Opinion
98-CV-6114
January 6, 2004
DECISION AND ORDER
Preliminary Statement
Exactly how much butter must a croissant blended with butter contain to be properly labeled a "butter blend" croissant? The answer to this slippery legal issue has consumed the attention of lawyers, scientists, testing laboratories, and ultimately this federal court. In a complicated legal proceeding that was long on organic chemistry and lamentably short on taste testing, the evidence adduced confirmed what noted British mathematician, logician and philosopher Bertrand Russell (1872-1970) once wrote about evaluating scientific proof: "Although this may seem a paradox, all exact science is dominated by the idea of approximation. When a man tells you that he knows the exact truth about anything, you are safe in inferring that he is an inexact man." Bertrand Russell,The Scientific Outlook, Chapter 2, 1st Ed. 1931.
FACTUAL BACKGROUND
The hard facts of this dispute center upon a soft roll known as the croissant. Made in the shape of a crescent similar to the emblem of Turkey, this small roll was first baked in Vienna in 1689 to celebrate the defeat of the Turks by Austria. The croissant was introduced into France by Marie Antoinette, who was originally the Princess of Austria. The French were later recognized as being responsible for re-formulating the dough resulting in the buttery flaky pastry which today is marketed and sold throughout the world.One of the major sales outlets for baked goods, including croissants, are supermarket bakeries. Industry publications estimate that the retail bakery business is a ten billion dollar industry. (Transcript of Contempt Proceedings, hereinafter "Tr." at 27). Aside from being lucrative, the industry is also intensely competitive. Retail bakeries, including supermarket chains, invite different manufacturers to bid on supplying frozen dough products. The bidding process often includes blind taste tests of the competing products. (Tr. at 41). It is a highly competitive process, frequently pitting large national companies against regional dough suppliers who are vying for the same frozen dough business. The "winner" usually gains the exclusive right to supply the retail outlet with the frozen dough that is ultimately baked and sold by the retail bakery as their own "freshly baked" goods.
Given the size of the industry and the profits at stake, the contenders for the various dough supply contracts are justifiably concerned that those competing against them play "by the rules." One of the most fundamental rules pertains to a supplier's obligation to truthfully label its product. The United States Food and Drug Administration (FDA) has promulgated extensive regulations concerning the accurate labeling of food products. The applicable FDA labeling rules are not in dispute here and can be simply stated. First, the regulations require that frozen croissant dough labeled and sold as a " butter croissant" must, in fact, actually be croissant dough made only with butter fat. Second, frozen croissant dough labeled and sold as a " butter blend" croissant may legally contain a blend of butter and margarine so long as the dominant shortening product used in the "butter blend" product is, in fact, butter.
FDA regulations generally require that ingredients in a food product be listed on the label "in descending order of predominance." See 21 C.F.R. § 101.4. Thus, if the ingredient label on a package of frozen croissant dough lists butter before margarine, butter must be the predominant fat ingredient in the dough.
Finally, by way of background to the current dispute, there is a meaningful link between the competitive bid process and the accurate labeling of the product by the competing suppliers. Butter is three to five times more expensive than margarine. Tr. at 716-717). Consequently, the type of shortening product used by a frozen dough supplier could dramatically affect the price for which the product could be offered for sale. A product labeled and marketed as a butter croissant but which contains margarine as a shortening product can be manufactured at a significantly lower cost than a butter croissant product that is accurately labeled and contains only butter fat. The same holds true for butter blend croissants where a manufacturer does not use butter as its dominant shortening product but nevertheless improperly labels the product as a "butter blend" croissant. Thus, an unscrupulous dough supplier who deliberately "fudges" the actual butter content of its product has a significant price advantage over a dough supplier who accurately labels the butter content of its product. By virtue of deceptive labeling, the unscrupulous supplier will be able to submit lower bids and will unfairly profit from that deception where price is a consideration in awarding the dough supply contract.
BACKGROUND OF THE BUTTER DISPUTE
The dispute between the parties here is straight forward. Pillsbury and Upper Crust are competitors in the frozen dough business. Pillsbury has accused Upper Crust of cheating by deliberately misrepresenting the butter content in its "butter blend" croissants. Upper Crust denies that the labels on its frozen croissant dough packages misrepresent the butter content of its products. Indeed, Upper Crust submits that the current litigation was commenced by Pillsbury not because of any product labeling dispute, but because Pillsbury just "lost another big customer" to Upper Crust and they "just can't believe a small unsophisticated business [like Upper Crust] could make a product that customers like better [than Pillsbury] and want to buy." (Tr. at 20).
This is not the first time that Upper Crust and Pillsbury have squared off over allegations that Upper Crust deliberately exaggerated the butter content of frozen croissant dough. In 1998, Pillsbury sued Upper Crust in federal court alleging that Upper Crust was selling "Butter Croissants" which contained less than 100% butter as is required for products labeled "butter" without qualification. In support of their claim, Pillsbury provided Upper Crust with test results it had obtained from both its own and an independent testing laboratory which revealed that the shortening used in Upper Crust's "Butter Croissants" contained an average of 35% butter.
Following discovery, this Court engaged the parties in settlement negotiations and a settlement of the case was achieved. Interestingly, in the earlier case, Upper Crust did not deny that its "Butter Croissants" contained less than 100% butter, but insisted that as a Canadian company it was simply unaware of PDA labeling regulations. However, as part of the settlement, Upper Crust acknowledged that it inaccurately labeled and sold "Butter Croissants" in violation of FDA regulations. In addition, Upper Crust agreed to have a Consent Order and Permanent Injunction entered by this Court forbidding it from: (1) "marketing or selling any croissants in the U.S. labeled "Butter Croissants unless 100% of the shortening ingredients is butter," and (2) from "marketing or labeling any croissants sold in the U.S. in any way that misrepresents the butter content of the croissant product." (Docket #14, 112). The Consent Judgment further provided that "[i]n the event any party breaches any terms of this Consent Judgment and Permanent Injunction, or any term of the incorporated Settlement Agreement, that party shall be subject to a contempt proceeding in which the Court shall award the prevailing party its attorney fees and costs, and in which the Court may, in its discretion, award damages and other relief." (Docket #14, 54). The Consent Judgment, Permanent Injunction and settlement agreement (Docket #14) were filed on January 15, 1999.
The dispute now before this Court pays tribute to the fact that Pillsbury believes Upper Crust has violated the Consent Judgment and Permanent Injunction. More specifically, Pillsbury contends that Upper Crust has intentionally overstated the butter content of its "butter blend" frozen croissant dough. The parties agree that a "butter blend" croissant which contains more than 50% butter (as a percentage of the total shortening in the dough) would comply with both FDA regulations and the Consent Judgment. However, Pillsbury maintains that Upper Crust's "butter blend" croissant dough contains less than 50% butter despite the fact that the product ingredient label lists butterbefore margarine.
In order to verify Upper Crust's alleged violation of the Consent Judgment, in January 2001, Pillsbury sent samples of Upper Crust's frozen dough to two separate independent laboratories (Rtech Laboratory and Covance Laboratory) for testing. Both labs determined that butter constituted less than 50% of the fat content of the butter blend product samples tested. (Hearing Exhibits 2 and 3). After receiving the test results, Pillsbury notified Upper Crust that "it was in violation of the [settlement] agreement and the injunction." (Tr. at 37). Unable to resolve their dispute, in April 2001 Pillsbury commenced the instant contempt proceeding.
Pillsbury concedes that the "precipitating fact" that caused Pillsbury to test the Upper Crust frozen dough was the loss of business from Kroger, its largest supermarket croissant customer, to Upper Crust. (Tr. at 41).
THE APPLICABLE STANDARD OF REVIEW
Before turning to the merits of the contempt proceeding commenced by Pillsbury, it is important to note the standard of proof needed to demonstrate civil contempt of court. Pillsbury bears the burden of proof to demonstrate contempt. Levin v. Tiber Holding Corp., 277 F.3d 243, 250 (2d Cir. 2002) ("The party seeking to hold another in civil contempt bears the burden of proof"). There are three requirements for a court to find a party in civil contempt: (1) the order being enforced must be clear and unambiguous; (2) proof of non-compliance is clear and convincing, and (3) the party charged with contempt has not been reasonably diligent and energetic in attempting to accomplish what was ordered. Equal Employment Opportunity Comm'n v. Local 580, 925 F.2d 588, 594 (2d Cir. 1991). See United States v. O'Rourke, 943 F.2d 180, 189 (2d Cir. 1991) (parties may be held in civil contempt for failure to comply with a court order if order is clear and unambiguous, proof of noncompliance is clear and convincing, and parties have not been reasonably diligent in attempting to adhere to court order). However, a finding of contempt need not be based on willful or deliberate conduct on the part of the party being charged with the violation. Equal Employment Opportunity Comm'n v. Local 638, 81 F.3d 1162, 1171 (2d Cir. 1996).Here, Upper Crust does not argue that the Court's order was unclear or ambiguous. Hence, the only issue for this Court to resolve is whether Pillsbury has demonstrated by clear and convincing evidence that Upper Crust violated the terms of the Consent Judgment and Permanent Injunction.
The parties disagree about what standard the Court should use to determine whether Upper Crust violated the Consent Judgment and Permanent Injunction. The Settlement Agreement provides, in pertinent part, that if test results from a specifically named independent laboratory showed a "Material Difference" in the percentage of shortening in an Upper Crust product, then Upper Crust would be deemed in violation of the Consent Judgment and Permanent Injunction. The term "Material Difference" was defined as a difference in the butter content of a particular Upper Crust product of fifteen percent (15%) or more from the butter content represented on the Upper Crust product label or any other Upper Crust marketing, selling or advertising material for the product. It was further agreed that a finding of "material difference" had to be "based on an average of the results of tests upon three (3) samples of the product, each sample having a different date code." (Plaintiff's Exhibit 8). The disagreement arises because the specifically named independent laboratory, FAS, went out of business after the settlement agreement was signed and the parties could not agree on an alternative testing laboratory to be the arbiter to determine whether Upper Crust was complying with the Injunction. Pillsbury argues that the Court should not employ a 15% margin of error and instead simply use a hard and fast 50% standard for deciding whether Upper Crust violated the Injunction. Pillsbury submits that the more lenient "material difference" standard that embraces a 15% margin of error in the test results is no longer applicable because the agreed upon testing laboratory is now defunct. Not surprisingly, Upper Crust urges the Court to utilize the "material difference" standard as an appropriate measure of compliance notwithstanding the unavailability of the agreed upon laboratory. Because the Court finds that Upper Crust violated the Consent Judgment and Preliminary Injunction under either standard, the Court will adopt the "Material Difference" standard urged by Upper Crust as contained in the settlement agreement negotiated by the parties.
DISCUSSION A. The Scientific Evidence
"Although this may seem a paradox, all exact science is dominated by the idea of approximation."— Bertrand Russell
Pillsbury's Evidence: As far as Pillsbury is concerned, the scientific evidence that Upper Crust violated the Consent Judgment and Permanent Injunction is unambiguous and irrefutable. Pillsbury's case in chief essentially consisted of introducing test results by two reputable independent laboratories who tested samples of Upper Crust's "butter blend" croissants. The first test was conducted by Covance Laboratories (hereinafter Covance) in February 2001. (Hearing Exhibit "2"). The second test was conducted by Ralston Laboratories (hereinafter Ralston) in September 2001. (Hearing Exhibit "1"). Covance separately tested three different samples of Upper Crust frozen dough product and Ralston separately tested four different product samples.
Both Covance and Ralston used gas chromatography to analyze the Upper Crust product samples. The Covance test results established the butter fat content (as a percentage of total fat) of the three samples as 30.39% butter, 16.67% butter and 36.08% butter. The Ralston tests established the butter fat content (as a percentage of total fat) of the four samples as 27.41% butter, 29.11% butter, 29.95% butter and 39.45% butter. (Hearing Exhibit 1, p. 1). Based on the test results, none of the Upper Crust products tested came anywhere near a 50% butterfat content. Applying the "material difference" margin of error, (7.5% on either end of the 50% butterfat requirement; i.e., 42.5%-57.5% butterfat), not one of the croissant samples tested came within this range. Accordingly, Pillsbury argues that it has met its burden of showing a violation of the Consent Judgment by clear and convincing evidence.
Upper Crust's Challenges to the Scientific Evidence: Upper Crust did not dispute that both Covance and Ralston are well-respected independent testing laboratories. (See Court Exhibit 4). What Upper Crust did dispute was whether the testing method used by Covance and Ralston (gas chromatography) could accurately measure the butterfat content of any product, including frozen croissant dough. According to Upper Crust, the testing methodology relied upon by Pillsbury was so scientifically flawed that this Court cannot rely on the testing results in determining whether Upper Crust violated the Consent Judgment and Permanent Injunction.
Given the importance of scientific evidence to Upper Crusts' defense, one would expect Upper Crust to retain and present an expert in the field of organic chemistry to challenge the reliability of the Covance and Ralston test results. Upper Crust elected a different tactic. The only expert testimony presented by Upper Crust was that of Dr. Jonathan DeVries, the Senior Principle Scientist atPillsbury (now General Mills). In a lengthy, complicated and intermittently entertaining duel (also known as direct examination), able counsel for Upper Crust attempted to convince Dr. DeVries that the testing methods upon which Pillsbury relied were so faulty that they were totally and completely unreliable.
General Mills acquired Pillsbury in November of 2001, subsequent to the filing of this lawsuit. (Tr. at 47).
Dr. DeVries holds a PhD in organic chemistry and has been the principal scientist at Pillsbury for the past twelve years. (Tr. at 100). At the time of his testimony DeVries was the President-Elect of the International Association of Official Analytical Chemists, a professional organization which formulates and issues official testing protocols and methods in the field of organic chemistry. (Tr. at 50). By any measure, Dr. DeVries is an expert in the scientific field of evaluating analytical methodologies and this Court found his testimony to be frank, factual and highly credible.
Before turning to the details of Upper Crust's claims, a brief discussion of the scientific methodology used to test the croissants is necessary. The particular test run on the Upper Crust products at issue is known as the C4 to C24 gas chromatography fatty acid analysis. The testing process involves several steps. First, the food product being tested must be homogenized. In this case, the frozen dough was "homogenized" by being ground up and thoroughly mixed. Next, the fat is extracted from the homogenized product through a scientific method called the acid hydrolysis process and methyl esters of each fatty acid are formed. After the extracted fat is methylated, the resulting methyl esters are injected into the gas chromatography device for separation. The resulting analysis yields a fatty acid profile for the extracted fats from carbon number four (C4) to carbon number twenty four (C24). The critical carbon number for purposes of both the Covance and the Ralston tests is C4 since, as Upper Crust concedes, butterfat should be the only source of C4 in the tested croissant dough. Once the C4 (also known as butyric acid) in the tested sample is segregated through gas chromatography, the resulting quantity of C4 is divided by a "standard reference value" for C4 and then multiplied by 100 to achieve the percentage of butterfat in the tested product.
The fatty acids of natural fats consists of four to twenty-four carbon atoms.
Depending on the source (USDA Handbook, O'Brien's Fats and Oils, Durkee tables), the Standard Reference Value for butyric acid varies from 3.2 to 3.8. (Tr. at 150-156).
Upper Crust does not dispute that determining the amount of butterfat in a product by isolating the amount of C4 it contains "theoretically should work." See Upper Crust Summation (Docket #57) at page 3. What Upper Crust disputes is whether the gas chromatography methodology used by Covance and Ralston can measure C4 "with sufficient accuracy to extrapolate the percentage of butterfat in a croissant." Id. at 4.
Through Dr. DeVries, Upper Crust attempted to expose a number of alleged deficiencies in the C4 to C24 gas chromatography fatty acid analysis used by Covance and Ralston. First, Upper Crust argued that the C4 to C24 gas chromatography fatty acid analysis was designed and is used primarily for nutritional labeling purposes and not to separate and measure C4 as a method of determining a butterfat percentage. Upper Crust's counsel devoted many hours with DeVries endeavoring to demonstrate that small variations of over-reporting or under-reporting of C4 would produce negligible effects for a nutritional labeling analysis, yet the exact same C4 variations would completely alter the results of testing performed for extrapolating butterfat by isolating C4. According to Upper Crust, the results of a testing methodology intended, devised and used to determine fatty acids for nutritional labeling purposes yields inaccurate data when used to extrapolate butterfat and hence cannot be relied upon as clear and convincing proof of a violation of the Consent Judgment and Permanent Injunction. Upper Crust's counsel argued that an alternative test which avoids the methylation process and only measures C4 (AOCS Official Method Ca 5c-87) should have been used. See Hearing Exhibit 421, AOCS test.
Second, Upper Crust contended, again through Dr. DeVries, that although the C4 to C24 gas chromatography fatty acid analysis is an inherently unreliable method of extrapolating butterfat from C4 alone, the Covance and Ralston test results can be scientifically interpreted in such a way that actually supports the presence of butter. Specifically, Upper Crust's counsel queried Dr. DeVries on other fatty acid values obtained in the tests — particularly those for C8, C10 and C12 — in order to substantiate Upper Crust's claim that its croissants contained adequate levels of butter. Using the gas chromatography fatty acid analysis, Upper Crust isolated the amounts of C8, C10 and C12 that were found in the croissant samples. Upper Crust's counsel then subtracted from those values the amount of C8, C10 and C12 that would be expected to be found in a product containing butter and scrutinized the remaining values and their ratios to one another. See Defendant's Hearing Exhibit 476. Counsel for Upper Crust then attempted to convince Dr. DeVries that an analysis of these three variables (C8, C10 and C12) produced ratios that were far more consistent with the presence of butter than to any other oil or fat substance.
To buttress this argument, Upper Crust presented the testimony of Claude DuFresne, the sales director of Margarine Thibault Ltd. (Thibault). DuFresne testified that the only oils used in Thibault's margarine were soy oil and palm oil. (Tr. at 281). According to Upper Crust, Thibault is their only supplier of margarine. (Tr. at 285). Because the remainder ratios for C8, C10 and C12 were not consistent with the oils used in Thibault's margarine, but were similar to the ratio of C8, C10 and C12 in butter, Upper Crust argued that the test results actually support a finding that the butterfat content in its croissants are in full compliance with the injunction and all FDA regulations.
Court Findings: Ultimately, the Court found Upper Crust's arguments intriguing but not very convincing. First of all, although Upper Crust now criticizes the methodology used by Covance and Ralston, the exact same methodology was used and relied upon when Upper Crust was found to be manufacturing "Butter Croissants" that contained substantially less than 100 percent butter. That labeling law violation resulted in the 1999 Consent Judgment and Permanent Injunction that forms the basis for the instant contempt action. Indeed, over the course of this lawsuit, no less than four testing laboratories (TPC, Rtech, Covance, and Ralston) have tested Upper Crust croissants to determine their butter content. Each laboratory based its results on the C4 to C24 fatty acid profile analysis and then extrapolated the butterfat content based on the presence of butyric acid. The fact that the scientific methodology used to test the croissants is most often used for nutritional labeling purposes simply does not render it inherently unreliable when used to isolate and calculate the presence of C4. In fact, it is logical to expect that FAS (Fish Analytical Services) would have used the same fatty acid profile methodology had it been a viable business when Pillsbury alleged a violation of the Consent Judgment and Permanent Injunction.
The Settlement Agreement signed by the parties designated FAS as the independent testing laboratory to be used to conduct tests on the croissant samples, however, as described infra at footnote 3, FAS has since gone out of business.
Second, all of the foregoing pays tribute to the fact that it was no secret that Pillsbury intended to rely on the C4 to C24 fatty acid profile in meeting its burden of proof in this contempt action. The test results were all provided to Upper Crust. Each independent laboratory found Upper Crust to be in violation of federal labeling laws. Upper Crust thus had ample notice that a successful defense to the scientific evidence would either have to: (1) impugn the Pillsbury test results to the point where the Court could not fairly rely upon them, and/or (2) offer alternative test results that supported their claim that Upper Crust "butter blend" croissants contained more than 50% butter. Given the clear direction this contempt proceeding was taking, Upper Crust's defense was odd, to say the least. The only expert testimony Upper Crust offered was the chief scientist of their adversary. And even with the ample latitude afforded by this Court to Upper Crust's counsel in questioning Dr. DeVries, DeVries was immovable in his expert opinion that the C4 to C24 fatty acid methodology was not only a scientifically sound way of determining the butterfat content of a food product, but that the test results here were accurate, reliable and indubitable.
Third, despite ample opportunity to do so, Upper Crust never submitted alternative testing results to support their defense. While the burden of proof rested with Pillsbury, Upper Crust's counsel affirmatively suggested to the Court that a test which avoids the methylation process and only measures C4 would be a far more accurate and reliable testing methodology. Yet, Upper Crust apparently decided not to have its "butter blend" croissant tested pursuant to AOCS Official Method Ca 5c-87 and this Court declines to divine what the results might have been had a butyric acid specific testing method been used.
Fourth, Upper Crust's attempt to demonstrate though Dr. DeVries that the C8, C10 and C12 values actually support the conclusion that the tested croissants had adequate butter content was unconvincing. Even if Claude DuFresne, the sales director of Thibault, was correct in his analysis of the oils used in manufacturing the Thibault margarine, the question remains whether Upper Crust adds shortening ingredients other than Thibault margarine to its croissant dough. As discussed hereafter, Upper Crust's record of accurately recording the ingredients used in its croissant dough is spotty at best and the testimony of its "Quality Assurance Manager" lacked credibility. Dr. DeVries did agree with Upper Crust's counsel that if a croissant was made with fats, oils or margarine that do not contain ingredients that are a source for C8, C10 or C12, then the values for C8, C10 and C12 would allow computation of the percentage of butter in the croissant. But based on the fatty acid profile, Dr. DeVries did not buy the "if" when it came to the Upper Crust croissants. DeVries explained that Upper Crust's theory was scientifically invalid because it ignored the levels of C14 present in the samples. According to DeVries, confidence in the results of a fatty acid profile stems from analyzing the entire profile and not isolated ratios. (Tr. at 470). To prove his point, Dr. DeVries examined the C14 value of the Covance and the Ralston test results. While the C8 to C10 to C12 ratios, if examined in isolation, might be consistent with the presence of butter, the C14 value was so far below what it should have been had the sample contained butterfat that it rendered Upper Crust's theory invalid. In fact, the amount of C14 present in the samples was consistent with the C4 levels reported and if Upper Crust's hypothesis was correct, then the C14 levels in the samples would have been much higher. (Tr. at 471). Even accepting Upper Crust's arguments using the values obtained for C8, C10, and C12, when viewed in context of all of the C-variables — and particularly C14 — the resulting ratios are not consistent with butter. And in fact, the C6 and C14 are consistent with the ratios for butter, which only lends further credence to the argument that the remainders of C8, C10 and C12 likely came from an outside source. It must be stressed that the burden was not on Pillsbury to determine exactly what other fat source was present; rather it had the burden of showing that the fat did not come from butter alone. Pillsbury met that burden.
Finally, defense counsel did satisfy the Court with respect to one scientific principle: There is a margin of error inherent in almost any method of calculating C4 and intelligent approximation may just be a necessary ingredient of scientific certainty. As Albert Einstein once explained: "Not everything that can be counted counts and not everything that counts can be counted." Nevertheless, it is important to point out that despite any margin of error, the tests results here are remarkably consistent. Numerous tests performed by reputable independent laboratories following established scientific protocols produced consistent findings that generally showed the tested croissants to have around 30% butterfat, well below the required levels. Even taking into account the 15% margin of error agreed upon by the parties when the Consent Judgment was entered into, the Upper Crust croissants failed miserably, as not one of the test results showed the croissant to have achieved 40% butterfat. Put simply, the scientific evidence presented amply supports a finding of contempt.
B. The Non-Scientific Evidence
"When a man tells you that he knows the exact truth about anything, you are safe in inferring that he is an inexact man"
— Bertand Russell
As set forth above, Upper Crust's challenge to the accuracy of the scientific methodology used did not persuade the Court that the results were inherently unreliable. In addition to the scientific proof presented, Upper Crust presented non-scientific evidence consisting of testimony from its President and Quality Assurance Manager. Upper Crust's non-scientific evidence fared no better than the scientific evidence did. In fact, it only served to corroborate the Court's conclusions about the accuracy of the scientific evidence. As detailed below, the testimony of Upper Crust's President and Quality Assurance Manager were shown to be so inaccurate, incomplete and inexact that the Court is unable to credit any claim, however forcefully made, that Upper Crust accurately reports the shortening ingredients in their croissant products.
Upper Crust's Quality Assurance Manager: Larissa Korkina has been employed by Upper Crust as its "Quality Assurance Manager" for six years. Korkina was present throughout the trial as Upper Crust's designated corporate representative pursuant to Rule 615 of the Federal Rules of Evidence. Korkina testified that she holds a Bachelor of Science degree in food science from McGill University in Canada.
On direct examination, Korkina confidently testified that the ratio of butter to margarine in Upper Crust's "butter blend" croissants was 65 percent butter and 35 percent margarine. Korkina testified that she was certain as to the amount of butter and margarine in the product because of "calculations" she personally made and because she regularly observes the manufacturing process of "butter blend" croissants. According to Korkina, to achieve the required ratio of butter, 75 kilograms of butter are used for every 40 kilograms of margarine.
"The Smoking Croissants" (Exhibits 44 and 45): On cross examination, Korkina confirmed that the shortening ingredients in Upper Crust's "butter blend" croissants had not changed during her tenure with Upper Crust. (Tr. at 561). Apparently unknown to Korkina, however, was that prior to trial Pillsbury had obtained sealed cartons of two products manufactured by Upper Crust: Upper Crust's product number 10017 labeled as 3.75 oz. "deluxe croissants" (Hearing Exhibit 44) and Upper Crust product number 10015 labeled as 3 oz. "plain croissants" (Hearing Exhibit 45).
(a) Hearing Exhibit 44: After establishing that the "composition of the 10017 product has been the same during [Korkina's] term as Quality Assurance Manager" at Upper Crust, (Tr. at 561), Pillsbury had Korkina identify the sealed box as containing a label identifying the contents as Upper Crust product number 10017 croissants. (Hearing Exhibit 44). After opening the box and establishing that it contained croissants, Pillsbury asked Korkina to explain why the ingredient label on the 10017 croissants showed margarine and not butter as being the primary shortening ingredient.
Clearly unsettled, Korkina explained that she uses product names and not product numbers to distinguish between various Upper Crust croissant products. Korkina declared: "I'm not aware of numbers. I'm aware of the name of croissants and the name has changed as well as the sequence of margarine and butter." (Tr. at 564). Korkina reported that although Exhibit 45's product number was identical (10017) to Upper Crust's "butter blend" croissant, the predominant shortening in the product was based not on the number assigned to the product, but by the name given the product. Thus, if Korkina is to be believed, Upper Crust has assigned identical product numbers to two different products: the 10017 (3.75 oz.) "butter blend" croissant (containing butter as the predominant shortening) and the 10017 (3.75 oz.) "deluxe" croissant (containing margarine as the predominant shortening).
Korkina's explanation with respect to Exhibit 44 was both illogical and inconsistent with other evidence. Upper Crust's official list of ingredients for its "butter blend" croissants (Hearing Exhibit 458) identified the 10017 product by number and not by name. Furthermore, although Upper Crust's "butter blend" croissant is produced in a variety of sizes (each differentiated by a different product number), Korkina testified that the 10017 is the only Upper Crust croissant that shares a product number with another croissant having different shortening ingredients. (Tr. at 566). Thus, according to Korkina, a "deluxe croissant" in any other size (i.e., a 1.75 oz. or the 2.5 oz. croissant) would not share the same product number as its "butter blend" counterpart. Given the marketing and distribution confusion that would seem to be inherent in such a haphazard product identification system, Korkina's account of Exhibit 44 seemed odd to say the least. Indeed, later in her testimony Korkina stated that sometime after March 2002, Upper Crust stopped manufacturing the 10017 "butter blend" croissant altogether. (Tr. at 603).
(b) Hearing Exhibit 45: As odd as Korkina's explanation was with respect to Exhibit 44, her explanation as to Exhibit 45 was even more peculiar. Hearing Exhibit 45 was another unopened box of Upper Crust croissants. A label on Exhibit 45 identified the product number as 10015 and described the contents as "plain croissants." However, the ingredient label on the side of Exhibit 45 did not list butter as a shortening ingredient. Rather, according to Upper Crust's label on Exhibit 45, the only shortening product in Upper Crust's 10015 "plain croissant" was margarine. The ingredient label on Exhibit 45 was problematic for Korkina and Upper Crust because Upper Crust Exhibit 14 included "plain croissant" labels that listed both butter and margarine as the shortening ingredients in Upper Crust's plain croissant. (See Hearing Exhibit 14). When asked to explain the apparent ingredient inconsistencies in Upper Crust's products utilizing the same name, Korkina's explanation was vague and evasive. Although she could not "remember the date," Korkina declared that at some point in time Upper Crust stopped putting butter in its "plain croissant." (Tr. at 590-592).
There were other ingredient inconsistencies in the Upper Crust labels. Korkina conceded that the Upper Crust labels on its "butter blend" croissants (product number 10017) inaccurately represented the amount of yeast in the product and falsely claimed that the product contains dextrose when it is clear from the recipe that dextrose is not an ingredient. (Tr. at 622-624). Also introduced in evidence was a Canadian Food Inspection Agency Report (Hearing Exhibit 12) which cited Upper Crust for inaccurate ingredient labeling of its "butter blend" croissant. (Tr. at 636-638).
Upper Crust's President: Victor Litinetsky, Upper Crust's President and largest stockholder, also testified during the defense case. Litinetsky spoke primarily about the circumstances surrounding Upper Crust's assignment of two different products with the same (10017) product identification number. According to Litinetsky, Upper Crust won a contract with Kroger in which Upper Crust croissants were required to be sold through Kroger owned stores. One part of Kroger's chain was Ralph's Supermarket. Litinetsky testified that Ralph's asked Upper Crust to make a croissant with a reduced amount of butter so that it would have a higher melting point and would be easier to handle. (Tr. at 676). Upper Crust agreed and began manufacturing a new croissant which, according to Litinetsky, was sold exclusively to Ralph's. Because the predominant shortening in the new croissant was no longer butter, Litinetsky testified that he changed the name of the croissant to the "deluxe croissant" specifically to comply with federal labeling laws. As to why he changed the name of the Ralph's croissant to "deluxe" but kept the product identification number the same as the "butter blend" croissant, Litinetsky claimed that Ralph's warehouse was using the 10017 as their own SKU number and if Upper Crust changed the product code it would delay delivery of the new "deluxe" croissant from Ralph's warehouse. (Tr. at 675-678). On cross-examination, Litinetsky conceded that it "doesn't make any sense to use the same product code for two different products." (Tr. at 704). But he denied that the 10017 "deluxe croissant" was rolled out by Upper Crust in direct response to Pillsbury's allegation that the 10017 "butter blend" violated the consent judgment because it did not contain butter as the predominant shortening ingredient. According to Litinetsky, the 10017 "deluxe croissant" was merely "a special croissant designed for a stubborn customer to satisfy their needs." (Tr. at 704-705).
As with Korkina, Litinetsky's testimony was, in many respects, unsettling and difficult to credit. According to Litinetsky, Upper Crust was specially manufacturing and labeling a "one of a kind" croissant to specifically satisfy the special needs of a "stubborn customer," yet he claimed there were no documents to corroborate his explanation. Despite the fact that Ralph's request resulted in the production of two different Upper Crust products that were assigned the same product number, Litinetsky insisted that the entire matter was decided "orally." (Tr. at 718). Further, if the problem Ralph's had with the Upper Crust croissant was truly a low melting point in the dough itself, why would Ralph's want the change made only in the 3.75 oz. version of the croissant? Upper Crust manufactured and distributed many size croissants for its customers. Yet, Litinetsky and Korkina both claimed that the only Upper Crust product that shared the same product number but was manufactured with different ingredients was the 10017 product. Korkina testified that Upper Crust stopped manufacturing the "butter blend" version of the 10017 in March, 2002 (Tr. at 602). Litinetsky testified that the "butter blend" croissant is still sold "full time" today in a variety of sizes with the exception of the 3.75 oz. version. (Tr. at 703). If their testimony is to be credited, the odd result would be that: (1) Upper Crust customers can buy a "butter blend" croissant in any size except for the 3.75 oz. croissant; (2) Ralph's is the only Upper Crust customer who sells a 3.75 oz. "deluxe croissant"; and (3) Upper Crust's 10017 is sold only as a "deluxe croissant" to Ralph's ostensibly to keep the same SKU code as the now discontinued 10017 3.75 oz. "butter blend" croissant.
That Litinetsky has a penchant for playing "fast and loose" with facts surrounding Upper Crust's croissant manufacturing process was confirmed when he decided to reveal the details of his "trade secret" precision mixing station to prove that Upper Crust puts more butter than margarine in its "butter blend" croissants. According to Litinetsky, "any time" his company manufactures a batch of "butter blend" croissants the dough is blended with exactly 60 kilograms of butter and 50 kilograms of margarine. (Tr. at 715). When asked on cross-examination if he was "absolutely sure" of the butter to margarine ratio used by Upper Crust, Litinetsky responded, somewhat indignantly, "100 percent." (Tr. at 720). Litinetsky was then shown Exhibit 48, a series of Upper Crust's "Batch Control Sheets." According to Korkina, "batch sheets" were business records created by Upper Crust to document the exact ingredients and blending times of every batch of croissants made by Upper Crust employees. Unfortunately for Litinetsky, his company's batch sheets contradicted his "100 percent" certainty over the amount of butter being used in Upper Crust's "butter blend" croissants. Indeed, each and every batch sheet indicated that 75 kilograms of butter and 40 kilograms of margarine were being placed in the "butter blend" croissant dough. Litinetsky's flippant response, ("You mean to tell me that my people are throwing one more cube of butter for no reason today?" (Tr. at 725)), did nothing to rehabilitate his credibility.
Neither Korkina or Litinetsky helped Upper Crust's defense. While their testimony was not scientific, the Court assumes they were called by Upper Crust to foster the viewpoint that Upper Crust takes seriously its obligation to comply with the Consent Judgment and accurately report the shortening ingredients in its croissants. Evidence that Upper Crust's manufacturing process was rigorous and precise and its labeling process was accurate and consistent would have supported Upper Crust's position that the Court should give pause before relying on Pillsbury's scientific evidence to find Upper Crust in violation of the Court's Orders. But such was not the case. The non-scientific evidence only demonstrated that Upper Crust is not committed to accurate labeling of ingredients in its products and does not have a production process, a record keeping method or a product identification system that promotes either (1) correct measurement of shortening products used in croissants, or (2) accurate ingredient labeling of those products in the marketplace. Instead of giving pause to the accuracy of the test results, the non-scientific evidence presented by Upper Crust provided powerful corroboration for reliability of the scientific tests run on Upper Crust croissants.
CONCLUSION
The Court began by posing the question: Exactly how much butter must a croissant blended with butter have to contain to be properly labeled a "butter blend" croissant? While the methodology employed by Pillsbury may not yield a calculation precise enough for Upper Crust's liking, it did yield a result that, when considered in conjunction with the testimony of Korkina and Litinetsky, allows the Court to answer the question posed with enough certainty to state: A lot more butter than Upper Crust puts in its "butter blend" croissants.
For the reasons set forth in this Decision and Order, the Court finds that Pillsbury has met its burden of showing that Upper Crust has violated the Consent Decree and Permanent Injunction by clear and convincing evidence. The parties shall participate in a telephone conference on January 20, 2004 at 11:00 a.m. to schedule a hearing to determine damages.
IT IS SO ORDERED.