Opinion
No. 82907-6-I
05-15-2023
Cathy ARROYO, Appellant, v. PACIFIC MARITIME ASSOCIATION ("PMA"); International Longshore and Warehouse Union (ILWU) Local 19; in their own capacities and d/b/a Seattle Joint Port Labor Relations Committee ("JPLRC"), Respondents.
Cynthia J. Heidelberg, Breskin, Johnson & Townsend, 1000 2nd Ave. Ste. 3670, Seattle, WA, 98104-1053, for Appellant. Clemens H. Barnes, Susan Kathleen Stahlfeld, Miller Nash LLP, 605 5th Ave. S., Ste. 900, Seattle, WA, 98104, April L. Upchurch Fredrickson, Miller Nash, LLP, 2801 Alaskan Way Ste. 300, Seattle, WA, 98121-1128, Robert H. Lavitt, Barnard Iglitzin & Lavitt LLP, 18 W Mercer Street, Suite 400, Seattle, WA, 98119-3971, for Respondents.
Cynthia J. Heidelberg, Breskin, Johnson & Townsend, 1000 2nd Ave. Ste. 3670, Seattle, WA, 98104-1053, for Appellant.
Clemens H. Barnes, Susan Kathleen Stahlfeld, Miller Nash LLP, 605 5th Ave. S., Ste. 900, Seattle, WA, 98104, April L. Upchurch Fredrickson, Miller Nash, LLP, 2801 Alaskan Way Ste. 300, Seattle, WA, 98121-1128, Robert H. Lavitt, Barnard Iglitzin & Lavitt LLP, 18 W Mercer Street, Suite 400, Seattle, WA, 98119-3971, for Respondents.
PUBLISHED OPINION
Chung, J. ¶1 Cathy Arroyo sued both the Pacific Maritime Association (PMA) and International Longshore and Warehouse Union Local 19 in their own capacities and together, as the Seattle Joint Port Labor Relations Committee (JPLRC), under the Healthy Starts Act (HSA) and the Washington Law Against Discrimination (WLAD). Arroyo alleges they failed to accommodate her pregnancies and that their leave and accommodation policies had a disparate impact on women longshore workers. The trial court dismissed all of her claims on summary judgment.
¶2 We affirm dismissal of Arroyo's disparate impact claim. However, we conclude issues of material fact remain as to whether Local 19 and JPLRC are employers under the HSA, whether PMA and Local 19 could provide the accommodation Arroyo requested, and whether such accommodation would pose an undue burden. Therefore, we reverse dismissal of her HSA claim and remand for further proceedings.
FACTS
I. Background on the Longshore Industry and Dispatch System
¶3 Longshore workers at the Port of Seattle perform the essential functions of loading, unloading, and managing cargo from cargo ships. Longshore workers move the cargo to and from ships, trains, and trucks. Marine clerks perform the necessary record keeping and ensure proper positioning of cargo for loading and unloading. Most longshore work involves physically demanding and often dangerous tasks performed in a hazardous environment.
¶4 The port and longshore industry consist of several entities with a complex relationship. The International Longshore and Warehouse Union (ILWU) is the exclusive bargaining representative on the West Coast and encompasses approximately 25,000 workers at 29 ports from San Diego, CA, to Bellingham, WA. Local 19 is the local branch of ILWU at the Port of Seattle. Pacific Maritime Association (PMA) is a multi-employer association of steamship lines, stevedoring companies, and marine terminal operators on the West Coast, including the Port of Seattle.
¶5 PMA employers and the ILWU have engaged in successive collective bargaining agreements for work at West Coast ports since 1937. The Pacific Coast Longshore and Clerks’ Agreement (PCL&CA) is the current agreement and consists of the Pacific Coast Longshore Contract Document (PCLCD) and the Pacific Coast Clerks’ Contract Document (PCCCD). Local 19 operates under the PCLCD.
Marine clerks operate under the PCCCD and are represented by Local 52.
¶6 The PCLCD details the operations of the joint dispatch halls, the complex method of dispatching daily jobs at PMA member terminals. A joint port labor relations committee (JPLRC) at each PMA port oversees the joint dispatch hall system. According to Local 19, the joint dispatch hall "has enabled the Union to maintain a system where earnings and work opportunity are generally equalized, and jobs are distributed in a fair and democratic system untarnished by prejudice or favoritism." Additionally, "[c]entralized hiring provides skilled and competent workers on an as-needed basis throughout the peaks and valleys of the shipping economy and maritime schedules, without the cost of maintaining a full-time steady workforce." The system also allows for promotion of members according to experience based on seniority.
¶7 Seniority plays an important role in dispatching and in the longshore industry in general. Four categories of workers make up the longshore workforce: fully-registered Class A workers, limited-registered Class B workers, and unregistered or casual workers comprised of Identified Casuals (ID casuals) and Unidentified Casuals (unID casuals). Class A and Class B registered longshore workers receive the benefits of continuous, regular employment, including pension, health care, vacation, and holiday pay. Casuals are an "auxiliary workforce" to supplement the fluctuating labor needs of the shipping industry. Casuals are not guaranteed jobs on a daily basis. Because work opportunities are unpredictable, many casuals have second jobs.
¶8 Progressing from ID casual to B-registered status depends on the number of B slots allocated at the port. Registrations do not occur at regular intervals, and registration may not take place for years if the industry is slow. New B slots are filled by casual workers based on the number of hours they have worked as a longshore worker at that port. The casuals with the most hours worked are selected for registration. This system results in what is referred to as the "hours race" among ID casuals.
The hours worked as an unID casual count toward this "hours race."
¶9 Dispatching of daily jobs depends on this seniority system and follows a defined process described in the PCLCD. Class A workers receive first preference in dispatch to available jobs, followed by Class B workers. After all Class A and B workers have received jobs, ID casuals are dispatched. Finally, unID casuals are assigned any remaining jobs.
¶10 The PCLCD states that dispatching "shall be under the principle of low-man, low-gang, first-to-be-dispatched." This means that if an employee did not work one day, they will be dispatched first on the next day. The PCLCD specifies "there shall be no favoritism or discrimination" and requires dispatching "to equalized work opportunities as nearly as practicable, having regard for the qualifications for the work they are required to do." To accomplish this goal, Local 19 uses a "rotational peg" system to dispatch jobs.
¶11 Workers arrive at the dispatch hall for their category—Class A, B, or Casual. Each person "pegs in" by placing a small wooden peg next to their name on a clear plexiglass window or "board" that contains the list of all the workers for that category. On the ID casual rotational peg board, workers’ names remain in a fixed position regardless of seniority. On the other side of the board, the dispatcher inside the dispatch booth can see the list of those workers who are "pegged in" and looking for work. During dispatch, the dispatcher proceeds down each column of names in sequence until all jobs have been filled for the shift. At the next dispatch, the process resumes where it left off, ensuring that individuals who did not work the previous day are dispatched first. On busy days, the dispatcher may have jobs for all the ID casuals pegged in for the shift. At other times, the dispatcher may have little or no work for casuals. Normally, at least some of the ID casuals seeking work do not receive dispatch to a job.
ID casuals do not get to move their names. Names are moved only with the rare addition of new ID casuals to the list.
¶12 Dispatchers dispatch jobs following an agreed sequence established by the JPLRC—clerk supervisors first, then longshore skilled jobs, followed by basic clerks, and finally, longshore unskilled jobs. This sequence ensures that skilled work is filled first to best serve the operational needs of the PMA employers. An ID casual must accept the job offered when it is their turn in the dispatch queue; otherwise they are subject to the "24-hour" rule or "flop" penalty and are not eligible for dispatch to work until the next day.
II. Accommodations Regarding Work Assignments
¶13 The JPLRC has had to find ways to provide accommodations in accordance with the Americans with Disabilities Act (ADA) within this system. The longshore industry did not create "light-duty" jobs that can be set aside for workers with medical limitations. Instead, pregnant workers or those with temporary disabilities can apply to the JPLRC for accommodations on an individual basis. ¶14 One such accommodation the JPLRC provides is exemption from the 24-hour rule. As the PMA area director explained, "pregnant and disabled workers, without penalty, are still eligible for dispatch to a different job that day if the jobs offered earlier are jobs they are entitled to pass on due to medical restrictions resulting from a disability or pregnancy." Then, "[i]f a job is available later during dispatch that fits their limitations (and it's their turn among casuals who have been allowed to pass on jobs that day), they will be offered it." This accommodation allows a pregnant worker to refuse work that exceeds their medical restriction and instead be dispatched to a job within those limitations, if available.
¶15 However, no lighter duty jobs are held for a pregnant worker. Dispatch continues in the set pattern of assigning jobs in order of supervisor clerks, skilled jobs, basic clerk, and unskilled jobs. If all jobs fitting within a pregnant worker's limitations have been dispatched before their place in line, that worker either takes a job outside their restrictions or does not work that day. For example, if only physically demanding, unskilled jobs remain when a pregnant worker's name comes up for dispatching, the worker can refuse those jobs but then will not work that day and will not add to their hours race total. JPLRC does not give and has never given credit hours for time not worked.
Those workers who take a leave of absence for military service may qualify for hours credit based on federal law. This is the only exception.
III. Arroyo's Work History
¶16 Cathy Arroyo began working in the longshore industry in 2004, beginning as an unID casual. She became an ID casual in 2007. Between 2012 and July 2017, few B-registrations occurred. In July 2017, a significant number of B slots opened at the Port of Seattle. At the time of qualification for that registration date, Arroyo had worked 3,921.75 hours in the hours race. Arroyo missed the July 2017 B-registration by 606.5 hours. There were 31 workers with more hours who also missed registration in July 2017. Arroyo became B-registered on July 28, 2018.
¶17 During her time in the industry, Arroyo gave birth to four children. The children were born March 2005, October 2006, July 2012, and January 2018. Arroyo worked throughout the four pregnancies. In February 2012, while pregnant with her third child, Arroyo presented a note from her doctor and requested exemption from lashing and semi driving, which are among the heaviest jobs at the ports. JPLRC granted this accommodation.
Lashing is the securing of cargo for transportation to minimize shifting at sea and requires anchoring the container under tension to the vessel. Lashing is physically tough and dangerous work "that is more available than other jobs for casual longshore workers, because A-and B-registered workers have job choice and don't prefer them," according to a former PMA manager.
¶18 On November 20, 2017, while pregnant with her fourth child, Arroyo provided a doctor's note stating that she was pregnant and requesting to "[p]lease minimize her being around diesel fumes and driving the semi trucks while pregnant." Despite the doctor's directions, Arroyo accepted the offered accommodation of exemption from the 24-hour rule. Having recently missed B-registration, Arroyo did not want to lose work hours toward the next opportunity. By the time she qualified for B-registration, Arroyo had worked 5,129.75 hours in the longshore industry. ¶19 In June 2020, Arroyo filed a lawsuit against PMA and Local 19, in their own capacities and as representatives of JPLRC, alleging disparate impact and disparate treatment of pregnant longshore workers and failure to accommodate in violation of the WLAD and the HSA. She requested declaratory and injunctive relief, as well as damages "including lost past and future wages, income and earnings, and out of pocket and emotional distress damages." Arroyo dropped the WLAD disability discrimination claim and continued to pursue only the WLAD disparate impact claim and the HSA claim for failure to accommodate.
When Arroyo was pregnant in 2012, her doctor wrote a note that she was not to drive semi trucks or do any lashing. The JPLRC granted this accommodation. For the 2017-18 pregnancy, the doctor requested only minimization of exposure to diesel fumes and semi-truck driving. According to Arroyo, when she met with the JPLRC again in December 2017 to discuss pregnancy accommodations, she asked that any lighter duty jobs available on a given shift be held for her. In response, JPLRC said she would be exempt from lashing and semi driving, but clerk and lighter duty jobs would not be held for her. In contrast, JPLRC meeting minutes note that Arroyo sought exemption from the 24-hour rule for her 2017-18 pregnancy. There was no mention of minimizing diesel fumes and semi-truck driving, or exemption from lashing. A union representative on the JPLRC recalled that he discussed various accommodation options but Arroyo wanted exemption only from the 24-hour rule.
¶20 PMA/Local 19 and Arroyo filed cross-motions for summary judgment on the HSA accommodation claim. PMA/Local 19 requested dismissal of the claim; Arroyo requested partial summary judgment on the issue of liability on the basis that PMA/Local 19 failed to reasonably accommodate her pregnancies as a matter of law. The trial court denied Arroyo's motion and granted PMA/Local 19's motion for summary judgment on the HSA claim. PMA/Local 19 then moved for summary judgment on the WLAD disparate impact claim. The trial court granted summary judgment on the disparate impact claim and dismissed all of Arroyo's claims with prejudice. She appeals the dismissal of both her HSA accommodation claim and her WLAD disparate impact claim.
For the purposes of this opinion, PMA/Local 19 refers to the respondents collectively (PMA, Local 19, and JPLRC) where the entities share a common argument. The entities will be addressed individually where appropriate.
Arroyo originally filed a motion for discretionary review on the HSA claim. After the trial court granted summary judgment on the WLAD claim and dismissed all claims with prejudice, the case became appealable as of right. Arroyo filed an amended notice of appeal and requested conversion from notice of discretionary review to notice of appeal, which was granted.
ANALYSIS
¶21 We review orders on summary judgment de novo. Kim v. Lakeside Adult Family Home, 185 Wash.2d 532, 547, 374 P.3d 121 (2016). Summary judgment is appropriate when there is no genuine issue of material fact and the moving party is entitled to judgment as a matter of law. Folsom v. Burger King, 135 Wash.2d 658, 663, 958 P.2d 301 (1998) (citing CR 56(c) ). We consider the evidence and reasonable inferences in the light most favorable to the nonmoving party. Kim, 185 Wash.2d at 547, 374 P.3d 121. To defeat summary judgment, the opposing party must set forth specific facts showing a genuine issue of material fact. In re Estate of Black, 153 Wash.2d 152, 161, 102 P.3d 796 (2004).
¶22 We discuss Arroyo's HSA accommodation claim first, then her WLAD disparate impact claim.
The HSA became effective on July 23, 2017. Because statutes are presumed to operate prospectively, absent evidence of legislative intent to the contrary, any alleged events that occurred before July 23, 2017, cannot be violations of the HSA. See Matter of Est. of Burns, 131 Wash.2d 104, 110, 928 P.2d 1094 (1997). Therefore, Arroyo's HSA claim pertains only to the pregnancy as to which she provided notice in November 2017. After announcing her pregnancy, there were at least 12 shifts during which Arroyo was entitled to a job, but because the clerk positions that were available to casual IDs were dispatched prior to her, in each instance, she accepted a physically demanding semi-truck driving or lasher job.
¶23 Arroyo contends the trial court erred by dismissing her HSA claim because PMA/Local 19 failed to provide her with her requested accommodation, which she contends was a statutorily defined reasonable accommodation. PMA/Local 19 claims the accommodations they provided were reasonable and Arroyo's requested accommodations would have created undue hardship.
¶24 Statutory interpretation is a question of law reviewed de novo. HomeStreet, Inc. v. Dep't of Revenue, 166 Wash.2d 444, 451, 210 P.3d 297 (2009). The primary objective of statutory construction is to ascertain and carry out the intent of the legislature. Id. When interpreting a statute, we look first to the plain language and end the inquiry if the plain language is subject to only one interpretation. Id. ¶25 The HSA states that is "an unfair practice" for an employer to "fail or refuse to make reasonable accommodation for an employee for pregnancy, unless the employer can demonstrate that doing so would impose an undue hardship on the employer's program, enterprise, or business." RCW 43.10.005(2)(a). The HSA defines "reasonable accommodation" for pregnancy as follows:
¶26 "Reasonable accommodation" means:
(i) Providing more frequent, longer, or flexible restroom breaks;
(ii) Modifying a no food or drink policy;
(iii) Job restructuring, part-time or modified work schedules, reassignment to a vacant position, or acquiring or modifying equipment, devices, or an employee's work station;
(iv) Providing seating or allowing the employee to sit more frequently if her job requires her to stand;
(v) Providing for a temporary transfer to a less strenuous or less hazardous position;
(vi) Providing assistance with manual labor and limits on lifting;
(vii) Scheduling flexibility for prenatal visits.
(viii) Providing reasonable [time and place] to express breast milk for two years after the child's birth ....
RCW 43.10.005(1)(c)(i)-(viii). The definition also includes a catch-all provision, so that a reasonable accommodation may also be "[a]ny further pregnancy accommodation an employee may request, and to which an employer must give reasonable consideration in consultation with information provided on pregnancy accommodation by the department of labor and industries or the attending health care provider of the employee." RCW 43.10.005(1)(c)(ix).
A. Local 19 and JPLRC as "Employers" under the HSA
¶27 Local 19 contends that neither it nor JPLRC qualifies as an employer that may be liable under the HSA. Arroyo argues that Local 19's role in dispatching means that it fits within the HSA's definition of employer. We agree with Arroyo's analysis.
Arroyo also claims that Local 19 should be collaterally estopped from arguing about its status as an employer due to a prior federal court case, Ross v. Pac. Mar. Ass'n, No. C19-1676 TSZ, 2020 WL 8996653, at *1 (W.D. Wash. Feb. 24, 2020) (court order). In that case, the federal district court concluded as a matter of law that Local 19 and the JPLRC "are employers pursuant to the WLAD because each entity had a ‘direct role’ in processing Ross's accommodation request before she set foot on the docks." However, as Arroyo fails to discuss or apply the requirements necessary for application of collateral estoppel, we will not consider this conclusory argument. See Christian v. Tohmeh, 191 Wash. App. 709, 728, 366 P.3d 16 (2015) ("We do not consider conclusory arguments. Passing treatment of an issue or lack of reasoned argument is insufficient to merit appellate review.").
¶28 The HSA refers to the WLAD for the applicable definition of "employer":
"Employer" has the same meaning as and shall be interpreted consistent with how that term is defined in RCW 49.60.040, except that for the purposes of this section only the threshold of employees must be fifteen or more.
RCW 43.10.005(1)(a). Under the WLAD,
"Employer" includes any person acting in the interest of an employer, directly or indirectly, who employs eight or more persons, and does not include any religious or sectarian organization not organized for private profit.
¶29 Local 19 is a labor union. The WLAD covers unfair practices of employers, RCW 49.60.180, separately from unfair practices of labor unions, RCW 49.60.190. WLAD defines a "labor organization" to "include[ ] any organization which exists for the purpose, in whole or in part, of dealing with employers concerning grievances or terms or conditions of employment, or for other mutual aid or protection in connection with employment." RCW 49.60.040(16). This definition defines a labor organization by its typical function of negotiating with an employer on behalf of its members. Yet nothing in the WLAD precludes labor unions from also being employers, for example, with regard to their own employees. ¶30 Like the WLAD, the HSA also does not specifically exclude labor unions from being employers. The HSA neither mentions labor unions nor incorporates the WLAD definition set out in RCW 49.60.040(16). While Local 19 fits the WLAD's definition of a "labor organization," that classification does not preclude it from also being an employer under either the WLAD, RCW 49.60.040(11), or under the HSA, RCW 43.10.005(1)(a), which incorporates the WLAD definition of "employer."
¶31 Local 19 is not the longshore workers’ employer in the traditional sense. But the WLAD definition includes "any person acting in the interest of an employer." Thus, for example, a supervisor acting in the interest of an employer can be held individually liable for discriminatory acts. Brown v. Scott Paper Worldwide Co., 143 Wash.2d 349, 358, 20 P.3d 921 (2001).
¶32 For purposes of the claims raised in this case, Local 19 acts in the interest of the employers of longshore workers. As the sole dispatcher of jobs for longshore workers, Local 19 fills the labor needs of PMA's member ports. Local 19's dispatch hall serves a critical function for the employers, as, according to Local 19, "[c]entralized hiring provides skilled and competent workers on an as-needed basis throughout the peaks and valleys of the shipping economy and maritime schedules, without the cost of maintaining a full-time steady workforce." In establishing and maintaining the dispatch system, Local 19 is "acting in the interest of an employer" as set out in RCW 49.60.040(11). Moreover, in its capacity as dispatcher, Local 19 is not performing the normal functions of a labor organization as set out in RCW 49.60.040(16).
¶33 The JPLRC, comprised of representatives from Local 19 and PMA, maintains and operates the labor dispatch hall, controls the registered list of longshore workers, and performs other duties described in the collective bargaining agreement. The JPLRC is also the entity that meets with longshore workers seeking accommodations and approves requested accommodations. As the entity that controls workplace accommodations, the JPLRC acts in the interest of the employers.
¶34 In their unique roles in the longshore worker employment and dispatch context, Local 19 and JPLRC are both "acting in the interest of [the] employer[s]" under the WLAD and HSA. It was error for the trial court to rely on Local 19's status as a labor union under the WLAD to conclude it was not an "employer" under the HSA.
The HSA's definition of "employer" also requires a threshold of fifteen or more employees. RCW 43.10.005(1)(a). Neither Arroyo nor Local 19/JPLRC addressed this element of the definition in their motions for summary judgment, and the record lacks evidence as to the number of people employed by the two entities. Thus, we do not resolve that question here.
B. Meaning of Reasonable Accommodation
¶35 Arroyo and PMA/Local 19 disagree as to whether the HSA establishes per se reasonable accommodations or allows employer discretion to determine reasonable accommodations. Arroyo contends that differences in the plain text of the HSA compared to the ADA support her claim that the HSA identifies certain accommodations for pregnant workers that are deemed reasonable. We agree with Arroyo.
¶36 When interpreting a statute, "[t]he court's fundamental objective is to ascertain and carry out the Legislature's intent, and if the statute's meaning is plain on its face, then the court must give effect to that plain meaning as an expression of legislative intent." Dep't of Ecology v. Campbell & Gwinn, L.L.C., 146 Wash.2d 1, 9-10, 43 P.3d 4 (2002). "We consider the ordinary meaning of the words, any statutory definitions provided, the context of the statute, related provisions, and the statutory scheme as a whole." Estate of Ackerley v. Dep't of Revenue, 187 Wash.2d 906, 910, 389 P.3d 583 (2017).
¶37 The HSA affirmatively states that " ‘Reasonable accommodation’ means" those accommodations listed in RCW 43.10.005(1)(c)(i)-(ix). This definition of "reasonable accommodation" in the HSA fundamentally differs from the ADA. Under the ADA:12 ¶38 The term "reasonable accommodation" may include
(A) making existing facilities used by employees readily accessible to and usable by individuals with disabilities; and
(B) job restructuring, part-time or modified work schedules, reassignment to a vacant position, acquisition or modification of equipment or devices, appropriate adjustment or modifications of examinations, training materials or policies, the provision of qualified readers or interpreters, and other similar accommodations for individuals with disabilities.
42 U.S.C. § 12111(9) (emphasis added).
¶39 The ADA's use of "may include" means the listed accommodations are examples of reasonable accommodations, but none is specifically required. Under the ADA and WLAD, "[a]n employer need not necessarily grant an employee's specific request for accommodation. Rather, an employer need only ‘reasonably’ accommodate the disability." Pulcino v. Fed. Express Corp., 141 Wash.2d 629, 643, 9 P.3d 787 (2000). For the purpose of accommodating a disability, the duty to accommodate is limited to "those steps reasonably necessary to enable the employee to perform his or her job." Doe v. Boeing Co., 121 Wash.2d 8, 18, 846 P.2d 531 (1993). "Generally, the best way for the employer and employee to determine a reasonable accommodation is through a flexible, interactive process." Frisino v. Seattle Sch. Dist. No. 1, 160 Wash. App. 765, 779, 249 P.3d 1044 (2011) (construing WLAD).
¶40 By contrast, the HSA defines certain kinds of accommodations intended to support a person's ability to continue to work during pregnancy. The accommodations listed in RCW 43.10.005(1)(c)(i)-(viii) are, by definition, reasonable. As such, the HSA provides a list of per se reasonable accommodations. In other words, the plain language of the HSA establishes a floor of accommodations that are deemed reasonable to assist employees in continuing to work throughout pregnancy, without requiring any individualized assessment as to whether the accommodation is reasonable.
As further indication that certain accommodations must be deemed reasonable, the HSA also provides that as to some of the identified accommodations, an employer may not even claim undue hardship. RCW 43.10.005(1)(d) (referencing restroom breaks, modifications to no food and drink policies, and providing seating or allowing more frequent sitting). As to other accommodations, the employer also may not condition the accommodation on the employee's provision of written certification from a treating health care professional. RCW 43.10.005(3).
Additionally, the accommodation regarding expressing breast milk assists employees in returning to work after pregnancy. See RCW 43.10.005(1)(viii).
¶41 Considering the language of the statute as a whole, we note that the HSA's catch-all provision allows for other accommodations also to be deemed reasonable. RCW 43.10.005(1)(c)(ix) requires an employer to give "reasonable consideration" to "[a]ny further pregnancy accommodation an employee may request." This provision thus leaves flexibility for employees to request additional types of accommodations in addition to the more common, defined "reasonable accommodations" for pregnant employees.
¶42 But if an employee requests one of the per se reasonable accommodations under RCW 43.10.005(1)(c)(i)-(viii), the employer must comply unless it can demonstrate undue hardship. RCW 43.10.005(2)(a). Unlike the ADA and WLAD, the HSA requires no interactive process to determine whether an accommodation is reasonable. In other words, the employer may not reject a per se reasonable accommodation and instead propose a different accommodation in its place. Indeed, the HSA expressly states it is an "unfair practice" to "[r]equire an employee to take leave if another reasonable accommodation can be provided for the employee's pregnancy." RCW 43.10.005(2)(d).
C. Arroyo's Requested Accommodation
¶43 The parties dispute whether Arroyo's requested pregnancy accommodation, that the employer hold for her any lighter duty jobs that were available when she was eligible to work, was either a "reassignment to a vacant position" or "a temporary transfer to a less strenuous or less hazardous position" and, therefore, a per se reasonable accommodation. See RCW 43.10.005(1)(c)(iii), (v). It is undisputed that PMA/Local 19 did not provide the requested accommodation. However, PMA/Local 19 claims it provided the statutorily required accommodation through its policy that allows employees in need of light duty to be exempt from the "flop" penalty, or "24-hour rule." According to PMA/Local 19, this accommodation "allows the pregnant worker to pass on unsuitable jobs in favor of light-duty jobs still slated for dispatch—that is, light-duty jobs that are ‘vacant’ when it's the pregnant worker's turn for a work assignment." PMA/Local 19 also points to other HSA accommodations that are afforded to casual workers, such as flexibility for prenatal visits and waiver of availability standards as evidence that it has met its obligations to pregnant workers. Arroyo contends the trial court erred by ruling that these accommodations were sufficient to satisfy the employer's obligations under the HSA, when she requested an accommodation that is per se reasonable and it was not provided.
The parties also disagree as to the accommodation Arroyo requested below. After oral argument, Local 19 filed a "notice" of additional authorities claiming that Arroyo had argued for priority for lighter duty jobs available at any given dispatch, without limiting her request to only the days that she was eligible on the pegboard, such that no casual ID would be displaced from a job. Arroyo submitted a response to Local 19's filing, asserting that she has consistently requested priority for lighter duty jobs when she would have been eligible to work that day. Indeed, Arroyo's motion for summary judgment argues that Arroyo's main concern was that available light duty jobs were already dispatched by the time they reached her name. Viewing the evidence in the summary judgment record in the light most favorable to Arroyo, as we must, her requested accommodation was not so expansive that it amounted to jumping the queue on days she would not have been eligible to work based on her position on the pegboard.
After Arroyo filed her response to Local 19's statement of additional authorities, PMA filed a response to Arroyo's response. Arroyo moved to strike PMA's response as improper under RAP 10.8(c), which requires that "any response to the statement of additional authorities must be filed within seven days after the statement is filed." PMA's filing was a response to a response, and RAP 10.8(c) does not establish a right to file such a motion. Moreover, if PMA's submission is a response to Local 19's statement of additional authorities, as it was filed 14 days after Local 19's filing, it is untimely. See RAP 10.8(c). Therefore, we grant Arroyo's motion to strike PMA's response.
We note that an accommodation in the form of an exemption from availability standards—i.e., allowing a worker to take penalty-free leave—could be an unfair practice if an employee is required to accept it even though another reasonable accommodation is available. RCW 43.10.005(2)(d).
¶44 As discussed above, under the HSA, accommodation either in the form of reassignment to a vacant position or a "temporary transfer to a less strenuous or less hazardous position" is per se reasonable under RCW 49.10.005(1)(c)(iii) and (v). Whether, under the dispatch system, there were "vacant" or available positions that could have been held is a separate question.
¶45 According to PMA/Local 19, the light duty jobs Arroyo wanted were not necessarily vacant, because "[f]or each such position, there is a longshore worker who wants and can claim it based on the collectively-bargained rotational dispatch system." Arroyo contends the jobs were vacant unless others are entitled to those positions through a bona fide seniority system. From Arroyo's perspective, all longshore jobs are vacant every day. After dispatch to Class A and Class B workers, "all jobs that are left and trickle down to casual dispatch on a given shift are, by definition, ‘vacant’ – they are ‘not occupied by an incumbent.’ "
Arroyo also acknowledged that while there is no requirement under the "temporary transfer" accommodation that a position be vacant, her requested accommodation was to be "transferred" only to jobs when they were available. Thus, in this case, Arroyo makes no distinction between positions that are "available" or "vacant."
¶46 The record shows that light duty work that is available to the group of ID casuals and has not yet been assigned to a particular person has been held for a person who needs an accommodation, whether due to pregnancy or disability. PMA's Pacific Northwest Area Director described the process for dispatching a pregnant worker with accommodations:
When her turn in the dispatch line has arrived, if the dispatchers are assigning
jobs she could do, medically, that job is offered. If not, the dispatcher might, if aware a "light job" is coming up that would fit her restrictions, assign that job out to her at the time, or else have her stand to the side until that job comes up in the dispatch sequence and then assign it to her.
According to this process, a dispatcher may pull a lighter duty job out of sequence if it is within the abilities of a worker with an accommodation who is up for dispatch.
¶47 The possibility for discretion within the process of dispatching raises questions of material fact as to whether the as-yet-unassigned light duty work is the equivalent of a "vacant" position or an available "less strenuous or less hazardous position" such that these jobs must be held for assignment to a pregnant person who is high enough on the list to be matched to a job on a particular day. It is, therefore, a question for the jury whether PMA/Local 19 could provide the particular reasonable accommodation requested by Arroyo—priority for a lighter duty job if she was eligible to work based on her position on the rotational peg board.
D. Undue Hardship
¶48 PMA/Local 19 claims that Arroyo's request to hold lighter duty jobs, such as clerk jobs, imposes undue hardship because doing so would violate the Collective Bargaining Agreement (CBA). Under the HSA, failure to provide one of the per se reasonable accommodations is an unfair practice unless the employer can demonstrate undue hardship. RCW 43.10.005(2)(a). Undue hardship "means an action requiring significant difficulty or expense." RCW 43.10.005(1)(d).
¶49 Arroyo contends that "at most, undue hardship can be proven based on conflicts with bona fide seniority systems." The HSA acknowledges this limitation: "This section does not require an employer to discharge any employee, transfer any employee with more seniority, or promote any employee who is not qualified to perform the job, unless the employer does so or would do so to accommodate other classes of employees who need accommodation." RCW 43.10.005(4)(b) (emphasis added). As a result, an accommodation that conflicts with an existing seniority system may amount to an undue hardship under the HSA.
Similarly, under the ADA, the U.S. Supreme Court has determined that an accommodation is not reasonable if it would "trump the rules of a seniority system." US Airways, Inc. v. Barnett, 535 U.S. 391, 403, 122 S. Ct. 1516, 152 L. Ed. 2d 589 (2002).
¶50 The ADA and WLAD include additional definitions of "undue hardship." For example, regulations interpreting WLAD's disability provisions provide that an accommodation is considered an "undue hardship" if the cost or difficulty is unreasonable in light of "[t]he requirements of other laws and contracts, and other appropriate considerations." WAC 162-22-075 . WAC 82-56-020(9) specifically includes "deprivation of another employee's job preference or benefit guaranteed by a bona fide seniority system or collective bargaining agreement" as an undue hardship for religious accommodations. Similarly, under the ADA, "[e]mployers are not required to create new jobs, displace existing employees from their positions, or violate other employees’ rights under a collective bargaining agreement or other non-discriminatory policy in order to accommodate a[n individual with a disability]." Burns v. Coca-Cola Enterprises, Inc., 222 F.3d 247, 257 (6th Cir. 2000).
The ADA and the regulations implementing the WLAD provide the same basic definition of "undue hardship" as the HSA. 42 U.S.C. § 12111(10)(A) ; WAC 82-56-020 . In interpreting the WLAD, we may look to federal cases construing analogous federal statutes such as the ADA. Kumar v. Gate Gourmet, Inc., 180 Wash.2d 481, 494 n.19, 325 P.3d 193 (2014).
¶51 In general, dispatch of longshore workers is managed through a hierarchical, seniority-based system. Class A workers receive first preference in dispatch to available jobs, followed by Class B workers, then ID casuals and unID casuals. Progress through these ranks depends on hours worked in the industry. However, seniority or ranking does not play a role in the ID casuals dispatch system. The location of names on the casuals peg board is not based on seniority. Thus, the order of dispatch for ID casuals does not occur based on seniority, but based on the random placement of names on the board who are then dispatched to jobs in a consistent rotation. Altering the sequence of jobs presented to a set order of the names on the list, which are not in order of seniority, would not violate a seniority system.
¶52 PMA/Local 19 contend that the entire rotational peg dispatch system is a product of a CBA and, thus, an accommodation that would interfere with the bargained rights of other employees constitutes an undue hardship. The PCLCD specifies that "[d]ispatching of men and gangs shall be under the principle of low-man, low-gang, first-to-be-dispatched, except where local dispatching rules provide otherwise for dispatching of special skilled men and gangs." Within this principle, JPLRC agreed to the sequencing order of dispatching casuals as clerk supervisors, skill work, basic clerk work, and unskilled work.
¶53 But the record suggests that this sequence is not as strict as PMA/Local 19 portrays. For example, one dispatcher stated that there are numerous situations where a dispatcher exercises discretion. "I have 90 jobs, I've got 80 bodies. I can't follow normal procedure to do the best job for the employer's interest. So you go out of sequence to fill gangs as agreed to in the tripartite, minimum manning requirements." As another example, the dispatcher explained that dispatchers know that semi truck drivers need to get to the Port as quickly as possible:
They've got to go, get a semi, check in, it takes them longer to get to the job site, get ready to work than it does a basic stevedore. So I will pass that person that does not drive that day, for numerous reasons, and I will get as many semis out the door as quickly as possible. Then when I'm done with those, I'll come back and get the babysitters, the nonpriority jobs.
Thus, the sequence for dispatching has flexibility as needed to best fill the needs of the Port employers.
¶54 Given that the CBA allows the above examples of dispatching out of order to properly fill employers’ staffing needs, altering the order to dispatch lighter duty work to accommodate pregnant workers who are entitled to work on a given day would not interfere with the CBA. In fact, the terms of this CBA include that "gangs and men not assigned to gangs shall be so dispatched as to equalize their work opportunities as nearly as practicable, having regard to their qualifications for the work they are required to do." Arguably, refusing to dispatch in a manner that takes a worker's accommodations into account thwarts the stated goal of equalization of work opportunities. When the number of available jobs would entitle a pregnant casual ID to work, but the remaining jobs are outside of that worker's restrictions when their turn arrives for dispatch, they lose the opportunity to work that day.
¶55 When a pregnant casual ID would be entitled to work a shift based on the available number of jobs, an accommodation of dispatching light duty jobs out of order for the pregnant worker, while still dispatching other casual IDs whose pegs are before the pregnant worker to different jobs, appears consistent with the CBA. However, we cannot say that the accommodation is not an undue hardship as a matter of law. Whether Arroyo's requested accommodation would place an undue burden on the employer is a question for the jury. See Phillips v. City of Seattle, 111 Wash.2d 903, 911, 766 P.2d 1099 (1989) (under WLAD, whether employee's requested accommodation would create undue burden is a question of fact).
¶56 Here, given the questions of material fact, the court properly denied Arroyo's partial summary judgment on liability under the HSA. However, the trial court erred in granting summary judgment to PMA/Local 19 and dismissing the HSA claim, as there are genuine issues of fact as to whether Local 19 and JPLRC are "employers," whether PMA/Local 19 could provide the requested accommodation, and whether doing so would constitute an undue hardship. Therefore, we reverse the trial court's dismissal of Arroyo's HSA claim against both PMA and Local 19, and we remand for further proceedings.
II. WLAD Disparate Impact Claim
¶57 In addition to her HSA claim, Arroyo claims that PMA/Local 19 engaged in "disparate impact discrimination of female longshore workers through their policies and practices related to job assignments, light duty, lack of credit for hours on pregnancy and childbirth related leave, and advancement in seniority." The trial court dismissed this claim on summary judgment.
¶58 Employment discrimination claims under RCW 49.60 may be brought under one of two theories—disparate impact or disparate treatment. Oliver v. Pac. Nw. Bell Tel. Co., Inc., 106 Wash.2d 675, 678, 724 P.2d 1003 (1986). "[D]isparate impact theory enables a plaintiff to address the consequences of seemingly objective employment practices by allowing the plaintiff to prevail in an employment discrimination suit without establishing discriminatory motive." Shannon v. Pay 'N Save Corp., 104 Wash.2d 722, 727, 709 P.2d 799 (1985), abrogated on other grounds by Blair v. Wash. State Univ., 108 Wash.2d 558, 740 P.2d 1379 (1987), as recognized by Louisiana-Pacific Corp. v. Asarco Inc., 131 Wash.2d 587, 934 P.2d 685 (1997). To establish a prima facie case of disparate impact, the plaintiff must show (1) a facially neutral employment practice that (2) falls more harshly on a protected class. Kumar v. Gate Gourmet Inc., 180 Wash.2d 481, 503, 325 P.3d 193 (2014). If a plaintiff meets this threshold showing, the burden shifts to the defendant to show that the challenged employment practice serves a business necessity. Id. at 499, 325 P.3d 193. The elements and evaluation of a disparate impact claim are the same under the WLAD and federal law. Howell v. Dep't of Soc. & Health Servs., 7 Wash. App. 2d 899, 908, 436 P.3d 368 (2019).
¶59 "The establishment of a prima facie case of disparate impact is a question of proof." Shannon, 104 Wash.2d at 733, 709 P.2d 799. The standard of proof necessary to establish a disparate impact on a protected class "is whether the plaintiff has produced evidence sufficient to justify an inference that the employment practices complained of caused a substantial disproportionate exclusionary impact on the protected class, other than by mere chance." Oliver, 106 Wash.2d at 681, 724 P.2d 1003. The primary means of proving a substantial disproportionate impact is through statistical evidence. Id. at 682, 724 P.2d 1003.
¶60 Arroyo argues that her expert provided sufficient statistical analysis to establish a "battle of the experts" for resolution by a jury. PMA/Local 19 contends the statistical evidence is unreliable based on its small sample size and reliance on incorrect data and also that it fails to support an inference of disparate impact on pregnant workers.
¶61 Arroyo introduced statistical analysis from expert Paul Torelli. He considered a group of five pregnant longshore workers who had requested accommodations and calculated an average of 9.8 years to progress from unID casual to Class B. A population of 562 comparators needed 6.7 years to achieve B-registration. Comparing the two groups showed a disparity of 3.1 years, which is statistically significant at a 90% level of confidence. Dr. Torelli concluded the result was "consistent with the allegations that the defendant engaged in policies and practices that constitute disparate impact discrimination towards female longshore workers who became pregnant."
This was the third set of calculations provided by Dr. Torelli. His first analysis calculated the averages for four women based on the time it took them to progress from ID casual to B-registration. He arrived at an average of 8.8 years for this progression for the pregnant longshore workers, but only 3.8 years for 998 non-pregnant workers. This amounted to a disparity of 5.0 years longer to reach B-registration, which he calculated at a 99% confidence level. After identifying an additional pregnant worker who requested accommodation, Dr. Torelli recalculated the averages. He determined that the five pregnant workers required an average of 7.9 years to progress from ID casual to B status. The 997 non-pregnant workers achieved B status in an average of 3.8 years. Dr. Torelli opined that the statistical disparity amounted to 4.1 years and was statistically significant at a 95% confidence level.
¶62 PMA/Local 19 submitted statistical analysis from its own expert, Peter Nickerson. Dr. Nickerson opined that Dr. Torelli's "statistical methodology and technique are flawed and are not accepted ways to analyze discrimination either in the courts or in the economics profession." He concluded that Dr. Torelli's calculation "shows that only something less than one percent of the time necessary to become a B worker could be associated with being pregnant and asking for accommodation." In performing his own calculations, Dr. Nickerson used a population of 946 people who attained B-registration and determined the mean time from unID casual to B-registration was 3,379 days. Of the five pregnant longshore workers requesting accommodations, two were below the mean time to B-registration and three were above. All five were within one standard deviation of the mean. His calculations also showed that being pregnant and asking for an accommodation, standing alone, had no impact on the length of time to reach B status.
A. Sample Size
¶63 PMA/Local 19 criticizes Arroyo's statistical evidence based on its small sample size and argues the result is unreliable. The trial court agreed that Dr. Torelli, Arroyo's expert, relied on a sample size too small "to lead to a meaningful analysis." Arroyo argues that statistical analysis should not be discounted merely due to small sample size that is beyond the control of plaintiff. We agree with Arroyo.
PMA also claims the statistical evidence was unhelpful and excludable under ER 702. But PMA fails to provide legal argument in support of this claim. Therefore, we decline to consider this conclusory argument. See Tohmeh, 191 Wash. App. at 728, 366 P.3d 16.
¶64 The Washington State Supreme Court has discussed the problematic nature of small sample sizes in disparate impact claims. "Because statistical evidence derived from an extremely small sample has little predictive value, and is therefore unreliable, the use of such evidence must be closely scrutinized to avoid inferences of disproportionality, which are based upon conjecture, speculation, or chance, rather than discriminatory practices." Oliver, 106 Wash.2d at 682, 724 P.2d 1003. While acknowledging concern about small sample sizes, neither Washington courts nor the federal courts have created a bright line rule for the size of a data set in a disparate impact discrimination case. See Watson v. Fort Worth Bank & Tr., 487 U.S. 977, 994-95, 108 S. Ct. 2777, 2788, 101 L. Ed. 2d 827 (1988) (the requirement for statistical evidence has "never been framed in terms of any rigid mathematical formula"); Freyd v. Univ. of Or., 990 F.3d 1211, 1225-26 (9th Cir. 2021) (probative value of small sample sizes is limited but no bright line rule to determine the adequacy of a data set). Rather, the size of the data set goes to the probative value of the evidence and "is a matter for the experts to debate and the jury to resolve." Freyd, 990 F.3d at 1226. This approach ensures that small sample size does not prevent consideration of discrimination claims involving underrepresented populations such as the number of pregnant longshore workers.
¶65 For disparate impact claims, statistical data is not inadmissible based merely on small sample size. Rather, sample size impacts the quality and probative value of the evidence. Here, the trial court erred to the extent it held that Dr. Torelli's statistical analysis failed because the sample size was too small. The sample size is one of several factors to consider to determine whether evidence establishes a prima facie case of disparate impact. Therefore, Dr. Torelli's statistical analysis is not deficient due only to sample size.
B. Statistical Proof of Causation
¶66 Even without considering the persuasive value of the small sample size, Dr. Torelli's analysis is not sufficient to demonstrate causation as required for a prima facie case of disparate impact.
¶67 "[T]he plaintiff's burden in establishing a prima facie case goes beyond the need to show that there are statistical disparities in the employer's work force." Watson, 487 U.S. at 994, 108 S.Ct. 2777. First, the plaintiff must identify the specific employment practice challenged. Id. After identifying the specific practice, "causation must be proved; that is, the plaintiff must offer statistical evidence of a kind and degree sufficient to show that the practice in question has caused the exclusion of applicants for jobs or promotions because of their membership in a protected group." Id.; see also Oliver, 106 Wash.2d at 681, 724 P.2d 1003 (standard of proof "is whether the plaintiff has produced evidence sufficient to justify an inference that the employment practices complained of caused a substantial disproportionate exclusionary impact on the protected class, other than by mere chance").
¶68 A disparate impact claim relying on statistical disparity fails if the plaintiff cannot point to a policy causing the disparity. Tex. Dep't of Hous. & Cmty. Affs. v. Inclusive Communities Project, Inc., 576 U.S. 519, 542, 135 S. Ct. 2507, 2523, 192 L. Ed. 2d 514 (2015). Summary judgment is appropriate when the statistics do not demonstrate causation as required to support a disparate impact analysis. See Katz v. Regents of the Univ. of Cal., 229 F.3d 831, 836 (9th Cir. 2000) ("the plaintiff's statistical evidence is insufficient to raise an inference that the disparate impact fell upon employees by virtue of their membership in a protected age group"); Rose v. Wells Fargo & Co., 902 F.2d 1417, 1425 (9th Cir. 1990) (affirming summary judgment where statistical evidence was insufficient to show a disproportionate impact).
¶69 Dr. Torelli calculated the length of time for pregnant workers who requested accommodation to reach B status compared to all other workers. After completing his statistical analysis, Dr. Torelli concluded the results were consistent with disparate impact discrimination toward female longshore workers who became pregnant. While calling it disparate impact, Dr. Torelli's analysis considers only two categories—pregnant and not pregnant. This comparison shows disparate results for pregnant people rather than disparate impact of a particular policy on pregnant people. Even though Dr. Torelli was aware of the policies at issue, he did not isolate for the specific employment practice or policy allegedly responsible for the disparate impact in his statistical analysis. Dr. Nickerson summarized the critique:
In this particular case, the policy defines your universe. So in his particular analysis, his universe is broader than what I think it should be. He's looking at the entire population instead of the people that could have been impacted by the policies.
¶70 Without providing statistics isolating for a particular policy, Arroyo cannot establish the causation needed for disparate impact. The statistics presented are too broad to show that the challenged policy was the cause for delay in B-registration for the pregnant workers. Again, Dr. Nickerson summarized the problem:
Even if Torelli had the data that looked only at individuals who were pregnant or otherwise unable to work, he does not have all the other data he needs. There is a myriad of other reasons that individuals might take longer to get to B status other than pregnancy or injury. These include working other jobs, travel, taking care of children or elders and time out of the labor market. He has no information on these variables and thus has a very under-specified model.
Using the single variable of "pregnant" yields an R-Square value of 0.0000, "meaning that the model describes none of the variation in hours it takes to reach B [status]."
The R-Square is the "coefficient of determination" which measures the proportion of the variation in the independent variable (days from unID casual to B status) that is explained by the dependent variable (being a pregnant woman who asks for accommodation). R-Square ranges from 0-1.0 with larger R-Squares preferred. The R-Square shows the percentage of variation in the data that can be explained by the dependent variable.
¶71 Arroyo's case illustrates the problem with the statistical proof of causation that arises from Dr. Torelli's failure to take into account other factors in the statistical analysis. The record produced by PMA/Local 19 shows that Arroyo fell behind in her hours worked toward B-registration between September 2012 and May 2017. These were all years during which Arroyo was neither pregnant nor taking parental leave after having a child. During those years, Arroyo took several positions working on pipelines throughout the country and for that reason, would not have been available to earn work hours towards the hours race. Dr. Torelli's statistical analysis did not consider whether alternative employment may have contributed to the lag in B-registrations.
Though we do not decide the issue on this basis, we note that this problem of individual causation also raises questions as to Arroyo's ability to bring a disparate impact claim. Standing to bring a lawsuit requires a plaintiff to have suffered an injury and have a "personal stake in the outcome of the controversy." Bacon v. Honda of Am. Mfg., Inc., 370 F.3d 565, 577 (6th Cir. 2004). As a result, "an individual plaintiff arguing a disparate impact theory must show that the challenged policy directly disadvantaged him in some fashion." Id. For individual relief, the plaintiff must establish that individual circumstances entitle them to that relief. Gilty v. Vill. of Oak Park, 919 F.2d 1247, 1255 (7th Cir. 1990).
A plaintiff must also show injury from the employment policy in order to demonstrate causation. There must be a causal nexus between the act complained of and the resulting discrimination in order for the act to be an unfair practice. Fell v. Spokane Transit Auth., 128 Wash.2d 618, 640, 911 P.2d 1319 (1996). "The causation requirement is based on the commonsense notion that if the alleged discrimination results from factors other than anything the defendant did, the defendant has not violated the Law Against Discrimination." Id.
The fact that this data may not have been available is not a reason to credit the analysis as sufficient to establish causation.
¶72 Arroyo's statistical evidence, even drawing all reasonable inferences in her favor, fails to demonstrate that PMA/Local 19's pregnancy accommodation policy caused a disparate impact on pregnant workers by increasing the length of time for them to reach B-registration. Being "consistent with discrimination" is not the same as providing proof of causation. As Dr. Nickerson noted, the statistical evidence "may also be consistent with women tak[ing] time off to be pregnant. They left the job for a particular period of time. They took other work ... There's lots of other things that could happen." The expert evidence does not use a particular challenged policy in its analysis, and, thus, fails to show a causal link between any delay in B-registrations and the particular policy. As a result, Arroyo cannot establish a prima facie case of disparate impact. No further analysis of the burden-shifting test is required. PMA/Local 19 is entitled to a judgment as a matter of law on the disparate impact claim.
CONCLUSION
¶73 We affirm the trial court's dismissal of the disparate impact claim under WLAD and reverse and remand for further proceedings on Arroyo's HSA claims.
WE CONCUR:
Diaz, J.
Mann, J.