9 demand is inelastic. Inelastic means an elasticity less 10 than 1.0 in absolute value, between zero and minus one, 11 meaning consumers reduce consumption less than 12 proportional -- less than proportionately in response to 13 higher prices. 14 Q. All right. 15 A. That implies Federal Milk Marketing Orders 16 increase farm revenue of milk. If demand for Class I milk 17 is inelastic, a higher price causes an increase in 18 revenue. 19 As I said, we need an elasticity of demand that 20 captures behavior of consumers in current and actually in 21 future markets. Studies using data from the middle of the 22 20th century or late 20th century likely do not capture 23 relevant behavior or market conditions. Data from the 24 1960s or the 1980s don't reflect the choices that 25 consumers face in the marketplace. They don't reflect 26 how -- where or how consumers shop. And so I would be 27 hesitant to use that data to reflect -- to estimate a 28 parameter that we want to reflect consumer behavior now or 10088 1 in the near future. 2 And in particular, there's been growth in 3 competition in the retail space for fluid milk. In 4 particular, in the last several years, call it eight or 5 ten years, growth of non-dairy substitutes. 6 Mintel Group estimated that non-dairy milk 7 accounted for 17% of all quote/unquote "milk sales" in 8 2022; now is up 67% from 2017. So it's a large percent -- 9 large and growing percentage of the quote/unquote "dairy 10 aisle" is non-dairy milk. 11 Son and Lusk using Nielsen data found non-dairy 12 share of milk expenditure a little bit lower. That's 13% 13 in 2002 -- 2022, excuse me. 14 Q. Next slide, page 22. 15 A. The growing availability of substitutes makes 16 demand more elastic. The main driver of consumer response 17 to higher prices is the sub- -- is what economists call 18 the substitution effect. When there are close substitutes 19 for a good, consumers respond to higher prices of that 20 good by switching to the close substitute. 21 Thus, when the presence of more sub- -- when 22 there's -- when there are more substitutes in the market, 23 there is greater consumer response to higher prices, in 24 other words, more elastic demand. Growing competition 25 within the dairy aisle and across the beverage category 26 means the demand for milk is likely more elastic today 27 than it was even ten years ago. 28 Demand studies -- because of that, demand studies 10089 1 using data that do not capture these market realities are 2 not relevant for analyzing Proposal 19, which would be 3 implemented in current or near future market conditions. 4 THE COURT: Do you want us to add those words on 5 the record copy, "current or near future"? 6 THE WITNESS: I think "current" communicates the 7 idea appropriately. 8 THE COURT: Thank you. 9 BY MR. ROSENBAUM: 10 Q. All right. Next page, page 23. 11 A. So what do recent milk demand studies find? And 12 I'm a little sloppy here, I don't necessarily mean 13 recently published. All right? The timing of the 14 publication of the study isn't as important as the timing 15 of the data used in those studies. 16 So each of the studies used here use relatively 17 recent data, from dating back to 2015 or 2017 depending on 18 the study. So those data better reflect modern market 19 conditions, and so I picked these three as representative 20 of analysis using this type of -- this more recent data. 21 What do they find? Professor Judd Capps -- Oral 22 Capps, Jr., estimates using IRI or Circana data, a milk 23 demand elasticity of minus 1.26. That's his estimate for 24 all milk, all fluid milk in the most recent period in his 25 study, what he called the post or moving -- 26 moving-out-of-COVID period, I believe it was. 27 THE COURT: And although we have Dr. Capps' data 28 and exhibits, would you spell the sources of his 10090 1 information so that we'll have it correct here? IRI, you 2 said. 3 THE WITNESS: Capps uses data from Circana. It's 4 C-I-R-C-A-N-A. 5 THE COURT: All right. I thought you had 6 indicated two words that I wanted to make sure we 7 captured. 8 THE WITNESS: I think I made might have said IRI, 9 which was the previous name of Circana before it merged 10 with another company and changed names. 11 THE COURT: That was it. Thank you. 12 THE WITNESS: A study by Ghazaryan, et al., found 13 milk demand elasticity with values between minus 1.3 and 14 minus 1.7, using scanner data. And I don't recall off the 15 top of my head if it's Nielsen or Circana data. 16 Son and Lusk analyze using Nielsen data, demand 17 for fluid milk, in this context of -- with non-dairy 18 substitutes, and found milk demand elasticity of minus 19 0.946. 20 BY MR. ROSENBAUM: 21 Q. Next slide, slide 24. 22 A. So what does that mean, compared to the previous 23 literature, and Professor Kaiser found a median elasticity 24 of minus 0.196, compared to that previous literature, 25 recent work suggests demand in current markets is more 26 elastic. 27 So that means Proposal 19 would be implemented in 28 a market where milk consumers will be more responsive, and 10091 1 the proposal would reduce milk consumption by more than 2 what is suggested by previous estimates. Proposal 19 3 would make a bigger contribution to declining milk 4 consumption than is suggested by previous estimates, and 5 Proposal 19 has bigger implications for manufacturing 6 class milk than is proposed -- than is suggested by 7 previous estimates of the fluid milk demand elasticity. 8 And that's a subject I'll turn to next in a minute. 9 Q. All right. Slide 25 then, please. 10 A. So using recent elasticity estimates, I look at 11 the effect of Proposal 19 on fluid milk consumption. As I 12 stated in one of my earlier slides, I calculate that 13 Proposal 19 raises Class I prices by 8.7%, using a price 14 transmission -- transmission elasticity of 0.55 from 15 Professor Kaiser. That 8.7% increase in Class I prices 16 translates to a 4.3% increase in retail milk prices. 17 Applying Professor Capps, Jr.'s demand elasticity 18 of minus 1.6 -- 1.26, excuse me, that 4.3% increase in 19 retail price translates to a 5.4% reduction in consumption 20 of fluid products. 21 THE COURT: And those fluid products you are 22 referencing in this third bullet point are fluid milk 23 products? 24 THE WITNESS: Correct. 25 THE COURT: Okay. 26 MR. ROSENBAUM: Okay. 27 BY MR. ROSENBAUM: 28 Q. Next slide, please. 10092 1 A. Those higher prices harm fluid milk consumers. By 2 causing higher retail prices, Proposal 19 makes consumers 3 worse off. As a measure of that cost to consumers, I use 4 the change in consumer surplus, which is approximately 5 equal to the change in consumer expenditure on milk. 6 Q. Is that an economic concept, consumer surplus? 7 A. Yes, it is a standard economic concept used to 8 measure consumer well-being in markets. Yes. So the idea 9 is if I have to pay a higher price for any given item, the 10 additional price I pay is money out of my pocket. That's 11 a cost to me. 12 If I use Capps, Jr.'s Circana data, which showed a 13 price of $4.95 per gallon and 56.9 million gallons per 14 week purchased by consumers, harm to consumers would be 15 $11.8 million per week. That's just taking that retail 16 price increase that we calculated, applying it to the 17 consumer's expenditure in his data. 18 If I further assume that that price increase and 19 Capps' data, applied to the 12% of untracked retail 20 purchases, that is retail purchases not tracked by 21 Circana, then harm to consumers is $14 million per week. 22 If I further assume that Professor Capps' data and 23 -- applies to the 24% of milk volume that is sold in 24 foodservice, I get a harm to consumers, a loss in consumer 25 surplus of $18.4 million per week. 26 Q. Dr. Balagtas, looking at this now, should we be 27 adding the word "and schools" after "food service" to get 28 up to the 24%? 10093 1 A. I'm including schools and foodservice, yes. 2 Q. Thank you. 3 THE COURT: Then let's do -- add the "schools" 4 just to make it perfectly clear. This is Exhibit 436, 5 page 26. How would you like it to read? Would you like 6 it to say after foodservice, parentheses, "including 7 schools"? 8 THE WITNESS: That would be accurate. 9 THE COURT: All right. Let's do that. So we're 10 just inserting "including schools" in that last bullet 11 point. That is closing the parentheses just before the 12 24%. 13 All right. Done. Thank you. 14 THE WITNESS: Your Honor, can I ask a question? I 15 didn't start my timer. Can you give me a sense of time? 16 MS. TAYLOR: 22 minutes -- 17 THE WITNESS: 22 minutes? 18 MS. TAYLOR: -- you have left. 19 BY MR. ROSENBAUM: 20 Q. We're doing fine, Dr. Balagtas. Take us on to the 21 next slide, please. 22 A. So in addition to -- where as a consequence of 23 reducing Class I milk consumption, Proposal 19 will 24 redirect farmers' milk, producer milk to manufacturing. 25 By -- that increased supply of milk to manufacturing 26 uses -- excuse me -- results in -- there's a typo there -- 27 increased production of manufactured dairy products, 28 reduced prices for those dairy commodities, and lower 10094 1 prices of milk components because milk components on -- 2 are priced on dairy commodity prices. 3 And next I turn to quantify these effects. 4 THE COURT: Before you go there, on page 27 of 5 Exhibit 436, do we just strike the word "uses"? 6 THE WITNESS: Yes, Your Honor. 7 THE COURT: All right. Do you see it there? Just 8 after manufacturing, we strike the word "uses." Thank 9 you. 10 THE WITNESS: And I want to clarify here what I'm 11 trying to do. I am trying to quantify what it means to 12 divert this quantity of milk to manufacturing uses, and 13 that exercise depends on some elasticities. Again, we're 14 going to get back to elasticities, different ones this 15 time, but there's uncertainty in my mind about what the 16 value of those elasticities are. 17 So I -- what I tried to do is provide a range of 18 plausible elasticities. I think that range is wide 19 because, to my knowledge, those -- there are not publicly 20 available published studies that -- that report such 21 elasticities. So I tried to be fair and to give you a 22 sense of the range of possible effects in the manufactured 23 dairy product market. 24 BY MR. ROSENBAUM: 25 Q. And just to orient ourselves before you get into 26 the numbers. You, in your analysis, suggested that an 27 8.7% increase in Class I prices would translate to 5.4% 28 reduction in consumption of flood products, correct? 10095 1 A. Correct. 2 Q. And obviously then farmers are not being paid a 3 Class I price on that milk that's no longer being sold to 4 Class I, correct? 5 A. Correct. That's milk that gets diverted to, for 6 example, butter and powder production and gets priced in 7 Class IV. 8 Q. So one way of conceptualizing it for a non-expert 9 like me is you start out with 8.7% more money through the 10 Class I increase, but you lose 5.4 percentage points, 11 roughly, through the decline in sales? 12 A. So you get a lower price on -- on that milk that's 13 diverted from Class I to Class IV. In addition, because 14 we produce -- because we increase production of, for 15 example, butter and powder, lower prices of butter -- of 16 butter and nonfat dry milk cause the whole structure of 17 Federal Milk Marketing Order prices to also decline. 18 Q. So please continue. 19 A. So we start with this 5.4% reduction in Class I 20 milk, in 2022, using Federal Milk Marketing Order 21 statistics data. Class I -- producer milk used in Class I 22 was 41 billion pounds, so 5.4% of that is 2.2 billion 23 pounds. 24 Where will that milk get absorbed? For the 25 purposes of this exercise, I assume all of that 26 2.2 billion pounds goes into Class IV. 2.2 billion pounds 27 translates to -- of -- of farmer milk, translates to 28 200 million pounds of nonfat dry milk production. That's 10096 1 a 7.6% increase, again, using 2022 data. And it -- that 2 2.2 billion pounds also would produce -- of milk would 3 produce 62.9 million pounds of butter. That's a 4 3.1% increase in U.S. butter production. 5 By the way, my Excel spreadsheet has all of this 6 done so that it's transparent, we can see how I calculate 7 these changes. 8 Q. Okay. And the hard copy of that is attached as 9 the last two pages of Exhibit 435; is that correct? 10 A. Exhibit 435. 11 Q. Yes. 12 A. Correct. 13 Q. Okay. 14 THE COURT: Now, on this bullet point, you said, 15 200 million pounds, and your slide says 201 million 16 pounds. 17 THE WITNESS: I'm being sloppy, Your Honor. It's 18 201 million pounds, additional pounds of nonfat dry milk, 19 correct. 20 THE COURT: Thank you. 21 THE WITNESS: So the question then is, what are 22 the effects of that increased production of nonfat dry 23 milk and butter on commodity prices? That depends on 24 demand elasticities, demand for U.S. nonfat dry milk, 25 demand for U.S. butter. Right? 26 In the absence of relevant demand elasticities, 27 estimates of those elasticities, I report effects for a 28 wide range of elasticity values. 10097 1 BY MR. ROSENBAUM: 2 Q. And if we go on to page 29. 3 A. I believe this is Table 5 in my -- in Exhibit 435. 4 And here what I do is take that increase in -- 5 those increased -- that increased production of butter and 6 powder, apply elasticities for -- demand elasticities for 7 butter and powder, and for those -- for each set of 8 elasticities, I calculate a change in the nonfat dry milk 9 price and the change in the butter price. And then I work 10 through the Federal Milk Marketing Order pricing formula 11 to get price -- price changes in the skim price and 12 butterfat price and therefore changes in the all-milk 13 price. 14 I look at three scenarios in this table. On the 15 left is a more inelastic scenario with a demand for -- for 16 U.S. nonfat dry milk an elasticity of demand for U.S. 17 nonfat dry milk of minus 4, and a demand -- an elasticity 18 of demand for U.S. butter of minus .25. 19 In all of these, the elasticity of demand for U.S. 20 nonfat dry milk is significantly more elastic than that 21 for butter. Because nonfat dry milk is traded, right, so 22 we're -- that -- that's going -- that production trades in 23 a global market. I think some large percent, 70-ish 24 percent, of nonfat -- U.S. nonfat dry milk is exported, 25 and so it's a smaller portion of world production, 26 demand's going to be more elastic. 27 THE COURT: "Demand is going to be more"? 28 THE WITNESS: More elastic for nonfat dry milk. 10098 1 So what we find is comparing across scenarios, in 2 the more inelastic scenarios, we find larger prices of 3 increased production of dairy commodities, larger drops in 4 those dairy commodity prices, a one -- a minus 1.9% 5 reduction in the nonfat dry milk price -- excuse me -- a 6 1.9% reduction in the nonfat dry milk price and a 12.23% 7 reduction in the butter price. 8 Those translate, again, through the Federal Milk 9 Marketing Order pricing formulas to a $0.20 per pound 10 reduction in the skim price and a $0.385 per pound 11 reduction in the butterfat price. Again, all of this done 12 in my Excel spreadsheet. 13 Those reductions in skim and butterfat result in a 14 $0.28 reduction in the all-milk price. That $0.28 15 reduction in the all-milk price includes increased Class I 16 revenue from higher Class I differentials. That is to 17 say, in this inelastic -- more inelastic scenario, the net 18 effect of the Proposal 19 on dairy farmers would be to 19 reduce the all-milk price, make farmers worse off. 20 When we move to the more elastic scenario, 21 increased production of dairy commodities has a smaller 22 effect on commodity prices, therefore, a smaller effect on 23 component prices, and a smaller effect on the all-milk 24 price. In that case, this is the far right column, in 25 that more elastic world, Proposal 19 would raise the 26 all-milk price by $0.12 per hundredweight. And then the 27 middle column shows a middle scenario. 28 I present this range of scenarios to show -- to 10099 1 reflect the uncertainty of the effect of this on -- of the 2 net effect of this proposal on farmers. We have increased 3 Class I revenue. We have decreased manufacturing milk 4 revenue. And the sign of that sum is maybe positive or 5 negative depending on elasticities of demand that I've 6 identified in my -- in my analysis. 7 Moreover, what I think are modest effects, 8 relatively small effects of -- on the all-milk price, 9 right, in my middle -- backing up one slide -- in my 10 middle scenario, we have a $0.03 per hundredweight change 11 in the all-milk price. So let's call that a modest effect 12 on the all-milk price, although it is real money for 13 farmers, I understand. That modest effect masks big 14 changes within the sector. There's big harm to fluid milk 15 consumers, as I calculated, and there's a disruption to 16 the manufacturing milk market. The relative size of 17 which -- or the implications of which, again, depends on 18 these demand elasticities. 19 And with that, I end my presentation. 20 MR. ROSENBAUM: Your Honor, the witness is 21 available for cross-examination. 22 MS. TAYLOR: Your Honor, I hate to interrupt, but 23 might I suggest a morning break, and then we might be able 24 to continue to lunch and not interrupt anyone's cross? 25 THE COURT: I think that's a great idea. Is that 26 okay? And it will also allow the witness to move about, 27 maybe get a little sunshine. I'm sure, yes. There's a 28 lot of work still to be done. 10100 1 Please be back and ready to go at 10:45. We go 2 off record at 10:29. 3 (Whereupon, a break was taken.) 4 THE COURT: Let's go back on record. 5 We're back on record at 10:47. 6 MS. HANCOCK: Thank you, Your Honor. 7 CROSS-EXAMINATION 8 BY MS. HANCOCK: 9 Q. Nicole Hancock for the National Milk Producers 10 Foundation. 11 Good morning, Dr. Balagtas. Thank you for being 12 here today. 13 A. Good morning. 14 Q. It is not a long commute for you, is it? 15 A. No, I'm local. 16 Q. You don't get to hear that much here. 17 I'm wondering if you can start off, I'm just going 18 to maybe go in chronological order in your exhibits, 436. 19 You started off by talking about your background. 20 I'm wondering if you can share a little bit more detail 21 about your background in the dairy industry in particular. 22 A. So I haven't been in the dairy industry. I have 23 been in academia all my life. I have done research on 24 dairy, starting with my Ph.D. dissertation was on Federal 25 Milk Marketing Orders. 26 During that time, while I was a grad student at 27 Davis, I wrote a couple other papers, just off the top of 28 my head remembering, related to -- one related to dairy 10101 1 trade, another related to the New England Compact. So 2 I -- I had written a few papers, academic papers on the 3 economics of dairy markets. At Purdue, I continued to do 4 some work in dairy. I have written several papers in -- 5 on the subject. And continue to have -- don't currently 6 have any active work that -- I think that's correct -- 7 related to dairy other than this, but continue an interest 8 and follow the research and literature on the topic. 9 Q. Prior to your engagement by IDFA -- and that began 10 in August of this year? 11 A. Correct. 12 Q. Prior to your engagement by IDFA to provide your 13 testimony today, when is the last time that you wrote 14 about the dairy industry or did any kind of analysis of 15 the dairy industry? 16 A. I don't recall off the top of my head. I'm not 17 sure. I'd have to look at my list of papers and to think 18 about that. 19 Q. Has it been some time? 20 A. That I have published on the dairy industry? I 21 believe so. Again, I'd have to look at my CV. 22 Q. And have you been involved in conducting any 23 studies yourself with respect to the dairy industry? 24 A. No, not recently. 25 Q. When is the last time? 26 A. I want to say maybe ten years ago. I have done 27 some work, but it has been a few years. 28 Q. What was the study that you did? 10102 1 A. Again, I don't recall off the top of my head. If 2 I could pull up a CV, I -- or we could look at papers I 3 have written. 4 Q. Okay. 5 THE COURT: Do you have it in your laptop? 6 THE WITNESS: Or I could pull up Google Scholar. 7 THE COURT: Would you like that information, 8 Ms. Hancock? We can do that. 9 MS. HANCOCK: Sure. 10 THE COURT: We learned about a new thing, Google 11 Scholar. 12 THE WITNESS: So if I sort these by -- if I can 13 sort these by date. 14 The last dairy-related work -- I was thinking 15 fluid milk, but I did have some work on ice cream package 16 sizes in the retail space: One published 2021, another 17 published 2014. 18 BY MS. HANCOCK: 19 Q. What was the 2014 article? 20 A. "Consumer Response to Packaged Downsizing: 21 Evidence From the Chicago Ice Cream Market." 22 Q. So related to your ice cream packaging? 23 A. Correct. 24 Q. What about Federal Orders in particular, ever do 25 any studies specific to the Federal Order system? 26 A. Yes. I think the most recent work that involved 27 Federal Orders would have been published in 2012. 28 Q. 2012? 10103 1 THE COURT: 2012? 2 THE WITNESS: 2012, Your Honor. 3 THE COURT: Oh, 2012. Thank you both. 4 BY MS. HANCOCK: 5 Q. What was the subject of that article, do you 6 recall? 7 A. That was Competition and Market Power in Fluid 8 Milk -- U.S. Fluid Milk Supply Chains. 9 Q. And I think we have a copy to look at. I think we 10 have handed that out and e-mailed that to everyone. 11 MS. HANCOCK: Your Honor, if we could have the 12 next exhibit number. 13 THE COURT: Thank you. All right. I have been 14 handed, Ms. Hancock, my copy of Exhibit 437, also shown as 15 Exhibit NMPF-110. 16 (Thereafter, Exhibit Number 437 was marked 17 for identification.) 18 MS. HANCOCK: Your Honor, I missed the NMPF 19 number. 20 THE COURT: It's 110. 21 MS. HANCOCK: You are providing it to me. It 22 should be the other way around. My copy didn't have it, 23 so thank you for that. 24 THE COURT: I have a black and white 110, freshly 25 minted. 26 BY MS. HANCOCK: 27 Q. Dr. Balagtas, is Exhibit 437 a copy of the last 28 article that you ever -- that you conducted a study on 10104 1 related to the Federal Order system? 2 A. I believe that's correct. 3 Q. And this was done in 2012? 4 A. It was published in 2012, correct. 5 Q. And since this time you haven't done any studies 6 related to the Federal Order system? 7 A. I have not published anything since this time that 8 I could recall on Federal Orders or related -- or work 9 related to Federal Orders. 10 Q. Okay. Since this time have you conducted any 11 studies that haven't been published related to the Federal 12 Order system? 13 A. That, I'm not sure of. There -- the way I -- we, 14 in academia, do research, have lots of projects going on. 15 But not that I recall. 16 Q. Okay. Fair to say that as you sit here today not 17 that you can think of? 18 A. Yes. Correct. 19 Q. And is it also fair to say that you have not 20 conducted any separate analysis or modeling related to the 21 Federal Order system? 22 A. Ever? Or since 2012, since this? 23 Q. Well, have you ever conducted any modeling of the 24 Federal Order pricing system? 25 A. Economic modeling of the Federal Order -- of Milk 26 Marketing Orders, I have, yes. 27 Q. When did you do that? 28 A. My -- there's a paper published in 2007 that 10105 1 models Milk Marketing Orders. The paper I referred to 2 with the Northeast Dairy Compact, we analyzed the compact 3 on top of Federal Order structure, which is in -- also in 4 the model. So it's not accurate to say I have never 5 modeled Federal Milk Marketing Orders. 6 Q. Okay. So since 2007, have you done any modeling 7 or analysis of the Federal Order system? 8 A. In the paper that you have handed me, published in 9 2012, analyzes milk markets, including Federal Milk 10 Marketing Orders. 11 Q. Okay. I was considering that to be a study, and 12 maybe my question was separate from that. I was asking 13 about in addition to the study, if you had done any actual 14 modeling work? 15 A. I'm not sure I understand the distinction between 16 a study and modeling work. 17 Q. Okay. I thought you were distinguishing it -- 18 A. Oh. 19 Q. -- so maybe that's why I was maybe confusing what 20 I -- what I was asking you. 21 So other than the 2007 modeling work that you did 22 and the analysis that you did in Exhibit 437, nothing else 23 that you can think of? 24 A. Not that I can think of. 25 Q. Okay. And in Exhibit 437, you were looking at the 26 competitive nature of cooperatives specifically in the 27 fluid milk market; is that right? 28 A. Correct. 10106 1 Q. And it's fair to say that you were somewhat 2 critical of what the cooperatives' ability in the -- to 3 sell products and the influence in selling fluid milk 4 products in the market; is that right? 5 A. No, I don't think I made a judgment of the 6 cooperatives in the paper. 7 Q. If you look at -- let's look at page 1 of 8 Exhibit 437. So this is the first page of the article. 9 And in that first column, you say, "One of the 10 stated goals" -- I'm about halfway through that -- or 11 maybe a little over halfway into that first paragraph. Do 12 you see the sentence that starts off with "one"? 13 A. I do. 14 Q. And you are talking about the 1937 Agricultural 15 Marketing Agreement Act. And you say, "One of the stated 16 goals of each of these policies is higher milk prices for 17 dairy farmers." 18 Do you see that? 19 A. I do. 20 Q. Is that still your understanding today, that 21 that's one of the stated goals for the Federal Order 22 system? 23 A. I think the wording is fair pricing and not higher 24 pricing, so that this -- the wording of marketing order 25 stated goals I think is fair pricing for dairy farmers, 26 which I interpret as higher prices for dairy farmers, yes. 27 Q. Okay. So that still accurately captures your 28 understanding of the goals of the Federal Order system 10107 1 today? 2 A. Yes. I don't think dairy farmers would want milk 3 Marketing Orders that lower their price. 4 Q. And the -- you wrote this some time prior to April 5 of 2012 when it was published? 6 A. Can you repeat the question, please? 7 Q. Sure. You wrote this article some time prior to 8 April of 2012 when it was published? 9 A. Correct. 10 Q. And you understand that price differentials that 11 we're talking about here at this national hearing have not 12 been adjusted since prior to the time that this article 13 was written? 14 A. I understand -- I'm sorry. Can you repeat? 15 Q. Yeah. You understand that the price differentials 16 that we're here talking about for the national -- at this 17 national hearing have not been adjusted or updated since 18 prior to the time that this article was written? 19 A. I understand that, yes. I think there was some 20 revisions, minor revisions in 2008. But, yeah. 21 Q. For a select region, right? 22 A. Correct. 23 Q. Not at the national level that we're talking about 24 here? 25 A. I -- I do understand that. 26 Q. And at some point when you were talking, you said, 27 we haven't updated -- we haven't updated this since 2000. 28 Do you recall that? You just -- 10108 1 A. I do recall that, yes. 2 Q. When you said, we haven't updated it, I made a 3 note here to ask you about whether you had any role in the 4 Federal Order Reform? 5 A. No. I meant we, the people. Yeah, I'm sorry. 6 The very grand we, not I. 7 Q. We have a lot of people who were involved, so I 8 just wanted to make sure we were clear on that. 9 A. I'm sorry. I did not have any role in that. 10 Q. Okay. Let's look at -- it says page 648, which 11 must be the page of the article, but it is the second page 12 of the article. 13 A. Yes. 14 Q. There's some industry background there I want to 15 ask about on that -- so I'm on the second column on the 16 right-hand side. The last sentence of the first paragraph 17 there says: "Marketing orders have three key effects." 18 And the first bullet point you have there is 19 "price discrimination." And you state: "Minimum 20 processor prices are set such that fluid milk plants pay a 21 higher price for farm milk than do other types of dairy 22 processors." 23 Do you see that? 24 A. Yes, I do. 25 Q. And is price discrimination your word as the 26 author? 27 A. I wrote those words, correct. And it -- yes. 28 Those are words that I wrote. 10109 1 Q. Do you believe that one of the effects of the 2 marketing order is that it creates price discrimination 3 for fluid milk? 4 A. I believe Class I differentials raise fluid milk 5 prices, Class I prices, relative to manufacturing uses, 6 which I -- which is the definition of price 7 discrimination. 8 Q. Do you understand that one of the purposes of the 9 Federal Order system is that dairy farmers get paid a 10 uniform price without regard to the end use of that milk? 11 A. I do understand that, correct. 12 Q. So when you say "price discrimination," are you 13 talking about it discriminates the price between and among 14 handlers? 15 A. No, it discriminates on end user. I do 16 understand, if you look at the next -- second bullet 17 below, "revenue pooling," that has to do with paying a 18 uniform price to farmers. Price discrimination based on 19 end use is a part of Class I differentials -- is a part of 20 Federal Milk Marketing Orders pricing. 21 Q. So when you say "end users," do you mean the 22 customers of -- ultimately the consumers of those 23 products? 24 A. Fluid milk plants versus, for example, a butter 25 powder plant. 26 Q. Okay. Meaning the different classes of -- pay 27 different prices? 28 A. Correct. 10110 1 Q. And that is what you believe is price 2 discrimination? 3 A. Yes. Price discrimination is a term used in the 4 literature -- the economic literature on Milk Marketing 5 Orders before this paper and continues to be I believe 6 used. 7 Q. Okay. Let's turn to page 657 of the article. 8 A. 6? 9 Q. 657. 10 A. Yeah. 11 Q. And this is under your conclusion. 12 A. Okay. 13 Q. And I want to go to the last sentence on the first 14 column on the left, under "Conclusion," and it starts off 15 with "a key finding." 16 A. I'm there. 17 Q. And you state that: "A key finding is that while 18 the estimated conduct parameter for dairy cooperatives is 19 small (e.g., .0027 for the Northeast region), the fact 20 that the derived demand for milk facing cooperatives is 21 very inelastic allows cooperatives to exact markups of 22 approximately 9%." 23 Do you see that? 24 A. I do. 25 Q. So in one of the conclusions that you reached is 26 that demand, at least at this time period, for fluid milk 27 is that it was very inelastic; is that correct? 28 A. Derived demand for milk facing cooperatives is 10111 1 very inelastic, correct. 2 Q. Okay. Do you remember what the inelasticity 3 measure was at this time? 4 A. I don't. We can look. I think it ranged across 5 the marketing order regions and was inelastic for all of 6 those. But it is inelastic, so between zero and minus one 7 for all of those. 8 Q. Minus 1.0? 9 A. Between minus 1.0 and zero. 10 Q. Okay. And -- and then you go on to say, "The 11 resulting estimate of annual income transfer from milk 12 buyers to dairy farmers, in the regions subject to Federal 13 Milk Marketing Order regulations, is approximately 14 $636 million, with a mass away from zero. Retail demand 15 for fluid milk is also quite inelastic." 16 Do you see that? 17 A. I do. 18 Q. Do you -- did you measure retail demand separately 19 from the derived demand? 20 A. I estimate -- we estimate those simultaneously in 21 this -- in this study. 22 Q. Was it the same number? 23 A. No. 24 Q. What was the difference between the derived demand 25 for fluid milk and the retail demand elasticities? 26 A. I don't recall I -- in general, derived demand -- 27 for derived demand for the upstream product, in this case 28 farmers' milk, would tend to be more inelastic than the 10112 1 retail demand. 2 Q. Okay. So -- 3 A. As a general rule. 4 Q. So the retail demand you -- you qualify by "quite 5 inelastic." 6 Can you give me what range you would conclude 7 would be quite inelastic? 8 A. Close to zero. 9 Q. Or something less than zero? 10 A. All of the elasticities are less than zero. 11 "Quite elastic" would be less than zero -- negative and 12 close to zero. 13 Q. Okay. 14 A. So negative .05, for example, I would say is quite 15 inelastic. 16 Q. Okay. So if we were looking at a scale, quite -- 17 the retail demand would be somewhere between zero and 18 negative .05, and the derived demand for fluid milk would 19 be somewhere between zero and negative -- or .1 -- or 20 negative 1, negative 1.0? 21 A. I'm sorry. Can you repeat that? 22 Q. Yeah. I just want to make sure we're clear on 23 this. 24 So retail demand elasticity that you concluded in 25 this article was somewhere between zero and negative .05? 26 A. Can we look at the results? 27 Q. Sure. Is that on 654? 28 A. 655. 10113 1 Q. Okay. What's the range that you have there? 2 A. So for the Northeast, I'm looking at the top -- 3 top number in the left column. That's retail demand of 4 minus 0.0445. 5 Q. Okay. And so you are looking at Table 3. And you 6 break it down by region; is that right? 7 A. Yep. 8 Q. And all of those regions show that retail prices 9 are inelastic? 10 A. Not retail prices. Retail demand. 11 Q. Okay. Retail demand is inelastic? 12 A. Correct. 13 Q. Okay. And after this article was written have you 14 done any further analysis since this time on the 15 elasticities? 16 A. Have I estimated demand elasticity for fluid milk 17 since this time? I have not. 18 Q. And the work that you did on behalf of IDFA, you 19 took the conclusions that were already in the record to 20 draw your conclusions for your presentation today; is that 21 right? 22 A. Can you repeat that question? 23 Q. Sure. The work that you did in Exhibit 436, your 24 presentation, your PowerPoint presentation that you 25 provided, you were analyzing the elasticity analysis that 26 was already performed by Dr. Capps for IDFA; is that 27 right? 28 A. I used -- yes. I used that estimate. 10114 1 Q. Okay. You didn't do any of your own independent 2 analysis or modeling work; is that right? 3 A. Correct. 4 Q. Okay. And to the extent that your conclusions are 5 based on his information, you would agree then that it 6 would also be true that your conclusions would be 7 different if you used Dr. Kaiser's elasticity results? 8 A. Correct. 9 Q. If we turn to -- I want to look at your page 10 436 -- or Exhibit 436, page 3. 11 A. Page 3. 12 Q. You talked about some funding to conduct research. 13 Is this for your work at Purdue, the funding that 14 you received here? 15 A. Which bullet? 16 Q. It just says "greater than" -- 17 A. Funding for research, correct. Yes, that's work 18 con- -- funded research conducted at Purdue, yes. 19 Q. Okay. Is any of that research that you have been 20 funded at Purdue related to the dairy industry? 21 A. I don't recall. 22 Q. And -- 23 A. I have been funded to do work for the dairy 24 industry -- not for the dairy industry, research of dairy 25 markets. Some of that dates back to my time at the 26 University of California. I don't recall if it spilled 27 over into -- if it came with me to Purdue or if I had new 28 grants at -- at the time at Purdue. But if I did, it's 10115 1 been some time. 2 Q. Okay. Fair to say not as you recall as you sit 3 here today? 4 A. Yep. 5 Q. Okay. And the work that you are doing as an 6 expert witness today, that's not through Purdue, is it? 7 A. No, it is not. 8 Q. You have been hired separately by IDFA to perform 9 this -- or to provide this expert opinion? 10 A. I have. 11 Q. And how much have you been paid for that service? 12 A. I will be paid about approximately $37,000. 13 Q. Okay. I want to turn to page 12 of your 14 PowerPoint presentation, Exhibit 436. 15 And you're here on this page 12 talking -- or 16 showing a bar chart showing the decline in fluid milk 17 consumption; is that accurate? 18 A. It is accurate. 19 Q. And it's fair to say that fluid milk consumption 20 in the U.S. has been declining since 1970 according to 21 this bar chart? 22 A. Correct. 23 Q. And in any of your work that you have done, have 24 you attempted to identify the reasons why consumers are 25 consuming less fluid milk than they did previously? 26 A. I have not evaluated that question, no. 27 Q. Have you done anything to draw any conclusions 28 about whether that correlates to the prices of those fluid 10116 1 milk prices? 2 A. I have not done any work on that question, no. 3 Q. Have you looked at, even just recently -- for 4 example, do you know that last year milk prices hit -- 5 well, strike that. Let me say that again. 6 Are you aware that last year in 2022 that Class I 7 milk prices hit a peak? 8 A. I'm -- I'm aware that Class I prices were high 9 last year. I don't know if they were a peak or how you 10 define that. But, yes. 11 Q. Or a historical high maybe i a better way to say 12 it? 13 A. I understand Class I prices were high last year, 14 yes. 15 Q. Do you know what they were? 16 A. I don't know. I think it was 20 -- I don't know 17 off the top of my head. 18 Q. Do you know what -- if there was a corresponding 19 effect at the retail level, either in the prices set by 20 retailers or in consumer buying behavior, when those 21 prices were high? 22 A. I do not. But those high prices occurred not in 23 isolation. High Class I prices did not occur in 24 isolation, right? So really hard to say without careful 25 analysis what the effect of those high prices last year 26 were on consumption. 27 Q. So when you say those prices didn't -- high prices 28 didn't occur "in isolation," do you mean overall the 10117 1 inflation rates have gone up and the cost of lots of 2 products have increased? 3 A. I mean, the elasticity, yes. The elasticity of 4 demand that we're all talking about measures the effect of 5 a price change on consumption, assuming nothing else 6 happens. Right? That assumption that nothing else 7 happened last year would be a bad assumption. 8 So I -- I wouldn't want to attribute changes in 9 consumption in the last year to changes in high prices 10 without carefully considering changes in lots of other 11 things, including general inflation, higher prices of 12 other food products, changes in the macro economy, 13 et cetera. 14 Q. Because all of those other factors also play into 15 a consumer's buying decision about whether they are 16 willing to pay a higher price for any one given product; 17 is that accurate? 18 A. That's correct. I also have a correction. I -- I 19 did think about, have thought about, the role of prices on 20 fluid milk consumption recently. And I reviewed a study 21 by my colleagues at the Center for Food Demand Analysis 22 who surveyed 1200 consumers. And I cite this in my study. 23 And the study was about alternative milks. 24 And they found a large portion of the respondents 25 never had tried alternative milks. And they asked those 26 respondents, if prices were the same, so they were asking 27 about how -- how would you -- would you be more willing to 28 try those products, those alternative milks, if prices 10118 1 were the same, in other words, if dairy milk prices were 2 higher. 3 And I don't remember off the top of my head. It's 4 cited in my testimony. A large portion said correct. A 5 large portion of those tended to be younger generations. 6 So I think price does affect, influence consumers' 7 behavior. All these estimates that we have cited, 8 including those by Dr. Kaiser, suggests prices do, in 9 fact, affects consumption. 10 Q. Yeah, my question was -- was talking about -- 11 so let me just be clear. There's no doubt that prices at 12 the retail level are going to affect buying decisions; is 13 that fair? 14 A. I think we agree. Yeah. 15 Q. Okay. And what I'm talking about is whether you 16 have done any analysis or seen any analysis related to an 17 increase in Class I prices, and whether that translates 18 all the way through the supply chain to reach the consumer 19 in a way that will also result in a change in consumer 20 buying behavior? 21 A. I used Professor Kaiser's price transmission 22 elasticity of 0.55. That quantifies the effect of changes 23 in the Class I price on retail milk prices. I have not 24 estimated that elasticity myself. I have -- recently I 25 have thought about it. I have worked on that number, and 26 it's a plausible number. 27 So I do think Class I prices influence retail milk 28 prices. 10119 1 Q. Okay. So you are saying that you -- based on the 2 work that you have done, you have been able to validate 3 the price transfer number that Dr. Kaiser used in his 4 report that he provided to USDA? 5 A. I believe Kaiser's -- Professor Kaiser's 6 elasticity of 0.55, that's a price transmission 7 elasticity, is plausible. 8 Q. Okay. And have you ever seen a different number 9 or a different study that would be more plausible than 10 what Dr. Kaiser offered? 11 A. Not a dramatically different number, no. 12 Q. Okay. But you did not use his elasticity 13 findings; is that right? 14 A. I didn't -- I used his price transmission 15 elasticity. I did not use his demand elasticity. 16 Q. Okay. Fair distinction. 17 A. Yeah. 18 Q. And instead you used Dr. Capps's demand 19 elasticity? 20 A. Correct. 21 Q. Okay. If you turn to page 17 of your testimony. 22 You have there a higher Class I utilization and a 23 title that says "has not resulted in higher retail milk 24 prices." 25 And can you help me understand how you read these 26 numbers? Can you just put it into -- maybe one example 27 into a complete sentence so I know how you are using it? 28 A. So I have three -- here three cities of 30 for 10120 1 which I have data, data reported by AMS. Three cities 2 located in the region's marketing order areas with the 3 highest Class I utilization rates. 4 And if Class I utilization rates were driving high 5 retail milk prices, I'd expect these cities to also have 6 the highest retail milk prices. 7 Q. Okay. 8 A. But I don't find that. And so I -- I don't think 9 high Class I utilization rates are driving high -- are 10 driving high retail milk prices. 11 Q. So if we took in 2019, the 30-city average is 12 $3.25 a gallon; is that right? 13 A. Correct. 14 Q. And then the price in Atlanta, Georgia, is $3.56 a 15 gallon; is that right? 16 A. Correct. 17 Q. And so what would be the correlation between those 18 two numbers? 19 A. It says Atlanta -- the retail -- retail price in 20 Atlanta is higher than the 30-city average, and that 21 retail price in Atlanta, or the 25% of the sample, so 22 eight cities, seven or eight cities, have higher prices 23 than Atlanta, Georgia. 24 Q. Okay. 25 A. That's the 75th percentile. So it is not among 26 the highest prices -- highest priced cities in that year. 27 Q. And then in 2023, for example, its 30-city average 28 is $4.39? 10121 1 A. Yes. 2 Q. But Miami, Florida, is less than that at $4.21; is 3 that right? 4 A. Correct. 5 Q. Does that tell us that the retail prices are not a 6 match for farm prices? 7 A. No. It says Miami -- the retail price in Miami in 8 that year is lower than the 30-city average in that year. 9 Q. And if you look at these prices as compared to 10 what was the Class I prices at the time, do you know if 11 these are reflective of the movement that happened at the 12 Class I price? 13 A. I don't -- I don't know that over time -- I don't 14 know what the Class I prices were in these regions over 15 time. 16 Q. Do you know what the -- how -- at the retail 17 level, do you know what the difference is, or the delta, 18 on average, between conventional milk and milk substitutes 19 or alternative milk products? 20 A. I don't have that number off the top of my head, 21 no. 22 Q. Do you -- how about for organic, do you know what 23 the delta would be at the retail level between 24 convictional milk and organic products? 25 A. I don't have that number off the top of my head. 26 Q. You didn't analyze that for any part of your 27 conclusions? 28 A. No, I did not. 10122 1 Q. Could you turn to page 23 of your PowerPoint. 2 You talk about the most recent milk studies. And 3 I think that you clarified Dr. Capps, the one that you are 4 talking about here, used the IRI data or the Circana data; 5 is that right? 6 A. Correct. 7 Q. And that's weekly data specifically limited to the 8 retail markets; is that right? 9 A. Correct. 10 Q. And you said you didn't recall what Ghazaryan had 11 used for the source of his data? 12 A. Correct. 13 Q. Do you know if it was weekly data that he used? 14 A. I believe it was weekly data. 15 Q. And do you know that if it was limited to retail? 16 A. It was retail. All three of these studies are 17 retail. 18 Q. Okay. Well, you fast tracked that for me. 19 Because the next one was the Nielsen was the data 20 source; is that what you said? Is that correct? 21 A. Correct. 22 Q. And so you are aware that that one is a weekly 23 data weekly measurement of just retail? 24 A. Correct. 25 Q. And for the Son and Lusk, that would be an 26 inelastic finding; is that right? 27 A. Yes. 28 Q. And so even with all three of these using weekly 10123 1 retail data, they reached different conclusions as to the 2 elasticity -- to the demand elasticity for fluid milk; is 3 that right? 4 A. Correct. 5 Q. Okay. 6 A. All of these are more elastic than Professor 7 Kaiser's median and more elastic than most of the papers 8 in that -- in that literature review. 9 Q. All of these are substantially more elastic than 10 the conclusions that you reached in your article in 437; 11 is that right? 12 A. Correct. 13 Q. You can turn to page 26. 14 This is where you extrapolate the elasticities 15 concluded by Dr. Capps and apply the -- I forgot what it 16 is called -- the price elasticity? What did you call it? 17 A. The price elasticity of demand? 18 Q. No. 19 A. Price transmission elasticity? 20 Q. I was missing the transmission, thank you. 21 A. Yeah. Price transmission, uh-huh. 22 Q. Let me start again. 23 Page 26 is where you take Dr. Capps' demand 24 elasticity and apply Dr. Kaiser's price transmission 25 elasticity to start your calculations that you have done 26 in the rest of the report? 27 A. Correct. 28 Q. And you start by using the three areas that 10124 1 Dr. Capps had analyzed, the first one being the actual 2 Circana/IRI reported retail data; is that right? 3 A. Correct. 4 Q. And that's the -- that's the weekly reported 5 retail data? 6 A. Yes. 7 Q. And then the next bucket that you have there is 8 untracked retail. 9 Is that reported by IRI or just extrapolated from 10 the reported IRI? 11 A. Untracked means not in the IRI or Circana dataset. 12 Q. So you have taken the results that are actually in 13 the IRI dataset, extrapolated them into, and projected 14 them onto unreported data to get that second bullet point 15 calculation? 16 A. Correct. And it's -- really what matters here is 17 the price. The elasticity is almost irrelevant. 18 Q. Okay. 19 A. But, yeah. 20 Q. And then under the third bullet point here, this 21 is the foodservice and schools. 22 And I think Dr. Capps also said military and other 23 government contracts? 24 A. Correct. 25 Q. Okay. And in his report, he said that this area 26 would be inelastic. 27 Were you aware of that? 28 A. I'm aware that he speculated that this segment of 10125 1 the market had more inelastic demand. 2 Q. And I think -- well, do you agree with that? 3 A. I think that's possible. I think it's also 4 possible that it's more elastic. I don't know. 5 Q. Do you have any idea which way it goes? 6 A. I -- I don't. 7 Q. Do you have any reason to disagree with what 8 Dr. Capps had concluded, that it was an inelastic 9 category? 10 A. I think that's possible, but I don't know what the 11 number is. 12 Q. Do you think it is probable? 13 A. I wouldn't say I think it's probable. I think 14 it's a number that I'd like to know. Yeah, I don't know 15 what it is. 16 Q. You haven't done anything to find out? 17 A. I have not estimated it. One of the reasons why 18 we don't estimate these things is because the prices are 19 hard to get, the data is hard to get. 20 Q. I'm so sorry, I didn't hear the -- one reason you 21 don't? 22 A. As ag economists, we don't typically have access 23 to the foodservice sector, so we don't -- data on the 24 foodservice sector. We don't have prices. The reason why 25 we have all these estimates in the grocery store is 26 because we have access to scanner data, so we can -- those 27 are relatively easy to do. We don't have data on the 28 foodservice sector, so we -- we need to speculate about 10126 1 what that elasticity is. 2 Q. And you state here that -- that this bucket where 3 you don't have that data input would harm consumers in the 4 amount that you have quantified to be $18.4 million per 5 week; is that accurate? 6 A. Yes. 7 Q. How would that harm be caused? From the increase 8 in price differentials? 9 A. Higher Class I prices cause higher prices for milk 10 in restaurants, in schools, in the military. 11 Q. So when you say "harm to consumers," you are just 12 quantifying that this is the amount that the end product 13 would increase in price, and you are saying that that 14 would be the harm to the consumers? 15 A. I'm taking the price transmission elasticity of 16 .55 and saying, there is a -- I forget off the top of my 17 head -- that gives me a change in the retail price -- a 18 percentage change in the retail price of I think it's 7.8% 19 I believe I said -- no, it's 7.8% of the Class I price, 20 and 55% of that gives me the change in the retail price. 21 So to quantify how much more consumers are paying 22 at retail, I need to know what the price was to apply the 23 percentage change. 24 Q. Okay. So this is -- would it be accurate then, if 25 I -- if each time that you say "harm to consumers" in each 26 one of these three bullet points, if I change that to say 27 "an increase at the retail level price for consumers"? 28 A. Yeah. It's not just a price -- it is increased 10127 1 expenditure by consumer -- by fluid milk consumers. 2 Q. So in each one of these, if I took the word "harm" 3 and I said "increased expenditure to consumers," it would 4 mean the same thing? 5 A. Correct. 6 Q. So every time there is an inflationary event at 7 the retail level, is that an increased expense to 8 consumers? 9 A. If I -- yes. If -- if I face a higher price by X 10 for an item that I purchase, I am worse off by X dollars. 11 Q. Okay. And so, for example, if feed costs were to 12 go up for dairy farmers, is that an increased expenditure 13 for farmers consuming those feed products? 14 A. Yes. 15 Q. Not consuming, ingesting, but purchasing? 16 A. Correct. Correct. Yes. Agreed. 17 Q. And so you're using "harm" here just meaning it 18 costs consumers more money? 19 A. Correct. 20 Q. And you would agree with me that since the last 21 time price differentials were updated, that dairy farmers 22 have had a significant increase in their costs of 23 supplying that fluid milk to the market; is that right? 24 A. I have not seen the data. But, yeah, I believe 25 that's correct. And if that is correct, then that's harm 26 that's a cost to producers, yes. 27 Q. Yeah. That would be a harm to producers under 28 your use of the word "harm"; is that right? 10128 1 A. Yes. 2 Q. And that harm to producers has not been remediated 3 since their price differentials were last updated more 4 than 20 years ago; is that right? 5 A. Class I differentials? Is that what you are 6 asking about? 7 Q. The Class I price differentials. 8 A. Yeah, I'm not sure I see the connection between 9 Class I differentials and feed costs. But I do think if 10 producers face higher costs, that's a burden to them, yes. 11 Q. And by "burden," you have used that 12 interchangeably with the word -- 13 A. Burden, harm, increased expenditure. Correct. 14 Q. Okay. I want to turn to page 29. 15 And on page 29 you have a chart here that's 16 describing the effects of a 7.6% increase in nonfat dry 17 milk and a 3.1% increase in butter production under an 18 alternative demand elasticity scenario; is that right? 19 A. Correct. 20 Q. And so are you -- here you are talking about the 21 effects on other classes of milk with the price one [sic] 22 differential increase? 23 A. Here we're talking about displaced milk because of 24 higher Class I -- displaced Class I milk because of higher 25 fluid milk prices. And the effect of that displaced milk 26 in -- in manufacturing product and milk price -- in 27 manufacturing product markets, those dairy commodity 28 prices, and farm milk prices. 10129 1 Q. So if it was an inelastic demand, you have 2 concluded that it would have a negative $0.28 3 hundredweight change in the all-milk price? 4 A. If -- yeah, I should be careful. I call that a 5 "more inelastic" scenario. It actually has elastic demand 6 for nonfat dry milk and inelastic demand for butter. But 7 across the scenarios, that's the more inelastic scenario, 8 yes, and that causes -- that would result in a drop in 9 manufacturing milk revenue that more than offsets the 10 additional Class I revenue from the higher differentials. 11 Q. And so then when you calculate that, you -- you 12 have concluded that there's a net change in the all-milk 13 price in that scenario, that would be a negative $0.28 per 14 hundredweight; is that correct? 15 A. Correct. 16 Q. And did you factor in here an increase in 17 Make Allowance? 18 A. No. 19 Q. You believe that National Milk's Proposal 19 20 asking for an increase in differentials will end up 21 resulting in retail prices being too high, and that would 22 decrease consumer demand; is that accurate? 23 A. I didn't say too high. I do think Class -- I do 24 conclude higher Class I differentials as proposed by 25 Proposal 19 would cause higher retail milk prices and 26 reduced milk consumption. 27 Q. Have you done any analysis to determine if Class I 28 price differentials could be set too low? 10130 1 A. I don't know what "too low" means. I don't know 2 what "too high" means. Those aren't terms that I have 3 used. 4 Q. Okay. Have you done any analysis to determine 5 if -- what the effects of the marketplace would be if 6 Class I differentials were not set high enough to allow 7 dairy farmers to stay in business? 8 Let me say it differently. Have you done any 9 analysis on the effects of a dairy farmer's ability to 10 supply the Class I market if differentials are not 11 increased? 12 A. So in my analysis, I assume milk supply is 13 perfectly inelastic. 14 THE COURT: "Is perfectly"? 15 THE WITNESS: Perfectly inelastic. So I don't 16 consider supply response here. Supply response by 17 dairy -- to farm milk price could change my results 18 somewhat, but it's -- I don't have supply response in 19 this -- in this analysis. Others, including myself, have 20 modeled supply as typically inelastic, but I don't have it 21 in here. 22 BY MS. HANCOCK: 23 Q. And is your assumption for your analysis that 24 demand is perfectly inelastic based solely on what you 25 described as Dr. Capps' elasticity analysis? 26 A. I'm sorry, can you repeat the question? 27 Q. Yeah. Is your demand analysis assumption, is it 28 premised on Dr. Capps' elasticity analysis? 10131 1 A. Yes. 2 Q. Okay. 3 MS. HANCOCK: I have no further questions. Thank 4 you so much. 5 THE WITNESS: Thank you. 6 MR. SIMS: Jeffrey Sims, Lone Star Milk Producers. 7 CROSS-EXAMINATION 8 BY MR. SIMS: 9 Q. We're going to go through this real slow because, 10 well, I'm slow. We're going to start with -- and I have 11 got a couple of math questions. 12 A. Sure. 13 Q. I'm going to start with Exhibit 435. That's the 14 prepared testimony or the prepared statement. And 15 although the page is not numbered, it would be I guess 16 page 21, the next to the last page where you start your 17 spreadsheet computations. 18 Are we there? 19 A. We are there. 20 Q. Okay. A straightforward question. What would be 21 Row 8, identified as in the Excel spreadsheet of Row 8, 22 Class I price, in Columns B, C, and D, you show $19.20 per 23 hundredweight? 24 A. Correct. 25 Q. I have looked at my statistics, and I see that 26 $19.20 is the simple average of what we colloquially refer 27 to as the Class I mover for 2023. 28 Is that where that $19.20 comes from? 10132 1 A. No. I believe it is the 2023 Class I price. 2 Q. Okay. So how -- how is the Class I price 3 computed? 4 A. Well -- so it's Class I mover plus a differential. 5 Q. Okay. So if I -- if we were to look at some 6 statistics that said that the average Class -- the average 7 Class I mover for 2023 -- and the December number has been 8 announced already -- averaged 19.20, that's the mover 9 without any differential, was 19.20, would you accept 10 that? 11 A. I would accept that. 12 Q. Okay. Great. We are making progress. 13 So now if you will just turn over to your 14 PowerPoint, that is Exhibit 436, and we're going to start 15 with page 5. 16 A. 436. 17 Q. In the first bullet there, I read that this 18 proposal -- you compute that Proposal 19 would raise 19 Class I differentials to an average -- and I presume 20 that's a national average? 21 A. Simple average across -- 22 Q. Simple average, great, across the country of 23 $4.07? 24 A. Yes. 25 Q. Per hundredweight? 26 THE COURT: Gentlemen? 27 MR. SIMS: Yes, Your Honor, I know where we're 28 going, and I'll be -- 10133 1 THE COURT: Too much caffeine, Mr. Sims. 2 MR. SIMS: And too little brain. 3 BY MR. SIMS: 4 Q. So we'll do a little math. 5 And we look at the second bullet point, and 6 said -- and this says -- and actually I agree with 7 these -- these numbers. I don't challenge your bullets 8 here -- that this would -- Proposal 19 would increase 9 those averages by $1.50 per hundredweight? 10 A. Yes. 11 Q. Okay. So if the new average Class I differential 12 under our proposal is $4.07 per hundredweight, and you say 13 that -- and I agree, that the differentials would increase 14 by an average of $1.50 per hundredweight, the current 15 average national Class I differential must be $2.57 or 16 thereabouts? 17 A. I'm sorry, I lost you. 18 Q. Okay. Okay. The new average differential you 19 state will be $4.07 per hundredweight? 20 A. Correct. 21 Q. And if that is an increase of $1.50 per 22 hundredweight -- and I would agree with that number too -- 23 the current average Class I differential nationally must 24 be about $2.57 per hundredweight, or 4.07 minus 1.50 25 equals 2.57? 26 A. I see what you are doing. Yes. 27 Q. Yes. Okay. 28 And I don't disagree with these numbers. In my 10134 1 head, I often use $2.60 as the kind of national average 2 differential. So we're not -- we don't have a problem 3 there. 4 But let's move to page 25. 5 Okay. So we have -- you have -- we have taken it 6 on faith that the $19.20 you used from Exhibit 435 on 7 Row 8, next to last page, is 19.20. That's the national 8 average Class I mover for 2023. Okay? 9 So actually the national Class I price unlike, 10 what you have got listed here, should be the sum of $19.20 11 plus the current average national Class I differential. 12 So actually, sir, the national average Class I 13 price from 2023 should be the sum of 19.20 per 14 hundredweight -- $19.20 per hundredweight, plus $2.57, the 15 average Class I differential for today. So an actual 16 average national Class I price of $21.77. 17 A. I'm following you. 18 Q. Okay. So if we are going to compute the 19 percentage increase in Class I prices resulting from 20 Proposal 19, we should actually be dividing the $1.50 21 increase not by 19.20, or $19.20, but rather by $21.77? 22 A. It's possible I looked at the wrong AMS Federal 23 Order Milk Marketing Orders statistics table and read 24 19.20 as the Class I price. I don't have those in front 25 of me. 26 Q. Okay. 27 A. But if -- if that's correct, then, yes, the 28 percent change in the retail price -- in the Class I price 10135 1 would be 6.8%, not 8.7. 2 Q. 6.8, 6.9, that's what I did on my little 3 calculator? 4 A. Yeah. 5 Q. Now, it seems to me that you used 7-point -- how 6 did you get 7.8%? 7 A. Excuse me. 8 Q. In Row 9 -- I'm sorry, I'm bouncing. 9 435, you -- there seems to be a conflict here. 10 THE COURT: 435 where? 11 THE WITNESS: So 7.8 is the correct number. I 12 inverted those on slide 25 of Exhibit 436. So that should 13 be 7.8 is what I meant to write. You are claiming that 14 should be in fact 6.9. 15 BY MR. SIMS: 16 Q. Yes. 17 A. I'm with you. 18 Q. Okay. So we're good there. 19 So basically -- 20 MR. SIMS: I'm sorry. 21 THE COURT: Mr. Sims, we're good where? 22 MR. SIMS: Oh, I'm sorry. 23 THE COURT: Where do I look? 24 MR. SIMS: Okay. Your Honor, we have agreed 25 that -- 26 THE COURT: What exhibit, what page? 27 MR. SIMS: Exhibit 435. 28 MR. HILL: 436, page 25. 10136 1 MR. SIMS: Okay. If we are looking at 2 Exhibit 435, the next to the last page -- it is an 3 unnumbered page, but it's the next to the last page -- 4 Row 8 and Row 9, we have agreed that the 1920 in those 5 three columns, B, C, and D, should in fact be $21.77 6 instead of, as listed, $19.20. 7 THE COURT: All right. And you are working on 8 your calculator to tell me how Row 9 would change? 9 MR. SIMS: Row 9 would change from -- correct me, 10 Dr. Balagtas -- the correct number should be actually -- 11 if the correct number in -- 9 is actually -- well, let's 12 see here, 6.9%. Excuse me. Are we about pretty close to 13 agreeing on 6.9% on Row 9? So 1.5 -- 14 THE WITNESS: If 21.77 is the Class I price, 15 correct. 16 BY MR. SIMS: 17 Q. So 1.5 divided by 21.77 yields about 6.9%? 18 A. Correct. 19 Q. I believe we have at least mathematically agreed 20 that the percent change in the Class I price would not be 21 7.8%, but rather would be 6.9%? 22 A. Correct. 23 Q. Okay. 24 THE COURT: Now, I'm not going to make any changes 25 on this until we have gone through redirect. 26 MR. SIM: Yes, ma'am. 27 THE COURT: Okay. 28 BY MR. SIMS: 10137 1 Q. And let's go through the same kind of exercise, 2 just take a real quick look at had you used the Class I 3 prices for 2023. As Ms. Hancock noted, the Class I price 4 was particularly high in 2020- -- excuse me -- 2022. 5 A. Okay. 6 Q. The simple average mover for the year, if my 7 statistics are correct, was 23.61, $23.61 per 8 hundredweight, an average of $2.57 annual national Class I 9 differential, for a total -- if I did my math right -- of 10 $26.18, national average Class I price. 11 So for -- if you had used 2022 instead of 2023, 12 you would have $1.50 per hundredweight divided by 26.18 is 13 about 5.7% increase. 14 So I guess my -- our point is that your 15 spreadsheet here, either way, whichever base year you use, 16 makes a presumption of an increase in the Federal Order 17 Class I price which is overstated, presuming I'm right 18 about how the -- 19 A. Correct. 20 Q. -- Class I prices are computed. 21 A. Correct. And the higher the Class I price, the 22 smaller -- $1.50 is a smaller share of a higher Class I 23 price, so that's correct. 24 Q. And so to the extent that the -- at least in these 25 two annual calculations that 7.8% is overstated, then the 26 impact of that -- even following the spreadsheet further 27 down, we -- certainly there's the possibility that the 28 impacts of this are overstated as it works through the 10138 1 spreadsheet? 2 A. Correct. 3 MR. SIMS: Thank you. 4 THE COURT: It's 11:55. Do I want to break for 5 lunch now and we'll continue cross-examination after 6 lunch? 7 All right. Good. 8 Please be back and ready to go at 1:00. We go off 9 record at 11:55. 10 (Whereupon, the lunch recess was taken.) 11 ---o0o--- 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 10139 1 THURSDAY, DECEMBER 7, 2023 - - AFTERNOON SESSION 2 THE COURT: All right. Let's go back on record. 3 We're back on record at 1:01. 4 All right. Who will next cross-examine? 5 MR. HILL: While Mr. Miltner -- this is Brian 6 Hill, USDA -- while Mr. Miltner gets to the lectern, I 7 would say that there is one change that is justified on 8 Exhibit 436 on page 25. I think the witness did say that 9 he transposed 8.7 for 7.8, if I remember correctly. 10 THE WITNESS: Correct. 11 THE COURT: I'm looking at Exhibit 436, which 12 page? 13 MR. HILL: Page 25, on the first bullet point, I 14 think he said that he just transposed those numbers. It's 15 supposed to read 7.8% instead of 8.7%. 16 THE COURT: All right. Very good. And we'll make 17 that change on the record copy. Have you located the 18 page? 19 All right. And so I'd just like you to read that 20 bullet point, Doctor, so that we have it exactly like you 21 want it. 22 THE WITNESS: The correct bullet should read: 23 "Proposal 19 raises Class I prices by 7.8%, $1.50 divided 24 by $19.20." 25 THE COURT: Thank you. 26 // 27 // 28 // 10140 1 CROSS-EXAMINATION 2 BY MR. MILTNER 3 Q. Good afternoon, Dr. Balagtas. 4 A. Good afternoon. 5 Q. My name is Ryan Miltner. I represent Select Milk 6 Producers, which is a cooperative of farmers in the 7 Midwest and Southwest. 8 In preparing your report and your testimony, how 9 closely did you need to look at the proposed Class I 10 differentials that are in Proposal 19? 11 A. How closely did I need to? I -- there is an 12 Excel spreadsheet dated from June, I believe, that lists 13 Class I differentials by county, proposed and current, and 14 a bunch of other columns, and those are the ones that I 15 consulted. 16 Q. You noted in your PowerPoint and your written 17 testimony, and in questions, that the average increase of 18 the differentials was $1.50. 19 Did you -- in order to perform your analysis did 20 you have to go and look at the individual Milk Marketing 21 Orders and figure out for the Mideast Order that the 22 average differential increase was X number? 23 A. I did not do that. It would -- I took a simple 24 average across all of the counties to calculate that $1.50 25 increase, and I think it would it would be a good idea to 26 think about looking -- that it would be a positive 27 direction, productive direction for further analysis if 28 you looked at a weighted average or if you looked at 10141 1 subnational state marketing order county effects. Yes. 2 Q. Would you think there would be value in looking at 3 the changes in the differentials, either on a marketing 4 order basis or a milk shed basis or something like that? 5 A. I -- I think that could be a productive direction, 6 yes. 7 Q. And as I was more listening to your testimony than 8 I was focusing on the written part. And I thought it was 9 pretty clearly presented, so thank you for that. 10 Sometimes the economic testimony gets a little thick. 11 But I heard you numerous times refer to the fact 12 that you had examined, interpreted, or drawn conclusions 13 from other research on elasticities and things like that. 14 And Ms. Hancock asked you if you had modeled anything. 15 Did you perform what we would -- what I would, I 16 guess, call an econometric analysis of any of the impacts 17 of the proposed Class I differential changes? 18 A. I did not conduct any econometrics, no. 19 Q. Okay. And I'm flipping through here because I 20 still see that a lot of my questions have been already 21 asked and I don't want to duplicate them. 22 A. That's fine. 23 Q. The Proposal 19 the increases there, have you had 24 a chance to look into what the basis for those increases 25 are, in other words, the rationale and methodology in 26 arriving at the numbers that are in Proposal 19? 27 A. Are you referring to the increase in the Class I 28 differential? 10142 1 Q. Yes. 2 A. If I had a chance to look at the basis? So I 3 have -- I'm not sure I understand. Sorry. 4 Q. That's okay. I'll try to phrase my question a 5 little better. 6 Do you understand that the Class I differentials 7 in Proposal 19 have some basis in an econometric model 8 known as the USDSS? 9 A. I understand that that's the case, yes. 10 Q. You also understand that the results from that 11 model were then examined and adjusted based on members of 12 a working group with the National Milk Producers 13 Federation? 14 A. I understand that's the case, yes. 15 Q. Okay. Did you look at just the results of the 16 USDSS model output and evaluate whether that would change 17 your assessment of the impacts of the -- of a Class I 18 differential adjustment? 19 A. So I did not see any USDSS modeling or results. 20 I -- my understanding of Proposal 19 is this spreadsheet 21 dated from June, so I only see the output, I think, just 22 the -- just the final Class I differential proposals, 23 which I understand to be coming from this model. 24 MR. MILTNER: Your Honor, could I get Exhibit 301? 25 And I can hand that to the witness unless you are planning 26 on doing so. 27 BY MR. MILTNER: 28 Q. Dr. Balagtas, that's a spreadsheet that was 10143 1 introduced as an exhibit. 2 Is that -- well, first of all, do you recognize 3 that spreadsheet at all? 4 A. Yes, I do. 5 Q. Does that look like the spreadsheet that you were 6 provided that gave the differentials you evaluated? 7 A. Correct. Yes, it does. 8 Q. And if you are looking at that spreadsheet, I 9 don't think you need to flip through the pages or 10 anything, but which letter column would you have looked at 11 to complete your analysis? 12 A. Column O as Proposal 19, and then Column I as the 13 current differentials. 14 Q. Did you look at all at Column -- Columns F? 15 A. I -- I have seen them. I didn't give them any 16 thought or I did not analyze any of them. I didn't. 17 Q. Okay. Did you know that those are the columns -- 18 F, G, L -- those were the numbers derived from the USDSS 19 model? 20 A. I did see, because the first row labels have model 21 estimates in the name, I did see that, and suspected that 22 they were from the model. Yes. 23 Q. Now, even though I was listening to your 24 testimony, I did read your written statement, so there was 25 not a waste of your time by any stretch. And on page 6 of 26 your written statement, which I believe is Exhibit 435. 27 A. Page 6? 28 Q. Yes. In the third paragraph there, it reads, "I 10144 1 start from the premise that Class I differentials set in 2 2000 were appropriate for market conditions during that 3 time to support adequate supply of fluid milk and dairy 4 farm income." 5 Now, my question is, you started from that as a 6 premise, which is different from a belief or a conclusion. 7 So do you believe that the differentials that were 8 set in 2000 were appropriate? 9 A. I don't have a view of that. 10 Q. You acknowledge that you have done some -- well, 11 not some, but a good bit of work in dairy and Federal 12 Orders during that period of Federal Order Reform. 13 Were you at all involved in research or kind of 14 watching that process? 15 A. I don't believe I was involved -- I -- I don't 16 believe I was involved in that process. 17 Q. Okay. Now, your statement there speaks 18 specifically to the differentials set in 2000. 19 Do you have an opinion as to whether the 20 differentials as they stand today are at appropriate 21 levels? 22 A. So based on the premise of 2000 -- based on the 23 premise that there are adequate supplies in 2000, combined 24 with the fact that we have had a 40% increase in milk 25 production since then, and a decrease in milk consumption 26 since then, I think it follows that there is adequate 27 supply for Class I use today. 28 Q. Okay. Speaking just to the differentials, do you 10145 1 believe that the Class I differentials today are set at an 2 appropriate level? And if you don't have an opinion, 3 that's fine. 4 A. I'm not sure I have an opinion on that. I do not 5 have an opinion on that. 6 Q. You have also provided some testimony on the 7 retail price of conventional reduced fat milk. 8 Have you as part of your work for your testimony, 9 or otherwise, looked at the price of retail conventional 10 milk in real terms? 11 A. So starting with your statement, I don't think I 12 looked at the price of conventional nonfat milk, so I'm 13 not sure about that. 14 But then can you restate the question that you 15 followed with? 16 Q. Sure. If I said nonfat, I'm sorry, I think your 17 testimony refers to reduced fat. If I misstated that, 18 that's my fault. 19 My question is, as part of the work for your 20 testimony here, or otherwise, have you studied the price 21 of conventional milk retail price in real terms? 22 A. Controlling for inflation? 23 Q. Yes. 24 A. Adjusted for inflation? No. 25 Q. If -- if CPI from 2000 to 2022 is up about 71.2%, 26 and the reported price of fluid milk, conventional fluid 27 milk, is up 47.1% over that same timeframe, would that 28 indicate that -- well, would you be able to draw any 10146 1 indications from that or draw any conclusions from that 2 data? 3 A. Yeah, that -- and to be clear, I don't think I 4 reported time series data on fluid prices as you are -- as 5 you discuss that I recall. But in general, inflation -- 6 inflation means that any growth in nominal prices results 7 in a smaller growth in real prices. That's what -- that's 8 what you mean, yes. 9 Q. I believe you acknowledged or stated previously 10 that the Class I differentials, other than some changes in 11 the Southeast, have remained -- have been unchanged since 12 2000; is that correct? 13 A. That's my understanding, yes. 14 Q. And I'm trying to find where the statement is so I 15 can give it verbatim, and I'm not finding it at the 16 moment. 17 But I believe you stated that one of the purposes 18 of the Federal Orders is to help equalize bargaining power 19 between farmers and milk buyers; is that correct? 20 A. Yeah. My understanding is one of the objectives 21 of Milk Marketing Orders is something along the lines of 22 fair treatment of farmers with respect to milk buyers. 23 Q. Do you understand that that fair treatment is -- 24 is specifically tied to the equalization of bargaining 25 power between farmers and their milk buyers? 26 A. That would make sense to me, yes. 27 Q. Do you have any understanding as to the genesis of 28 that statutory goal? 10147 1 A. I don't know the specific history of the statute. 2 I do understand looking back on, you know, turn of the 3 last century, we were looking at a world with -- so lots 4 of relatively small farmers selling to a relatively few 5 and large milk processors. And that -- that put farmers 6 at a disadvantage in terms of negotiating power. Yes, I 7 understand that. 8 Q. Do you understand part of the analysis was also 9 due to the perishability of milk? 10 A. Yes. 11 Q. Okay. Do you believe that the -- that in an 12 unregulated market, dairy farmers and dairy handlers have 13 equivalent or acceptedly equivalent bargaining power? 14 A. I'm not sure what the world would look like in an 15 unregulated market. We have had Marketing Orders and 16 other policies for a hundred years. And so I'm not sure 17 I'm -- I'd speculate on what the world would look like in 18 their absence. 19 Q. Do you believe that minimum regulated prices -- 20 minimum regulated classified prices and marketwide pooling 21 are advisable in today's dairy marketplace? 22 A. Again, it would -- relative to what would be the 23 question, what would the world look like in their absence, 24 what other institutions might emerge, contracting, for 25 example. And so would -- yeah, whether -- whether the 26 effect of Marketing Orders relative to that unobserved, 27 hypothetical, unregulated world is -- is really difficult 28 to gauge. I'm not trying to be evasive. I think it is a 10148 1 hard question to wrestle with. 2 Q. It might be the question, right? Or one of the 3 questions. 4 I'm going to give you a series of hopefully brief 5 hypotheticals and ask your opinion on it, if you are 6 you -- if would indulge me. 7 I want you to assume that, in fact, CPI is up 72% 8 between 2000 and 2022, and that over that same period, the 9 price of a gallon of milk at retail conventional is up 10 47%. And I want you to assume that the Class I 11 differentials over that same period have remained 12 stagnant. I don't think those are hypotheticals, but 13 accept them as true if you would. 14 THE COURT: Accept what? 15 BY MR. MILTNER: 16 Q. Accept those statements as true, if you would. 17 And then I want you to assume that, in fact, 18 farmers do not have bargaining power versus their milk 19 buyers. 20 Does the fact that the regulated component of the 21 Class I price having remained constant for 22 years 22 account for at least some of the 23% gap between inflation 23 and the shelf price of conventional milk? 24 A. If Class I differentials had grown over that 25 time -- I'm sorry, what's the gap we're talking about, the 26 gap between inflation -- inflation and growth in milk 27 prices? 28 Q. And growth in the retail price of conventional 10149 1 milk. 2 A. Yeah. So if Class I differentials had grown over 3 that time, retail prices would have grown by more than 4 what we have observed. I think that's what you are asking 5 me. I'm not sure. 6 Q. Not quite what I'm asking. Maybe let me approach 7 this somewhat differently. 8 The Class I differentials you would agree 9 establish a regulated minimum price, correct? 10 A. The Class I differentials together with the mover, 11 yeah. 12 Q. And sellers of milk, producers, and their 13 cooperatives can negotiate prices higher than that 14 minimum, correct? 15 A. Correct. 16 Q. Okay. One of the issues in the hearing has been 17 how effective are those over-order premiums. 18 And so my question is, given that the Class I 19 differentials have not changed in 22 years, and there is a 20 gap between overall inflation and the inflation of the 21 price of a gallon of milk, conventional milk, is that an 22 indication that over-order premiums, while an important 23 part of the marketplace, may not adequately cover 24 inflationary pressures? 25 A. I -- so I think there's a premise in the question 26 that retail milk prices should have grown at the rate of 27 inflation, that's why there's a gap -- that's what you are 28 calling a gap. I'm not sure that that's -- I think that's 10150 1 a judgment that I'm not willing to make. 2 I do think if Class I differentials had grown over 3 that time period, Class I prices -- or retail fluid milk 4 prices would have grown faster relative to inflation than 5 what we have observed. So I -- I'll agree with that part. 6 But I -- yeah. I think -- yeah. Real milk 7 prices, retail prices of milk have fallen over time, I 8 think -- I agree with you with that. I don't -- I don't 9 agree that it's -- it's obvious that they should not have, 10 I guess is what I'm saying. 11 Q. That's fair. And if you are not willing or you 12 don't feel comfortable drawing that conclusion, would you 13 agree it is a valid hypothesis? 14 A. Sure. Yes. 15 Q. Okay. I wanted to ask a few questions about the 16 Excel spreadsheet or the printouts you have at the end of 17 your written statement if I could. 18 You stated in your PowerPoints that "in the 19 absence of relevant demand elasticities, I report effects 20 for a wide range of elasticity values." 21 And that was the elasticities of demand for nonfat 22 dry milk and butter, correct? 23 A. Correct. 24 Q. You describe there's an absence of relevant demand 25 elasticities. 26 Were you able to start from a point of any demand 27 elasticity reference that you thought was a fair place to 28 start? 10151 1 A. I have seen a few elasticity estimates. There are 2 some in the documentation from the ERS dairy model. There 3 are some in the documentation for the FAPRI model. So 4 there are some out there, yeah. 5 Q. So if the -- if those were your reference points, 6 do you -- where -- where do the elasticities that you 7 plugged in here relate to those that you found referenced 8 elsewhere? 9 A. I believe -- and this is off the top of my head -- 10 that that powder and nonfat dry milk elasticities in both 11 of those studies are more inelastic than my inelastic 12 case. I believe that that's the case. I also think it's 13 not clear that they are relevant to what I'm trying to do 14 here. 15 Q. They are not relevant? 16 A. Well, so in the case of the ERS study, they report 17 a demand elasticity -- a U.S. demand elasticity for nonfat 18 dry milk, for example. That's demand for nonfat dry milk 19 from consumers in the United States. 20 Q. Right. 21 A. So that -- that's not relevant for a market for 22 which 75% of the product is exported. So I need a 23 demand -- global demand elasticity, if you will, for U.S. 24 nonfat dry milk. 25 Q. And you -- I think what you have stated, and 26 more -- well, regardless of the words you have used -- 27 essentially what you have said is that I had to put some 28 elasticities in here, and I gave a range, but I -- you 10152 1 don't know -- you don't know which column is more likely 2 than another of being reality? 3 A. I think that's accurate. 4 Q. I don't mean to sound flippant when I ask -- 5 A. No, I think it is absolutely accurate. 6 I do think the elasticities I have here are -- are 7 all plausible. If you told me that demand elasticity for 8 nonfat dry milk was minus 3 or minus 12, I wouldn't argue 9 with you. I think those are plausible. 10 And -- and my point here is, within that plausible 11 range, the outcome -- you know, the net effect on farmers 12 could flip, could go black or red, right. And that's -- 13 that's the -- so I -- I present a plausible range. It 14 might be a little wider; it might be a little narrower, 15 right? But within that range, the net effect of 16 Proposal 19 might make farmers worse off or better off, 17 and that for me would be -- would be troubling. 18 Q. Were you here for Dr. Brown's testimony yesterday? 19 A. I -- I watched it as I was driving in the morning, 20 and then I watched some of the cross in person, yes. 21 Q. Did you happen to hear his statement that -- when 22 people ask, he says that, well, the milk price will be 23 between 15 and $30 per hundredweight? Did you hear that 24 statement? 25 A. Yeah, and I heard some statements about not making 26 bets on his -- yeah, not -- not gambling on his -- or 27 investing on the basis of his model. 28 Q. But you are -- you're a little bit in the same 10153 1 boat because you are saying that the total change in farm 2 milk revenue in the all-milk price is between a loss of 3 $637 million and a gain of $262 million, and it could be 4 outside of that range, we just don't know? 5 A. And that is exactly the point I wanted to 6 everybody to understand here is that there's uncertainty 7 on the net effect on -- on dairy markets. I think there 8 is a lot less uncertainty about what's happening in the 9 fluid markets. If you had told me, you know, that the 10 demand elasticity for fluid milk is minus point -- I 11 think -- so I have minus 1.2. If you told me it was minus 12 1.8 or minus .9, I wouldn't argue too much with you, and 13 it doesn't change my results a lot. It changes the 14 quantity of milk that gets diverted from Class I and to 15 manufacturing, so it changes the magnitude, does not 16 change the direction of the effects. Right? 17 The uncertainty I'm talking about here in this 18 table changes the direction of effects, and so there's -- 19 I think we ought to be cautious, right, about -- I think 20 AMS ought to be cautious about pursuing something when 21 we're not -- it's not clear even the direction of the 22 effect on -- on the all-milk price. 23 Q. You have presciently steered me back to fluid milk 24 elasticity, which is where I want to ask my last set of 25 questions. 26 A. All right. 27 Q. I don't know that anybody -- well, most of the 28 people in here had not seen Dr. Capps' study until he 10154 1 presented it. 2 When did you first have the opportunity to see it? 3 A. I saw a draft maybe a week before his testimony. 4 Q. Was that the same time that IDFA reached out to 5 you about presenting additional testimony on the topic? 6 A. No. I had already started my work in -- for IDFA 7 in August. 8 Q. Okay. Now, there was another study that Dr. Capps 9 was I believe lead author on that gets sent to AMS and 10 then to Congress. The most recent one isn't fully public 11 yet, but the one prior, which was published late summer, 12 early fall, reported an own-price elasticity for milk 13 of -- 14 (Court Reporter clarification.) 15 BY MR. MILTNER: 16 Q. -- own-price elasticity -- reported an own-price 17 elasticity of negative 0.037, if I remember correctly. 18 What do you think of that number? 19 A. It's quite inelastic. I don't know the data he 20 used to estimate that parameter. But it's -- it's not 21 surprising in the context of the long history of -- of 22 published estimates in this area. 23 Q. And, in fact, it is those studies that show an 24 inelasticity of own-price -- own-price elasticity, it is 25 elastic, which tend to be inconsistent historically, 26 correct? 27 A. The -- these recent studies that I have cited tend 28 to be more elastic than the body of work that's -- yes, 10155 1 than the body of work. 2 Q. Does it surprise you that Dr. Capps did two 3 studies in such a close proximity of time and ended up 4 with such different results? 5 A. It's -- it's not surprising that you get different 6 estimates. One, this work is tricky, hard to do, and I 7 respect Dr. Capps' work in the area. But also, even if 8 you use the same methods, estimates on using different 9 data will create different -- will give you different 10 estimates. 11 So, for example, I think in those promotion 12 studies, I use annual historical disappearance data, I 13 think -- I think that's case I'm not sure. Right? And so 14 that's -- that's -- that's different than the type of 15 scanner data we're talking about that that -- these recent 16 studies that I have cited use. And -- and so not too 17 surprising given that we're looking at different data 18 sets, different methods, that you -- that you get 19 different estimates. 20 Q. And the three studies cited that show elasticity 21 greater than negative 1 are all using weekly scanner data, 22 are they not? 23 A. Correct. 24 Q. And I asked Dr. Capps if he thought that weekly 25 data on sales and sales prices and sales volumes might 26 capture large promotional changes in price and volume that 27 might not be captured when measuring monthly, quarterly, 28 or annual data. 10156 1 Do you think that that might be the case? 2 A. Yes. So the weekly data will capture variation 3 week to week, or even month-to-month variation, that 4 annual data would not -- that is not observed in annual 5 data, yes. 6 Q. This morning we had testimony from a gentleman 7 from Maple Hill Creamery I think is the name, if I have it 8 correct, and he described a promotion that lowered his 9 price by 7 to 10% resulting in a 30% change in his sales 10 for that week. 11 So that would appear to be very elastic, wouldn't 12 it? 13 A. That looks like responsive consumer behavior, yes. 14 Q. And over a month period, or a yearlong period, 15 that might not be the -- you may have the same sales data, 16 but measured a different way, showing a different 17 elasticity, correct? 18 A. Correct. 19 MR. MILTNER: Thank you. That's all I have. 20 THE COURT: Mr. English. 21 CROSS-EXAMINATION 22 BY MR. ENGLISH: 23 Q. Good afternoon. Chip English for the Milk 24 Innovation Group. 25 Dr. Balagtas, thank you for being here. 26 I have two sets of questions. One is a direct 27 follow-up on Mr. Miltner -- 28 THE COURT: Slowly, please. 10157 1 MR. ENGLISH: It's sunny outside. I'm trying to 2 get done. But I will slow down. Thank you, Your Honor. 3 BY MR. ENGLISH: 4 Q. To the extent you may have looked at the most 5 recent Dr. Capps study, are you aware that he studied 6 cross-elasticities with substitutes for fluid milk such as 7 beverage, water, juice, and plant-based beverages that 8 claim to be milk? 9 A. I am aware, yes. 10 Q. And are you aware that that is different from any 11 study he has done for Congress? 12 A. I am aware, yes. 13 Q. Turning to page 26 of Exhibit 436, you had a 14 conversation with counsel for National Milk on this page. 15 And I -- I think I understand it, but I'm not sure the 16 record is clear. 17 So first, what is consumer surplus? What does the 18 term consumer surplus mean? 19 A. It's the difference between the maximum price 20 consumers are willing to pay and the price that they do 21 pay in markets. It's a -- it's an economic concept that 22 measures well-being of consumers in markets. 23 Q. And why is that important to your analysis? 24 A. Because in this case, consumers are affected by 25 the fluid milk price, a change in the fluid milk price in 26 retail prices, and it is important to quantify the effects 27 of that change on consumer well-being. And consumer 28 surplus is the standard way to measure that effect on 10158 1 their well-being. 2 Q. All right. Thank you. 3 My remaining questions are about the three bullet 4 points, and my understanding, and see if you agree with my 5 understanding. 6 Your first bullet point simply used Dr. Capps' 7 results and concluded that there was -- and using the 8 economic term -- harm to consumers of $11.8 million per 9 week, correct? 10 A. Correct. 11 THE COURT: And to be clear, that's million with 12 an "M"? 13 MR. ENGLISH: Million with an "M." 14 THE COURT: Thank you. 15 BY MR. ENGLISH: 16 Q. Your second bullet point assumes that Dr. Capps' 17 data can also be applied to the untracked retail of 12%, 18 correct? 19 A. Correct. 20 Q. And when you say harm to consumers is 14 million, 21 my understanding is that is 2.2 million added on to the 22 11.8 million to get to a total of 14 million if you took 23 the first two bullet points together; is that correct? 24 A. Correct. 25 Q. So it's cumulative, correct? 26 A. Correct. 27 Q. And so the third bullet point, which is then 28 assuming Dr. Capps' data -- 10159 1 A. I'm sorry, can you go back and say that? 2 Q. So as I understood it -- 3 A. Yes. 4 Q. -- the first bullet point is, just applying 5 Dr. Capps, you conclude that there's this harm to 6 consumers of $11.8 million per week -- 7 A. Yes. 8 Q. -- correct? 9 A. Yes. 10 Q. And the second bullet point is cumulative, that is 11 to say if you take 12 -- the 12% that is untracked retail, 12 and you add that in, you end up with $14 million harm per 13 week taking both of those data points, correct? 14 A. Correct. $14 million per week is the loss in 15 consumer surplus for all consumers at retail, the 76% in 16 Capps' data. 17 Q. Thank you. That's what I'm trying to get at. I'm 18 trying to get at this cumulative concept as opposed to 19 somehow it's additive. 20 A. Correct. 21 Q. Does that question make sense? 22 A. Yes. 23 Q. So then that's -- my -- my third point then is 24 that point too is cumulative, that is to say the 24% that 25 is foodservice, schools, military, whatever, when you add 26 that to the first -- to the all retail, that's where you 27 end up with 18.4 million, correct? 28 A. Correct. 10160 1 MR. ENGLISH: Thank you. That's all I have. 2 THE COURT: Who next has cross-examination? 3 MR. SLEPER: Good afternoon, Judge. 4 CROSS-EXAMINATION 5 BY MR. SLEPER: 6 Q. Good afternoon, Professor. 7 A. Good afternoon. 8 Q. I'll call you "Professor," because I'll butcher 9 your name up. 10 I just have a few questions, a couple of them just 11 for clarification. 12 THE COURT: Please identify yourself. 13 MR. SLEPER: Okay. 14 BY MR. SLEPER: 15 Q. I'm primarily going to focus on Exhibit 435 -- 16 THE COURT: No, no. Say who you are. 17 MR. SLEPER: Oh, I'm sorry. Jim Sleper, 18 S-L-E-P-E-R, Sleper Consulting, LLC. 19 Thank you, Judge. I forgot about that. 20 BY MR. SLEPER: 21 Q. In your exhibits in both 435 and 436, you discuss 22 the impact an increase in Class I differential is going to 23 have on consumption, additional milk going into 24 manufacturing purposes. You back that up with some of 25 your analysis with elasticities and so forth. I'm just 26 paraphrasing if I can. 27 I didn't see anything in terms of what one of the 28 major tenets or premises in which National Milk actually 10161 1 increased -- or asks in this proposal to increase Class I 2 differentials, and it had to do with the impact of 3 additional costs associated with servicing the Class I 4 market today as compared to back in 2000 Federal Order 5 Reform; is that fair? 6 A. I would say the first third or so or quarter of my 7 testimony, both written and in the slides, addresses that 8 question. 9 Q. Okay. But I didn't see a whole lot of detail on 10 that aspect. 11 And all I'm simply asking is: Do you recognize 12 that there are additional services, additional costs in 13 servicing the fluid market? 14 A. I understand that that's the case, yes. 15 Q. Okay. So you do understand that balancing costs 16 have most likely increased 2023 as compared to 2000; 17 plants do not operate seven days a week; fluid plants take 18 special quality requirements to receive their milk and so 19 forth. 20 But you do recognize those are additional costs, 21 and those are just some examples; is that fair? 22 A. I recognize that those are additional costs. 23 Q. Okay. One of the major aspects of it has to do 24 with hauling costs as well: Diesel fuel, labor, terms of 25 haulers and so forth. Okay. Good to know. 26 It's often that stated dairy farmers are price 27 takers, and so when I see your exhibit, and I think it was 28 437, which was your Marketing Power of Co-Ops, would you 10162 1 agree with my statement that dairy farmers are indeed 2 price takers or not? 3 A. I think individual dairy farms are price takers, 4 yes. 5 Q. So that being the case, are co-ops price takers or 6 not? 7 A. My analysis found that they have a small amount of 8 market power. 9 Q. Okay. So was a co-op a buyer or a seller of milk 10 then? 11 A. Both. 12 Q. Both. 13 A. I think of -- if you think of a co-op as a 14 producer, then they are a seller of milk, yes. 15 Q. So even if I am a member of a co-op and I own 16 those facilities, am I a buyer then or am I a seller? 17 A. I think of co-ops as a seller of milk and also a 18 processor of milk. 19 Q. Okay. If I can go to your -- and I think it's 20 Exhibit 435, which is your written testimony, page 13 21 specifically. And let's see here. The second paragraph, 22 and I think it's the second line of that particular 23 paragraph, you state: "Thus, in aggregate, U.S. milk 24 production is more than adequate to supply national fluid 25 needs. Over the same period, Class I utilization is low 26 and falling in all but three of the Federal Milk Marketing 27 Orders regions, which leads me to conclude that milk 28 supplies in those markets are also adequate to serve the 10163 1 fluid milk market." 2 Okay? 3 A. Yes. 4 Q. So there's three markets that when I look at, I 5 believe, it's Table 3, which is a couple lines up -- or a 6 couple pages up, and it specifically the three markets 7 that you are referencing, that this change in utilization 8 I think has to do with the Appalachian, which is -- and 9 the Central market, and let's see here, if I have got it 10 right, the Southeast. 11 A. Yes. 12 Q. That would be Federal Orders 5, 30, and 7. 13 Why not Florida? Why don't you think Florida -- 14 THE COURT: Let me stop you. Are you on page 10? 15 MR. SLEPER: Yes. I'm on page 10, Table 3, Judge. 16 THE COURT: And did you mean to skip over Central? 17 MR. SLEPER: No, I stated Central. 18 THE COURT: You said Appalachian -- 19 MR. SLEPER: So there's three that show a positive 20 number, Your Honor: Appalachia, Central, and the 21 Southeast. 22 THE COURT: Thank you. 23 MR. SLEPER: And those are Federal Orders 5, 30 24 and 7, respectively. 25 BY MR. SLEPER: 26 Q. Why would you -- why wouldn't Florida be one of 27 those changes, just speculating? 28 A. Because I followed -- I reported data in which the 10164 1 Class I utilization rate in Florida, between 2001 and 2022 2 fell by 7.7 percentage -- percent, excuse me. 3 Q. Okay. So when I go back to page 13, and if I 4 understand the second sentence that I read, in essence, 5 you are saying in all of the Federal Orders, including 6 those that actually increased, that there's an adequate 7 supply of milk for the fluid market; is that fair? 8 A. I believe that there's an adequate supply of milk 9 in all Federal Orders for fluid uses. 10 Q. Okay. Are you familiar with the milk or the 11 supplemental milk that's brought into Florida as a good 12 example in the fall months? 13 A. I'm familiar that milk is shipped into Florida. 14 Yes. 15 Q. So you also make the statement that there is -- 16 sounds like, and I'm paraphrasing, you know, there's 17 sufficient quantities. 18 But I would be here to say, and I know a little 19 bit about it, that during the months of let's say 20 August 15th through December 15th, it's pretty durn short 21 in Florida. There's a lot of supplemental milk coming 22 into the market. So there would be dairy farmers down 23 there specifically who would say, there's not even close 24 to an adequate supply to meet the fluid market. 25 So would you help me to understand why there's -- 26 in your general statement, there seems to be no problem, 27 when, in fact, in real life, there are really shortages, 28 especially in the Florida market, as well as Federal 10165 1 Orders 5 and 7? I'm just trying to understand the paradox 2 here. 3 A. Yeah, I don't see a paradox. I think the shipped 4 milk is part of the adequate supply. 5 Q. Okay. Even though the dairy farmers are incurring 6 tremendous cost to bring that supplemental milk in, when 7 you, in fact, say there's plenty of milk. Okay. 8 MR. ROSENBAUM: Is there a question, Your Honor? 9 Is that a question? 10 MR. SLEPER: No, that was just a comment. 11 BY MR. SLEPER: 12 Q. Let's go to page 15, if we could. And I don't 13 have but just a one or two more questions, Professor. 14 Yeah. Just trying to get clarification. 15 The next to the last paragraph in page 15, the 16 paragraph that starts with "finally." So: "Finally, it 17 is important to note the Federal Milk Marketing Orders' 18 objective of ensuring adequate supply of fluid milk for 19 consumers implies that encouraging consumption of fluid 20 milk is a goal of the regulation." 21 So does the Federal Milk Marketing regulations or 22 the Act itself actually state that there is a goal of 23 encouraging consumption of milk? 24 A. I don't believe so. I think I infer that from the 25 fact that we're worried about adequate supplies, in which 26 case I'm -- I think adequate means we want enough for 27 consumers to be able to drink milk. 28 Q. Okay. I got you. And that's what I assumed. The 10166 1 word "implies" was written in that particular sentence. 2 Just getting clarification. Okay. 3 My last area would be on page 18, next to the last 4 paragraph. The paragraph that is: "Notwithstanding the 5 ambiguous effect on milk producers, Proposal 9 [sic] would 6 cause significant disruption in dairy markets," and then 7 you go through some of those. 8 THE COURT: Did you mean Proposal 19? 9 MR. SLEPER: Yes. If I misstated that, I 10 apologize, Your Honor. 11 THE COURT: Thank you. 12 BY MR. SLEPER: 13 Q. Help me understand "significant disruption," if 14 you could, please. Give me some examples, give me some 15 real life, what do you mean by significant disruption. 16 A. Oh, a 7% increase in the retail price of milk, a 17 5% decrease in milk consumption, 2 billion extra pounds -- 18 2 billion pounds of producer milk moved from Class I to 19 manufacturing classes, lower prices of dairy commodities. 20 Those that's what I mean by disruption. 21 Q. Okay. In it you say "ambiguous effect on milk 22 producers." It sounded like you at least heard some of 23 Dr. Scott Brown's testimony yesterday. 24 Do you believe in some of his -- or do you believe 25 in the FAPRI analysis where it does show there would be an 26 impact on dairy producers? 27 A. I -- if I recall, he shows a long-run effect of a 28 $0.01 -- $0.01 per hundredweight -- I can't remember if it 10167 1 was in percentages or a hundredweight, but it is $0.01 per 2 hundredweight increase in the all-milk price. 3 Q. Okay. I was just trying to get an understanding 4 when you stick the word "ambiguous" in there versus real 5 life study showing specific numbers, I was just trying to 6 get a little understanding of the word "ambiguous." 7 A. I'm not sure what you mean by "real life study." 8 But so he has a point estimate that says $0.01 increase in 9 the all-milk price. Right? I have a result that is 10 similar, $0.03 per hundredweight increase in the all-milk 11 price. I further go on to say, I'm not certain about some 12 of the key important -- the key parameters that you need 13 to know to estimate that effect. And so over a plausible 14 range of parameter values, that $0.03, or $0.01 in 15 Scott's -- in Professor Brown's case, might be negative or 16 positive. 17 Q. Okay. In the very, very last phrase of that 18 particular sentence, you know, you are going through these 19 various examples of significant disruptions of dairy 20 market, and you mention the diversion of milk from Class I 21 uses to manufacturing uses. 22 So when you say the "diversion," you are talking 23 about the price of manufacturing products and so forth. 24 You are not talking about the actual physical movement of 25 milk to manufacturing, are you? 26 A. I'm talking about the actual physical movement of 27 milk from Class I uses to manufacturing uses. 28 Q. So it's the actual additional cost of moving that 10168 1 milk to manufacturing? 2 A. No, it's -- it's the fact that you are taking milk 3 that was Class I, and now selling it, in my scenarios, in 4 Class IV uses. 5 Q. Okay. I -- I think I have a better, clear 6 understanding. 7 MR. SLEPER: Thank you, Professor. 8 THE WITNESS: You're welcome. 9 THE COURT: Thank you, Mr. Sleper. 10 Who next has cross-examination questions? 11 Now, does the Agricultural Marketing Service want 12 redirect before you ask your questions or do you want to 13 ask your questions now? 14 MS. TAYLOR: I think we're fine going now. 15 THE COURT: All right. I invite the Agricultural 16 Marketing Service to ask questions. 17 CROSS-EXAMINATION 18 BY MS. TAYLOR: 19 Q. Good afternoon. 20 A. Good afternoon. 21 Q. Thank you for joining us here today. 22 A. It's good to be here. 23 Q. Is this the first Federal Order hearing you have 24 done? 25 A. It is. 26 Q. We are getting a parade of all of the economists 27 in this lovely, lengthy hearing, so it's very nice to meet 28 you. 10169 1 A little bit of context. I have been kind of 2 telling each of the economists that have testified before 3 and have done that, here at USDA, our job, make sure the 4 record's clear, ask you questions about your work, maybe 5 take the opportunity to ask you a little bit more broader 6 questions that you could use your professional experience 7 to give an answer to, because once this is over, we can't 8 come back and ask you any questions. So this is our 9 one-time opportunity to make sure we're clear, so to give 10 you that context. 11 And I'm going to try not to be repetitive with 12 what's been asked. 13 If we could turn to page 13. Some of these are 14 going to be quick and simple. 15 A. Of 435? 16 Q. Yes, of your PowerPoint. Oh, 436, excuse me. And 17 I'm probably just going to stick to Exhibit 436. 18 THE COURT: Exhibit 436, page? 19 MS. TAYLOR: 13. 20 THE COURT: 13. 21 BY MS. TAYLOR: 22 Q. You have the list of orders on here, and the same 23 list it looks like on 14? 24 A. 14, correct. 25 Q. I see orders -- the Arizona order and California 26 not on this list. Just wanted some clarification why they 27 are not there. 28 A. Yeah. Because I was trying to compare Class I 10170 1 milk and Class I utilization rates over the 20 years since 2 the last revisions to Class I differentials. The 3 California Federal Order didn't exist in 2000, 2001, nor 4 did Arizona, I believe, so I left those out. Both of 5 those are low Class I utilization rate Marketing Orders, 6 though, and so I don't think that -- and they are included 7 in the all-markets combined numbers. 8 Q. They are included? 9 A. Yes. 10 Q. So 2022 number also includes California and 11 Arizona? 12 A. Yeah. That's why we have 41 billion pounds is 13 total Class I producer milk across all market orders. 14 Q. Okay. I didn't sum these, so I guess if I did 15 that, I would figure out -- 16 A. Yeah, you know what? I probably should have just 17 included the two just for completeness. 18 Q. Okay. I appreciate that. 19 And the percentage change, that's a weighted 20 average? 21 A. For all markets combined? 22 Q. Uh-huh. 23 A. Yes. 24 Q. Okay. And then the same thing on the next page, 25 that is a weighted average? 26 A. Correct. 27 Q. Okay. 28 A. So it's, yeah, a percentage change in the total, 10171 1 which gives you the weighted average. Yes. 2 Q. Okay. Thank you. 3 On page 15, and you have talked about it some, and 4 I think maybe you just were asked a question along this 5 line. You talked about how in six of the regions, Class I 6 use rates have gone down, and you take that as that 7 there's more than adequate supply of milk for fluid uses. 8 And that doesn't necessarily consider, as we have heard 9 here in other testimony presented, the willingness of that 10 milk to supply. 11 A. I'm sorry, the -- 12 Q. Willingness. 13 A. Willingness. Of? 14 Q. The milk to supply these areas. We have heard 15 testimony about difficulties in getting milk to Class I, 16 and there are producers and cooperatives providing that 17 service now, but at a great cost, according to them, and 18 that might not continue in the future. 19 But I wanted to see if you considered that at all 20 while you were doing your analysis. 21 A. The willingness, I -- I think producers, including 22 cooperatives -- hmm, the willingness... 23 Q. New term for you, maybe. 24 A. Yeah. 25 Q. They say -- I'll summarize -- I'm willing to do it 26 right now, but I might not be willing to do it in the 27 future. 28 A. I think the fact that -- looking at, again, on 10172 1 page 13, the fact that Class I utilization rates have been 2 falling suggests that farmers have been willing to deliver 3 milk to those -- to these areas. 4 Q. Okay. And did you consider how Federal Order 5 pooling provisions might impact that Class I percentage 6 number? I mean, there's pooling provisions that allow for 7 diversions, which can lower that number in the grand 8 scheme of things. There's other pooling provisions in 9 orders that kind of keep that milk pooled all the time. 10 A. Right. And I don't -- you know, if you look on 11 page 14, of 27%, that's -- that's the Class I utilization 12 of pooled milk. So I'm not including here the fact that 13 there's an additional amount of milk that's not pooled. I 14 think the Class I is 20% of all milk produced, right? So 15 even lower. But I'm aware of that and that there are 16 pooling conditions and that, but I've not here evaluated 17 them. 18 Q. Okay. On page -- or 16, slide 16, in your 19 highlighting the three Southeast orders. In those 20 markets, a lot of Class II products are often manufactured 21 at Class I plants. 22 So did you consider how that combined Class I/II 23 utilization might impact the milk needs in that market and 24 thus kind of your evaluation? 25 A. I -- I did not do that here. I just took the 26 monthly Class I utilization rates and looked at peak -- 27 peak months. So, no, I did not. 28 Q. Okay. I want to turn to slide 16. 10173 1 A. 16? 2 Q. 17. Thank you. I haven't gotten my afternoon 3 Starbucks yet, so I might be slow. 4 I think in cross helped clarify how those 5 numbers -- you know, you use these numbers to draw the 6 conclusion that the Class I -- the high Class I 7 utilization percentages haven't resulted in higher retail 8 milk prices. 9 What struck me I think looking at that is, well, 10 Federal Orders don't regulate retail prices, one. And we 11 have had handlers testify at this hearing about how they 12 don't control the retail prices and how those things are 13 set, et cetera. And I just -- you know, and a lot of 14 times -- or there's been testimony about how it's still 15 used -- milk can still be used as a loss header in the 16 retail space. 17 So I was just -- wanted to get your opinion on how 18 that also might affect this analysis and if those things 19 need to be considered. 20 A. Yeah. I'm not familiar with various state 21 regulations that might be in effect in these areas. 22 Let me say what I would have expected to find if 23 there were inadequate supplies of Class I milk in these 24 areas. Right? One measure of inadequate is that there's 25 so little milk that we end up with high prices of milk. 26 Right? And that's why I decided to look at these retail 27 milk prices. These three cities -- these cities in these 28 three Marketing Orders with high Class I utilization rates 10174 1 don't in this sample have a particularly high retail milk 2 price. I'm not sure if that's -- helps you understand 3 what I'm trying to communicate here. 4 Q. Okay. And did you look at any orders that have 5 low utilization -- Class I utilization rates and look at 6 how those compare to the retail prices in those markets? 7 A. Those low-utilization-rate cities would be in the 8 in the 30- -- in the whole sample, so -- 9 Q. Sure. 10 A. -- so they show up in the 30-city average or the 11 75th percentile. So comparing to all cities, including 12 those low -- including cities in low utilization marketing 13 areas, in that sample, these three cities don't have the 14 highest retail prices. Let's say it that way. 15 Q. Okay. 16 A. Don't have systematically higher retail prices. 17 Q. And so from that, the opposite -- I don't know how 18 else to say this -- so in areas where there's lower 19 utilization rates, this data would show us that they may 20 have higher retail prices? 21 A. I'm -- I don't think that's -- I don't think that 22 follows. I'm not saying that there's an inverse 23 correlation. 24 Q. Okay. 25 A. I'm saying there's -- that doesn't look like 26 there's much correlation at all. Does that make sense? 27 Q. Yes. 28 A. So I'm not saying that high Class I utilization 10175 1 rates are driving low retail prices. I say the high 2 Class I utilization rates appear to be unrelated to retail 3 milk prices. 4 Q. Okay. And I -- 5 A. I'm sorry, I probably should have made that -- 6 could have made that point clearer. 7 Q. That's okay. 8 And the next slide in -- so in some of these 9 slides you talk about the three Southeast orders, and in 10 other places you talk about Appalachia, Central, and the 11 Southeast. I'm on slide 18, for example, rising Class I 12 utilization rates. 13 Just wondering why you didn't look at the -- 14 A. Central. Yeah. 15 Q. Between those comparisons. 16 A. Yeah, I should have picked a city from the 17 Central. 18 Q. Okay. 19 A. You are right. 20 Q. I'm going to turn to slide 26. 21 A. Slide? 22 Q. 26. 23 A. I think I -- I'm sorry, going back one. I 24 remember now why I picked Florida, not Central. 25 Q. Uh-huh. 26 A. So if you look at Central, it did have a rising 27 Class I utilization rate, but it is a Class I utilization 28 rate of 27% compared to Florida, which is 83%. So I was 10176 1 looking -- I was trying to look at regions with high 2 Class I utilization rates. 3 Does that make sense? 4 Q. Yes. And then -- 5 A. Yeah. 6 Q. The other slide was looking at the -- 7 A. I should have said -- I agree. 8 Q. The other slide was then looking at the increase? 9 A. Right. 10 Q. Okay. 11 A. So 26? 12 Q. Yes, please. 13 So I think I gathered from some cross-examination 14 maybe from Mr. Sleper -- I think I wrote it down 15 somewhere -- on your paper that you imply that the -- an 16 increase in -- that one of the goals of the Federal Order 17 system is to ensure an adequate supply of fluid milk, 18 which is in the Act, and from that you imply encouraging 19 the consumption of fluid milk. And the Act also talks 20 about and how the Department must look at kind of the 21 three-legged stool and how the Federal Order impacts both 22 consumers and producers and handlers, and we do our best 23 to find a balance between those competing interests. 24 So did you take a -- try to do a look at the 25 impact to those two other legs of the stool? And I ask, 26 because later on you talk about the net-net, or maybe just 27 the net. Or maybe somebody else used net-net in this 28 hearing. 10177 1 A. I think you said slide 26, and I'm not 2 following -- 3 Q. Well, I said that, and then I asked a different 4 question -- 5 A. Okay. So let's go back to the question then. 6 Q. Yeah. So -- well, I put my note on this slide 7 because you're particularly talking about the impact to 8 consumers. 9 A. Right. 10 Q. And the Act does mention consumers as providing an 11 adequate supply of milk for consumers, but it also talks 12 about uniform payments to producers and uniform costs 13 among handlers. So we look at the balance of those three 14 competing interests and try -- do our best to find that 15 balance. And so this looks at the consumer side of 16 things. 17 Did you look at the impact to the handler and the 18 producer sides of the stool? Later on you talk about kind 19 of what -- what the net impact is, which to us would look 20 at what -- how that -- how does that impact all three, not 21 just the one. 22 A. Right. I do look if -- slide 29, I do look at the 23 change in the all-milk price, which I think of as the 24 ultimate effect -- or a big indicator of the effect on 25 producers. Right? Higher milk prices are better for 26 producers; lower milk prices make producers worse off. So 27 I do look at that one. I don't look at returns to milk 28 processors. 10178 1 Q. I have a few questions on that slide you just 2 mentioned. 3 But before I get there, you say one of your 4 assumptions you used is that the decrease in Class I 5 pounds you assign to Class IV. 6 So we have had testimony, Dr. Capps' study, which 7 you have talked about a few times, showed that yogurt was 8 a competing product to fluid milk. 9 So would -- another analysis, would you consider 10 assigning some of that to Class II instead of IV? Is it 11 problem to put everything in IV? 12 A. So there's two different types of substitution, I 13 think. And the diversion I'm talking about in my study 14 takes the lost Class I sales, pounds, and puts it 15 somewhere. I put it in Class IV. Right? 16 The substitution you are talking about would 17 affect how much lost Class I sales we have. Right? I use 18 his all -- yeah, so if -- to the extent that some of 19 this -- some of the lost Class I sales goes into Class II, 20 depending on that substitution pattern between fluid milk 21 products and Class II, yes, I think, yes, that would -- 22 that would reduce the net effect -- that would reduce the 23 quantity that goes into Class IV, and there would be some 24 additional milk in Class II I think that -- yes, that's 25 correct. 26 Q. How come you didn't consider any of the milk to go 27 into III when III seems to be a growing market? 28 A. Yeah. I -- I -- somewhat arbitrary. I have a 10179 1 sense that Class IV is also growing. I think it would be 2 fair to evaluate the impact of some of this milk going 3 into Class III, and I think that does not change the 4 qualitative effects of my analysis. You would have 5 increased cheese production, which also affects the 6 complex of -- well, cheese commodity prices, and there- -- 7 and therefore the complex of FMMO prices, but through 8 Class III rather than Class IV, I think. That would be an 9 interesting and valid way to think about this as well. 10 Q. Okay. We have also had discussion or some talk on 11 the -- at the hearing from other witnesses that 12 cooperatives have -- some cooperatives have base/excess 13 plants to help control the supply that they have to 14 manage. 15 Would you expect if there was such a decline in 16 Class I consumption, that instead of putting milk into IV, 17 they might tighten up their supply to help mitigate that? 18 Is that -- I mean, I'm just looking at other 19 considerations to look at other than all this milk goes in 20 Class IV. 21 A. Yeah. I'd hate to speculate on what cooperatives 22 might do. I do think if you -- yeah, I guess it's a 23 possibility, if -- if cooperatives are trying to feel like 24 they have too much milk, that they might -- it's feasible 25 to me that they might reinstitute or institute or expand 26 base/excess plants. 27 Q. Okay. And on your slide 29 -- let's see. I 28 wanted to talk about your elasticity assumptions you made. 10180 1 And in some discussions with, I think, Mr. Sleper, you 2 said that you wanted your elasticities to account or 3 exports since that's where a lot of the powder goes, which 4 is why you didn't use elasticities maybe that other 5 studies have used. 6 And you did talk about ERS, and we did look that 7 up, and that elasticity for dry milk products is negative 8 .124. And the one Dr. Brown testified in his study he 9 used was negative 0.13. And you talked about how those 10 are elasticities for domestic demand of powder, which 11 makes sense. But you tried to choose -- you say that the 12 ones in your studies are plausible based on exports. 13 And I just wanted to know why you think those are 14 plausible. Those are certainly significantly different 15 than the numbers in these other two studies. And so kind 16 of why should we look at those and say, yep, those are 17 plausible ones to assume? 18 A. I think of this -- I think of it this way, right? 19 If -- if the U.S. increased nonfat dry milk by 7%, right, 20 as I estimate here, 7.6%, do I think that the effect on 21 global nonfat dry milk prices would be 30%? That would be 22 the effect of applying a demand elasticity for nonfat dry 23 milk of minus .25. That's -- it's not plausible. We have 24 seen change -- variation in nonfat -- U.S. nonfat dry milk 25 prices in the past, and we don't see those types of 26 effects on global prices of nonfat dry milk. So I think 27 larger elasticities that account for this -- this -- the 28 large export market makes sense for nonfat dry milk. 10181 1 Q. Yeah, I don't know if I followed the math, a 7.6 2 increase in nonfat dry milk to I think you said a 25% 3 increase in -- 4 A. So if you -- if you have the -- if you know the 5 quantity and the percentage change in the quantity, you 6 divide by the elasticity to get the change in the price. 7 Right? So in my case, I'm -- I have got a quantity -- 8 percentage change of 7.6, right, 7.6% increase. If the 9 demand elasticity is minus .25, I divide by a quarter, 10 which is the same as multiplying by four, and I get a 30% 11 decrease in the price. Right? And it's just not -- 12 Q. So -- 13 A. I don't think anyone here -- all of you know 14 dairy -- believe that that's within the realm of 15 possibility. 16 Q. I think we probably would agree with that. 17 But I just want to make sure. You said .25, and 18 maybe you are pulling that from somewhere else. I always 19 need to make sure the math is correct. I see a negative 20 .25 on your elasticity for butter. 21 A. Yeah, I -- you -- you cited an elasticity from ERS 22 of minus -- 23 Q. 1.24. 24 A. -- minus point -- 25 Q. Minus.124. 26 A. Yeah. And I used .25 because it's easier to 27 divide by a quarter than it is to divide by .124 -- 28 Q. Okay. 10182 1 A. -- if that makes sense. 2 Q. It does. 3 A. Yep. 4 Q. If we don't ask you the math now, then I'll never 5 figure it out back in DC. 6 A. So if you applied that elasticity of minus .124, 7 right? I actually think then you will get double -- 8 Q. You are right, okay. 9 A. -- the price effect. And that's -- that's -- I 10 don't -- I don't think ERS is saying that either, right? 11 They have trade equations in there. So somewhere in that 12 model there's an implied -- implied total demand 13 elasticity for U.S. nonfat dry milk, and that would be the 14 number, I think, that's relevant. But I -- they don't 15 publish that number. 16 Q. Okay. If I wanted to look at a price change for 17 powder, I would take -- under your numbers -- 7.6% divided 18 by negative 4, in the first example, and I would get a 19 price change of 1.92%. Okay. 20 A. Correct. 21 Q. Thank you. That's helpful. 22 A. I'm sorry that's confusing. 23 Q. I'm sure I don't help, honestly. 24 A. My undergraduate students have the same complaint 25 when we talk about elasticities. 26 Q. And can you just walk us through then how you get 27 your change in Federal Order skim price and Federal Order 28 butterfat price? 10183 1 A. Yes. So maybe it would be helpful to go to 435, 2 the first un- -- the first unnumbered page, Excel 3 spreadsheet at the back. 4 So starting on line 21, that's the quantity of 5 diverted milk, 2.2 billion pounds, right? 6 Q. Uh-huh. 7 A. I take the fat out of that to calculate the 8 quantity of skim that goes to nonfat dry milk. You see 9 the elasticities there in green. And that gives me the 10 net price change. 11 Follow? 12 Q. Yeah. I'm just curious, because you are using 13 these elasticities that account for exports. The Federal 14 Order is a domestic program. 15 Are those -- my undergraduate question to you is: 16 Is that the right elasticity -- 17 A. Oh, I did not mean to imply that you are like -- 18 I'm sorry if I -- 19 Q. It's okay. 20 A. I'm sorry. Say this again? 21 Q. You are using your elasticities that include your 22 assumptions on exports, that accounted for exports? 23 A. To get a net change -- to get the change in the 24 price of nonfat dry milk. Yes. 25 Q. Okay. I'll let you finish your math before I ask 26 it. 27 A. So that gives me the minus 1.9 that we just saw in 28 the slide that I was looking at. 10184 1 Q. Uh-huh. 2 A. Right? So I take that change in the price of 3 nonfat dry milk. If we flip over to the next page. And 4 on line 45, I have a change in the skim milk price, which 5 is that price change. So basically I take your formula, 6 I -- if I recall, it's in the Excel spreadsheet, but 7 it's -- I multiply that price change -- the change in the 8 price of nonfat dry milk by .9 and then .99 again, right, 9 to get the price -- the change in the price of skim -- in 10 the skim price. 11 Q. Okay. 12 A. Does that make sense? 13 Q. My number crunchers are telling me they follow. 14 Is that per pound or per hundredweight? 15 A. So that is a -- 16 Q. The -- 17 A. The skim price would be per hundredweight. 18 Q. Okay. And the change in butterfat price would be 19 per pound? 20 A. Correct. 21 Q. Then might I suggest we can make that change on 22 the record copy to make sure the units are correct? And 23 I'm looking at slide 29. 24 A. Slide 29. Slide 29? 25 Q. Yes. You have the negative $0.20 per pound. 26 A. That's correct. 27 THE COURT: So we're about to make a change on the 28 record copy. We're in Exhibit 436, and we're on page 29. 10185 1 And the line that we are going to is the "Change in FMMO 2 Skim Price"? 3 MS. TAYLOR: Yes. 4 THE COURT: And where it says negative $0.20 per 5 pound, is that what it's supposed to be, Doctor? 6 THE WITNESS: Should be minus $0.20 per c-weight. 7 MS. TAYLOR: CWT. And I think everywhere on 8 that -- just that line, where it says pounds, should be 9 CWT. So it would be a change in three different places. 10 THE WITNESS: Across that entire row, yes. 11 MS. TAYLOR: Yep. Across that row. 12 THE COURT: All right. So, Doctor, just to be 13 sure the record's clear, I want you to read that row 14 across. 15 THE WITNESS: This is slide 29, in the row labeled 16 "Change in FMMO Skim Price," I have a value of minus $0.20 17 per hundredweight; moving over a column, a value of minus 18 $0.10 per hundredweight; moving over to the right one more 19 column, a value of minus $0.08 per hundredweight. I 20 apologize. 21 THE COURT: No worries. I tell you, it is a treat 22 to watch these people work. 23 And we need a break. 24 MS. TAYLOR: Well, I'm finished, so AMS has no 25 more questions. So that's perfect. 26 THE COURT: You went out with a flourish. 27 Mr. Rosenbaum, let's take our break, and then 28 we'll have you come back. 10186 1 All right. Let's take -- do you want 15 or 10? 2 15. Please come back ready to go at 2:45. 3 (Whereupon, a break was taken.) 4 THE COURT: Let's go back on record. 5 We're back on record at 2:48. 6 MR. ROSENBAUM: Your Honor, at this time I would 7 like to move Hearing Exhibits 435 and 436 into the record. 8 THE COURT: Is there any objection of the 9 admission into evidence of Exhibit 435, which is also 10 Exhibit IDFA-61? 11 There is none. Exhibit 435 is admitted into 12 evidence. 13 (Thereafter, Exhibit Number 435 was received 14 into evidence.) 15 THE COURT: Is there any objection to the 16 admission into evidence of Exhibit 436, also shown as IDFA 17 Exhibit 62? 18 There is none. Exhibit 436 is admitted into 19 evidence. 20 (Thereafter, Exhibit Number 436 was received 21 into evidence.) 22 THE COURT: And do you have further questions of 23 this witness? 24 MR. ROSENBAUM: No, Your Honor. 25 I'm sorry, did you admit 435? 26 No further questions, Your Honor. 27 THE COURT: All right. 28 Ms. Hancock, do you wish to have admitted into 10187 1 evidence Exhibit 437? 2 MS. HANCOCK: No, Your Honor. We were just 3 referencing it. I think it is in the record for reference 4 purposes, but he testified to it. That's fine. Thanks. 5 THE COURT: All right. Does anyone else want to 6 move the admission of Exhibit 437? 7 I see no one does. Do we consider it withdrawn? 8 I have to account for it in some manner. It's never -- it 9 was never moved into evidence, so it's not withdrawn. 10 What can I call it? 11 MS. HANCOCK: I just think it is noted. It is not 12 admitted as evidence, but it is noted for the record. 13 THE COURT: Okay. Good. That's how we'll show 14 it. 437 is noted, not moved into evidence. And we keep 15 it -- we'll keep it as part of the record. It will show 16 as those that are admitted, those that are rejected, and 17 in this case, one that is merely noted, but it will not be 18 considered part of the evidence. All right. 19 I take it no one else has questions for 20 Dr. Balagtas? 21 Your work here is done. And I thank you. It's 22 been extremely interesting testimony. I enjoyed it very 23 much. 24 THE WITNESS: Thank you, Your Honor. 25 THE COURT: Who will be the next witness? 26 MR. ENGLISH: Good afternoon, Your Honor. My name 27 is Chip English for the Milk Innovation Group. 28 Continuing in a spirit of witnesses that are not 10188 1 technically mine, I am actually now calling Dr. Mark 2 Stephenson to the stand, S-T-E-P-H-E-N-S-O-N. And I have 3 a document that was sent to most of the parties and USDA 4 yesterday around 8:30 in the morning to be marked as an 5 exhibit, which is his testimony. 6 THE COURT: Very good. We'll go off record while 7 you mark and distribute those. 8 (An off-the-record discussion took place.) 9 THE COURT: Let's go back on record. 10 We're back on record at 2:53. 11 I have before me Exhibit 438, also shown as 12 Stephenson Exhibit 2. 13 (Thereafter, Exhibit Number 438 was marked 14 for identification.) 15 THE COURT: Shall I swear in the witness? 16 MR. ENGLISH: Yes, Your Honor, please. 17 THE COURT: Dr. Stephenson, would you state and 18 spell your name? 19 THE WITNESS: My name is Mark Stephenson, that's 20 M-A-R-K, S-T-E-P-H-E-N-S-O-N. 21 THE COURT: Have you previously testified in this 22 proceeding? 23 THE WITNESS: I have, Your Honor. 24 THE COURT: You remain sworn. 25 MARK STEPHENSON, 26 Having been previously sworn, was examined 27 and testified as follows: 28 THE COURT: Mr. English. 10189 1 DIRECT EXAMINATION 2 BY MR. ENGLISH: 3 Q. So, Dr. Stephenson, you're appearing today not on 4 behalf of any party, correct? 5 A. That's correct. 6 Q. You do expect -- or at least I expect you to 7 return sometime in January to testify on MIG 20, correct? 8 A. That's my understanding, sadly. 9 Q. Not sure how to take that, Doctor, but I assume it 10 is because you are going to have to come back in January. 11 Today you are not testifying about MIG 20, 12 correct? 13 A. No. 14 Q. Earlier in this proceeding Dr. Nicholson testified 15 and provided information about what is known as the USDSS 16 model, correct? 17 A. That's correct. 18 Q. And you happened to tune in and watch on that 19 particular day, correct? 20 A. I did. Yes. 21 Q. And you noted that on more than one occasion 22 Dr. Nicholson suggested that you might have information to 23 add to the record, correct? 24 A. That's correct. 25 Q. And since that time, you have also periodically 26 listened in and from time to time you have heard that 27 somebody from industry or USDA might have further 28 questions of you about the USDSS model, correct? 10190 1 A. I have heard that, yes. And I did come here, 2 Mr. English, to answer some of those questions, and I have 3 tried to put my comments here today in the form of bullet 4 points that address some of those that I have heard. And 5 I'm certainly willing to answer any questions we have 6 about the model. 7 Q. And with that introduction, Your Honor, I am 8 perfectly happy to have Dr. Stephenson take over, and I 9 will sit down. 10 THE COURT: All right. 11 I think it's wonderful that you are willing to do 12 this. So many questions have been to what extent what 13 factors are in the model. 14 THE WITNESS: Uh-huh. 15 THE COURT: So I appreciate very much your coming. 16 THE WITNESS: Well, I'm -- I'm happy to do that, 17 and I think that people should know, absolutely, what is 18 in the model, what is not in the model, and, you know, to 19 try to capture the understanding about what these results 20 really are what they imply. So I will talk about that. 21 And my comments, they're in the form of bullet 22 points rather than just a written narrative, but I will 23 read them nonetheless. 24 The USDSS, which is our acronym for the United 25 States Dairy Sector Stimulator, is a large and 26 computationally complex model which solves a very simple 27 task: Assemble milk at farms and move it to plants to be 28 manufactured into dairy products that are distributed to 10191 1 consumers across the 48 states. 2 The model is constrained by the location of milk 3 production and the volume and components at the county 4 level. A few states, such as Wisconsin and California, 5 report milk production at the county level, but most 6 states don't. We estimate county milk production for 7 states that don't report by using the National Ag 8 Statistic Service ag census data for dairy cow numbers at 9 the county level, and we apportion the NASS state milk 10 production using those. 11 The USDSS accounts for component levels, which 12 vary by region. NASS reports butterfat at the state 13 level, and that butterfat is used in the counties within 14 the state. Protein and other solids levels are estimated 15 using the FMMO data to establish a relationship between 16 butterfat and another components. We use regression 17 analyses employed to estimate the other component levels. 18 And as a final check on milk volumes and components, state 19 and national totals are calculated and calibrated to be 20 precisely equal to the NASS data for the month and year of 21 interest. 22 Dairy product processing is constrained by actual 23 plant locations, the approximate size or capacity of the 24 plants, and the products produced there. A proprietary 25 database of these plants is maintained and updated with 26 popular press news items, Federal Milk Marketing Orders, 27 Pasteurized Milk Ordinance, AMS, and personal 28 communication with industry participants. The plant 10192 1 capacity values that we do have, and that's for a little 2 more than half of the plants, account for about 90% of the 3 NASS milk volume in the 48 states. So we don't have 4 capacity estimates for every plant, but we do have them 5 for the majority of the larger plants. 6 Dairy products are constrained by the components 7 required for their manufacture. We have established an 8 extensive spreadsheet which details final product 9 components and the make procedure which produces them from 10 either raw milk or intermediate dairy products, such as 11 cream, skim, condensed, filtered milks in various forms, 12 and skim or nonfat dry milk powder. 13 Domestic and export use of final products are 14 distributed to the counties of the 48 states or ports for 15 overseas shipments. The volume of these products are 16 constrained to the volumes sold to consumption -- that's 17 what we might call demand at locations -- or exported 18 through those ports. We utilize per capita demand as 19 calculated by the Economic Research Service and multiply 20 the county -- by the county population. 21 In previous iterations of the model, ERS had done 22 per capita demand estimates by region, by state, and 23 ethnic strata. That hasn't been updated in many years, so 24 we are now using just a national per capita value. 25 THE COURT: Let me ask you, "in previous 26 iterations," would you read that sentence again? 27 THE WITNESS: Sure. 28 In previous iterations of the model -- and maybe 10193 1 parenthetically I should mention that we have been 2 building this model over a 30-year time period, so we have 3 had less sophisticated versions of the model, and today it 4 is, we think, relatively sophisticated. 5 But at any rate, in previous iterations of the 6 model, ERS had done per capita demand estimates by region, 7 age, and ethnic strata. That hasn't been updated in many 8 years, so now we're using just the national per capita 9 average value. 10 However, California has higher solids nonfat 11 standards for lowfat milk, and we enforce that standard on 12 fluid sales in that state. We also have data from AMS 13 which indicates variable preferences for butterfat content 14 in fluid milk by region, and we also utilize that in our 15 estimates. 16 Transportation associated with raw milk assembly, 17 final product distribution, or interplant shipments are 18 constrained to take place by the shortest distance over 19 actual road networks. This is not the shortest distance 20 calculated by "the great arc" of the earth, but rather the 21 actual miles that a truck must travel over named roads. 22 There are in our model 9,436,323 of these arcs or roads 23 that the model can traverse, which connects all geographic 24 points in the model. 25 The cost of transportation is calculated using a 26 highly detailed economic engineering model. The model 27 begins with a concept of a hauling firm, which describes 28 their vehicle fleet of both active and reserve trucks. 10194 1 Fuel, oil, tire, and interest rate costs per unit are also 2 accounted for. It also cost accounts for overhead 3 maintenance of the fleet. 4 Individual tractor values are identified, such as 5 how many axles, or tires, type of fuel, unloaded and 6 loaded mileage on those trucks, insurance, fees, 7 et cetera. Tanks on street chassis trucks and trailers 8 pulled by tractors are similarly input. Employees are 9 identified, and their wage and overtime, if applicable, 10 rates, as well as benefits are accounted for. 11 From the individual data -- and I should say firm 12 data there -- various routes are assembled which must use 13 one of the trucks (a tractor-trailer or a straight 14 chassis), one of the employees, and it describes the 15 route: As in how far from the firm to the first farm? 16 How many farms will be loaded on that route before the 17 truck is full? What distance and how long does it take to 18 get from the first farm to the last farm on that route? 19 What distance and time does it take to get from the last 20 farm to the plant? How long does it take to unload and 21 wash the tank? And how long does it take to get back to 22 the garage? Does this truck and/or driver, make more than 23 one route in a day? Are there any tolls or fees along 24 that route? Does the loaded truck switch drivers or 25 tanks? Overtime pay is calculated for any employee on a 26 route that exceeds the normal workday time. 27 We sample from a variety of the engineered firms, 28 including small one- and two-truck haulers to very large 10195 1 fleets. We also assemble a variety of routes from 2 multiple small farm pickups to switching trailers at very 3 large farms. We also look at short routes close to 4 plants, to longer haul routes with a distant plant. We 5 also calculate plant-to-plant hauls of intermediate 6 products, like cream or skim milk, and we can look at 7 distribution costs from plants to population centers. 8 Dozens of our example route costs are used to 9 estimate a nonlinear function of hauling costs per mile 10 based on the length of the route, driver wages, and fuel 11 costs per gallon. 12 As I mentioned before, those items, the wages and 13 fuel costs per gallon, can, and do, vary by region of the 14 country. 15 Cost of transportation differs for bulk raw milk 16 or fluid intermediate product, refrigerated, and 17 unrefrigerated trailers. Costs differ regionally by fuel 18 and labor wages. Road weight limits are restricted to the 19 most constrained state that the route passes through. 20 For example, Michigan has the least restrictive 21 weight limits, and it allows gross vehicle weight of 22 164,000 pounds. However, if the truck passes from 23 Michigan into Indiana or Ohio, the gross vehicle weight is 24 now restricted to those state limits of 129,400 pounds. 25 If the truck further passes into Illinois or Pennsylvania, 26 their gross vehicle weight limit is only 80,000 pounds. 27 The model can take a cost advantage of the supertankers 28 within a state like Michigan and other high-gross vehicle 10196 1 weight states, but if the route crosses into a more 2 restricted state, then that more restricted state's limit 3 becomes the limiting weight. 4 The model's task -- and this is in bold in here -- 5 is to minimize the costs of milk assembly, dairy product 6 processing, and final product distribution while 7 respecting all of the constraints. 8 Just as a side note, cost minimization yields the 9 same outcome as profit maximization in a perfectly 10 competitive market. 11 The model's primary solution is one of physical 12 flows, as in X pounds of milk were shipped to plant Y and 13 made into Z pounds of that product, A, which was then 14 distributed to distribution points or consumption points, 15 I and J. This is referred to as the "primal" solution. 16 An optimization model like the USDSS can also 17 express the "dual" solution, which is in terms of dollars. 18 A dual, sometimes called a "shadow price," really tells us 19 how much could be saved if a constraint was relaxed by one 20 unit. 21 If you think about a fluid plant, you could ask, 22 how much would the next 100 pounds of milk at that 23 location be worth if it just showed up at the plant? That 24 relaxed constraint may let the model move milk and dairy 25 products around the country in a different way that saves 26 the entire system some money. That is what the dual 27 values are at fluid plants, that's what they are 28 reflecting. 10197 1 Another interpretation of that dual is "at what 2 price would the processor at that location be indifferent 3 to receiving the next hundredweight of milk"? If you were 4 asking more than that amount, the model knows that it 5 could go elsewhere and procure milk from another source 6 for less cost. 7 A shadow price is calculated for any constrained 8 value in the model. We are usually only reporting on the 9 values at Class I fluid plants, but there are also values 10 for the other classes. Further, there are dual values for 11 farm milk at all locations. 12 It should be noted that these dual values will be 13 qualitatively related, but not equal across the country or 14 even within proximity to one another across different 15 constraints. That is, farm value of milk will be somewhat 16 different to a nearby plant value for milk based on what 17 the model can do with another unit. 18 The optimization model can only report dual values 19 at points of constraint. For example, Class I dual values 20 are only calculated at fluid milk plants. Values in 21 locations where there is no fluid plants are being 22 estimated post-processing model solution with a geographic 23 interpolation known as "Kriging," or a Gaussian process 24 regression. A raster image -- 25 THE COURT: Could you stop and spell those terms 26 for us? 27 THE WITNESS: I would. Kriging is named after the 28 man who invented the process called Kriging, 10198 1 K-R-I-G-I-N-G. And Gaussian process regression is 2 G-A-U-S-S-I-A-N, P-R-O-C-E-S-S, R-E-G-R-E-S-S-I-O-N. 3 A raster image is -- 4 THE COURT: And, again, help us with that. 5 THE WITNESS: Sure. A raster is spelled 6 R-A-S-T-E-R, which is just an infinite number of points in 7 an area that can be differentiated by color. When we see 8 things like heat maps for temperature differences and 9 things like those, those are raster maps. 10 A raster image is created which estimates a 11 weighted value from the nearest 12 points (known as the 12 dual values at Class I plants). This value is weighted by 13 distance from the point of interest. The smooth surface 14 (commonly referred to as a "heat map") can then be 15 outlined by isoclines -- that's I-S-O-C-L-I-N-E-S -- which 16 are lines of equal value, or the values of the raster can 17 be projected back down onto a geographic area, like a 18 county, and the average of those values can be calculated. 19 That's what we are doing, the county values, and 20 rounding them to the nearest $0.10. This also explains 21 why we might not see the minimum value reported in the 22 county values that were attached to Dr. Nicholson's 23 documentation, because the average of the raster points -- 24 one of them will contain a zero value -- but the average 25 of the entire county may not average to zero, but might 26 round to $0.10. We then add a fixed value, as specified 27 by the group asking for the model values, to get our 28 Class I values. In recent years, that amount has been 10199 1 $1.60. 2 We do try to be responsive to the concerns and 3 observations of the folks looking at model results. It 4 has been these comments that have pushed us to refine the 5 model over the last 30 years. For example, recently, in 6 these model runs, it was observed that the spatial values 7 in Michigan seem as though they wouldn't move milk in the 8 way it was needed. 9 Further reflection of the USDSS plant shadow 10 prices appeared as expected, but the county interpolation 11 values were not. We realized that the Kriging algorithm 12 was using points in Western New York and in Wisconsin as 13 being the nearest 12 plant locations. The Great Lakes are 14 not navigable by tanker truck, and we needed to make a 15 change to our post-processing estimate. This was done by 16 constructing a geographic fence down the Great Lakes that 17 the Kriging algorithm must go around. When that was done, 18 Michigan's county values looked appropriate, and our 19 thanks go to that NMPF committee for pointing out a 20 shortcoming of the model, which has now been corrected. 21 Any model is a simplification of reality, but in 22 my opinion, the USDSS model is the most complete and 23 systematic means that we have of considering spatial milk 24 values across the country. We have been developing this 25 model for more than 30 years. Over that time we have 26 refined the model and made it much more sophisticated, and 27 we have addressed concerns that folks have expressed 28 through many iterations. 10200 1 For instance, the model now accounts for milk and 2 dairy products at the component level and not at the milk 3 equivalent value that it was originally built around. The 4 cost of transportation are quite detailed and include 5 things like tires and insurance, capital replacements, 6 fees, et cetera, as well as fuel and labor that differ by 7 region of the country. 8 If we're going to have questions or concerns about 9 the model results, we need to talk about the inadequacies 10 of the model structure or about the quality of the data 11 that was used in the model. The rest of the results are 12 just math, which I believe are being done correctly. 13 The model does not include items such as 14 restriction of bridges and tunnels during certain hours of 15 the day. This can add legitimate cost to servicing an 16 area like New York City. We have not incorporated 17 instances like the bridge and tunnel example because the 18 added complexity may not be worth the effort. But this is 19 a place where professional judgments might be made that 20 would supersede model results, and this kind of price 21 alignment may alter the dual values by nickels, dimes, or 22 possibly quarters over small areas. 23 The model could be further refined in many ways. 24 Currently we identify 20 final dairy products and 11 25 intermediate products, dairy products that can be used in 26 the manufacture of other final dairy products. Our fluid 27 milk category include both conventional fluid products, as 28 well as other products like organic, A2, lactose-reduced, 10201 1 et cetera. These could further be broken out into 2 separate products, but the Federal Milk Marketing Order 3 recognizes all of these as Class I, and hence, our 4 aggregation into the one category. 5 There are other items that the model does not 6 consider, including the Federal Milk Marketing Order 7 regulation itself. There are important reasons why the 8 model does not, and we believe should not, consider this 9 regulatory system. The USDSS was designed to inform about 10 an efficient marketplace, oftentimes for the purposes of 11 developing regulations themselves. Imposing those 12 regulations on the model could cause a departure from 13 market efficiency, which is not the stated goal of Federal 14 Orders. 15 Larger value changes imposed over larger regions 16 suggest a significant shortcoming in the model structure 17 or data. Such shortcomings should be brought to the 18 attention of the researchers for correction in current or 19 future model use. We have gladly and willingly considered 20 changes in the past which have resulted in improvements 21 employed in the present model iteration. 22 If we were requested to rerun the model with 23 larger value changes imposed over larger regions, we would 24 need to understand the reasons for the change so we could 25 adjustment the model to assign this additional cost. We 26 also would want to run the model with those changes so 27 that we could ensure the surrounding counties and states 28 adjust appropriately to changes in a certain area. We 10202 1 have not been asked to do that. 2 The USDSS model results in an "efficient" market. 3 That is, milk movements are optimal to achieve the lowest 4 cost to the system. Any market will have some friction 5 which results in departure from the optimal solution. 6 This can happen when there are contractual obligations 7 between parties which move milk from regions to plant 8 where the model would rather access other supply 9 locations. 10 In my opinion, the price surface represented by 11 regulation should reflect an efficient market and not have 12 market inefficiencies hardwired into them. A minimum 13 price regulation allows higher prices to accommodate 14 inefficiencies while encouraging and rewarding movement 15 toward a more efficient solution. 16 The price surface of the USDSS model reflects an 17 economic current which is analogous to an ocean current. 18 It's possible to move against the current, but it's more 19 difficult, and the current will try to move product in a 20 market efficient direct. 21 Price differences from any two points in the model 22 will not cover full cost of transportation. If price 23 incentives greater than full costs occur in the model, 24 then more milk than is needed would be enticed to move to 25 capture the rewards. The price surface reflects 26 incentives to move milk in the direction of greatest need. 27 Just as a thought experiment, consider a farm that 28 is located 100 miles from two processing plants: One of 10203 1 the plants is west of the farm and one is southeast of the 2 farm. The 100-mile hauling charge is the same to supply 3 either plant, but the plant to the west has a zone price 4 that is $3, and the plant to the southeast has a zone 5 price that is $3.10. The farm should choose to sell milk 6 to the southeast plant to net a larger price. This moves 7 milk in the efficient market direction. 8 Contractual obligations that move milk in a 9 non-optimal or non-market efficient way can and do happen. 10 This is not disallowed in Federal Milk Marketing Orders, 11 but in my opinion, it shouldn't be encouraged either. At 12 a micro level, there's been a criticism of multiple milk 13 haulers driving past farms, carrying milk from farms to 14 plants. The additional cost to the system for this 15 behavior has been voluntarily reduced by swapping farm 16 milk loads going to plants and only having a single hauler 17 traversing the roads. If the original contractual 18 relationship had been reinforced in the regulations, the 19 firms would not have had the incentive to find more 20 efficient solution of doing a swap. Market inefficiencies 21 can and do move toward a more efficient market with 22 economic incentives. Without incentives, the markets will 23 not achieve efficient milk movements. 24 Dr. Nicholson and I have worked hard to provide a 25 sophisticated and detailed analysis of efficient milk and 26 dairy product flow movements that I hope will be of use to 27 the participants in the industry in this proceeding. I 28 would be happy to answer any questions or provide any 10204 1 further insights into the model's design and outputs so it 2 can be of the highest possible use for these proceedings. 3 THE COURT: That is excellent, and I would like a 4 five-minute stretch break before we begin questions. 5 So please be back and ready to go at, let's see, 6 5:28 -- I mean, 3:28. 3:28. 7 (Whereupon, a break was taken.) 8 THE COURT: Let's go back on record. 9 We're back on record at 3:28. 10 Who first has questions? 11 CROSS-EXAMINATION 12 BY MR. ROSENBAUM: 13 Q. Steve Rosenbaum for the International Dairy Foods 14 Association. 15 Good afternoon, Dr. Stephenson. 16 A. Good afternoon. 17 Q. You and I may both be of enough the same age so 18 that we remember the Prego spaghetti sauce commercials 19 from about 30 years ago, where the question was always, 20 "Well, what about tomatoes? It's in there. What about X? 21 It's in there, it's in there." So I'm going to ask you 22 about a bunch of questions about your model, and you can 23 tell me, yeah, it's in there, so -- 24 A. Or I might not. 25 Q. Or you might not, is the -- exactly. Well, that's 26 the question, that's why I'm asking the questions. I 27 really just have a couple. 28 I know that there have been restrictions placed as 10205 1 to how many miles or hours -- I think actually, it's not 2 miles -- hours truck drivers can drive. 3 Is that included in your model, those constraints? 4 A. When we do the economic engineering of the hauling 5 costs, which is an input to the optimization model, we do 6 include that type of thing. So there is a place in that 7 economic engineering framework where we can say that a 8 milk load is assembled and maybe hauled to a different 9 point and the driver is switched out. So we know that 10 that happens with drivers, and so we have example routes 11 that -- that do have that kind of thing occurring. 12 Q. Okay. And then what about -- I assume you 13 don't -- the model doesn't assume that the milk gets 14 unloaded instantaneously when a truck shows up at the 15 plant. 16 Is there some time, in fact, that's built in for 17 that? 18 A. There is. And I think I mentioned that in here, 19 that one of the questions that when we're assembling those 20 routes is, how long does it take you at the plant to 21 unload and be washed? So, you know, there's leaving from 22 the last farm to the plant, how far is it? How long does 23 that take? And then there's, how long are you at the 24 plant before you get turned around and that truck is 25 capable of being reused for perhaps a second route that 26 day, or just back to the garage? 27 MR. ROSENBAUM: That's all I have. Thank you. 28 THE COURT: Who goes next? 10206 1 CROSS-EXAMINATION 2 BY MR. MILTNER: 3 Q. Good afternoon, Dr. Stephenson. 4 A. Good afternoon. 5 Q. My name is Ryan Miltner. I represent Select Milk 6 Producers. 7 You made a reference to the base differential, or, 8 as you say on page 4, "A fixed value as specified by the 9 group asking for the model values to get Class I values." 10 In recent years that amount has been $1.60. When 11 you -- in your referring to in recent years, does that 12 include the model run that was presented by National Milk 13 for their Proposal 19? 14 A. I believe that in the spreadsheets, it may be this 15 very large one that's here, there are a couple of columns 16 that say USDSS May results and October results. Those do 17 include the $1.60. 18 Q. So if in those columns that show USDSS results, if 19 the lowest number is $2.20, does that mean that the model 20 has a value of at least $0.60 at every point? 21 A. No. We -- we added $1.60 as a uniform price to 22 the shadow prices everywhere. There will be at one or 23 more locations a zero shadow price value. That simply 24 reflects the fact that the model can't do anything with 25 the additional milk at this location, it has no extra 26 value to the model. So in our USDSS results, the smallest 27 amount value that you should expect to see there is $1.60. 28 But the 2.20 that you are suggesting, those were 10207 1 the price alignment results, I believe, that had come from 2 the National Milk committee. 3 Q. Do you recall off the top of your head where you 4 would expect to see the $1.60 values, or the zones where 5 there essentially was no incremental value? 6 A. Yeah. You can imagine that those locations are 7 going to be in places where there's a fair amount of milk 8 and not much population. If I remember correctly, we had 9 a location like that in Idaho, and we had a location like 10 that in North Dakota, I believe. I'd have to go back to 11 look to make sure. 12 Q. Okay. 13 A. If you look actually in that large table of 14 results, in one of the months, the smallest value that you 15 find is not $1.60, it's actually $1.70. And I did try to 16 explain that in here, because the model actually gives us 17 infinitesimally small point values. So at some point in a 18 county there is $1.60 value or a zero shadow price value 19 at that location, but by the time the Kriging algorithm 20 finds the 12 closest plants to that and does its 21 interpolation, it may round to an additional dime. 22 Q. I wrote myself a question here, and I can read the 23 words, but I don't know what I'm trying to convey. 24 THE COURT: If I might just ask? 25 MR. MILTNER: Please. 26 THE COURT: Are 12 closest plants always used even 27 if one suffices? 28 THE WITNESS: We always look for 12 closest plants 10208 1 to fill in the areas where there is no plant. The Kriging 2 algorithm provides weighted average values. So, for 3 example, you're closer to me than Ryan is, and it would 4 give you greater Kriging, as that should be the case, than 5 it would Ryan. 6 THE COURT: And that's a good explanation. 7 BY MR. MILTNER: 8 Q. Now, what if we were at the previous location 9 where my podium might have been closer to you than the 10 judge? 11 A. You could aspire to judgeship. 12 Q. That I could. 13 And I have since figured out what I want to ask. 14 Before I get to that, on page 5, and a few other 15 places throughout here, you -- you are using a -- you are 16 referring to "we" working in the model. 17 And is that just broadly speaking, you and anyone 18 else that's working with you on the model, including 19 perhaps Dr. Nicholson, or was there something else you are 20 conveying there? 21 A. Dr. Nicholson and I have exclusively worked on 22 this round of the model. In times past, we have had 23 graduate students that have also worked on different 24 aspects of the model, but, you know, it's been the two of 25 us primarily this time. 26 Now, I shouldn't say that. I have also had the 27 benefit of working with geographers who helped us to do 28 mass calculations of road distance networks, that when we 10209 1 went from a multi-county area to individual county areas, 2 it was a great mushrooming of the data that were there. 3 So they provided help to us in terms of understanding how 4 to do that with the software, but they didn't do the 5 calculations. 6 Q. Thank you. 7 When the model was first developed and utilized 8 during order reform, during that notice and comment 9 rulemaking phase? 10 A. Before that, we had aspects of the model. There 11 existed a Northeast Dairy Sector Simulator before we had 12 the U.S.-wide one. But the U.S.-wide one was built and 13 constructed for the Federal Order Reform in that 1998 to 14 2000 time period. 15 Q. And during the Federal Order Reform process, 16 there's some language in the Federal Register that 17 explains, I would say in very broad terms, that the model 18 results were adjusted by USDA to reflect certain -- 19 certain things. Certain witnesses throughout the course 20 of the hearing have testified about that. 21 And I wondered if you are able to shed any light 22 on what specifically was changed in order reform between 23 the model's results and the differentials we ended up 24 with? 25 A. The changes that were made were done internally. 26 For example, National Milk had their room of folks or, you 27 know, by regions, that assembled their professional 28 judgment, you know, to make the changes here. In Federal 10210 1 Order Reform, USDA had their group that did that. And so 2 it was a group of -- internally to AMS that made the 3 changes to the model structure last time. 4 And, you know, those were relatively small changes 5 in there. But, you know, again, I always suggest that 6 professional judgment will or may need to be used in some 7 locations to capture things that the model simply didn't 8 capture and wasn't built to capture. 9 Q. Do you have an opinion as to whether the changes 10 that happened during order reform between the model and 11 regulation, how that looks compared to the changes between 12 the model run and what we have in Proposal 19, if there's 13 a similarity in magnitude or effect? 14 A. Ryan, are you talking about what we actually did 15 versus what is being proposed in 19, or are you asking me 16 to go back to 2000 model results and make comparisons? 17 Q. I suppose I'm asking a little bit of both. And 18 what I want to ask is, the changes between the model run, 19 which was used as the base for Proposal 19, and what 20 Proposal 19 looks like, are those changes similar in 21 magnitude to the type of adjustments that were made during 22 order reform? 23 A. Oh, boy. I don't really recall the magnitude of 24 all the changes that were made back in order reform. But 25 it was my memory that, you know, they were relatively 26 small changes, as I said, something like nickels, dimes, 27 quarters, and they were in places where a Market 28 Administrator simply said, there's something here that 10211 1 wasn't considered in the model that really ought to be 2 considered. 3 And, you know, the example I used of New York City 4 bridges and tunnels and restricted times of deliveries, 5 those kind of things can and will make real dollar 6 differences to the transportation issues that you have in 7 some areas. 8 Q. You also noted there have been a number of 9 improvements to your model since it was first rolled out, 10 I suppose, and now. And you believe it is more accurate 11 today, correct? 12 A. Yes. 13 Q. Do you believe that a -- I'll call it a raw model 14 run to optimize for milk movement today is a more accurate 15 reflection of absolute efficiency than you would have 16 achieved in 1999, per se? 17 A. Yes. There's no question about that in my mind. 18 A good example was, back at that point in time, the model 19 solved in terms of milk equivalents, it didn't have 20 components. 21 And the other thing to recognize is that because 22 of that, you had places where you had to have surplus 23 components and just recognize that we were going to freely 24 dispose of them in that area. You had to have more than 25 was needed or the model wouldn't solve. And when we 26 achieved that, then there would always be a small amount 27 of surplus in some regions. 28 What we were able to do, and this happened after 10212 1 model reform, before -- we got down to the component level 2 and we said, we have to mass balance and account for 3 components at the farm and the plant and the finished 4 product level. So we now have about 2% surplus, which 5 doesn't have to be there, it just is, but that's about 6 what we -- and it's a little different, 2% for fat and a 7 little less than that for solids nonfat. But that's what 8 we attribute to shrink in the system. 9 Q. And so when you are doing that type of mass 10 balancing, I assume that also helps to take into account 11 the effective capacities of the plants that are in the 12 model? 13 A. Yes. We have what we assume are the plant 14 capacities from our knowledge. Now, we don't have 15 intimate knowledge, you know, of every plant, but we have 16 our own point estimates of what we think that plant is 17 capable of doing, and we give it a plus or minus 10% on 18 each plant. So that plant can process a little bit more 19 than our point estimate is or we can constrain it so that 20 it must be at least that 10% lower value or above. 21 We also have, as I said, a number of plants where 22 we don't have capacity knowledge, but that's only about 23 10% of milk volume that's unaccounted for in the U.S., and 24 we let the model choose the plant location as long as it 25 can make the products that are needed there at the sizes 26 that are necessary. 27 MR. MILTNER: Thank you very much, Dr. Stephenson. 28 // 10213 1 CROSS-EXAMINATION 2 BY MS. HANCOCK: 3 Q. Good afternoon, Dr. Stephenson. 4 A. Good afternoon. 5 Q. Did you -- 6 THE COURT: I know he knows who you are, but I'd 7 like you to identify yourself for the transcript. 8 MS. HANCOCK: Nicole Hancock with National Milk. 9 BY MS. HANCOCK: 10 Q. Did you work with Dr. Nicholson in compiling this 11 document? 12 A. In compiling his? 13 Q. In compiling Exhibit 438. 14 A. No, not specifically. Although I did author the 15 document that was given to National Milk Producers 16 Federation. 17 Q. For Dr. Nicholson? 18 A. Yes. 19 Q. And you didn't include this in that document that 20 was used? 21 A. With this document here? 22 Q. Right, these details. 23 A. Not all the details were probably there. And 24 that -- I had heard questions about a number of these from 25 observing on the Zoom conferences that we have had 26 available to us, and so I tried to make bullet points of 27 those things which appeared to be coming up again. 28 Q. Okay. And you listened to Dr. Nicholson's 10214 1 testimony? 2 A. I did. 3 Q. Okay. How come you didn't vet this with him 4 before you finalized it? 5 A. Well, first of all, there are a number of the 6 things that, as Dr. Nicholson indicated, you would have to 7 ask Mark, for instance, those would be such things as the 8 cost of transportation portions of it, the Kriging 9 algorithms, those types of things. The mapping that was 10 done. 11 Dr. Nicholson, I felt, did a very good job of 12 explaining the basic construct of the model and what was 13 run and done. But most of these were things that was not 14 in Dr. Nicholson's testimony and were the kind of 15 questions that have been reflected back to me. 16 Q. Okay. And is there anything that you can recall 17 that Dr. Nicholson testified to when you were listening 18 that was inaccurate or incomplete, other than the items 19 that he referred to you? 20 A. Not much. There was one thing that I thought, 21 "Chuck, you probably better check the data on that," that 22 he had referred to, but that's with future testimony and 23 not with current. 24 Q. What was that pertaining to? 25 A. He had talked about the fact that shadow prices he 26 felt should be close to the same at the same location. 27 And in point of fact, when you go in and take a look at 28 the data from the model runs, they can be different and by 10215 1 a fair amount. 2 Q. Okay. And that just depends on the location and 3 the factors that go into establishing that spatial 4 difference? 5 A. And the products and -- and just a host of other 6 things. But, yes, you are correct. 7 Q. Okay. And you -- you said that you're here today, 8 you are wearing a neutral hat; is that fair? 9 A. Yes. 10 Q. What does that mean? 11 A. Pardon, what does -- 12 Q. Yeah, what does that mean, that today you are 13 neutral? 14 A. Well, when -- I come here, I don't want to be a 15 proponent of any one proposal or not unless I'm explicitly 16 saying that that's the case. 17 When I have come to testify on Make Allowances, 18 for example, or on something like this, this is to provide 19 what I think are questions, answers to questions that 20 people may have from work that we have done. I'm not a 21 proponent or an opponent of any of the proposals that have 22 been here. 23 Q. Okay. And you contrasted that with when you 24 testified on Make Allowance. 25 Is that because in that role you were being paid 26 by IDFA? 27 A. Yeah. You -- I can ask for compensation, I feel. 28 But, you know, if you had a particular desire for any 10216 1 answer, I will never guarantee somebody that. I will 2 guarantee that I will do the work and I will provide you 3 with -- with an answer, and you can decide if that's what 4 you want to use or not. 5 Q. And I appreciate that. I recall when we had the 6 examination on Make Allowance that you were very 7 transparent, even to the detriment of IDFA, so I 8 appreciate that. 9 My question, though, was I was just clarifying 10 when you said today you are neutral, is that -- did you 11 mean that it's because today you are not being paid by 12 anyone to be here? 13 A. No. Although, I could hope on my poor fixed 14 retirement income that maybe my mileage and room is picked 15 up. 16 Q. And who would you expect to pick that up for you? 17 A. I would expect the MIF group to pick that up. 18 Q. Is that MIG? 19 A. Or MIG, sorry about that, yes. 20 Q. That's okay. 21 And, in fact, Mr. English's -- his law firm is the 22 one that submitted your testimony for you; is that right? 23 A. Yes. 24 Q. Okay. I want to ask just a couple questions so 25 I'm clear on this. 26 The hours that Mr. Rosenbaum just asked you about, 27 you said that the model does account for the limitations 28 on a driver's number of hours. 10217 1 Do you recall that? 2 A. Yes. 3 Q. And you said yes. And if they -- if it -- if it 4 looks like they are going to exceed, then it picks up a 5 new driver; is that right? 6 A. We have routes that are modeled that are multiple 7 driver routes. So, yes, that's correct. 8 Q. And that doesn't take into account things like 9 traffic congestion, does it? 10 A. No, it doesn't. 11 Q. And it doesn't take into account things like 12 weather events that might stop a truck from proceeding but 13 burn up some time for a driver? 14 A. No, it doesn't account for any of those possible 15 one-off events. We try to account for what we think 16 whatever, in air quotes, "normal" might be. 17 Q. And it doesn't account for things like labor 18 shortages or drivers being unwilling or unavailable to be 19 hired for different roles to transport milk, does it? 20 A. Not directly. Although we did see a pretty 21 substantial increase in driver wages with the 2021 data 22 that we had in here, and part of this was just trying to 23 attract over-the-road truckers, you know, milk haulers, 24 that wage rates increased fairly substantially from 25 previous times we have run the model. 26 Q. And it doesn't account for the Federal Order 27 regulations? 28 A. No, it doesn't. 10218 1 Q. It doesn't account for commercial relationships 2 that exist, such as contracts where milk is required to be 3 delivered to a farther away plant? 4 A. No, it doesn't do that. It -- it -- it will 5 absolutely meet the needs of consumers by processing all 6 the dairy products that we think are being demanded and 7 consumed, and it will source that from milk from farms and 8 run that through plants. But, no, we -- we do not 9 identify those relationships. 10 Q. And it doesn't -- it doesn't account for plant 11 efficiencies or inefficiencies either, does it? 12 A. It doesn't. We do have some scale efficiencies in 13 the plants. So in our observations over the years of 14 plant costs of processing, we do see that larger plants 15 tend to have lower costs, but that's not absolute. We 16 have some small plants that are very competitive, and the 17 large plants that probably could improve. 18 Q. And it's not just the cost built into those 19 efficiencies, but also the effect that that would have on 20 the volumes that the plant can take in as well and 21 process? 22 A. Well, again, we -- we try to respect -- first of 23 all, we don't make any decisions about how much milk is 24 produced and where. We take that as a given, a starting 25 point, that dairy farmers have made their decisions about 26 where their farm is located and how much milk they 27 produce. That's given. We don't take consumer demand as 28 a calculated value, we take that as observed. And we do 10219 1 the same thing with plants. Their locations, products 2 produced, and the approximate capacities of at least half 3 the plants. 4 Q. Okay. And it's fair to say that those kind of 5 boots-on-the-ground experiences that would deviate from 6 the most efficient movement of milk would be a relevant 7 experience to overlay the model results? 8 A. Well, I have to be a little careful about that. 9 You know, I guess the qualifications that you are making 10 there, I try to give some examples that I think, I believe 11 personally, that what we want to do with regulation is to 12 provide a goal that needs to be strived toward, not easily 13 accomplished, that these minimum prices should move milk 14 in the direction where it's most needed, not compensate 15 100% of the movements or necessarily respect those price 16 relationships, that if you are having to, as I phrased in 17 the document, swim a bit against the current, you know, to 18 get there, that over time we hope that that current 19 probably moves you toward a more efficient market and 20 proper movements. 21 This model does accomplish all of the needs of 22 dairy product consumption and moving milk from farms to 23 plants to accomplish that. 24 Q. In the most efficient way possible. 25 A. In the most efficient way possible. 26 Q. And -- and that also assumes full truckloads of 27 milk? 28 A. It does assume full truckloads of milk, or at 10220 1 least weight limit truckloads, yes. 2 Q. And the -- and which would mean if there's 3 multiple stops on a route, the commingling of milk between 4 various farms as well? 5 A. That's correct. That -- if -- I'm trying to 6 recall what one of the routes is. I think we have a 7 seven-farm pickup that I might just call hilly route or 8 something, because it -- it takes the better part of the 9 day to pick up that route and to get it to the plant to do 10 that. So, I mean, it's not a terribly efficient route, 11 but it's a route that has to happen. 12 Q. And you've actually talked about that, that you 13 have to use some of your professional judgments when you 14 are looking at these model results, and you said that it 15 can alter those values by nickels, dimes, and possibly 16 quarters; is that right? 17 A. Sure. I think that there are going to be places 18 where you could very well say the model was not 19 sophisticated enough to incorporate this particular 20 problem that we have in an area, and, you know, we need to 21 account for that. 22 But as I also mentioned, that if the problem is 23 large and systemic, I would really want to know about it, 24 because I do think that the model should try to capture 25 that stuff. It's awfully difficult for our humanness to 26 sit down with data and I think try to draw these things or 27 to try to make efficient routes. 28 We do what we do a lot of the time because it's 10221 1 what we have been doing, not necessarily what -- what we 2 ought to do. 3 MS. HANCOCK: Thank you for your time. Appreciate 4 it. 5 THE WITNESS: You're welcome. 6 THE COURT: Mr. English. 7 REDIRECT EXAMINATION 8 BY MR. ENGLISH: 9 Q. Dr. Stephenson, I have just a couple questions. 10 THE COURT: Please identify yourself, even though 11 we all know who you are. 12 MR. ENGLISH: I am Chip English for the Milk 13 Innovation Group. I thought about saying something funny, 14 but I couldn't come up with it. 15 BY MR. ENGLISH: 16 Q. So I want -- I just want to do a little math for a 17 second, and this has to do with the number of plants, and 18 then on page 1 you talk about the plant capacity you have 19 represents, I think you said 90% of NASS milk volume in 20 the 48 states. 21 But you said you think you have 50% of the plants, 22 correct? 23 A. Correct. 24 Q. But if 50% of the plants represent 90% of the NASS 25 milk, then 50% of the plants you don't have then 26 represents the 10%? 27 A. Yes. And, you know, I'm going to allow that the 28 volume that we have on those plants that I say account for 10222 1 90%, that's at our midpoint estimate on each of those 2 plants, and we do allow the plants to -- when we're doing 3 this, to be constrained plus or minus 10% from that. 4 Q. Thank you. 5 So then one other set of questions, or one 6 question. Earlier this week I asked some questions of a 7 witness from Michigan Milk Producers Association, and he 8 didn't know, or at least wasn't able to answer, I think, 9 the question of whether you and Dr. Nicholson consulted 10 with Michigan Milk Producers Association about closed 11 plants and new Class I or II operations. 12 Do you have any information on that issue? 13 A. We -- we did. We got a document from the National 14 Milk Producers group that had gone through our plant list 15 and had identified a number of changes that they felt 16 should be made. 17 And in Michigan, I don't remember the names of all 18 the plants, but there was one producer dealer, a 19 relatively small operation, and two relatively small ice 20 cream operators that -- that were added to the list as a 21 result. 22 Q. Thank you. 23 MR. ENGLISH: That's all I have. 24 THE COURT: Dr. Stephenson, what number of shifts 25 do you assume the plants utilize, and what days of the 26 week do you assume they work, or does that matter for the 27 model? 28 THE WITNESS: It -- it doesn't matter as much for 10223 1 the model, because the model is -- is solving for a 2 one-month observation. So we take the milk production in 3 the month, in this case May or October, and the demands 4 that we estimate for that -- those two months. 5 I should also say that one of the other items that 6 is a part of this model is the change in stocks, dairy 7 product stocks, too, because in that spring period of the 8 year, typically our stocks build, in the fall of the year 9 we draw on them. So those are additional supplies, I 10 guess, you could call them. 11 But when you are thinking about a monthly model 12 looking at those plants, we have an idea about how many 13 days a week those are processing, and sometimes about, you 14 know, the shifts. But that's -- that's just information 15 that we have that is not used in the model calculations. 16 What we do have is a weekly processing number for 17 each plant, and we then multiply that to get up to the 18 monthly value that the plant can process. 19 THE COURT: Did you hear the testimony about 20 certain processors wanting each load of milk delivered to 21 them to come from a single source from one farm because 22 that's how the customer wants to observe quality of milk? 23 THE WITNESS: No, I wasn't listening on that day. 24 I do know that that kind of thing does happen. And, you 25 know, the plants that are doing some maybe co-packing or 26 something may also have to be sure that we run this milk 27 that's coming in the first thing on Monday, you know, when 28 we're starting the plant up, and after that we run our 10224 1 more conventional milks. Those are added costs in a plant 2 as well, but we don't account for those. 3 THE COURT: There were a number of questions about 4 how much of the increased feed costs are in the model. I 5 have no idea what all feed costs are, but there was 6 uniform recognition that feed costs have escalated lately. 7 THE WITNESS: Yes. It's my understanding that 8 they have as well. But I can tell you with certainty that 9 there are zero feed costs in the model. 10 We don't try to calculate what supply might do or 11 react to. We are simply looking backward and saying, on 12 this month of this particular year, farmers chose to 13 produce this much milk, regardless of what the costs were 14 or anything else. So we -- we take milk supply as a 15 given. 16 THE COURT: How does the transportation cost, the 17 hauling costs, take into account hours lost in traffic? 18 THE WITNESS: It doesn't, except the economic 19 engineering model that we use on there has these steps 20 that are ordinary for a plant and a particular route, and 21 if it's typical that that route runs into traffic, you 22 know, on a daily basis, then that's something that is 23 asked there. If it's unusual, if there was a traffic 24 accident or something else of the sort, we don't capture 25 that and don't have any real way of doing that. 26 THE COURT: We had a witness this week whose 27 deliveries are primarily to the north, in a part of the 28 country that incents transportation of milk toward the 10225 1 south. He complained of the differentials going north 2 compared to what they would be if they went south. 3 Do you have any comments about that? 4 THE WITNESS: Well, he apparently has business 5 relationships that are asking him to service a customer 6 that, again, would be going against this economic current 7 where we, in general, are short of milk and dairy products 8 in the South and Southeast in particular. And as I said, 9 the model doesn't preclude that that could be done, but it 10 wouldn't necessarily want to try to incentivize those 11 either. 12 I mean, if this were the normal thing, if we 13 expected that to be happening on a widespread and regular 14 basis, then we would try to look -- the model would find 15 its most efficient way to service that particular market. 16 Apparently, the model is not doing that because those 17 price relationships are incentivizing milk to move to the 18 South and Southeast. 19 And perhaps I should say, you know, when we look 20 at what's changed over time, or what are the factors that 21 the model results would provide or would yield today that 22 might have been different than it was 20-some years ago 23 when we were doing it, or in the intermediate time period, 24 there are some pretty large things that can move this 25 model. 26 But it takes pretty large things to get very 27 different results from it. One of them is population. So 28 do we have more people in a given area or do we have less? 10226 1 What is the per capita demand for dairy products, is it 2 greater or is it declining, as we have seen with fluid 3 milk? Since about 2010 we have had declining per capita 4 consumption -- or total consumption per capita has been 5 declining since before that. So there have been a number 6 of things that are changing the model. 7 And the other thing is, where are farms choosing 8 to locate to produce milk? The Southeast is a difficult 9 region, partly because cows now produce very much more 10 milk than they did on a per-cow basis, and a byproduct of 11 being something equivalent to a high-performing athlete is 12 that there's a lot of excess body heat that has to be 13 dissipated. Harder to do in those hot and humid climates. 14 So we have seen a lot of more growth in some of the either 15 dry or regions -- or the cooler regions of the northern 16 tier states. 17 So milk production changes, demand changes, and to 18 some extent plant locations change. 19 THE COURT: When you consider the population, do 20 you consider areas in which there's a large tourist influx 21 that increases the demand in fluid milk at times? 22 THE WITNESS: No. But we do look at the -- well, 23 I shouldn't -- no. The real answer is no. We don't look 24 at the tourist influx. 25 If -- if that tourist influx does cause a change 26 in the quality of milk that is required, in other words, 27 the butterfat levels, you know, change because we have 28 people from different parts of the country that are down 10227 1 there, then that would be captured because we do look at 2 the AMS sales data in the regions to better understand 3 that. 4 But, no, the per capita -- the number that we use 5 for fluid and other dairy products are per capita on a 6 national basis, and we multiply that by the population 7 that has residency there. 8 THE COURT: If you're transporting to a large 9 population, does it make a difference in the model whether 10 what's being transported is fluid, raw milk, or packaged 11 milk product? 12 THE WITNESS: It does make a difference. The raw 13 milk bulk transport, or even bulk transport of skim milk 14 or cream from a plant, has a different cost per mile than 15 a refrigerated boxcar does, or that a non-refrigerated dry 16 car does for other dairy products like either nonfat dry 17 milk or something. So all those different product 18 possibilities do have different cost structures associated 19 with them. 20 The model also has non-linear components in the 21 transportation. And one of the things that the model has 22 actually projected for about 20, 25 years that we have 23 been doing this was that we might expect to see more 24 packaged long-distance movements of packaged fluid milk, 25 because the probability of a backhaul in a reefer or a 26 boxcar is much greater than it is for a backhaul in a 27 tanker, which essentially can't happen anymore. And that 28 does reduce the cost of transportation on that particular 10228 1 route or unit. 2 We do see more of that happening. For a while it 3 was a package limitation that packages of fluid milk just 4 didn't survive the jostle and jiggle of a long truck ride, 5 but we have managed to overcome that now. 6 THE COURT: Did you hear the testimony about the 7 complaints because the drivers' hauling hours are reported 8 by electronic equipment which is unreliable and fails? 9 THE WITNESS: No, I didn't hear that, and not sure 10 what I would do with it even if I did. 11 THE COURT: Who else has questions for 12 Dr. Stephenson? 13 I would ask for Agricultural Marketing Service 14 questions. 15 MS. TAYLOR: Your Honor, might I ask for a 16 five-minute stretch break? And in that time I'm going to 17 have my colleague hand the witness exhibit, in case 18 anybody else wants to look, 302, which was the testimony 19 from Dr. Nicholson a bit ago. 20 THE COURT: Very good. 21 So, please, we'll take a five-minute stretch 22 break. Please be back and ready to go at 4:15. And in 23 the meantime, we will obtain that exhibit. 24 We go off record at 4:10. 25 (Whereupon, a break was taken.) 26 THE COURT: Let's go back on record. 27 We are back on record at 4:16. Just turned 4:16. 28 Mr. English, while we were on break you mentioned 10229 1 that there's one function that you would like to do with 2 the exhibit, and I agreed with you. 3 Would you identify what that is? 4 MR. ENGLISH: Yes, Your Honor, I also discussed 5 this with Ms. Taylor who, because she mentioned an 6 exhibit, reminded me. 7 There is an exhibit that was marked earlier in the 8 hearing, Exhibit 291, that several witnesses have referred 9 to. I think it -- I think two attempts were made to admit 10 it, and it was objected to on authenticity grounds. 11 THE COURT: Say it again. 12 MR. ENGLISH: Authenticity grounds is my 13 recollection for the objection. And whether or not I 14 agreed with the objection, we just put it on hold, because 15 as it turns out, one of the two authors of 291 is on the 16 witness stand. And I would like to just clear up that 17 issue and see if we can get 291 admitted. 18 THE COURT: And do you expect to ask -- 19 MR. ENGLISH: I do not intend to ask questions. 20 Others may, having recalled it now, intend to ask the 21 witness questions. That's up to them. 22 But, no, I am merely trying to clear up for the 23 record a marked, not admitted document. I have confirmed 24 with USDA it has never been admitted. And so all I would 25 like to do is have USDA provide the official copy of 291 26 to Dr. Stephenson. If you need to see a copy, Your Honor, 27 I certainly have no problem with that. 28 It is merely, I hope, ministerial and quick. 10230 1 Quite frankly, I intended to do it, and I forgot. I was 2 not the person who submitted it, I think it was Edge. 3 It was under MIG-291, which is -- I guess that was 4 the same number as the exhibit. I believe it's actually 5 Exhibit 291 as well. 6 MR. HILL: It is. 7 MR. ENGLISH: All right. 8 THE COURT: I don't promise you we'll have time 9 for this, because I'm doing the other first, but I'll do 10 my best. 11 We're going to start with 302. 12 MR. ENGLISH: That's fine, Your Honor. That's 13 fine. All I want to do is have them authenticated and 14 admitted. 15 THE COURT: All right. 16 Now, I have in front of me Exhibit 302, also 17 marked as NMPF-36, and you may proceed, Agricultural 18 Marketing Service. 19 MS. TAYLOR: Thank you, Your Honor. 20 RECROSS-EXAMINATION 21 BY MS. TAYLOR: 22 Q. I think the -- well, first, Dr. Stephenson, I 23 would like to thank you for making the trip down here to 24 answer our questions. Glad to know somebody's listening 25 out there in Zoom. 26 302, which we just handed you, is the testimony 27 submitted by Dr. Nicholson weeks ago on the model you are 28 discussing today. 10231 1 You indicated to Ms. Hancock that you did help him 2 with this testimony; is that correct? 3 A. I didn't help him with this testimony. I had 4 written the original report that was submitted to National 5 Milk, and he drew heavily from that for many parts of 6 this. 7 Q. Okay. And I know you have answered some of the 8 questions we had on this, and I -- I'm probably going to 9 go through this with some additional questions since 10 you're here, so this will help you get to the same page 11 I'm on. 12 A. Okay. 13 Q. I think I'm first going to start, though, with 14 what you submitted today, which was Exhibit 438. We'll go 15 through this one first. 16 A. Okay. 17 Q. Okay. I promise I'm not going to take too long. 18 You mentioned that the model accounts for 19 components now, which is different than what it did in 20 2000. 21 Is that done for the fat-skim order as well? How 22 did you -- 23 A. We make the estimates for the fat-skim orders. We 24 are primarily concerned with the butterfat values that 25 differ by region, and we do -- have done -- regression 26 analyses to look at the relationship between butterfat and 27 other solids and nonfat, or solids nonfat in milk 28 production. So when we have the butterfat values, we 10232 1 estimate the non-butterfat components in milk. 2 Q. Okay. And you mention that your dairy product 3 plant spreadsheet, you estimated size capacity and 4 products produced there, and you drew that from another 5 number of sources, and it covers about 90% of the NASS 6 milk volume. 7 So that other 10%, you said you just let the model 8 allocate that by the constraint on the plant capacity at 9 plus or minus 10%. 10 Is that basically how it was handled? 11 A. Yeah. The plus or minus 10% gives bounds on what 12 those plants must do and may only do up to. What is left 13 is available to be processed by those remaining plants 14 that we don't have capacities for, and they are 15 uncapacitated. 16 So in other words, you know, those -- those plants 17 have to account for that additional 10%. And this is 18 largely across the country, so it's not like we see all of 19 that 10% of volume going into one plant that doesn't 20 really have that kind of capacity. 21 Q. So you have -- you have capacity limits on some 22 plants, but not all plants? 23 A. That's correct. Just because we don't have the 24 knowledge of, you know, the capacity at those plants. 25 Q. I just want to make sure I understand this. 26 So there -- the plants that are in your 27 spreadsheet that you don't have capacity numbers for, and 28 would you say -- and that's beca- -- is that the 10%? 10233 1 A. That's the volume accounts for about 10%. 2 I mean, we tried to, first of all, identify the 3 plant capacities going through our lists and using sources 4 that we have multiple sources. 5 And then there were still quite a few plants that 6 we didn't have information for, but out of curiosity we 7 wanted to see what was accounted for by the volume of the 8 plants, which we thought we knew something about. But it 9 turned about to be about 90% of the milk volume in the 10 country that was accounted for with those plants. And 11 even by region, when you look at dairy products reports 12 and they break out some regional totals of products, the 13 sum was very close, so -- 14 Q. Okay. In transportation questions, you do account 15 for the cost of supertankers. 16 And when you say "cost," that's the purchase cost; 17 is that right? 18 A. I'm sorry, Erin, I didn't quite hear what you 19 said. 20 Q. Sure. So for the transportation -- 21 A. Yes. 22 Q. -- portion, you say you account for the cost of 23 supertankers. 24 Is the cost the purchase price of the supertankers 25 or are you also including maintenance, et cetera? We have 26 heard some testimony that it's more expensive. 27 A. Yeah. We -- we've tried to have a sample of a 28 variety of different things in the plants, and that 10234 1 includes the different types of tractors that might be 2 used. Gee, I don't remember what it was, maybe a decade 3 ago we started to see different kinds of high-mileage 4 tires that were being used on trailers and tractors, so we 5 began to incorporate that and acknowledge those things. 6 And the cost of tankers is, in fact, one of the inputs. 7 Again, what we try to do is to build this up with 8 examples of what we think firms are, so a relatively small 9 hauling firm, relatively larger firm, what kind of trucks 10 do they have, what kind of routes do they have, what's 11 their labor source. And that includes things like the 12 tankers. So, yes, those large tankers are there. 13 Q. Okay. And another question on hauling. Can 14 hauling markets differ regionally in terms of market 15 structure? We have heard some testimony from producers 16 that have few options for hauling, maybe a hauler exited, 17 so they maybe have one option at this point in where they 18 live. 19 How might this -- but that might not be the case 20 for other producers in other markets who have multiple 21 options. 22 So is that reflected at all in your model? 23 A. Yes. I mean, we account at the county level for 24 the milk that's produced there. We do have a good idea 25 about, you know, how many farms that might be, but that's 26 not a part of the model structure in there. We just 27 account for the total volume of milk. And that total 28 value of milk has to get to a plant somewhere, and it has 10235 1 to be processed into different dairy products. 2 So the relatively small amount of milk in a single 3 county might go to a single plant that produces, you know, 4 single products. Is that going to replicate what actually 5 happens? Maybe not, but it is a possibility that it could 6 happen in the way that the model would project it does. 7 And if there's really only one option for a farm that you 8 have indicated somebody may have testified for, that's 9 probably what the model's going to do. It wouldn't try to 10 pick a far distant plant. 11 Q. But would the additional increased hauling costs 12 because of that reality be reflected? 13 A. That's reflected in there, sure. 14 Q. Okay. 15 A. The distance from that farm, or that center, or 16 that county at least, to the nearest plant, if it's, you 17 know, 20 miles, it's going to be charged for the hauling 18 cost for 20 miles. If it's 200 miles, it will be charged 19 for 200 miles. 20 Q. Okay. Let's see. In the next page you talk 21 about -- and I'm on page 3 of Exhibit 438. You did 22 address some of the questions we had on road weight 23 limits. 24 A. Is this page 3 of my testimony or -- 25 Q. Yes. 26 A. -- Dr. Nicholson's? 27 Q. Yes. Your testimony, which is marked as 28 Exhibit -- 10236 1 A. Okay. Yep. 2 Q. -- 438. 3 A. Yes. 4 Q. And I just want to make sure I understand, and 5 I'll use your Michigan truck example that goes into Ohio 6 or Pennsylvania that has a lower vehicle weight limit. 7 Does your model assume the lower weight for the 8 entire length of that route or -- 9 A. It does. 10 Q. Okay. 11 A. It may be possible for plants to unload a portion, 12 you know, of a supertanker on to something else, but I'm 13 not aware. I mean, maybe Carl could correct me if he's 14 here, that that actually happens. We don't. 15 We would just say that if your route that you 16 intend to take is going to pass through one of these more 17 constricted weight limit states, then the starting point 18 of that route can only be 80,000 pounds or 129,000 pounds. 19 Q. Okay. So in your transportation model that models 20 a bunch of different routes, how many routes is it looking 21 at? 22 A. We look at, oh, about 18 different routes, I 23 believe it is. It's either 15 or 18. I don't remember 24 off the top of my head, Erin. But those 15 routes or 18 25 routes are meant to be representative of the different 26 kinds of things, you know, that we do see. So they could 27 be picking up a large number of relatively small farms. 28 They could be picking up, you know, a single tanker at a 10237 1 very large operation, maybe multiple tankers at that 2 operation, but we try to cover the span of what's 3 observed. 4 Q. Okay. 5 A. And I should also say that once we have got those 6 costs to move 100 pounds of milk 100 miles down the road, 7 those estimates from those different routes and different 8 firms, we then do regression analysis to fit the curves to 9 see how they have changed. 10 Q. Okay. We have had some discussion about -- and on 11 page -- page 4 of your testimony in the middle bullet, 12 that's where you talk about kind of how you -- how it 13 works, is there's a differential at each plant location, 14 and then you use some regression to kind of fill in the 15 map. 16 And so there's 3,000-some counties in the United 17 States, so the model just looks to find an estimate at 18 each of the fluid milk plants; is that correct? 19 A. When we report this -- this is only Class I 20 values. We can do other classes. You know, we can do 21 farm milk values. There are a variety of things that 22 could be done. But we have only reported on Class I 23 values in the past. 24 Q. Okay. So there's been some questions about 25 Proposal 19 that say there's 3,000 counties, and 2,000 of 26 them under Proposal 19 are different than what was -- the 27 model says or -- for example, my numbers could be off, but 28 let's take that for an example. 10238 1 What I'm hearing and taking away from your 2 testimony is to be less concerned with the number 2,000 3 because there's some art to the computer of filling in the 4 areas and the counties in between the plants. 5 Would that be accurate? 6 A. That is correct. But I also hear that, you know 7 we -- we have made changes because we really need to get 8 milk from this milk supply region to this plant, and we 9 identified and we use this plant in the model. You may 10 drive past a whole lot of counties that don't have any 11 plant that it could drop it at, but that plant will have a 12 value that, you know, is where we are calculating from for 13 those other counties that don't have plants. 14 Q. Okay. 15 A. And that is a good point that you bring up, 16 though. That Kriging algorithm that we use is not used to 17 soften, change, morph, any of the actual values from the 18 model. It only uses actual model values to be able to 19 fill in places where we do not have plants. 20 Q. Right. And the only places you do have plants 21 that's producing more out of the model is the fluid plant? 22 A. That's all that we have reported on. 23 Q. Yeah. On your model, when you talk about the 24 transportation cost piece of it doesn't include 25 restrictions such as bridges and tunnels, you talk, too, 26 it doesn't look at traffic congestion beyond what would be 27 considered normal. 28 Does it include wait times at plants? We have 10239 1 heard discussion about how a hauler could have to sit at a 2 plant for a while before it can get a back on the road. 3 Does it capture that? 4 A. In the data that goes into calculating those route 5 costs from the economic engineering model, yes, it does 6 have the place to say how long are you at the plant to 7 unload and get washed and get out of there. So it does 8 include that in those kind of observations. 9 Now, to the extent that that may be just a serious 10 problem at some one plant or, you know, in a particular 11 region, I mean, more serious there than it is elsewhere, I 12 don't know that and I have not included that. But we do 13 include the wait time at a plant for unloading. 14 And, again, we -- we do have some months of the 15 year when it can be -- or days of the month, when it could 16 be a particular problem at plants. Thanksgiving is a 17 notorious time for a lot of plants to have truckers 18 sitting and waiting to get in and unloaded. 19 Q. Okay. Are there any other limitations to what 20 costs are included, other than the ones we just kind of 21 discussed that you think are things should be -- are still 22 part of the cost of transportation and getting milk to the 23 plant? 24 A. We tried to capture all of the ones that we can 25 think of, and the ones that we think are the major ones. 26 I have heard people say things like, you know, the spine 27 of the mountains, you know, down the western third of the 28 country is -- is a problem getting trucks over that. 10240 1 I haven't accounted for that, if it is, in fact, a 2 really big problem. My guess is it's maybe not as big a 3 problem as, you know, I have heard, but, you know, I have 4 no question that on a snowy conditions day, getting up and 5 through the mountains is a problem. 6 Q. On page 6, in the second bullet you say, "Larger 7 value changes imposed over larger regions suggests a 8 significant shortcoming in the model structure or data." 9 How would you define "large" as you use it? 10 A. Well, again, in my opinion, I think that 11 professional judgment changes can and should be made to 12 the model results. This model's a simplification of 13 reality. But if there are really large changes, and I 14 have given you some idea because I think that nickels, 15 dimes, and quarters, you know, are probably the kind of 16 thing we can look at and say, you know, that's possible, 17 but if it's, you know, in that quarters to more than that 18 range, why? What are we missing? If we are missing that 19 and it's really something that is impacting the way milk 20 and dairy markets can and should work, that ought to be a 21 part of the model. That would be, in my opinion, a pretty 22 big shortcoming. 23 Q. Would you say that maybe the business 24 relationships, the contractual relationships that exist 25 that are not accounted for in the model, would be one of 26 the reasons that people see these costs they say that need 27 to be increased, that you don't necessarily want to 28 incorporate into the model, according to your testimony? 10241 1 A. No, I don't have that in here. And maybe two 2 comments about that. 3 Number one, it would be very difficult for me to 4 know and understand all of that. Well, there are some 5 that, you know, I would certainly know and could put in 6 there. 7 But the other is that I'm not really sure that 8 just because they're a business relationship that -- and I 9 have a hard time moving milk, or selling milk, or making 10 product on there, that that should be incorporated into a 11 model like this. If it's not an efficient flow, then it 12 doesn't mean you can't do it, but it ought to mean that 13 maybe somebody else should. 14 And, you know, I tried to give the example of the 15 milk swaps that we saw happening in several regions of the 16 country. Those didn't happen immediately. It was a 17 complaint that we heard from a lot of farmers over quite a 18 long period of time, and pretty soon, as those began to 19 happen, you began to have more efficient movements of 20 milk, and everybody still kept their business 21 relationships. So I think that there are ways to do this. 22 Q. Okay. Thank you. 23 I want to turn to 302. And I'm sure I don't have 24 as many questions as I started with, so let me flip 25 through. 26 You talked about your plant data side, the demand 27 side, and you had National Milk look through this list, so 28 that's probably pretty up-to-date. 10242 1 What's the most recent supply data that you used 2 for the runs? 3 A. 2021. Oh, supply data. Yeah, 2021, right. 4 Q. Okay. 5 A. That was where we used the state level values. 6 And at the time we did the modeling, the NASS data that 7 allowed us to do that, I think had only been released a 8 month earlier or something to that effect, so it was 9 pretty fresh at the time. 10 Q. Okay. If we can turn to page 7 of Exhibit 302 in 11 Figure 4. 12 And if I remember Dr. Nicholson's testimony about 13 this, this isn't -- well, let me ask you: Are these dots 14 all of the plant locations that were in the model? 15 A. I would have to check to make sure if that were 16 the case. I think that this is an up-to-date one, but it 17 may not be. We did have a few additions on final model 18 run and a couple plant closures, so there may be, you 19 know, very few plants that weren't actually shown here. 20 Q. And is this -- would you -- are these just fluid 21 plants on these dots, or fluid and manufacturing plants on 22 those dots? 23 A. I believe these are -- these are not fluid, these 24 are all plants. 25 Q. All plants. 26 A. Yeah. No, Wisconsin doesn't have that many fluid 27 plants. 28 Q. And then if we notice dots in areas where there's 10243 1 not a plant there anymore operating, how does that impact 2 your results versus what's actually going on now? 3 A. We have looked at those specific questions, and we 4 have been asked about that before. So a good example was 5 a particular plant closing in Northern New York quite a 6 few years ago. And people were concerned about what this 7 would do to the value of milk there, farms would have to 8 have an increased hauling cost to get to the next possible 9 plant. 10 And so we ran a baseline model which had that 11 plant in there, its capacity, and we then shut that plant 12 off and said, you are not there anymore, that milk has to 13 go elsewhere, and then we compared what that actually did 14 to milk prices. And in this particular case, the question 15 was, you know, what does that do to farm milk prices? So 16 that's what we looked at was farm milk prices. 17 You can absolutely see what the color scheme we 18 used was red to green on there. The impact right at that 19 plant location to the value of milk for those farms, but 20 recognized that that's also a positive contribution to 21 some other areas that have now got milk, you know, that 22 they didn't have access to before. 23 But it's like throwing a rock in a pond. You see 24 these ripples that go out, you know, in the value of milk. 25 They are intense toward the center, and they get less and 26 less and less as they go out, and pretty soon it's a 27 relatively insignificant change. 28 Q. And might that be a reason one might look to make 10244 1 some changes from the model? As you indicated, USDA did 2 that in reform, took the model results and then -- 3 A. Well, we did make a few changes for the final 4 model runs that you have the results for, in that a couple 5 of plants were added, a few plants, I think four plants 6 were added, that were pretty sure to be online. Either 7 they had already broken ground and they were going to be 8 taking milk within a year or something to that effect, and 9 we did close a couple of plants that we knew were going to 10 be closed in the not-to-distant future. 11 So in -- you know, I do say that we tried to make 12 these runs for a very specific month and year, and we take 13 those things as given. We can make some changes because 14 by the time this is implemented, if indeed there are 15 changes recommended and voted on, you will probably want 16 to be dealing with the closest thing to the markets you 17 have then, not what existed a few years ago. 18 Q. Uh-huh. I want to turn to page 15 at the top. 19 You say -- there's a sentence, the top paragraph, the 20 second sentence from the end: "Moreover, the model 21 results are not sensitive to changes of plus or minus 5% 22 in demand values or estimated transportation costs." 23 I just want to make sure the record is clear. So 24 if -- if there's any change that happens, but it doesn't 25 hit that threshold, then the model doesn't recognize it 26 and account for that in any way? 27 A. No, it -- it does. Obviously, it will take that 28 in. But it may be that if you have changes in estimated 10245 1 transportation costs -- let me -- maybe I should read this 2 to see what the 5% was referring to. 3 Yeah. I think that what Chuck is trying to say 4 and account for there is that the model is sensitive to 5 things that have changed, but it's not hypersensitive. 6 It's kind of like a supertanker in the ocean. It takes 7 quite a bit of water, you know, to turn this thing around. 8 So relatively small changes are not going to give you a 9 big change in outcome, because, frankly, there aren't that 10 many things that -- different that the model can do, you 11 know, for change in a given area or something. But big 12 changes over time make real differences. 13 And you can see some of that, I think, in the maps 14 that were shown later on when we compared our results to 15 the kind of differentials that we currently had in place. 16 There were some regions, even those that have already had 17 some attention paid to them since Federal Order Reform 18 back in 2000, where it was still suggesting, you know, 19 there were some increased prices that should be made. 20 Q. If you can turn to page 17, and there's four 21 regions on that, in that Figure 7. And we asked Mark if 22 he could define the regions, and he said -- we asked 23 Dr. Nicholson if he could define those regions, and he 24 said, "Ask Mark." 25 So you are here now. I was wondering if you could 26 define those four regions, just so we know exactly where 27 you are talking about. 28 A. I wish you would have asked me before I came down. 10246 1 That file is on a different laptop. 2 Q. Okay. 3 A. But I can tell you approximately what those 4 regions are. The Northeast includes certainly over to 5 Ohio, and I believe down through Virginia, just as an 6 example. 7 The Southeast is the region south of that, and at 8 least on over through Louisiana. 9 The West is Texas on up through Colorado, and I 10 believe on up -- I'd have to -- I could give you the 11 precise states, I'll be happy to do that. But there are 12 regions of the country where I think the divisions make 13 some sense, you know, that they have characteristics that 14 are somewhat homogeneous. 15 Q. And do you know -- 16 THE COURT: Let me just ask, what is the best way 17 for us to get that from him? Which would be very useful, 18 I think, Mr. English. 19 MR. ENGLISH: He's coming back in January. 20 THE COURT: Okay. 21 MS. TAYLOR: He's going to be our favorite witness 22 by the end of this hearing, which is not a title he 23 wanted. But if he could possibly bring it down when he -- 24 back when he comes and enter it, and I promise not to ask 25 very many questions on that. 26 THE WITNESS: So noted. 27 BY MS. TAYLOR: 28 Q. If you will do me the kindness, I would appreciate 10247 1 it. 2 A. I would happily do that. 3 Q. Okay. And then the Figure 7 change, it's titled, 4 "County-Level Change in Milk Production." 5 What do the colors -- I want to make sure this is 6 straight, because there's not a color legend, so could you 7 tell us what the color shading indicates? 8 A. Sure. And the indication here was -- that I 9 wanted you to get away with is that shades of green 10 indicate degrees of increase in milk production; shades of 11 pink to red indicate contraction; and those gray-colored 12 counties are maybe a little up, a little down, but 13 basically no change. 14 Q. Okay. Thank you. 15 If we can move to page 21. I think you have 16 answered some of them, so let me look here. 17 On the pie charts that are on that page, and the 18 percentages, I assume percentages of the total -- you 19 know, you write something, like, six weeks ago, down on a 20 note, and you might not know what it means anymore. 21 Let me ask you this. The percentages from 2011 22 and 2021 shown, does that come from that ATRI study -- 23 A. I believe that it did, yes. 24 Q. -- and footnoted in the previous page? 25 A. Yes. 26 Q. I believe that was my question. 27 And do you know how they got there? How that 28 study worked? 10248 1 A. Do I know how that study works? 2 Q. Yeah. I mean, where did they get their 3 percentages from? 4 A. They survey their members, and I don't remember 5 what the actual number is, but it's a very large number of 6 trucks and firms that they get their statistical survey 7 data from. It's something like 100,000 vehicles or, you 8 know -- I believe it was something like 100,000, so it's a 9 large thing. 10 The ATRI stands for American Transportation 11 Research Institute, I think, something like that. They 12 have been around for quite a while. 13 Q. Okay. And these charts are supposed to show us 14 that maybe the overall price increased, but within that 15 the allocation of costs have changed a little bit. For 16 example, fuel was 35% in 2011, and it's -- it was found to 17 be 25% in 2021. 18 A. That's right. And wages had gone up rather 19 considerably. 20 Q. And I have a question on Figure 13. If you could 21 explain that figure. And is that in any way related to 22 what was in the Figure 12 pie charts? 23 A. Yeah, it's a way -- I mean, the pie chart is, of 24 course, showing you the relative proportions of the cost 25 categories, but it doesn't impart an idea about the 26 overall change in total costs. So, for example, it 27 indicates that fuel costs had declined from 35% to 25%, 28 but perhaps total fuel costs actually remain the same, or 10249 1 might have even gone up a little bit. And you can see 2 from that Figure 13 below there that when you are looking 3 some of the total costs of transportation that, from 2020, 4 there was a substantial increase in costs. 5 Q. And does this -- where -- do these numbers come 6 from the ATRI report as well? 7 A. From which report? 8 Q. So Figure 13, I'm -- I was wondering where is the 9 source of these numbers in this chart? 10 A. Where the data came from? I believe that the data 11 from this also came from the ATRI report. I did pull 12 numbers from a few different sources. Those, you know, 13 wages, fuel, and that type of thing, can be pulled from 14 BLS data specific to transportation companies. You can 15 also pull data from a couple of other trucking reports 16 that are done. And so we do look at that when we're 17 trying to update the costs in that economic engineering 18 model. 19 Q. Okay. And the last question, and I only have like 20 two more, and they are very simple. 21 But the last question on the transportation side 22 of things: Does it take into account lack of competition 23 in the hauling industry in some regions of the country? 24 A. No. All this is doing is trying to synthesize the 25 costs of -- of what it costs to assemble and transport 26 milk. 27 And I should just tell you maybe that the genesis 28 of this economic engineering model had actually come from 10250 1 work that predated me out of Cornell University, and we 2 had relatively small haulers that were so competitive with 3 one another that, you know, they would sometimes bid for 4 milk pickups that didn't cover their full costs of 5 operation. And they may have covered the variable costs, 6 or hope they did that, but maybe not the full cost. 7 And over time the co-ops realized that even though 8 they were contracting with these haulers, they were going 9 out of business because they weren't covering their costs. 10 And so they had asked whether we would help them better 11 understand what those costs were. 12 And we worked with the haulers very closely. We 13 worked with co-ops and others, and that way when we had an 14 opportunity to -- when the co-ops had the opportunity to 15 sit down and negotiate for hauling costs, they wanted as 16 much as anything to assure their haulers that their total 17 costs were being covered. 18 So that's the way we began to do that. Even that 19 economic engineering model has been substantially re-done 20 and rewritten from those early days, but that was the 21 genesis of it. 22 Q. Okay. If we could turn to page 24. This was a 23 quick question on the map, the lines. 24 Is it $0.10 increments? What is the increments 25 between the lines? 26 A. Yeah. Those -- those are the isocline lines that 27 I talked about in my comments today, and they are shown 28 here just for clarity at every $0.10. If you put them in 10251 1 every $0.05, it would be just difficult to read in some 2 areas. 3 Q. Okay. 4 A. And I believe that you had an earlier question 5 about where are those low dual value regions in the 6 country. 7 Q. Yes. 8 A. And you can see from a map like this pretty 9 clearly that those are going to be in the areas where we 10 have -- oh, I see 1.8, for example, over there in Idaho. 11 There's -- looks like sunglasses on edge there. 12 Q. That 1.8 includes the $1.60 differential as 13 discussed earlier with somebody? 14 A. Yeah, it would in there somewhere. 15 Q. Okay. 16 A. And likewise, I believe it -- right up there in 17 that Red River Valley area, that there was a low spot. 18 Q. Okay. This is perfect timing. It's 4:57 and AMS 19 has no more questions. 20 MS. TAYLOR: Thank you very much. 21 THE COURT: Let me ask, does anyone else have 22 questions about Exhibit 302 that you want to ask 23 Dr. Stephenson? 24 There is no one. 25 Mr. English, you may approach. And we have enough 26 time -- 27 MR. ENGLISH: Your Honor, I would like 28 Dr. Stephenson to identify what has been previously marked 10252 1 as Exhibit 291. 2 THE WITNESS: Yes. This is a document that I 3 co-authored with Dr. Novakovic, and this was done at the 4 time of COVID when we were all locked at home worried 5 about such things as depooling and negative PPDs, and 6 we -- we authored this document. It's a document that is 7 still available on that DairyMarkets.org website. 8 MR. ENGLISH: It is authentic? 9 THE WITNESS: Pardon? 10 MR. ENGLISH: It is authentic? 11 THE WITNESS: Yes, it is authentic, at least from 12 what I looked at. 13 MR. ENGLISH: Your Honor, at this time I would 14 move admission of both Exhibits 291 and 302. 15 THE COURT: Is there any objection? 16 There is none. I admit into evidence Exhibit 291. 17 (Thereafter, Exhibit Number 291 was received 18 into evidence.) 19 MR. ENGLISH: Then Exhibit 302. 20 MS. TAYLOR: That's already in. Your exhibit is 21 438. 22 MR. ENGLISH: Sorry, 438, I apologize. 23 THE COURT: Is there any objection to the 24 admission into evidence of Exhibit 438? 25 There is none. Exhibit 438, also known as 26 Stephenson Exhibit 2, is admitted into evidence. 27 (Thereafter, Exhibit Number 438 was received 28 into evidence.) 10253 1 THE COURT: Dr. Stephenson, is there anything you 2 would like to add before we invite you to step down? 3 THE WITNESS: No. Thank you. 4 THE COURT: Thank you. It was a real pleasure to 5 have you here. 6 THE WITNESS: Thank you. 7 THE COURT: I look forward to January. 8 Now, let's have AMS collect its record copies. 9 Mr. English is helping, but I have one. I have actually 10 two, don't I? I have 302, and I have 291. Thank you so 11 much. 12 And now do we want to talk a little bit about 13 tomorrow before we recess? 14 MR. ROSENBAUM: Yes, Your Honor. Steve Rosenbaum 15 for the International Dairy Foods Association. 16 Our first witness tomorrow is going to be Mr. Tim 17 Galloway. Mr. Galloway will have a short statement that's 18 going to be solely devoted to Proposal 21, the proposal to 19 increase the Class II differential, at least that's what 20 his written testimony covers. 21 THE COURT: All right. 22 MR. ROSENBAUM: And then we have two other 23 witnesses, Mike Brown and Sally Keefe. And -- well, I say 24 "we." There are -- Ms. Keefe is here for MIG, not for 25 IDFA. And I'm not exactly certain that right now which is 26 going to be the order of the two for those two, but those 27 are the three witnesses. 28 THE COURT: Mr. English? 10254 1 MR. ENGLISH: And given that, we will, when it is 2 available, as soon as possible, post Ms. Keefe's testimony 3 and e-mail it to the parties. I cannot promise when that 4 will be. 5 MS. TAYLOR: And Mr. Galloway? 6 MR. ENGLISH: It has not been submitted yet, but 7 it will be submitted shortly. 8 THE COURT: You are going to keep everybody up all 9 night, aren't you? 10 MR. ENGLISH: I hope not. It's not my fault. 11 THE COURT: All right. Is there anything further 12 before we recess? 13 MS. TAYLOR: We end early tomorrow. 14 THE COURT: Yes, I will say that. I will say 15 that. Did you -- let's see what Ms. Hancock's question 16 is. 17 (Inaudible question by Ms. Hancock.) 18 THE COURT: She wants to know if there is any 19 chance we will get through those and have someone else. 20 And I'm seeing negative responses, that it is -- that is 21 unlikely. 22 MR. ENGLISH: If we get through all of those, as I 23 said earlier this week, Ms. Keefe would be prepared to 24 give her testimony on MIG-20. But I can tell you that if 25 that happens with only, I think, five hours and 40 minutes 26 of actual time for Mr. Brown and Ms. Keefe plus 27 Mr. Galloway, I just don't see that happening. 28 THE COURT: Yes, you are correct, Mr. English. 10255 1 Our plan is to end by 2:40 p.m. tomorrow so that we can be 2 out of this room by 3:00. All right. 3 I will see you all at 8:00 tomorrow morning. We 4 go off record at 5:02 p.m. 5 (Whereupon, the proceedings concluded.) 6 ---o0o--- 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 10256 1 STATE OF CALIFORNIA ) ) ss 2 COUNTY OF FRESNO ) 3 4 I, MYRA A. PISH, Certified Shorthand Reporter, do 5 hereby certify that the foregoing pages comprise a full, 6 true and correct transcript of my shorthand notes, and a 7 full, true and correct statement of the proceedings held 8 at the time and place heretofore stated. 9 10 DATED: January 28, 2024 11 FRESNO, CALIFORNIA 12 13 14 15 16 MYRA A. PISH, RPR CSR Certificate No. 11613 17 18 19 20 21 22 23 24 25 26 27 28