Winn-Dixie Stores, Inc. v. Dolgencorp, LLC

OMNIBUS ORDER CONCERNING EXPERT TESTIMONY OF DR. PATRICIA PACEY

DONALD M. MIDDLEBROOKS, District Judge.

THIS CAUSE comes before the Court upon the following motions filed in these consolidated eases: (1) Dolgencorp, LLC’s Motion to Strike and Exclude Proffered Expert Testimony of Patricia Pacey, Ph.D. (“Dr. Pacey”) Pursuant to Federal Rules of Evidence 702 and 703 (DE 56 in Case No. 11-cv-80601) filed on November 21, 2011; (2) Dollar Tree Store, Ine.’s Motion to Exclude Proffered Testimony of Patricia Pacey (DE 83 in Case No. 11-cv-80638) filed on December 5, 2011; and (3) Big Lots Stores, Inc.’s Motion to Strike and Exclude Proffered Expert Testimony of Patricia Pacey, Ph.D. Pursuant to Federal Rules of Evidence 702 and 703 (DE 131 in Case No. 11-ev-80641) filed on January 17, 2012 (collectively “Motions”). A Daubert hearing was held to discuss Defendants’ Motions on February 29, 2012. (DE 255). At the hearing I heard testimony from Dr. Pacey, her colleague Dr. McNulty, as well as Dr. Frederick Raffa, Dollar General and Big Lots’ expert witness, and Dr. Andrew Abere, Dollar Tree’s expert witness. After the hearing the parties submitted post-hearing memorandums outlining their arguments for and against excluding Dr. Pacey’s testimony. (DE 301, 303, and DE 313). I have reviewed the matter and am advised in the premises.

I. BACKGROUND

Dr. Pacey was hired by Plaintiffs to prepare an Economic Damage Appraisal Report (collectively the “Reports”) for three different cases in this consolidated action: (1) Winn-Dixie Stores, Inc. v. Dolgencorp, Inc. (DE 56-1 in Case No. 11-cv-80601); (2) Winn-Dixie Stores, Inc. v. Dollar Tree, Inc. (DE 83-1 in Case No. 11-cv-80638); and (3) Winn-Dixie Stores, Inc. v. Big Lots, Inc. (DE 131-1 in Case No. 11-cv-80641).¹ The Reports attempt to describe and analyze the damages arising from Dollar Tree’s, Dollar General’s, and Big Lots’ violations of Winn-Dixie’s grocery exclusives. (Reports at 3).

A. THE DATA

All of Dr. Pacey’s Reports utilize the same data. Dr. Pacey’s Reports:

• Consider sales data from June 26, 2003 through May 4, 2011 from 482 of the 484 Winn-Dixie grocery stores currently operating. (Id. at 6).

• Identify over 12,000 stores as competitors located within three miles of a Winn-Dixie store and over 700 competitors located within 0.2 miles of a Winn-Dixie store. (Id. at 7).

• Aggregate Winn-Dixie’s competitors into five categories: supermarkets, other supermarkets, variety stores, drug stores, and others. (Id.).

• Consider Dollar Tree, Dollar General, and Big Lots separately from the other variety stores in some of the analyses. (Id.).

• Identify 73 Dollar General stores, 42 Dollar Tree Stores, and 24 Big Lots stores within 0.2 miles of a WinnDixie. (Id. at 8).

• Identify 145 to 207 other variety stores (depending on which of the Defendants is separated in the analysis), 52 supermarkets, 26 other supermarkets, 241 drug stores, and 154 others within 0.2 miles. (Id.).

• Identify 829 Dollar General stores, 440 Dollar Tree Stores, 194 Big Lots stores, 1,485 to 2,120 other variety stores (depending on which of the Defendants is separated from the analysis), 1,307 supermarkets, 1,166 other supermarkets, 2,384 drug stores, and 4,866 others within three miles of a Winn-Dixie store. (Id. at 9).

Dr. Pacey’s analysis considers four broad categories of data: Winn-Dixie store characteristics;² demographics surrounding each Winn-Dixie store;³ competitors by type and distance;⁴ and Defendant store characteristics.⁵ (Id.). Dr. Pacey notes that:

Presently, information on [Defendant stores’] square footage and sales, whether total store and/or grocery allocated, are not complete and, as such, cannot be included in the analysis at this time. A supplement to this report will be provided if and when this data becomes available. Of note, it is anticipated square footage ... could assist in sorting lost profits across stores ... as it is understood from industry experts ... that the larger the store and/or the larger the amount of space within a store assigned to “grocery products,” the greater the store sales.

(Id. at 12).

B. FORMULATING THE MODELS

Dr. Pacey’s Reports rely upon data obtained from the annual American Community Survey and the Nielsen Company, which “provide basic information regarding Winn-Dixie sales and store characteristics, area demographics, and type and prevalence of competitors within the Winn-Dixie trading area.” (Id.). Dr. Pacey used this data to determine “the proper specification of the econometric model____” (Id.). This data showed that the typical Winn-Dixie store is approximately 45,000 square feet with annual grocery sales of approximately $12.5 million. (Id.). The data also showed that most store characteristics, such as store age, and area demographics, such as per capita income, were fairly symmetrical, with the exception of the distribution of population densities, since Winn-Dixie stores are heavily located in more rural areas. (Id. at 13). Dr. Pacey argues that this data indicates Winn-Dixie stores with a variety store within 0.2 miles have, on average, lower grocery sales than their similarly sized counterparts without a competitor. (Id. at 15-17). Dr. Pacey states that the descriptive statistics in her Reports “provide insight and reasonable crosschecks of the data and are often times used to assist in the measurement of economic damages.” (Id. at 18).

C. REGRESSION MODEL

Dr. Pacey used a cross-sectional regression analysis in attempting to determine Winn-Dixie’s damages from Defendants’ alleged violations of the terms of their leases. (Id.). The cross-sectional analysis examines the sales of many Winn-Dixie stores at a single point in time. (Id.). Dr. Pacey’s Reports account for differences in substitutability by grouping competitors into five separate categories: variety stores, supermarkets, other supermarkets, drug stores, and others. (Id. at 20). Dr. Pacey then isolates Dollar General, Dollar Tree, and Big Lots from the other variety stores depending on which Defendant she is estimating damages for. (Id.). Proximity is accounted for by using three different distance classifications: less than 0.2 miles, 0.2 to 1.0 miles, and 1.0 to 3.0 miles. (Id.). Dr. Pacey’s Reports do not take into account store-specific size data because it is only available for a few of the competitors and is not expected to have an overall material impact on the aggregate lost profits. (Id.). Dr. Pacey’s Reports also use a fresh meat covariate. Dr. Pacey states that the meat covariate is:

included to account for location conditions (e.g. traffic patters) impacting the sales of the Winn-Dixie store. As local conditions change that potentially impact the number of shoppers and/or individual shopper bills, it is anticipated that the sales on non-meat grocery items will be impacted in a similar manner. For example, if road construction limits access to the store, both sales will tend to decrease. Of note, because variety stores do not sell fresh meat, the use of this variable as a proxy for local conditions does not change the relationship between “Sales” and the presence of a co-located variety store to be estimated more precisely. The “logarithm of meat” is significant and indicates a strong relationship between non-meat and meat sales, as is anticipated.

(Id. at 22, 23).

Dr. Pacey’s cross-sectional regression analysis utilizes sales data for 2009 and 2010 for 482 Winn-Dixie stores. (Id. at 21). The dependent variable in the analysis is a logarithm of total non-meat grocery sales,⁶ while the independent variables include meat sales, year, state, store age, store square feet, store type, whether remodeled, population density, age, race, income, education, Defendant stores, and distance, whether 0.2 miles, 0.2-1.0, or 1.0-3.0 miles from other variety, supermarket, other supermarkets, drug stores, and other. (Id. at 24-25). Additionally, Dr. Pacey separates the Defendant from the variety store category for each Report. (Id.). Dr. Pacey states that the results of her regression model suggest the following:

• The probable presence of a co-located Big Lots store reduces Winn-Dixie’s “Sales” by 7.7%. (Id. at 23).

• The probable presence of a co-located Dollar General store reduces WinnDixie’s “Sales” by 6.5%. (Id.).

• The probable presence of a co-located Dollar Tree store reduces Winn-Dixie’s “Sales” by 5.0% (Id. at 23-24).

On December 12, 2011, Dr. Pacey issued “an updated report (“Updated Report”) to incorporate additional information received and to clarify issues that have arisen since issuing the November 1, 2011 report” in the Dollar General case.⁷ (DE 182-2 at 1). Dr. Pacey’s Updated Report distinguished her econometric model and damage model and explained why she included certain variables in her Report. (Id. at 1-4). Dr. Pacey explained that while 73 Dollar General stores located within 0.2 miles of Winn-Dixie were included in here regression model, only those Dollar General stores alleged to be in violation of WinnDixie’s grocery exclusive were included in her damage model. (Id. at 6). Dr. Pacey also updated a couple of variables in her regression analysis. First, Dr. Pacey altered the “supermarkets” category, to include three tiers: major supermarket chains, wholesale food clubs, and local supermarket chains and stores. (Id. at 5). Dr. Pacey states that this refinement had a small, but increased effect on the magnitude of the impact from close proximity Dollar General Stores, from 6.5% to 6.7%. (Id.). Second, Dr. Pacey differentiated between violating and non-violating Dollar General stores and found that the negative effect of Dollar General stores’ proximity rises to 7.3% of Winn-Dixie sales when considering alleged violators of Winn-Dixie’s grocery exclusive. (Id. at 10).

D. DAMAGES MODEL

After completing her regression model, Dr. Pacey then used this information to estimate the sales Winn-Dixie could reasonably have expected but for the co-located Defendants being in violation of its grocery exclusives. (Id. at 28). Dr. Pacey claims that her data “allows for highly probable estimates of the additional sales lost by Winn-Dixie due to the presence of a co-located” Defendant. (Id. at 29). Dr. Pacey notes, however, that not all of the Defendant stores are in violation of its lease agreement. (Id.). Since lost sales are different than lost profits, Dr. Pacey converts the lost volumes of sales used in her analysis to lost profits by utilizing Winn-Dixie’s operating margin in order to calculate damages. (Id. at 29). Additionally, Dr. Pacey’s Reports alter the time-frame for recoverable damages by each state depending on the applicable statute of limitations. (Id. at 30). Dr. Pacey then applies this data to the Defendant stores understood to be in violation of its lease agreement to reach her final calculation of damages: $54,589,200 for Dollar General; $20,792,100 for Big Lots; and $16,101,000 for Dollar Tree. (Id. at 31, 32, 34). The Reports contain the following note:

[N]o allocation has been made to damages for the square footage in excess of the grocery store exclusive, as it is expected that reduction in Winn-Dixie sales resulting from [Defendant’s] sales based on conforming footage is either offset by the increased traffic to the shopping center generated by the presence of [Defendant] (in part the basis for Winn-Dixie’s willingness to accept a certain level of competing grocery sales) or because [Defendant] has closed stores rather than comply with a 500 square foot grocery restriction.

(Id. at 34).

E. CAUSATION

To establish causation, Dr. Pacey states that

The a priori economic logic is a competitor located near a Winn-Dixie store and selling competing products will negatively impact Winn-Dixie sales.... The fundamental task in measuring this potential negative economic impact is to understand the relationship between Winn-Dixie sales and the sales of nearby competitors, along with other facts that will reasonably explain grocery purchase behavior.

(Id. at 19). According to Dr. Pacey, if all other factors remain constant, “the impact of a competitor on Winn-Dixie sales depends upon three factors: proximity, substitutability, and size.” (Id.).

II. LEGAL STANDARD

The admissibility of expert testimony is governed by the framework set out in Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). The party seeking to have the expert testimony admitted bears the burden of demonstrating its admissibility by a preponderance of proof. Davidson v. U.S. Dep’t of Health & Human Servs., No. 7:06-129-DCR, 2007 WL 3251921 at *2 (E.D.Ky. Nov. 2, 2007) (internal citations omitted). See also United States v. Frazier, 387 F.3d 1244, 1260 (11th Cir.2004) (“The burden of establishing qualification, reliability, and helpfulness rests on the proponent of the expert opinion.”). Rule 702 provides:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.

Fed.R.Evid. 702. According to the Supreme Court, the inquiry envisioned by Rule 702 is a flexible one, in which federal judges perform a “gatekeeping role” to ensure that speculative and unreliable opinions do not reach the jury. Daubert, 509 U.S. at 594-95, 597, 113 S.Ct. 2786 (“[Rule 702’s] overarching subject is the scientific validity and thus the evidentiary relevance and reliability of — the principles that underlie a proposed submission. The focus, of course, must be solely on principles and methodology, not on the conclusions that they generate.”).

In Daubert, the Supreme Court listed several factors federal judges may consider in determining whether to admit expert scientific testimony under Rule 702: whether an expert’s theory or technique can be and has been tested; whether the theory or technique has been subjected to peer review and publication; whether the known or potential rate of error is acceptable; and whether the expert’s theory or technique is generally accepted in the scientific community.⁸ 509 U.S. at 593-94, 113 S.Ct. 2786 (declining to set forth a “definitive checklist or test”).

The Supreme Court subsequently held that the Daubert factors “may or may not be pertinent in assessing reliability, depending on the nature of the issue, the expert’s particular expertise, and the subject of his testimony Too much depends upon the particular circumstances of the particular ease at issue.” Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 150, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999) (internal citations and quotations omitted). Accordingly, “the trial judge must have considerable leeway in deciding in a particular case how to go about determining whether particular expert testimony is reliable.... [A] trial court should consider the specific factors identified in Daubert where they are reasonable measures of the reliability of expert testimony.” Id. at 152, 119 S.Ct. 1167. The trial court has the same kind of latitude in deciding how to test an expert’s reliability as it enjoys when it decides whether or not that expert’s relevant testimony is reliable. Id.

The Eleventh Circuit engages in a three part inquiry to determine the admissibility of expert testimony under Rule 702, considering whether: Quiet Tech. v. Hurel-Dubois UK Ltd., 326 F.3d 1333, 1340-41 (11th Cir.2003) (internal citations omitted).

(1) [T]he expert is qualified to testify competently regarding the matters he intends to address; (2) the methodology by which the expert reaches his conclusions is sufficiently reliable as determined by the sort of inquiry mandated in Daubert; and (3) the testimony assists the trier of fact, through the application of scientific, technical, or specialized expertise, to understand the evidence or to determine a fact in issue.

III. DISCUSSION

As stated by the Eleventh Circuit in Quiet Tech., three criteria must be met for expert testimony to be admitted under Rule 702: (1) the expert is qualified to testify competently regarding the matters she intends to address; (2) the methodology by which the expert reaches his conclusions is sufficiently reliable; and (3) the testimony assists the trier of fact, through the application of scientific, technical, or specialized expertise, to understand the evidence or to determine a fact in issue. Quiet Tech., 326 F.3d at 1340-41.

With regard to the first criteria, I find that Dr. Pacey is qualified to testify competently regarding economic damages caused by Defendants’ violation of Plaintiffs’ grocery exclusives. Dr. Pacey received her Ph.D. in Economics from the University of Florida in 1976. (DE 61-58 at 58). She is the president of her own economic and business consulting firm. (Id.). She has experience providing economic evaluations of lost profits and/or other economic damages in a number of areas, including specific business interruptions such as breach of contract. (Id.). Prior to being president of her own consulting firm, Dr. Pacey was a professor of economics and finance for over fifteen years. (Id. at 59). She has also published nineteen journal articles and publications on topics relating to economics. (Id. at 60-61). Accordingly, it is evident that Dr. Pacey is qualified to testify in this case regarding economic damages caused by Defendants’ violation of Plaintiffs’ grocery exclusives.

Next, I must consider whether Dr. Pacey’s testimony would assist the trier of fact to determine a fact in issue and whether her methodology is sufficiently reliable. For the following reasons, I find that Dr. Pacey’s Reports would not assist the trier of fact in determining a fact in issue in this case and that her methodology is not sufficiently reliable.

A. DR. PACEY’S REPORTS MEASURE COMPETITION AND DO NOT ACCOUNT FOR PERMITTED SALES UNDER THE GROCERY EXCLUSIVES

At the outset, Dr. Pacey agrees that her regression analysis does not account for the fact that most of the grocery exclusives at issue in this matter permit Defendants to sell a certain amount of groceries.⁹ Rather, Dr. Pacey states that the results of her regression analysis show that “the probable presence of a co-located [Defendant] store reduces Winn-Dixie’s ‘sales’.... ” (Reports at 23-24). Dr. Pacey stated at the Daubert hearing that the regression model “has nothing to do with a violation. It has to do with close proximity, the consumer’s choice because of close proximity.” (DE 276 at 55). In other words, the regression model measures “consumer behavior.” (Id. at 54). Dr. Pacey contends that the damage model measures the violating stores by multiplying grocery sales for each relevant WinnDixie store times the reduction (from the regression model) times the relevant lost profit margin. (Reports at 31). Dr. Pacey states that she is measuring “the average impact of the presence” of Defendants. (DE 276 at 57).¹⁰

Dr. Pacey provides three reasons why she believes it is unnecessary to account for the fact that most of Defendants’ stores are allowed to sell a certain amount of groceries. First, Dr. Pacey states that it is understood that Defendants do not operate any of their stores “with 500 square feet or less of sales area devoted to grocery products.” (Reports at 34). Second, “Winn-Dixie management and the grocery industry experts have indicated the ‘allowable’ 500 square feet reflects the benefits of increased traffic associated with additional retailers ... offsets the detriment of such incidental sales.” (Id.). In Plaintiffs’ Post-Hearing Submission, Plaintiffs reiterate that “the sale of grocery products from 500 square feet of sales area is too small an offering of overlapping substitute goods to have a significant impact on Winn-Dixie. The impact from 500 square feet would be zero or negligible. This is the logical assumption underlying the exception allowing 500 square feet: Winn-Dixie is obviously willing and able to tolerate some limited competition that has a negligible effect.” (DE 313 at 29). Third, Dr. Pacey’s “own research utilizing small sized Dollar Tree stores suggests no statistically significant impact from incidental grocery store sales.”¹¹ (Reports at 34).

In this case, Plaintiffs seek to recover damages from Defendants for violating Winn-Dixie’s grocery exclusives by selling more groceries than are permitted under the respective leases. Therefore, I think it is critical that Dr. Pacey’s analysis account for the fact that nearly all of the grocery exclusives allow Defendants to sell a certain amount of groceries rather than merely state that she, nor Winn-Dixie’s management, does not think these limited sales make a difference. By simply multiplying the reduction from the regression model times grocery sales for the relevant WinnDixie store times the relevant lost profit margin, she fails to take into account permitted grocery sales. As a result, all of Defendants’ stores are treated equally as violators regardless of the degree of their purported violation.

As discussed further, infra Part III.B, I find that Dr. Pacey’s focus on competition as opposed to Defendants’ violations is the fatal flaw in her Reports. As Dr. Raffa stated at this Court’s Daubert hearing, “competition is not illegal. This lawsuit is not about competition. Dr. Pacey’s model is about competition. She’s measuring competition.... She is not measuring the impact of a restrictive covenant.” (DE 276 at 342). I raised the issue of why the correct measurement is not the effect of the violation as opposed to any other competition with Dr. Pacey at this Court’s Daubert hearing. She responded “if you want to measure it that way, its not a regression measure. Then don’t use regression. Use some other avenue. But regression you can’t — we’re measuring causality, the economic reasons for the rational relationships of those variables.” (DE 276 at 191). Rule 702 “requires that the evidence or testimony ‘assist the trier of fact to understand the evidence or to determine a fact in issue.’ ” Daubert, 509 U.S. at 591, 113 S.Ct. at 2795. By measuring overall competition and consumer behavior instead of Winn-Dixie’s damages caused by Defendants’ violations of the grocery exclusives, I find that Dr. Pacey’s Reports are analyzing the wrong problem and therefore do not assist the trier of fact to determine a fact in issue in this case.

B. DR. PACEY’S REPORTS DO NOT ESTABLISH CAUSATION

It is a well-established theory of economics that

Causality cannot be inferred by data analysis alone; rather, one must infer that a causal relationship exists on the basis of an underlying causal theory that explains the relationship between the two variables. Even when an appropriate theory has been identified, causality can never be inferred directly. One must also look for empirical evidence that there is a causal relationship.

Daniel L. Rubinfeld, Reference Guide on Multiple Regression, in Reference Manual on Scientific Evidence 303, 310 (3d ed.2011). Thus, to prove causation Dr. Pacey must provide empirical evidence that there is a causal relationship: that Defendants’ stores are decreasing WinnDixie’s sales by selling more groceries than permitted in the grocery exclusives. However, Dr. Pacey stated at this Court’s Daubert hearing that competition is the underlying causal theory in this case. (DE 276 at 113). Dr. Pacey articulated this theory further in her Reports:

To be sure, we do not rely on the regression analysis to show causality.... The causal theory underlying our regression analysis is a simple application of well-known business and economic principles (and of common sense), as described below:

• Two stores that are located near one another can impact each other’s sales in two primary ways.

• The first is that having multiple stores in close proximity makes the area more attractive to customers as it reduces travel and shopping time and hence, increases customer traffic. This is understood to be one reason underlying the development of shopping centers.

• The second way stores in close proximity can impact each other’s sales is when the stores sell similar products; as customers buy from one store, the demand for similar products decreases at the other store. This is understood to be one of the reasons for the “grocery exclusive” clause in the Winn-Dixie lease agreements.

It is these logic chains, and not regression analysis, that provide the causal basis for a relationship between WinnDixie sales and “co-located” [Defendant] stores. The function of the regression analysis is to measure the strength of the relationship.

(DE 182-2 at 7-8).

In their Post-Hearing Submission, Plaintiffs argue that Defendants are competing with Plaintiffs by selling overlapping substitute goods and this demonstrates causation. (DE 313 at 6). Plaintiffs assert, and Defendants’ experts agree, that as a general theory, competition suggests that if a competitor is selling substitute goods in close proximity to a store, this would have a negative impact on the sales of the first store. (Id. at 7). Plaintiffs point out that corporate representatives from Dollar General, Dollar Tree, and Big Lots include exclusives in their own leases to prevent competitors that impact their business from locating near them. (Id. at 8-9). Plaintiffs offer Defendants’ respective 10K Reports and Plaintiffs’ investigative reports as empirical evidence to support Dr. Pacey’s theory of causation in this case. (DE 313 at 12-22).

While I do not question the general theory of competition, the problem is that Dr. Pacey provides no empirical evidence that the Dollar General, Dollar Tree, and Big Lots stores caused Winn-Dixie damages by violating their grocery exclusives. As stated previously, the issue in this case is not whether Defendants are competing with Winn-Dixie. The issue is what damages, if any, Defendants caused Plaintiffs by selling more groceries than allowed in the respective stores’ grocery exclusives. Since Dr. Pacey’s Reports do not provide any empirical evidence on this issue, I find that allowing her to testify may cause a jury to believe that Defendants are causing Plaintiffs’ damages by selling more groceries than allowed, when her regression model and other empirical data is only focused on competition.

The empirical evidence produced by Dollar General and Big Lots’ expert, Dr. Raffa, also gives me pause when considering Dr. Pacey’s theory of causation in this case. Dr. Raffa analyzed Winn-Dixie’s net sales data from 12 locations where a Dollar General store was co-located with a WinnDixie store, but later closed. (DE 301 at 18). By comparing Winn-Dixie’s net sales from 12 months prior to Dollar General’s closing to 12 months after Dollar General’s closing, Dr. Raffa found that Winn-Dixie’s sales decreased by a total of 1.6% across these stores during the year after the co-located Dollar General stores closed. (DE 301 at Ex. G). Although this analysis does not take external factors into account, the data suggests that Dollar General’s closings resulted in decreased sales for WinnDixie. (Id.). Dr. Raffa also presented data indicating that the alleged lost sales at two Winn-Dixie stores caused by Dollar General’s violation of the grocery exclusives is actually greater than Dollar General’s total sales. (Id. at Ex. F and G).¹² Utilizing the theories of competition and substitutable goods, which would suggest that Winn-Dixie sales decrease when consumers purchase goods at Dollar General that they would otherwise purchase at Winn-Dixie, it seems implausible that Winn-Dixie’s lost sales could be greater than Dollar General store’s total sales. Accordingly, I find that Dr. Pacey’s underlying causal theory of competition does not prove causation in this case. Although I find Dr. Pacey’s regression analysis irrelevant to this case because it measures competition rather than the effect of Defendant’s violations of Winn-Dixie’s grocery exclusives, I also find that since Dr. Pacey does not establish causation, the probative value of her evidence is substantially outweighed by the danger of misleading the jury under Fed.R.Evid. 403.

C. DR. PACEY’S REPORTS HAVE NOT BEEN PEER-REVIEWED

In Daubert, the Supreme Court stated that when considering whether certain expert testimony is admissible,

Another pertinent consideration is whether the theory or technique has been subjected to peer review and publication. Publication ... is not a sine qua non of admissibility The fact of publication (or lack thereof) in a peer reviewed journal will be a relevant, though not dispositive, consideration in assessing the scientific validity of a particular technique or methodology on which an opinion is premised.

Daubert, 509 U.S. at 593-94, 113 S.Ct. at 2797. Winn-Dixie notes that regression analysis has long been accepted in the field of economics and in the courtroom. (DE 313 at 34-35). However, Dr. Pacey’s regression analysis for this case has not been subjected to peer review. While not dis-positive of whether Dr. Pacey’s testimony is admissible, “one very significant fact to be considered is whether the experts are proposing to testify about matters growing naturally and directly out of research they have conducted independent of the litigation, or whether they have developed their opinions expressly for purposes of testifying. ...” Daubert v. Merrell Dow Pharmaceuticals, Inc., 43 F.3d 1311, 1317 (9th Cir.), cert. denied, 516 U.S. 869, 116 S.Ct. 189, 133 L.Ed.2d 126 (1995).

D. FLAWED VARIABLES

In addition to my concerns regarding whether Dr. Pacey’s testimony would assist the trier of fact, I also find that Dr. Pacey’s regression model contains flawed variables leading me to question whether her methodology is sufficiently reliable.

1 DR. PACEY CALCULATES WINNDIXIE’S DAMAGES FROM 2005-2008 BY USING SALES DATA FROM 2009 AND 2010

Dr. Pacey’s Reports use Winn-Dixie’s sales data from 2009 and 2010 to calculate its damages from 2005 through 2008. (DE 313 at 30). Plaintiffs state that Dr. Pacey utilized this sales data because it “was the most recent, up to date, available data for Winn-Dixie store sales ....” (Id.). Plaintiffs state in their Post-Hearing Submission that “Dr. Pacey’s opinion is that there is not enough differences to render the results inapplicable” to the period from 2005 through 2008 (Id. at 31). However, Dr. Pacey does not provide any support for this assumption. This type of “leap of faith” is condemned by Daubert. See Rink v. Cheminova, Inc., 400 F.3d 1286, 1292 (11th Cir.2005); McGlinchy v. Shell Chemical Co., 845 F.2d 802, 807 (9th Cir. 1988) (finding a study “hopelessly flawed” where the expert witness “did not confirm that relevant market conditions were the same before and after the time the injury was alleged to have occurred”). Dr. Pacey’s leap of faith is particularly troublesome considering the economic conditions in 2009 and 2010, during the worst recession since the Great Depression, were remarkably different than they were from 2005 through 2008 and I think it is likely, if not certain, that these conditions affected sales at grocery stores. For instance, Dr. Abere testified regarding the “triple trade down” in the supermarket business during the recession, which resulted in consumers who typically shopped in supermarkets going to places like Wal-Mart where prices are cheaper. (DE 276 at 282). Dr. Abere argued that the triple trade down has bolstered Dollar Tree’s sales during the recession, while it has hurt sales at supermarkets such as WinnDixie. (Id.).

2. THE MEAT COVARIATE IS UNRELIABLE

As stated previously, Dr. Pacey’s Reports utilize a meat covariate, which she argues helps track general increases or decreases in Winn-Dixie store sales due to external factors. (DE 313 at 30). Since Defendants do not sell fresh meat, Dr. Pacey argues that Winn-Dixie’s meat sales are correlated to non-meat grocery sales. (Id.). Dr. Pacey asserts that “[m]eat captures foot traffic but in a different way.” (DE 276 at 37). Dr. Pacey argues that the meat covariate works because-meat is “a big portion of your budget for groceries____[PJeople go to the grocery store— when you’re shopping, you typically plan your meal around meat. Not always, but typically that’s how you do that.” (Id. at 38). Therefore implicit in Dr. Pacey’s theory is that consumers are purchasing their meat and other groceries at the same store: However, as Dr. Abere suggested at this Court’s Daubert hearing, that may not be true. Dr. Abere cited a 2011 article from Progressive Grocer which stated “consumers say that meat needs aren’t truly satisfied by a single store. When asked how much of their meat is purchased in the store where they do most of their food shopping, two-thirds say they purchase 75 percent or more of their meat at the store where they do most of their grocery shopping. For the remaining third the figure is even lower.” (Id. at 296). Thus, Dr. Pacey’s assumption regarding meat sales appears flawed. Dr. Pacey’s decision to use meat sales to measure foot traffic seems even more questionable considering she did not ask Winn-Dixie for information regarding foot traffic in its stores. (Id. at 216).

3. DR. PACEY’S REPORTS ARBITRARILY USE A THREE-MILE RADIUS TO MEASURE COMPETITION

In her Reports, Dr. Pacey states that “[t]he names and locations ... for all competitors identified within 10 miles of any of the 482 Winn-Dixie stores are provided by Nielsen Company (via a contract with Winn-Dixie management).” However, in her regression analysis, Dr. Pacey only looked at Winn-Dixie’s competitors within a three mile radius. (Reports at 11). She states that “[t]he three (3) mile measure is based on the understanding that this distance is a reasonable measure for the ‘trading area’.” (Id.). Dr. Pacey added that “[w]e could have gone further because Winn-Dixie did ultimately give us out to ten miles, but by three miles out it was so small enough that there was really not much reason to include that....” (DE 276 at 50). While not necessarily fatal, I find it questionable that Dr. Pacey would only include Winn-Dixie’s competitors within a three mile radius when Winn-Dixie’s management receives data regarding all competitors within a ten mile radius. If Winn-Dixie’s management finds that its competitors are within a ten mile radius of its stores, it seems logical to include these competitors in a regression model that is measuring consumer behavior and competition.

At this Court’s Daubert hearing, Dr. Abere testified regarding the problem of using a three-mile radius in this regression model. Dr. Abere stated that a 2009 Department of Transportation survey found that the average shopping trip was just over six miles. (DE 276 at 290). While Dr. Paeey’s model utilizes a three-mile radius in measuring competition, the model does not take into account that a competitor may be within three miles of a consumer, but not within three miles of Winn-Dixie. (Id. at 291). As. Dr. Abere stated, “if you limit yourself to three miles, you could be excluding important competitors and unfairly attributing effects to those firms that you do include in the model.” (DE 276 at 291).¹³ Other research also suggests that a three-mile radius may not be sufficient to capture all of Winn-Dixie’s competition. See, e.g., Edward J. Fox, et al., The Impact of Retail Location of Retailer Revenues: An Empirical Investigation (January 2007), available at efox.cox.smu.edu/personal/retaillocation.pdf. (“Proximity of stores to consumers is defined as the time it takes to travel from the consumer’s home to the retailer’s closest store.... [W]e specify travel times rather than distances because they more closely approximate travel costs. Unlike distances, travel times incorporate geographic factors that facilitate ... or hamper travel....”). See also Summon Datta & K. Sudhir, The Agglomeration-Differentiation Tradeoff in Spatial Location Choice 11 (June 2011), available at http://faculty.som.yale.edu/ ksudhir/workingpapers/Agglomeration% 20Differentiation% 20Tradeoff-June.pdf. (“A consumer who wants to buy, let’s say, groceries, may be attracted to a particular grocery store in location l if the location also consists of other commercial activities that cater to the consumer’s non-grocery needs (e.g. electronics and apparel stores).”)

4. DR. PACEY’S REPORTS ALLEGE DAMAGES FOR ITEMS THAT DEFENDANTS DO NOT SELL

As stated previously, supra Part III.B, Dr. Pacey’s Reports assume that Defendants compete with Plaintiffs in the form of substitutable goods. “If there’s two retailers selling the same products, they’ll have to share that market so each will have less than they would if the other retailer wasn’t there.” (DE 276 at 25). To calculate Winn-Dixie’s damages, Dr. Pacey’s Reports define Winn-Dixie’s non-meat grocery sales to include bakery, dairy, deli, floral, frozen food, grocery, package meat, produce, seafood, and tobacco. (Reports at 21). However, Dr. Pacey admitted at this Court’s Daubert hearing that none of the Dollar Tree stores at issue in this case sell alcohol, produce, seafood, or tobacco. (DE 276 at 202-03). She maintains that the reduced sales of such items should be included to calculate Winn-Dixie’s damages because “they’re part of the basket of goods that consumers would buy when they went into the store .... ” (Id. at 203). Again, Dr. Pacey provides no support for her “basket of goods” theory. Furthermore, while it is unclear what Dr. Pacey includes as “groceries” in her regression model, I previously determined that groceries includes only food items (DE 317 at 14 in Case No. 11-cv-80601). Therefore it is likely that Dr. Pacey’s model claims damages for goods that are not at issue in this case.

IV. CONCLUSION

Having reviewed the matter, I find that Dr. Pacey’s testimony in this matter would not assist the trier of fact and that her methodology is not sufficiently reliable. As detailed herein, Dr. Pacey’s Reports contain numerous omissions, miscalculations, and unsupported assumptions. Most importantly, Dr. 'Pacey’s analysis focuses on the impact of Defendants’ competition with Winn-Dixie without considering the main issue in this matter: the damages resulting from Defendants’ violations of Winn-Dixie’s grocery exclusives. “The burden of establishing qualification, reliability, and helpfulness rests on the proponent of the expert opinion.” United States v. Frazier, 387 F.3d 1244, 1260 (11th Cir. 2004). I find that Plaintiffs failed to meet this burden.

In their Post-Hearing Submission, Plaintiffs ask this Court to allow Dr. Pacey an opportunity to address any “fatal flaw” that I found in her regression model. (DE 313 at 38). While some of the flaws discussed in this Order could be addressed by Dr. Pacey, the critical issues of whether her Reports address the right issue and demonstrate causation cannot. When I asked Dr. Pacey at this Court’s Daubert hearing why the correct measurement is not the effect of the violation as opposed to competition, she stated “if you want to measure it that way, its not a regression measure. Then don’t use regression. Use some other avenue.” (DE 276 at 191). Accordingly, by Dr. Pacey’s own admission and this Court’s finding that Dr. Pacey should have focused on grocery sales that are not permitted under the grocery exclusives, a regression analysis is not appropriate for this case and no tweaking can fix that problem. Therefore, it is hereby

ORDERED AND ADJUDGED that Dolgencorp, LLC’s Motion to Strike and Exclude Proffered Expert Testimony of Patricia Pacey, Ph.D. (“Dr. Pacey”) Pursuant to Federal Rules of Evidence 702 and 703 (DE 56 in Case No. ll-cv-80601); Dollar Tree Store, Inc.’s Motion to Exclude Proffered Testimony of Patricia Pacey (DE 83 in Case No. ll-ev-80638), and Big Lots Stores, Inc.’s Motion to Strike and Exclude Proffered Expert Testimony of Patricia Pacey, Ph.D. Pursuant to Federal Rules of Evidence 702 and 703 (DE 131 in Case No. ll-cv-80641) are GRANTED.

ORDER DENYING MOTION FOR RECONSIDERATION AND CLARIFICATION OF ORDER CONCERNING EXPERT TESTIMONY OF DR. PATRICIA PACEY

THIS CAUSE comes before the Court upon Winn-Dixie’s Motion for Reconsideration and Clarification of Order Concerning Expert Testimony of Dr. Patricia Pacey (“Motion”) (DE 383) filed on April 23, 2012. In that Order, I found that Dr. Pacey’s testimony would not assist the trier of fact and that her methodology is not sufficiently reliable. (DE 358 at 23). As a result, I granted Defendants’ motions to strike and exclude her testimony. (Id. at 23-24). Winn-Dixie now asks that I: (1) clarify whether Dr. Pacey may testify regarding relevant economic issues, competition, and the effects of competition, without reliance upon or reference to the regression model; (2) reconsider my Order striking and excluding Dr. Pacey’s testimony at those store locations where no grocery sales are permitted; and (3) reconsider my Order regarding the perceived flaws in the variables and assumptions utilized by Dr. Pacey in the regression model and in regard to the Court’s refusal to allow Dr. Pacey the opportunity to modify the model and correct the perceived flaws. (DE 383 at 2-3). Having considered the matter, it is hereby

ORDERED AND ADJUDGED that Winn-Dixie’s Motion (DE 383) is DENIED. As I stated in my Omnibus Order Concerning Expert Testimony of Dr. Patricia Pacey (DE 358), Dr. Pacey may not testify at trial. Furthermore, reconsideration of my previous order is unwarranted. See Mannings v. School Bd. of Hillsborough County, Fla., 149 F.R.D. 235, 235 (M.D.Fla.1993) (“When issues have been carefully analyzed and decisions rendered, only a change in the law, or the facts upon which decision is based, will justify a reconsideration of the Court’s previous order.”).

1 . Dr. Pacey also issued updated reports for her Dollar General and Dollar Tree Reports. (DE 131-1 in Case No. 11-cv-80638, DE 182-2 in Case No. 11-cv-80601).

2 . This data includes Winn-Dixie store identification number, address/location, store type, square footage, age, dates of any remodeling events, and sales by major product category from July 2003 through April 2011. (Id. at 9-10).

3 . This data includes population density, per “P1* Percentage by gender, percent-by of education, percentage by race, and median age. (Id. at 10).

4 . This data aggregates the data by competitor type (Dollar General, Dollar Tree, Big Lots, other variety stores, supermarkets, other supermarkets, drug stores, and others) and distance (within 0.2 miles, greater than 0.2 miles but less than 1 mile, or greater than 1 mile but less than 3 miles from the nearest WinnDixie). (Id. at 11).

5 . This data includes Dollar General stores' identification numbers, address/location, square footage and square footage allocated to groceries, sales, and grocery sales. (Id. at 11-12).

6 . Non-meat groceries includes alcohol, bakery, deli, floral, frozen food, grocery, package meat, produce, seafood, and tobacco. (Reports at 21, 22). The items sold by WinnDixie excluded from its dependent variable are meat, general merchandise, pharmacy, photo, and fuel. (Id.).

7 . A similar update was issued for the Dollar Tree case. (DE 131-1 in Case No. 11-cv-80638).

8 . In Daubert, the Supreme Court considered the federal judge’s gatekeeping role in ensuring that all scientific expert testimony is not only relevant, but reliable. The Supreme Court later held that this basic gatekeeping obligation and Daubert’s general principles apply to all expert testimony, not just testimony that is classified as scientific. Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 147, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999).

9 . In fact, 82 of the 108 leases at issue in this case allow Defendants to sell 500 square feet of groceries, while only twelve leases prohibit tenants from selling any groceries. (DE 168 in Case No. ll-cv-80641 at 3-4).

10 . Perhaps indicative of Dr. Pacey's focus in her analysis, 29 of the 62 Dollar General stores that Winn-Dixie is claiming damages for are not included in her regression analysis because complete data was not available. (DE 182-2 at 6).

11 . In their Post-Hearing Submission, Plaintiffs state that “the sample size is too small and the results are not statistically significant.” (DE 313 at 30).

12 . Plaintiffs allege damages of $2,867,466 at Winn-Dixie Store # 1540, while co-located Dollar General Store #770 had only $2,781,276 of sales over that same period. (DE 301 at Ex. E). Plaintiffs allege damages of $4,748,424 at Winn-Dixie Store # 1537, while co-located Dollar General Store # 626 had only $3,974,294 of sales over that same period. (Id. at Ex. F).

13 . Dr. Abere noted that a related problem in Dr. Pacey’s regression model is that she groups all supermarkets together regardless of their size or what they sell. (DE 276 at 292). Since consumers may prefer larger supermarkets, or may shop at Publix, but not Whole Foods, the model’s failure to account for size and offerings may skew its results.