Opinion
16-10444
10-11-2023
In re Flint Water Cases. v. Snyder et al. This Relates To: Carthan, et al.
OPINION AND ORDER GRANTING DEFENDANTS VEOLIA NORTH AMERICA, LLC, VEOLIA NORTH AMERICA, INC., AND VEOLIA WATER NORTH AMERICA OPERATING SERVICES, LLC'S MOTION TO EXCLUDE THE OPINIONS AND TESTIMONY OF DR. ROBERT A. SIMONS [2462]
JUDITH E. LEVY UNITED STATES DISTRICT JUDGE
This opinion is yet another in a series of opinions addressing the admissibility of the testimony and reports of nine experts retained by Plaintiffs in anticipation of the issues class trial, set to begin on February 13, 2024. (ECF No. 2435.) Defendants argue that these experts cannot meet the standards set by Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993).
See ECF No. 2454 (VNA's motion to exclude opinions and testimony of Dr. Larry Russell); ECF No. 2455 (VNA's motion to exclude opinions and testimony of Dr. Clifford P. Weisel); ECF No. 2456 (VNA's motion to exclude testimony and reports of Robert A. Michaels); ECF No. 2458 (VNA's motion to exclude opinions and testimony of Dr. David Keiser); ECF No. 2459 (VNA's motion to exclude opinions and testimony of Dr. Daryn Reicherter); ECF No. 2460 (VNA's motion to exclude opinions and testimony of Dr. Paolo Gardoni); ECF No. 2461 (VNA's motion to exclude opinions and testimony of Dr. Howard Hu); ECF No. 2483 (VNA's motion to exclude opinions and testimony of Dr. Panagiotis (Panos) G. Georgopoulos); and ECF No. 2462 (VNA's motion to exclude opinions and testimony of Dr. Robert A. Simons) (in ECF Nos. 2606 and 2617 the Court inadvertently failed to include the motion related to Dr. Simons' testimony in this list).
Currently before the Court is the motion by Veolia North America, LLC, Veolia North America, Inc., and Veolia Water North America Operating Services, LLC (collectively “VNA”) to exclude the testimony and opinions of Dr. Robert A. Simons. (ECF No. 2462.) On September 13, 2023, a hearing was held and argument was heard. For the reasons set forth below, VNA's motion is granted.
I. Background
Dr. Simons is Professor of Urban Planning and Real Estate Development in the Maxine Goodman Levin College of Urban Affairs at Cleveland State University. (ECF No. 1208-95, PageID.36136.) He has a Ph.D. in City and Regional Planning, an M.S. in Economics, and an M.R.P. in City and Regional Planning from the University of North Carolina at Chapel Hill. (Id. at PageID.36155.) He received a B.A. in Anthropology from Colorado State University. (Id.) Dr. Simons has published a variety of books and peer-reviewed articles, including research “quantifying the effect of environmental contamination on property values.” (ECF No. 1208-95, PageID.36136.)
In this case, Plaintiffs retained Dr. Simons to opine on whether “the contaminated water conditions” could cause “harm to Flint businesses,” and he concluded that they “in fact did cause such harm.” (ECF No. 24625, PageID.80338.) On May 19, 2023, VNA filed this motion to exclude the opinions and testimony Dr. Simons. (ECF No. 2462.)
II. Dr. Simons' Reports
Dr. Simons submitted three expert reports, each of which employ different methodologies. VNA challenges the opinions expressed in all three.
A. Dr. Simons' Initial Report
Dr. Simons submitted his initial report (“Initial Report”) on June 29, 2020, in support of certification of a subclass of Flint business owners who suffered losses attributable to the Flint Water Crisis. (ECF No. 120895.)
In his Initial Report, Dr. Simons relies on the theory that Flint residents diverted money they would have spent on goods and services to ensure they had clean water. (ECF No. 1208-95, PageID.36137.) His view is that in periods of economic contraction, such as the one he hypothesizes occurred in Flint as a result of the Water Crisis, declines in spending will vary depending on whether goods and services are necessities, luxuries, or something in-between. (Id. at PageID.36137.) The idea that spending declines can relate to how necessary a good is-known as “income elasticity of demand”-is a theory rooted in scholarly literature. (Id. at PageID.36140-36141.)
In his Initial Report, Dr. Simons employed a four-step process to test this theory. His goal was to identify economic sectors that experienced declines in sales volume that could not be explained by population loss or by economic and wage trends that impacted communities like Flint. (Id. at PageID.36141.) Insofar as these losses could not be explained by these other factors, Dr. Simons attributes them to the Flint Water Crisis. (Id. at PageID.36140.)
First, he identified commercial subsectors of Flint's economy that would likely “be sensitive to reductions in consumer discretionary income.” (Id. at PageID.36139.) Dr. Simons found these sectors by looking at which sectors in Michigan had experienced wage declines during the “great recession” that started in 2008. (ECF No. 1208-95, PageID.36141.) He looked at this wage data as a “proxy” for businesses' total receipts, hoping to identify sectors that would be likely to experience losses in an economic contraction. (Id.)
Second, within sectors Dr. Simons viewed as likely to be sensitive to reductions in consumer income, he identified those that experienced a decline in wages relative to Grand Rapids and Saginaw, cities he viewed as “comparable” to Flint. (Id. at PageID.36141-36142.) At this step of Dr. Simons' analysis, wages continue to function as a proxy for gross receipts. (Id. at PageID.36142.) Insofar as wages in a sector of Flint failed to perform comparably to the corresponding sector of these two other cities, he viewed that worse performance as an “indicator” of “potential commercial enterprise revenue declines in Flint following discovery of the water crisis.” (Id.) Dr. Simons' logic here is that, given that these three cities face similar economic trends, any worse performance by an economic sector in Flint requires explanation and is, at least potentially, a sign that the sector was impacted by the Flint Water Crisis.
At the third step, Dr. Simons then looked at the sectors identified at step two and proceeded to identify subsectors where “individual enterprises typically experienced sales volume declines between 2014 and 2018.” (ECF No. 1208-95, PageID.36142.) To do so, Dr. Simons relied on a database called Reference USA, which provides data about individual businesses, including estimates of their sales volume. (Id.) On this basis, he “calculate[d] an approximate measure of the change in [a business's] sales volume between 2014 and 2018, and thus [identified] ‘typical' sales volume percentage changes” in economic subsectors. (Id.) He found that 36 subsectors experienced such declines. (Id.)
Fourth, he filtered out subsectors whose losses were attributable to population loss in Flint or otherwise had unclear losses. (Id. at PageID.36143.) He then calculated damages for businesses in these subsectors, as well as failed businesses that went out of business “since the beginning of the Flint water crisis.” (Id. at PageID.36145-36146.)
B. Dr. Simons' Supplemental Report and Rebuttal Report
On October 2, 2020, Dr. Simons filed a Supplemental Report (“Supplemental Report”) that modified his original methodology. (ECF No. 2462-4.) There, he begins by explaining his motivation to shift approaches as being grounded in the fact that his Initial Report did not examine restaurants' losses, because that report only detected a “relatively minor impact” on their revenue streams. (ECF No. 2462-4, PageID.80308.) Dr. Simons explains that this result is not intuitive, given that “a deterioration in the quality of the city's public water supply would be likely to lead to a reduction in consumer activity at such businesses, due to fear of adverse health effects.” (Id. at PageID.80307) Suggesting that the exclusion of restaurants might have been because his previous method hid losses in the restaurant sector, he developed a new approach using a comparison to Genesee County, which he then applied to other economic sectors besides restaurants. (Id.)
Dr. Simons' hypothesis in the Supplemental Report differs from his hypotheses related to diversion of money from certain goods and services, which formed the basis for his Initial Report. The Supplemental Report predicts that consumers will also redirect spending away from businesses in the City of Flint to avoid exposure to contaminated water. (Id.) Dr. Simons predicts substitution of some restaurants for others, including people eating outside Flint instead of within it. (Id.) His idea is, in part, that “Flint residents would migrate to restaurants in neighboring communities when they choose to eat out, because of their awareness that the water at those restaurants is safer than the water at restaurants in Flint.” (ECF No. 2462-4, PageID.80312 n.7.) His Rebuttal Report adopts this stance, too, suggesting that there will be a shift away from Flint establishments to those in Genesee County, especially in those where contact with water is more likely. (ECF No. 2462-5, PageID.80321.)
In the Rebuttal Report, Dr. Simons explains: “businesses were harmed not merely as a result of reduced discretionary spending by Flint residents due to higher costs for clean water, but also result from redirected discretionary spending (both by Flint residents and by residents of nearby, non-Flint communities) from businesses located in Flint to businesses located outside the city. These harms thus include a substitution effect related to the fairly widespread perception, among residents of Flint and other Genesee County communities, that Flint city water remained unsafe for human contact after discovery of the lead crisis.” (ECF No. 2462-5, PageID.80322.)
In part because the Initial Report did not find a significant impact on restaurants in Flint, Dr. Simons begins by comparing Reference USA data from restaurants in Flint to restaurants in the rest of Genesee County, the county where Flint is located. (ECF No. 2462-4, PageID.80308.) Using Genesee County as a “control group,” Dr. Simons aims to “highlight local differences within the same regional economy” through the comparison of Flint to the surrounding County. (Id. at PageID.80309.) He categorized different types of restaurants by type based on the rates at which they closed between 2014 and 2018 and the percentage of them that had a decline in annual sales volume during that period. (ECF No. 2462-4, PageID.80310.) He found a “statistically significant greater percentage of ‘failed' restaurants in Flint than in the rest of Genesee County” but did not find significant differences “between Flint and the rest of Genesee County in the percentage of ‘surviving' businesses whose annual sales volume appears to have declined from 2014 to 2018.” (Id. at PageID.80308) Within the group of surviving restaurants, he then calculated the amount of profit lost by those restaurants in Flint that experienced a loss. (Id. at PageID.80309.) Dr. Simons then applied this same comparative methodology to the subsectors he studied in his Initial Report. (Id.)
A third report (“Rebuttal Report”), filed March 1, 2023, responded to VNA's expert. (ECF No. 2462-5.) In addition to directly answering VNA's expert's criticism, Dr. Simons presented four models that incorporated VNA's expert's critiques to “demonstrate that regardless of the various calculation assumptions used, the existence of substantial amounts of class-wide damages sustained by similarly affected subclass members across numerous business subsectors is confirmed, and those damages remain capable of calculation in a formulaic manner.” (Id. at PageID.80334.)
III. Legal Standard
Federal Rule of Evidence 702 governs the admissibility of expert testimony and requires that: (1) the witness must be qualified, (2) the testimony must be relevant, and (3) the testimony must be reliable. Fed.R.Evid. 702; In re Scrap Metal Antitrust Litig., 527 F.3d 517, 528-29 (6th Cir. 2008). As the Supreme Court explained in Daubert, Rule 702 imposes a “gatekeeping” obligation on the courts to ensure that scientific testimony “is not only relevant, but reliable.” Daubert, 509 U.S. at 589; See also Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 147 (1999).
“Rejection of expert testimony is the exception, rather than the rule.” United States v. LaVictor, 848 F.3d 428, 442 (6th Cir. 2017) (quoting In re Scrap Metal, 527 F.3d at 529-30)). But the burden is on Plaintiffs to show by a “preponderance of proof” that the proffered expert meets the standards of Rule 702 as interpreted by Daubert. Pride v. BIC Corp., 218 F.3d 566, 578 (6th Cir. 2000) (quoting Daubert, 509 U.S. at 592).
IV. Analysis
VNA argues that (1) Dr. Simons' opinions quantifying business losses are irrelevant to the issues trial; (2) his quantitative opinions are unreliable; and (3) his economic theories about how contaminated water harmed Flint businesses are unreliable. (ECF No. 2462, PageID.80209.)
Because the Court holds that his quantitative opinions and economic theories are unreliable, there is no need to reach VNA's arguments about relevance, though the Court's earlier opinion related to Dr. David Keiser has already addressed similar arguments and rejected them. (ECF No. 2617, PageID.85801.)
A. The Reliability of Dr. Simons' Quantitative Opinions
VNA asserts that Dr. Simons' quantitative opinions should be excluded under Rule 702, because “they are based on unreliable control groups, unreliable data, and an unreliable methodology.” (ECF No. 2462, PageID.80219.) Additionally, VNA argues that he fails to account for alternative explanations for his results.
i. The Reliability of Dr. Simons' Control Groups
VNA critiques the control groups Dr. Simons chose for his reports, both his comparison to Grand Rapids and Saginaw, and to Genesee County. Because Dr. Simons chose these control groups without sufficient justification, they do not support the conclusion that the differences he found between them and Flint are attributable to the contaminated water, VNA argues. (ECF No. 2462, PageID.80222.)
While challenges to the selection of control groups often are not enough to undermine an expert's reliability, if choices about control groups mean that testimony does not rest on a “reliable foundation,” then it must not be admitted. Daubert, 509 U.S. at 597. Plaintiffs cite several cases where courts treated experts' choices about control groups as impacting the weight of testimony rather than its reliability. (ECF No. 2508, PageID.83122.) One case finds that a survey without a control group is not so flawed as to render it unreliable but emphasizes that the survey was not intended to show causation, in contrast to the testimony offered by Dr. Simons. Zarinebaf v. Champion Petfoods USA Inc., No. 18 C 6951, 2022 WL 910638, at *3 (N.D.Ill. Mar. 29, 2022). Another case Plaintiffs cite also addresses surveys specifically. Takeguma v. Freedom of Expression LLC, No. CV-18-02552, 2021 WL 487884, at *3 (D. Ariz. Feb. 10, 2021). In a third case, the court treated objections related to control groups in a study as involving issues of weight of the evidence and not its admissibility. Phoenix Light SF Ltd. v. Bank of N.Y. Mellon, No. 14-CV-10104, 2020 WL 1322856, at *10 (S.D.N.Y. Mar. 20, 2020). In that case, however, the objections to the control group relied on analysis that was less precise than the challenged expert's selections, and the opposing party's expert performed analysis that showed they believed the model at issue was reliable. Id. In a fourth case, the court found control group issues solely impacted the testimony's weight but did so in a context where the expert's opinion was “premised on relevant medical literature.” Mathews v. Novartis Pharms. Corp., No. 3: 12-cv-314, 2013 WL 5780415, at *18 (S.D. Ohio Oct. 25, 2013). Here, the testimony, which is offered to prove general causation, is not survey-based. Additionally, these cases do not support a presumption that objections related to control groups speak only to the weight of Dr. Simons' testimony. His choice of controls is imprecise and his grounding of those choices in the relevant literature is insufficient. Ultimately, “any step that renders the analysis unreliable under the Daubert factors renders the expert's testimony inadmissible.” Little Hocking Water Ass'n, Inc. v. E.I. du Pont de Nemours and Co., 90 F.Supp.3d 746, 760 (S.D. Ohio 2015) (quoting In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 745 (3d Cir. 1994)). As a result, the Court must assess whether Dr. Simons' choices related to control groups provide a reliable foundation for his opinions.
In Dr. Simons' Initial Report, he uses the metropolitan statistical areas (“MSA”) of Grand Rapids and Saginaw as comparators. He calls them “comparable communities” to Flint. (ECF No. 1208-95, PageID.36141.) He offers no further explanation there for why he chose these cities in his Initial Report. Plaintiffs suggest that these cities are not traditional controls, because Dr. Simons is using them to identify business subsectors in Flint for further analysis, not to directly compare revenues between the cities to determine losses, if any, in Flint. (ECF No. 2508, PageID.83119.) Dr. Simons makes a similar argument in his Rebuttal Report. (ECF No. 2462-5, PageID.80329.) In his deposition, Dr. Simons speaks in terms of controls, however, so the Court will refer to these cities as such while recognizing that there may be a substantive difference from traditional control groups. For instance, Dr. Simons explains that he chose these “two controls,” Grand Rapids and Saginaw, on the following basis:
Both in Michigan. Both similar sizes. I had East Lansing in there for a while, but it was the state capital. I thought that was a pretty important factor to take out. And I thought that
Saginaw is pretty close to Flint and maybe not economically stronger. And Grand Rapids, other side of the state, maybe a little economically stronger. Both about the same population size, same order of magnitude. So I thought two controls was good enough.(ECF No. 2462-3, PageID.8026.) Dr. Simons emphasizes that he was concerned about “the inherent differences in state-to-state business environments.” (ECF No. 2462-5, PageID.80329.) The reason why it was imperative to do an in-state comparison is not further explained or supported with a citation in his Rebuttal Report. Dr. Simons' notion that economic environments differ from state to state is plausible, but he does nothing to explain the significance of this factor relative to other factors (such as economic trends, population trends, common industries, and so on).
Dr. Simons likely misspoke and intended to refer to Lansing, which is the state capital of Michigan. In his Rebuttal Report, he references Lansing rather than East Lansing. (ECF No. 2462-5, PageID.80329.)
As he acknowledges, the choice to focus his comparisons within the state left him with limited options. VNA criticizes how Dr. Simons handled those options, specifically by not following Dr. Keiser, another of Plaintiffs' general causation experts, in comparing Flint to Pontiac. (ECF No. 2462, PageID.80220.) The fact that two experts studying different issues choose different controls is not concerning on its own. What is problematic is the apparently arbitrary choices Dr. Simons made in his Initial Report, which were not validated in any way. Dr. Simons considers the fact that Grand Rapids and Saginaw are both in Michigan, are not the state capitol, had populations that bracketed Flint (meaning Grand Rapids is bigger and Saginaw is smaller, presumably), and were not suffering from contaminated water. (ECF No. 2462-3, PageID.80263, 80265; ECF No. 2462-5, PageID.80329.) It is unclear whether he considered population trends in these cities, and he did not consider the relative economic trends in these three cities, which are relevant to the type of opinion Dr. Simons is offering. (ECF No. 2462-3, PageID.80266.) As VNA points out, Grand Rapids is, on its face, very different from Flint. (ECF No. 2462, PageID.80220-80221 n.4.)
Dr. Simons provides insufficient justifications for the weight he gave to different factors in choosing controls. He does not explain, for instance, why Lansing being the seat of government is more of a concern than Grand Rapids being a much larger city with an expanding economy. Given the obvious, relevant differences between the comparators and a lack of a compelling justification for the choice of these comparators, these controls create reliability concerns sufficient to justify exclusion of the quantitative opinions in the Initial Report.
VNA also raises concerns about Dr. Simons' reliance on Genesee County as a control in his later reports. Dr. Simons' aim is to test whether “substitutionary spending changes by both city and county residents, combined with changes in available discretionary income for city residents, led to business losses for certain subsectors in Flint, with a theorized expectation of associated improvement in revenue patterns for similar nearby businesses in the county.” (ECF No. 2462-5, PageID.80330.) Dr. Simons says that it is “intuitively correct” that Genesee County is a good control. (ECF No. 2462-3, PageID.80302.) He indicates that “using an area that's closely proximate as a control group is standard operating practice.” (Id.)
VNA argues that because Genesee County is affected by the variable being tested (“redirected spending caused by the contaminated water”), it is an invalid control. (ECF No. 2462, PageID.80221.) Plaintiffs respond that relying on Genesee County as a control holds “changes in the economic base constant,” so that if a factory closes or an economic development project happens, such changes will impact Genesee County and Flint equally. (ECF No. 2508, PageID.83121.) The idea is to eliminate such local economic shifts as a confounding variable. Plaintiffs acknowledge, though, that there is “a trade-off between the need to hold the economic base constant in a comparison and having some degree of economic crossover between the city and the rest of the county.” (ECF No. 2508, PageID.83121.) Dr. Simons does not explain how he weighed this trade-off and why it is valid to think that one factor outweighs the other. Troublingly, he also testifies in his deposition that he did not analyze differences between the economic base of the City of Flint and Genesee County in detail, noting only that they both have shopping and services, in addition to being part of the same MSA where there is likely to be commuting. (ECF No. 2462-6, PageID.80352-80353.) This analytic gap is especially concerning, given the apparent differences between Genesee County and the City of Flint. (ECF No. 2462, PageID.80221.) Plaintiffs' expert, Dr. Keiser, specifically enumerates the many ways Genesee County and Flint differ: size of cities, population trends, employment rates, racial demographics, wealth, economic growth, as well as other factors. (ECF No. 1208-128, PageID.37518-37519.) As noted above, Dr. Keiser addresses a different issue, whether the Flint Water Crisis had an impact on the value of homes in Flint. Nonetheless, these differences between Genesee County and Flint are relevant for the analysis Dr. Simons offers on the impact of the crisis on businesses. Dr. Simons need not take the same approach as Dr. Keiser, but his failure to consider or address these differences in choosing a control group raises serious reliability concerns, insofar as his choice is not “supported by appropriate validation-i.e., ‘good grounds,' based on what is known.” Daubert, 509 U.S. at 590. That reliability issue is sufficient to warrant exclusion of his quantitative opinions in these later two reports (the Supplemental Report and Rebuttal Report).
All of Dr. Simons' reports rely heavily on comparators or control groups to support his quantitative conclusions. However, he does not ground his choices regarding these controls in sufficient justifications or in peer-reviewed literature. Appeals to intuition are not enough when there are well-founded concerns about the comparators chosen as the key foundation for an opinion. Accordingly, the reliability concerns prompted by these choices of controls render Dr. Simons' quantitative opinions inadmissible.
ii. The Reliability of Dr. Simons' Data
Further, VNA argues that because Dr. Simons' data are not reliable, his quantitative opinions are unreliable as well. (ECF No. 2462, PageID.80223.) Specifically, VNA raises questions about Reference USA, the resource Dr. Simons uses to gather enterprise-level data about businesses in Flint. (ECF No. 1208-95, PageID.36142.) Reference USA is “library-based, online database of U.S. businesses, maintained by Infogroup.com,” which includes data from the period at issue in this case. (ECF No. 1208-95, PageID.36142 n.14.)
First, VNA argues that Reference USA's data is “opaque,” because it is proprietary and therefore it is unclear how it was generated. (ECF No. 2462, PageID.80223.) Additionally, VNA argues that Dr. Simons cannot explain how Reference USA arrives at their numbers for business data. (Id.) These objections on their own do not suggest that this data is unreliable, only that it is hard to assess its reliability. This data is not based in groundless assumptions or data generated by the Plaintiffs themselves without any independent verification. See Bruno v. Bozzuto's, Inc., 311 F.R.D. 124, 137 (M.D. Penn. 2015). Rather, Reference USA is an independent source. This objection about opacity speaks to the testimony's weight and not its reliability.
Second, VNA argues that Reference USA is not used by Dr. Simons in his other work, nor is it used by other economists. To the extent that this objection is relevant under Rule 702, VNA would have to show that Dr. Simons used unreliable data that experts would normally reject. Experts should rely on the facts or data upon “which experts in the field would reasonably rely.” Gopalratnam v. Hewlett-Packard Co., 877 F.3d 771, 780 (7th Cir. 2017). In response, Plaintiffs note several law review articles that use Reference USA. (ECF No. 2508, PageID.83123.) One of these articles uses this data source for business locations and not gross receipts data. Colleen V. Chien & Michael Risch, Recalibrating Patent Venue, 77 Md. L. Rev. 47 (2017). Another one, Predatory Lending and the Military, uses Reference USA to identify the number of businesses engaged in payday lending. Steven M. Graves & Christopher L. Peterson, Predatory Lending and the Military: The Law and Geography of “Payday” Loans in Military Towns, 66 Ohio St. L.J. 653 (2005). A third article used the database to identify businesses' year of founding. Colleen V. Chien, Software Patents as a Currency, Not Tax, on Innovation, 31 Berkeley Tech. L.J. 1669 (2016). These are not precisely the same type of data that Dr. Simons retrieved from Reference USA for his research. VNA argues that this renders them irrelevant as support for Dr. Simons' approach. (ECF No. 2528, PageID.83871.) The first article refers to Reference USA as a widely used scholarly resource, however. Chien & Risch, Recalibrating Patent Venue, supra, at 81. Dr. Simons also points to other scholars' use of the database. (ECF No. 2462-5, PageID.80330.) This resource appears to be the sort of tool that scholars reasonably rely upon.
Third, VNA argues that there are systematic problems with the data in Reference USA. VNA points to issues like identical estimates for different businesses, the use of ranges rather than actual revenue figures, and potential issues with accuracy. (ECF No. 2462, PageID.80223-80224.) They also challenge the reliability of Reference USA as a means for identifying business failures, relying on the findings of their own expert. (ECF No. 2462, PageID.80225.) It is not clear that any such issues with data inputs create systematic unreliability sufficient to exclude Dr. Simon's opinion, though. Courts exclude data when it is unverified and comes from a particularly suspect source such as the litigant who has hired the expert. Dreyer v. Ryder Auto. Carrier Grp., Inc., 367 F.Supp.2d 413, 446 (W.D.N.Y. 2005). Here, Reference USA is an independent resource, one that VNA is free to subject to scrutiny via cross-examination. At this stage, however, the concerns about data inputs raised by VNA speak more to weight than admissibility. Choosing which expert's view of data inputs is correct is appropriate for a fact finder to adjudicate; it should not be resolved at this stage absent strong reasons to suspect the fundamental validity of a data resource.
iii. The Reliability of Dr. Simons' Methodology
VNA further challenges Dr. Simons' methodology, arguing that it is “results-oriented.” (ECF No. 2462, PageID.80226.) His approach, they argue, is so questionable that it should be excluded as unreliable.
VNA objects first to the methodology in Dr. Simons' Initial Report. They note that at Step 1 of his process, he predicted that 23 sectors would be likely to experience losses in Flint due to the contaminated water, but only found losses in 11 sectors at Step 2. (Id.) Rather than questioning his hypothesis due to less than half of the sectors experiencing worse performance, VNA suggests Dr. Simons assumes that the 11 sectors experienced losses. (Id.) In response, Plaintiffs argue that Dr. Simons did not assume these 11 sectors in the Flint MSA, which performed worse than corresponding sectors in the Saginaw and Grand Rapids MSAs, experienced losses-only that they potentially did. (ECF No. 2508, PageID.83126.) That conclusion comes later, after steps 3 and 4, where he determines whether there was loss, having considered alternative factors “like discount rates and population decline.” (Id.)
It is true that Dr. Simons' methodology did not guarantee any result and could have found that no sector in the Flint MSA experienced a loss, though that does not mean it was reliable. Every sector in Flint might have performed the same or better than the corresponding sectors in Grand Rapids and Saginaw. But if any one sector performed worse than the controls, Dr. Simons concluded sales declines found in those sectors were due to water contamination (unless the decline was unclear, including potentially because it was explicable due to population decline). (ECF No. 1208-95, PageID.36143.)
Even though Dr. Simons does not ignore the need to exclude some alternative causal factors, he falters at an earlier stage of his analysis when he fails to account for results that do not align with his hypothesis. VNA's point is that given that less than half of the Flint MSA's sectors performed worse than the control group, that raises a question about whether Dr. Simons' underlying hypothesis (that water contamination explained the difference) is valid. Dr. Simons' methodology builds in a presumption that differential results found at Step 2 likely indicate that any losses detected in a sector at Step 3 are caused by water contamination, so long as those losses are not too unclear or small. The support for that presumption is unclear, particularly given the finding of worse performance in only 11 out of 23 sectors at Step 2. The problem is not that Dr. Simons' methodology could not fail, as VNA argues, but that it puts a thumb on the scale in favor of finding losses attributable to the water crisis. (ECF No. 2528, PageID.83873.)
To assess reliability, Daubert provides a “flexible” set of factors to consider when assessing a method: whether it can be (and has been) tested, whether it has been subject to peer review, whether there are known error rates and standards controlling a method's operation, and general acceptance of the approach. 509 U.S. at 593-94. Plaintiffs do not indicate whether Dr. Simons' filtering method has been tested or subjected to peer review, nor is there evidence about whether it is an accepted approach. Dr. Simons develops a “model” in his Initial Report, which he says involves a “logical, sequential process.” (ECF No. 1208-95, PageID.36141.) It is unclear whether such a process is widely used or whether it was developed specifically for the Initial Report. Dr. Simons does, however, include an estimate of error rates in his Initial Report. (ECF No. 1208-95, PageID.36143.) Given the general issues with Dr. Simons' methodology set forth above, in addition to not fulfilling many of the key Daubert factors, Plaintiffs have not shown the methodology in the Initial Report to be reliable.
VNA also objects to the methodology in Dr. Simons' later reports, making a similar argument about presuming the truth of the hypothesis Dr. Simons is testing. They note that he “found that there was a statistically significant underperformance by Flint businesses in only 4 out of 26 subsectors for business failure rate and only 2 out of 26 subsectors for revenue.” (ECF No. 2462, PageID.80226.) Again, this small percentage of subsectors that experienced underperformance would seem to raise questions about whether Flint had indeed experienced unexplained underperformance relative to Genesee County. Dr. Simons responds that the lack of statistical significance is likely due to small sample sizes and notes that the differences between Flint and Genesee County were statistically significant in the aggregate. (ECF No. 2462-5, PageID.80326.)
Here, the problems are less clear than with the Initial Report. Accordingly, the methodology Dr. Simons uses in his later reports (the Supplemental Report and Rebuttal Report) is sufficiently reliable. Nonetheless, his quantitative opinions must still be excluded due to the problems with the control group set forth above.
iv. Ruling Out Alternative Explanations
VNA further argues that Dr. Simons fails to rule out alternative factors that could also explain his results. (ECF No. 2462, PageID.80227.) They point to precedent suggesting that “failure to rule out alternative causes” is a “red flag.” Newell Rubbermaid, Inc. v. Raymond Corp., 676 F.3d 521, 527 (6th Cir. 2012). Plaintiffs rightly note that experts need not eliminate all other causes, however. See, e.g., Jahn v. Equine Servs., PSC, 233 F.3d 382 (6th Cir. 2000). And Dr. Simons clearly considers issues like regional variation and population loss and incorporates them into his methodology. (See, e.g., ECF No. 1208-95, PageID.36141-36143.) At times, Dr. Simons does not take the need to eliminate alternate explanations seriously enough, calling such concerns a “smokescreen” and insisting that the Water Crisis is the “elephant in the room.” (ECF No. 2462-5, PageID.80327.) As an expert, he cannot simply take such a point as obvious. His methodology does, however, make attempts to address alternative factors, despite the rhetoric he uses elsewhere. Any limitations in his attempts to eliminate alternative factors are issues that impact the weight of his testimony and not its admissibility.
As set forth above, however, his quantitative opinions are nonetheless inadmissible, because his reports depend on unreliable control groups and his Initial Report also has an unreliable methodology.
B. The Reliability of Dr. Simons' Theories
Beyond Dr. Simons' quantitative opinions, VNA takes issue with the theories of general causation he offers, arguing for the exclusion of his testimony more broadly.
i. Identifying Dr. Simons' Theories
The litigants do not agree on what theory Dr. Simons is offering, however. VNA asserts that Dr. Simons offers two hypotheses: (1) the aversion-cost theory, the view that Flint residents spent what would have been discretionary income on water-related costs, which, if true would result in less spending at businesses offering discretionary goods and services; and (2) the redirected-spending theory, that spending on discretionary goods and services was directed outside of the City of Flint due to consumers' perception that Flint water was unsafe.
Plaintiffs respond that Dr. Simons' underlying theory is something more fundamental: “the income elasticity of demand-the idea that an economic contraction will affect the sale of certain goods or services based on their respective degree of necessity such that certain sectors are more sensitive than others to possible economic downturns.” (ECF No. 2508, PageID.83131.) They suggest that VNA's depiction of Dr. Simons' work is too narrow, and that he instead relies on general knowledge about economic contractions and uses his comparative methodology and other analytic tools to determine whether there were losses. (Id.)
The Court finds that VNA's delineation of these theories is useful for analyzing the admissibility of Dr. Simons' opinions. As a result, the analysis below speaks in terms of the aversion-cost theory and the redirected-spending theory, though it also considers income elasticity of demand. VNA attacks both the aversion-cost theory and the redirectedspending theory, but they do not attack Dr. Simons' more general reliance on the income elasticity of demand. The question before the Court is whether his two more specific theories are reliable.
ii. The Aversion-Cost Theory
VNA contends that Dr. Simons' theory that Flint residents had to “divert funds from their normal spending patterns to ensure that” they had clean water, causing losses to businesses that provide non-discretionary goods and services, is not reliable. (ECF No. 1208-95, PageID.36137.)
In objecting to the aversion-cost theory, VNA emphasizes that part of the diversion of residents' discretionary spending would likely be bottled water or water filters, but that there was less spending on bottled water (because it was provided for free), according to one of Plaintiffs' other experts, Dr. Keiser. (ECF No. 2462, PageID.80230.) In his deposition, Dr. Simons does suggest that bottled water might be one place where spending was diverted, in addition to water filters and other costs, though he acknowledges not having studied that issue. (ECF No. 2462-3, PageID.80258-80259.) He also suggests the need to purchase bottled water is something that decreased consumers' discretionary income. (ECF No. 2462-4, PageID.80307.) In his Rebuttal Report, he concedes that bottled water is not a large factor in Flint businesses' revenue declines. (ECF No. 2462-5, PageID.80322.) On its own, this point does not undermine the reliability of Dr. Simons' theory, even if it is contradicted by the testimony of a different expert offered by Plaintiffs. It simply shows that Dr. Simons' findings about specific ways that consumers might have diverted their spending are likely incorrect, not that his overall theory that consumers diverted their spending in ways that impacted businesses is clearly unreliable.
More worrisome for this theory is that it is dependent on the proposition that there was an economic contraction in some sectors in Flint, due to water contamination that caused diverted spending. (ECF No. 2508, PageID.83112.) Dr. Simons has not presented support for that proposition. The Court has already held that Dr. Simons' methodology, which is the basis of his finding that there were contractions in some sectors is unreliable, at least as a basis for his quantitative opinions. Yet that is how Dr. Simons attempts to show that economic contractions occurred. For his theory to be reliable, Dr. Simons would need some other way to show that the contaminated water can cause such contractions with resultant losses for Flint businesses. He does not, for instance, offer secondary literature showing water contamination or other crises like the one in Flint have a tendency to cause economic contractions and business losses. Dr. Simons instead provides references to research that indicates there is likely to be economic loss in some sectors during negative economic shocks. (ECF No. 1208-95, PageID.36140-36141.) The question, though, is whether such a shock occurred here because of the contaminated water. He briefly cites to his own work outlining certain criteria for when environmental contamination can cause losses, but his application of these criteria here depends upon the methodology already deemed unreliable. (ECF No. 2462-5, PageID.80327-80328.) Accordingly, because Dr. Simons does not show that there was an economic contraction or that an event like the Water Crisis was likely to cause such a contraction in certain sectors of Flint's economy, his theory that Flint residents diverted their discretionary spending in a manner that harmed local businesses is not reliable and must be excluded.
iii. The Redirected-Spending Theory
VNA critiques as unreliable Dr. Simons' theory that stigma related to Flint's water would lead consumers to direct their spending outside Flint and, resultantly, harm Flint businesses. (ECF No. 2462, PageID.80230.) Central to their objection is the notion that Dr. Simons' results undermine his theory. Dr. Simons' general prediction was that:
Such substitutionary decisions would be likely to affect businesses in which customers could anticipate coming into contact with contaminated water conditions directly (e.g., eating and drinking establishments, hotels, hair salons, selfserve car washes, pet grooming businesses); indirectly through contact with goods which had been exposed to contaminated water (e.g., laundry services); and businesses where the amount of time spent shopping might result in the need to use the facilities' public restrooms.(ECF No. 2462-5, PageID.80321.) VNA points out that several of the sectors predicted to experience losses did not experience such losses while others had losses despite not seeming to be related to water. (ECF No. 2462, PageID.80231-80232.)
For instance, VNA notes: “Dr. Simons's rebuttal report mentions six types of businesses for which he predicts losses. He did not even look at two of those categories; he found no losses in two others; and he had inconsistent results in another (the ‘eating and drinking establishments' subsector).” (ECF No. 2462, PageID.80232 n.11.) They also offer a “conservative” list of sectors that Dr. Simons lists in Model D of his Rebuttal Report as experiencing losses but that do not involve contact with water: “Cosmetics, Beauty Supplies & Perfume Stores; Insurance Agencies & Brokerages; Claims Adjusting; Hobby, Toy & Game Stores; Musical Instrument & Supplies Stores; Book Stores; Agents & Managers For Public Figures; Paint & Wallpaper Stores; Other Building Material Dealers: Glass/Auto Glass; Outdoor Power Equipment Stores; Photofinishing Laboratories (Except One-Hour); and Electronic Stores.” (Id. at PageID.80232 n.12.)
Plaintiffs respond that “Defendants' arguments that Dr. Simons' theory is supposedly not in line with his ‘results' is just a disagreement with his conclusions rather than a substantive critique of his methodology.” (ECF No. 2508, PageID.83132.) Plaintiffs are correct that mere disagreement with results is not a basis for exclusion; reliability is about the principles and methodology underlying an expert opinion. Daubert, 509 U.S. at 595. VNA's argument, however, is that Dr. Simons' results do not match his own theory, which is a not a claim that his results are implausible or objectionable, but that there is a lack of validation of his conclusions. The argument is that he makes a prediction in his theory aimed at explaining why the losses he detects are attributable to the contaminated water. (ECF No. 2462-5, PageID.80321.) However, when that prediction is not borne out, he does not reexamine his premises or otherwise explain why it is legitimate to conclude that the losses he finds are due to the contaminated water. Rather, he concludes that his methodology is “consistent with the theoretical expectations regarding the likelihood of such substitutionary purchasing and patronage behavior in response to environmental contamination, such as the contaminated water conditions in Flint discovered in 2014.” (ECF No. 2462-5, PageID.80337.) Part of the requirement that an expert “reliably appl[y] the principles and methods to the facts of the case” is the need to adjust one's theory or hypothesis based on the results one discovers, insofar as those results do not support one's hypothesis. Fed.R.Evid. 702. Daubert depicts science as “a process for proposing and refining theoretical explanations about the world.” 509 U.S. at 590. Here, Dr. Simons fails to adequately refine his theoretical explanation in response to the data he analyzes, which speaks to his methodology. Accordingly, his redirected-spending theory is not admissible.
iv. Income Elasticity of Demand
The Court has already noted that VNA does not challenge the general economic principle of income elasticity of demand, “the idea that an economic contraction will affect the sale of certain goods or services based on their respective degree of necessity such that certain sectors are more sensitive than others to possible economic downturns.” (ECF No. 2508, PageID.83131.) Nonetheless, Dr. Simons cannot testify to this principle, because its relevance to this case is conditioned on showing that there was an economic contraction due to the contaminated water. Fed.R.Evid. 104(b). As set forth above, he has not done that. As a result, there is also a concern that any such testimony would be prejudicial or confusing, insofar as it would encourage the fact finder to assume such a contraction caused by contaminated water had occurred. Fed.R.Evid. 403. Therefore, Dr. Simons may not testify about the economic principle of income elasticity of demand.
In reaching this conclusion, the Court is not weighing in on the truth or falsity of this principle, just as it is not drawing any conclusions about the truth of whether businesses experienced losses due to contaminated water in Flint. The Court's role here is not to evaluate experts' conclusions, but to assess whether their methodology is reliable and, ultimately, whether their testimony is admissible. Fed.R.Evid. 702. Even if the Flint Water Crisis is the “elephant in the room,” as Dr. Simons puts it, the Court cannot take for granted that the Crisis caused an economic contraction that harmed businesses in Flint. (ECF No. 2462-5, PageID.80327.) The burden, rather, lies with the party offering an expert's testimony, to show by a “preponderance of proof” that an expert's opinions and testimony are admissible. Pride, 218 F.3d. at 578 (quoting Daubert, 509 U.S. at 592). The general conclusion that businesses suffered economic losses due to contaminated water may well be true. All the Court holds here is that Dr. Simons has failed to offer a reliable scientific basis to draw such a conclusion, and, under the applicable rules, his opinions and testimony are therefore inadmissible.
V. Conclusion
For the reasons set forth above, VNA's motion to exclude Dr. Simon's opinions and testimony is granted.
IT IS SO ORDERED.