4477 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 Channing D. Strother, Judge 15 16 ---o0o--- 17 18 Carmel, Indiana 19 September 19, 2023 20 21 ---o0o--- 22 23 24 25 26 Reported by: 27 MYRA A. PISH, RPR, C.S.R. Certificate No. 11613 28 4478 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 AMERICAN FARM BUREAU FEDERATION: 7 Roger Cryan Danny Munch 8 FOR THE INTERNATIONAL DAIRY FOODS ASSOCIATION: 9 Steve Rosenbaum 10 FOR THE MILK INNOVATION GROUP: 11 Ashley Vulin 12 Charles "Chip" English 13 FOR THE NATIONAL MILK PRODUCERS FEDERATION: 14 Nicole Hancock Brad Prowant 15 FOR SELECT MILK PRODUCERS, INC.: 16 Ryan Miltner 17 FOR EDGE DAIRY COOPERATIVES: 18 Dr. Marin Bozic 19 FOR LEPRINO FOODS: 20 Erik G. Nielsen 21 22 ---o0o--- 23 24 (Please note: Appearances for all parties are subject to 25 change daily, and may not be reported or listed on 26 subsequent days' transcripts.) 27 28 ---o0o--- 4479 1 M A S T E R I N D E X 2 SESSIONS 3 TUESDAY, SEPTEMBER 19, 2023 PAGE 4 MORNING SESSION 4483 AFTERNOON SESSION 4646 5 6 7 ---o0o--- 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 4480 1 M A S T E R I N D E X 2 WITNESSES IN CHRONOLOGICAL ORDER 3 WITNESSES: PAGE 4 Danny Munch: 5 Direct Examination by Dr. Cryan 4494 Cross-Examination by Ms. Hancock 4500 6 Cross-Examination by Mr. Rosenbaum 4504 Cross-Examination by Mr. English 4524 7 Cross-Examination by Mr. Miltner 4529 Cross-Examination by Ms. Hancock 4535 8 Cross-Examination by Mr. Rosenbaum 4536 Cross-Examination by Ms. Taylor 4537 9 Redirect Examination by Dr. Cryan 4541 10 Steve Cooper: 11 Direct Examination by Mr. Miltner 4547 Cross-Examination by Mr. Rosenbaum 4563 12 Cross-Examination by Ms. Hancock 4565 Cross-Examination by Mr. English 4570 13 Cross-Examination by Mr. Rosenbaum 4571 Cross-Examination by Mr. Wilson 4572 14 Redirect Examination by Mr. Miltner 4578 15 Nana Y. Farkye: 16 Direct Examination by Mr. Miltner 4580 Cross-Examination by Mr. English 4602 17 Cross-Examination by Ms. Taylor 4610 Redirect Examination by Mr. Miltner 4614 18 Chris Allen: 19 Direct Examination by Mr. Miltner 4617 20 Cross-Examination by Ms. Taylor 4620 Redirect Examination by Mr. Miltner 4621 21 Alison Krebs: 22 Direct Examination by Mr. Nielsen 4623 23 Cross-Examination by Dr. Cryan 4632 Cross-Examination by Mr. Miltner 4633 24 Cross-Examination by Ms. Taylor 4643 Cross-Examination by Mr. Rosenbaum 4646 25 26 27 28 4481 1 M A S T E R I N D E X 2 WITNESSES IN CHRONOLOGICAL ORDER 3 WITNESSES: PAGE 4 Mike Brown: 5 Direct Examination by Mr. Rosenbaum 4648 Cross-Examination by Mr. Miltner 4657 6 Cross-Examination by Dr. Cryan 4664 Cross-Examination by Mr. Miltner 4665 7 Cross-Examination by Ms. Taylor 4666 Redirect Examination by Mr. Rosenbaum 4667 8 Dr. Peter Vitaliano: 9 Direct Examination by Ms. Hancock 4668 10 Cross-Examination by Ms. Vulin 4693 11 ---o0o--- 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 4482 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. 5 222 Testimony of 4485 4546 Danny Munch 6 223 Testimony of 4548 4580 7 Steven Cooper 8 224 Testimony of 4582 4615 Nana Y. Farkye 9 225 Exhibit Select-8 4582 4615 10 205 Testimony of 4616 11 Amy Penterman 12 210 Testimony of 4616 Nicole Barlass 13 211 Testimony of 4616 14 Christina Zuiderveen 15 213 Testimony of 4616 Paul Cornette 16 226 Testimony of Chris Allen 4617 4622 17 227 Testimony of 4623 4648 18 Alison Krebs 19 228 Testimony of 4649 4668 Mike Brown 20 229 Testimony of 4669 21 Dr. Peter Vitaliano 22 230 Spreadsheet 4669 23 231 Pandemic Market 4717 4733 Volatility Act Document 24 ---o0o--- 25 26 27 28 4483 1 TUESDAY, SEPTEMBER 19, 2023 - - MORNING SESSION 2 THE COURT: On the record. 3 I'll swear you in again. Raise your right hand. 4 DANIEL MUNCH, 5 Being first duly sworn, was examined and 6 testified as follows: 7 THE COURT: I understand you're self-represented 8 for a few minutes here, so you may -- I guess you gave 9 your background and all previously. 10 THE WITNESS: I have not been up here, yet, so -- 11 THE COURT: All right. Introduce yourself, who 12 you work for, business address. 13 THE WITNESS: Sounds good. My name is Daniel 14 Munch, D-A-N-I-E-L, M-U-N-C-H. Business address is 600 15 Maryland Avenue, Southwest, Suite 1000W, Washington, DC, 16 20024. 17 I'm not attempting to characterize myself as an 18 expert, although I would like to provide some background 19 information on myself for context. 20 I am the economist for the American Farm Bureau 21 Federation, the largest general farm organization in the 22 nation, which gets to boast itself as the voice of 23 agriculture due to its diverse membership in all 50 states 24 and Puerto Rico, and nearly 6 million farmer and rancher 25 families. 26 As a general farm organization, we represent dairy 27 farmers, but also the farmers who grow the feed for dairy 28 farmers. The farmers and ranchers who grow beef, poultry, 4484 1 pork, lamb, fish and all other forms of animal protein, as 2 well as major row crops, specialty crops, horticulture, 3 forestry, and everything in between that provides us and 4 the rest of the world, food, fiber, and fuel. 5 As such, in my position I not only cover dairy 6 markets and policy, I also cover specialty crop markets, 7 transportation infrastructure, invasive and endangered 8 species, public lands, disaster assistance, aquaculture, 9 and a number of other topics. 10 I have been on the economics team of AFBF for over 11 two and a half years. I received my Bachelor's of science 12 from the University of Connecticut -- go Huskies -- with a 13 double major in livestock management and policy and 14 resource economics. I then went on to receive my Master's 15 of science from Cornell University in applied economics 16 and management, with a focus on agricultural and food 17 economics. 18 My Master's thesis research was in dairy marketing 19 topics, estimating the value of cooperative membership, 20 and running analysis on milk pricing topics in the 21 Northeast. Sorry. 22 THE COURT: Do you want to keep going or do we 23 want to switch off? Do you have a statement that you 24 would like to present for the record today? 25 THE WITNESS: Oh, I wasn't -- I still had a little 26 bit more. 27 THE COURT: Oh, I'm sorry. 28 THE WITNESS: They were asking me to slow down. 4485 1 THE COURT: Oh, okay. 2 THE WITNESS: I paused for a second. 3 I've worked on Capital Hill in agricultural policy 4 issues in the House of Representatives for Farm Credit 5 East, and as a protein commodity analyst for Urner Barry, 6 a leading animal product price supporting company. 7 And just to put things into context, I took one of 8 the last iterations of Dr. Novakovic's virtual dairy 9 markets classes. I was -- Dr. Chris Wolf came in in my 10 last semester, another witness, and then I was also a 11 student of Dr. Harry Kaiser's in his -- 12 (Court Reporter clarification.) 13 THE WITNESS: And I was also a student of 14 Dr. Harry Kaiser's, another previous witness, in my first 15 semester of grad school. All to say, that there's a 16 wealth of knowledge out there, and I appreciate the 17 opportunity to be up here on behalf of the American Farm 18 Bureau Federation. 19 I will now be reading AFBF's statement listed 20 as -- is it AFBF-3, Roger? 21 DR. CRYAN: Yes. 22 THE COURT: Let's mark AFBF-3, Exhibit 222, for 23 identification. 24 (Thereafter, Exhibit Number 222 was marked 25 for identification.) 26 THE COURT: You may continue with your statement. 27 THE WITNESS: Thank you. 28 The American Farm Bureau Federation has nearly 4486 1 6 million members in all 50 states and Puerto Rico, 2 including many thousands of cooperative and independent 3 dairy farmers. All of these dairy farmers are indirectly 4 or mostly directly affected by the pricing provisions of 5 the Federal Milk Marketing Orders. These dairy farmers 6 play a crucial role in the development of AFBF dairy 7 policy. Every Farm Bureau position and proposal is based 8 explicitly on that policy, developed through a grassroots 9 process in which farmers make the decisions every step of 10 the way. 11 AFBF submitted nine proposals for consideration in 12 this hearing, and appreciates the opportunity to address 13 the four that were accepted by USDA, as well as the clear 14 direction on what may be needed to advance the rest. 15 A fundamental focus of AFBF's proposals is a 16 reduction or elimination of negative producer price 17 differentials and the depooling they cause. We believe 18 that an orderly pool is the key to an orderly marketing -- 19 a key to orderly marketing and ensuring Federal Orders 20 continue to benefit farmers, cooperatives, processors, and 21 consumers. The key to an orderly pool, in turn, is above 22 all, the proper alignment of the four class prices. 23 This statement covers Category 3, Class III, and 24 Class IV formula factors, and includes AFBF's response to 25 Proposal 7 made by the National Milk Producers Federation; 26 Proposal 8 made by the Wisconsin Cheese Makers 27 Association; and Proposal 9 made by the International 28 Dairy Foods Association. 4487 1 AFBF supports adjusting Make Allowances to reflect 2 the changes in costs and technology, following the same 3 general logic as NMPF's petition. We believe, however, 4 that such adjustments cannot be fairly undertaken except 5 using the data from a mandatory and audited USDA survey of 6 at least the plants participating in the NDPSR survey. 7 At the time of Order Reform, product formula 8 prices were instituted using a combination of voluntary 9 survey and a mandatory and audited survey. 10 The voluntary survey conducted by Dr. Stephenson, 11 among others, and used as the primary source for Order 12 Reform, was one of a series of studies that had been 13 undertaken as a means of evaluating and benchmarking plant 14 costs for the benefit of plant operators. 15 Because that survey's purpose had not previously 16 been the setting of regulatory parameters, there was no 17 obvious bias in the self-selection of participants. Each 18 participant was, presumably, interested in a full picture 19 of costs, including seeing how they stacked up. 20 However, more recent surveys, particularly the 21 2021 update conducted by Dr. Stephenson, was commissioned 22 by USDA with the clear intention of making its results 23 available for proposals to update the Make Allowance, and 24 its update in 2023 was also explicitly commissioned for 25 regulatory purposes. 26 This, unfortunately, creates an equally clear 27 incentive for dairy manufacturers to be selective in their 28 choice to participate, and an unfortunate temptation to be 4488 1 creative in the accuracy of their reporting. As a result, 2 whatever value the original voluntary survey had for the 3 original development of Make Allowances in the price 4 formulas, has been substantially undermined by potential 5 bias in the survey. 6 The publicly released 2021 survey, for example, 7 represents only 60% of the nonfat dry milk plants 8 participating in the NDPSR, 29% of the dry whey plants, 9 24% of cheddar cheese plants, and 20% of butter plants. 10 The conclusion must be that it will be unfair to increase 11 the Make Allowances based on this survey. 12 And those values are from a webinar that Dr. Mark 13 Stephenson did for USDA on the first survey. Those 14 statistics were directly from that webinar posted on 15 USDA's website. 16 IDFA contends that the 2023 update to 17 Dr. Stephenson's study captures a higher percentage of 18 product plants and volume, therefore, nullifying our 19 concerns. Even with the improved sample size, nearly 20 45% of cheese and nearly 50% of whey volume are still not 21 captured. Our members have expressed ardent concerns over 22 plants who elect not to participate in voluntary surveys 23 that are used to inform regulatory decisions. Even small 24 variations in reported cost numbers could lead to 25 Make Allowance changes that unfairly substantially reduce 26 the price paid to farmers. 27 AFBF also has concerns with the projections made 28 by Dr. Schiek on behalf of the International Dairy Foods 4489 1 Association. The regression analysis used to estimate the 2 source of changes in the labor, utility, and other costs, 3 are based only on 15 annual observations for 2000 through 4 2016, which is a relatively small sample size for any 5 regression analysis, but especially for equations with 6 three to five explanatory variables, including the 7 constant term, and is even smaller when multiple 8 specifications have been explored. 9 In addition, the use of dummy variables, which are 10 often applied to explain data in the years that the 11 underlying estimation doesn't fit, raises further concerns 12 about the real fit of the regression analysis. 13 And as problematic as Dr. Schiek's estimation of 14 the pattern of cost growth within 15 years of observed 15 data, more problematic is the extrapolation of such 16 results beyond the data period to project costs in 2022. 17 The estimation is to find the best fit within 18 15 years, which often leads to parameters that help fit 19 the end years, but often becomes unreliable when extended 20 to years before or after the study period. A simpler 21 analysis would have been easier to interpret and would 22 have allowed better evaluation of how reasonable the 23 extrapolated results may be. In this case, a relatively 24 complicated model based on 15 years of data is projected 25 out for six years, generating questionable results. 26 Regarding the USDA tradition of using two 27 different cost surveys, there is an old saying, "A person 28 with one watch always knows what time it is; a person with 4490 1 two watches never knows what time it is." 2 Such is the problem with using two significantly 3 different sets of survey results and blending them into 4 one result. This is more an art than a science, and USDA 5 was put in the difficult position of applying such an art 6 in the past. It is better that we have one very accurate 7 watch, such as a mandatory and audited survey of 8 processing plants. 9 The last time we knew what time it was, was in 10 California in 2016, as the last mandatory audited survey 11 of U.S. dairy processing costs were those of all 12 manufacturing plants in the state of California in 2016, 13 conducted and audited by the California Department of Food 14 and Agriculture. 15 This full accounting of processing costs was the 16 useful component of the overall data used to set 17 Make Allowances at the time of Order Reform, because 18 California has been the largest milk producing state since 19 1993, with over 18% of U.S. production in 2022. It's been 20 the largest butter producing state since about the same 21 time with roughly a third of current U.S. production, and 22 the largest nonfat milk producing state with 44% of U.S. 23 production in 2022. California is also the second largest 24 cheese producing state, with 17% of U.S. production in 25 2022. 26 Since the 2016 California survey was a 27 mandatory -- was mandatory, a representative sample of 28 commodity dairy products was captured providing an 4491 1 important check to voluntary surveys. The CDFA survey was 2 discontinued in 2017 after the promulgation of a Federal 3 Order in California. 4 Although it did not evaluate product yields, this 5 survey would provide a basis for a conservative one-time 6 increase in Federal Order Make Allowances, preferable to 7 those proposed by NMPF and IDFA. These numbers are nearly 8 in line with both NMPF's proposal and with IDFA's 9 proposal, proposed initial Make Allowance increases. 10 Our conclusion is that any fair update of the 11 Make Allowance must be based on a mandatory and audited 12 survey of costs and yields of at least the plants 13 participating in the National Dairy Product Sales Report. 14 Currently, only the 2016 CDFA survey comes close 15 to this. We at Farm Bureau are working with NMPF and IDFA 16 to pursue language in the upcoming Farm Bill that would 17 direct USDA to conduct such a survey. AFBF, NMPF, and 18 IDFA, all by their own testimony, hope to have the 19 official data -- survey data as soon as it practicable. 20 Given the continued investment in dairy processing 21 capacity, it is a real question whether the 22 Make Allowances are too low at all. Moving forward with 23 increases now could easily go too far. 24 Handler groups have often argued that they cannot 25 reap the benefits of charging higher prices in the 26 marketplace because those prices get looped into the NDPSR 27 survey, meaning the Make Allowance is the only monetary 28 value they can operate off of. This point holds less 4492 1 water when less than 10% of butter, about 10% of all 2 cheese, 28% of whey, and 52% of nonfat dry milk volume is 3 captured in the NDPSR. 4 Combined for the presented commodities, the NDPSR 5 captured 19.6% of total volume in 2000; 18.3% of 6 production volume in 2011; and 14.8 of total production 7 volume in 2022, for an average decline in 1.1% in total 8 production captured annually. 9 On average, less than 20% of the total production 10 of cheese, butter, nonfat dry milk, and whey, have been 11 captured in the NDPSR survey, with a clear decline in the 12 percentage of butter, whey, and nonfat dry milk captured. 13 Considering the various other dairy products that 14 are sold and not included in the NDPSR, the true volume of 15 dairy products captured by the NDPSR is likely much lower 16 than 14.8%. 17 Additionally, a comparison of the 2023 NDPSR 18 survey to the latest National Agricultural Statistic 19 Survey, Dairy Product Survey, reveals that 17.2% of butter 20 processing plants, 61.4% of nonfat dry milk plants, 53.3% 21 of dry whey plants, and 12% of cheddar cheese plants are 22 captured by the NDPSR survey. 23 In total, 7.2%, a drop from 7.8% in 2018, of all 24 manufacturers that produced one or more dairy products are 25 captured in USDA's mandatory price reporting. This means 26 that more than 92% of dairy processing plants are not 27 required to report the prices for the dairy products they 28 manufacture and sell. 4493 1 Given these statistics, one could easily argue 2 that handlers can benefit from the sale of the substantial 3 product volume and product varieties not currently 4 captured within the NDPSR and are not exclusively reliant 5 on the Make Allowance to make ends meet. This is 6 especially true of handlers to diversify their operations, 7 a tactic that many farmers are told to use to protect 8 against revenue uncertainty. 9 This does not mean -- this does not mean 10 Make Allowances are not important. Our members recognize 11 they are. But they also recognize the system does not 12 restrict all handlers in terms of covering costs. Only a 13 mandatory and audited survey of costs and processing 14 yields can provide a fair basis for adjusting 15 Make Allowances and yield factors within the current 16 pricing structure, just as it has been clearly established 17 that only a mandatory and audited survey of manufacturers' 18 prices can provide a fair basis for setting the monthly 19 milk and component prices used in the Federal Orders. 20 We believe that such a survey should be conducted 21 once every two years in order to appropriately balance the 22 value of the data with the burden on processors. 23 This is a -- this is close to a realistic estimate 24 of the time it takes to undertake a Federal Order hearing 25 from petition to implementation. More frequent surveys 26 would be unproductive, although the biannual survey could 27 collect two years of data. 28 And a final note on using input price indices. 4494 1 Proposal 9 uses processing input costs to update mandatory 2 audited survey data from California from 2003 to 2016. 3 AFBF opposes using indexing to adjust Make Allowances. 4 Over time, input price increases tend to be at least 5 partly offset by productivity increases. This was 6 observed in the record of 2007 price hearing, in which it 7 was suggested that labor productivity growth, for example, 8 more or less matched wage increases. This is why the full 9 plant costs and yield accounting is critical to any fair 10 adjustment of the Make Allowance. 11 And we have some resources linked there. 12 THE COURT: Do you want to take over, Dr. Cryan? 13 Actually, I did neglect to ask whether the 14 statement was prepared by you, under your supervision? 15 THE WITNESS: Yes, Roger and I worked on it 16 together. 17 THE COURT: Close enough. 18 And did you have any corrections? 19 THE WITNESS: There was one small typo where I 20 think there was an "as," I think he meant "of." But -- on 21 page 4 in the second paragraph. I believe it said in the 22 first sentence, "and yields at at least," when it should 23 say "of at least." 24 THE COURT: Very good. 25 DR. CRYAN: Well, that works either way. 26 THE COURT: You may take over. 27 DIRECT EXAMINATION 28 BY DR. CRYAN: 4495 1 Q. Okay. Thank you for testifying, Mr. Munch. It's 2 good to see you. 3 Could you talk a little bit about the policy 4 process that led this beginning in -- well, just 5 generally, including the event in Kansas City? 6 A. Yeah, absolutely. 7 So AFBF is a grassroots organization. We have 8 over 2,600 county farm bureaus who are all members of the 9 50 State Farm Bureaus and Puerto Rico, and all of those 10 farm bureaus are members of AFBF. They all have 11 county-level policies that are developed and voted on at 12 the county level. The counties submit policies to be 13 reviewed at the state level. And if they are state 14 specific, they will vote and approve policy at the state 15 level. And then any policy that is federal in nature will 16 make it up through the counties, through the states, and 17 get approved by both the presidents of all the states and 18 Puerto Rico, in December resolutions, followed by delegate 19 voting session, which has a few hundred farmers at our 20 annual meeting every January. 21 Q. And those farmers are all elected to those 22 positions? 23 A. Yeah. All those farmers are elected to those 24 positions by their peers, by other members. 25 Back in 2019, one of the -- one of the -- one 26 other thing that the members can do is provide 27 recommendations to the board, and one of the 28 recommendations was to hold a dairy working group. And 4496 1 the dairy working group would have equal representation of 2 each Farm Bureau region. The Farm Bureau regions, being 3 the Midwest, Northeast, West, and South, have two dairy 4 farmers from each of those regions -- sorry, excuse me -- 5 three dairy farmers from each of those regions, to review 6 work with industry experts and come up with a list of 7 recommendations to give to our states for policy 8 development. That dairy working group was re-recommended 9 both in 2021 and 2022, so we have had a total of three 10 years of dairy working groups. 11 In the 2022 dairy working group, one of their big 12 recommendations coming off of the 2021 working group was 13 that the industry needed to get in a room, get together, 14 and talk about Federal Order issues. And this was in 15 direct response to Secretary Vilsack's request for 16 everybody to get into a room. 17 So they saw that need, and they requested in a 18 board recommendation -- in a recommendation to the board, 19 that AFBF hold a national dairy Federal Order forum that 20 brought together processors, co-ops, other industry 21 adjacent officials, USDA, and most importantly, farmers, 22 to discuss Federal Order topics. So it was meant to be 23 very educational as well, but also to come up with 24 consensus. 25 So we had over 200 participants, most of which 26 were dairy farmers, over 150 were dairy farmers, from 35 27 states. And it was set up in such a way that we had about 28 21 tables in the room, each with eight to nine people at 4497 1 them, and they were specifically assigned seating based on 2 regional diversity, making sure you have co-op and 3 processor representation at each table, making sure you 4 had people who sold to an independent co-op at each table. 5 And they were all given discussion questions throughout 6 the conference, related to many of the topics we're 7 discussing at this hearing. 8 Consensus was measured by how many tables reported 9 consensus on a particular item. So, for instance, we had 10 21 tables at the forum. All 21 tables supported switching 11 back to the higher-of unanimously, about 19 supported 12 increasing Class I differentials, and the same amount a 13 mandatory and audited cost survey. 14 And it is important to note that not all tables 15 discussed all the issues. So just because there is less 16 tables doesn't mean that they didn't support it, they just 17 might not have discussed it. 18 So the forum brought a lot of folks together. A 19 lot of the people that have been in this room attended 20 that forum, spoke at that forum. 21 And we really -- our members utilized the outcome 22 of that forum for this January's policy process. So many 23 of the new policies related to dairy that we have were 24 directly an outcome of what the industry at our forum came 25 up with. 26 So I think it really represents the strong 27 grassroots nature, our policy is our farmer's policy, and 28 you know, we get to work on behalf of them. 4498 1 Q. And for this topic, Farm Bureau policy 2 representing producers is what? 3 A. Yes. So for Make Allowance specifically, we only 4 support Make Allowance increases based on mandatory and 5 audited surveying. They also -- it's also very important 6 to them that yields are updated in accordance with 7 Make Allowances. 8 We do not oppose updates to Make Allowances. They 9 clearly want Make Allowances that are fair. We heard last 10 week in testimony, that even small cent changes have a 11 major impact on dairy farmers' bottom line, so they want 12 to buffer against any wrongful increases. 13 Q. And what's the significance of Farm Bureau 14 presenting the -- and presenting and presenting an 15 overview of the California -- the CDFA numbers from 2016? 16 A. So our main -- our main goal of including the 2016 17 CDFA survey and in this testimony, was to show that that's 18 the last time we had an audited and mandatory survey. We 19 understand and we realize it's only California, so it is 20 not representative national costs. But in our mind it's 21 more preferable than the other methodologies that have 22 been -- that have been expressed by other groups. 23 We don't intend -- you know, we realize some of 24 the other testimony that was brought up last week in terms 25 of labor cost differences, that it's not national in 26 nature, but it's still preferable to the other 27 methodologies we have seen. 28 Q. And would you just say a few words again, 4499 1 elaborate on the importance of yields and why that 2 matters? 3 A. Yeah. You know, as we're well aware, as 4 technology improves, you know, even as farmers get better 5 at their jobs, often yields are increased. And in order 6 to offset, you know, and compensate for those increases in 7 yields in the formula, any increases in Make Allowances 8 should be offset by updated yields. Without updating 9 both, you sort of have an inequality in which piece of the 10 industry you're favoring in our members' opinions. So we 11 want to make sure that yields and Make Allowances are 12 updated together. 13 And I recall in Dr. Stephenson's testimony, he 14 also agreed that any sort of survey that we have should 15 account for all factors that are used in -- to set a 16 pricing formula. So we agree with Dr. Stephenson there. 17 Q. And you saw the letter that Mr. Miltner provided 18 the hearing about, from IDFA made very clear that the 19 intention of the last update of the survey was for 20 regulatory purposes, right? 21 A. Yes. So I believe that was Exhibit 179, but the 22 first sentence of that paragraph was in anticipation of a 23 possible USDA hearing to consider possible adjustments to 24 Make Allowances, and that was in their e-mail to call for 25 survey participants. So that -- that e-mail was sent out 26 specifically for regulatory purposes. 27 So -- so our concern is, folks might not have 28 participated or might have chosen maybe the lower side of 4500 1 averages when they reported to that survey because they 2 knew it might be used in this context. 3 And we have already heard from a number of 4 witnesses, very large processing witnesses, that they did 5 not even participate in that updated survey. So that is 6 concerning to our members. 7 Q. And even without being dishonest, a processor 8 could choose any of a number of legitimate cost accounting 9 approaches. 10 A. Correct. And I think it's important to note, you 11 know, he mentioned any outliers would be a red flag and he 12 would reach out. But it's easy to have a lower number and 13 not an outlier number, so I think that's a very different 14 issue. 15 Q. Do you have anything else you would like to add? 16 A. No. 17 Q. Well, thank you. 18 Mr. Munch is available for cross-examination. 19 THE COURT: Any questions for this witness? 20 CROSS-EXAMINATION 21 BY MS. HANCOCK: 22 Q. Good morning, Mr. Munch -- Dr. Munch. 23 A. Not doctor. 24 Q. Just Mr. I don't think I have ever been in a room 25 with so many Ph.D.s, so it's better to err on the side of 26 confidence. 27 Okay. So thank you so much for your testimony. I 28 just had a couple of questions. 4501 1 When I go to page 3 of your testimony, you have 2 included there a comparison of the makes and costs. But 3 this is you just tracking the current numbers compared to 4 where CDFA was in 2016, and then the proposals that have 5 been put forth by National Milk and IDFA? 6 A. Correct. 7 Q. Okay. And this was just a way that you could 8 visually see what the differences were in comparing the 9 data? 10 A. Yes, it was solely a reference for the readers. 11 Q. And I think Dr. Cryan just clarified this, but the 12 proposal or the position of AFBF is that if you are going 13 to use the data, you want it to be mandatory audited cost 14 survey data, and the last one that you are aware of is the 15 2016 CDFA survey. Is that fair? 16 A. Correct. 17 Q. Do you know what the CDFA's process was in 18 collecting that data that they used from that 2016 survey 19 going backward? 20 A. I believe Dr. Schiek or -- or another witness had 21 gone over some of their process. I can't recall the 22 specifics, but just from what he mentioned, that's my 23 knowledge of it. 24 Q. Okay. And did you hear in the description of the 25 process that CDFA undertook that not only did they go out 26 and do the -- the cost -- the mandatory cost survey, but 27 in addition to that, they -- once they collected that 28 data, that they were able to do a subjective analysis of 4502 1 it and apply kind of a totality perspective around that 2 data in order to set their Make Allowances? 3 A. I remember the first part. The subjective part, I 4 do not recall. But I don't oppose that that was in there. 5 Q. Okay. Do you think it would be important to be 6 able to put actual context around the numbers collected in 7 order to make sure that the data reflects actual market 8 conditions that are currently existing? 9 A. Absolutely. 10 Q. And we have had some recent examples that we have 11 talked about throughout this hearing, but one of them 12 would be the global pandemic and the rise in inflationary 13 costs that we have seen over the last few years. 14 A. Agreed. 15 Q. Okay. And then other things like, you know, wars 16 or conflicts with Ukraine and Russia, that's another 17 example of things that would affect or impact numbers at 18 least on a -- on a temporary, if not a long-term temporary 19 basis? 20 A. Agreed. 21 Q. Okay. Do you think that would be important to 22 overlay against the data that's received to be able to put 23 it in the context of what's happening in the market? 24 A. I think it's important. I think it's difficult in 25 any existing methodology that we have to account for those 26 changes until we have a mandatory survey. So obviously 27 those 2016 numbers haven't been updated for those market 28 disrupting events, but neither have any other options that 4503 1 we have. So our stance is a mandatory audited new survey. 2 But in the scheme of things, that methodology was more 3 preferable than the rest. Just that's a background 4 example. 5 Q. Okay. And it's fair to say that even with that 6 methodology from 2016 that you recognized, that there are 7 also some limitations in using that and applying that 8 across the board throughout the country in 2023 or 9 forward? 10 A. We do. 11 Q. And one of them would be that that survey is only 12 reflective of what was happening in California? 13 A. Absolutely. As I mentioned a little bit, you 14 know, we know the labor costs, energy costs, and taxes 15 even, for instance, in California, are much different than 16 the rest of the country. 17 Q. And another limitation would be that it's 18 outdated. It is already seven, eight years old based on 19 the data that was collected. 20 A. Correct. 21 Q. And then another limitation would be that 22 California is not necessarily reflective of what the cost 23 conditions are that exist throughout the country -- that's 24 just the end of it. 25 A. Correct. We were just -- as far as methodology 26 goes, the CDFA survey is the most preferable option that 27 AFBF sees based on its methodology compared to the other 28 methods. We are not contradicting or, you know, arguing 4504 1 against any of the faults that you mentioned. 2 Q. Yeah. And then, I think you make clear in your 3 statement in Exhibit 222 that AFBF's preferred route would 4 be to get a current standardized audited mandatory cost 5 survey of the entire country. 6 A. Absolutely. And we have led the charge on the 7 language of that. The language that NMPF, IDFA agreed on 8 was drafted originally by Roger, and we have been working 9 with our lobbyists to get that agreed upon by both your 10 groups. 11 Q. Okay. 12 MS. HANCOCK: Thank you so much for your time 13 today. Appreciate it. 14 THE COURT: Mr. Rosenbaum. 15 CROSS-EXAMINATION 16 BY MR. ROSENBAUM: 17 Q. Steve Rosenbaum, International Dairy Foods 18 Association. 19 Is -- are Land O'Lakes personnel, are they liars? 20 A. I don't think so. 21 Q. Do they submit false information under oath? 22 A. I don't believe so. 23 Q. Do you believe the information they submitted 24 under oath in this hearing as to what increases they have 25 personally experienced since Make Allowances were last 26 set? 27 A. I -- I do not oppose anything that Land O'Lakes 28 has said. 4505 1 Q. You don't? 2 A. No. 3 Q. How about their participation in Dr. Stephenson's 4 survey, you believe them when they said they did it? 5 A. I do. 6 Q. Okay. Because I mean, you have said people are 7 manipulating their submissions to the survey. That's a 8 pretty extreme accusation, isn't it, sir? 9 A. I don't think we're accusing any processor of 10 doing so. All we're stating is from our farmers' 11 perspective, without a USDA audit of the data, we have no 12 way of knowing that did not occur. 13 We -- we trust in many of the processor groups, 14 especially the ones that are farmer-owned, to report data. 15 But as in any industry, there are bad players, and I'm not 16 saying any of the witnesses we have seen here have been a 17 bad player. But there are players out there that may not 18 participate or who participate who could put forth lower 19 data. 20 Q. Well, sir, if you don't consider this an 21 accusation, I don't know what is. Look at page 2 of your 22 statement. "This, unfortunately, creates an equally 23 center incentive for dairy manufacturers to be selective 24 in their choice to participate, and an unfortunate 25 temptation to be creative in the accuracy of their 26 reporting." 27 I mean, you're saying people are lying. That's 28 what you are saying, sir. 4506 1 A. I disagree with that premise completely. 2 Q. Really? An unfortunate temptation to be 3 inaccurate? To be creative in the accuracy of their 4 reporting? You don't think that's an accusation that 5 people lied in their submittals? 6 A. No. 7 Q. Really? 8 So you do accept that Land O'Lakes' data is 9 correct and it was submitted correctly to -- to 10 Dr. Stephenson? Is that your testimony? 11 A. I believe that Land O'Lakes, to the best of their 12 ability, completed that survey. I do not have any 13 knowledge, nor was there auditing to know, you know, 14 exactly what the results were. But I have no reason to 15 believe that Land O'Lakes was predatory in any way. 16 Q. Okay. And you have no reason to believe that 17 their data is incorrect in showing that they have 18 experienced an 81% increase over Dr. Stephenson's 2000 19 survey -- 2007 survey reports? 20 A. I have no reason to believe that that is false. 21 Q. Okay. How about AMPI, are they -- are they liars? 22 A. I'm not going to assume any -- any processor group 23 that submitted data was a liar. 24 Q. Well -- 25 A. The way that the survey -- voluntary survey is set 26 up, unaudited, allows for fluctuation in how people can 27 report data. 28 Q. Yeah, but I'm trying -- 4507 1 A. I'm not accusing any processor. We are clearly 2 just saying that our farmer members have concerns with how 3 the data can be reported. 4 Q. Are some -- 5 A. And everybody in this room, every witness has 6 agreed to that extent. The mandatory survey, audited, is 7 the most preferred option. 8 Q. Are some of your farmer members, members of 9 Land O'Lakes? 10 A. Absolutely. 11 Q. Okay. Have you heard from any of them suggesting 12 that their, you know, employees, submitted false data? 13 A. We have heard from many of our members of the 14 concerns they have with their processors, and not 15 particularly specific processors. But this policy 16 directly comes from them. 17 I was on a call yesterday with a farmer for 18 30 minutes concerned about behavior of handlers. And I -- 19 I'm on those calls, traveling around the country is one of 20 the perks of the job, every week listening from farmers. 21 They are concerned about the data put forth by processors 22 and want the audited data. 23 Q. Are they con- -- is -- are the members of AMPI 24 concerned about the data they themselves submitted, 25 whether it was, whether they had given in to the 26 unfortunate temptation to be creative in the accuracy of 27 the reporting? Anyone from AMPI express that concern 28 about their own data? 4508 1 A. I cannot recall any of my members from AMPI. 2 Q. Are you aware that AMPI and Land O'Lakes both 3 reported that they had experienced increases in 4 manufacturing costs that exceed the increases that -- that 5 International Dairy Foods Association is seeking in its 6 proposals? 7 A. I do not. 8 Q. How about -- how about Darigold, they are a 9 cooperative too, right? 10 A. Yes. I do not have reason to believe anything 11 that they put forth personally was incorrect. 12 Q. Okay. 13 A. I think we're kind of going in circles now. 14 Q. Well, no, no. We're trying to pin down the 15 reality, sir. 16 Are you aware that they testified that with 17 respect to all four commodities combined, they have 18 experienced an 80% increase in costs since the last 19 Make Allowances were set? 20 A. We do not oppose that processors have experienced 21 increases. We do not know to an exact extent on average 22 what that is from the current methodologies. 23 Q. Do you have suspicions about the accuracy of the 24 Darigold information? 25 A. Not specific to Darigold. We have suspicions with 26 the methodologies used in both of the proposals. 27 Q. Well, I'm -- I'm -- I'm focusing specifically on 28 your accusation that there is an unfortunate temptation to 4509 1 be creative in the accuracy of reporting -- 2 A. There is an opportunity for processors to be 3 creative. That's -- that's true. 4 Q. But you think -- do you think Darigold was 5 creative? 6 A. I do not have enough knowledge of Darigold to make 7 a characterization. 8 Q. How about California Dairies, they provided oral 9 testimony on this question that indicated -- we'll have to 10 check the transcript to make sure I have this exactly 11 right -- but I believe their testimony was that the IDFA 12 proposal reflected percentage increases in Make Allowances 13 that were consistent with the increases they had 14 experienced in their own costs of manufacture. 15 Do you have reason to doubt the accuracy of that 16 information? 17 A. I have no reason to doubt it, but it was 18 unaudited. And just like with every other processor, we 19 do not know. 20 Q. Okay. And, by the way, every -- I mean, ever 21 since -- ever since USDA relied upon the Cornell studies 22 to set the Make Allowances in the 2000 Order Reform, 23 everyone's always known the Stephenson studies were 24 something USDA relies upon for setting Make Allowances. 25 Isn't that right? 26 A. Yes. But just because there's a precedence of 27 using a flawed path or option doesn't mean we can't 28 improve the system. 4510 1 Q. Yeah, but I'm focusing now on your separate 2 accusation, which is that because when IDFA undertook the 3 Yeoman's effort to secure the most robust survey ever 4 conducted for setting Make Allowances, that that somehow 5 was a tip-off about how the Stephenson survey might be 6 used. I mean, my God, everyone knew that since 2000, 7 didn't they, that that's what Stephenson surveys are used 8 to do, set Make Allowances? 9 A. Yes. We do not think IDFA was ill-intentioned 10 when they requested this data. But as we have seen, there 11 are numerous witnesses of very large co-ops, processors 12 that did not participate in that survey. 13 Q. You don't question -- I can go through the 14 numbers. I mean, you don't question that this is in fact 15 the most robust survey ever submitted for purposes of 16 setting Make Allowances, do you? 17 A. I don't believe Dr. Stephenson was asked that 18 question. I would trust his judgment. But he put a 19 mandatory audited survey at above the quality of that 20 survey. 21 Q. And if we could -- and everyone would like to have 22 that, and everyone knows we're not going to have that for 23 years, so -- 24 A. I disagree. 25 Q. -- so we're facing that reality. 26 Do you -- but you still haven't answered my 27 question. I mean, I asked whether this is the most robust 28 survey that's ever been used or submitted for use for 4511 1 purposes of setting Make Allowance in terms of the 2 percentage of the commodities covered. 3 Do you agree with that formulation? 4 A. The last time Make Allowances were updated was in 5 2008, which was more than half my life. So I am not aware 6 of the other surveys or how much more robust they would be 7 than the Stephenson one. 8 Q. Well, I mean, we just had extensive testimony on 9 that very question yesterday. Are you not aware of that 10 comparison that was provided as to the percentage of total 11 production that was captured by the surveys used to set 12 Make Allowances the last time as compared to the 13 percentage captured in the surveys that underlie the IDFA 14 proposals? Are you not aware of that testimony? 15 A. I am aware of the testimony. I'm also aware of 16 the testimony from Dr. Mark Stephenson that there were 17 variability and variance concerns amongst both surveys 18 that he ran. 19 Q. Well, okay. Are you aware that the percentage of 20 commodities captured by the surveys that underlie the IDFA 21 study are materially higher than the percentage of -- 22 captured by the survey used to set the current 23 Make Allowances? 24 A. Yes, I said in my testimony, Dr. Stephenson's 25 newer study captures a higher percentage of product plants 26 and volume. 27 Q. Okay. Materially higher. Do you agree with that? 28 I don't think you did -- actually, I don't think 4512 1 you did a comparison to the 2000 -- to the surveys that we 2 used to set the current Make Allowances, did you? 3 A. No, I did not. 4 Q. I missed that if it was there. Okay. 5 So -- but you do acknowledge that the percentages 6 are materially higher in the proposals now before USDA, 7 correct? 8 A. To my knowledge. 9 Q. And by the way, I mean, just -- do you know that 10 there were back-to-back hearings held in 2006 and 2007 11 that resulted in -- each resulting in increases in the 12 Make Allowances? 13 A. Yes. I -- in part of my research I went back and 14 dug into the Federal Register for that -- 15 Q. I mean, do you -- 16 A. -- to understand. 17 Q. Are you aware that the survey that underlay the 18 second Make Allowance revisions was conducted at a time 19 when everyone was aware of the use of the Stephenson 20 surveys for that purpose because the Stephenson surveys -- 21 an earlier Stephenson survey had just been used to set the 22 first of those two Make Allowance increases, correct? 23 A. Our position is just because those processes were 24 used in the past does not mean they were not flawed and 25 want to improve the system. 26 Q. No, but I'm focusing on your specific accusation 27 that, you know, oh, processors knew this time that the -- 28 when they were participating in the 2023 Stephenson 4513 1 survey, this was going to be used for Make Allowance 2 purposes. That's the accusation I'm focusing on. 3 I mean, are you aware that when -- when 4 Make Allowances were increased the second time in the 2006 5 through 2008 time period, that everyone knew that's what 6 the Stephenson survey was going to be used for? 7 A. I do not -- you know, I do not think that that was 8 not the case. I know, in that case, USDA might have made 9 final adjustments to a combination of numbers. We believe 10 that those numbers should be based on a mandatory audited 11 survey. 12 Q. Well, I know that. But, you know, you keep going 13 to that. I'm asking a different question. I'm asking 14 whether everyone was aware, when the second of the two 15 Make Allowances increased in the 2006, 2008 timeframe, 16 based upon a second Stephenson survey that picked up some 17 additional, more recent data, that everyone knew that's 18 what the survey would be used for? 19 A. I mean, I will say then I was not actively 20 involved in the dairy industry in middle school, but I 21 was -- I will not oppose that characterization. 22 Q. Did you go back and look at the transcripts or 23 anything like that before you started making these 24 statements? 25 A. I did. 26 Q. You did? 27 A. I reviewed the transcripts in the Federal 28 Register, but I did not -- well, I should say, I reviewed 4514 1 the rule in the Federal Register. The transcripts were 2 not available on the same web page. 3 Q. Did you -- okay. Did you try to reach out to 4 people who did participate and some of whom are still in 5 this room and have copies of those transcripts? 6 A. I'm not really sure how this relates. 7 Q. I'm just asking you whether you made an effort to 8 get hold of the transcripts. 9 A. In our drafting and in my research I communicated 10 with numerous individuals in this room to help 11 substantiate our research. 12 Q. And -- and did you ask any of them, hey, I would 13 like to -- before I start talking about whether people -- 14 I mean, you know -- what -- whether people were -- for the 15 first time in 2023 knew that the purpose for which the 16 survey was going to be used, did you say to yourself, gee, 17 I better check and see what people actually knew last time 18 around? 19 A. I mean, I think we're getting -- you know, that 20 part of the testimony was related to when Make Allowances 21 were first started. The knowledge was that they were -- 22 people were in -- participated in the survey wanting to 23 know all those costs. I do not contend that in later 24 iterations that they did not. 25 Q. Oh, so you -- your -- that part of your testimony 26 only relates to the survey that was done in, I don't know, 27 1997 or something like that, but not to the surveys that 28 were done in 2006 and 2007? 4515 1 A. I believe my testimony says "as used in" -- "as a 2 primary source for Order Reform." And when we refer to 3 Order Reform, I believe that's the 2000 or around 2000 4 time period. 5 Q. Now, you are aware that -- I mean, we have had 6 testimony here from Glanbia and Hilmar, which are the two 7 largest cheddar cheese manufacturers in the United States, 8 correct? 9 A. I don't oppose that characterization. 10 Q. And you -- you are aware that they testified that 11 they submitted their data from all their cheddar cheese 12 manufacturing plants for inclusion in the 2023 Stephenson 13 study. Are you aware of that? 14 A. I'm aware. 15 Q. I mean, you say, quote, "our members have 16 expressed ardent concern over plants who elect not to 17 participate in voluntary surveys that are used to inform 18 regulatory decisions," end quote, right? You say that? 19 A. That is true, they do. 20 Q. And have you ever said to them, oh, well, gee, 21 that's -- don't worry, the biggest plants with the most 22 efficient plants, they are in the survey? 23 A. They are not at these hearings, most of our 24 members. 25 Q. Okay. Well, have you told them -- I mean, you say 26 you have had recent discussions, like in the last week 27 with your members that -- where this particular issue has 28 been raised. Isn't that right? 4516 1 A. Yes. Our members are very concerned that the 2 survey does not encapsulate -- 3 Q. And have you -- 4 A. -- the majority of processors. 5 Q. And in those discussions, have you said to them, 6 we have sworn testimony from the two largest cheddar 7 cheese manufacturers in the United States, Hilmar and 8 Glanbia, that they submitted accurate testimony with 9 respect to the costs of manufacture of every single one of 10 their cheddar cheese plants? 11 A. I did not specifically go to them and say that, 12 no. 13 Q. I mean, have -- I mean, I would have thought your 14 members would want to have their concerns assuaged. 15 A. We are -- we are a not a top-down organization. 16 The intel from our organization comes up, not down. All 17 of our members interact with their processors, interact 18 with their trucking, interact with all the aspects of 19 their business more than I could ever know. I respect and 20 I believe in their characterization and the concerns that 21 they have. 22 Q. I mean, you have on page 3 a comparison of makes 23 and -- and costs, correct? 24 A. Yes. That is just for reference. 25 Q. Okay. I mean, obviously, the current is the 26 current, right? 27 A. Yes. 28 Q. It's not -- you are not suggesting that's -- 4517 1 reflects -- strike that. 2 That's the current Make Allowance, correct? 3 A. Yes, those are the current in the Federal Order 4 system. We're not suggesting that's what currently should 5 be. 6 Q. Okay. So that's like data from 2006, 2007? 7 A. That's what's in regulations right now. 8 Q. Okay. So I mean, it wouldn't be an expect- -- I 9 mean, you know, you have this chart that seems to show all 10 this variability, this Figure 2, in makes and costs and -- 11 but -- 12 A. The chart was honestly specifically for reference, 13 and honestly just to show the difference in the numbers 14 between the proposals. There was nothing really intended 15 by the variability beyond to compare the CDFA 2016 survey 16 that we believe was a more preferable methodology than 17 some of the other options. There was no other intention 18 by including the chart. 19 Q. Like IDFA-1 and IDFA-5, though, that simply 20 reflects the fact that IDFA, as an accommodation to 21 farmers, is asking that its Make Allowances be implemented 22 over four years, correct? 23 A. I believe so. 24 Q. Now, turn to page 4 where you talk about the NDPSR 25 only covering a certain percentage of the commodity 26 production, correct? 27 A. Correct. 28 Q. I mean, that's simply a reflection that USDA 4518 1 after, I'm sure, very careful consideration established 2 criteria for inclusion in that survey, correct? 3 A. Correct. The sole reason for me including this 4 section is that our members are constantly told the story 5 of basically a positive feedback loop of if you increase 6 prices, it is going to be in the survey, and it can't go 7 to the farmer. 8 This is only there to show that in all cases that 9 is not exactly true. In some cases it is. Our members 10 appreciate and understand the importance of 11 Make Allowances, but there are plants who can reap the 12 benefits of higher prices. 13 Q. Well, let's assume -- you know, I'm sure, that, 14 for example, the NDPSR survey for cheese has cutoff dates 15 for how old the cheese can be, correct? 16 A. Correct. 17 Q. And if it's over, I think it is 30 days, you don't 18 get -- it's not a reportable transaction, correct? 19 A. Correct. 20 Q. If it's sold more than 30 days after the date of 21 manufacture, correct? 22 A. Correct. 23 Q. I mean, so I take it you're not suggesting the 24 NDPSR survey fails accurately to capture the transaction 25 prices for the cheese that's within the survey, correct? 26 A. We are not saying that at all. 27 Q. I mean, are you -- I mean, do you -- do you -- 28 really -- and so everyone knows what that reported price 4519 1 is, right? I mean, buyers of cheese, sellers of cheese, 2 everyone knows the reported price, correct? 3 A. Yes. It's in the survey. 4 Q. I mean, are you suggesting that a cheese 5 manufacturer in the real world is somehow able to garner 6 some really higher price if it holds onto its cheese for 7 31 days? 8 A. We are suggesting that there's a large proportion 9 of varieties of cheese and volume not captured in the 10 survey, and that the prices on those goods are not 11 included in any regulatory process to set minimum pricing 12 for farmers. 13 Q. And -- 14 A. So they can reap benefits of higher prices, yes. 15 Q. In my specific example, are you seriously 16 suggesting that a maker of commodity cheddar cheese can 17 find a buyer to pay a materially higher price than the 18 NDPSR price simply by holding on to the cheese for 19 31 days, causing that cheese to fall outside the survey? 20 A. I mean, I'm sure there's nuance around the 21 situation that you describe. I couldn't say yes or no. 22 Q. Would you hazard a best guess? 23 A. I would say in one day after, not as likely to be 24 different. But that's not really the point of the 25 paragraph in the testimony. 26 Q. Well, I mean, presumably, to the extent that -- I 27 mean, I don't doubt other cheeses are sold for other 28 prices, higher or lower, but I mean, don't you assume that 4520 1 that -- the extent there's a higher price that could be 2 garnered, that's going to be reflected in higher costs, in 3 all likelihood? 4 A. Can you repeat that question? 5 Q. Yes. I mean, if you are going to take that 6 cheddar cheese and age it or do something else with it, 7 that causes it to fall outside the NDPSR definitions? 8 A. Right. I'm not opposing -- 9 Q. Isn't there likely to be a cost associated with 10 that? 11 A. I'm not opposing that there aren't increased costs 12 associated with other varieties of specialty cheeses, but 13 they can also sell those products at much higher prices 14 that are not looped into the regulatory price for farmers. 15 That's all -- that's the purpose of the section. 16 Q. And -- and you haven't done any examination 17 suggesting there's not a competitive market for those 18 other cheeses, are you? 19 A. I believe there is competitive market depending on 20 where you are for some of those areas. In some regions of 21 the country, you have farms producing specialty cheese, in 22 a very isolated area, and it is not a competitive market, 23 depending on the region. There's a lot of other factors 24 at play here, wealth income of a certain region, the 25 consumers you are targeting. There's a lot of other 26 factors at play. 27 Again, the sole purpose of this was to say that 28 there are other options for some plants to get the 4521 1 benefits of higher prices, and our farmers are 2 consistently told that no one can increase prices unless 3 the Make Allowances are higher, which is not the case. 4 Q. I mean, you are aware, obviously, that there have 5 been proposals along the way as to whether or not other 6 cheeses should be included in setting Make Allowances and 7 USDA has consistently rejected those proposals? 8 A. USDA's precedent in the past doesn't mean that 9 they can -- I mean, that's why we're here today. We're 10 here to amend Federal Orders to make them better. That 11 means that everybody agrees that there's a fault in them, 12 so I believe they can do better. 13 Q. And do you -- and you say in reference to the 14 California survey on page 4, you say, quote, "Although it 15 did not" -- let me start that question again. 16 I believe at the top of page 4 you are referring 17 to the CDFA survey that was last conducted in 2016, 18 correct? 19 A. Correct. 20 Q. And you say, and I quote, "Although it did not 21 evaluate product yields, this survey would provide a basis 22 for a conservative one-time increase in FMMO 23 Make Allowances, preferable to those proposed by NMPF or 24 IDFA." 25 Do you see that? 26 A. Yes. 27 Q. So your suggestion is that that USDA use 2016 data 28 to set current Make Allowances? 4522 1 A. Ms. Hancock previously asked questions about this. 2 Our sole intention of including information on the 2016 3 CDFA survey was to say that it was preferable in 4 methodology to the other options. 5 Q. Are you say -- 6 A. AFBF believes that of all the options, the CDFA 7 survey from 2016 is the optimal option that we have in 8 terms of methodology. 9 Q. The optimal option is to use seven-year-old data; 10 that's your view? 11 A. It's the only mandatory and audited survey. 12 That's -- our -- that's the only purpose of including 13 that. That's the last time there was a mandatory and 14 audited survey. We have already expressed our concerns 15 with the regional isolation of the survey. 16 Q. But -- 17 A. But, yes, we believe in that. 18 Q. That's the -- that's the -- that's what they 19 should -- that's what USDA should do, use seven-year-old 20 data? 21 A. It's the preferable methodology. 22 Q. I mean, you -- you -- okay. 23 THE COURT: Wait a minute, I'm not sure we got a 24 clear answer to that. 25 Are you talking -- you are talking about data; 26 you're talking about methodology. Are we overlapping 27 here? I mean -- 28 MR. ROSENBAUM: Let's be -- well, your Honor -- 4523 1 THE COURT: Nail it down. 2 MR. ROSENBAUM: Your Honor -- 3 THE COURT: I understand there's a certain 4 repetition here. 5 THE WITNESS: The option -- 6 MR. ROSENBAUM: I should -- 7 (Court Reporter clarification.) 8 THE COURT: I'm sorry, one at a time. 9 MR. ROSENBAUM: I'm -- you know, please. 10 THE COURT: I want to hear what you just said. 11 MR. ROSENBAUM: I should allow your Honor to take 12 over. 13 THE COURT: No, I don't want to -- you're both 14 working hard. 15 BY MR. ROSENBAUM: 16 Q. The -- okay. You -- your -- your view is that 17 USDA should adopt, as the cheese Make Allowance, the 18 survey report from 2016 -- 19 A. Our position -- 20 Q. -- conducted by the state of California? 21 A. Our position is ardently in support of a mandatory 22 and audited cost survey. Of the existing other options, 23 we believe that that is the best option. But our position 24 is for data from a current national mandatory and audited 25 survey as I have stated. 26 MR. ROSENBAUM: That's all I have. Thank you. 27 THE COURT: Further questions? 28 /// 4524 1 CROSS-EXAMINATION 2 BY MR. ENGLISH: 3 Q. Good morning. Chip English for the Milk 4 Innovation Group. 5 So in response to questions from Mr. Rosenbaum, 6 you -- and also Ms. Hancock, you said that you included 7 Figure 2 for comparison purposes, correct? 8 A. Correct. 9 Q. I just want to note and see if you would like to 10 confirm, for CDFA 2016, under cheese, you have 0.2354. 11 If the exhibit submitted by Dr. Schiek has current 12 for 2016 of .2454, the record will speak for itself, but 13 if that's the case, then that should be increased by 14 $0.01, correct? 15 A. I accept that revision to that. 16 Q. Okay. So even though you are talking about using 17 that for a reference point -- and I'm going to focus a lot 18 of my attention this morning on cheese -- the CDFA 2016 19 number of .2454 is higher than IDFA for the first year, 20 correct? 21 A. Correct. 22 Q. Okay. And I understand your responses, and I'm 23 not going to go back over them, to Ms. Hancock and 24 Mr. Rosenbaum, about the theoretical approach. 25 But leaving aside the theoretical approach 26 comment, you are certainly not suggesting the costs have 27 gone down since 2016, are you? 28 A. No, I do not have that data available. 4525 1 Q. Well, but you have seen data from a number of 2 witnesses that costs have gone up, correct, since 2016? 3 A. Correct. I have seen that. 4 Q. Did Dr. Schiek find in his testimony that the 5 information you provided had quote/unquote statistical 6 significance? 7 A. It did. 8 Q. Okay. 9 A. I will note that in a regression analysis, a high 10 R-squared value with a small number of observations often 11 is a sign of overfitting. So that's -- that often is not 12 necessarily just because it has high statistical 13 significance, and the regression analysis doesn't mean 14 that it is not an overfitted model. 15 Q. But nonetheless, he found statistical 16 significance, correct? 17 A. Correct. 18 Q. Thank you. 19 So I want to go back and focus on a couple things. 20 First, in response to a question from Mr. Rosenbaum about 21 the timing for what might happen in the future with 22 respect to audited surveys, I think I heard you say you 23 disagreed with him on how long it would take. 24 A. I disagree with the characterization that a Farm 25 Bill will not happen in four years, say. I think it will 26 happen before then. 27 Q. Well, but isn't -- well, how long would you think 28 a Farm Bill will happen? 4526 1 A. Our hope is to have it done by the end of the 2 year, if not by -- before June. 3 Q. Okay. So when was the last time a Farm Bill was 4 passed, you know, without an extension, on time? 5 A. Well, the timeframe I mentioned does not say that 6 it was passed on time. If it's passed before June of next 7 year -- you know, the 2019 Farm Bill was not passed 8 exactly on time. 9 Q. Okay. So, now, did you follow some of the 10 discussion, and Mr. Brown talked about this yesterday, 11 assuming even that a Farm Bill were passed say in June of 12 year, you understand, of course, that USDA then has to put 13 the program together, correct? 14 A. Correct. 15 Q. And they have to end up surveying plants that have 16 never been surveyed before, correct? 17 A. Correct. 18 Q. And then they have to audit all that, correct? 19 A. Correct. 20 Q. And then they have to come up with results of that 21 study, correct? 22 A. Correct. 23 Q. And then the industry has to talk about it, 24 correct? 25 A. Correct. 26 Q. And then somebody has to submit a hearing 27 proposal, correct? 28 A. Yes. 4527 1 Q. Okay. So USDA hired Dr. Stephenson originally for 2 the study that came out in 2021 in 2018, correct? 3 A. Correct. 4 Q. And we're now five years past that, correct? 5 A. Correct. 6 Q. So I want to go back because I don't think you 7 actually answered the question. And I apologize that was 8 my fault. 9 When was the last time a Farm Bill was passed on 10 time? 11 A. I do not recall off the top of my head. I want to 12 say 2002, but that could be incorrect. 13 Q. Okay. So, now, I want to go back to your comments 14 that you made a couple times that a number of large 15 processors did not participate in the Stephenson survey of 16 2023. I want to focus on cheese. 17 You agree that Hilmar testified that it 18 participated, correct? 19 A. Yes. 20 Q. You agree that Saputo testified that it 21 participated, correct? 22 A. Yes. 23 Q. You agree that Leprino testified and participated, 24 correct? 25 A. Yes. 26 Q. You agree that Glanbia testified and participated, 27 correct? 28 A. Yes. 4528 1 Q. And you agree that AMPI testified and 2 participated, correct? 3 A. Correct. 4 Q. And we also heard that Land O'Lakes participated, 5 correct? 6 A. Yes. 7 Q. Okay. Which large cheese manufacturer did not 8 participate? 9 A. I do not know off the top of my head, but there's 10 a substantial amount of cheese not in the volume of his 11 survey. It said, I have in my testimony, what was it, 50 12 or 45%. 13 Q. But that doesn't necessarily mean large ones 14 didn't participate, did it? 15 A. No, but it means that a substantial amount of 16 volume of all those plants was not represented. 17 Q. But your testimony was that large processors did 18 not participate. 19 A. I'm sure some of those processors are very large. 20 I mean, we have heard of very large co-ops that did not 21 participate that produce cheese. 22 Q. You have heard, what -- 23 A. I believe DFA did not participate, and they 24 produce cheese. 25 Q. Do you know if they are large compared to these 26 other entities for their plants? 27 A. I do not have the volumetrics on the top of my 28 head. 4529 1 Q. And speaking about motivations to participate or 2 not participate, since DFA opposes IDFA's proposal, your 3 own surmise about intentions for not participating could 4 lead to a negative inference that if they participated, 5 their costs would have been higher, correct? 6 A. Can you repeat the question? 7 Q. You, in your statement, and in response to 8 Mr. Rosenbaum, discussed what you viewed as an 9 "unfortunate" -- I'm sorry. I apologize. I went to the 10 previous paragraph. 11 "Manufacturers to be selective in their choice to 12 participate." Given the fact that DFA opposes the IDFA 13 proposal, isn't it a logical conclusion that had they 14 participated, their higher costs would have raised the 15 survey results? 16 A. If they participated and they had higher cost data 17 that went into the survey, yes, it would have, 18 hypothetically, increased those numbers. 19 MR. ENGLISH: I have no further questions. 20 CROSS-EXAMINATION 21 BY MR. MILTNER: 22 Q. Good morning, Mr. Munch. 23 A. Good morning. 24 Q. I'm Ryan Miltner. I represent Select Milk 25 Producers. You knew that. 26 Three areas that I think I wanted to ask you 27 about, and I wanted to start on page 4 of your testimony, 28 the second paragraph from the bottom. And you -- you 4530 1 speak about handlers who diversify their operations, a 2 tactic that many farmers are told to use to protect 3 against revenue uncertainties. 4 So my first question on that sentence is, tell me 5 a little bit about the farmer side of that equation 6 because I'm not sure -- not sure what you fully intend to 7 convey there. 8 A. Right. So the main point of the inclusion in that 9 sentence is, our farmers have often reported to us that, 10 you know, one of the things that they are told by 11 processors, by, you know, consultants, by other experts 12 or, you know, people providing advice to their operation, 13 is that if they want to protect against revenue 14 uncertainty, then they should diversify their operation. 15 And that could mean maybe investing in on-site bottling. 16 Maybe that means growing a different crop. Maybe that 17 means now you grow your own feed, or you invest in higher 18 technology, or you increase your variety of products sold, 19 you add a farm store. 20 So they are told often that they need to invest in 21 this diversification on farm to protect against revenue 22 uncertainty. 23 The main -- the main point to include in that 24 sentence is that we believe processors should also be held 25 to the same standard, if they want to, you know, protect 26 against revenue uncertainty, like farmers, they should 27 have to do the same sort of diversification to compete. 28 Q. So now on the processor side, have you been here 4531 1 or listening to most of -- most of the hearing testimony? 2 A. I was here all of last week and the first three 3 days. I have been in and out listening the rest of the 4 time. 5 Q. Okay. So do you recall any processor that did not 6 have some sort of diversification in their products 7 produced? 8 A. I believe most testified on the fact that they can 9 produce other products. So I believe many have 10 diversified. There are some very large and, you know, 11 recently that, you know, they -- they specialize in 12 particular commodity products, and we have heard many of 13 those testify that they are restricted by the system. So 14 when I speak of this, those are sort of some of the ones 15 that I'm speaking about specifically. 16 Q. The secondary I wanted to ask about related to the 17 timing of any mandatory survey, and I think you have -- 18 Mr. English asked a good bit about that. But is it -- is 19 it your testimony that once the Farm Bill is passed, and 20 the regulations and procedures are adopted, you would 21 expect it to be a two-year process to -- from the 22 beginning of plant surveys until -- until a change was 23 adopted or until a hearing could begin to confirm those 24 changes? 25 A. I believe the intention was for when a hearing 26 would begin, they could collect that data within that time 27 period. 28 Q. Okay. So I'd now like to ask you some questions 4532 1 about the various studies on Make Allowances that have 2 been discussed that were presented at this hearing. 3 And so where I wanted to start was by asking you, 4 as an economist, is there a difference in your mind among 5 these three terms, which have all been talked about: A 6 survey, a sample, and a census? 7 A. There is a difference. I mean, a sample is the 8 sample of thing that you are surveying. A census, you are 9 utilizing a much broader subset of data. I mean, Census 10 of Agriculture is one that I'm in every single day. I 11 don't have the exact definitions, but they are different. 12 Q. So I mean, when you are doing a -- when you are 13 trying to obtain data about a population, am I correct 14 that a census is intended to gather that data from every 15 member of that entire population? 16 A. To my knowledge, I would not disagree with that. 17 Q. And then if you are going -- if you are 18 surveying -- or if you are gathering data from a smaller 19 set of the total population, would that be a survey? 20 A. I would not disagree if that's the definition you 21 found. 22 Q. Okay. 23 A. I believe that a survey can be of the majority of 24 a population. So it kind of depends on the situation. 25 Q. It could be a majority but -- but a subset of the 26 total -- 27 A. Right. 28 Q. -- population? 4533 1 A. It kind of -- yeah. You could have a very small 2 population and have the bulk of it within a survey, just 3 because of that's how the project works out. But, yes, I 4 agree with that characterization. 5 Q. So if you wanted to obtain the best information 6 about the total population, between a census and a survey, 7 if you are looking at this for economic analysis or other 8 purposes, which of those two is going to give you the most 9 reliable outcome? 10 A. Census. 11 Q. And was the California data, to your knowledge, a 12 census or a survey? 13 A. I believe it was a survey. 14 Q. You believe it was a survey? 15 A. Yes. 16 Q. Okay. Within the world of surveys, when you are 17 going to select a sample, are there -- are there benefits 18 to taking a random sampling or a stratified sampling 19 versus just opening the doors to whoever wants to 20 participate? 21 A. Yes. 22 Q. What would -- in terms of the validity of the 23 results, what -- as an economist, what are the benefits to 24 a random sampling versus just opening the doors? 25 A. It gets rid of potential for bias. I mean, that's 26 the main reason. 27 Q. And what about -- 28 A. In your sample size. 4534 1 Q. And what about a stratified sampling versus just 2 opening the doors? 3 A. It kind of allows you to see different subsets 4 of -- in sort of an unbiased manner. 5 Q. There were some questions asked of you about 6 Dr. Stephenson's report from the 2006, 2007 hearings. 7 Those, as I believe Dr. Stephenson testified, were 8 stratified samples from his pool of data. And the 2023 9 report was one where it was by invitation to a subset of 10 the population. 11 Now, putting aside the fact that there are two 12 very different time periods, as an economist, would you 13 find one of those two approaches more or less valid than 14 the other? 15 A. I mean, I'm not intimately aware of the 16 methodologies, but the stratification piece is preferred, 17 I would say. 18 Q. And so if in relying on a stratified survey versus 19 one that was non-stratified, and you were a regulator, 20 would you feel quite as comfortable relying on one over 21 the other? 22 A. I mean, I believe the stratified data is more 23 appropriate as a basis. But, again we're in support of a 24 mandatory audited survey amongst all processors. That's 25 AFBF's position. So I don't really want to comment on if 26 USDA has to decide, well, you know, which -- because 27 that's our stance. That's where our members stand. 28 MR. MILTNER: I think that's all the questions I 4535 1 had. Thank you. 2 THE WITNESS: Thank you. 3 THE COURT: Questions from anyone else? 4 AMS -- oh, I'm sorry, Ms. Hancock. 5 CROSS-EXAMINATION 6 BY MS. HANCOCK: 7 Q. Let's see. Good morning, Mr. Munch. I just want 8 to first start off by thanking you for your level of 9 professionalism and maintaining your professional demeanor 10 throughout your examination. I think it's important for 11 this process that we do that. 12 You had some questions about whether you were 13 aware of Land O'Lakes' testimony on -- on its increases to 14 its own Make Allowance. 15 Do you recall those questions? 16 A. I do. 17 Q. And were you here when Mr. Edmiston testified? 18 A. I was not, but I might have been listening. 19 Q. Okay. Mr. Edmiston testified that Land O'Lakes 20 had experienced an increase in its own Make Allowances and 21 the percentages that it provided with respect to its own 22 increases. 23 Is that how you understood the percentage to 24 increase to be characterized to you in your questions? 25 A. Not specifically, no. 26 Q. Okay. And I'll represent to you that Land O'Lakes 27 never testified about how much, if any, its own costs had 28 increased over the current ly set Make Allowance. 4536 1 Does that make sense what I'm comparing there? 2 A. Yes. 3 Q. And we did ask some other IDFA members at what 4 point in time their costs had exceeded the currently set 5 Make Allowance. 6 Were you here for that testimony? 7 A. I was, yes. 8 Q. And did you hear them say that it was within the 9 last four years? 10 A. Yes. 11 Q. Would that suggest to you that if the 12 Make Allowances haven't been increased in 15 years, and it 13 was only the last four years that -- that their actual 14 costs had exceeded the Make Allowance, that they enjoyed 15 the benefits of the profits that they could obtain by 16 beating that Make Allowance for 11 years? 17 A. That would make sense. 18 Q. Would it also suggest to you that perhaps 15 years 19 ago, those Make Allowances numbers were set too high? 20 A. I would not disagree with that. 21 MS. HANCOCK: Thank you. That's all I have. 22 THE COURT: Yes, Mr. Rosenbaum. 23 CROSS-EXAMINATION 24 BY MR. ROSENBAUM: 25 Q. Steve Rosenbaum from the International Dairy Foods 26 Association. 27 I take it you have not examined the 28 Land O'Lakes -- 4537 1 A. I have not. I'm not intimately familiar with 2 either of the questioning or -- 3 Q. So if, in fact -- so you are not in a position to 4 say whether in fact that exhibit explicitly says that he's 5 tracking costs as compared to the Stephenson 2007 survey? 6 A. I am not familiar enough with the testimony to 7 make -- but I feel like the record will stand for itself. 8 Q. Or to the 2008 Make Allowances, you don't know 9 what -- whether he -- that that's what the document shows? 10 A. No. Not off the top of my head. 11 MR. ROSENBAUM: That's all I have. 12 THE COURT: AMS? 13 Should we take -- we have been going an hour and a 14 half, I guess. Should we take a break of ten minutes? 15 Let's come back at 9:40. 16 (Whereupon, a break was taken.) 17 THE COURT: Let's come to order. Let's get 18 started. 19 AMS. 20 CROSS-EXAMINATION 21 BY MS. TAYLOR: 22 Q. Good morning. 23 A. Good morning, Erin. 24 Q. I want to thank you for being flexible from last 25 week and coming back to testify today. So thank you very 26 much. 27 A. Thankfully there's lots of flights from DCA. 28 Q. I really don't have a lot of questions for you 4538 1 that haven't been asked. 2 So you -- I do have a few. 3 On page 3, the paragraph before the chart, you 4 talk about using different -- the different surveys that 5 have been done in the past. "This is more art than 6 science, and USDA was put in the difficult position of 7 applying such art in the past." 8 I was wondering if you could just talk a little 9 bit about what you mean there for the record. 10 A. Yes. So, to my knowledge, when they -- 11 Make Allowances were last adjusted, USDA basically -- I 12 mean, like you are doing now, is pull this different 13 information from what's been presented, and all the other 14 information that's been on the record, and make a final 15 decision. So we're kind of characterizing that as an art 16 of what's the best method to do that that results in the 17 most optimal conclusion or -- or proposed rule. 18 Q. Okay. And your preference, as you said 19 throughout, is instead of doing that, is an audited 20 mandatory survey? 21 A. Right, have that information based on that. 22 Q. Okay. On the same page, Chart 2, you had a note 23 at the bottom of the chart: "'CDFA 2016 Whey' is the CDFA 24 2016 NDM plus current difference." 25 Could you explain what that is? What the current 26 difference is? 27 A. I have to be honest, that was a Roger plop in, 28 so... 4539 1 Q. Okay. Well, if Roger gets up on the stand later, 2 I'll put a note to ask him that question. 3 A. I apologize. I should have clarified that. 4 Q. No problem. 5 On page 4 in the middle, that big paragraph in the 6 middle, you are talking about, generally, the amount of 7 commodity that is covered by the NDPSR. You say in the 8 one sentence, 10% of butter, about 10% cheese, 28% of 9 whey, and 52% of nonfat dry milk is captured in the NDPSR. 10 Just could you illuminate what you used to get 11 those calcula- -- get those percentages? 12 A. Yes. So for those ones in particular -- are those 13 ones -- I believe since 2000 I used -- when NASS covered 14 the survey, I took that volume, and then I used once it 15 switched over to AMS, I took that volume, and then divided 16 what's in the NDPSR survey volume by each of those 17 categories. The cheese ones included both the 40-pound 18 and obviously 500-pound barrels, against all cheese. 19 Q. And "all cheese" being the NASS all cheese 20 volumes? 21 A. Yes. 22 Q. Okay. So it is NDPSR over the NASS numbers? 23 A. Correct. 24 Q. Okay. And then on the last page of your 25 statement, when you are talking about a note on using 26 input price indices, in the middle it says, "Over time, 27 input price increases tend to be at least partially offset 28 by productivity increases." 4540 1 Could you just explain a little bit more about 2 that? 3 A. Yes. So as -- as technology advances, as, you 4 know, you have new optimal workplace processes depending 5 on the particular operation that you have, a lot of those 6 efficiencies offset some of the inputs that you have. So 7 in -- in many cases you will see sort of a deflation in 8 certain costs associated with the same amount of effort by 9 a particular amount of labor. 10 Q. And -- sorry. 11 A. No, you go ahead. 12 Q. So in using those indices, then it's your position 13 that those efficiencies aren't captured fully because you 14 are still using the energy index, for example, or -- 15 A. Right. 16 Q. But I ask because I think I have heard previous 17 testimony that tends to say that those efficiencies are 18 captured in the labor index. But you would argue not 19 fully? 20 A. Well, and -- and particular in Dr. Schiek's study, 21 he uses broad indices instead of specific indices, such as 22 dairy plants, so there's a concern there. But also, just 23 broadly, we have policy opposed to indexing just because 24 we don't believe that it takes in all the factors into 25 consideration. 26 Q. Okay. That is a Farm Bureau policy position? 27 A. Yes. 28 Q. Okay. 4541 1 A. That is directly from our book. 2 Q. Okay. Thank you. 3 And just one last question for clar- -- to 4 clarify. Farm Bureau supports an audited mandatory cost 5 survey? 6 A. Correct. 7 Q. But that is not a proposal here at this hearing 8 that's been discussed. 9 So do you have a position or are you not having a 10 position on Proposals 7, 8, and 9? 11 A. We oppose those proposals. 12 Q. You do oppose them all? 13 A. Yes. 14 Q. Okay. 15 MS. TAYLOR: That's it from AMS. Thank you. 16 THE COURT: Redirect. 17 REDIRECT EXAMINATION 18 BY DR. CRYAN: 19 Q. Roger Cryan for American Farm Bureau Federation. 20 Thank you, Mr. Munch, for your testimony. 21 I would like to thank Mr. English for pointing out 22 the error. As you said, that was -- that was my 23 assignment. That chart was my assignment, and that's an 24 error. That number is a cent higher. 25 But would you -- does that change -- change the 26 testimony in any -- in substance? 27 A. No. The change in the number, you are saying? 28 Q. Right. 4542 1 A. No, we think it still reflects what was found and 2 determined in that survey. 3 Q. Because it is a matter of principle, not a matter 4 of the individual number? 5 A. Correct. 6 DR. CRYAN: And just for the record, this is -- 7 this testimony was a joint product. It is -- it could 8 have been equally delivered by Mr. Munch or myself, but it 9 had the stamp of each of us. 10 BY DR. CRYAN: 11 Q. Are you accusing anyone of lying or deliberate 12 deception in any part of this process? 13 A. I am not. 14 Q. I think it's from the record, the -- from the 15 record, from the previous hearings, from this record, it 16 is relatively clear that when Dr. Stephenson collected 17 data in 1998, there was -- there was a -- it was based on 18 a tradition of benchmarking. 19 And as -- as time goes on and folks presumably 20 begin to understand more and more that it is going to be 21 used for regulatory purposes, how would you describe the 22 way survey results change based on the understood purpose 23 of the survey? 24 A. You mean -- no. As sort of referred to earlier, 25 depending on what a plant might see as an optimal result 26 in sort of their long-term revenue goals or how that 27 impacts their individual business, they are going to 28 choose whether or not to par- -- I mean, obviously, a 4543 1 plant with a very high cost, they have a strong incentive 2 to participate because they want those numbers reflected, 3 and they want the highest costs that they have to be 4 reflected. 5 On the other hand, there is incentive for the 6 lowest cost, highest efficiency to not participate or to 7 present numbers on the lower spectrum in their books. 8 So it really just comes up as, you know, there is 9 a strategy, people are going -- you know, this is -- you 10 know, people are trying to make money, and this is a 11 regulatory consideration that impacts the bottom line. So 12 they are going to choose to participate and choose to the 13 extent of which that they participate depending on their 14 intended goals and... 15 Q. So if you have a voluntary survey, it is not -- 16 it's a reasonable business decision for a plant to choose 17 to participate or not to participate? 18 A. I believe it is. 19 Q. And it doesn't represent -- that's not a lie or a 20 deception, it is simply a business decision to engage in 21 something or not? 22 A. Correct. It is a voluntary survey. Nobody is 23 forced to participate, as with any voluntary survey. 24 Q. There's been some talk about the proceedings in 25 2006 and 2007. 26 At that time, surveys by Dr. Stephenson were used, 27 but the CDFA numbers were available. 28 And what did that -- what did that mean in terms 4544 1 of the application of the Stephenson results? 2 A. So I believe USDA was utilizing the CDFA survey as 3 sort of a check to the voluntary survey to make sure the 4 numbers were in -- in -- in an appropriate range. So you 5 are using the improved preferable methodology that we 6 believe is preferable to the voluntary survey. 7 Q. You talked about in the survey -- in the 8 testimony, you talked about the temptation for creative 9 activity. 10 Would you say that has an impact on the perception 11 of the proceeding, whether anyone succumbs to the 12 temptation or not? 13 A. Right. So, you know, our farmers have expressed 14 extreme concerns over, you know, the reporting in these 15 voluntary surveys. It doesn't mean that the numbers are 16 wrong. That's not what we're saying. There is a concern 17 and there is perception that people are being deceptive. 18 And I think that's the problem that our members have. And 19 having an audited survey gets rid of -- you know, gets rid 20 of that perception, gets rid of the chance for people to 21 be deceptive. So that's really what our farmers want is 22 an assurance. It is not saying that the numbers aren't 23 right now. It is saying that now we know the numbers are 24 right, and we can go to our farmers with confidence that 25 they are right. And I think that's an assurance that they 26 deserve. 27 Q. The -- there was previous testimony, I believe it 28 was by Mr. Edmiston, that their co-op farmer members were 4545 1 asking them not to go too far repeatedly, that they -- 2 that that was a message they heard a number of times. And 3 I don't think it was only Mr. Edmiston. And then I asked 4 whether in seeking these changes they tried to find the 5 highest make increase -- the highest make that they're 6 assured isn't too high. 7 Would you say this is -- this is kind of the 8 objective in suggesting the CDFA numbers as an 9 alternative? 10 A. Yes. You know, we're looking at a decision that 11 even at the smallest percent of being off has a large 12 magnitude impact on our farmers' bottom lines. And some 13 of our farmers from last week, like Brian Henrichs, he 14 answered a question that said even a 1 to 2% difference in 15 the Make Allowance has an extreme impact on their bottom 16 line. 17 So we want to make sure whatever numbers are 18 utilized are definitely not too high, and given the -- you 19 know, the timing and the methodology behind that CDFA 20 survey, we are under the understanding that currently it 21 is not too high. 22 Q. And finally, Mr. Miltner asked some questions 23 about surveys and the nature of surveys and stratification 24 and random and self- -- self-selected, and the -- and the 25 concept that you kind of were walking around was 26 self-selection bias. 27 Would you talk about self-selection bias a little 28 bit? 4546 1 A. Yeah. So self-selection bias is really, you know, 2 when a particular population or group decides to 3 participate in a survey, and that's what happens when you 4 have a voluntary survey. You know, they might find some 5 sort of benefit out of having the information. It could 6 just be benchmarking, or it could be -- and, you know, we 7 argue that it could be to skew data or just have excess 8 data. And folks might not participate for the opposite 9 reason. 10 So any sort of situation or surveying situation 11 where people are choosing based on a decision or an 12 outcome or to have information to participate, you have 13 the self-selection bias, which negatively impacts the 14 quality of the results. 15 Q. Thank you. 16 DR. CRYAN: And I believe this is when I ask for 17 the exhibit to be recognized -- 18 THE COURT: Yes. 19 DR. CRYAN: -- as -- what was the number? 20 THE COURT: Exhibit 222. 21 Any objections? 22 Seeing none, Exhibit 222 is admitted into the 23 record. 24 (Thereafter, Exhibit Number 222 was received 25 into evidence.) 26 DR. CRYAN: Thank you, your Honor. That's all for 27 me. 28 THE COURT: Okay. You are dismissed. Thank 4547 1 you -- 2 THE WITNESS: Thank you. 3 THE COURT: -- Mr. Munch. 4 MR. MILTNER: Your Honor, we would call Steve 5 Cooper to the stand. 6 THE COURT: Please raise your right hand. 7 STEVEN COOPER, 8 Being first duly sworn, was examined and 9 testified as follows: 10 THE WITNESS: Good morning. My name is Steven 11 Cooper. I'm the president and chief operating officer of 12 Continental Dairy Facilities and Continental Dairy 13 Facilities Southwest. 14 DIRECT EXAMINATION 15 BY MR. MILTNER: 16 Q. Thank you, Steve. 17 Before we go further in your statement, which you 18 have got in front of you, and at the top that's marked 19 Exhibit Select-5, correct? 20 A. Correct. 21 Q. And you have seen this before? 22 A. Yes. 23 Q. This is the testimony that you have prepared on 24 Select's proposal regarding the nonfat dry milk yield, 25 correct? 26 A. Correct. 27 Q. Could you -- you have given your name, but could 28 you spell your name for the record and then give your 4548 1 business address as well. 2 A. Steven, S-T-E-V-E-N, Cooper, C-O-O-P-E-R. 3 Business address is 999 West Randall Street, Coopersville, 4 Michigan, 49404. 5 Q. And Randall was R-A-N-D-A-L-L, correct? 6 A. That is correct. 7 Q. So if you would go ahead and give your 8 introduction, and then I'll have some questions for you, 9 and then we can finish giving your statement. Okay? 10 A. Okay. 11 THE COURT: Let's go ahead and mark Select -- 12 Exhibit Select-5 as Exhibit 223 for identification. 13 (Thereafter, Exhibit Number 223 was marked 14 for identification.) 15 MR. MILTNER: Thank you, your Honor. 16 THE COURT: You may continue, Mr. Cooper. 17 THE WITNESS: CDF operates a butter/powder plant 18 in Coopersville, Michigan, and CDF Southwest operates a 19 butter/powder plant in Littlefield, Texas. Both CDF and 20 CDF Southwest are wholly-owned subsidiaries of Select Milk 21 Producers, Inc. 22 I have a Bachelor of Science degree in dairy 23 science and technology from California State University 24 Fresno, and before earning my Bachelor's degree, I earned 25 an associate's degree in dairy production from Mount San 26 Antonio College in Walnut, California. I have worked in 27 the dairy industry either on a dairy farm or in dairy 28 manufacturing since I was 16 years old. 4549 1 Following my graduation from Fresno State, I 2 worked for three years as an assistant plant manager with 3 Sonoma Cheese, and I was then hired by California Milk 4 Producers, which is now California Dairies, Inc., as a 5 cheese plant supervisor, working in a plant manufacturing 6 40-pound block cheddar cheese. 7 And for the last 15 years of my time with CDI, I 8 worked in its butter/powder plants, ultimately holding the 9 title of vice president of manufacturing operations, where 10 I had oversight responsibility of CDI's manufacturing 11 facilities. 12 I joined CDF in 2009, initially working with 13 Continental Dairy Products, Inc., a cooperative that has 14 since been merged with Select, and I was hired to work 15 with Continental as they were designing and building a new 16 Class IV balancing plant to serve its owner producers in 17 the Midwest marketing area. I was involved in the design, 18 construction, and commissioning of the plant, and have 19 been its general manager and COO since its inception. 20 In 2015, after Continental had merged into Select, 21 I served the same role in designing, constructing and 22 commissioning and operating a sister plant in Littlefield, 23 Texas to serve Select's owners in the Southwest Marketing 24 Area. 25 I am ultimately responsible for all dairy -- or 26 daily operations at both CDF and CDF Southwest, including 27 product manufacturing, milk sourcing, product sales, and 28 food safety. In conjunction with Select's management and 4550 1 our producer directors, I'm also responsible for 2 short-term and long-term strategic planning for both 3 facilities and related activities of the cooperative. 4 BY MR. MILTNER: 5 Q. Thank you. 6 So in -- in total, Mr. Cooper, how many years now 7 have you spent working in dairy manufacturing? 8 A. In manufacturing, alone? 47 years. 9 Q. And during that time, what types of facilities 10 have you worked in in terms of the types of products they 11 produce? 12 A. I started off in cheese, working in an 13 artisanal-style cheese plant in Sonoma, doing artisanal 14 cheeses. And then went to a block cheddar manufacturing 15 operation at California Milk Producers. And then from 16 there, went into butter and powder. And then the 17 facilities I've worked in have also produced cream, 18 condensed, ice cream mixes. 19 Q. And how much of your time has been in the 20 butter/powder plant part of that equation? 21 A. Most of my career. So all about -- except for the 22 six years I was in cheese, probably 40-plus years in 23 butter/powder. 24 Q. And the most recent 40 years, consecutively 25 dealing with -- 26 A. That's correct. 27 Q. -- butter/powder operations? 28 A. Yes. 4551 1 Q. Do you also serve on any industry committees? 2 A. Yes, I do. 3 Q. What are those currently? 4 A. I serve on ADPI, IDFA, U.S. Dec, Dairy Products 5 Institute of Texas. I think that's it. 6 Q. And is Continental Dairy Facilities or Select Milk 7 Producers a member of IDFA? 8 A. Continental Dairies is a member of IDFA. 9 MR. MILTNER: Your Honor, we would ask that 10 Mr. Cooper be recognized as an expert in the area of dairy 11 product manufacturing. 12 THE COURT: Any objections? 13 I find this witness to be an expert in that area. 14 MR. MILTNER: Thank you. 15 BY MR. MILTNER: 16 Q. Mr. Cooper, if you would go ahead and present the 17 rest of your testimony. And just for our court reporter's 18 sake, if you can slow down just a touch. We have all had 19 the tendency to read at pace, and it's hard for her to 20 take it all down. So thank you very much. 21 A. I'll slow it down. 22 The scope of my testimony today is related to the 23 production and sales of butter/powder in both the Michigan 24 CDF plant and the Texas CDF Southwest plant. In addition 25 to the analysis presented in this testimony, I have 26 reviewed and am familiar with testimony of Chris Allen on 27 this proposal, and my testimony is intended to supplement 28 and build upon his. 4552 1 Both the Michigan and the Texas plant produce a 2 variety of products. The predominant products produced 3 are nonfat dry milk, butter, and buttermilk powder. 4 Depending on market conditions, the plants will also 5 produce condensed skim milk, whole milk powder, and bulk 6 cream. 7 I was asked by Select to analyze and provide 8 testimony on the sales prices of low-heat nonfat dry milk 9 and buttermilk powder for the period of January 2021 10 through June 2023. Specifically, I was asked to analyze 11 the alignment of these prices at each plant for each month 12 in that period. I was also asked to discuss the costs of 13 manufacturing low-heat nonfat dry milk compared with 14 buttermilk powder. 15 CDF and CDF Southwest maintain records of the 16 prices received for all products sold, both on an 17 individual sale basis and aggregated data for the purposes 18 of internal reporting and analysis. I asked my staff to 19 compile the sales data described above for each facility. 20 I was provided the monthly average sales prices for 21 low-heat nonfat dry milk and buttermilk powder at each 22 plant. That data was then compared monthly for the entire 23 period. 24 Included in this statement are a series of tables 25 and graphs. The first table lists for the CDF Michigan 26 plant, for each month, the price received for buttermilk 27 powder as a percentage of the price received for low-heat 28 nonfat dry milk in the same month. The average, maximum, 4553 1 and minimum is also reported. Over the observed period, 2 buttermilk powder prices averaged 96.72% of nonfat dry 3 milk prices. The maximum buttermilk powder price was 4 115.54%, and the minimum was 82.51%. 5 The second table provides the same information for 6 the CDF's Southwest Texas plant. Over the observed 7 period, buttermilk powder prices averaged 96.59% of nonfat 8 dry milk prices. The maximum buttermilk price was 9 114.67%, and the minimum was 82.89%. In certain months, 10 the plants did not sell any buttermilk powder. No data is 11 reported for those months. 12 As an aside, for those months where no buttermilk 13 powder was sold, I compared the Dairy Market News reported 14 price relationship between low-heat nonfat dry milk and 15 buttermilk powder. For the four months when CDF did not 16 sell buttermilk powder, the price relationships for the 17 Central/Eastern reports were 95.90%, 89.17%, 119.18%, and 18 120.88%. 19 For the four months where CDF Southwest did not 20 sell buttermilk powder, the price relationships for the 21 Western reports were 90.59%, 94.44%, 96.45%, and 82.46%. 22 We present this data to demonstrate that the 23 decision not to sell buttermilk in those months was not a 24 function of a weak buttermilk market relative to nonfat 25 dry milk. I also note that the four months that CDF did 26 not sell buttermilk powder are not the same four months 27 when CDF Southwest did not sell buttermilk powder. 28 The next graph shows the buttermilk price index on 4554 1 a month-to-month basis for CDF Michigan. The next page 2 also shows the relationship for Continental Dairy 3 Facilities Southwest. 4 Related information is presented graphically. The 5 first graph plots the monthly sales price of buttermilk 6 powder for the Michigan and Texas plants expressed as a 7 percentage of the nonfat dry milk prices for the plants. 8 You will see close alignment in months where both plants 9 sold buttermilk powder. Graphs 2 and 3 plot the nonfat 10 dry milk price for the Michigan plant against the prices 11 for the Texas plant and the same analysis for buttermilk 12 powder. Again, you will see a close alignment between the 13 two plants. 14 In addition to examining the sales prices of 15 nonfat dry milk and buttermilk powder, I also examined the 16 plants' manufacturing costs for low-heat nonfat dry milk 17 and buttermilk powder. 18 I note that neither Continental Dairy Facilities 19 nor Continental Dairy Facilities Southwest participated in 20 the price surveys conducted by Mark Stephenson or Bill 21 Schiek. The reasons for our decision not to participate 22 will be testified to separately when Select addresses 23 Proposals 7, 8, and 9. Nor will I discuss during my 24 testimony the actual manufacturing costs of CDF or CDF 25 Southwest for nonfat dry milk or buttermilk powder. 26 However, I did examine whether the stated 27 manufacturing cost relationship noted by USDA during order 28 reform was accurate in my experience and in the operations 4555 1 of CDF and CDF Southwest. The order reform decision found 2 that it costs $0.02 more to make buttermilk powder than to 3 make nonfat dry milk powder. The actual process of drying 4 buttermilk and drying skim milk are essentially the same. 5 We utilize the same equipment and processes to make both 6 products. 7 The only difference is that it takes somewhat 8 longer to dry buttermilk than to dry skim milk. That 9 additional drying time requires additional natural gas. 10 While the specific additional cost would vary directly 11 with the actual costs of natural gas, the incremental fuel 12 cost to CDF and CDF Southwest in 2023 would be 13 approximately $0.02. 14 For both the CDF plant in Michigan and the CDF 15 Southwest plant in Texas, the actual prices received for 16 the sale of nonfat dry milk and buttermilk powder are 17 closely aligned. In addition, in no month was the price 18 for buttermilk powder more than 18% less than the prices 19 received for nonfat dry milk. And for CDF and CDF 20 Southwest, the reality is that buttermilk receives nearly 21 the same price as low-heat nonfat dry milk. 22 The prices received for nonfat dry milk produced 23 in Michigan and Texas, as well as the prices received for 24 buttermilk produced in each location, are also aligned. 25 There is very little difference between the prices 26 received for either product that could be attributed to 27 geography. Those differences that do occur on a 28 month-to-month basis are virtually nonexistent on an 4556 1 annual basis. There appears to be little regional 2 difference in the prices received. 3 The price information from Continental Dairy 4 Facilities and Continental Dairy Facilities Southwest is 5 consistent with the Dairy Market News data comparing 6 nonfat dry milk and buttermilk powder prices in both the 7 Western market and the Eastern/Central Market. The 8 average of the buttermilk powder price relationship of 9 97.5% testified to by Chris Allen is consistent with the 10 CDF/CDF Southwest price relationship of 96.7%. 11 CDF and CDF Southwest's costs to manufacture 12 buttermilk powder is higher than the cost to manufacture 13 nonfat dry milk. That additional cost is due to 14 additional drying time and associated fuel. Although the 15 costs will vary with natural gas prices, the Department’s 16 estimate that the make costs for buttermilk powder are 17 $0.02 higher than the make costs for nonfat dry milk are 18 appropriate. 19 Q. Thank you, Mr. Cooper. 20 A. Thank you. 21 Q. So I'd like to ask you a few more questions about 22 your statement. And I would like to start actually with 23 something you stated on page 8, where you stated the 24 reasons for our decision not to participate in those 25 surveys will be testified to separately. 26 Now, Select may have a statement at the end of the 27 proceeding addressing all the proposals in which that 28 issue would be addressed, but since you are here, and we 4557 1 can talk about it, I thought I would ask now. 2 Dr. Stephenson's study that was conducted in 2018 3 or so, do you recall receiving an invitation to 4 participate in that survey? 5 A. Yes. 6 Q. Did you participate in that survey? 7 A. No, we did not. 8 Q. And was the reason for your participation that it 9 wasn't a full census of all applicable plants, that all 10 plants would be required to participate in? 11 A. That, and it wasn't a mandatory, so we chose not 12 to participate. 13 Q. Do you recall if your plants were invited to 14 participate in the 2023 update to those surveys that IDFA 15 had commissioned? 16 A. I believe we were. 17 Q. And did you participate in those studies? 18 A. No, we did not. 19 Q. Was the reasoning the same? 20 A. Yes. 21 Q. Now, I don't want to ask about their operations. 22 But when you worked for CDI, did CDI participate in 23 those -- in the surveys of the California Department of 24 Food and Agriculture? 25 A. Yes. 26 Q. And why did they participate in those studies? 27 A. Because they were mandatory. 28 Q. Did you have a concern about the conclusions that 4558 1 were drawn by the California surveys when you worked for 2 CDI? 3 A. No, I did not. 4 Q. Would you have a concern with another survey that 5 was neither mandatory nor comprehensive? 6 A. I would. The California surveys were mandatory, 7 and they were audited, and they were very detailed. I 8 participated in them. So I know that they were extremely 9 detailed and -- I -- if another survey was done the same 10 way, I would -- I would participate in it, but if it was 11 not done the same way, I would not. 12 Q. I'd like to ask also about the production of 13 buttermilk powder at your two plants. 14 Now, you testified that in some months there was 15 no buttermilk powder sold. 16 A. Correct. 17 Q. Now, why would that occur? 18 A. In those months we did not produce butter. We 19 looked at cream, and we run a formula based on cream where 20 we sell cream as fluid cream or we sell -- if the 21 multiples are high enough, we'll sell the majority or all 22 of our cream for fluid cream. If the -- if the multiples 23 do not dictate that it's a good enough return, we'll run 24 butter. We typically run butter in months where the 25 multiples don't support selling fluid cream. We put that 26 butter away, and we use it for our retail operations. 27 Q. So I think during the course of this hearing there 28 may have been a couple passing references to cream 4559 1 multiples but -- so the record is clear, can you share 2 with us what a cream multiple is? 3 A. Basically a cream multiple reflects the market 4 conditions for the cream. So we take the CME butter price 5 and we take it times a multiple based on market demand. 6 And if we see that that return on -- and we take 7 buttermilk into that process as well. We look at the 8 return on buttermilk -- on the cream and on butter, and we 9 look at what brings the best margin to the plant. And we 10 make a decision whether we move that to fluid cream or we 11 move it into the butter operations. 12 Q. So let me walk through this, kind of step by step 13 and make sure that we have got this correct. 14 So raw milk arrives at one of the plants, and you 15 skim that -- that raw milk, correct? 16 A. Correct. 17 Q. And the skim then becomes nonfat dry milk, 18 correct? 19 A. Correct. 20 Q. And that for the most part happens every day? 21 A. Correct. 22 Q. The cream then, you and your team make a decision 23 about the best economic outlet for that cream; is that 24 right? 25 A. Yes, we do that. 26 Q. And when the cream multiple is high enough, the 27 economic decision is to sell the bulk cream; would that be 28 right? 4560 1 A. That is correct. 2 Q. And at a lower multiple, the better economic 3 decision is to make butter, correct? 4 A. Correct. 5 Q. So if you are not making butter, there's no 6 buttermilk to dry; is that right? 7 A. That's correct. 8 Q. So -- now, when you do make butter, do you -- do 9 you always make buttermilk powder? 10 A. Most of the time we make buttermilk powder, but we 11 have made buttermilk condensed. If the market conditions 12 are better to sell buttermilk condensed, we'll sell 13 buttermilk condensed, but that's very seldom. 14 Q. Now, you have had a chance -- I don't know if 15 you -- I don't think you were here when Chris Allen 16 testified yesterday. I think you were still in transit, 17 correct? 18 A. That's correct. 19 Q. He, in his testimony -- which I think you have 20 seen -- he talked about how the relationship between the 21 price of buttermilk powder and the price of nonfat dry 22 milk has converged, I suppose, over time. 23 Given your experience in dairy product 24 manufacturing and sales, do you have any observations or 25 reasoning as to why that price relationship might have 26 changed? 27 A. Yes. Buttermilk used to be used primarily in 28 baking, and it was really kind of a byproduct of butter 4561 1 making. In recent years, the food industry has realized 2 the attributes and nutritional attributes of using 3 buttermilk powder in formulations, such as infant formula. 4 One of our biggest customers is an infant formula customer 5 for buttermilk powder. 6 And what we used to sell buttermilk powder for was 7 a very -- at times, very, very cheap. We would lose money 8 on it. Now we're seeing a relationship where buttermilk 9 powder is, at times, more valuable than nonfat dry milk 10 because of the food industry starting to see those 11 nutritional attributes in buttermilk powder and 12 nutritional products, and they put formula in other things 13 rather than just baking. 14 Q. So the demand profile for that product has changed 15 over the last 20 years or so? 16 A. Yes. 17 Q. The difference -- 18 A. Especially in the last two to three years, that 19 demand -- that demand profile and the pricing has changed 20 dramatically. 21 Q. Now, based on the fact that the solids in 22 buttermilk or buttermilk powder are not included in the 23 price formulas right now, other than the cost to process 24 the buttermilk, does that represent pure profit to your 25 plant? 26 A. It certainly helps it. I'm not sure how to 27 comment on that. 28 Q. You can stick with "it helps," if you like. 4562 1 A. Yes. 2 Q. And just a couple of things to clarify in your 3 statement, as you have presented it. 4 On page 4 and page 5, where you see the word 5 "null," are those the months where there was no buttermilk 6 sold by the plants? 7 A. That's correct. 8 Q. And then on your graphs on page 7, there is no 9 legend for the Y axis. And I understand that is because 10 you did not want to particularly peg the exact sales 11 prices on that -- those graphs; is that correct? 12 A. That is correct. 13 Q. Okay. So I have one more question for you, and it 14 relates not so much to this statement, but to a question 15 that was asked of Ms. Stehouwer yesterday. And it has to 16 do with the component tests of the farm milk coming into 17 your plants and the component tests of the silos at your 18 plants. 19 Now, I'm not asking you for the numbers 20 themselves, but she stated that the -- those numbers, 21 whatever they are, are aligned, that the components you 22 realize from the farm tests are the same components you 23 realize when you test your silos. 24 Is that your experience? 25 A. Yes. We track that, and we are very confident 26 that those are aligned. 27 Q. Now, similarly, through whatever other measures 28 you utilize in making products, to the extent there are 4563 1 solids losses through the process, do you find that 2 butterfat losses exceed those of solids in total? 3 A. No. We -- we have set KPIs that we look at and 4 the plant is measured on, the management staff is measured 5 on. We have set guidelines for solids nonfat, and we have 6 set guidelines for butterfat. And the butterfat solids -- 7 the butterfat losses that are acceptable are lower, the 8 standard is lower than the -- on a percentage basis than 9 solids nonfat. 10 Q. So when you are comparing farm components and silo 11 components, they line up? 12 A. Yes. 13 Q. And then once you start processing, you find that 14 solids -- other solids are actually lost at a higher rate 15 than butterfat. Is that what your experience is? 16 A. That's my experience, yes. 17 Q. Okay. 18 MR. MILTNER: Your Honor, I don't have any other 19 questions, and we would make Mr. Cooper available for 20 questioning from others. 21 THE COURT: Questions for this witness? 22 Mr. Rosenbaum. 23 CROSS-EXAMINATION 24 BY MR. ROSENBAUM: 25 Q. Steve Rosenbaum for the International Dairy Foods 26 Association. 27 You testified that you chose not to participate in 28 either the 2019 Stephenson survey or the 2022 Stephenson 4564 1 survey; is that correct? 2 A. That is correct. 3 Q. So we have heard testimony that the 2023 4 Stephenson survey of 2022 costs covered 91.2% of total 5 nonfat dry milk production in the United States. I'd ask 6 you to accept that number. Not -- obviously you're not in 7 a position to verify that. 8 A. No. 9 Q. Would -- would Select be a material component of 10 the -- of the seven -- excuse me -- of the 8.8% that 11 didn't participate? 12 A. Most likely. 13 Q. Okay. Okay. And it was simply, if you will, a 14 voluntary decision on your part not to -- not to have 15 those numbers included; is that right? 16 A. Yes. 17 Q. Okay. And -- okay. Were you -- and I take it you 18 were aware, obviously, because you got a specific 19 invitation, that the results of the survey would be 20 submitted as part of this Make Allowance hearing, correct? 21 A. Yes. 22 Q. And in also choosing not to participate in the 23 survey of 2019 costs, had you been aware that 24 Dr. Stephenson had actually been commissioned by USDA 25 itself to -- to make that study? 26 A. At the time I was not, no. 27 MR. ROSENBAUM: That's all I have. Thank you. 28 /// 4565 1 CROSS-EXAMINATION 2 BY MS. HANCOCK: 3 Q. Good morning, Mr. Cooper. Just a couple 4 questions. 5 You have talked about your plants that you have in 6 Texas and Michigan. I'm wondering if you have -- well, 7 and you also have some experiences at CDI, so with 8 California plants as well; is that fair? 9 A. That is fair. 10 Q. So given your experiences, how representative do 11 you believe that the California cost survey would be as 12 extrapolated and applied to the rest of the country? 13 A. There are going to be some nuances, as was earlier 14 discussed. Energy costs are higher in California. Labor 15 costs are higher. Benefits are higher. All those things 16 are quite a bit higher. We -- I have compared the 17 California study to our -- our current costs that we see 18 at our plants right now, and when you look at the 19 subsets -- those subsets, there is a difference. 20 Q. And are you willing to share with us which way 21 that difference goes? 22 A. Labor costs and energy are lower in the Midwest 23 and in Texas than in California. 24 Q. And are you comparing it to that 2016 survey? 25 A. Yes. 26 Q. Okay. So even comparing 2016 to today, you still 27 see some differences? 28 A. Some differences. They are very similar with 4566 1 today's costs, especially since COVID, because all of our 2 costs went up since COVID, especially in labor. To retain 3 our people, we had to pay more in labor. 4 Q. Okay. Have you ever dumped your butterfat solids 5 or your buttermilk powder? 6 A. Yes, we have. 7 Q. And how frequently does that occur? 8 A. As seldom as possible. We -- we know the value of 9 buttermilk. The only time that we dump buttermilk is when 10 we have a choice whether we're dumping milk on the farm 11 because there's more production than the plant can handle. 12 And since we're owned by a cooperative, we do not dump our 13 members' milk. So at that point we make the decision to 14 dump buttermilk to avoid dumping milk on the farm. 15 Q. Okay. And how frequently does that occur? 16 A. Not very frequently. 17 Q. Okay. Is it seasonal or due to some one-off 18 situation? 19 A. It's seasonal, and it really is a decision made 20 whether we can -- like I said, it goes back to how -- you 21 know, what kind of -- what kind of volume do we have 22 coming in from the farm, how much can we balance on a 23 daily basis within the plants. We tend to run the plants 24 at overcapacity as much as possible. So once we get to 25 the overcapacity limits and we see that we can't handle 26 the buttermilk, then we'll dump it. But we do that very 27 seldomly. 28 Q. And that's because of the cooperative, you owe 4567 1 your members a responsibility to take all the milk that 2 they produce; is that fair? 3 A. That is fair. That is correct. 4 Q. And have you been able to honor that commitment to 5 your members? 6 A. Yes, we have. 7 Q. When a butter plant purchases cream to make 8 butter, how does that butter plant pay for the buttermilk 9 solids? 10 A. Basically when we purchase cream, it is in the 11 cream -- it's -- it's calculated in the cream price. So 12 when we buy the cream either -- when we buy outside cream 13 from others, because we do handle cream from others, that 14 is based on the -- on the -- on the multiple that we pay 15 for the cream. 16 Q. And you talked about finding the opportunity or 17 the outlet for using the buttermilk powder for use in an 18 infant formula. I know as a very protective mom I'm 19 sensitive about what I put into my kids' bodies, and I 20 know that there's lots of specifications. I'm just 21 wondering if there are any additional or more strict 22 classifications for manufacturing infant formula than you 23 would have for other uses? 24 A. Yes. 25 Q. Can you share with us what those are? 26 A. It's a bacterial analysis is primarily what it is. 27 Q. So additional testing and monitoring? 28 A. Yes. 4568 1 Q. Do you have to build any other features into the 2 process to make sure that -- that you hit or achieve or 3 exceed those bacterial thresholds for testing? 4 A. No. Basically the -- if the -- if the plant is 5 doing everything it is supposed to be doing from a food 6 safety standpoint, you do not have to build that in. So 7 as long as you are adhering to a very strict food safety 8 program, you don't have to do that. 9 Q. Okay. And I think you said this already. I just 10 want to make sure I understood. On your -- on page 4 and 11 5 when you have tracked the indexes for your two plants 12 for those sales, where you have a -- an index indicator 13 there that says "null," that just means that you did not 14 sell any -- any in that month? 15 A. That is correct. 16 Q. Okay. Can you -- can you share with us what 17 volumes we're talking about here? I mean, is that a huge 18 drop-off when you get to a zero or -- or is it kind of 19 nominal sales up until that point so that it's not a huge 20 drop at that point? 21 A. Well, it's usually a pretty good drop-off at that 22 point, because our -- we have very large churns that when 23 we produce butter, we produce quite a bit of buttermilk 24 powder. There are some months when we will start ramping 25 down. If we're looking at the months there, like August 26 and September, usually, we'll start ramping down towards 27 June, July. But that really depends on the demand for 28 cream and how we start ramping down the butter plants. 4569 1 Q. Okay. So those are just intentional business 2 decision that you make as you ramp down in those summer 3 months? 4 A. Yes. 5 Q. And then what about in this example on page 4, 6 we're looking at for Michigan, you had -- June of '21 and 7 November of '21? 8 A. Those were both months when we -- we weren't 9 churning. 10 Q. Okay. 11 A. I mean, by the time we get to November, and 12 usually into November, you know, the butter season is over 13 because if you haven't fulfilled the pipeline with butter 14 by October, it doesn't make it to the store shelves by 15 Thanksgiving and Christmas. So in those months we'll 16 start ramping down as well. 17 Q. Okay. So just seasonal responsiveness? 18 A. That's exactly what it is. 19 Q. Okay. And then I think you noted that you -- you 20 try not to have an overlap between the months, in the same 21 months, between Michigan and Texas. Was that right? 22 A. That's just random. It depends on the -- on the 23 cream sales and the cream demand in those regions and 24 butter demand in those regions. 25 Q. So it would be coincidental if it did overlap, not 26 an intentional business decision? 27 A. That's correct. 28 Q. Okay. 4570 1 MS. HANCOCK: That's all I have. Thank you so 2 much for your time. 3 THE WITNESS: Okay. 4 CROSS-EXAMINATION 5 BY MR. ENGLISH: 6 Q. Good morning, Mr. Cooper. My name is Chip 7 English. I represent the Milk Innovation Group. 8 So I was trying to follow along and at the same 9 time was jotting down questions, so if I got lost, I 10 apologize, but I also don't want to repeat from your 11 questions from your counsel. 12 Is some fat lost from when farm level protein and 13 fat components are first tested through the end of the 14 cheese manufacturing system? 15 A. We don't make cheese. 16 Q. Do you know from your experience in making cheese 17 whether fat is lost from farm level protein and fat 18 components when they are first tested through the cheese 19 manufacturing system? 20 A. I mean, you are asking me something I haven't done 21 in over probably 40 years, so... 22 Q. Okay. 23 A. You know, technologies have changed quite a bit 24 from when I was making cheese, so I -- 25 Q. Well, I will move along. 26 A. -- I'll refrain from commenting on that. 27 Q. I apologize, I thought some witnesses seemed to 28 get you to talk about cheese, so I thought maybe you had 4571 1 more experience recent. But that's fine. 2 So for your butter plant, do you make Grade AA 3 butter of the type reported to the NDPSR? 4 A. Yes, we do. 5 Q. Do you use any whey cream in that butter? 6 A. No, we do not. 7 Q. Is that because whey cream isn't treated as 8 Grade AA? 9 A. That's correct. 10 Q. How much does whey cream sell for? 11 A. You know, we don't really participate in that 12 market, so I don't -- I have no idea. 13 Q. Okay. 14 MR. ENGLISH: I have no further questions then. 15 Thank you. 16 THE COURT: Further questions from anyone else 17 aside from AMS? 18 Yes, Mr. Rosenbaum. 19 CROSS-EXAMINATION 20 BY MR. ROSENBAUM: 21 Q. Steve Rosenbaum for the International Dairy Foods 22 Association. 23 You made a comparison between California labor 24 costs and labor costs where your plants are located, 25 correct? 26 A. Yes. 27 Q. Do you -- do you pay at the minimum wage? 28 A. No, we do not. 4572 1 Q. Do you pay substantially higher than the minimum 2 wage? 3 A. Yes, we do. 4 MR. ROSENBAUM: That's all I have. 5 THE COURT: Seeing no one else, AMS? 6 CROSS-EXAMINATION 7 BY MR. WILSON: 8 Q. Good morning, Mr. Cooper. 9 A. Good morning. 10 Q. Thank you for coming and testifying today. This 11 is Todd Wilson, USDA. 12 I have a few questions about -- about the -- your 13 testimony and wanted to kind of dig into some of the 14 information, expertise-wise, that you have. 15 So when we talk about the nonfat solids yield 16 factor in our formulas, what -- what is it that you feel 17 like should be included in that -- in that factor? 18 A. Specifically? 19 Q. Specifically should we be looking at all powders? 20 Should we be looking at nonfat dry milk powders? Should 21 we be looking at buttermilk powders? 22 A. Yes. 23 Q. All of the above? 24 A. (Shakes head.) 25 Q. Each one of those powders have different -- 26 (Court Reporter clarification.) 27 BY MR. WILSON: 28 Q. I'm sorry, can you verbally respond? 4573 1 A. Yes. I'm sorry. 2 Q. Each one of those products have unique yields 3 inherent to themselves? 4 A. Correct. 5 Q. Do you have any information as to what those 6 different yields should be? 7 A. We -- we follow those yields pretty closely, and, 8 again, it depends on butterfat. It depends on the 9 components. So your moisture in buttermilk powder is 10 lower than your moisture in nonfat dry milk, so you are 11 going to get a lower yield factor on buttermilk powder 12 than you would on nonfat. 13 Q. So there's been some historical yield factors on 14 buttermilk powder dating back to some earlier hearings 15 that we had in the early 2000s. 16 Is it -- do you feel like those yield factors are 17 still appropriate given today's environment and components 18 and things? 19 A. I think they have to be looked at based on today's 20 processing capabilities and the efficiency of today's 21 equipment. 22 Q. The next question is, when you have 100 pounds of 23 nonfat milk solids, you get 100 pounds of nonfat milk 24 solids into your plant, do you track or do you know what 25 should go into those products, how much is going into 26 nonfat dry milk, how much is going into butter, how much 27 is going into buttermilk powder? 28 A. We track it -- yes, we do track it. And -- and we 4574 1 track on how much should go into powder, how much should 2 go into either cream or butter. If it goes into cream, 3 obviously, that -- the solids and the butterfat follow the 4 cream. If it goes into butter, obviously how much gets 5 into the butter and how much gets into the buttermilk 6 powder, those yields are followed. We also calculate our 7 plant loss. And then we compare -- that is basically what 8 we put in our KPIs as to our plant performance. 9 Q. What would -- what is a KPI? 10 A. Key process indicators. 11 Q. Key process indicators. Okay. Thank you. 12 So do you have any -- I don't want to ask you for 13 proprietary information, but do you have any information 14 from an industry standard about if you had 100 pounds of 15 nonfat milk solids, how much should go into making nonfat 16 dry milk, how much should go into buttermilk powder? Is 17 there a standard? 18 A. There's standards that I have used over the years 19 based on my experience and based on my former employers as 20 well, and that's -- and we look at those standards, and 21 then we compare our -- we set our own standards as well. 22 So we have our own internal standards that we set. 23 Q. Do you have any figures that you might be able to 24 share with us from an industry standard, not a proprietary 25 standard? 26 A. They are pretty close, so I would probably refrain 27 from sharing those. 28 Q. So those are going to be out there in the public 4575 1 domain? 2 A. They should be out in the public domain. 3 Q. That's what I'm looking for. 4 So a couple of questions by some other parties 5 asked about how a butter/powder plant pays for their 6 nonfat milk solids. 7 Is it your understanding that all nonfat milk 8 solids coming into the plant are paid for at -- or valued 9 at the Class IV nonfat solids price? 10 A. Right. 11 Q. Do you make whole milk powder? 12 A. No. 13 Q. You do not. 14 When Mr. Miltner was asking you about some of the 15 prior testimony with -- with some plant accountants, there 16 was some discussion with him about you tracking your 17 solids losses versus your butterfat losses. Can you 18 elaborate a little bit more about those -- about those two 19 components and how you -- how they track, how you -- how 20 your plant handles those? 21 A. Certainly. Basically it's -- from us -- our 22 standpoint, we look at a mass balance. With milk coming 23 in the door, we look at our inventories, we look at what 24 we produce on a daily basis, and then we look at 25 components. And then at the end of the day when we 26 compare everything that's came in the door, what we have 27 produced and the component levels, you know, what we're 28 missing is our loss. 4576 1 And so we track those. There's standards that I 2 have always used in my career through former employers 3 and -- and what we have established internally as to 4 what's acceptable. Usually -- and I'll -- I won't give 5 you our standards, but I can tell you, on a standards 6 basis, if you look at butterfat losses as compared to 7 solids nonfat losses, solids nonfat losses usually run 8 about a percent higher than butterfat losses. 9 So we -- we set two guidelines, and the guidelines 10 for our managers are, here's your acceptable butterfat 11 loss, here's your acceptable solids nonfat loss. And if 12 you exceed those, obviously it's going to affect your 13 paycheck. 14 Q. Okay. On page 8 of your testimony, you -- in 15 your -- in that middle paragraph, you talked about costs 16 of making buttermilk powder versus costs of making nonfat 17 dry milk, and you talked about the incremental fuel 18 increase. 19 Could you talk a little bit more, elaborate more 20 about what that -- what causes that? I'm thinking it's in 21 the drying time, how long it takes more? 22 A. When we -- when we dry buttermilk, there's two 23 things that affect our efficiencies there. When we go -- 24 when we switch from nonfat to buttermilk, we have to have 25 a purge time in between to where we have no product, so 26 that we do not intermix the two products, because it's all 27 dried on the same equipment. So that's calculated in it. 28 But the other thing that's calculated in is 4577 1 because there's a higher fat percentage in the buttermilk 2 than there is in nonfat. We have to run at a slower speed 3 because it -- buttermilk has certain properties that are, 4 I would say a little stickier, so you have to run a higher 5 temperature and a lower throughput to the dryer to make 6 sure that you do not plug your bag houses or plug your 7 cyclones on a dryer that has cyclones. 8 So the components in the buttermilk dictate on how 9 you dry -- how you run the dryer. Similar to whole milk. 10 I have dried whole milk powder in the past. You run -- 11 when you run whole milk versus nonfat, you lose 25% 12 efficiency in your dryer because you have higher solids, 13 higher fat, and it runs completely different. 14 So how that equipment runs between the two 15 different products and the constituents in the product 16 dictate on how you run. And when you run hotter, 17 obviously, you have to run -- you to have burn more gas. 18 Q. So the 25% that you mentioned of decreased 19 efficiency, was that whole milk powder? 20 A. Yes. 21 Q. What's that in relation with the buttermilk 22 powder? What kind of inefficiencies do you experience? 23 A. It's not quite that much. It's -- like I said, we 24 looked at that $0.02 cost price in there that was USDA 25 established, and we agreed that was about the difference 26 between the two. 27 Q. All right. Thank you. 28 MR. WILSON: That's all from AMS. Thank you, 4578 1 Mr. Cooper. 2 REDIRECT EXAMINATION 3 BY MR. MILTNER: 4 Q. Mr. Cooper, just a few questions to tie up some 5 things. 6 I understand your protection of confidential 7 information. I do want to see if we can help with some 8 yield data on buttermilk. 9 In its prior decision USDA stated that there are 10 0.0479 pounds of -- let me read this correctly -- let me 11 just read it instead of trying to summarize it. 12 "According to the Economic Research Services 13 publication, nonfat milk solids in dry buttermilk are 14 0.0479 pounds per pound of nonfat milk solids." 15 Does that sound reasonable to you? 16 A. Yes. 17 Q. And for every pound of dry buttermilk, there are 18 0.919 pounds of nonfat milk solids. 19 Does that sound reasonable to you? 20 A. Yes. 21 Q. So that is from a 2002 USDA Final Decision. 22 In the intervening 20 years or so, those numbers 23 still sound reasonable to you? 24 A. They sound reasonable, but I think that it 25 would -- it would bear being -- it would -- it would 26 dictate we probably ought to look at it based on newer 27 production facilities that are a lot more efficient. And 28 we're capturing more -- more solids and more butterfat on 4579 1 our equipment now than we ever have due to changes in 2 technology. 3 Q. Okay. But it certainly wouldn't be less than what 4 USDA said 20 years ago? 5 A. No. 6 Q. Now, if we just think about nonfat dry milk. If 7 you take a pound of nonfat solids, and you dried it, would 8 the yield be higher than 0.99? 9 A. Yes. 10 Q. Would it be reasonable for a plant of average 11 efficiency to yield something more like 1.02? 12 A. Yes. 13 Q. And that's just on the nonfat dry milk part of it, 14 correct? 15 A. That is correct. 16 Q. And that assumes that you skimmed off the cream, 17 and all those solids that would end up in buttermilk 18 powder are still there to be dried or condensed or 19 utilized in some fashion? 20 A. That's correct. 21 Q. And when you -- okay, so -- and then last question 22 would be, you -- you testified to it, and I -- you 23 clarified it I think with AMS, but just to make sure it's 24 completely clear. 25 That relationship of two additional cents to dry 26 buttermilk powder, which was pulled from that same 2002 27 decision, that $0.02 difference, even in today's price 28 environment to you, is still a reasonable difference 4580 1 between the costs of drying those two products? 2 A. That's correct. 3 Q. Thank you. 4 MR. MILTNER: I don't have any other questions, 5 your Honor. And I would move the admission of 6 Exhibit 223. 7 THE COURT: Any objections? 8 Seeing none, Exhibit 223 is entered into the 9 record. 10 (Thereafter, Exhibit Number 223 was received 11 into evidence.) 12 THE COURT: Thank you, Mr. Cooper. 13 MS. TAYLOR: We have been going for about an hour 14 and 15. Let's come back at 11:00. 15 (Whereupon, a break was taken.) 16 THE COURT: Let's come to order. I'll go ahead 17 and swear in the witness, I guess. 18 Please raise your right hand. 19 NANA FARKYE, 20 Being first duly sworn, was examined and 21 testified as follows: 22 THE COURT: Your witness. 23 DIRECT EXAMINATION 24 BY MR. MILTNER: 25 Q. Good morning, Dr. Farkye. How are you? 26 A. I'm well. Thank you. 27 Q. You may have to get a little closer to the mic, 28 mostly so that our folks listening online can hear you. 4581 1 Before we get going, could I ask you to give your 2 name, spell it, and then give your business address, 3 please? 4 A. My name is Nana Farkye or Farkye. People say it 5 in different ways, but I prefer Farkye. Nana, N-A-N-A, 6 Farkye, F as in Frank, A-R-K-Y-E. Business address, 7 2929 Floyd Avenue, Number 330, Modesto, California, 95535. 8 Q. And, Dr. Farkye, do you have two documents with 9 you, one of them at the top reads Exhibit Select-7 and the 10 other in the upper right reads Select-8? Do you have 11 those with you? 12 A. That is correct. 13 Q. And have you seen these before? 14 A. Yes, I have. 15 Q. Select Exhibit 7, is this the report and testimony 16 that you prepared to present today at this hearing? 17 A. That is correct. 18 Q. And I understand that Exhibit Select-8 is some 19 information that you relied upon in preparing your report 20 that wasn't easily available to the world, so you -- we 21 wanted to provide that as an exhibit for the hearing; is 22 that correct? 23 A. That is correct. 24 Q. And I think for perhaps part of your testimony you 25 are going to put some information on the screen from 26 your -- from your testimony; is that correct? 27 A. That is correct. 28 Q. Great. If you could, Doctor, provide us -- go 4582 1 through the introduction materials on your background, and 2 then pause, and I can ask you a couple questions, and then 3 we'll go from there. Okay? 4 A. Sure. 5 THE COURT: Let's mark the exhibits. 6 Select-7, top right-hand corner, will be marked as 7 Exhibit 224. 8 (Thereafter, Exhibit Number 224 was marked 9 for identification.) 10 THE COURT: Exhibit Select-8 will be marked 225 11 for identification. 12 (Thereafter, Exhibit Number 225 was marked 13 for identification.) 14 THE COURT: Thank you. Your witness. 15 MR. MILTNER: And before you start, just for the 16 record, I'm not good at putting things on the record, but 17 I want to put on the record, I wrote myself a note to get 18 these marked before we started. And I'm really not good 19 at that. But Dr. Farkye -- 20 THE COURT: No problems at all. 21 THE WITNESS: So I have been retained by Select 22 Milk Producers to provide a written report expressing my 23 independent opinion on milkfat recovery or milkfat 24 retention during cheese making. I have a Bachelor of 25 science degree in biochemistry and nutrition from the 26 University of Ghana. Came out to do my Master's and Ph.D. 27 at Utah State University in nutrition and food sciences. 28 Then I worked as a post doctoral scientist at the 4583 1 University College, Cork, before joining California 2 Polytechnic State University where I worked as an 3 assistant professor, associate professor, full professor, 4 over a 25-year period, rising through the ranks. I taught 5 dairy chemistry, dairy processing, cheese and fermented 6 dairy foods, at both undergraduate and graduate levels. 7 My research primarily focused on emphasis on 8 cheese technology and cheese ripening. I have also worked 9 an enzymes in cheese and enzymes in milk -- milk. 10 I served as a judge at the world and U.S. 11 championship cheese contest and the ACS cheese contest. 12 I'm an author or co-author of several 13 peer-reviewed articles, industry presentations, book 14 chapters, encyclopedia articles in areas of dairy 15 chemistry, microbiology, and cheese technology. 16 I took an early retirement in 1995 to serve as a 17 consultant for the dairy industry, which I have done 18 since. My role as a consultant, I advise, troubleshoot, 19 and find solutions to problems and develop new products 20 and streamline processes for the dairy industry. 21 My CV attached in Appendix 1. 22 BY MR. MILTNER: 23 Q. Thank you, Dr. Farkye. 24 Now, you noted that your CV is attached. It's 25 pages 7 and 8 of your testimony. 26 I notice at the end that you said there were over 27 a hundred different research publications that you were 28 involved with; is that correct? 4584 1 A. That is correct. 2 Q. And you provided a smaller subset of those that 3 more directly relate to butterfat retention or butterfat 4 recovery; is that right? 5 A. That is correct. 6 MR. MILTNER: Your Honor, we would ask that 7 Dr. Farkye be recognized as an expert in the areas of 8 cheese processing, cheese manufacturing, and food science. 9 THE COURT: Seeing no objections, I so find that 10 this witness is qualified to testify as an expert on those 11 topics. 12 BY MR. MILTNER: 13 Q. So, Dr. Farkye, I think perhaps the best way to 14 proceed with your testimony is to have you present it as 15 you would to folks you wanted to educate, and if that 16 involves putting some information up on the screen so 17 folks can follow along, that would be great. And then we 18 can ask some additional questions either throughout or at 19 the conclusion. 20 Does that sound all right with you? 21 A. Sure. 22 Q. Excellent. I'll let you take the keys from here. 23 A. So if it's okay with you, I would just project my 24 talk on the screen, and then I would just point along as 25 we read it, or at least talk about it. 26 So the first thing is to understand cheese making. 27 Cheese making involves a number of steps, which I've 28 outlined here. First you are going to bring the milk into 4585 1 your plant, and you are going to standardize the milk to a 2 protein of casein-to-fat ratio. 3 Then you are going to pasteurize the milk. Of 4 course, some people choose to use raw milk, but we're 5 going to focus on pasteurized milk here. 6 Then you transfer the milk into a cheese vat, and 7 then you add your ingredients which are color for -- if 8 you are making yellow cheddar, you are going to add color. 9 Then you are going to add your starter culture, which is 10 harmless lactic acid bacteria that you are adding to the 11 milk. Then you are going to add calcium chloride -- or 12 may or may not add it depending on the type of milk you 13 are using. Then you are going to add the coagulants that 14 we call rennet or chymosin. 15 Then you are going to allow the milk to settle for 16 about 30 minutes, and that's when we call coagulation or 17 curd formation. And then after the curd is firm enough, 18 you are going to cut the curd into cubes. 19 Okay. And then after cutting that curd into 20 cubes, you are going to allow the curd to heal for about 21 five minutes. Then you are going to begin to stir and 22 cook the curd in the whey. So now you have curd and whey, 23 and you are going to cook. 24 Then after you cook to the desired temperature, 25 here we are talking about cheddar cheese, so many times we 26 are going to cook to about 100 to 102 degrees Farenheit. 27 And then when -- this allows the cultures to grow and give 28 you the right acidity that you desire. 4586 1 Then after that, you are going to drain the whey. 2 So at this point that we are going to think of fat 3 recovery or fat retention. So when you cut the curd and 4 you drain the whey, some of the milk components are 5 retained in the curd and others are lost in the whey. And 6 so how you treat the curd during cheese making can affect 7 how much retention of butterfat or casein that you have. 8 Then after that, after you drain the whey, you are 9 going to matt and cheddar the curd, and you are going to 10 mill the curd or cut it into small finger size chips. And 11 then you are going to salt it, hoop or mold it, and 12 package it. 13 And it can be done either manually or using 14 towers. Current cheese plants, large cheese plants will 15 use block formers or cheese towers. 16 Okay. So that's the basic steps of cheese making. 17 Now, to make that cheese efficiently, you want to 18 recover as many -- or as much of the milk components as 19 possible. Because really the whole intent is to preserve 20 milk, and you don't want to lose the components -- every 21 pound of milk solids that you retain in the cheese, the 22 more profitable you are. Because if it's lost in the 23 whey, and you don't have at market for the whey, that's 24 it. Okay. Even though there are ways to recover that. 25 Okay. So the amount of fats that is lost in the 26 whey is very small, but it's -- it can all be very 27 significant. So when fat is lost in the whey, it's called 28 whey fat or you get whey cream. Now, some companies may 4587 1 choose to add the whey cream back, but it is not a 2 standard practice in industry, because when you add whey 3 cream, you ought to think about bacteriophage and 4 contamination. So you -- you have to treat as well. So 5 it is not a standard practice, but that's not to say that 6 it's not done. 7 Okay. Now, to make cheese we are going to start 8 with milk. So what I have here is a table of milk 9 components. And we have different breeds of cattle in the 10 U.S. or worldwide. The left column shows the components 11 of fat, protein, lactose, ash, total solids, water, 12 protein-to-fat ratio. 13 Then the second column shows the composition of 14 Holstein milk. Holstein is the largest breed of cattle we 15 have in the United States. 16 Then in the fourth column -- the fifth column is 17 Jersey cattle. That's the second largest breed. So if 18 you can -- we can focus on those two there. Holstein milk 19 has about 3.5% fat; Jersey milk has about 5.05% fat. 20 Protein in Jersey -- Holstein milk is 3.2% on the 21 average; protein in Jersey milk is about 3.79 or 3.8%. 22 So if you look at those, the protein-to-fat 23 ratios, Holstein milk has a protein-to-fat ratio on 24 average about 0.91 and Jersey milk is about 0.75. 25 Now, USDA calculates the average protein-to-fat 26 ratio in standard milk, which contains 2.99% protein or 27 3.5% fat, with a ratio of .85. So it is closer to 28 Holstein milk than Jersey milk. But if you blend the two, 4588 1 you are going to get something very, very close. 2 Actually, in my Master's thesis which was done in 3 1984, on the effect of casein-to-fat ratio on fat 4 retention in cheese making, I came up with a chart here, 5 which holds true up till today. 6 So hopefully I can -- so if we look at the chart 7 that shows clusters of three triangles. Essentially what 8 we did was we took Holstein milk, which is the cluster of 9 three triangles in the bottom right corner. The 10 protein-to-fat ratio or the casein-to-fat ratio was 11 average of .71. 12 Then we took the cluster in the top left corner is 13 All-Jersey milk. So those had -- because of the high fat 14 and high protein, the casein-to-fat ratio is about .64 15 average. 16 And the cluster in the middle is a blend of both 17 milks. So we blended Jersey milk and Holstein milk to get 18 what is typical in the industry. And that's the cluster 19 in the middle. 20 And we made cheddar cheese to see if we are going 21 to get differences in fat recovery. 22 Now, what we found is there's no difference in fat 23 recovery, so the higher protein-to-fat ratio, the more fat 24 you get in the cheese, but as a percentage of recovery, it 25 was about the same. 26 And why do we standardize milk for fat recovery? 27 We do standardize milk to be able to meet the standards of 28 identity for cheese. For instance, cheddar cheese must 4589 1 have a minimum of 50% fat in it, dry matter, and no more 2 than 39% moisture. So if you do not standardize the milk 3 to the right casein-to-fat ratio, you are going to lose 4 the milk components, and you are not going to recover. 5 And then you are also not going to meet the standards of 6 identity. And that's the primary reason why we 7 standardize milk for cheese making. 8 Okay. Now, every cheese maker is making cheese to 9 maximize cheese yields. 10 How do we calculate cheese yields? Well, there is 11 an old formula called the Van Slyke and Price cheese yield 12 formula. This formula was developed in 1894 from work 13 that was done in New York. And this holds true up till 14 today. 15 So this formula says that for cheese yield, 16 cheddar cheese yield, you are going to retain about 93% of 17 the fat. You are going to lose a tenth of a pound of 18 casein, which is when you clot milk for cheese making, 19 casein is hydrolyzed at a specific bond, and some of that 20 casein is lost in the whey, and most of it is retained as 21 part of the curd. So the formula accounts for the loss of 22 casein in the whey. 23 Now, the formula also says that when you add salt 24 to cheese during cheese making, you retain some of the 25 salt and other solids in the milk, which is nonfat and 26 protein, like the lactose, the minerals. And that is 27 about 9% of the sum of the casein and the fat, so that's 28 where the 1.09 comes in. And then you divide by the 4590 1 solids in the milk, so one minus the moisture composition. 2 Say you had a cheese of 39% moisture, it would be one 3 minus .39. If you have 37% moisture, it would be one 4 minus .37. Okay. 5 So this has what worked so well, but what happened 6 is Van Slyke and Price, their cheese making was done in 7 open vats. And I will show you pictures of that in a 8 minute. 9 And in the '80s cheese plants started going to -- 10 to automate, and they were going for what we call enclosed 11 cheese vats for -- to improve safety, to improve yields, 12 also. Then the first set of enclosed cheese vats that 13 were introduced were called the Double O vats. Okay? And 14 I will show a picture of that in a minute. 15 And when these vats were introduced, we couldn't 16 attain the 93% fat recovery. So there was petition to 17 reduce the Van Slyke yield formula and reduce the fat 18 recovery to 90%. And that's how the whole thing started. 19 So we had the modified Van Slyke formula where it 20 was .90 times F plus C minus .1, or some people -- instead 21 of C minus .1, they'll do .78 times P because casein, 22 which is C, was not easily measurable. So they'll measure 23 protein, and since 78 to 80% of the protein in milk is 24 casein, they'll use .78P instead of C. 25 Okay. So if I may, I can just pause here and show 26 pictures of vats. 27 So -- so once Van Slyke did his work, or Van Slyke 28 and Price did their work, we have open vats, which is 4591 1 shown here. 2 Then we went to the enclosed cheese vats. So on 3 the middle left there we have -- it looks like two Os if 4 you look at that, right? Two Os, it is called Double O 5 vats. That's what we went to, Double O vats. 6 And in the box there as -- it says that Double O 7 side vat -- from the side view, style vats on the side 8 view. 9 Then we went to horizontal cheese vats. So 10 instead of having the vertical agitation, we have 11 horizontal agitation. 12 And then the new cheese vats are -- instead of 13 having a single shaft, we have double shaft. So that's 14 been an improvement in the technology for cheese making 15 equipment, and this affects cheese making -- efficiencies 16 of cheese making. 17 Okay. So I'll go back, toggle back to... 18 So Van Slyke and Price did an excellent job to get 19 the 95 -- 93% fat recovery. We couldn't achieve it when 20 we went to the Double O vats. It was more difficult. 21 Yes, you can attain it, but you have to work hard to 22 attain that. Now that we have newer vats, we are able to 23 even go beyond that. 24 And in my research to write this paper, I sent 25 e-mails out to equipment manufacturers, and I have notes 26 up here from Tetra Pak, this -- they are saying that, 27 using their cheese vats, which is what most of the large 28 cheese plants use today, you can attain as high as 97% fat 4592 1 recovery. This is using the twin screw cheese vats, so 2 it's horizontal ACV. I have seen that in cheese plants 3 myself, and I know that it's -- it's doable. Okay? 4 Now, if you are using the old Double O vats, yes, 5 you can achieve it, but it's a little bit more difficult 6 to do. 7 Okay. Now, cheese making and cheese yield is not 8 only about fat recovery. It's about protein also. 9 Because during milk clotting, you're curdling the milk, 10 and as you curdle the milk, it's forming a network and 11 it's trapping the fat in it. So if you have a weak curd, 12 you are going to -- that curd is going to shatter very 13 easily, and so when you -- when you stir it, you are going 14 to lose more fat. If you have a firm curd, because of the 15 strong protein network, you are able to hold more fat. 16 So it's not just about equipment, but it's about, 17 say, the ingredients that you are using in cheese making, 18 which the coagulant plays a big role. 19 So Christian Hansen, that's the leading coagulant 20 manufacturer in the world, actually has introduced enzymes 21 that allow you to increase yield based on protein content. 22 So essentially as you are trapping more protein, you are 23 also trapping more fat. 24 And this picture here shows how different 25 coagulants can actually result in higher cheese yield. 26 And there's more information in the exhibit that is 27 already included, Exhibit 8. 28 So as you use more efficient milk coagulants, you 4593 1 are going to increase your cheese yield by creating a 2 better network, and by so doing you are also trapping more 3 fat. Okay. 4 So in my experience as a consultant, I'm aware 5 that to -- to optimize cheese yield, we need to 6 standardize milk by blending -- by introducing protein or 7 fat, from either a source where you are using UF milk, RO 8 milk, skim milk, nonfat dry milk, or cream. You need to 9 standardize to the right protein-to-fat ratio or 10 casein-to-fat ratio. And you have to control the 11 coagulants used, the curd firmness, the stirring, that's 12 where the equipment comes in. And for large commodity 13 cheese plants, you can achieve 93% or higher fat recovery, 14 if you do a good job. 15 Okay. So in conclusion, consistent milk 16 composition, which we know now through breeding programs 17 and good milk -- milk composition, and improvements in 18 equipment, and cheese making ingredients, particularly 19 coagulants, have all contributed to increased cheese 20 making efficiency and higher milk solids recovery, 21 including milk fat in cheese. Okay. 22 Therefore, in my opinion, fat recoveries of 93% in 23 the original Van Slyke and Price cheese yield equation, 24 93% is achievable. And direct observations and my direct 25 view of cheddar cheese plants are achieving these fat 26 recoveries, and even above these levels. So I would say 27 that it was done in 1894, it can be done now -- or 28 actually it is being done now. 4594 1 And on that note I will conclude. 2 Q. Thank you, Dr. Farkye. I want to ask you some 3 more questions about what you just presented. 4 So you talked about when Van Slyke and Price 5 developed their formula. And when you and I were talking 6 last night, you told me you have a copy of this book that 7 I found last week. 8 A. That is correct. 9 Q. From whatever, 1916, by Mr. Van Slyke -- or I 10 don't know if he was Dr. Van Slyke or not. But Van Slyke, 11 he was a Ph.D. 12 A. Yes. 13 Q. He published this book in 1916, and he talked 14 about the formula you have discussed. 15 Now, you have talked about the impact of vats and 16 how that affects the butterfat recovery, so I want to go 17 through that just a little bit more with you. 18 A. Okay. 19 Q. His research, his observations were done with the 20 open vats that you showed, correct? 21 A. That is correct. 22 Q. Now, the next kind of evolution in vats, those are 23 the Double O vats you talked about? 24 A. That is correct. 25 Q. And do you recall when those types of Double O 26 vats started being manufactured? 27 A. I will say in the '80s. Early '80s. 28 Q. And so you described those vats as a vertical vat, 4595 1 correct? 2 A. Vertical agitation. 3 Q. Okay. So when you talk about vertical agitation, 4 that's -- that means that the shaft in the vat is 5 vertically oriented? 6 A. That's correct. 7 Q. And that's what stirs the curd, correct? 8 A. That is correct. 9 Q. And in an open vat, would that be done more 10 manually, originally? 11 A. So it will cut manually, but it's also vertical 12 agitation. 13 Q. Okay. In the Double O vat? 14 A. In the open vat. So, for instance, if you look at 15 this here -- 16 Q. Okay. 17 A. -- if I may. 18 Q. Yeah. Oh, you have to ask him -- 19 A. That's fine. He doesn't have it on. I'm sorry. 20 Q. That's okay. There we go. 21 A. So if you look there, there is an agitator right 22 up there in the top right corner. And that's traveling 23 along the cheese. So as it's traveling, it is stirring. 24 Q. Okay. 25 A. So that's a traveling agitator, whereas in the 26 enclosed cheese vats, it is stationary, so it is more 27 rotating. 28 Q. And so am I correct that the primary benefit of 4596 1 the vertical Double O vat was to automate the cheese 2 making process a little more? 3 A. That is correct. 4 Q. But what was found in that automation, the 5 agitation resulted in a lower butterfat retention than 6 with an open vat; is that correct? 7 A. That is correct. 8 Q. Now, after vertical vats, after Double O vats came 9 the horizontal vats that you described, correct? 10 A. That is correct. 11 Q. How are those different than the Double O vats? 12 A. So if you can see the horizontal vats, they have 13 the shafts laid horizontally, and they are rotating, and 14 they are -- it is more gentle. That's the first 15 generation. And the second generation you have twin 16 screws, so they are kind of stirring in a counterclockwise 17 manner. 18 Q. So it's the same agitation function, but the shaft 19 is oriented differently? 20 A. That's correct. 21 Q. And you describe it as a gentler action? 22 A. That is correct. 23 Q. Now, with the adoption of those horizontal vats, 24 have cheese makers realized higher butterfat retentions? 25 A. That is correct. Yes. 26 Q. If I go back to your initial description of the 27 cheese making process, this is the cutting of the curd, 28 the stirring, and the cooking, that's the process that the 4597 1 vats have automated, correct? 2 A. Yes. 3 Q. Now, if we go back to when Dr. Van Slyke was 4 observing cheese manufacturing in the 1890s, all of those 5 processes would have been manual, would they not? 6 A. The cutting was manual, just like I did in my 7 thesis work. 8 Q. And I think when you were describing those 9 processes, you -- you indicated that you have to be 10 careful when you cut the curd because you said -- you 11 described it as it has to heal, correct? 12 A. That is correct. 13 Q. And that was to keep the protein and fat matrix 14 intact, correct? 15 A. That is correct. So you don't lose the fat. 16 Q. So if I can summarize, we went from a mostly 17 manual process where people could be very careful and 18 develop an expertise in how to stir and cut cheese, to a 19 more automated process where we lost some of that human 20 touch and precision maybe? 21 A. That is true. 22 Q. To a new iteration of vats, which improve upon, I 23 guess, the efficiencies lost through that initial 24 generation of automation? 25 A. That is correct. 26 Q. So now I want to ask you about cheese plants in 27 general. I don't want to talk about really specific 28 plants, but -- 4598 1 A. Okay. 2 Q. -- were there some significant large commodity 3 plants built after the introduction of the Double O vats? 4 A. Yes. 5 Q. Now, those plants, if they were operating at a 6 reasonable level of skill and efficiency, with those 7 Double O vats, what do you think a plant like that could 8 expect in terms of butterfat retention? 9 A. If they do a very good job, they will get the 93%, 10 but most times they will not. That's why many of the 11 plants and the newer plants went away from the Double O 12 vats. 13 Q. Now, to get 93%, would that be using what I will 14 call a standard rennet? 15 A. Yes. You can get a standard rennet, yes. 16 Q. Okay. If those plants using a Double O vat were 17 to use a more advanced modern rennet, like the one you 18 have described, would they achieve greater than 93% 19 butterfat retention? 20 A. They will achieve it based on the fabric of better 21 curd formation. 22 Q. Now, my understanding -- and I was -- like 23 Mr. Munch, I was not doing much during this time period in 24 the industry. In the '90s, let's say, was there a lot of 25 growth in large commodity cheese manufacturing in the 26 United States during that period? 27 A. Yes. 28 Q. When would you have started to see the 4599 1 introduction of the horizontal vats that you have talked 2 about? 3 A. In the '90s. 4 Q. So does the introduction of the horizontal vats 5 kind of coincide with the more expansion within the 6 commodity cheese industry? 7 A. That is correct. That's a fair assumption. 8 Q. Now, with that first generation, single shaft, 9 horizontal vats, again, a plant of, let's say -- not a top 10 operator but an average cheese manufacturer, using an 11 average rennet, what type of butterfat retention should 12 that plant be achieving? 13 A. They should strive to get at least 93% fat. 14 Q. They are striving to achieve that, but let's say 15 that they are -- say that they are -- they are still 16 striving. What might they be realizing on a -- 17 A. Somewhere between 87 and 91, maybe. 18 Q. Okay. 19 A. That's what they will be getting. 20 Q. And their equipment allows them to achieve higher, 21 though? 22 A. That's correct. 23 Q. And if they are not achieving that, is that a 24 deficiency in their perhaps training and processes? 25 A. It could be both. 26 Q. But it would be something in the control of the 27 manufacturer that they could correct? 28 A. Yes. If they are careful, they would be able to 4600 1 control it, yes. 2 Q. So now take that plant that's striving for 93, and 3 let's say they have got the 93. And now they want to use 4 the best rennets or best coagulants they could find. 5 What do you think that would do to their butterfat 6 retention? 7 A. It will improve because of better curd formation. 8 Q. Now -- and I know you -- you listened to part of 9 the hearing, and you have been here for a bit but -- maybe 10 you have heard the testimony on some of these issues and 11 maybe not. But in the last, say, five years, maybe ten 12 years, has the industry seen quite a number of large 13 commodity cheddar cheese plants constructed and 14 commissioned? 15 A. Yes. 16 Q. If you were building a plant in that period, what 17 type of vat would you recommend they install? 18 A. The -- what we call the ACV double, ACV horizontal 19 cheese vats with the twin screw. And also install the 20 drain mat cheddar system to finish the cheese. 21 Q. Okay. So if you are installing the double shaft 22 horizontal vat system, that's the one that you described 23 as being, I think, more gentle with its action? 24 A. That's correct. 25 Q. And not top-of-the-line achievement, but 26 reasonable operation of a vat like that, what would you be 27 advising that they could expect to get in terms of 28 butterfat retention from those vats? 4601 1 A. 93-plus. 2 Q. And if they were -- if they were firing on all 3 cylinders, so to speak, and everything was going right, 4 what might they expect to achieve? 5 A. As the Tetra Pak engineer said, they can go as 6 high as 96, 97. 7 Q. And would that go even higher if they were using 8 the best coagulants available? 9 A. Yeah. After you get to 97, it's -- I mean, 10 some -- it's like having a cut, you are going to lose some 11 blood anyway, right? So there's -- you are trying to trap 12 the fat, but you are -- since you are cutting that fat, 13 you are going to lose some of that fat. 14 Q. I know your goal here, your role today is not to 15 tell USDA what proposal they should or shouldn't adopt. 16 A. No. 17 Q. But do you have an opinion about whether the 18 majority of cheddar cheese produced in the U.S. today, 19 commodity cheddar cheese produced in the U.S. today, comes 20 from plants that can achieve a butterfat retention of 93%? 21 A. Yes. Very much so. If you look at the amount of 22 milk that is produced in the United States and the number 23 of cheese plants that make cheddar, there are probably the 24 top maybe ten would use -- would have modern cheese making 25 equipment and achieve higher. 26 Q. Achieve higher? 27 A. Yes. 28 Q. And is your conclusion based both on the 4602 1 theoretical science as well as your observations of the 2 industry? 3 A. Yes. 4 Q. Thank you, Dr. Farkye. 5 MR. MILTNER: We would make the witness available 6 for additional examination. 7 THE COURT: Any questions for this witness aside 8 from AMS? 9 CROSS-EXAMINATION 10 BY MR. ENGLISH: 11 Q. Checking to see it is still morning. Good 12 morning, Dr. Farkye. My name is Chip English. I 13 represent the Milk Innovation Group. I think you got here 14 sometime yesterday, although I think you may have taken a 15 red eye, so I don't know how much you were able to follow 16 yesterday. But I do appreciate very much, I think 17 everyone appreciates your appearance. 18 A. Thank you. 19 Q. And the fact that you are not taking a position, 20 but you have articulated the science. 21 I actually want to start maybe at the end of what 22 you just said in response to additional questions from 23 Mr. Miltner. 24 And as I understood it, the -- he got you to say 25 it was the top ten cheese companies or cheese plants could 26 achieve this 93% butterfat, correct? 27 A. If they do a good job, yes. 28 Q. If they do a good job? 4603 1 A. Uh-huh. 2 Q. So even within the top ten, if they didn't do a 3 good job, they might not hit 93%, correct? 4 A. I would say with modern equipment, most of them 5 would right now. 6 Q. Okay. So that modern equipment comes with a cost, 7 though, correct? I mean, Tetra Pak is in the business of 8 making money, correct? 9 A. That is correct. 10 Q. So you have talked about sort of the majority of 11 cheddar cheese. Would that mean volume as opposed to 12 plants? When you talk about the majority of cheddar 13 cheese could be produced at 93%, would that be a volume of 14 cheese as opposed to a majority of plants? 15 A. So it's both. Because if you look at California, 16 for instance, ninety- -- about -- the last data I saw from 17 CDFA, about five plants use about 95% of the milk that is 18 produced in cheese making. And I mean, it may not be 19 cheddar alone. It may be cheddar, mozzarella. And if you 20 look at the U.S. as a whole, I believe there are about 150 21 cheddar cheese plants, maybe 50 in Wisconsin and 25 in 22 Idaho and -- I mean, making large commodity cheddars. 23 So your -- to answer your question, it's both 24 volume and number. 25 Q. But there are a significant number of smaller 26 plants in other parts of the country, such as Wisconsin, 27 correct? 28 A. Yes. But the amount of milk they utilize may be 4604 1 proportionally -- proportionately lower. 2 Q. Nonetheless, those plants do provide an outlet for 3 producers' milk, correct? If those plants didn't exist -- 4 A. Yeah, they do. 5 Q. Okay. And the e-mail that you received from 6 Mr. Steffens from Tetra Pak, is that an e-mail you got? 7 A. Yes, that's an e-mail I got. 8 Q. Is the text you provided, the entire text of his 9 e-mail? 10 A. Yes. 11 Q. Okay. So nowhere in that text does he say, for 12 instance, how much the Tetra Pak YieldMaster costs, does 13 he? 14 A. No. 15 Q. Nowhere in his e-mail does he say how much the 16 Tetra Pak HCV costs, correct? 17 A. No. 18 Q. And nowhere does he say how much the current Tetra 19 Pak Double O costs, correct? 20 A. No, sir. 21 Q. And when you talked about the results, I am 22 wondering, because the e-mail, the last part of it, which 23 carries over to page 5, says, we can increase the Van 24 Slyke rFat factor in a relational way; moving from rF 25 equals 0.90 hyphen 0.91 up to 0.91 to 0.925. 26 Is that what he says? 27 A. Yeah. So what he is saying, fat retention or fat 28 recovery, that's what he's saying. 4605 1 Q. Right. But nowhere there does he say 93%, does 2 he? 3 A. No, he doesn't say that. 4 Q. Okay. And then the next sentence says, "Much of 5 our data has been based on a very large plant where three 6 different style vats are located," correct? 7 A. That's what he says, yes. 8 Q. So his e-mail is basically limited to one very 9 large plant, correct? 10 A. I would -- I don't know what he was referring to 11 because I didn't ask him. 12 Q. And he's not here to tell us, is he? 13 A. No. But I do know from my experience that most of 14 the large plants would use the Schpering vats or Tetra 15 Scherping vats. 16 (Court Reporter clarification.) 17 BY MR. ENGLISH: 18 Q. And similar to the e-mail that you obtained from 19 Mr. Steffens, the Chr. Hansen document, which you called a 20 study, is it maybe more correct to say it's a marketing 21 piece by Chr. Hansen? 22 A. Well, I didn't -- it's -- so some of -- Chris 23 Hansen is a very reputable company, and they would -- they 24 would do the work before they print out something like 25 this. And we -- and I do know from experience that using 26 some of these cheese rennets would give you high yield 27 because I have done it. 28 Q. Does the Chr. Hansen document discuss what the 4606 1 costs are for this marketing piece of 20 pages? 2 A. No. 3 Q. I may have misheard, but did you hear some 4 testimony from -- did you follow last week and follow some 5 of the testimony last week or just when you got here? 6 A. No, I did not follow. 7 Q. Okay. If there was a witness who said that, yeah, 8 that's what the salespeople would say, but, you know, they 9 don't necessarily believe it about the coagulants, would 10 you disagree with that witness? 11 A. I don't know what the witness said, so I cannot. 12 Q. If a witness appeared last week and testified that 13 they were still using Double O, you would have no reason 14 to disbelieve them, correct? 15 A. There are some plants using the Double Os. It's 16 been sold. Again, it's more volume, throughput, 17 efficiency. And there are some that have gotten away from 18 Double O -- the newer Double Os to go to the ACVs because 19 if you do the math of the fat recovery, it will pay for 20 itself. 21 Q. But you don't have that math here because you 22 don't have the costs, do you? 23 A. No, I don't have the math, but I have spoken to 24 cheese plants overseas who have actually expressed that 25 sentiment. 26 Q. But it could very well be that a plant, say, in 27 Wisconsin hasn't been able to make that conversion 28 because -- for reasons that go beyond your testimony, the 4607 1 Federal Order pricing system may limit their investment 2 ability? 3 A. I have no knowledge of that. 4 Q. Now, you were here, yesterday, though, when I had 5 my conversation with Mr. Allen in the afternoon, correct? 6 A. I may have listened about it, yes. 7 Q. Okay. If he agreed with me that inevitably there 8 is some fat loss, through, say, wastewater, you would have 9 no reason to disbelieve that, correct? 10 A. So there is some fat loss in wastewater, but it's 11 very, very, very minimal, because you are going to wash 12 the vat or the line anyway, and all that is going to go to 13 waste. But the amount is very, very low. Most of the fat 14 loss occurs during whey drainage. 15 Q. But there is some fat loss? 16 A. Very low. 17 Q. But it is above zero, correct? 18 A. I don't have a number, but it's very, very, very, 19 very low. 20 Q. I understand. 21 But would you agree with me it is above zero? 22 A. It depends on how many decimal places and how 23 significant it is. 24 Q. Okay. Nonetheless, your own statement says, on 25 page 2, "Most of the fat losses during cheese making occur 26 during whey drainage"? 27 A. That's what I just said, yes. 28 Q. Okay. Now, the word "most" doesn't mean "all." 4608 1 So are there other fat losses other than through whey 2 drainage? 3 A. Yes, manipulation of the curd can occur anywhere 4 where you are -- you are losing components. 5 Q. Okay. And going back to the Double O vats, you 6 would agree that butterfat recovery from Double O vats 7 would not be at 93%, correct? 8 A. It is achievable if you do a good job, yes. 9 Q. But achievable is not the same thing as achieved, 10 is it? 11 A. It's become semantics, sir. 12 Q. Okay. So I realize you are not here to talk about 13 Federal Order policy. But are you familiar with the 14 current Class III milk formula in Federal Orders and how 15 it accounts for fat? 16 A. No. 17 Q. Okay. If -- if I were to represent to you that my 18 understanding of the two proposals from Select, which are 19 Proposals 10 and 11, if both were adopted, that the cheese 20 yield formula would use 93% milk fat going to cheese and 21 7% of fat recovered as sweet cream for use in Grade AA 22 butter, so essentially it's 100% fat recovery, does that 23 make sense to you? 24 A. Please come again? 25 Q. Okay. If combination of the proposals submitted 26 by Select were to achieve in one part of the formula 93% 27 of the milk fat going to cheese and 7% of fat being 28 recovered as sweet cream for use in Grade AA butter, that 4609 1 adds up to 100%, correct? 2 A. Yes. 3 Q. Okay. Does that make sense to you that it would 4 be 100%? 5 A. I don't know the values for the mass balance, but 6 93 plus 7 is 100. 7 Q. But you have already mentioned that there are some 8 losses, although you can't define them, correct? 9 A. Again, it depends on how many decimal places you 10 are going to. 11 Q. Do you know -- and if this gets beyond what you 12 are talking about that, that's fine -- but do you know how 13 whey cream is marketed? 14 A. I know about whey cream, but I'm not into 15 marketing, so I can't comment on how it's marketed. 16 Q. So would you know whether it could typically be 17 use in Grade AA butter products? 18 A. Can whey cream be used in butter? There are 19 people who use whey cream in butter making. 20 Q. But can it be used -- 21 A. It is called whey cream butter. 22 Q. Yes. It's called whey cream butter. 23 Can it be used in Grade AA butter products -- 24 A. I don't know the regulations there, so I cannot 25 answer that. 26 MR. ENGLISH: That's all the questions I have. 27 Thank you. 28 THE COURT: Anyone else have questions other than 4610 1 AMS? 2 AMS. 3 CROSS-EXAMINATION 4 BY MS. TAYLOR: 5 Q. Good morning. 6 A. Good morning. 7 Q. Thank you for coming and testifying today. Just a 8 couple questions. 9 I want to turn to your statement on page 2. And 10 you have a Table 1 listing the average composition of milk 11 by different breeds of dairy cattle. I just wanted to 12 know the source of that. 13 A. That's any dairy chemistry book would have that. 14 So a good one would be Pat Fox's books. 15 Q. And these are -- so these are average 16 compositions, recent? 17 A. Yes. 18 Q. And so on your figure on the next page, what I 19 think the take-away here is you looked at All-Jersey milk, 20 and you looked at all Holstein milk? 21 A. That's correct. 22 Q. And while the fat levels in the cheeses differed, 23 as a percentage basis, the recovery was the same? 24 A. That's correct. 25 Q. And then the middle cluster is where you blended 26 the two, and that is -- and is it right that that is what 27 is more realistic of what is currently done in order to 28 standardize to meet the standard of identity? 4611 1 A. No. So if a plant does not have milk coming from 2 two sources, that's not -- this was a study that was done 3 because at that time the sponsor was -- had both Jersey 4 and Holstein and wanted to find the optimum. 5 Q. Okay. And when was that study done? 6 A. 1983, '84. 7 Q. Okay. So on page 4 when we're talking about -- 8 you're talking about cheese equipment, and I know you have 9 been asked a few questions on this. But you talk about 10 how 93% is achievable using Tetra Pak equipment. 11 My first question would be, what other types of 12 equipment are out there or how many cheese manufacturers 13 do or do not use this type of equipment? 14 A. So there are probably about -- for cheddar cheese 15 making, there are probably maybe two or three major 16 players there in equipment making, and they all have, 17 different versions. Because some of these equipments are 18 patented. Just because of the design, they may have 19 slight variations, but they do similar work. 20 Q. And you mean slight variations in the butterfat 21 recovery? 22 A. No, in the design. 23 Q. In the design. 24 So do you know anything about the butterfat 25 recovery in those? 26 A. Fairly similar. 27 Q. Okay. 28 A. And I can bet you that most, if not all, the large 4612 1 commodity cheese plants are going for the more efficient 2 equipment. 3 Q. The newer -- the newer plants being built? 4 A. Yeah. 5 Q. Can you speak to the -- about the plants that are 6 maybe still using Double O vats, the number of plants or 7 the age of the plants or -- 8 A. Very little. I mean, as a side note, I wasn't 9 using -- and it's maybe in the -- I would say in the early 10 '90s this plant had Double O vats that they had taken out, 11 was a year old. And I asked, why did you take it out? 12 The answer was, just do the math with butterfat recovery. 13 It will pay for itself. And when I did the math, within a 14 year, it will pay for itself. 15 Q. On page 5 when you talk about coagulants and the 16 use of better coagulants can create stronger protein bonds 17 and then trapping more fat. So can you talk about the 18 prevalence of industry cheese makers in using these better 19 coagulants? 20 A. Most cheese plants, I mean, use the better 21 coagulants. In the United States. 22 Q. Okay. And so when you talk about "most cheese 23 plants" use that or you use Tetra Pak equipment, can you 24 speak a little bit about how you have that knowledge? 25 A. As a consultant in the industry, as a reviewer of 26 articles, either as an expert reviewer or as an author, 27 and of course just knowing the industry. 28 Q. Is some of that going into plants to observe and 4613 1 studying their data to see that's what they achieve? 2 A. Going to plants as a consultant. 3 Q. Right. 4 A. Uh-huh. 5 Q. On the last page, the last sentence ends, "The 6 majority of commodity cheddar cheese manufacturers," and I 7 think you clarified in response to Mr. English that 8 that's, in your opinion, both in volume in cheese and 9 number of plants. Then you say, "are achieving fat 10 recoveries at or above these levels." 11 Are these levels 90% or 93%? 12 A. 93%. 13 Q. Okay. So is it of your opinion that the current 14 modified Van Slyke formula, which assumes a 90% fat 15 recovery, is not valid anymore based on current technology 16 and ingredients available to use in cheese making? 17 A. I think the 93% is a more appropriate number. 18 Q. And there's been -- when we look at cheddar cheese 19 and we survey that for both -- for our formulas, we look 20 at both barrels and blocks. 21 So is there any difference in the process in the 22 fat recovery in producing those two products? 23 A. No. Up to the point of packaging, the processes 24 are the same. 25 Q. And generally manufacturers who make either of 26 those, they use all the same equipment? 27 A. They use the same equipment up to the time of 28 packaging. Then they -- it's diverted into the barrel or 4614 1 block format. 2 MS. TAYLOR: I think that's it from AMS. Thank 3 you. 4 THE WITNESS: Thank you. 5 REDIRECT-EXAMINATION 6 BY MR. MILTNER: 7 Q. Dr. Farkye, just a follow-up question or two on 8 the questions you received from USDA. 9 Where you are not a consultant visiting a plant 10 but instead are reviewing someone else's papers or 11 research, would the author of that paper have visited the 12 plant and seen the data firsthand before you stepped in as 13 a reviewer? 14 A. So I have done both. I have been a consultant in 15 the plant where I have been involved in cheese making, 16 observed their data, and made my comments. So it's been 17 both. 18 Q. It's been both. 19 So your conclusions about what plants are 20 achieving, and not just what is theoretically achievable, 21 is that conclusion drawn on your own personal observations 22 or your review of the direct observations of others in the 23 industry? 24 A. It's both. 25 Q. It is both. 26 And then Mr. English asked you about the e-mail 27 from Mr. Steffens and the information from Chris Hansen. 28 Those were not the sole sources for your 4615 1 conclusions today, were they? 2 A. Oh, no. 3 Q. There are -- there are -- I guess are they a 4 confirmatory data point in your analysis? 5 A. They are one of several. 6 Q. Okay. 7 MR. MILTNER: I don't have any other questions 8 then. 9 And, your Honor, we would ask the admission of 10 Exhibit 224 and 225. 11 THE COURT: Seeing no objections, Exhibits 224 and 12 225 are admitted into the record. 13 (Thereafter, Exhibit Numbers 224 and 225 were 14 received into evidence.) 15 MR. MILTNER: Thank you. 16 Thank you, Dr. Farkye. 17 THE WITNESS: Thank you. 18 THE COURT: Is now a good time for lunch? Let's 19 come back at 1:05. 20 (Whereupon, a luncheon break was taken.) 21 22 23 24 25 26 27 28 4616 1 TUESDAY, SEPTEMBER 19, 2023 - - AFTERNOON SESSION 2 THE COURT: Let's go on the record. 3 We're on the record. Off the record we discussed 4 four-originally marked as Edge exhibits that we discussed 5 on September 15th, and we were going to wait until we had 6 hard copies to move them. We marked them, but we wanted 7 to have hard copies to move them into the record. And AMS 8 I think has provided hard copies now, so we'll consider 9 these to go into the record. 10 I don't know that we need to take them one by one. 11 I'll take them as a group. We have Exhibit 205, which was 12 Edge-9; Exhibit 210, which was Edge-7; Exhibit 211, which 13 was Edge-10; and Exhibit 213, which was Edge-8. 14 Any objections? 15 Seeing none, those four stated exhibits are made a 16 part of the record. 17 (Thereafter, Exhibit Numbers 205, 210, 211, 18 and 213 were received into evidence.) 19 THE COURT: Mr. Miltner. 20 MR. MILTNER: Thank you, your Honor. We would 21 call Chris Allen to the stand. 22 THE COURT: Welcome. I guess we'll swear you in. 23 CHRIS ALLEN, 24 Being first duly sworn, was examined and 25 testified as follows: 26 THE COURT: Your witness. 27 DIRECT EXAMINATION 28 BY MR. MILTNER: 4617 1 Q. Mr. Allen, do you have in front of you a document 2 marked in the upper right as revised Exhibit Select-9? 3 A. Yes. 4 MR. MILTNER: Your Honor, could we mark that as a 5 Hearing Exhibit for identification? 6 THE COURT: Yes, we can. Select-9 is marked for 7 identification as Exhibit 226. 8 (Thereafter, Exhibit Number 226 was marked 9 for identification.) 10 MR. MILTNER: Thank you. 11 BY MR. MILTNER: 12 Q. Mr. Allen, you have seen this document before, 13 correct? 14 A. Yes. 15 Q. And it is a written statement which summarizes 16 the -- primarily summarizes the regulatory text changes 17 for the adoption of Select's proposals, correct? 18 A. Yes. 19 Q. Now, this is a revised exhibit, it differs from 20 the one submitted in advance of your testimony, correct? 21 A. Yes. 22 Q. And we distributed that to counsel this morning, 23 as well as AMS. 24 What are the two principal -- really the two only 25 changes to the exhibit? 26 A. Well, first we found that in our initial 27 calculations of the outcomes of the changes to the 28 formula, that we had incorrectly entered the monthly 4618 1 commodity price data. We had entered the prices as two 2 decimals, rounded to the first two decimals. If we had 3 been more thoughtful, we would have noticed that those 4 should have been entered as rounded to four decimals for 5 each month. So those five-year and ten-year periods. We 6 went back and corrected the average calculation to rounded 7 to four decimals, and I sorted those commodity prices, 8 those revised commodity prices into formulas, in the 9 original formulas, the revised formulas for each proposal, 10 and established a new outcome of those prices. 11 Q. And that revised data is reflected in the two, I 12 guess, tables on page 1 and page 2 of Hearing Exhibit 226, 13 correct? 14 A. That is correct. 15 Q. And the calculations of the effects of the 16 proposals, there were modest changes in the terms of a few 17 pennies -- 18 A. Correct. 19 Q. -- that's correct? 20 Okay. How about the second change to -- to this 21 exhibit? 22 A. Again, just reviewing our process, we recognized 23 that we had failed to indicate regulatory text change 24 related to the advanced price. A portion of the 25 regulatory text addresses the butterfat price without 26 using references to the other portions that we had already 27 suggested edits for, so we added in Section 7 of 28 CFR 1000.50 paragraph (q)(3), we made an adjustment there, 4619 1 a revision there to the regulatory text. 2 Q. And the change to the multiplication factor there, 3 it's the same -- same number as we have proposed a change 4 in 7 CFR 1000.50(l); is that correct? 5 A. Yes. 6 Q. And just so the record's clear, the changes that 7 Select has proposed in Proposals 10, 11, and 12, they 8 change yield factors for the various milk components, 9 correct? 10 A. Yes. 11 Q. And those components primarily affect the formulas 12 for Class III and IV milk, correct? 13 A. Correct. 14 Q. But butterfat, there's an advanced butterfat 15 factor that goes into the advanced pricing, correct? 16 A. Yes. 17 Q. And the other advanced factors, their changes 18 refer to CFR sections without restating the particular 19 mathematics, correct? 20 A. Yes. 21 Q. So there's no change in what Select intends to 22 accomplish with any of its proposals, correct? 23 A. Correct. 24 Q. Okay. Is there anything in this statement that 25 you feel the need to read into the record? 26 A. I do not. I mean, we have covered the key points 27 of the statement. 28 Q. Are there anything else that you would like to 4620 1 address with respect to this exhibit? 2 A. No, I don't -- I don't believe so. 3 MR. MILTNER: Okay. We would make Mr. Allen 4 available for additional questioning, your Honor. 5 THE COURT: Questions for this witness? Other 6 than AMS? 7 Looks like you are up AMS if you are ready -- or 8 even if you aren't. 9 MS. TAYLOR: Lucky for us, we are. 10 CROSS-EXAMINATION 11 BY MS. TAYLOR: 12 Q. I don't think we -- I have -- we have one 13 question, as you gave us a lot of food for thought to go 14 back last night and think about on the buttermilk piece. 15 So since you are back up here, I want to get your 16 thoughts on from a policy perspective, why does Select 17 feel it's important to account for buttermilk powder in 18 the yield formula as opposed to other, including other 19 possible products? Like why are we focusing just on 20 buttermilk powder I guess is the question? 21 A. I believe it's just because that's how the initial 22 price formula was established, with that focus on 23 buttermilk powder, and we just believe that those 24 underlying assumptions were not established properly. 25 Q. Okay. 26 MS. TAYLOR: Okay. Okay. That's it. 27 THE WITNESS: Thank you. 28 /// 4621 1 REDIRECT EXAMINATION 2 BY MR. MILTNER: 3 Q. Since AMS asked the question, let me add a little 4 bit -- or ask you a few more about that. 5 When -- when a producer sells milk to a Class III 6 manufacturing plant, to your knowledge, all of the 7 components that are delivered are priced in one product or 8 another, aren't they? 9 A. Yes. 10 Q. And if you deliver milk to a Class IV plant that 11 produces butter and nonfat dry milk, are all of the 12 components delivered to that plant priced? 13 A. Yes. 14 Q. Even the components that end up in buttermilk 15 powder? 16 A. Yes. 17 Q. Correct? Like now they are? 18 A. Can you please restate your question? 19 Q. Sure. Sure. 20 Under the current price formulas, milk delivered 21 that a Class IV plant -- 22 A. Yes. 23 Q. -- solids that end up in buttermilk powder -- 24 A. Right. 25 Q. -- do producers get paid for it? 26 A. Assume that the value is negligible. 27 Q. Okay. But those plants are indeed selling 28 buttermilk powder in many cases, correct? 4622 1 A. They are. 2 Q. Okay. Sorry my questions weren't as clear as they 3 probably should have been. 4 A. I understood your question to be how the producer 5 was paid on the components of the milk received by the 6 plant. That's why I wanted you to restate it. So I'm 7 sorry about that. 8 Q. No, I'm glad you -- I'm glad you asked me to 9 clarify. 10 MR. MILTNER: I don't have any other questions, 11 and so we would move admission of Exhibit 226. 12 THE COURT: Okay. Seeing no objections, 13 Exhibit 226 is admitted into the record. 14 (Thereafter, Exhibit Number 226 was received 15 into evidence.) 16 MR. MILTNER: Thank you. 17 THE COURT: Thank you, Mr. Allen. 18 MR. NIELSEN: Good afternoon. Eric Nielsen, 19 counsel for Leprino Foods Company. 20 THE COURT: Okay. I need to swear in the witness. 21 Please raise your right hand. 22 ALISON KREBS, 23 Being first duly sworn, was examined and 24 testified as follows: 25 THE COURT: Your witness. 26 MR. NIELSEN: Thank you, your Honor. 27 I have just circulated a document marked 28 IDFA Exhibit 43. I'd like to have that document marked 4623 1 Exhibit 227 for identification purposes. 2 THE COURT: Yes. That exhibit is so marked. 3 (Thereafter, Exhibit Number 227 was marked 4 for identification.) 5 MR. NIELSEN: Thank you, your Honor. 6 DIRECT EXAMINATION 7 BY MR. NIELSEN: 8 Q. Please state your name for the record. 9 A. Alison Krebs. 10 Q. Ms. Krebs, you have previously testified in this 11 hearing, on September 5th and September 14th, correct? 12 A. Yes. 13 Q. And has anything changed materially in your 14 professional or educational background that you provided 15 in previous testimony? 16 A. No. 17 Q. Okay. Ms. Krebs, the document in front of you 18 marked Exhibit 227 for identification purposes, is that a 19 true and accurate representation of the testimony you 20 intend to present today? 21 A. Yes. 22 Q. Great. Please proceed with your testimony. 23 A. Thank you. 24 I'm Alison Krebs, director of dairy and trade 25 policy, for Leprino Foods Company (Leprino), headquartered 26 in Denver, Colorado, as I have previously provided my full 27 introduction in prior testimony during this hearing. That 28 information has already been entered into the record, so I 4624 1 will not repeat it here. In this tranche of testimony I 2 will address Proposals 10 through 12. 3 Opposition to Proposals 10, 11, and 12, yield 4 factor adjustments. 5 Leprino Foods opposes the three proposals from 6 Select Milk: Proposal 10, to update butterfat recovery to 7 93%; Proposal 11, to update specified yield factors to 8 reflect actual farm-to-plant shrink; and Proposal 12, to 9 update the nonfat solids factor from 0.99 to 1.03. 10 Leprino's opposition to these proposals rests 11 primarily on the fact that a more comprehensive review of 12 the yield assumptions and the losses throughout the 13 balance of the manufacturing process must be completed in 14 conjunction with any changes. 15 The Van Slyke yield formulas (which form the basis 16 of the current cheddar yield factors) and the Select Milk 17 proposal are premised on components in a vat. That yield 18 formula does not address the other losses that occur 19 throughout the production process. 20 While we do not dispute that some cheddar plants 21 achieve the 93% fat retention that is proposed, this was 22 also the case at the time the current factors were 23 established. The question is whether the vats that 24 facilitate this higher fat capture have been fully 25 implemented and whether the proposed capture rate is 26 achieved across the broader industry. The broad industry 27 data needed to make such updates is not currently 28 available. 4625 1 Even more importantly, Select Milk does not 2 address the fact that the current formula assumes that all 3 fat not captured in finished cheddar is processed into 4 Grade AA butter. This assumption neither recognizes 5 in-plant milk component losses nor that butter 6 manufactured from whey cream is not legal under standards 7 for Grade AA butter. 8 In summary, these proposals essentially 9 “cherry-pick” yield factors within the formula. If some 10 of the yield factors are to be evaluated, then all yield 11 factors should be considered. 12 Finally, if this hand-picked group of factors is 13 updated without broad, publicly available data, it would 14 directly conflict with the logic USDA provided in the 15 following quote from the 2013 Final Decision regarding the 16 valuation of whey cream (p. 9274), quote: "While there is 17 record evidence from some manufacturers as to their 18 individual saleable volumes and values of whey cream, that 19 limited data does not provide for a reasonably complete 20 assessment of the national market for whey cream and its 21 various competing uses. Accordingly, Proposals 9 and 10 22 are not proposed to be adopted." 23 If, after a thorough vetting of all yield 24 assumptions in the Class III formula in a future 25 rulemaking proceeding record evidence supports the 26 proposed increases in fat retention to 93%, Leprino Foods 27 would not oppose that the butterfat recovery factor be 28 moved to 93%. However, that change must be accompanied by 4626 1 a broader vetting of data and recognition of in-plant 2 losses, along with proper valuation of whey cream rather 3 than following Select Milk's proposals that cherry-picks 4 factors to update. 5 If Congress grants USDA the authority to conduct 6 regular, mandatory cost of processing studies, yield data 7 (including butterfat recovery) could become part of this 8 process as well. When that study data is available, the 9 industry would then have broad, publicly available data 10 from which to update these factors. 11 There is an important caveat for cheese if 12 mandatory studies are used to standardize yield factors. 13 Vat component data needs to be detailed in order to 14 accurately identify yield drivers, including from 15 fortification ingredients, rather than assuming that the 16 vat components mirror those of the incoming raw milk. 17 Fortification is the process of including other, more 18 concentrated milk products such as nonfat dry milk, 19 condensed skim, or ultra-filtered milk in the cheese vat 20 along with milk. Fortifying the cheese make process with 21 these products enables cheese makers to improve 22 productivity and plant utilization, manage raw ingredient 23 inventories, and manage input economics based on market 24 price relationships. 25 If sufficiently detailed vat component and yield 26 information is captured as part of a mandatory industry 27 survey for the purpose of updating these factors in milk 28 pricing formulas, Leprino Foods would be open to 4627 1 considering use of such mandatory study data to update 2 relevant formula factors going forward. At a minimum, 3 such data, if accurately collected, could validate both 4 the yields and the losses that are inherent to 5 manufacturing cheese and its related proposals. 6 Specific to Proposal 11, the proposal to eliminate 7 the allowance for farm-to-plant shrink, many of the same 8 principles noted immediately prior also apply here. The 9 key difference being that instead of yields, we're 10 considering the difference between the components and 11 volume that are measured at the farm bulk tank versus what 12 is delivered to the manufacturer. 13 The starting point of the Van Slyke yield formula 14 is the dairy components in a cheese vat at the start of 15 cheese making. However, milk priced under Federal Milk 16 Marketing Orders is sampled for components and measured 17 for volume at the farm. Elimination of the allowance for 18 the farm-to-plant shrink denies the reality that not all 19 volume or components measured at the farm make it into 20 cheese vats. Losses occur prior to delivery to the 21 manufacturer's milk silos, in addition to within the 22 production process. 23 We applaud that Select Milk Producers has limited 24 their own farm-to-plant volume losses. A simple 25 calculation of average farm size using the data on their 26 website suggests that their members deliver on average 27 231,898 pounds and assemblers are shipping multiple full 28 truckloads from single locations daily. This contributes 4628 1 to significantly lower losses than the industry norm. 2 Additionally, many Select Milk members scale their 3 milk, weighing the trucks before and after loading, and 4 eliminating the measurement of milk that is lost in the 5 transfer process between the milk bulk tank (or silo) and 6 the truck. 7 While Select Milk's performance in this regard is 8 laudable and aspirational, it is not reflective of the 9 broader dairy industry. The average farm size in most 10 milk sheds is significantly smaller than that of the 11 Select Milk's dairies. The 2017 Census of Agriculture 12 noted that just 8.8% of farms produced at least 39,500 13 pounds of milk per day. Trucks hauling milk from multiple 14 farms per load continue to suffer the same losses that 15 existed at the time the farm-to-plant loss assumptions in 16 the formula were first established. 17 Milk volume and fat loss may differ significantly 18 between the largest farms and smaller operations. For 19 much of the equipment that is used even today, a hose full 20 of milk is still lost on every farm between the farm's 21 bulk tank and the truck. For cheese makers buying milk 22 from smaller farms where a load includes multiple stops, 23 this volume loss remains significant. 24 Some milk sheds are solely comprised of small 25 farms, and those losses are consistent. Others have more 26 diversity in size. If the current volume allowance is 27 removed, this would incentivize cheese makers to buy from 28 larger farms or penalize farms that fail to provide a full 4629 1 load of milk. Creating this motivation would be 2 detrimental to the smaller farms across our rural 3 communities. 4 The characteristic of fat clinging to the inside 5 of stainless is no different today for most farms than 6 when the farm-to-plant loss was first acknowledged in the 7 formula. Many milk sheds are still dominated by smaller 8 farms where the fat that remains on the inside walls of 9 the farm bulk tank is meaningful relative to the volume of 10 milk. Flushing farm bulk tanks with water is considered 11 adulteration and is therefore illegal, so the fat clinging 12 to the inside of the bulk tank remains at the farm. 13 Similar to the volume loss differences across farm 14 sizes, Select Milk can be considered an anomaly with 15 regard to fat losses. Many of these large dairies sample 16 each tanker for components directly from the tanker 17 immediately after loading since the tanker is either being 18 direct-filled or may represent a portion of the volume of 19 a milk silo. 20 Consequently, one would expect lower differences 21 in fat tests than typically occur when components are 22 sampled in the bulk tank and fat is left clinging to the 23 interior surface of the tank, as is the case across most 24 farms in the US. 25 There is no evidence that volume and fat losses do 26 not occur between the farms and plants. While milk sheds 27 dominated by large dairies shipping full truckloads of 28 milk tend to have less significant losses than their 4630 1 smaller counterparts, those reduced losses are not 2 universal across all milk sheds or orders. 3 The evidence clearly does not support adoption of 4 Proposal 11; volume and fat loss still exist across the 5 industry, even at today's most efficient and innovative 6 plants. It is important that the farm-to-plant loss 7 assumptions embedded in the cheddar yield calculation 8 continue to recognize these losses to maintain orderly 9 marketing. 10 Leprino also opposes Proposal 12 because it does 11 not reflect the realities of manufacturing. Similar to 12 the cheddar yield factor, however, it is based upon a 13 theoretical yield approach that assumes a perfect system 14 with no losses before or after the conversion of solids 15 non-fat ("SNF") into nonfat dry milk. In-plant losses 16 exist not just with average, but with even the best 17 manufacturing practices. 18 For example, it is well known that cream includes 19 some SNF in addition to butterfat and water. Therefore, 20 one cannot assume all SNF is captured in nonfat dry milk. 21 Since cream is sold on fat value, there is no direct value 22 assigned to the skim solids in cream. Therefore, milk 23 could be overpriced relative to its value leaving the 24 market ripe for disorderly marketing. This was 25 well-stated in the February 7th, 2013 Final Decision (p. 26 9273): "It is important that the product-price formulas 27 reflect current plant conditions, not plant conditions 28 that may be possible but not reflective of general 4631 1 industry wide conditions." 2 For these reasons, Leprino Foods opposes Proposals 3 10, 11 and 12. 4 Q. Thank you, Ms. Krebs. One follow-up question. 5 Why is cherry-picking yield factors a risk from a policy 6 perspective? 7 A. Yep. Thank you. 8 I think -- yeah, I'm certainly not a deep expert 9 on cheese manufacturing. My expertise is more in the 10 realm of policy. And I think the aspect that we're 11 looking at here is if you are only adjusting yield factors 12 that potentially increase the price of milk, and you are 13 not looking at, for example, whey cream and that pricing, 14 you could get into a situation where you are 15 unintentionally adjusting things just in one direction, 16 and that may not be where the marketplace is right now. 17 And so, you know, one analogy that I come up with 18 is folks are familiar with the Far Side cartoon where you 19 have got the guy is rowing the Viking boat. And on one of 20 the boat you have these great, big, burly, strong guys, 21 and on the other side of the boat you kind of had these 22 skinny, scrawny guys. And one guy says to the other, "You 23 ever feel like we're rowing in circles?" 24 And so it really comes down to, do we have that 25 sense of balance within the industry in the formulas. And 26 if we just look at those factors that could increase that 27 price of milk, are we in a situation where we could risk 28 having some more disorderly marketing. 4632 1 You know, for example, I think it was -- it was 2 one of the two cheese makers from Wisconsin last week that 3 testified that they used Double O vats. And then another 4 one of them talked about buying milk from several smaller 5 farms. And, you know, those situations, are you getting 6 that butterfat retention? What is the farm-to-plant loss 7 that's happening? 8 And so, you know, maybe it's just pricing 9 adjustments that have to happen at the farm level for 10 those -- those particular plants to remain competitive or 11 do adjustments like this potentially run those types of 12 plants out of business, and then a farmer loses a market 13 for their milk. 14 So I think we need to be cautious about just 15 updating and selecting individual aspects of the formula 16 to update. And then I think from the other perspective 17 it's the whole notion of do we really have broad 18 representative data that goes sufficiently across the 19 industry to have solid evidence to build that policy on. 20 Q. Great. Thank you. 21 MR. NIELSEN: I will yield the witness for 22 cross-examination. 23 CROSS-EXAMINATION 24 BY DR. CRYAN: 25 Q. Roger Cryan for the American Farm Bureau 26 Federation. I'm sorry, I forgot to say that earlier 27 today. 28 Hello, Ms. Krebs, nice to see you. 4633 1 A. Likewise. 2 Q. Just to understand, you're -- what you are -- you 3 are saying is that if you are going to adjust the yields 4 in the formulas, you have to have a mandatory audited 5 survey to -- to get the best data in order to do that 6 across the board? 7 A. That's not quite what I said. I think with yield 8 factors, we haven't had a way to get representative data 9 across the industry. For other proposals, such as 10 Make Allowances, we do have a history of generally 11 accepted studies that have been produced that provide us 12 with evidence and data. 13 Q. Is it easier to measure yields directly than costs 14 for across the whole plant? 15 A. You know, I'm not an expert for plants, so I can't 16 answer that. 17 DR. CRYAN: Okay. Thank you. 18 THE WITNESS: Uh-huh. 19 CROSS-EXAMINATION 20 BY MR. MILTNER: 21 Q. Good afternoon, Ms. Krebs. 22 A. Good afternoon. 23 Q. Ryan Miltner representing Select Milk Producers. 24 So if we oversimplified the whole end product 25 pricing schema, we'd start with a surveyed price of a 26 commodity, we subtract the cost of manufacturing that 27 commodity, and we'd multiply that times a yield factor, 28 and we get some value, correct? 4634 1 A. Correct. 2 Q. And in this hearing, you would agree IDFA has one 3 of the proposals to change the allowance, the 4 Make Allowance part of that formula, correct? 5 A. Uh-huh. Yes. 6 Q. And there are other proposals that would change 7 what prices, what surveyed commodities we'd plug into that 8 formula, correct? 9 A. Yes. 10 Q. Okay. And yet I think your statement says that we 11 should not look at any individual aspect of the formula, 12 we should look at it in total. 13 Did I get that correct? 14 A. I think that's a matter of nuance. I -- I wasn't 15 intending to say that you can't look at aspects of the 16 formula. 17 Q. You -- IDFA thinks we should look at the 18 allowances, correct? 19 A. Yes. 20 Q. IDFA does not think we should look at the yields, 21 though, correct? 22 A. That's a concern that -- and I'm here representing 23 Leprino Foods Company. But, yeah, we -- we have a concern 24 that you just don't have the data that's needed to make a 25 good judgment at this time. 26 Q. So Leprino believes we should look at the 27 allowances, though? 28 A. Yes. 4635 1 Q. But Leprino's position is that we should not at 2 this time look at the yield factors, correct? 3 A. Correct. It's just limitations on data. 4 Q. Okay. Let's start looking at your statement 5 because I have a few questions at different parts of it. 6 A. Okay. 7 Q. On the first page, third paragraph, in the middle 8 of the paragraph, you state, "While we do not dispute that 9 some cheddar plants achieve a 93% fat retention that is 10 proposed, this was also the case at the time the current 11 factors were established." 12 Do you recall USDA's comments on a higher 13 butterfat retention at the time of that decision? 14 A. No, I do not. 15 Q. If the comments were something along the lines of, 16 that might be achievable, but there's no evidence that 17 it's broadly achieved, would that surprise you? 18 A. No, that would not. 19 Q. And so if that number were now broadly achieved, 20 would that change Leprino's position on whether a 93% fat 21 retention is appropriate to incorporate in the formulas? 22 A. I think if we had that data that demonstrated 23 that, that that would certainly be worth looking at. 24 Q. Okay. And ultimately the Department's decision to 25 determine whether there's sufficient data to change that, 26 correct? 27 A. Yes. 28 Q. On page 2, at the first full paragraph, you refer 4636 1 to Congressional grants of authority to conduct mandatory 2 cost of processing studies, and you note that yield data, 3 which includes butterfat recovery, "could become part of 4 this process as well." 5 I don't want to belabor the discussion we have had 6 at the hearing about how long that process might take 7 and -- and whether Congress would actually pass something, 8 but do you know if inclusion of yields is currently in the 9 proposed legislative language being discussed? 10 A. My understanding is that, yes, it is. 11 Q. Okay. Further on page 2 you state that "the 2017 12 Census of Agriculture noted that just 8.8% of farms 13 produced at least 39,500 pounds of milk per day." 14 Did you calculate at all the volume of milk 15 production that comes off of farms that can produce at 16 least 39,500 pounds of milk per day? 17 A. I did not calculate the volume. 18 Q. It's -- it's a super majority of the volume of 19 milk produced, though, isn't it? 20 A. I wouldn't doubt that at all. 21 Q. And so for Leprino's plants, a super majority of 22 the milk you receive comes on tankers that have 23 single-farm deliveries or single-farm pickups, correct? 24 A. For some of our plants, yes. I would say for 25 three of our plants, one in particular, that that 26 certainly is not the case. And so I think it varies from 27 milk shed to milk shed. 28 Q. Do you have the data for Leprino as a system or as 4637 1 a total company? 2 A. Not with me, no. 3 Q. Would that have been useful to provide in your 4 statement, do you think? 5 A. It could have been. I -- I don't know if we 6 directly have access to that since we purchase the vast 7 majority of our milk through cooperatives or other 8 handlers. 9 Q. On page 3, in the third full paragraph, the last 10 sentence, "It is important that the farm-to-plant loss 11 assumptions embedded in the cheddar yield calculation 12 continue to recognize these losses to maintain orderly 13 marketing." 14 Are your comments limited to the cheddar yield 15 calculation? 16 A. In that particular case, it's addressing the 17 Proposal 11. Is that your question or -- 18 Q. I'm curious if your criticism of Proposal 11 is 19 limited to the cheddar yield portion of that. 20 A. It was intended to address the -- well, I think 21 the risk that you have across all the proposals is if you 22 are raising milk price to a level above a market-clearing 23 level, regardless of what the proposal -- or calculation 24 is, then you have some risk of disorderly market. But 25 this particular statement was primarily aimed at the 26 farm-to-plant shrink proposal. 27 Q. Okay. Thank you. 28 Your next paragraph, you -- you are discussing 4638 1 Proposal 12, and you state that the proposal is based on 2 "a theoretical yield approach that assumes a perfect 3 system with no losses." 4 Can you expand on that what you mean by that? 5 A. My understanding of that portion of the formula is 6 that it basically assumes that there's no losses, that 7 your mass balance is -- your milk you take in, you're 8 paying for in the products that are produced. 9 Q. Do you believe that the current Class IV formula 10 makes any payment to producers for the solids that end up 11 in buttermilk or buttermilk powder? 12 A. I don't know that I have the expertise to answer 13 that question. 14 Q. You further state in that same paragraph that 15 "since cream is sold on fat value, there is no direct 16 value assigned to the skim solids in cream." 17 Explain what you mean by that sentence, please. 18 A. Yeah. We do sell cream as a company, and we are 19 paid on a multiple of the butterfat price. And it doesn't 20 take into account any solids that are part of that cream 21 beyond just the fat. 22 Q. So do you track the cream multiples yourself? 23 A. I do not myself. 24 Q. Are you generally aware of what the cream multiple 25 range is in a typical year? 26 A. I have some idea of that, but I wouldn't want to 27 quote you a number because I'm sure I would be off. 28 Q. Have you -- during the past, say, six months have 4639 1 you seen cream multiples, you know, reaching 140 or 2 higher? 3 A. I honestly don't know, and I know that they can 4 vary quite a bit, depending on regions. 5 Q. So when you say a cream multiple, that means that 6 the cream is priced off a multiple of a butter price -- 7 A. Right. 8 Q. -- whether it is a CME price or an NDPSR price, 9 correct? 10 A. Correct. Yes. 11 Q. And so the sale of that cream, you take that 12 butter price, and you multiply it by some factor, and 13 that's the sales price, correct? 14 A. Yes. 15 Q. So if a buyer of that cream is utilizing all of 16 its components, cream and other solids, that multiple 17 captures the value of the whole tanker, correct? 18 A. It's priced off of fat though, so... 19 Q. But they are buying a tanker, correct? They are 20 buying a tanker load -- 21 A. Yes. 22 Q. -- or a tote of cream, correct? 23 A. Yes. They are. 24 Q. And the purchasers understand that they are not 25 buying a tanker full of butterfat, they are buying 36% 26 butterfat with some water and some other stuff in it, 27 correct? 28 A. I assume they have some knowledge of that, but the 4640 1 pricing is always off of the fat level -- or off of the 2 fat price. 3 Q. Does Leprino intend to introduce any data on its 4 own losses from your farms to your plants? 5 A. We don't have a lot of data available on that. 6 But, no, we are not planning to introduce that. 7 Q. Were you here when Mr. Allen delivered his 8 testimony on Proposal 11? 9 A. Yes. 10 Q. Do you believe he told the truth? 11 A. Yes. 12 Q. So when he testified that Select's proposals were 13 not aimed at increasing prices but getting the formulas 14 more precise, do you believe him? 15 A. I don't have any reason not to. 16 Q. And when he said that Select's philosophy is to 17 help make the formulas more precise so that producers 18 receive what Select perceives as the fair value for the 19 milk that they produce, you don't believe he was 20 misrepresenting his beliefs or that of Select, do you? 21 A. I don't believe he is. However, I think it comes 22 down -- potentially comes down to Select Milk and their 23 producers as opposed to the entire industry. 24 Q. And I don't -- I'm not trying to be theatrical. 25 A. Yep. 26 Q. But where you suggest that the proposals were 27 cherry-picked, do you believe that the proposals Select 28 submitted were specifically limited or targeted to those 4641 1 that would raise prices versus perhaps lower prices? 2 A. I don't know that it was intentional, but those 3 are the proposals that came forward. And I think that 4 it's -- regardless of how they are affecting milk prices, 5 we just need to be very careful that we don't end up with 6 distortions in the marketplace. 7 Q. Now, either you or others -- and I think you -- 8 have raised the issue of the value of whey cream and how 9 that's valued. IDFA and Agri-Mark proposed the last time 10 we did this to reevaluate cream. 11 Are you familiar at least that that occurred? 12 A. Yes, I am. 13 Q. Did -- and I may have asked you this before when 14 you were on the stand -- and if I did, I hope your answer 15 is the same, and I actually hope more I'm not repeating 16 myself -- did Leprino consider submitting a proposal on 17 the value of whey cream to this proceeding? 18 A. We had some consideration around it. And, no, you 19 didn't ask me this question before. But we -- we had some 20 consideration around it. But then after re-reading the 21 prior decision, the decision basically said there's no 22 publicly available data on whey cream pricing, and so 23 therefore, that would not be adopted. And we looked at 24 what data is available today, and there still is no 25 publicly available data on price of whey cream. And so we 26 figured we would hit the exact same roadblock again, so 27 why put the effort into it. I think there may be an 28 opportunity for going forward, is that something that USDA 4642 1 could get added to Dairy Market News to try and get a 2 dataset established so that that can be revisited in a 3 future hearing. 4 Q. If Select looked at the same decisions on issues 5 like butterfat recovery, farm-to-plant shrink, and nonfat 6 yield, and came to a different conclusion that, perhaps, 7 there was sufficient rationale for USDA to reconsider its 8 decisions, you wouldn't consider that cherry-picking would 9 you? 10 A. I think if there is sufficient broad data to 11 support that, that that could be balanced. It's just a 12 matter of looking across all of the different factors of 13 the formula. 14 Q. When it comes -- thank you. I appreciate that. 15 When it comes to the issue of the yields and the 16 formulas, do Leprino and Select at least agree that the 17 yields should be as accurate as they could be? 18 A. I believe all of the formulas would be best served 19 for the industry if it were accurate. It's just a matter 20 of, do we have good data, broad enough data to represent 21 that. 22 Q. So a disagreement on that issue of yields really 23 comes down to perhaps a policy decision as well as a data 24 quality decision? 25 A. Yeah, to me it's the -- do you have the breadth of 26 data to make a policy decision that is going to ensure 27 that we have the market-clearing price for milk. 28 MR. MILTNER: Thank you. 4643 1 THE WITNESS: You're welcome. 2 THE COURT: Additional questions for this witness? 3 Other than AMS I mean? 4 AMS, you are up. 5 CROSS-EXAMINATION 6 BY MS. TAYLOR: 7 Q. Good afternoon. 8 A. Hey, good afternoon. 9 Q. Just a few questions. On the first page when you 10 talk, you want a "more comprehensive review," I think 11 Mr. Cryan -- Dr. Cryan might have covered this, and I 12 missed the answer. 13 Can you elaborate -- is that the mandatory survey 14 that has been mentioned here many times? 15 A. Yeah, mandatory survey. Yeah, we do have -- for 16 Make Allowances, you do have some other sources of data 17 and information that are relatively broad across the 18 industry. Unfortunately, we don't have that breadth of 19 data available for the factors that are part of these 20 proposals. 21 Q. Okay. When it comes to your discussion on page 1 22 about how ninety -- excuse me -- 93% fat retention is -- 23 was achievable back in 2000. 24 But I think your statement is, generally, until we 25 have data to prove that out, we should keep the 90% 26 factor; is that correct? 27 A. Or whatever factor we can show is achievable on a 28 broad basis at this point. 4644 1 Q. And achievable is different -- you mean achievable 2 or is being achieved? 3 A. Broadly being achieved. 4 Q. Okay. And do you have any data to talk about what 5 is being achieved currently? 6 A. Unfortunately I do not. 7 Q. And just to make clear for the record, I'm -- is 8 it correct that mozzarella fat retention would not be the 9 same as cheddar fat retention? 10 A. That's my understanding, but I'm not an expert in 11 that area. 12 Q. You're more of an expert than I am in that area. 13 I was wondering on the last page, page 3, that 14 bottom paragraph that talks about your opposition to 15 Proposal 3. I don't think we're following kind of the -- 16 all of the logic that you have in that. So if you could 17 just try to summarize your opposition to that proposal for 18 us. 19 And I state that because, as I'm understanding 20 that proposal, is to account for buttermilk powder in the 21 nonfat dry milk yield, but yet I don't see that particular 22 point. 23 A. Yeah. You know, quite frankly, I'm not an expert 24 in the buttermilk powder realm of things. I think it's 25 really more about what I'm intending to say is sort of in 26 that -- again, that mass balance view of the world, there 27 historically is some demonstration that there are in-plant 28 losses or losses throughout the processing -- processing 4645 1 process. And so you have got milk going in, you have got 2 product coming out, and those don't necessarily match 3 perfectly. So if we can get a better understanding of how 4 those pieces fit together, then I think we would be better 5 informed to make those decisions from a policy 6 perspective. 7 Q. Okay. I'll ask another kind of question on that 8 realm. 9 Why would you think it's appropriate to continue 10 the policy that in-plant losses are accounted for in the 11 formulas and by extension, then, are accounted for by the 12 price that the farmer receives? 13 A. Could you ask that again, please? 14 Q. Sure. So the farm-to-plant losses are intended to 15 realize that all the milk from the farm doesn't 16 necessarily make it to the manufacturer. It gets lost in 17 that. And we can debate whether that happens now or what 18 the right percentage is, but theoretically that's what it 19 represents. 20 A. Sure. 21 Q. And in-plant losses represent the losses in the 22 manufacturing process. 23 A. Uh-huh. 24 Q. Those losses are not in the control of the farmer. 25 A. Correct. 26 Q. So my question is, why is it appropriate -- or 27 your opinion on the appropriateness of continuing 28 accounting for those losses in the formulas that do impact 4646 1 what the farmer gets paid, even though they don't actually 2 have any control over those in-plant losses? 3 A. Okay. If I understand you correctly -- 4 Q. Uh-huh. 5 A. -- you're -- on the in-plant losses, it may not be 6 something that over which the farmer has direct control, 7 but it's a reality of the industry. And it's a cost of 8 doing business and a reality that we experience as 9 processors. So I think, again, that's valuable 10 information for the industry to understand as we look at 11 how we value products, how we value and go through the 12 manufacturing or formula process because it's part of our 13 system. 14 Q. Yeah. That answers it. 15 A. Am I close? 16 Q. So this question always pops in my mind when I 17 think about things, and you are just the lucky person on 18 the stand to be here when it popped in my mind today. But 19 I always think to myself kind of why is that not accounted 20 for on the price end from the processor and instead is 21 accounted for on the back end to the farmer? 22 A. Well, for -- I mean, the farm-to-plant losses, 23 yes. 24 Q. I -- yeah, I'm not talking about that. 25 A. Okay. I don't know that I have anything more to 26 offer. 27 Q. Another conversation for when we can talk about it 28 when this hearing never gets done. 4647 1 A. There you go. 2 MS. TAYLOR: I think that's it. Thank you. 3 THE WITNESS: Sure. 4 THE COURT: Anyone else? 5 Okay, Mr. Rosenbaum. 6 CROSS-EXAMINATION 7 BY MR. ROSENBAUM: 8 Q. Steve Rosenbaum, International Dairy Foods 9 Association. 10 Is it fair to say the Federal Order system is 11 predicated on the notion that processors should pay for 12 milk based upon what money they can derive from selling 13 the products that they milk -- make with that milk, 14 correct? 15 A. Yes. 16 Q. And if, in fact, there's routinely a loss of milk 17 in the processing itself, obviously it doesn't result in 18 product that can then be sold, correct? 19 A. Yes, that's correct. 20 Q. And does the present formula, when it comes to 21 in-plant losses, essentially assume that, in fact, you 22 have been able to turn 100% of the milk that you received 23 into a useful, saleable product? 24 A. Yes. 25 Q. Is that -- is that -- and does that appear to be 26 an inaccurate assumption to the extent that, in fact, 27 there are inherent losses of milk during the processing 28 itself? 4648 1 A. Yes. 2 MR. ROSENBAUM: That's all I have. 3 THE COURT: Is that it for cross? 4 MR. NIELSEN: Thank you, Ms. Krebs. 5 Your Honor, at this time I would move to admit 6 Exhibit 227 into the record. 7 THE COURT: Objections? 8 Seeing none, Exhibit 227 is made a part of this 9 hearing record. 10 (Thereafter, Exhibit Number 227 was received 11 into evidence.) 12 THE COURT: Thank you, Ms. Krebs. 13 I'll swear you in while we have a minute. Please 14 raise your right hand. 15 MIKE BROWN, 16 Being first duly sworn, was examined and 17 testified as follows: 18 DIRECT EXAMINATION 19 BY MR. ROSENBAUM: 20 Q. Good afternoon, Mr. Brown. Good to see you again. 21 I have placed before you a document that's been marked as 22 IDFA Exhibit 44. 23 Is this your testimony regarding Proposals 10, 11, 24 and 12? 25 A. Yes, it is. 26 MR. ROSENBAUM: Your Honor, I would ask that this 27 be marked as Hearing Exhibit 228. 28 THE COURT: Yes. So marked. 4649 1 (Thereafter, Exhibit Number 228 was marked 2 for identification.) 3 BY MR. ROSENBAUM: 4 Q. Mr. Brown, could you please read the testimony? 5 A. Yes. 6 This testimony is submitted on behalf of the 7 International Dairy Foods Association, or IDFA, in 8 opposition to Select Milk Producers, Inc., Proposals 10 9 through 12. 10 Q. Mr. Brown, I think you can skip your discussion -- 11 A. Okay. 12 Q. -- about who IDFA is and who you are. You have 13 testified several times. 14 A. Yeah. There may be one new person, though, you 15 never know. 16 Q. Sure. 17 A. Summary of IDFA's objections to Proposals 10, 11, 18 and 12. 19 IDFA opposes Proposals 10 through 12. These 20 proposals seek to raise the butterfat recovery in the 21 Class III formula, eliminate farm-to-plant shrink, and 22 change the nonfat solids factor. While Select Milk 23 Producers has supplied some internal data in support of 24 these proposals, USDA does not have the benefit of any 25 broader industry or USDA studies relevant to the 26 consideration of these proposals. 27 When examined, it becomes clear that the proposals 28 would require pool handlers to pay for butterfat that 4650 1 cannot be uniformly recovered or valued at the Grade AA 2 price, pay for milk that may not actually be received in 3 the cheese vat for manufacture, and for nonfat solids 4 presumes a theoretical, rather than a real world system, 5 where there are no losses before or after the conversion 6 of solids nonfat into nonfat dry milk. 7 Further, the proposals choose to update only 8 factors that are revenue enhancing and ignore others, like 9 the current overstated whey cream valuation in cheese 10 making. 11 Select estimates that the added revenue from these 12 changes totals $0.12 to $0.13 per hundredweight to the 13 Class III price, and $0.41 to $0.42 per hundredweight to 14 the Class IV price. 15 But yield factors should not be addressed 16 piecemeal, but rather in a more holistic fashion, 17 examining all factors that impact product yields, 18 including factors not discussed in the proposals that 19 counterbalance Select's chosen factors for evaluation. 20 The proposals should be denied. 21 How Proposals 10, 11, and 12 would operate. 22 A: Proposal 10 would increase butterfat recovery 23 in the Class III formula to 93%. Proposal 10 would 24 increase the butterfat recovery in the Class III formula 25 to 93%, which results in a corresponding increase in the 26 butterfat yield in cheese to 1.624. 27 According to Select's analysis, adoption of the 28 this proposal would have increased the Class III price by 4651 1 $0.04 per hundredweight as compared to both the five- and 2 ten-year average. 3 MS. TAYLOR: Mr. Brown, could you slow down a 4 little bit? 5 THE WITNESS: Yes, I can. 6 MS. TAYLOR: Thank you. 7 THE WITNESS: Proposal 11 would update specified 8 yield factors to eliminate farm-to-plant shrink. 9 Proposal 11 would update the yield factors for butterfat 10 to 1.22 -- 11 (Off-the-record discussion took place.) 12 THE WITNESS: So I'll start with B again. 13 Proposal 11 would update specified yield factors 14 to eliminate farm-to-plant shrink. Proposal 11 would 15 update the yield factors for butterfat to 1.22, for the 16 protein value in cheese to 1.386, and for the butterfat 17 value in cheese to 1.582. Select asserts that the yield 18 factors for nonfat solids and other solids remain 19 unchanged due to rounding. 20 C: Proposal 12 would update the nonfat solids 21 factor from .99 to 1.03. Proposal 12 would replace the 22 current nonfat solids yield factor of .99 with 1.03. 23 According to Select's analysis, adoption of this proposal 24 would have increased the Class IV price from $0.35 to 25 $0.36 per hundredweight, as compared to both a five- and 26 ten-year average. 27 D: Select Milk Proposals 10 through 12 are taking 28 a piecemeal rather than a comprehensive approach to 4652 1 formula yield changes. Unlike multiple studies over the 2 past several decades that collected data from multiple 3 different manufacturing facilities owned by many different 4 companies with respect to the cost of manufacture for the 5 purpose of setting Make Allowances, Select presents no 6 such studies with respect to its yield assumptions and 7 losses, both before and after plant receipt and throughout 8 the production process. Instead, Select simply relies on 9 its own internal data regarding its own facilities. 10 IDFA supports maintaining the status quo until a 11 much broader base plant study is completed that establish 12 real world yields, shrinkage, and dairy solids recovery, 13 including values for that recovery. There are many 14 complicated issues, including fat recovery, plant loss, 15 and other factors across the dairy industry. Studies will 16 need to take into account plant ages, investments, and 17 processing techniques. USDA should first conduct 18 comprehensive reviews of the product yield assumptions and 19 losses. This would facilitate making yield adjustments in 20 a comprehensive, rather than piecemeal fashion. 21 E: The proposals selectively focus on 22 revenue-enhancing elements of the yield formulas. Related 23 to the lack of any industry studies is the fact that 24 Select has focused on dairy farmer revenue enhancements, 25 excluding other considerations. 26 For example, today's Class III formula presumes 27 that all excess fat from cheese manufacturing is 28 successfully recovered, 90% in the cheese and the 4653 1 remaining 10% ending up in the whey but valued as sweet 2 cream. 3 This presumption ignores the reality that: A, 4 every manufacturing system incurs losses in the form of 5 lost milk solids; and B, whey cream does not have the same 6 value as sweet cream, despite the wishes of all cheese 7 makers to the contrary. 8 Regardless of plant efficiency and full tanker 9 loads, many in the industry, especially in the Upper 10 Midwest and the Northeast, do not achieve full tanker 11 loads. Furthermore, unlike Select's assumption of no 12 farm-to-plant loss, it is likely that some purchased milk 13 solids are lost in that -- actually, in that 14 transportation, and data from all types of farms need to 15 be included in any analysis that would change the current 16 assumption of farm-to-plant loss. 17 We expect these yield studies can be accomplished 18 through the widely supported surveys for inclusion in the 19 upcoming Farm Bill. Simply put, Proposals 10 through 12 20 incorrectly assume that after applying the yield formulas 21 to milk processing, there is no need to account for other 22 losses that occur throughout the process from farm to 23 finish product. 24 F: Proponents' experience is not indicative of 25 broader experience. Select is known for its innovative 26 approaches and very large farms that likely generate more 27 efficient results and lower losses that are found in 28 industrywide. 4654 1 Just as with Make Allowances, it is critical that 2 AMS examine yields across the entire dairy industry, 3 recognizing that others do not experience the same 4 efficiencies and likely experience greater losses. And it 5 costs money to achieve many of these efficiencies, which 6 in turn impacts plant costs, although we acknowledge that 7 there would also be an adjustment for per pound product 8 costs resulting from these investments. 9 Three: Proposal 10's specific flaws. In addition 10 to the overall flaws applied to all of Select's proposals, 11 the 93% butterfat recovery proposal assumes, first, that 12 higher fat capture has been implemented by everyone, and 13 two, that all butterfat recovery has equal value. Without 14 the new yield studies mentioned above, there is no way for 15 USDA to conclude, hat the first point is accurate. 16 As to the value of butterfat, the butter not going 17 into cheese is valued under the current formula at 18 Grade AA butter, even though USDA by regulation assigns 19 such butterfat, known as whey cream, to Grade B butter. 20 With 20% or greater discounts on whey cream compared to 21 fresh cream, the Class III fat assigned to whey cream is 22 simply overvalued under the current formula. This is in 23 addition to in-plant losses of milk fat during processing, 24 which the current formula does not recognize. These 25 defects would need to be fixed as part of any revisions to 26 current formula yield factors. 27 Select seeks an increase in Class III prices in 28 Proposals 10 and 11 of approximately $0.12 per 4655 1 hundredweight, but they fail to recognize the greater, 2 more than offsetting decreases that would result from 3 accurately accounting for both processing losses and whey 4 cream values. 5 As with the butterfat recovery issue, Proposal 11 6 also assumes there is no farm-to-plant shrink. FMMOs 7 price milk based on components and volumes measured at the 8 farm, but losses occur prior to delivery to our member 9 plants. Select, again, may be an industry leader in 10 reducing farm-to-plant loss, but AMS should not base yield 11 factors on one company's experience, especially given the 12 fact that Select's dairy farmer members are large enough 13 that they can and do deliver full tanker loads of milk, 14 reducing the risk of leakage from farm tank to plant silo. 15 But less than 10% of all farms produce enough milk 16 to fill tanker loads of milk, meaning the vast majority of 17 trucks hauling milk are still delivering multiple loads of 18 milk. It is therefore reasonable to conclude that the 19 losses experienced when formulas were adopted are still 20 happening today. 21 Failure to account for the diversity of farm size 22 and the implications for farm-to-plant loss based upon 23 less than full tanker loads of milk would further 24 incentivize manufacturing to prefer large farms over 25 smaller farms. 26 The implications to USDA's necessary small 27 business regulatory analysis we leave to USDA, but it 28 appears to be detrimental to smaller farms and the rural 4656 1 communities that depend on those farms. As discussed in 2 DFA member testimony, fat clings to stainless tankers just 3 the same today as it did when the formulas were last 4 updated. 5 Again, milk sheds dominated by smaller farms 6 continue to experience larger loss of fat as a result. 7 Proposal 11 assumes away farm-to-plant losses in both 8 solids and fat. Until AMS conducts studies of these 9 issues, the proposals should not be adopted. 10 Proposal 12's specific flaws. Proposal 12 would 11 treat solids nonfat in the nonfat dry milk pricing formula 12 the same as proposed for cheese, a theoretical yield 13 approach relying on a perfect loss-less system. This is 14 not true even with the most modern and efficient 15 facilities, let alone average plants, often today 16 operating without the margin necessary to make the 17 investments that would be industry leading. 18 As just one example, after cream separation, some 19 portion of the solids nonfat remains together with the 20 butterfat and water. That lost SNF cannot then be 21 processed into nonfat dry milk, and nearly all cream is 22 priced on a multiple of the butter market, with no direct 23 value assigned to the skim solids in cream. Overvaluing 24 the volume of SNF and thus nonfat dry milk that can be 25 manufactured will overvalue and overprice the nonfat dry 26 milk that is market clearing and contributes to disorderly 27 marketing. 28 Conclusion: Adoption of Select’s proposals would 4657 1 at best be premature before widely supported AMS studies 2 are conducted, and likely would be a step backwards 3 because only producer revenue enhancing factors are 4 examined. This could overstate the impact of any yield 5 changes. Indeed, a comprehensive review would likely 6 result in revised yields factors that subtract from, not 7 add to, dairy farmer revenue. 8 MR. ROSENBAUM: Your Honor, Mr. Brown is available 9 for cross-examination. 10 CROSS-EXAMINATION 11 BY MR. MILTNER: 12 Q. Good afternoon, Mr. Brown. 13 Ryan Miltner representing Select Milk. 14 A. Good afternoon, Ryan. 15 Q. Mr. Brown, you are pretty familiar with the whole 16 Federal Order amendment process, aren't you? 17 A. It's been 15 years since I dealt with it, but 18 fairly -- fairly -- yes, fairly familiar. I'm old enough 19 to have been through a couple. 20 Q. What is the purpose when USDA issues an invitation 21 to submit proposals? 22 A. Anyone can submit proposals they think need to be 23 considered, and then USDA will decide which ones are 24 appropriate for a hearing. 25 Q. So were you -- where you have criticisms -- where 26 you have criticisms of the fact that the formulas don't 27 assume sufficient in-plant shrink, was that not an issue 28 sufficient enough for IDFA to submit a proposal? 4658 1 A. If you don't have good data, you can't submit the 2 proposal. And that was discussed, but we simply don't 3 have the data to support it. 4 Q. Your members did not have data to document their 5 in-plant shrink? 6 A. Some do, some don't. It's surprising the 7 differences. 8 Q. And similarly, the issue of the valuation of whey 9 cream was not important enough for IDFA to submit a 10 proposal for USDA to consider. 11 A. Again, we just -- we discussed that very issue. 12 And where we really came down is, is there a way that, for 13 example, to get Ag Marketing Service to survey whey cream 14 as well as sweet cream so we have numbers for the next 15 round when we have an issue again. We didn't think we 16 could provide industry data again from people that market 17 whey cream, but we did not feel that it would be adequate 18 proof or support for a change in recommended decision 19 based on our past experience. 20 Q. And, similarly, I assume your members did not have 21 sufficient data within their own operations to provide 22 support for that type of proposal? 23 A. It didn't seem to work before. We really wanted 24 to make sure it was something that there would be comfort 25 with. Quite honestly, that it would be adequate. I mean, 26 those of us who have bought and sell cream know what whey 27 cream is worth. But I wish I had a database to share, and 28 I don't. 4659 1 Q. Similarly, what's your understanding of the 2 purpose of a Hearing Notice issued by USDA? 3 A. It is a notice to consider the proposals they have 4 decided to bring to a formal rulemaking hearing. 5 Q. And isn't part of the reason to let everybody 6 within the industry or any other affected party know that 7 if you have opinions on these issues, here's where you 8 need to show up and let yourself be heard? 9 A. Yes. 10 Q. So where you criticize Select for only offering 11 its own data, the whole industry knows that if they have 12 different data that conflicts or supports Select's 13 position, they can show up and take the stand just like 14 you did, correct? 15 A. Yes, they can. 16 Q. Were you in the room for Mr. Allen's testimony 17 yesterday? 18 A. Yes. 19 Q. Did you hear his testimony on farm-to-plant shrink 20 specifically and state that if USDA felt that a reduction 21 in the amount of farm-to-plant shrink rather than its 22 wholesale elimination was better supported by the data, 23 Select would accept that decision? 24 A. Yes. 25 Q. Do you believe that the current farm-to-plant 26 shrink factors -- and by "you," I mean IDFA -- does IDFA 27 believe that those factors, as current, are appropriate? 28 A. We have no data to support or -- to support or 4660 1 oppose making those changes. We do know the shrink 2 exists. It's the same challenge quantifying the 3 information. I appreciate Select's effort to do that. 4 Q. Do you -- so can I restate that as IDFA doesn't 5 know if the current factors are the right factors or not? 6 A. We know that they were appropriate when adopted. 7 We do not know how they need to be adjusted today. 8 Q. The current butterfat retention factor in the 9 formula is 90%, correct? 10 A. Yes. 11 Q. Does that mean that USDA assumes that that 90% has 12 been achieved by every cheese plant? 13 A. No. 14 Q. So where you state that Select assumes that its 15 higher fat capture has been implemented by everyone, 16 that's not exactly precise, is it? 17 A. No, it is not. 18 Q. If, in fact, plants are achieving a 93% butterfat 19 recovery, wouldn't that necessarily mean that they are not 20 penalized by paying for 10% of their butterfat as whey 21 cream, accepting your characterization of whey cream 22 valuation? 23 A. Please repeat that. I'm sorry. 24 Q. Sure. 25 A. A couple moving parts there. 26 Q. And I probably made a compound question there. 27 You argue that currently the formula makes cheese 28 plants pay for 10% of their butterfat at Grade AA value, 4661 1 correct? 2 A. Yes. 3 Q. And that that 10% is actually more properly valued 4 as whey cream, correct? 5 A. Yes. 6 Q. Now, accept that characterization for purposes of 7 our questioning. If the formula were to move to 8 93% butterfat recovery, those plants would only be 9 overpaying on 7% of their butterfat; would that be 10 correct? 11 A. Yes. Assuming the 93% recovery is accurate for 12 the cheese, you are correct. 13 Q. And accepting that the whey cream is overvalued, 14 right? 15 A. Yes. That is true. 16 Q. On page 5, section 3, first paragraph, last 17 sentence, this -- you are summarizing -- I understand you 18 to be summarizing two flaws that you allege: The 19 valuation of whey cream and in-plant losses; is that 20 correct? 21 A. Yes. 22 Q. And you categorize those as defects, correct? 23 A. Yes. 24 Q. So where you say, "These defects would need to be 25 fixed as part of any revision to current formula yield 26 facts," again, that's IDFA's opinion on those points, 27 correct? 28 A. Yes. But en masse, not just certain parts. 4662 1 Q. But whether those need to be fixed, that sets an 2 opinion of IDFA, correct? 3 A. Well, yeah, it does. It also -- and I think we 4 all agree -- the need for good data to have these 5 discussions, which sadly we don't currently have. Let me 6 just say, you guys did a great job putting together 7 information. It's just too bad it's not broader. I guess 8 that's what I would say. 9 Q. And, again, any other co-op can come in and give 10 the same data that Harmoni Campbell did, and any plant can 11 come in and provide the same data that Cheslie Stehouwer 12 did, right? 13 A. That is correct. Yes. 14 Q. Including IDFA's members, correct? 15 A. Anyone who has a role in the Federal Orders, from 16 my understanding, yes, could testify. 17 Q. And the Federal Register let the whole world know 18 that now is the time, right? 19 A. Yeah. Times. They were going. 20 Q. Yeah. Times. 21 At the bottom of page 7 you say, "Select seeks an 22 increase in Class III prices." 23 You heard Mr. Allen testify that price increases 24 are the result of, and not the impetus for, the proposals, 25 correct? 26 A. Yes. 27 Q. So -- you also heard him say Select's intent is to 28 make the formulas more accurate, correct? 4663 1 A. Yes. 2 Q. And he -- he also testified that the butterfat 3 recovery factor might actually decrease prices in certain 4 circumstances, correct? 5 A. Yes. 6 Q. As a matter of fact, I think you and I had 7 conversations about that, correct? 8 A. Oh, yeah, we know the percentage, yes. 9 Q. We know the percentage. We figured that out. 10 A. Hopefully we came up with the same number looking 11 at it independently, which makes me feel better about my 12 educated guess. 13 Q. I'm more concerned that we figured it out. I'll 14 finish by asking you a similar question to that I asked of 15 Ms. Krebs. 16 Do IDFA and Select share the same goals in trying 17 to make these formulas as accurate as we can be? 18 A. Yes. I think we all -- well, certainly most of us 19 would like that to be the case, yes. 20 Q. And if to the extent there's a difference in how 21 we get to accuracy, those are primarily issues of policy 22 and data, correct? 23 A. Yeah. And I might add, it isn't good versus evil, 24 it's just what's the best way to get there. 25 Q. I would not have tried to make that 26 characterization, but I appreciate you clarifying it. 27 MR. MILTNER: Thank you. That's all I have. 28 THE COURT: Anyone else besides AMS? 4664 1 CROSS-EXAMINATION 2 BY DR. CRYAN: 3 Q. Hello, Mr. Brown. 4 A. Good afternoon. 5 Q. You are concerned about lack of data for the yield 6 changes. Is that -- would that be addressed through -- 7 best addressed through mandatory and audited survey of 8 costs and yields at plants? 9 A. A couple comments first. 10 We're delighted with your leadership. Working 11 with you and National Milk to get that in the Farm Bill 12 was one of our primary goals. So, yes, I think that is 13 important. 14 But I don't think you can let perfection be enemy 15 of the good, if you have got good strong data that's 16 reasonably broad covering a fair amount of product. And 17 that would be true with yields as well, but unfortunately, 18 we don't currently have that. 19 So that's -- that's where I come down. 20 That is the ideal. That doesn't mean it's the 21 only path. Because we haven't had it in the past, and we 22 still made decisions, which I think for the most part 23 we're pretty effective. 24 Q. But you believe that would meet your objections to 25 the -- 26 A. That -- oh, absolutely. I think we all -- that's 27 discovery we have needed since the start of Order Reform, 28 quite honestly. 4665 1 DR. CRYAN: Thank you very much. 2 THE WITNESS: You're welcome. 3 THE COURT: Any further questions? I'm mindful to 4 take a break. 5 Did you raise your hand earlier? Very good, sir. 6 AMS? 7 MS. TAYLOR: I think Mr. Miltner. 8 THE COURT: Oh, I'm sorry. 9 CROSS-EXAMINATION 10 BY MR. MILTNER: 11 Q. I left a note at my seat that had one more 12 question on it. 13 Mr. Brown, where you talk about the majority of 14 trucks on the road coming from multi-farm stops, how did 15 you come up with that number? 16 A. That -- that -- it shouldn't -- it's not majority 17 of trucks, it's majority of farms. Again, the volume 18 statements you made earlier are probably accurate. I 19 don't have the data. 20 Q. Okay. 21 A. Yes. 22 Q. I didn't want to try to do the calculation, but 23 that didn't make sense to me. 24 A. No. Hopefully that helped clarify. 25 Q. It did. Thank you. 26 A. You're welcome. 27 THE COURT: Okay. AMS? 28 MS. TAYLOR: Good afternoon. 4666 1 CROSS-EXAMINATION 2 BY MR. WILSON: 3 Q. Good morning, Mr. -- good afternoon, Mr. Brown. 4 A. Good afternoon. 5 Q. Todd Wilson, AMS. 6 I have a question on page 7 in discussing 7 Proposal 12's flaws. And we heard this from Ms. Krebs a 8 few minutes ago about theoretical yield approach. 9 Is it your understanding that the theoretical 10 yield approach that you reference is the concept of 11 accounting for buttermilk powder in the milk solids nonfat 12 part of the Class IV nonfat solids price calculation? 13 A. Currently? Not directly. I mean, it's inferred. 14 We read recommended decisions that this is discussed -- 15 Q. Can you speak closer to the mic? 16 A. I'm sorry, yes. 17 If you -- if you -- if you -- from my 18 recollection -- again, my brain doesn't work like it used 19 to -- but, yes, it was discussed as part of that entire 20 discussion on how you value nonfat solids in milk. But 21 it -- it's -- and I should remember because I was with 22 Darigold when this was discussed the last time -- but as 23 far as the detail of it, honestly, I don't remember. 24 Q. So in your statement you have, on the -- in the 25 first sentence after the hyphen, "a theoretical yield 26 approach relying on a perfect loss-less system." 27 Can you elaborate a little about more about what 28 you mean by those? 4667 1 A. What I mean is that the yields -- when you're 2 looking at yields in a plant, your yield should include 3 loss. I mean, it should, because that's the -- I mean, if 4 we're looking at value of milk, it's price times quantity, 5 more or less, on the different components, and so it 6 should be part of that consideration. 7 Does that mean that we have work to do on formulas 8 to make them more complete? Yes, I think that we do as 9 far as product yields. 10 MR. WILSON: Okay. Thank you. 11 MS. TAYLOR: That's it from AMS. 12 THE COURT: Thank you. 13 REDIRECT EXAMINATION 14 BY MR. ROSENBAUM: 15 Q. I mean, Mr. Brown, just to clarify, maybe it's 16 clear already, but when you refer to a loss-less system, 17 are you simply referring to the fact that the current 18 pricing system assumes that all the milk that comes into 19 the plant goes into a useable product? 20 A. Yeah, the only direct adjustments from 21 farm-to-plant loss, there are no plant losses in the 22 current formulas. 23 Q. Can you say that one more time? 24 A. I said, the only adjustments of the farm-to-plant 25 loss, there are no implicit plant loss as part of the 26 formulas currently. 27 MR. ROSENBAUM: That's all I have. 28 THE COURT: Okay. Let's move the exhibit into 4668 1 evidence. 2 MR. ROSENBAUM: Yes, your Honor. 3 THE COURT: Exhibit 228? 4 Seeing no objections, it is made a part of the 5 record. 6 (Thereafter, Exhibit Number 228 was received 7 into evidence.) 8 THE COURT: Let's take an afternoon break. Come 9 back at -- let's come back at 2:50. 10 Off the record. 11 (Whereupon, a break was taken.) 12 THE COURT: Back on the record. 13 Raise your right hand. 14 DR. PETER VITALIANO, 15 Being first duly sworn, was examined and 16 testified as follows: 17 THE COURT: Your witness. 18 DIRECT EXAMINATION 19 BY MS. HANCOCK: 20 Q. Good afternoon, Dr. Vitaliano. 21 A. Good afternoon. 22 Q. Thank you for round three on the stand at this 23 hearing -- no, round four. Math is getting more and more 24 difficult as we go along. Is it five? 25 A. I did a short one on the barrel cheese issue. 26 Q. Okay. Well, nonetheless, you're here on 27 higher-of; is that fair? 28 A. Yes. 4669 1 Q. Okay. And did you prepare Exhibit NMPF-30 in 2 support of your testimony today? 3 A. I did. 4 MS. HANCOCK: Your Honor, if we could mark that as 5 Exhibit 229? 6 THE COURT: Yes. 7 (Thereafter, Exhibit Number 229 was marked 8 for identification.) 9 BY MS. HANCOCK: 10 Q. Did you also prepare a spreadsheet that we have to 11 talk about with the higher-of as well? 12 A. Yes, I did. 13 Q. And is that identified as NMPF-30A, as in Adam? 14 A. Yes. 15 MS. HANCOCK: Your Honor, if we could mark that as 16 Exhibit 230. 17 THE COURT: Yes. Marked 230. Thank you. 18 (Thereafter, Exhibit Number 230 was marked 19 for identification.) 20 MS. HANCOCK: Thank you. 21 BY MS. HANCOCK: 22 Q. Dr. Vitaliano, would you please proceed with your 23 testimony? 24 A. Okay. 25 Hi. I'm Peter Vitaliano for the National Milk 26 Producers Federation, vice president of economic policy 27 and market research. I have been here before. I think I 28 have put my identifying information and background 4670 1 information into the record, but if there's anything that 2 I should repeat, let me know. 3 I have -- my statement consists of some 4 introductory background information on the federation and 5 the process that our organization went through to develop 6 a balanced set of proposals we brought to this hearing. 7 That's already in the record, and it's in my written 8 statement. I will not repeat it here. 9 The same with a later section on the overall 10 economic impacts of our suite of proposals. Likewise, 11 that's already been repeated -- been in the record and in 12 our written statement. I will not repeat that. 13 This testimony is in support of Proposal 13 14 concerning the base Class I skim milk price. Proposal 13 15 is to restore the original Federal Order Reform Class I 16 skim milk price mover. 17 NMPF requests the Secretary amend 7 CFR, paragraph 18 1000.50(b), applicable to all Federal Orders as specified 19 at the conclusion of this testimony, which would replace 20 the current Class I skim milk price mover with the 21 original Class I skim milk price mover in effect from 22 January 2000 through April 2019. The current language in 23 7 CFR 1000.50(b) is the product of two rulemaking 24 decisions: One, Federal Order Reform; and two, the final 25 rule implementing Section 1403 of the Agriculture 26 Improvement Act of 2015. Understanding both of these 27 actions is important to understanding the deficiencies of 28 the Class I mover -- the current Class I mover during 4671 1 periods of market instability since its implementation in 2 May 2019. 3 In Federal Order Reform, USDA adopted a new 4 Class I mover for the newly consolidated 11 Federal Orders 5 to replace the basic formula price, BFP. The BFP was 6 derived from a survey of prices paid for Grade B milk by 7 dairy manufacturing plants processing primarily butter, 8 nonfat dry milk, and cheese. It was, therefore, reported 9 as a single price which blended the value of Grade B milk 10 used to manufacture those products. 11 The BFP was discontinued at the end of 1999 due to 12 the declining and increasingly unrepresentative volume of 13 Grade B milk, and the Federal Order system subsequently 14 adopted product price formulas, or PPFs, to determine 15 minimum price, class prices. The transition to these new 16 class price formulas involved the adoption of four classes 17 of milk, including two full manufacturing use classes, III 18 and IV, with IV to be considered to be the full graduation 19 to full class status of the prior Class III A skim milk 20 price. 21 When a new Class I mover needed to be identified, 22 the question arose as to which manufacturing milk class 23 price to use as its basis. The Department determined the 24 mover should be the higher of the most currently 25 calculated Advanced Class III or Class IV skim milk 26 pricing factors. Federal Order Reform identifies at least 27 four reasons for using the higher-of Class III or Class IV 28 as the mover and base value for Class I skim milk prices. 4672 1 First, basing Class I on the higher of III or IV 2 would -- this is a quote from the Federal Order Reform 3 proposed rule -- quote, "more accurately reflect the value 4 of milk in those these different categories of use," in a 5 four-class system. 6 Furthermore, given the separation of manufacturing 7 milk into two classes, using the higher-of Class III and 8 IV would, quote, "assure that shifts in demand for any one 9 manufactured product would not lower...Class I prices," 10 end of that quote. 11 Second, using the higher-of the two classes, 12 quote, "to move Class I prices [will help] to reduce the 13 volatility in milk prices," end of quote. 14 Third, a major consideration was to address class 15 price inversions and depooling. The decision stated, and 16 this is an extended quote: "Class price inversion occurs 17 when a market's regulated price for milk used in 18 manufacturing exceeds the Class I fluid milk price in a 19 given month and causes serious competitive inequities 20 among dairy farmers and regulated handlers... Thus, an 21 inequitable situation has developed where milk for 22 manufacturing is pooled only when associating it with a 23 marketwide pool increases returns. Illustrative of the 24 worsening class price inversion problem are the growing 25 volumes of milk that, while normally associated with 26 Federal Orders, are not being pooled due to price 27 inversion problems ... Since volatility in the 28 manufacturing product markets is expected to continue -- 4673 1 very good call there -- the Class I price mover developed 2 as part of this Federal Milk Order Reform process should 3 address this disorderly marketing situation." 4 And finally, the purpose was to assist Class I 5 handlers in competing for a milk supply. With a quote 6 starting here: "In some markets the use of a simple or 7 even a weighted average of the various manufacturing 8 values may inhibit the ability of Class I handlers to 9 procure milk supplies in competition with those plants 10 that make the higher-valued of the manufactured products. 11 Use of the higher-of the Class III or Class IV price will 12 make it more difficult to draw milk away from Class I uses 13 for manufacturing," end of that quote. 14 And finally, indeed the Department recognized, one 15 final quote here: "The provisions adopted in the [Federal 16 Order Reform] best fulfill the requirements of the 17 Agricultural Marketing Agreement Act," end of quote. 18 Accordingly the Department concluded that the 19 higher-of the most current Class III or Class IV value 20 should be the mover for Class I prices. The pricing for 21 the Class I mover prevailed in all orders of the Federal 22 Order system until the 2018 legislation. 23 Section 1403 of the Agriculture Improvement Act of 24 2018, which was implemented in the 2019 Final Rule, 25 changed the Class I mover to the current language, which 26 uses the average-of Class III and Class IV prices, plus a 27 fixed differential of $0.74 per hundredweight. This 28 legislative change in the mover resulted from a request by 4674 1 Class I handler representatives to change the mover to one 2 that would better allow them to hedge the cost of Class I 3 milk in the dairy product futures markets. 4 NMPF (sic) to this request, subject to the 5 incorporation of the $0.74 per hundredweight fixed 6 differential. This differential represented the average 7 value that the higher-of Class III and IV contributed to 8 the Class I mover, above the average-of Class III and 9 Class IV from 2000 through August 2017. 10 Thus, the intention of both Class I milk buyers 11 and dairy farmer sellers was that the change would be 12 revenue neutral and would accommodate the buyers' desires 13 to better manage their price risk without harming the 14 sellers. The Department reflected this understanding of 15 the amendatory language when promulgating the Final Rule. 16 This is a quote: "The change in the Class I price 17 formula applies uniformly to both large and small 18 businesses. The dairy industry has calculated that 19 applying the higher-of provisions to skim milk prices has 20 returned a price $0.74 per hundredweight above the average 21 of the two factors since the pricing formulas were 22 implemented in 2000. Thus, the inclusion of the $0.74 in 23 the calculation should make the change roughly revenue 24 neutral [emphasis added]. 25 "At that same time, it is anticipated that using 26 the average of the Class III and Class IV advanced pricing 27 factors in the Class I skim milk price formula will allow 28 handlers to better manage volatility in monthly Class I 4675 1 skim milk prices, using Class III and Class IV milk 2 futures and options. Until now, uncertainty about which 3 class price will end up being higher each month has made 4 effective hedging difficult. Amending the Class I skim 5 milk price provisions may help small businesses better 6 utilize currently available risk management tools," end of 7 quote. 8 This was effectively an early recognition by the 9 Department of the growing importance of price risk 10 management and the potential need for the Federal Order 11 price mechanisms to accommodate this. But notably, this 12 statement did not reference nor discredit the four reasons 13 for originally adopting the higher-of mover elucidated in 14 the 1999 Final Decision. Because the 2019 amendment has 15 not functioned as intended, or anticipated by NMPF, has 16 exacerbated disorderly marketing conditions, has not been 17 revenue neutral, and will continue to have deleterious 18 effects on the dairy industry so long as it is in place, 19 the change contained in Proposal 13 is requested. 20 Disorder caused by the average-of plus $0.74 per 21 hundredweight Class I mover: Comparing the higher-of 22 Class I formula in operation from January 2000 to 23 April 2019 to the average-of plus $0.74 per hundredweight 24 Class I formula, in operation since May 2019, reveals a 25 clear asymmetrical impact. The higher-of Class I mover 26 will exceed the average-of Class I mover whenever the 27 Class III and IV advanced skim milk pricing factors differ 28 by more than $1.48 per hundredweight. It does not matter 4676 1 which of the advanced skim pricing factors is higher. The 2 reverse will be true whenever the advanced skim pricing 3 factors differ by less than $1.48 per hundredweight. 4 Thus the maximum amount by which the average-of 5 Class I mover can exceed the higher-of Class I mover is 6 $0.74 per hundredweight, which occurs when the two 7 advanced skim milk pricing factors are equal. However, 8 there is no practical limit by which the average-of 9 Class I mover can fall below the higher-of Class I mover. 10 The asymmetric price risk inherent in the current 11 Class I mover became evident during the second half of 12 2020 and then again during much of 2022. During these 13 periods the current Class I mover fell mostly and 14 significantly below the higher-of mover. 15 NMPF calculates that since it became effective in 16 May 2019, the cumulative market losses in pooled Class I 17 skim milk values and all Federal Orders have reached 18 941.1 million through July 2023. Now that we have August 19 in, that number is 998.3 million through August 2023. 20 NMPF greatly appreciates the Secretary's partial 21 compensation of these losses through the two rounds of 22 Pandemic Market Volatility Assistance program, or PMVAP 23 payments. However, this would not have been needed if the 24 amended Class I mover had performed as expected. 25 More specifically, during the four and one-third 26 years since the current Class I mover has been in place, 27 there have been three episodes when the higher-of mover 28 exceeded the average-of mover, by close to $1 per 4677 1 hundredweight or more. By contrast, as noted, the current 2 average-of mover can never exceed the higher-of mover by 3 more than $0.74 per hundredweight. 4 During the first of these episodes, the six months 5 from July through December 2020, the difference between 6 the two averaged minus $3.56 per hundredweight, generating 7 total losses of pooled Class I skim milk value of 8 $753.2 million, or an average of $125.5 million per month. 9 During the second of these episodes, the four 10 months from August through November 2022, the difference 11 averaged minus $1.47 per hundredweight, generating total 12 losses of pooled Class I skim milk value of 13 $197.8 million, or an average of $49.4 million per month. 14 During the third, and smallest of them, the two 15 months of July and August this year, the difference 16 averaged minus $1.40 per hundredweight and will generate 17 an estimated $58 million of total pooled Class I skim milk 18 values, or an average of $44 million per month. With the 19 August numbers in, that is $89.6 million for the two 20 months, for an average of $44.8 million per month. 21 In contrast to those three monthly loss figures 22 during these three episodes -- to recap, $125.5 million, 23 the second half of 2020, $49.9 million per month in the 24 second half of 2022, and $44.8 million in the last 25 months -- the maximum positive difference of $0.74 per 26 hundredweight would generate a gain in total pooled Class 27 I skim milk values of $25.4 million per month, based on 28 average monthly producer milk volumes during May 2019 4678 1 through July 2023. 2 Figure 1 illustrates the history of cumulative 3 losses of Class I skim milk values from all Federal Order 4 pools during the entire time the average-of mover has been 5 in effect, through this past July. 6 Actually the one in my testimony is through this 7 past July, the one that will hopefully show up on the 8 screen shortly -- maybe I plugged it in wrong -- is 9 through August. 10 It does not include offsets from PMVAP payments 11 nor is it an economic analysis. But it illustrates the 12 pattern that is generated by the increasing volatility of 13 the Federal Order manufacturing class prices. This 14 pattern consists of periods of relative stability during 15 which the average-of -- thank you -- mover generates 16 modest gains over the higher-of mover, followed by periods 17 of volatility, described in the preceding paragraph, that 18 generate losses that more than offset the previous modest 19 gains. The result is mounting cumulative market losses to 20 producers over time. 21 When last month is added to the analysis, the 22 cumulative losses will amount to just about $1 billion 23 dollars, as mentioned earlier. More detailed information 24 relative to this analysis is provided in Exhibit NMPF-30A 25 that should also have been handed out, the-one pager. 26 The change in Class I movers has increased the 27 level of disorderly marketing during this period by 28 reducing Class I prices relative to the other classes and 4679 1 thus creating greater incentives to depool milk. 2 Increased depooling is inconsistent with the Federal Order 3 Reform justification that the Class I mover should reduce 4 the disorderly marketing conditions created by class price 5 inversions and depooling. 6 The enhanced demand for cheese generated in 2020 7 by the Farmers to Families Food Box Program, relative to 8 the demand for butter and nonfat dry milk, widened the 9 spread between Class III and Class IV prices well in 10 excess of the $1.48 break point. This substantially 11 lowered Class I prices compared to where the previous 12 higher-of would have established them, created class price 13 inversions, and generated extensive depooling of Class III 14 milk during the second half of 2020. This was 15 inconsistent with the Federal Order Reform justification 16 that the higher-of mover would “assure that shifts in 17 demand for any one manufactured product will not lower ... 18 Class I prices." That was a quote, that last phrase. 19 Class price inversions recurred in 2022 because of 20 an unusually long period of tight milk supplies. This led 21 to relatively high Class IV skim milk prices, as cheese 22 and whey plants continued to receive relatively adequate 23 milk supplies while butter and nonfat dry milk plants 24 played their traditional balancing roles, producing 25 reduced volumes during periods of tight milk supplies. 26 The result was, again, price volatility and substantial 27 depooling of Class IV skim milk. 28 The third, shorter incident in the summer of 2023 4680 1 resulted when cheese and whey prices fell due to excessive 2 milk supplies relative to domestic and reduced export 3 demand, while butter prices remained robust, pressuring 4 Class III skim milk prices relative to those for Class IV 5 skim milk. 6 Hence, a wide variety of market conditions have 7 proven to be capable, on a seemingly regular basis, of 8 generating market volatility that drives Class III and 9 Class IV skim milk prices sufficiently far apart to drop 10 the current Class I skim milk price mover more than $1 a 11 hundredweight below the higher of the two, while periods 12 of relative market stability are needed to allow the 13 current mover to fall within its strictly limited range of 14 $0.01 to $0.74 per hundredweight above its Federal Order 15 Reform predecessor. 16 In sum, the average-of Class I mover is 17 inconsistent with USDA's Federal Order Reform 18 justifications for the higher-of and does not operate as 19 intended because it builds in an unintended asymmetric 20 risk to producer income, which has resulted in nearly 21 $1 billion of losses in producer income in little more 22 than four years of operation. 23 The current Class I mover dramatically increases 24 the marketing disorder represented by volatile volumes of 25 depooled milk. Market and price volatility continue to be 26 a basic feature of dairy markets and can be anticipated to 27 occur in the future. Little to no data has yet been 28 provided to suggest that the average-of Class I mover has 4681 1 facilitated actual risk management activity with a total 2 value to produce fluid milk processors anywhere near the 3 magnitudes of the quantifiable losses it has dealt to the 4 nation's dairy farmers. 5 The experiment with the average-of Class I mover 6 must therefore be deemed a failure, and the Federal Orders 7 should be amended to return to the higher-of formula. 8 That is the proposed solution. NMPF proposes to amend the 9 Class I skim milk price mover to return it to its original 10 form, as initially adopted in Federal Order Reform; 11 namely, the higher-of the Class III and Class IV Skim Milk 12 Pricing Factors. All of the reasons the Department cited 13 for its original decision, as previously summarized, still 14 apply -- and likely even more so -- to contemporary dairy 15 markets and will doubtless continue to do so going 16 forward. 17 In its lengthy and thorough deliberations and 18 analyses, the group of experts that developed NMPF's 19 package of Federal Order modernization proposals 20 deliberately maintained, and included in its 21 recommendations to NMPF's policy-making bodies, an 22 alternative to returning to the higher-of mover that 23 retained the basic average-of mover mechanism and 24 incorporated a periodic recapture of any lost Class I skim 25 milk pool revenues relative to the higher-of mover. 26 This alternative was unanimously rejected in favor 27 of returning to the higher-of mover. While this 28 alternative, and the similar Proposal 14 submitted by IDFA 4682 1 and Proposal 15 submitted by the Milk Innovation Group, 2 all effectively adopt the higher-of as the standard for 3 generating Class I skim milk price revenue to dairy 4 farmers through Federal Order pools, they all do so in an 5 after-the-fact manner that fails to maintain the maximum 6 monthly separation between the advanced Class I and the 7 manufacturing class prices that generates the best 8 performance for a Class I mover identified by the 9 Department in Federal Order Reform. 10 This testimony provides an overview of NMPF's 11 justification for adoption of Proposal 13. More detailed 12 testimony will follow that supports all, or key provisions 13 of, Proposal 13, including testimony provided by Craig 14 Alexander, representing NMPF member cooperative Upstate 15 Niagara Cooperative, other members of the NMPF task force 16 that developed NMPF's Federal Order modernization 17 proposals, an expert witness from another organization, 18 and producers who are members of NMPF member dairy 19 cooperatives. 20 NMPF sincerely wishes to thank Secretary Vilsack 21 and the Department for holding this important hearing and 22 for thoughtfully considering adoption of its proposed and 23 balanced amendments to the Federal Milk Marketing Order 24 regulations. NMPF has devoted considerable time and 25 resources to thoughtfully considering and recommending the 26 important changes it considers necessary to correct the 27 growing misalignment between the dynamic changes in the 28 U.S. dairy industry since Federal Order Reform and the 4683 1 largely unchanged factors in the critical Federal Order 2 component and class price formulas originally adopted at 3 that time. 4 Together, NMPF is requesting the Secretary to 5 amend certain provisions of 7 CFR 1000.50-52, applicable 6 to all Federal Milk Marketing Orders. The changes to 7 these regulations that Proposal 13 would entail are as 8 follows, as shown on the screen. This is the simplest of 9 all our recommended language changes. And it would simply 10 change the 1000.50(b), the Class I skim milk price. The 11 Class I skim milk price per hundredweight shall be the 12 adjusted Class I differential specified in 1052, strike 13 the next section. As mentioned, all of my recommended 14 Federal Order regulatory language changes incorporate all 15 five of our recommendations. So this -- this next 16 strike-out is pertinent to our support for Proposal 19 on 17 Class I differentials. But then continuing, plus the 18 higher-of, instead of the simple average-of, the 19 advantaged pricing factors computed in paragraph (q)(1) 20 and (2) of this section, rounded to the nearest cent. 21 Q. Thank you, Dr. Vitaliano. 22 I just want to highlight a couple of things in 23 your testimony to make sure that we're clear on what we're 24 doing. 25 On page 4 of your testimony, this -- this is in 26 the middle of the section where you are citing some of the 27 historical changes that had evolved to get to the 28 average-of mover; is that fair? 4684 1 A. Which paragraph are you citing there? 2 Q. I'm just putting this into context that this is 3 the section that we're in; is that right? 4 A. Okay. Yeah. And are you talking about how 5 National Milk and the processor groups came to the 6 agreement to make the change? 7 Q. Well, and just that you are providing the 8 historical context -- 9 A. Yes. 10 Q. -- on page 4 there? 11 A. That's correct. 12 Q. And so when you are in -- on the last half of this 13 page where it starts off and it says, "Federal Order 14 Reform identifies at least four reasons for using the 15 higher-of," starting there in that section, that's what 16 you have cited down below in the footnotes there where you 17 are citing to the actual Federal Register; is that right? 18 A. That is correct. 19 Q. Okay. And so when you are talking about Federal 20 Order Reform, that's -- what you are citing there was 21 first in 1999, the Federal Order, and from that April 2nd, 22 1999, Federal Register that's cited there? 23 A. That's correct. 24 Q. And then you go on to talk about the legislative 25 change as well on the next page. 26 A. Yes. And the legislative change was also 27 footnoted on page 4 for the March 11, 2019, Federal 28 Register notice. And that continues through the first 4685 1 full paragraph on page 5, also stemming from that 1999 2 Federal Order Reform Federal Register notice. 3 Q. Okay. So I want to jump ahead to page 6, and I'm 4 hoping that you can expand a little bit more on the 5 calculation that -- I think sometimes when the economists 6 do math, at least for me, it takes me a couple of times to 7 hear it to make sure I understand what you are saying. 8 But you have done a calculation there that shows the 9 effect of that -- the average-of essentially providing a 10 cap on the -- on the delta that can occur between the 11 prices. Is that fair? 12 A. That's correct. 13 Q. Can you expand on that or maybe even just 14 reiterate it in a more anecdotal way that can really help 15 capture what you are talking about there? 16 A. Certainly. It's not economics, it's arithmetic. 17 Q. Well, it's all the same to me. 18 A. When -- if you take two numbers and take their 19 average, the average between two simple real numbers is 20 going to be halfway between those two, and the higher-of 21 is going to be the higher-of those two. So as -- consider 22 taking two numbers, when they are both the same, the 23 average of them is the same as both of them, which is 24 obviously, therefore, the same as the higher-of the two. 25 As the two numbers separate, no matter which one 26 goes up or down, the further apart they get, let's say you 27 have -- you know, both numbers are 10 and 10. If one of 28 them drops to 8, the average is 18 divided by 2, which is 4686 1 9, which is halfway in between. So the average-of two 2 numbers will always be halfway between the difference 3 between the two numbers. 4 So if the -- in this case, when we take the 5 average-of two numbers and add $0.74 to it, the average-of 6 plus $0.74 will be $0.74 higher than the higher-of the two 7 numbers when they are the same. 8 As the two numbers deviate, once they get to be 9 twice $0.74 apart, $1.48, then the average will be 10 basically the same as the higher-of, Because the 11 average-of will be $0.74 below the higher-of, you add 12 $0.74 back into the average, and it will equate. 13 As the two numbers fall more than $1.48, or twice 14 $0.74 apart, then the average-of plus the $0.74 falls 15 increasingly below the higher-of. Just -- does that -- 16 arithmetic. 17 Q. Arithmetic by an economist. Okay. I appreciate 18 that. 19 And then -- so -- so explain how the cap works 20 then in comparison to the higher-of. 21 A. The cap? You mean the -- 22 Q. How it becomes an effective cap. 23 A. That the higher-of -- arithmetically, the 24 average-of plus $0.74 can never be more than $0.74 above 25 the higher-of. It can fall unlimitedly below. 26 In the case of milk prices, since Class III and 27 Class IV prices are likely to be not zero, maybe, let's -- 28 let's say $10, maybe -- well, actually during the -- 4687 1 during the worst month of the second half of 2020, 2 Class III and Class IV advanced skim milk pricing factors 3 were, as I recall, about $11 and something apart. And the 4 difference between the higher-of mover and the average-of 5 plus $0.74 mover, the average-of plus $0.74 fell, as I 6 recall, $5.16 below the higher-of. 7 Again, contrast that to the average-of could never 8 fall -- could never rise more than $0.74 above. As I 9 explained there, the average-of plus $0.74 is forever 10 confined at the highest level to the range of $0.01 to 11 $0.74 above the higher-of. 12 Q. Okay. So if I am just an average buyer of milk, 13 does that mean that as the buyer, I am -- I can see -- if 14 I'm using the average-of pricing mechanism, as the 15 price -- or the spread between the higher-of and the lower 16 one becomes increasingly larger, I can reap a benefit of 17 that because the average-of is pulling that -- the lower 18 one is pulling the price down? 19 A. There would be a benefit to a buyer of milk 20 compared to having to buy milk at the higher-of price. 21 Q. Okay. But also as the buyer, that cap essentially 22 comes into play because, if I'm looking at the spread -- 23 if I'm capped essentially by that $0.74 above the 24 higher-of, I know that there's a cap that above which I'm 25 not going to have to pay if I'm paying under the 26 average-of pricing? 27 A. That's correct. It's -- it's kind of like a 28 regulatory-induced option for buyers of milk. 4688 1 Q. Wasn't this something that was predictable back 2 when it was established? 3 A. It was to some extent. But, again, the background 4 of it was producer representatives from National Milk were 5 very interested in -- when asked, in accommodating -- to 6 the extent practicable, accommodating the general -- the 7 interest of processor groups in finding something that was 8 more hedgeable. It was an early indication, as I 9 mentioned, of the rise of risk management becoming 10 important in the context of Federal Order regulation and 11 how those two did or did not fit together. 12 During those discussions there was a discussion 13 about what revenue-neutral really meant. There was -- one 14 suggestion was made that since producers were giving up 15 something -- that it was intuitively understood that the 16 higher-of had a certain security factor. We didn't really 17 fully understand that arithmetic that I just went through. 18 You know, the markets gave us a rather soon and rather 19 brutal lesson in arithmetic on that score. 20 But there was a sense that we're -- you know, 21 producers would be -- in accommodating that request from 22 processors for better hedging ability, there was a-- 23 there was a sense that we're kind of giving up something 24 that has a little bit more security to it than the 25 average. It was an intuitive understanding. 26 But the response was, no, this needed to be 27 strictly revenue neutral in the sense that that $0.74 was 28 developed, which was ultimately to look back over the 4689 1 difference between the higher-of and the average-of for 2 the entire period starting with Federal Order Reform in 3 January 2000, up to the time those discussions were taking 4 place, which was in the summer of 2017. 5 Q. Okay. But it's fair to say that at the time it 6 was implemented, it was not only just intended, but that 7 was the goal, was that it would be revenue neutral to the 8 dairy farmer; is that fair? 9 A. That was the understanding, as well as, again, the 10 discussion settled on an average of base mechanism fairly 11 quickly. And in the interest of genuinely trying to 12 facilitate risk management for the folks that, you know, 13 our cooperatives sold milk to, there was a sense that that 14 was -- that was probably, with maybe more trust in the 15 history of price movements, Class III and IV, continuing 16 in the future. In a sense, we -- we kind of sort of 17 disregarded our own sense that the industry is evolving 18 and capable of showing us surprises that we have not seen 19 before. And, again, unfortunately, that was to our -- to 20 our chagrin, actually, was -- was shown to be the case 21 very quickly after that -- that change was made. 22 Q. And so if -- if -- now that we see what the 23 effects are in the proposal by National Milk to go back to 24 the higher-of, the opponents of that, as you understand 25 it, are those who benefit from that price cap; is that 26 fair? 27 A. Yeah. We will see what -- what transpires in the 28 cross-examination. We recognize, because National Milk, 4690 1 its original -- you know, after that -- in fact, before 2 the second half of 2020 was done, while we were still 3 shocked by the change, we did -- National Milk started 4 looking into, as I had mentioned, a modification to the 5 mover that would retain the average-of mechanism and its 6 hedgability, but incorporating a recoup -- basically a 7 revenue recoup mechanism after the fact, similar to that 8 in the -- you know, the two proposals from IDFA and MIG. 9 But we approached IDFA, you know, with that idea and were 10 pretty soundly rebuffed. 11 We continued to go ahead and develop that into a 12 proposal to USDA for an emergency hearing to make the 13 change, but then received word that the Secretary of 14 Agriculture, as I understand it, would prefer to see 15 Congressional funds, through what turned out to be the 16 PMVAP, and did not want to go to a hearing given the 17 differences between the two groups. 18 Q. Okay. And -- and that's what you -- that 19 alternative that you had considered and talked with IDFA 20 about previously is what you believe is -- is similar to 21 what they have proposed in Proposal 14; is that right? 22 A. That's correct. 23 Q. And -- and the downside to dairy farmers in that 24 situation is that while they might ultimately receive the 25 higher-of, it's not going to be contemporaneous with the 26 actual sale of the milk? 27 A. That's correct. It would not meet the criteria 28 that USDA -- the numerous criteria that USDA correctly 4691 1 laid out in Federal Order Reform. 2 Q. And then the buyers in those situations would 3 receive the float or the time value of a delayed payment 4 in return for the dairy farmers not having that 5 contemporaneous payment? 6 A. That was part of it, but it was more that the 7 producers -- even though we -- as I explained, we 8 initially came up with that idea, approached the 9 processors, did not receive much of a welcome on that, 10 proceeded to try to seek it on our own through USDA, were 11 told that, let's try the PMVAP approach. 12 We continued to look at that all the way through 13 our subsequent, much more detailed examination of all of 14 the Federal Order Reform product price formulas and 15 maintained the -- what we call the best possible 16 average-of, all the way up to the time when we presented 17 that and returning to the higher-of to our decision-making 18 body. 19 So we kept that option of the -- you know, the 20 recapture average-of base mover, all the way through up 21 until we came to our -- basically our decision-making 22 bodies chose the final package of reforms. So we gave it 23 a thorough vetting. 24 Q. Before coming up with National Milk's current 25 proposal? 26 A. That's correct. 27 And I must say that the -- the idea of returning 28 to the higher-of is probably one of the most, I'm going to 4692 1 say passionately embraced parts amongst dairy farmers in 2 this country of our package of proposals. We have heard 3 that testimony from several producers already. 4 Q. We have. 5 I want to talk briefly about your Exhibit 230, 6 which is your spreadsheet showing the difference in the 7 Class I. 8 A. Yes. 9 Q. And I don't know if you can pull that up on your 10 screen. 11 A. I don't have it on here, it's rather busy. But I 12 think we handed out the one page to everybody. 13 Q. Yeah. Do you need a copy? 14 A. Yeah. I didn't -- forget to bring one up. But I 15 am pretty familiar with it, so I can probably answer 16 without it. But if you want to ask something specific... 17 Q. Well, you had, during your testimony, mentioned 18 that you have since had the opportunity to look at August 19 numbers, and I wanted to just get those added to this 20 total here. 21 Is it fair to say you don't know the market 22 breakdown for August as you sit here today? 23 A. I have it in my computer here, but I can -- I can 24 produce this with an updated -- I'm getting a little short 25 of space on my -- on putting the spreadsheet on a piece of 26 paper. But I can produce an amended version of this 27 exhibit -- 28 Q. Okay. 4693 1 A. -- that would have August. 2 And what it does shows is that in the -- the 3 second page -- page 2 of 2, on the bottom right-hand 4 corner where it says minus 941.1, that will say minus 5 998-point -- whatever it was in my testimony. Basically, 6 for all practical purposes, you know, right around a 7 billion dollars. 8 Q. When you say -- did you add that on the fly when 9 you were on page 6 of your testimony? 10 A. I got that from the spreadsheet and put it in so 11 that the -- because I had -- I needed to submit this 12 testimony in advance before I had those numbers from the 13 Market Administrator's websites. 14 Q. Okay. So if we don't end up adding in a new 15 exhibit, if we just modified Exhibit 230, it would be to 16 change that total from 941 million to 998 million? 17 A. .3, yes. 18 Q. 998.3. Okay. Thank you. 19 MS. HANCOCK: That's all I have. 20 Your Honor, he is available for cross-examination. 21 THE COURT: Questions for this witness? 22 CROSS-EXAMINATION 23 BY MS. VULIN: 24 Q. Good afternoon, Dr. Vitaliano. How are you? 25 A. Good afternoon. 26 Q. Good. My name is Ashley Vulin. I am an attorney 27 with the Milk Innovation Group. 28 I'd like to start on page 2 of your testimony, 4694 1 please. 2 A. This is the part that has been read into the 3 record previously, but not read into it this time. 4 Q. And that's -- that's fine. No problem. 5 About two-thirds of the way down you say, 6 "Additionally, the United States currently sells about 7 18% of its milk production as manufactured products in 8 export markets compared to about 5% in 2000." 9 Do you see that? 10 A. Yes. 11 Q. And that's correct? 12 A. Yes. I have -- for years and years, I have been 13 calculating the percentage of U.S. milk solids that are 14 exported and that are imported, mostly for the purpose of 15 getting the industry, particularly the dairy producer side 16 of the industry that has historically been more focused on 17 imports, to realize that that's yesterday's problem, and 18 expanding exports is the futures problem -- or challenge. 19 Q. Opportunity maybe? 20 A. And all of that, yes. 21 Q. You agree exports are fairly significant now of 22 dairy products? 23 A. That is correct. 24 Q. And what products are the most commonly exported 25 dairy products today? 26 A. The easiest way is to say that about -- and very 27 consistently throughout this entire period -- about 80% of 28 the milk solids that we export consist of what I call skim 4695 1 milk ingredient product. That's milk -- that's skim milk 2 powder, nonfat dry milk, dry whey, whey protein 3 concentrate, whey protein isolates, lactose, the things 4 that basically don't have very much butterfat in them, and 5 they are ingredients that are used to reconstitute dairy 6 products in usually state-of-the-art plants in foreign 7 countries. 8 Q. And for farmer milk to be available for this 9 export market, it has to be Grade A; is that right? 10 A. I'm not sure. I -- I -- I'm not sure about that. 11 In fact, I -- I'm unable to answer it because I just don't 12 know. 13 Q. And is -- 14 A. Most of the milk produced in this country is 15 Grade A, but I would assume that there are Grade B 16 products -- or Grade B milk could be used in products that 17 are exported. I'm not sure. A lot of other countries 18 don't demand that -- you know, that our products are PMO 19 compliant. But I don't know that. You would have to ask 20 some experts, say, in the U.S. Dairy Export Council. 21 Q. And what class of milk products are the largest 22 portion of exports? 23 A. I would assume that it's Class IV. We export 24 about 7% of our cheese production, which would, you know, 25 be Class III. But my guess is if -- I have never seen 26 a -- sort of an export Federal Order pool. It would be 27 interesting to see one of those to see what the 28 utilization is, but it would be mostly Class IV. 4696 1 Q. And who are the U.S.'s biggest trading partners on 2 the export market? 3 A. Generally, and I can't give you an exact numerical 4 order, but it would be Mexico, China, Japan, South Korea. 5 Q. And who are the U.S.'s biggest competitors on the 6 export market? 7 A. The European Union, now 27 countries, that 8 basically export kind of as a block in terms of their 9 export policies, and New Zealand, sort of secondarily 10 Australia, maybe Argentina. But it's basically the EU and 11 New Zealand. They are numbers one and two; we are number 12 three in volume. 13 Q. And you'd agree that the export market is a 14 meaningfully larger portion of milk production sales now 15 than it was in 2000? 16 A. Yes. My calculations are the volume of milk 17 solids, although a different mix of fat and skim, the 18 volume of milk solids that goes to the export market, 19 particularly in the last two years when we hit that high 20 of 18%, it's down a bit this year with -- but the volume 21 of -- during those two years, I -- my -- I estimate we 22 exported more milk solids than we consumed milk solids 23 domestically in fluid milk products. 24 Q. Well, thank you for indulging my interests in the 25 export market. 26 A. Certainly. 27 Q. I'd like to talk a little bit about your thoughts 28 on just the theory behind pricing Class I in FMMOs. 4697 1 So is it NMPF's position that Class I should 2 always be priced off of both Class III and IV? 3 A. We're kind of taking our page from USDA. As I 4 have testified many times, we support the Federal Order 5 program. We support the product price formulas, but we 6 think that they needed to be updated, because so many of 7 them were kind of fixed in place with fixed numbers, you 8 know, fixed skim milk component composition factors, fixed 9 Class I differentials, the assumption that barrel cheese 10 was, you know, block cheese in sort of a barrel's 11 clothing, so to speak. Those things needed to be changed 12 because the industry is changing, and in order for the 13 Federal Order system to continue to be effective and 14 functional, which we really want it to be, that those 15 things need to change. 16 With respect to the Class I mover, simply because 17 it was so flexible, unlike so many of these other things 18 that had a fixed differential -- fixed coefficients in 19 them, we think USDA and to use the colloquial -- USDA got 20 it right the first time in terms of the Class I mover, and 21 all of those factors that I quoted from the Federal Order 22 decision on page -- basically page 6 and spilling over 23 to -- no, on page 4 going over to page 5 -- all of those 24 things that I quoted I think still apply, and in some 25 cases even more so. 26 So that's one area where we felt USDA got it right 27 the first time. Let's go -- and we deviated from that 28 with the best of intentions. It turned out, 4698 1 unfortunately, not to work out very well for dairy 2 farmers, and therefore, we're saying, let's go back to 3 what -- you know, what we -- what was right in the first 4 place. That's our position on that one. 5 Q. Okay. I want to make sure I tracked the answer to 6 my question in that. 7 So it is NMPF's position that, yes, Class I should 8 always be priced off of both Class III and IV? 9 A. Yes. Yes. 10 Q. And is the only reason NMPF believes that, because 11 it was USDA's prior policy, or what are the reasons why 12 NMPF today believes that Class I should be priced off the 13 both Class III and IV? 14 A. Well, actually, our belief that Class I should be 15 priced off of manufacturing classes goes way beyond -- 16 it's much earlier than Federal Order Reform, back when we 17 just had the MW. The -- it's our understanding that a 18 basic feature of the Federal Order formula is that a 19 certain distance needs to be maintained between Class I 20 prices above the manufacturing prices, especially with 21 Advanced Class I prices that processors definitely like, 22 and we support. That gives rise to the potential for 23 price inversions, and even more so with Class I 24 utilizations dropping. 25 Q. So why does there need to be a certain distance 26 between the Class I price and the manufacturing classes? 27 A. Basically for all the reasons that USDA specified 28 in -- in Federal Order Reform. I basically couldn't have 4699 1 said it better myself. That, you know, to prevent class 2 price inversions, to prevent changes in demand for one -- 3 you know, one type of dairy product that would be Class -- 4 priced off of say Class IV, from being impacted by drops 5 in, you know, supply/demand conditions for Class IV. 6 Again, on page 4, with the series of quotes that I 7 sort of put together with -- over several pages of the 8 decision, with some elisions and insertions, I thought 9 that was probably as good a summary of the reasons that 10 USDA put forth back in 1999 for choosing the higher-of 11 Class I mover, and we basically believe that those still 12 apply. 13 Q. And to the extent that market conditions change, 14 do you think those reasons should change? 15 A. The market conditions that are changing seem to -- 16 as far as they affect the Class I mover, seems to be 17 increasing volatility. And -- and because addressing 18 Class I volatility and its -- its manifestation in 19 depooling, we think the way the market is changed, unlike 20 in some of the other provisions in our proposals where it 21 requires a change in the Federal Order pricing provisions 22 to keep them updated, we think that the changes in market 23 conditions are actually making it more pertinent to -- for 24 the mover to be the higher-of Class I and Class II and 25 Class III and Class IV. 26 Q. So it's your position that the Class I minimum 27 price always has to be set above both manufacturing class 28 prices? 4700 1 A. To the extent possible. That often -- with 2 advanced pricing, Class I pricing, that does not always 3 happen. 4 Q. And so -- 5 A. But -- 6 Q. I'm sorry. Go ahead. 7 A. Yeah. But not necessarily always above each one, 8 but above the highest possible -- the highest possible 9 one. 10 Q. And what about the frequency, does it need to be 11 above each one monthly, biannually, on average yearly? 12 A. Ideally it would be on a monthly basis that you 13 would have the -- we recognize that the average-of plus 14 $0.74, or even an adjusted differential above the 15 average-of, would occasionally be, with some frequency, 16 above the higher-of. But nonetheless, the higher-of is 17 one that dairy farmers are most comfortable with, most 18 trusting of, and we think that would be -- you know, we 19 would be content to return to the higher-of as opposed to 20 something that was more complicated. 21 Q. And to the extent that setting the Class I minimum 22 price, would you -- would you agree it's supposed to 23 reflect the value of milk used for Class I purposes? 24 A. Can you repeat that question again? 25 Q. Is the Class I minimum price supposed to reflect 26 the value of milk used for Class I purposes? 27 A. That's the -- that's the -- well, it's supposed 28 to -- the Class I price is supposed to be set at a level 4701 1 that is sufficient to assure Class I processors of an 2 adequate supply of milk for fluid milk purposes. 3 Q. And from a consumer perspective, if the price of 4 cheese goes up, is the consumer going to find themselves 5 more willing to pay higher prices for their fluid milk? 6 A. Well, we had an extensive discussion on the 7 elasticity of demand for fluid milk earlier in this 8 hearing. Without -- without extensive volatility, 9 generally having a Class I price that is higher than the 10 manufacturing prices is pretty much a -- an important 11 feature of the operation of the entire Federal Order 12 program. 13 Q. And I'm not asking about the elasticity. I'm 14 saying in your experience, does a consumer find more value 15 in milk, fluid milk, when their cheese is more expensive? 16 A. Well, you are asking a question that could have 17 been asked at any time in the history of the Federal Order 18 program. And the Federal Order program has priced 19 Class I -- or attempted to price Class I above cheese 20 prices, to use your example, and the system seems to have 21 functioned for decades and decades under that principle. 22 Q. Okay. And I'm asking the question though now to 23 you, do you believe that's true today? 24 A. Well, if that is a -- if that is an outcome of the 25 way the Federal Order class pricing works, then that is 26 something that -- you know, that basically is a factor of 27 the system that we have lived with for a long time. 28 We do not discount or dismiss the importance of a 4702 1 Class I mover that is more hedgeable than, you know, if -- 2 if that is possible, without being too extensively at the 3 expense of all the other features of a Class I mover, 4 which is what we say has basically happened at this point. 5 We are not dismissing -- we would like to see some 6 data from those who wish to -- who claim they are hedging 7 Class I with the average-of to see what the extent of that 8 is, what a rough economic value of that ability is. We 9 have advanced, you know, our numbers about what we have -- 10 we feel we have suffered in terms of losses so far under 11 the change. 12 But, yes. Your question, I'm not -- I'm aware 13 that it's -- I guess where I come down on that is, if that 14 is a feature, if the price changes and the price of cheese 15 affect the price of Class I, that is basically something 16 that is a factor in the operation of the Federal Order 17 program, and it has been since the beginning of the 18 program. 19 Q. So I'd like to ask a little bit more about this 20 issue of disorderly marketing. So you say disorder was 21 caused by the average-of because it resulted in a 22 reduction in revenue to producers, correct? 23 A. It was not because it -- well, that was part of 24 it. But the disorderly marketing that has happened 25 basically, in a sense, violated many of the features that 26 USDA clearly spelled out in Federal Order Reform. 27 Q. Are you saying that participants in the pool did 28 not comply with the regulations? 4703 1 A. No. Let me read you this quote. "In some 2 markets, the use of simple or even weighted average of the 3 various manufacturing values may inhibit the ability of 4 Class I handlers to procure milk supplies in competition 5 with plants that make the higher value to the manufactured 6 product." 7 Earlier -- let me see. "Given" -- this is another 8 quote -- "given the separation of manufacturing milk into 9 two classes using the higher-of Class III and IV would," 10 quote, "assure that shifts in demand for any one 11 manufactured product will not lower Class I prices." 12 So this would be an incident going back to your 13 cheese example, where shifts in Class IV demand, product 14 demand, would not necessarily affect the Class III price 15 and would not transmit the -- well, no, let me do the 16 opposite. 17 Shifts in the price of cheese reflected in the 18 price of Class III would not be transmitted to fluid milk 19 processors if Class IV price was the higher-of and was 20 setting the Class I price and was stable. Whereas cheese, 21 to use your example, may have changed price. 22 Q. But what you are saying then is the Class I price 23 wasn't high enough, correct? That that pricing signal 24 didn't transmit to a higher Class I minimum price? 25 A. When there's a separation of the Class III and 26 Class IV prices, that can bring the higher -- the 27 average-of down, you know, almost penny for penny, penny 28 for -- half penny for penny, whereas the higher-of has a 4704 1 certain amount of stability, which was reflected in USDA's 2 statement that using the higher-of the two classes to move 3 Class I prices will help to reduce the volatility in milk 4 prices. 5 Q. Understood. 6 But, again, the results of that was lower producer 7 revenue, correct? 8 A. That was -- that was, yes. That was the case. 9 Q. And the results were not that fluid plants could 10 not acquire their supplies, correct? 11 A. As far as I know. 12 Q. So the disorderly marketing that you have 13 identified, it is not that fluid plants were unable to get 14 a sufficient amount of milk for their needs, correct? It 15 was that producers got paid less money? 16 A. Producers lost money, but there was depooling 17 and -- 18 Q. The system is designed to -- 19 MS. HANCOCK: Let him answer the question. 20 THE WITNESS: Yeah -- 21 MS. VULIN: I am having a little bit of trouble, 22 though, getting direct answers to my questions with pretty 23 long answers unrelated. 24 THE WITNESS: Go ahead. 25 MS. VULIN: So I understand, but I would ask that 26 the witness try very hard to answer the question posed. 27 MS. HANCOCK: I think you're entitled to follow 28 up, but you got to let him finish his answer. 4705 1 THE COURT: Yeah, I appreciate that. 2 THE WITNESS: I'll try to keep it short. Go ahead 3 and ask your question again. 4 BY MS. VULIN: 5 Q. So the issue that arose under the average-of was 6 not that there were insufficient supplies of milk for 7 fluid use, it was that producers made less money than they 8 would have made under a different calculation, correct? 9 A. There was an issue with producers receiving less 10 money, that they clearly identified as disorderly 11 marketing. I cannot speak to whether or not fluid milk 12 processors did not have a problem with attracting adequate 13 supplies. 14 We will have experts testifying on that with 15 Proposal -- in connection with Proposal 19. 16 Q. And then you identify depooling, correct? 17 A. Yes. 18 Q. Is depooling disorderly marketing? 19 A. In USDA's eyes, and ours, it is, yes. And we 20 agree with USDA's characterization of depooling and price 21 inversions as disorderly. 22 Q. So anytime a Class II, III, or IV handler chooses 23 not to pool their milk, that is disorderly marketing? 24 A. It is, yes. We do not dispute the right of 25 Class II, III, and IV handlers to depool when it's in 26 their interest. But we recognize that a system in which 27 the price relationships discourage depooling would be 28 preferable to one where depooling was -- was common. 4706 1 Q. You would agree with me that Federal Milk 2 Marketing Orders can only require participation by Class I 3 fluid milk handlers, correct? 4 A. That is correct, yes. 5 Q. So the decision by a manufacturing Class I handler 6 to not pool is an inherent design of the system, correct? 7 A. It is an outcome of the system, yes. 8 Q. So it is your position, and NMPF's position, that 9 the FMMO system, as designed, inherently causes disorderly 10 marketing? 11 A. I cannot speak for National Milk on that issue. 12 Q. For yourself? 13 A. Can you repeat the question again? 14 Q. If the FMMO system is designed to allow 15 manufacturing class processors to not pool, and you 16 believe that any decision by a manufacturing class 17 processor to not pool is disorderly marketing, then it's 18 your position that the FMMO system, as designed, causes 19 disorderly marketing? 20 A. I wouldn't state quite so boldly. 21 Q. How would you state it? 22 A. I would state it that the Federal Order system is 23 probably correct in not requiring Class II, III, and IV 24 processors to be pooled. There are orders, such as 25 Order 1, where the depooling requirements are pretty 26 strict, and that happens, and that is fine. If those -- 27 you know, if an individual Federal Order wishes to adopt 28 those provisions, that's fine for them. That's their 4707 1 right. Most orders do not have strict, you know, 2 depooling, anti-depooling regulations, and as a result, 3 processors other than Class I can make those decisions. 4 And that's, you know, that's an inherent part of 5 the system. It would be good to the extent that price 6 formulas could be arranged to discourage the conditions, 7 minimize the conditions which permit depooling, which 8 would basically mean increasing the difference between 9 Class I prices and the manufacturing class prices so that 10 depooling would occur less frequently, that would be good. 11 But we're not prepared to judge the system as deliberately 12 encouraging depooling. That's -- that's not a 13 characterization I would use. 14 Q. So you said there are instances in which it is 15 rational for a manufacturing class handler to choose not 16 to pool, correct? 17 A. Of course. Yes. 18 Q. What are those? 19 A. That is when basically the processor would receive 20 more income from the market by depooling than they would 21 from the market with, you know, basically paying into the 22 pool. 23 Q. And would you agree with me that's an important 24 signal from the market that the industry should respond 25 to? 26 A. Could you define an important signal that the 27 industry should respond to? 28 Q. If the marketplace is telling processors that that 4708 1 milk is more valuable if manufactured and sold outside of 2 the FMMO system, shouldn't processors respond to that? 3 A. Generally, it is in the interest of processors, 4 whether they are cooperative or not, to maximize their 5 income. That's -- that's the kind of system that 6 economists would -- would designate as a healthy industry. 7 Q. And are you familiar with what a 9(c) handler is? 8 A. I've heard the term, but I would need a 9 definition. 10 Q. A cooperative that is defined as a handler. Does 11 that sound familiar? 12 A. Yes, cooperatives are designed -- are designated 13 as handlers under the system. 14 Q. Are you aware if any cooperatives have chosen not 15 to pool their milk since the average-of was in place? 16 A. Yes. 17 Q. And so your position is that those cooperatives 18 contributed to disorderly marketing by choosing not to 19 pool their milk? 20 A. They responded to market signals in the interest 21 of their members, and they caused consternation in many 22 cases to other cooperatives that did not have that 23 opportunity because of where they sold their milk. 24 Q. And if we aim to peg the Class I minimum price 25 always above manufactured prices, aren't we disrupting 26 those external market signals that are telling the 27 marketplace that milk is more valuable in a manufacturing 28 plant outside of the system? 4709 1 A. Can you state again the conditions you predicated 2 your statement on? 3 Q. So in an instance, in the circumstances in which 4 you agree it would be rational for a manufacturing class 5 processor to not pool their milk, where it is more 6 valuable being sold outside the FMMO system, do you -- do 7 you recall that premise? 8 A. Yes. 9 Q. So in that situation, if we are pegging the 10 Class I price above that manufacturing class price, aren't 11 we distorting the signals that the marketplace needs to 12 follow to know what the highest and best use of that milk 13 is? 14 A. No. 15 Q. Why not? 16 A. Because the system in which the class prices would 17 be set at appropriate levels, so that depooling was not a 18 rational decision, that would not cause a problem. 19 Processors, both proprietary and cooperatives, that pool 20 their milk in Federal Orders and find themselves 21 confronted by the circumstances in which it is rational 22 for them to depool, and pooling regulations of their 23 particular order do not prohibit it or discourage it, as 24 for example in Order 1, it is a rational decision for them 25 to depool. It's not their right to depool, in a sense, it 26 is their -- it is in their best interest to depool. 27 If the system and the price formulas and the 28 Federal Order pooling regulations were adopted to make it 4710 1 difficult, more difficult let's say, to depool, that's 2 fine, too. I'm not -- I'm not -- I'm not specifying an 3 ideal Federal Order where depooling is a right or a 4 necessity. I think it is disorderly if there is 5 increasing depooling, even though it is a rational 6 decision by the depooling actors. 7 Q. So it is disorderly for a market to have a -- for 8 milk -- I'll start that again. Strike that. 9 You believe it is disorderly for the marketplace 10 to value milk at a higher value for a manufacturing use 11 than a Class I use? 12 A. It is disorderly if that happens on a regular 13 basis, yes. And we have several dairy farmers who have 14 testified and will testify that they consider depooling to 15 be disruptive and disorderly. 16 MS. VULIN: I still have a number of questions, 17 and I believe it might be time for a break. 18 THE COURT: Yes, I think so. Let's come back at 19 4:18. 20 (Whereupon, a break was taken.) 21 THE COURT: Let's reconvene. On the record. 22 Ms. Vulin, your witness. 23 MS. VULIN: Thank you, your Honor. 24 BY MS. VULIN: 25 Q. So you had talked a little bit earlier about 26 this -- this disparity, the spread between III and IV that 27 can develop, correct? 28 A. That's -- yes. 4711 1 Q. And you describe that as price volatility; is that 2 right? 3 A. The price volatility would be the forces that 4 cause those two to diverge in unpredictable and sometimes 5 in -- in rapid and extreme fashion, that's -- that creates 6 the volatility. 7 Q. I think you described it as market volatility; is 8 that right? 9 A. Yes. 10 Q. And so what are the market conditions that drive 11 Class III and IV apart? 12 A. Basically, the supply and demand for Class III and 13 Class IV products, and how they intersect and how they may 14 differ. 15 Q. So does USDA regulate the supply and demand for 16 Class III and IV products? 17 A. No. 18 Q. Okay. So USDA can't control the conditions that 19 could lead to the spread between III and IV? 20 A. No. 21 Q. But you agree that it's -- or you maintain that 22 it's disorderly for Class III and IV to have significant 23 price diversion from each other? 24 A. That contributes to disorderly marketing, yes. 25 Q. That is disorderly marketing or contributes to 26 disorderly marketing? 27 A. Well, depends on how you define disorderly 28 marketing. But, okay, let's say it is disorderly 4712 1 marketing. 2 Q. Is that how you -- 3 A. I would defer to USDA for the definition of 4 disorderly marketing because that's -- that's their key 5 term. 6 Q. What -- how do you believe USDA defines disorderly 7 marketing? 8 A. I don't know. But -- 9 Q. And you would agree that raising the Class I price 10 doesn't change the supply and demand market conditions 11 that lead to the spread between III and IV, correct? 12 A. Are you asking if Class I demand does not? 13 Q. Does raising the Class I minimum price change the 14 supply and demand factors or market conditions that drive 15 III and IV apart? 16 A. Generally. No. 17 Q. Do you believe that USDA should interfere with the 18 supply and demand market conditions that drive III and IV 19 apart? 20 A. I don't think USDA has the authority to do so. 21 Q. So this disparity between III and IV, that was the 22 source of the losses that you have described in 2020 that 23 producers experienced, correct? 24 A. Correct. 25 Q. And if we could go, I believe it's page 6 of your 26 testimony where you say that those losses as of July were 27 $941.1 million, correct? 28 A. Correct. 4713 1 Q. And how did you calculate that amount? 2 A. That's laid out in Exhibit NMPF-30A, which was 3 Exhibit -- was it 230? Those are the detail calculations 4 behind that. If you want me to go through that. 5 Q. No. Nope. Just wanted to know the source. Thank 6 you. 7 A. That's the -- that's the source. 8 Q. So this is $941 million that you believe producers 9 should have received under the FMMOs? 10 A. It is an arithmetic calculation, not an economic 11 analysis. 12 Q. You believe that -- are you saying that producers 13 would have had $941 million more dollars in their pockets 14 had the formula not been changed? 15 A. As an economist, if I were doing an economic 16 analysis of particularly the losses in the second half of 17 2020, I would have factored in a supply response where 18 producers may have produced less milk and a lower price 19 may have encouraged additional Class I consumption. 20 Dr. Scott Brown will be here sometime in the near 21 future to present an economic analysis of our proposals. 22 But this is, again, just an arithmetic. And we have seen 23 a lot of arithmetic calculations, which consists of simply 24 plugging the formula changes without taking into account 25 any economic adjustments in supply and demand. 26 Q. But when I look at your Figure 4 on page 7, it is 27 entitled cumulative producer losses due to change in 28 Class I movers. So you are saying that is inaccurate 4714 1 those were not producer losses? 2 A. Those are arithmetic producer losses as I 3 identified in the -- in the text. 4 Q. And so then my question is, if that $941 million 5 did not go to farmers, where did it go? 6 A. This was the -- basically the calculated 7 arithmetic losses in this sheet, which is simply that 8 chart graphs the far right column on page 2. That's those 9 numbers. 10 Q. And if those are losses to farmers, then they 11 would have been gains somewhere else, correct? 12 A. The implication is those are gains that accrued to 13 Class I processors, in the sense that these are arithmetic 14 calculations of monthly losses, they would have been -- 15 that because of the nature of the Class I mover and the 16 change, those would have been, by your characterization, 17 gains to the Class I processors, savings to them. 18 Q. And had the formula not been changed, Class I 19 processors would have owed $941 million more to the pool 20 than they would have otherwise? 21 A. In that same arithmetic analysis -- analytical 22 framework, yes. 23 Q. And where would Class I processors have gotten 24 nearly a billion dollars to put into the pool during this 25 time period? 26 A. They would have gotten that from -- basically from 27 the marketplace. They -- they would have had -- had the 28 Class I mover not been changed and continued to be the 4715 1 higher-of throughout this whole period, that revenue would 2 have been forthcoming from the market and would have been 3 paid into pools, just like it had been from January 2000 4 though up through April 2029 -- 2019. 5 Q. So the billion dollars would have come from 6 consumers, ultimately? 7 A. Yes. 8 Q. And I would like to talk about the pandemic market 9 volatility assistance program, the PMVAP, correct? And 10 I'll probably get that acronym mixed up at some point, so 11 please do correct me. 12 You said on page 6 that -- it's at the second full 13 paragraph up from the bottom, the second to the last 14 sentence you said, "NMPF greatly appreciates the 15 Secretary's partial compensation of these losses through 16 the two rounds of the Pandemic Market Volatility 17 Assistance Program." 18 Do you see that? 19 A. Yes. 20 Q. And so were these PMVAP payments meant to 21 compensate or reimburse farmers for their lost revenue due 22 to the base Class I skim price using the average-of 23 formula instead of the higher-of? 24 A. Partly compensate. 25 Q. But the purpose was to compensate farmers because 26 of the change in the base Class I skim formula? 27 A. That's my understanding. 28 Q. I -- I had thought those payments were to assist 4716 1 farmers because of the impacts of the pandemic, not 2 because of the change in the formula. 3 A. They were due to the pandemic market volatility, 4 in other words, the volatility created by the pandemic 5 which was, in terms of the severe underperformance of the 6 current mover compared to the higher-of, was due to a 7 great extent to the extensive Farmers to Families Food 8 Box -- as I recall, Farmers to Family Food Box Program 9 that -- that basically purchased very large quantities of 10 cheese, compared to Class IV products, and created that -- 11 that -- that severe separation of Class III and Class IV 12 prices. 13 MS. VULIN: Your Honor, I would like to introduce 14 an exhibit, and I have copies for everyone here. We'll 15 get them passed out. 16 May I approach the witness with a copy, your 17 Honor? 18 THE COURT: Yes, you may. 19 MS. VULIN: And do I give you your copy or does 20 that come from the official copy? 21 THE COURT: Either way. I think it comes from 22 sometimes one way, sometimes the other. Make sure counsel 23 for this witness gets one, though. 24 MS. VULIN: And the official USDA copy is 25 single-sided, but everyone else has a double-sided copy. 26 This is a document with the USDA logo at the top, 27 entitled Pandemic Market Volatility Assistance Program for 28 Dairy, and at the bottom has the date of November 2021. 4717 1 And I'd ask that it be marked Exhibit 230, if I have 2 counted correctly. 3 THE COURT: No, I have it 231. 4 MS. VULIN: 231. 5 THE COURT: Yes. So this exhibit will be marked. 6 231 for identification. 7 (Thereafter, Exhibit Number 231 was marked 8 for identification.) 9 BY MS. VULIN: 10 Q. Dr. Vitaliano, do you recognize this document? 11 A. Yes. I have seen it. 12 Q. What is it? 13 A. It is a fact sheet the USDA gave out by -- on the 14 PMVAP program. 15 Q. And where on this fact sheet does it identify that 16 the PMVAP is in response to the base Class I skim mover 17 price? 18 A. Well, take a look at the second bullet: "USDA 19 payments" -- "the Pandemic Market Volatility Assistance 20 Program provides" -- the second bullet point -- "USDA 21 payments to dairy farmers through their handlers and 22 cooperatives based on fluid milk sales from July through 23 December 2020." 24 Based on fluid milk sales. Not on milk sales. 25 Fluid milk sales. 26 Q. And how does that specify that it's based on the 27 base Class I skim formula? 28 A. You would have to ask USDA which administered the 4718 1 program. 2 Q. And was this assistance provided only to producers 3 with pooled milk? 4 A. That's my understanding. Yes. 5 Q. Do you know how USDA determined the amount of milk 6 eligible for the program? 7 A. I remember the limits they placed on it, but I 8 think it was based on producer shipments of milk that were 9 pooled on Federal Orders. 10 Q. You said the limits. What were those? 11 A. There were limits on 5 billion -- 5 million pounds 12 for the six-month period, and no payments were made of 13 that. I believe that did not pay 100% of the amount below 14 5 million pounds. And there were some adjusted gross 15 income limits of, as I recall, of $900,000. That's pretty 16 standard for a lot of assistance programs, direct payments 17 programs. 18 Q. So you said the six months. Are you talking about 19 July to December 2020? 20 A. That's correct. 21 Q. So the program was meant to provide assistance to 22 farmers based on fluid milk sales for that six-month 23 period, correct? 24 A. That's correct. 25 Q. But the average-of base Class I skim formula was 26 in place for multiple years; isn't that right? 27 A. At that time it was in place for a little over one 28 year. 4719 1 Q. By November 2021. 2 A. Where do you see 2021 on here? 3 Q. In the green bar at the bottom. 4 A. That was the date of -- that this fact sheet was 5 published. 6 Q. So the -- the average-of would have been in place 7 for two years, correct? Approximately? 8 A. A year and a half. Yeah. Let's call it two 9 years. 10 Q. Okay. 11 A. No, two and a half years. 12 Q. So USDA did not provide compensation to farmers 13 for the entire period the average-of was in place, 14 correct? 15 A. No. They only -- they only made compensation for 16 the July to December 2020 period when the two movers 17 seriously -- when the current mover seriously 18 underperformed the previous higher-of mover. 19 And I'm not aware, I'd have to check my records, 20 whether or not milk prices overall were particularly lower 21 during that period compared to -- I know in 2021 milk 22 prices were significantly low. 23 And so my -- my guess is that the July through 24 December period was distinguished as a hardship period for 25 dairy farmers, particularly because of the 26 underperformance of the recently implemented Class I mover 27 compared to its predecessor. 28 Q. And what caused the mover to underperform in your 4720 1 opinion? 2 A. The -- 3 Q. Actually, I have a better question. Let me 4 rephrase that if I may. 5 What were market forces that drove Class III and 6 IV apart during this six-month period? 7 A. They were, to a great extent, pandemic related, 8 through the Family -- Families -- Farmers to Families Food 9 Box Program and its extensive purchases of cheese. 10 Q. So the base Class I skim formula in place did not 11 cause disorderly marketing, it's that it didn't work as 12 intended as the result of unusual market circumstances? 13 A. Yeah. Right. The change in the movers resulted 14 in the loss given the -- basically USDA's other pandemic 15 activities. 16 Q. And you said that the PMVAP only partially 17 reimbursed producers for their losses under the average-of 18 formula, correct? 19 A. That's correct. 20 Q. What portion of the their $941 million losses did 21 the PMVAP reimburse? 22 A. In the aggregate, I believe that the -- the number 23 here is roughly 360 million. In the context of the 24 calculations I show in Exhibit NMPF-30A is about 25 1 billion, arithmetically calculated. That would make the 26 PMVAP paying in the aggregate a little over a third, 27 compensating a little over a third of the losses for 28 individual producers, particularly large producers subject 4721 1 to the payment limitations. It was a much smaller 2 proportion of their actual losses. 3 Q. There was a second round of PVMA- -- 4 A. Yeah, I don't know. The numbers -- the round 5 numbers I recall is 300 for the first go round and then 6 another hundred. I don't know whether the 360 -- you may 7 know the timing of this. I think the second round, was it 8 after November 2021 or before? 9 Q. We're up here to hear your testimony on that. 10 A. Yeah. No, I don't know whether that second round 11 is in these numbers. 12 Q. Okay. And the cumulative losses that you 13 detailed, you said you explicitly excluded the PMVAP 14 payments, correct? 15 A. That's correct. 16 Q. And you talked about a cap on those. 17 So the cap means that small producers would have 18 likely benefitted to a greater degree than large producers 19 under the PMVAP payments, correct? 20 A. Yes. 21 Q. And looking again at your Figure 1 on page 7, 22 these are cumulative producer losses -- 23 A. That's correct. 24 Q. -- because of the change, correct? 25 A. That's correct. 26 Q. So despite this chart detailing, you know, very 27 negative numbers after May of 2020, the base Class I skim 28 price formula did not always generate less revenue for 4722 1 producers than the higher-of would have, correct? 2 A. No. If you look at the right-hand column on 3 page 2 of Exhibit NMPF-30A, you will see some of those 4 numbers are positive and some of them are negative. 5 If you look at the July through December 2020 6 period, you will see very large negative numbers. You 7 will never see a number in here larger than, roughly, 8 25 million. 9 So, for example, if you look at October '21 where 10 the current mover was $0.73 above the higher-of, that's 11 one penny short of its maximum possible, that generated 12 $25.5 million. That's kind of the maximum -- you know, 13 the volumes of Class I skim milk change monthly in here. 14 But basically you will not see a number in that right-hand 15 column that's much over 25 million. You will see quite a 16 few negative numbers that are quite a bit lower than a 17 negative 25 million. 18 Q. And there are multiple months, it looks like even 19 the first six, that the average-of was in place that it 20 generated higher producer revenue than the higher-of would 21 have, correct? 22 A. Right. The first year after it was in place, I 23 was thinking, hey, we made a good deal here. That was 24 short lived. 25 Q. Did NMPF think it was disorderly when the formula 26 generated higher returns than intended for farmers? 27 A. The differences were not that great. I don't 28 recall huge discussion on it. The discussion on the 4723 1 change happened obviously in the second half of 2020. 2 Q. So it's only if prices are negative that you 3 believe it's disorderly marketing? 4 A. That is an attention-catching feature of dairy 5 markets in the minds of dairy farmers, yes. 6 Q. That's not quite an answer to my question. 7 Is it only when the formula produces less revenue 8 than the higher-of that you consider it disorderly? 9 A. The instances in which the revenue from the change 10 was positive were relatively modest and not disorderly, 11 whereas the huge drop from -- in the second half of 2020, 12 that's of a magnitude that I would characterize as 13 disorderly. 14 The subsequent plateauing at the lower level from 15 roughly the end of 2020 until the middle of 2022 was -- 16 kind of mirrored the first several months where it was a 17 modest recovery, but would not -- I would not characterize 18 the times in which prices were increasing as disorderly, 19 but only those periods that seem to be built into the 20 current mover. When the difference was negative, it 21 seemed to generate much more accelerated changes to the 22 downside than to the upside. 23 Q. And this inherent kind of structural bias that you 24 have talked about, FMMOs set minimum prices, correct? 25 A. That's correct. 26 Q. They are not meant to be market-clearing, correct, 27 for Class I? 28 A. Yes. They are designed to be minimum prices. 4724 1 Q. And it is the cardinal sin of minimum prices to 2 set them too high, correct? 3 A. That is a basic feature of every description I 4 have seen from USDA of the concept of minimum prices. 5 Q. Do you agree? 6 A. Do not set them too high to guarantee that every 7 processor, manufacturer can make a profit. 8 Q. Same is true for -- 9 A. Make Allowances. 10 Q. Yeah. But the same is true here for producer 11 prices, correct? 12 A. That's correct. 13 Q. And so when you are looking at this asymmetrical 14 risk, isn't it right that there is an upper limit, as you 15 describe it, but not a lower limit, consistent with 16 ensuring that prices do not get too far above a minimum, 17 correct? 18 A. Can you repeat that question again? I'm not sure 19 I fully understood it. 20 Q. If you are trying to ensure a minimum price, 21 aren't you better off ensuring that the price does not go 22 too high as opposed to ensuring that the price does not go 23 too low? 24 A. Yes, but looking at Figure 1, it's a little bit 25 difficult to -- to intuit or to conclude that the price 26 has been set too high. 27 Q. And would you agree that a goal of the system 28 should be price stability? 4725 1 A. Price stability in -- in terms of it being the 2 opposite of disorderly marketing. You cannot make dairy 3 prices stop moving. You know, price -- price volatility, 4 price changes is an inherent feature of a commodity 5 industry like dairy, and could be even be interpreted as 6 kind of a healthy sign. But it has -- there are extremes. 7 Q. And the proposed change here in NMPF Proposal 13, 8 what impact will that have on minimum prices for Class IV? 9 A. What impact will going back to the higher-of have 10 on Class IV? 11 Q. Yes. 12 A. I'm not sure it will have any impact. 13 Q. How about -- 14 A. Class IV is being moved by forces relative to the 15 markets for butter and nonfat dry milk and other things. 16 I'm not sure that Class I supply and demand and price 17 issues really affect Class IV prices that much. It's kind 18 of the opposite. 19 Q. Thank you. I want to just to make sure I 20 understand that. 21 So Class I supply and demand prices don't impact 22 Class IV supply and demand market forces, correct? 23 A. Well, as an economist, if -- if Class IV prices 24 were subject to, you know, some extreme on the upside, 25 while usually those episodes don't last too long. But 26 anything that would have an impact on Class I demand, 27 given the inelasticity of demand for Class I, that's 28 pretty small effect. 4726 1 But to -- but, theoretically, yes, if there's 2 something that -- about Class IV prices that moved Class I 3 prices up, and effect had -- whatever effect it would have 4 on Class I demand, much of that lost Class I sales would 5 temporarily probably fall into Class IV. 6 Q. But the change in the base Class I skim formula 7 doesn't have a direct impact on the minimum Class IV 8 classified price, correct? 9 A. I don't think the higher-of versus the average-of 10 would have that much of an effect -- impact. 11 Q. I'm sorry. 12 A. No. 13 Q. It wouldn't impact Class III minimum prices 14 either, correct? 15 A. Probably not. I'm not saying zero impact. But a 16 modest impact. The causation is a little different the 17 other way. Class III and Class IV affect Class I prices. 18 Q. And it wouldn't have a direct impact on the 19 Class II minimum price either, right? 20 A. What would not have a -- 21 Q. Adopting Proposal 13. 22 A. It would have the same effect or non-effect as it 23 has on Class IV. 24 Q. So the only direct impact it has on minimum 25 classified pricing is to raise the Class I minimum price, 26 right? 27 A. Yes, on average. Historically, the change in 28 movers was not supposed to change the long-run difference 4727 1 between the two -- the two movers. They were to be 2 revenue neutral, and we expected that was the case. 3 It turned out that the market churned up its level 4 of volatility, and that exploited that asymmetric risk 5 that we would just soon have been happy to see not happen. 6 Q. And I believe in your testimony you said that 7 these changes that you propose -- I should say this change 8 in Proposal 13 will only have a modest positive impact on 9 the average price of milk received by producers; is that 10 right? 11 A. That's correct. 12 Q. But it's not going to be a modest negative impact 13 on Class I, correct? 14 A. I'm not sure I fully understand the question. The 15 modest impact on -- on fluid milk prices at retail is 16 what -- is that what -- can you repeat that -- 17 Q. I'll start that again. 18 A. -- part of the question? 19 Q. Yeah. So the change, adopting Proposal 13, will 20 only have a modest positive impact on the average price of 21 milk received by most small -- by dairy farmers; is that 22 right? 23 A. In general, yeah. If you look at Figure 1, you 24 will see that except for that one period in the second 25 half of 2020, which, as I testified, we do not 26 characterize as a once-only negative impact from the 27 change in Class I movers, but it is -- will undoubtedly be 28 the largest single episode of negative impact from the 4728 1 change in movers. The others is more a -- sort of a 2 continuing ratcheting down, ratcheting up the cumulative 3 losses. 4 Q. But in terms of the money that producers will make 5 from the change, you are saying it will just be a modest 6 increase in the -- 7 A. Over a span of time. But as Figure 1 8 demonstrates, the bias toward lower -- toward losses will 9 accumulate over time. We would expect that this -- if you 10 could extend this chart five years in the future, you 11 would see that line heading down below a billion dollars. 12 I would not expect to see it increase significantly, and I 13 would not expect -- under the current Class I mover, I 14 would never expect it to go back to zero. 15 Q. And so what do you estimate the impact is on 16 Class I in moving from the current formula to your 17 Proposal 13 on the price that Class I will have to pay 18 their obligation to the pool? 19 A. I would be surprised to find in ten years or so 20 that market researchers would be able to detect a 21 significant change in the Class I sales from this change 22 in movers. Class I sales are being affected by much 23 bigger forces than the price of -- the price of milk. The 24 Class I price of milk. 25 Q. So if the impact is only modest on producers and, 26 as you claim, modest on processors, why are we here? 27 A. Because Class -- the change in Class I mover was 28 disruptive from a producer standpoint. I mean, you are 4729 1 asking me to say, okay, is a billion dollars of loss a 2 minor thing for producers given the amount of money that 3 goes -- goes to producers overall? Which is, you know, in 4 the -- you know, what, 30 or $40 billion per year? Just 5 talk to any dairy farmer and see whether they found the -- 6 you know, the change in the movers to be a problem. We 7 have heard some testimony at this hearing from dairy 8 farmers to that effect. We're going to probably hear some 9 more. 10 Q. So it's not a modest impact on farmers? 11 A. Not in their -- yeah. Not -- not from their 12 perspective, and not from our perspective, either. 13 But in the grand scheme of things, I would expect, 14 you know, after weathering -- because the big drop in the 15 second half of 2020 will never be forgotten. That has 16 permanently -- that is indelibly imprinted on dairy 17 farmers. And -- but from now on, I would expect to be -- 18 you know, to see a more modest but -- but steadily 19 downward track in the cumulative losses, just like you see 20 in the second -- the second more recent part of Figure 1. 21 Q. But the drop you are describing that no one will 22 forget was partial ly reimbursed by taxpayers, correct? 23 A. Partially reimbursed, yes. 24 Q. I would like to talk for a moment about hedging. 25 You had mentioned that earlier. 26 My understanding is that NMPF does not find the 27 need by Class I processors to hedge as warranting keeping 28 the average-of or some variation proposed by MIG or IDFA? 4730 1 A. I would say that achieving multiple objectives 2 through the Federal Order program, such as -- and please 3 note that National Milk first introduced at this hearing 4 the notion of the importance of hedging and risk 5 management in terms of how that can be accommodated in the 6 Federal Order pricing formulas by our suggestion of a 7 12-month implementation delay for our Proposal Number 1. 8 And likewise, when we talked with the -- with the 9 processors in the summer of 2017, we took totally 10 seriously the processors' express desire to be able to 11 better hedge Class I. There was no bad faith in that at 12 all. 13 But it has turned out that in the case of 14 Proposal 1, we found that there were -- that taking risk 15 management factors into account is worth, let's call it 16 some -- some level of deviation from strict Federal Order, 17 you know, pricing regulations. The way it was explained 18 to me is that if USDA makes a decision that a current 19 feature of Federal Orders needs to be changed, they are 20 under almost an obligation to change it as quickly as 21 possible. 22 We're trying to introduce some nuances that if it 23 is not too great an imposition on the normal order of 24 business in Federal Orders to accommodate the growing 25 importance of risk management, then it would be 26 appropriate to do so. 27 In the case of the -- the importance of hedging 28 some Class I milk by processors, the costs -- we were more 4731 1 than willing to countenance the original change and to 2 keep alive a modification of the average-of base movers to 3 preserve that hedging as long as we possibly could have. 4 But in the end, our decision-making body 5 determined that it was -- that the importance -- the 6 downside cost of continuing an average-of base mover of 7 whatever form was -- did not outweigh the -- you know, the 8 problems that it caused and would likely continue to cause 9 in the future. 10 Q. So despite your conclusion -- and this is -- I am 11 quoting you here -- that the changes proposed by NMPF will 12 have a modest positive impact on the average price of 13 milk -- 14 A. Yes. 15 Q. -- you still believe that modest positive impact 16 does not outweigh the need for declining Class I to try 17 and mitigate its risks on the market? 18 A. Yes. And because we -- we do not believe, as 19 economists, and we have had testimony to that effect, that 20 the price of Class I milk is not a primary causative 21 factor in the decline of Class I sales. We are fully 22 aware of the decline in Class I sales and how long that 23 has gone on and the various factors that are causing it. 24 We had an expert witness testify to that effect. 25 So on balance, life is full of trade-offs, and 26 this is one trade-off that our decision-makers have 27 decided, you know, dictates that we go back to the 28 higher-of. 4732 1 Q. And what efforts did your decision-makers 2 undertake to determine the scope of risk management 3 activities that fluid milk processors are using? 4 A. Well, I'm not sure that it was their 5 responsibility to determine that. That it was, instead -- 6 because these discussions have been going on for a while. 7 We initiated them back in late 2020. We have been open to 8 receiving any information on the importance and the 9 growing -- the growing amount and the economic value to 10 processors of having that risk management tool. We have 11 not seen very much information on that. 12 Q. If you did receive information on that, would you 13 support Proposals 14 or 15? 14 A. I cannot -- I cannot tell you that we would change 15 our position. 16 Q. Even if you received information -- I'm sorry, did 17 you -- 18 A. If you have information, we would be happy to 19 receive it. 20 Q. But even if you received it, you don't believe 21 that would change your position on these proposals? 22 A. Probably not, because it's not my decision to make 23 that change. 24 MS. VULIN: Nothing further. Thank you. 25 THE COURT: Questions of this witness? 26 MR. ROSENBAUM: Your Honor, it is almost 27 5 o'clock. It is three minutes to 5:00. I think we -- 28 THE COURT: Can't get done in three minutes, huh? 4733 1 MR. ROSENBAUM: Well, it depends how many 2 admissions Dr. Vitaliano is willing to make in three 3 minutes, but I think -- I think that based on past 4 history, I don't think it's going to work. 5 THE WITNESS: I think you're right. 6 THE COURT: I would concur with that. 7 Let's remember to do that at the end, unless we 8 want to make use of the three minutes. 9 MS. VULIN: That's what I thought. Perhaps -- so 10 if the witness has testified to Exhibit 231, I would like 11 to move it into admission in the record. 12 THE COURT: Objections? 13 MR. HILL: It's a USDA document. 14 THE COURT: It is. I think it's 15 self-authenticating. The witness basically authenticated 16 it. 17 MR. HILL: You can take notice of it. 18 THE COURT: I think I could. So Exhibit 231 is 19 admitted into the record. 20 MS. VULIN: Thank you. 21 (Thereafter, Exhibit Number 231 was received 22 into evidence.) 23 THE COURT: You're welcome. 24 All right. Yeah. Let's -- thank you, 25 Mr. Rosenbaum. 26 We're doing so well, I got excited about it. It's 27 now is a good I think we should wrap for the day. 28 Anything we need to take up on or off the record 4734 1 before we leave? What witnesses are up tomorrow? Do we 2 need any discussion or can you just -- 3 MS. HANCOCK: I think we have already done it, 4 your Honor. 5 THE COURT: Yeah, I think so, too. Bring us up to 6 date. 7 MS. HANCOCK: Your Honor, tomorrow we will 8 complete Dr. Vitaliano's testimony. Rob Vandenheuvel will 9 provide rebuttal testimony on the proposals from Select. 10 We will proceed -- and it will not be in this 11 order, but we will have Craig Alexander, Chris Hoeger, 12 Sarah Stevens, and that is all the witnesses that we're 13 going to have available tomorrow. If we are ambitious, we 14 might finish a little bit early, but we have checked with 15 all the parties, and as we understand it, no one has 16 anybody else available for tomorrow. So we'll finish the 17 day wherever we land with those witnesses. 18 THE COURT: Very well. We were going to go to 19 5:00 if necessary, right? 20 MS. HANCOCK: Oh, did I say Sarah Stevens? I 21 meant Sarah Dorland. Jeez, I don't even know where 22 Stevens came from. 23 THE COURT: All right. We're adjourned. See 24 everyone tomorrow at 8:00. 25 (Whereupon, the proceedings concluded.) 26 ---o0o--- 27 28 4735 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: November 6, 2023 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