Economics Expert's Industry Benchmark Opinion Admitted

The Equal Employment Opportunity Commission brought this action against Gypsum Express, Ltd., alleging that Gypsum "has engaged, and continues to engage, in a pattern or practice of hiring discrimination because of sex" since at least 2014, in violation of federal law.

Specifically, the EEOC alleged that Gypsum, a regional transportation-for-hire company with a focus on flatbed freight transportation, "has hired female drivers into flatbed positions at statistically significant lower rates than male drivers." The EEOC further alleged that such disparity resulted from Gypsum's formal policy requiring inexperienced drivers to be paired with an experienced driver of the same sex. More specifically, the EEOC contended that because Gypsum did not employ any experienced female drivers, the ultimate result of that policy was that Gypsum recruiters were precluded from hiring qualified female applicants for flatbed driver roles.

Plaintiff Equal Employment Opportunity Commission, filed a motion to limit the testimony of Defendant Gypsum Express's proffered expert, Dr. Chen Song. In return, Defendant Gypsum Express sought to exclude the testimony of the Plaintiff's proffered expert, Dr. Erich Cromwell.

Economics Expert Witnesses

Chen Song, Ph.D., CFA, holds a Ph.D. in economics from the University of Chicago. Since receiving her Ph.D. in economics from the University of Chicago, she has provided consulting and expert testimony in single-plaintiff, multiple-plaintiff, and class action discrimination disputes.

Get the full story on challenges to Chen Song’s expert opinions and testimony with an in-depth Challenge Study.

Erich Cromwell graduated with a B.A. in Communication from American University and received a B.S., M.S., and Ph.D. in Economics from Florida State University.

He used to be employed by the U.S. Equal Employment Opportunity Commission and served as an Economist in the Research and Analytic Services unit of the Office of General Counsel.

Want to know more about the challenges Erich Cromwell has faced? Get the full details with our Challenge Study report.

Discussion by the Court

The parties submitted competing expert reports concerning Gypsum's flatbed driver hiring rates. Erich Cromwell, an in-house labor economist employed with the EEOC, prepared an initial report on behalf of the EEOC analyzing Gypsum's hiring practices from March 25, 2014, to December 13, 2023. Cromwell's report indicates that the results still showed a statistically significant lower hiring rate for female applicants at 8.20 standard deviations.

Gypsum then proffered a report by its retained expert, Dr. Chen Song, in response to Dr. Cromwell's report. Song concluded that Gypsum's rate of female hires has increased in recent years. Song further concluded that in 2021, the female hiring rate exceeded the male hiring rate and used that data to opine that "the historical lower female hiring rate to Gypsum's flatbed positions no longer existed in 2021 when the female hiring rate exceeded the male hiring rate." It should be noted that Song developed her findings from a paper titled, "Driving in a man's world: examining gender disparity in the trucking industry," which was authored by Alex Scott and Beth Davis-Sramek and published in the International Journal of Physical Distribution & Logistics Management in 2023 (the "Scott/Davis-Sramek Study").

A. Dr. Chen Song's Testimony

The EEOC sought to limit two specific portions of Chen Song's testimony. First, the EEOC sought exclusion of the "industry benchmark" figure, which Song compared to Gypsum's hiring rate and concluded that Gypsum's female hiring rate tracked that of the trucking industry at large. The EEOC contended that the industry benchmark has no relevance because Gypsum's actual applicant data is available. The EEOC further took issue with the single study Song relied upon to calculate the benchmark. Second, the EEOC sought exclusion of Song's "hiring rate opinion" that in 2021, claims Gypsum's female hiring rate exceeded the male hiring rate. The EEOC argued that Song did not base this opinion upon reliable data or methods.

1. Industry Benchmark

I. The EEOC asked the Court to exclude Song's industry benchmark opinion

Song utilized the Scott/Davis-Sramek Study to infer the benchmark range. To estimate the proportion of female truck drivers, the Scott/Davis-Sramek Study relied on roadside inspection data from the FMCSA and inferred the driver's gender based on the first name from the inspection file and the historical likelihood of someone with that name being either male or female, as reported by the Social Security Administration.

Although Song noted that because the study "did not specifically identify flatbed trucking operations, no single estimate is directly comparable to Gypsum," Song used these figures to estimate an industry benchmark range for a trucking company comparable to Gypsum.

Song considered the industry benchmark range for female drivers to be between 1.28% and 4.05% in 2019. Song then concluded that Gypsum's relevant hiring rates fell within this range.

The Court held that Song's industry benchmark clears the "liberal" relevance standard. Gypsum could offer the industry benchmark in an attempt to minimize the potency of the EEOC's statistics, a permissible and relevant purpose, or to provide an explanation, other than discrimination, for the disparity in its female hiring rate. Certainly, evidence that seeks to disprove the existence of discrimination or provide a nondiscriminatory explanation for the apparent discrimination would be relevant to a pattern-or-practice claim. And Song's proposed testimony about the industry benchmark does just that with dots that are easy to connect.

The chain of reasoning goes like this: There is evidence showing that flatbed driving jobs are more labor-intensive and more physically demanding—and therefore less desired—than other driving jobs. In light of the greater physical demands on flatbed truck drivers, Song theorizes that female representation among flatbed truck drivers might be even lower than among the general population of truck drivers. So one would expect even a non-discriminatory company to employ female flatbed truck drivers at a rate lower than the overall rate of employment of female truck drivers generally. Thus, if Gypsum's employment of female flatbed truck drivers is on par with the trucking industry's overall rate of employment of female truck drivers generally, then it would appear all the less likely that Gypsum is engaged in discrimination. Of course, the EEOC would remain free to argue that other inferences should be drawn from the data and would also remain free to convince the fact finder that Gypsum's actual applicant data is preferable to the industry benchmark. But that does not call for the exclusion of the benchmark altogether. The industry benchmark is relevant information, and Gypsum should be able to present it and argue its significance to the jury.

II. The EEOC contended that the industry benchmark opinion did not result from reliable principles and methods

Next, the EEOC contended that the industry benchmark opinion did not result from reliable principles and methods because the Scott/Davis-Sramek Study is not a reliable authority, and even if it were, it does not prove a reliable basis for the industry benchmark opinion.

Upon review, Song's industry benchmark easily clears the reliability threshold. The EEOC acknowledged that Song based the industry benchmark on a peer-reviewed publication, the Scott/Davis-Sramek Study, and the EEOC does not call into question the methodologies or data used by the study.

Further, the EEOC has provided no reason to question the reliability of the Scott/Davis-Sramek Study. On the contrary, Gypsum notes that the study, in addition to being published in a peer-reviewed journal itself, has been cited approximately twenty times in other peer-reviewed journals.

III. The EEOC contended that the industry benchmark opinion does not reflect a reliable application of the Scott/Davis-Sramek Study to the facts

Finally, the EEOC contended that the industry benchmark opinion does not reflect a reliable application of the Scott/Davis-Sramek Study to the facts of the case because Gypsum is not "truly comparable" to any of the data provided by the study. The EEOC noted that, aside from grouping flatbed and non-flatbed drivers into the same dataset, Song relied upon the study's "broad findings, from industries as diverse as livestock carriers, U.S. mail carriers and passenger carriers."

In essence, the EEOC argued that the industry benchmark compares "apples to oranges." Gypsum retorts that the EEOC's arguments relate to the weight of Song's industry benchmark, not its admissibility. Gypsum is right. First, the EEOC's argument as to the proper anchor for the benchmark seeks, in effect, to disprove Song's conclusion and thereby show that her methodology in reaching those conclusions must be flawed. However, the Court's role here is not to determine the correctness of Song's opinion but instead simply whether it is based upon a reliable foundation.

2. Hiring Rate Opinion

The EEOC next sought exclusion of Song's "hiring rate" opinion, in which she opined that the female hiring rate at Gypsum exceeded the male hiring rate by more than 50% for a portion of the relevant time period.

Rather than presenting a basis to exclude the testimony, the EEOC's arguments are better suited for cross-examination of Song. The EEOC does not call into question the methods used by Song or the data on which she based her opinion. Instead, the EEOC essentially contended that Song's hiring rate opinion is misleading and offers only a portion of the truth. If the Court were to exclude the opinion on that basis, it would overstep its narrow "gatekeeper" function and impinge on the role of the jury.

B. Dr. Erich Cromwell's Testimony

Gypsum did not dispute that Cromwell is qualified, nor did Gypsum contest the reliability of the methodology he used. Instead, Gypsum contended that Cromwell's testimony failed to meet the third requirement of the Rule 702 test. Gypsum argued that Cromwell's testimony is not based on sufficient facts or data and thus not the product of reliable principles and methods.

Gypsum's arguments essentially boil down to a contention that Cromwell constructed an over-inclusive applicant pool, so the resulting testimony is not based on sufficient facts or data, reliable, or likely to assist the trier of fact. More specifically, Gypsum contended that Cromwell improperly included applicants that submitted less than a full application ("leads") and individuals who live outside of Gypsum's hiring areas, thus producing misleading results. Gypsum contended that Cromwell's inclusion of "leads" does not represent applicants with actual interest in working as a flatbed hire for Gypsum.

The EEOC, on the other hand, casts these "leads" in a different light. The EEOC asserted that applicants using AllTruckJobs select certain companies of interest, and then—when prompted to submit an application—the candidates select "Companies of My Choice," which sends their application to the previously selected companies of interest.

The EEOC also posits that Cromwell expressly considered the nuance of lead sources in constructing the applicant pool, specifically considering that recruiters were incentivized to solicit complete applications from candidates most likely to be hired.

Cromwell's statistical data is not so fatally flawed that the Court must exclude it. Much like the EEOC's objections to Song's opinions, Gypsum essentially lodges complaints as to the weight of Cromwell's testimony, not its admissibility.

Cromwell's applicant pool might include applicants who were never interested in a flatbed role at Gypsum, but Gypsum is welcome to argue that to the jury. Whether an expert did or did not consider all possible contributors to the disparity in hiring rate is an issue that can be addressed by cross-examination, which is "the traditional and appropriate means of attacking" statistics that Gypsum believes are not credible.

Further, Cromwell explicitly relied upon deposition testimony of Gypsum employees and documented hiring practices to construct his applicant pool. Cromwell included applicants who did not specifically indicate interest in a flatbed role because the records provided by Gypsum—and relied upon by Gypsum's own expert—indicated that applicants were hired into flatbed roles even without indicating specific interest in a flatbed role. And testimony from Gypsum's Recruiting Administrator indicated that Gypsum may have hired "leads" if they were male, so Cromwell included them. Thus, his decision as to the applicant pool has factual support in the record.

The same logic applies to Gypsum's argument regarding the inclusion of applicants outside of Gypsum's hiring areas, among other qualifications. For example, Cromwell limited his applicant pool to applicants who resided in the 22 states where Gypsum's flatbed drivers reside. Gypsum takes issue with Dr. Cromwell's inclusion of applicants that lived further than 300 miles from the nearest Gypsum terminal, contending that Gypsum would not hire such an employee. But, again, Cromwell based his assumption on facts that had support in the record. The record indicates that, although Gypsum set certain "minimum hiring standards," Gypsum deviated from these standards at times depending on its hiring needs. Thus, Cromwell's inclusion of applicants that may not have met Gypsum's stated hiring standards is supported by Gypsum's actual hiring practices. All this is to say that. Cromwell relied on far more than guesswork or "unsupported speculation."

Held

• The Court denied the Plaintiff Equal Employment Opportunity Commission's motion to limit the testimony of Dr. Chen Song.

• The Court denied Defendant Gypsum Express, Ltd.'s motion to exclude the testimony of Dr. Erich Cromwell.

Key Takeaway

The task for the district court in deciding whether an expert's opinion is reliable is not to determine whether it is correct, but rather to determine whether it rests upon a reliable foundation, as opposed to, say, unsupported speculation.

Case Details: